TRC
PREHISTORIC SITE POTENTIAL AND
HISTORIC SHIPWRECKS ON THE
ATLANTIC OUTER CONTINENTAL SHELF
Authors
Ervan G. Garrison
Jessica Cook Hale
Jeffrey L. Holland
Alice R. Kelley
Joseph T. Kelley
Darrin Lowery,
Daria E. Merwin
David S. Robinson
Christopher A. Schaefer
Brian W. Thomas
Larissa A. Thomas
Gordon P. Watts
February 2011
Prepared under MMS Contract
GS-10F-0401M M09PD00024
by
TRC Environmental Corporation
Norcross, GA 30093
Published by
U.S. Department of the Interior
Bureau of Ocean Energy Management, Regulation and Enforcement
Atlantic OCS Region
iii
TABLE OF CONTENTS
LIST OF FIGURES ................................................................................................................................... vi
LIST OF TABLES ................................................................................................................................... viii
SECTION 1 – INTRODUCTION AND BACKGROUND ...................................................................... 0
1. INTRODUCTION ................................................................................................................................ 1
1.1
1.2
1.3
1.4.
1.5.
1.6.
1.7.
Prehistoric Probabilistic Model ..................................................................................................... 2
Shipwreck Inventory ...................................................................................................................... 2
Geographic and Temporal Divisions ............................................................................................. 3
Paleoshorelines and Sea Level Determination Framework ........................................................... 7
Coastal Response to Sea Level Change ....................................................................................... 12
Archaeological Sensitivity for Prehistoric Sites .......................................................................... 13
Report Structure ........................................................................................................................... 13
2. PREHISTORIC SETTLEMENT PATTERNS ............................................................................... 16
2.1. Introduction.................................................................................................................................. 16
2.2. New England ............................................................................................................................... 18
2.3.1. Paleoindian Period........................................................................................................... 21
2.3.2. Early Archaic Period ....................................................................................................... 26
2.3.3. Middle Archaic Period .................................................................................................... 27
2.4. Mid Coast..................................................................................................................................... 28
2.4.1. Paleoindian Period........................................................................................................... 29
2.4.2. Archaic Period ................................................................................................................. 33
2.5 Southeast ...................................................................................................................................... 37
2.5.1. Paleoindian Period........................................................................................................... 37
2.5.2. Archaic Period ................................................................................................................. 40
2.6. Florida .......................................................................................................................................... 44
2.6.1. Paleoindian Period........................................................................................................... 44
2.6.2. Post-Paleoindian Period .................................................................................................. 45
SECTION 2 – CURRENT RESEARCH ................................................................................................. 48
3. GULF OF MAINE.............................................................................................................................. 50
3.1.
3.2.
3.3.
3.4.
Regional Geology ........................................................................................................................ 50
Relative Sea Level Changes ........................................................................................................ 50
Marine Transgression and Site Preservation................................................................................ 53
Archaeological Sensitivity and Preservation Potential ................................................................ 60
4. SOUTHERN NEW ENGLAND AND THE GEORGES BANK .................................................... 66
4.1.
4.2.
4.3.
4.4.
Regional Geological History ....................................................................................................... 66
Relative Sea Level Changes ........................................................................................................ 71
Marine Transgression and Site Preservation................................................................................ 72
Archaeological Sensitivity and Preservation Potential ................................................................ 74
ii
5. NEW YORK AND NEW JERSEY ................................................................................................... 80
5.1.
5.2.
5.3.
5.4.
Regional Geology ........................................................................................................................ 80
Relative Sea Level Changes ........................................................................................................ 87
Marine Transgression and Site Preservation................................................................................ 88
Archaeological Sensitivity and Preservation Potential ................................................................ 90
6. MIDDLE ATLANTIC ....................................................................................................................... 96
6.1.
6.2.
6.3.
6.4.
Regional Geology ........................................................................................................................ 96
Relative Sea Level Rise ............................................................................................................... 99
Marine Transgression and Site Preservation.............................................................................. 100
Archaeological Sensitivity and Preservation Potential .............................................................. 109
7. GEORGIA BIGHT........................................................................................................................... 114
7.1.
7.2
7.3
7.4.
Regional Geology ...................................................................................................................... 114
Relative Sea Level Rise ............................................................................................................. 124
Marine Transgression and Site Preservation.............................................................................. 124
Archaeological Sensitivity and Preservation Potential .............................................................. 127
8. FLORIDA.......................................................................................................................................... 130
8.1.
8.2
8.3
8.4.
Regional Geology ...................................................................................................................... 130
Relative Sea Level Changes ...................................................................................................... 136
Marine Transgression and Site Preservation.............................................................................. 138
Archaeological Sensitivity and Preservation Potential .............................................................. 140
SECTION 3 – DISCUSSION ................................................................................................................. 145
9. DISCUSSION–SYNTHESIS OF MODEL..................................................................................... 146
9.1. Timing of Human Settlement ...................................................................................................... 146
9.2. Use of Sea Level Curves ............................................................................................................ 146
9.3. Site Location and Survival ......................................................................................................... 148
9.4. Summary .................................................................................................................................... 152
10. RECOMMENDED FIELD SURVEY METHODS ....................................................................... 155
10.1 Introduction................................................................................................................................ 155
10.2. Underwater Survey Methods ..................................................................................................... 156
10.2.1 Multibeam Bathymetry and Backscatter Intensity Data................................................ 156
10.2.2. Side Scan Sonar............................................................................................................. 156
10.2.3. Seismic Reflection Profiling ......................................................................................... 156
10.2.4. Vibracoring.................................................................................................................... 157
10.2.5. Remotely Operated Vehicles (ROVs), Autonomous Underwater Vehicles (AUVs),
Video Surveys and Submersibles .................................................................................. 158
10.2.6. Geophysical Survey Planning ....................................................................................... 158
10.3. Summary .................................................................................................................................... 159
iii
SECTION 4 – HISTORIC SHIPPING AND SHIPWRECKS ............................................................ 162
11. ATLANTIC OCS SHIPWRECK DATABASE ............................................................................. 164
11.1. Introduction................................................................................................................................ 164
11.2. Methods and Sources ................................................................................................................. 164
11.1.1. Existing Shipwreck Databases ...................................................................................... 166
11.2. U.S. Government Documents .................................................................................................... 167
11.2.1. U.S. Coast Guard, Record Group 26 ............................................................................. 167
11.2.2. U.S. Customs Service, Record Group 36 ...................................................................... 169
11.3. STATE AND FEDERAL AGENCIES ...................................................................................... 170
11.3.1. U.S. Army Corps of Engineers...................................................................................... 170
11.3.2. State Historic Preservation Offices ............................................................................... 170
11.4. PUBLISHED SOURCES AND CONTEMPORARY DOCUMENTS ..................................... 171
11.4.1. Newspapers and Magazines .......................................................................................... 171
11.4.2. Lloyd’s Lists.................................................................................................................. 171
11.4.3. Business Records........................................................................................................... 172
11.4.4. Published and Manuscript Accounts ............................................................................. 172
11.4.5. Published Shipwreck Inventories .................................................................................. 173
12. HISTORIC SHIPPING AND SHIPWRECKS ON THE ATLANTIC SEABOARD ................ 174
12.1. Seafaring During the Age of Exploration (1000–1600 A.D.) .................................................... 174
12.2. Shipping and Seafaring in the English and Dutch Colonies of the Atlantic Seaboard (1600–
1884) .......................................................................................................................................... 177
12.2.1. The Development of Shipping and Shipbuilding in Colonial America......................... 181
12.2.2. Naval Action in the Revolutionary War ........................................................................ 183
12.3. The Rise of the United States as a Maritime Power (1790–1865) ............................................. 185
12.3.1. The Birth of the U.S. Navy ........................................................................................... 185
12.3.2. American Shipbuilding Comes of Age ......................................................................... 186
12.3.3. The Transatlantic Trade ................................................................................................ 187
12.3.4. The American Schooner ................................................................................................ 188
12.3.5. The Slave Trade in the United States ............................................................................ 189
12.3.6. The Rise of Steam ......................................................................................................... 190
12.3.7. The Civil War at Sea ..................................................................................................... 193
12.4 Decline of U.S. Merchant Shipping ........................................................................................... 195
12.4.1. Losses in U.S. Waters in World War II ......................................................................... 198
12.4.2. U.S. Shipping and Maritime Activity since World War II ............................................ 199
12.5 Discussion of Vessel Types ....................................................................................................... 200
12.5.1 Sailing Vessels of the Age of Exploration ...................................................................... 201
12.5.3 Coastal vessels ................................................................................................................ 204
12.5.4 Sailing Vessels of the Mercantile Era ............................................................................. 205
12.5.5 Unpowered Vessels......................................................................................................... 208
13.5.6 Steamships of the 19th and 20th Centuries ....................................................................... 208
13.5.7 Modern Motor Vessels.................................................................................................... 210
13.5.8 Modern Sailing Vessels .................................................................................................. 212
12.5.9 Modern Work Vessels .................................................................................................... 212
12.6 Shipwrecks ................................................................................................................................. 213
12.6.1 Number of Shipwrecks in the Atlantic OCS Survey Area and Perservation Conditions 213
12.6.2 Shipwrecks by Vessel Type and Period .......................................................................... 215
12.6.3 Analysis of Shipwreck Locations ................................................................................... 217
iv
12.7 Recommendations for Cultural Resources Management of Shipwrecks in the Atlantic OCS ... 232
12.7.1 Implications of ASD for Survey Approaches ................................................................. 232
13. CONCLUSION ................................................................................................................................. 236
REFERENCES CITED .......................................................................................................................... 238
APPENDIX A – SOURCES USED FOR SHIPWRECK DATABASE
v
LIST OF FIGURES
Figure 1.1.
Figure 1.2.
Figure 1.3.
Figure 1.4.
Figure 2.1.
Figure 3.1.
Figure 3.2.
Figure 3.3.
Figure 3.4.
Figure 3.5.
Figure 3.6.
Figure 3.7.
Figure 3.8.
Figure 3.9.
Figure 3.10.
Figure 4.1.
Figure 4.2.
Figure 5.1.
Figure 5.2.
Figure 5.3.
Figure 6.1.
Figure 6.2.
Figure 6.3.
Figure 6.4.
Figure 6.5.
Prehistoric divisions and study regions used in this study. ................................................. 4
Chronological events associated with this study................................................................. 6
Illustration of isostatic rebound, whereby the land mass that was depressed by the
weight of glaciers rebounds in adjustment as the glaciers melt and the weight is
removed. ............................................................................................................................. 9
Factors influencing relative sea level rise. ........................................................................ 10
Locations of sites mentioned in this section. .................................................................... 19
Gulf of Maine study region. .............................................................................................. 68
Maine relative sea level curve modified from Kelley et al. (2010). ................................. 52
Inner Continental Shelf map of southern Maine (Barnhardt et al. 1996). The dark
blue line marks the state 3-mile boundary. Yellow represents sand deposits; blue,
mud; red, rock; and green, gravel. Location A points to the paleodelta of the
Kennebec River. A seismic line near point A is shown in Figure 3.4. B points to a
seismic line off Saco Bay and is shown in Figures 3.5 and 3.6. Location C points
to the Wells Embayment. .................................................................................................. 54
Seismic reflection profile located near Point A in Figure 1 (from Barnhardt et al.
1997:Figure 8). BR indicates bedrock; TGL represents Thin Gravel Layer; D
indicates Delta; SG indicates Sand and Gravel. VC93-02 and VC93-03 are
vibracores that contained all sand and gravel. .................................................................. 55
Inner Continental Shelf map off Saco Bay in southern Maine (Barnhardt et al.
1996). Location B is the same location represented as B in Figure 3.3. The red
line beside B is the location of a seismic line shown in Figure 3.6. Colors are the
same as in Figure 3.3. ....................................................................................................... 56
Seismic line indicated near B in Figure 3.5 (from Lee 2006). .......................................... 58
Multibeam image of submerged Bass Harbor morainal spit complex. Middle
Archaic artifacts have been recovered by draggers in this area. Modified from
Kelley et al. (2010). .......................................................................................................... 59
Side scan sonar image of moraine complex approximately 5 km offshore of
Wells, Maine. Dark areas show highly reflective bottom types (till, gravel).
Similar features are likely to exist beyond Maine’s 3-mile limit in federal waters.
Modified from Kelley and Belknap (2003). ..................................................................... 61
Archaeological Sensitivity map for the Gulf of Maine study region. ............................... 78
High Preservation Potential areas for the Gulf of Maine study region. ............................ 82
Southern New England–Georges Bank study region........................................................ 92
Archaeological Sensitivity map for the Southern New England–Georges Bank
study region....................................................................................................................... 98
New York–New Jersey study region. ............................................................................. 103
Glacial lakes in the region, ca. 18,000–12,000 B.P. ......................................................... 85
Archaeological Sensitivity map for the New York–New Jersey study region................ 110
Middle Atlantic study region. ......................................................................................... 111
High priority area off the Delmarva Peninsula flanking the Norfolk and
Washington canyons showing high potential areas for Paleoindian habitation,
including the area inundated during the MWP 1b event. ................................................ 115
Head of the Norfolk Canyon imaged with multi-beam 2008 (NOAA Chart No.
12200 used as base map)................................................................................................. 105
High potential, terrace-associated area south-southeast of Cape Lookout, North
Carolina........................................................................................................................... 106
High priority feature identified as the Cape Fear Terrace. ............................................. 107
vi
Figure 6.6.
Figure 6.7.
Archaeological Sensitivity map for Middle Atlantic study region. ................................ 119
Sea levels for specified periods, including High Potential area corresponding to
MWP 1b in the Middle Atlantic study region................................................................. 122
Figure 7.1.
Georgia Bight study region. ............................................................................................ 131
Figure 7.2.
Generalized geological section for coastal Georgia from Oligocene through
Pleistocene (adapted from Weems and Edwards [2001] and published in Garrison
et al. [2008]).................................................................................................................... 117
Figure 7.3.
A multibeam sonar image of hardbottom exposures at Gray's Reef National
Marine Sanctuary off the Georgia coast, with archaeological locales (16, 20)
shown. ...................................................................................................................................
Figure 7.4.
Thin section (left) and hand section (right) of medium-to-coarse quartzitic sands
cemented in a carbonate matrix of spar or micrite from the OCS of the Georgia
Bight. .............................................................................................................................. 120
Figure 7.5.
Model showing location of paleochannels within the Georgia Bight. ..................................
Figure 7.6.
Seismic data for the inner-to-mid shelf north of Sapelo Island, Georgia, to Hilton
Head Island, South Carolina (Henry n.d.)....................................................................... 123
Figure 7.7.
Profile of sediment core placed into a paleochannel in the Georgia Bight showing
estuarine/fluvial sediments below the transgressive sand sediments. ............................. 126
Figure 7.8.
Archaeological Sensitivity map for the Georgia Bight study region. ............................. 128
Figure 7.9.
Sea levels for specified periods, including High Potential area corresponding to
MWP 1b in the Georgia Bight study region. .................................................................. 129
Figure 8.1.
Project area along the Florida coast. .....................................................................................
Figure 8.2.
Offshore depths around Florida’s coastline today. ......................................................... 132
Figure 8.3.
Archaeological Sensitivity map for the Florida study region. ........................................ 141
Figure 8.4.
Site potential areas for archaeological sites within the High Sensitivity area in the
Florida study region. ....................................................................................................... 142
Figure 12.1. Components of a historic sailing vessel................................................................................ 201
Figure 12.2. A replica of Santa Maria, Columbus’s flagship (Creative Commons AttributionShare Alike 2.0 Generic license). ................................................................................... 202
Figure 12.3. The frigate USS Boston in the Mediterranean in 1802 (Public Domain). ............................ 203
Figure 12.4. Four-masted, double topsail ship, commonly refered to as a “tall ship.” ............................. 206
Figure 12.5. A typical schooner-rigged sailing vessel of the 19th century (Public Domain). ................... 206
Figure 12.6. Bark-rigged vessel (Public Domain). ................................................................................... 207
Figure 12.7. A typical brig of the late 19th century (Public Domain). ...................................................... 207
Figure 12.8. SS Sirius, which crossed the Atlantic in 1838 in 18 days (Public Domain). ........................ 209
Figure 12.9. Coast Guard cutter Hamilton, a typical coastal patrol boat, in the Navy Yard in
Norfolk in 1898 (courtesy of U.S. Coast Guard Historian’s Office). ............................. 212
Figure 12.10. Location of shipwrecks with coordinate data in AWOIS database. ................................... 219
Figure 12.11. Location of shipwrecks with coordinate data in the Global Wrecks database. .................. 220
Figure 12.12. Location of shipwrecks with coordinate data from primary and secondary sources. ......... 221
Figure 12.13. Location of shipwrecks with coordinate data from the existing BOEMRE database......... 222
Figure 12.14. Location of shipwrecks with coordinate data from all sources in the ASD........................ 224
Figure 12.15. Shipwreck density map for BOEMRE lease blocks in the Atlantic OCS study area. ........ 225
Figure 12.16. High, medium, and low probability zones for shipwrecks in the Atlantic OCS based
on shipwreck density. ..................................................................................................... 226
Figure 12.17. Major trade routes of colonial North America, 1769 (from Johnson 1922). ...................... 229
Figure 12.18. The steamer Metropolis was lost off Currituck Beach in 1878, 3 years after the
lighthouse was put in operation there (Library of Congress).......................................... 231
vii
LIST OF TABLES
Table 8.1.
Table 9.1.
Table 12.1.
Table 12.2.
Florida Sea Level Curves (based on Balsillie and Donoghue 2004). ............................. 136
Sea Level Curves for Atlantic OCS. ............................................................................... 147
Vessel Types in the ASD by Historical Periods. ............................................................ 216
Distribution of Shipwrecks in the ASD by State and Region. ........................................ 227
viii
ABBREVIATIONS AND ACRONYMS
EFC
ACHP
Advisory Council on Historic
Preservation
AMS
Accelerator Mass Spectrometry
ArcGIS
geographic information system
software
ASD
Atlantic OCS Shipwreck
Database
ARIF
Archaeological Resource
Information Form
AUVs
Autonomous Underwater
Vehicles
AWOIS
Automated Wreck and
Obstructions Information
System, a database of wrecks
and obstructions compiled from
hydrographic surveys and field
reports maintained by NOAA.
B.P.
years before present
BOEMRE Bureau of Ocean Energy
Management, Regulation and
Enforcement (formerly Minerals
Management Service, within the
Department of Interior)
C14
radioactive isotope of carbon
measured to calculate
radiocarbon dates.
CALYBP calendar years before present, a
format used when presenting
dates obtained from dating
techniques such as
thermoluminescence or optically
stimulated luminescence (as
opposed to radiocarbon years
before present).
CHIRP
Shallow multi-frequency
seismic sub-bottom profiling
system
CSS
Confederate States Ship,
designation for ship names in
the Confederate Navy.
DEM
Digital Elevation Model
DSF
Data Source Form
DWT
deadweight tons
GIA
GIS
GOMR
GMWD
HMS
h.p.
ICA
IRSS
LIS
LGM
LORAN
MAB
ix
Emergency Fleet Corporation,
body created by the U.S.
Shipping Board to build up the
merchant marine fleet.
Glacial Isostatic Adjustment
Geographic Information
System, a tool for integrating
spatial data with other
information to allow
sophisticated spatial and
statistical analysis and
cartography.
Gulf of Mexico region
Global Maritime Wrecks
Database, published by General
Dynamics Advanced
Information Systems with more
than 250,000 shipwreck
locations worldwide currently
included.
Her Majesty’s Ship, designation
for British royal navy ship
names.
horsepower
Institute for Conservation
Archaeology, consultant that
prepared the previous Bay of
Fundy to Cape Hatteras study of
the OCS for MMS.
International Registry of Sunken
Ships, a commercial database of
shipwrecks.
Laurentide Ice Sheet
Last glacial maximum
long range navigation, a
terrestrial radio navigation
system using low
frequency radio transmitters to
determine the location and
speed of the receiver from
multiple stations to generate
coordinate locations.
Mid-Atlantic Bight
MMS
mtDNA
MWP
NAD
NARA
NOAA
NGA
NSC
NTL
NYSCL
OCS
OCSLA
OSL
optically stimulated
luminescence, a dating
technique employing ionizing
radiation to date geological
sediments and some other
mineral based materials.
PIDBA
Paleoindian Database of the
Americas, a collection of
locational data and
measurements of attributes for
late Pleistocene and early
Holocene projectile points
reported from across North and
South America.
ROVs
remotely operated vehicles
RSL
relative sea level
RV
research vessel
SAI
Science Applications, Inc.,
consultant that prepared the
previous Cape Hatteras to Key
West study of the OCS for
MMS.
SEAMAP Spatial Ecological Analysis of
Megavertebrate Populations
SHPO
State Historic Preservation
Officer
SNE-GB Southern New England–
Georges Bank
SRTM
Shuttle Radar Topography
Mission
SS
steam ship, designation used for
commercial ships.
SWATH Small Waterline Area Twin Hull
vessel
ULCC
Ultra Large Crude Carrier
USACE
U.S. Army Corps of Engineers
USGS
U.S. Geological Survey
USS
United States Ship, designation
for commissioned ships in the
U.S. Navy.
U-Th
uranium-thorium, a radiometric
dating technique used
on calcium carbonate materials.
Minerals Management Service,
body within the Department of
Interior that regulates and
permits mineral leases and
energy development on the
outer continental shelf.
deoxyribonucleic acid contained
within mitochondria, cellular
organelles with their own
genetic material inherited only
from the female line.
Melt water pulse (period
believed to represent a rapid rise
in global sea levels)
North American Datum
National Archives and Records
Administration
National Oceanic and
Atmospheric Administration
National Geospatial-Intelligence
Agency
Non-Submarine Contact List, a
database of shipwrecks, debris,
seafloor pinnacles, and other
features maintained by the NGA
from U.S. Navy data.
Notice to Lessees and
Operators, issued from
BOEMRE on any number of
topics.
New York Shipping and
Commercial List
Outer Continental Shelf
(generally referring to the
Atlantic Outer Continental
Shelf)
Outer Continental Shelf Lands
Act, 1953 legislation that
authorized the Department of
the Interior to regulate and
permit mineral leases on the
outer continental shelf.
x
ACKNOWLEDGEMENTS
This study is the result of work by numerous individuals, without whom it would not have
been possible, including both TRC employees and outside contributors who provided valuable
data—particularly with the prehistoric modeling component of the study. Contributors to this
report include (in alphabetical order):
•
•
•
•
•
•
•
•
•
•
•
Ervan G. Garrison, Department of Anthropology and Department of Geology,
University of Georgia
Jeffrey L. Holland, TRC Environmental Corporation
Alice R. Kelley, Department of Geological Sciences, Climate Change Institute,
Department of Anthropology, University of Maine, Orono
Joseph T. Kelley, Department of Geological Sciences, Climate Change Institute,
University of Maine, Orono
Darrin Lowery, Smithsonian Institution
Daria E. Merwin, Institute of Long Island Archaeology, Department of Anthropology,
SUNY Stony Brook
David S. Robinson, Fathom Research, LLC
Christopher A. Schaefer, Southeastern Archaeological Research, Inc.
Brian W. Thomas, TRC Environmental Corporation
Larissa A. Thomas, TRC Environmental Corporation
Gordon P. Watts, Tidewater Atlantic Research, Inc.
xi
SECTION 1 – INTRODUCTION AND BACKGROUND
PAGE LEFT INTENTIONALLY BLANK
1.
INTRODUCTION
The sea is often described as the last great frontier. It is vast and opaque, and exploring its
depths requires ingenuity and daring comparable to that summoned for space flight. The sea can
be inhospitable and indifferent to human survival. From the vantage point of a ship far from
shore, one can experience vistas of tremendous emptiness in every direction, with expanses of
water meeting expanses of sky. Within this watery void, your ship is the only remnant of human
culture as far as the eye can see—a floating outpost of cultural landscape. In this light, it can be
difficult to imagine that the submerged continental shelf holds a rich archaeological record,
documenting not just the history of maritime exploration, trade, and warfare in ships that never
reached their ports, but thousands of years of prehistoric human settlement when sea levels were
lower and coastlines were miles from the modern shore. This submerged archaeological record
within federal waters along the Atlantic Seaboard is the subject of the current study.
The Atlantic Outer Continental Shelf (OCS) off the east coast of the United States extends
from the Bay of Fundy in eastern Maine to Key West at the southern tip of Florida, and
encompasses the area from the outside edge of state lands (established by the Submerged Lands
Act of 1953 as 3 miles from the shoreline) out to the edge of the Exclusive Economic Zone. The
Bureau of Ocean Energy Management, Regulation and Enforcement (BOEMRE) has
responsibility for permitting undertakings within such waters under the Outer Continental Shelf
Lands Act (OCSLA), which was signed by President Eisenhower in 1953 and authorized the
Secretary of the Interior to grant mineral leases on the outer continental shelf and create
regulations necessary to carry out the provisions of the act (Austin et al. 2004:34). The
Department of Interior assigned its OCSLA responsibilities in 1982 to the Minerals Management
Service, now BOEMRE, giving it the authority to lease areas of the OCS for activities focused
on oil and gas and non-energy minerals including sand and gravel (ELI 2009:18). The Energy
Policy Act of 2005 extended that authority to include offshore alternative energy development
such as wind, solar, and hydrokinetic projects (ELI 2009:19).
BOEMRE permitted undertakings require consultation on cultural resource stewardship
under the provisions of Section 106 of the National Historic Preservation Act. To better manage
known and potential cultural resources, BOEMRE has requested an updated study for the
Atlantic OCS that gathers information on historic shipwrecks and models the potential for
prehistoric sites based on reconstruction of past landscapes, human settlement patterns, and site
formation and preservation conditions, particularly during the period of coastal transgression.
The current study supplements two previous studies of portions of the Atlantic OCS carried out
approximately 30 years ago (Institute for Conservation Archaeology [ICA] 1979; Science
Applications, Inc. [SAI] 1981). The ICA study covered the area from the Bay of Fundy to Cape
Hatteras, while the SAI study covered the area from Cape Hatteras to Key West. Both studies
provided an overview of the geology, prehistory, and sea level rise data that may impact on
preservation of submerged prehistoric sites, as well as a predictive model on locating historic
shipwrecks. This study builds on this body of work by exploring more recent research on
prehistoric settlement patterns, archaeological research, and relative sea level curves to refine the
predictive model for locating intact, submerged prehistoric archaeological sites on the OCS. It
supplements the research on historic shipping and shipwrecks by creating a database of known
and suspected shipwrecks.
1
The goals of this study are two-fold. First, it sets out to evaluate current theories on
prehistoric settlement patterns, paleoshoreline positions, relative sea level rise, and regional
geology in order to identify potential areas on the Atlantic OCS where submerged prehistoric
sites might be located. The second goal is to provide historic context for and construct a database
of historic shipwrecks within the Atlantic OCS region. While the database has been provided
under separate cover, this report documents the research involved in assembling the database, as
well as an overview of historic shipping that provides context for the many submerged historic
resources within the project area.
To tackle this investigation, characterized by sweeping geography and topical scope, TRC
assembled a team of in-house and outside experts in history, prehistory, underwater archaeology,
geoarchaeology, and marine geology from respected institutions across the East Coast. The
product of the team’s research is a database of historic shipwreck locations, accompanied by a
summary of maritime history for the study area, and a presentation of current thinking on
submerged prehistoric sites and the coastal landscape they occupied within what is now the
Atlantic OCS, with expectations for where sites might be found and how to go about finding
them. Each component of the study, along with an introduction of key concepts, is provided
below.
1.1
PREHISTORIC PROBABILISTIC MODEL
The first component of the study involved evaluating current theories on prehistoric
settlement patterns, paleoshoreline positions, relative sea level rise, and regional geology in order
to identify potential areas on the Atlantic OCS where submerged prehistoric sites might be
located. As part of this discussion, the TRC team provides information on state-of-the-art
techniques and equipment that can be used to locate and investigate such sites.
The portion of the technical report presenting the model for submerged prehistoric sites
draws on late Pleistocene/early Holocene site and settlement pattern information from terrestrial
areas to establish “terrestrial analogs” for coastal areas submerged during the Holocene
transgression. The limited available information on known submerged sites also is discussed.
Important to the discussion is a consideration of what the coastline looked like at particular
points in the past and how sea level rose in a given area. Recent research on paleolandscapes and
sea levels is assembled for each region. The section contains a general discussion of potential site
integrity in view of the nature of various types of landforms, how they might have changed
during transgression, and taphonomy during and after inundation.
1.2
SHIPWRECK INVENTORY
The second component of the study consists of documentary and database research to
identify confirmed, reported, and potential historical archaeological resources, centered on
sunken vessels within the boundaries of the Atlantic OCS. The sources for this database included
both primary and secondary sources from a large number of repositories, institutions, and
agencies with an interest in maritime history. Available information about each wreck was
assembled into a Microsoft Access database to serve as a searchable tool that BOEMRE can use
to identify known and likely historic sites within an area of concern. A geodatabase for all entries
with coordinates within the Atlantic OCS also has been generated to work with ArcGIS.
2
The current shipwreck inventory used existing government and commercial databases. It
built upon the data previously assembled, using additional primary historic sources, as well as
secondary sources and commercial and governmental databases that catalog known shipwrecks
and unidentified obstructions that could correspond to a historic wreck. The research on primary
documents concentrated on sources with the highest potential to yield useful information, such as
life saving station records, but other less productive sources were sampled as well—documents
like newspapers, admiralty court records, and insurance claims. The quality of information
contained in the records varied, both in terms of the location of the wreck and in the amount of
historical detail and context about the ship and the circumstances of its demise. All of the
information collected was entered into a database that referenced source information. That
database is one of the primary deliverables for this study. As context for the shipwreck
inventory, TRC’s senior historian prepared a general, abbreviated maritime history of the
Atlantic Seaboard, assembled information on vessel types, and offered a model for shipwreck
potential across the OCS, as well as recommendations for survey strategies tied to the probability
model.
1.3
GEOGRAPHIC AND TEMPORAL DIVISIONS
Given the breadth of the Atlantic Seaboard, it is necessary to break the OCS down into
smaller study regions to address the changes that occurred during both the prehistoric and
geologic past (Figures 1.1 and 1.2 depict the temporal and spatial divisions referenced in this
report). Any such divisions are, by their nature, arbitrary to a certain extent. Chapter 2, which
provides an overview of the prehistory of the Atlantic Seaboard, uses divisions that follow the
cultural and environmental differences noted in the archaeological literature. While it is obvious
that prehistoric cultures did not recognize state boundaries, much of the research conducted
along the East Coast has tended to follow such divisions, with similar projectile points
sometimes having different names in neighboring states, for example. The discussion of
prehistory is grouped into four parts: New England (encompassing Maine to Connecticut), the
Mid Coast (encompassing New York to Virginia), the Southeast (the Carolinas and Georgia),
and Florida. Certainly it is possible that one could group these areas differently, but these
divisions provide a reasonable way to encompass what are broad areas of shared cultural patterns
through the Paleoindian and Archaic periods, tied in some measure to environmental variables.
The cultural history in Chapter 2 also focuses on two key temporal divisions: the Paleoindian and
Archaic periods. Combined, these periods range from approximately 13,000 to 3,000 years ago,
and encompass the limits of when human settlement would have been possible on the OCS,
which varies region to region.
Section 2 of this report discusses current research on marine transgression and archaeological
site preservation, and is also presented within a regional framework. These study regions are
somewhat arbitrary, but they are employed here because they reflect a combination of
geographical/geological distinctiveness and research histories. Chapter 3 focuses on the Gulf of
Maine along the coast of Maine, which due to glaciation has a geologic history distinct from the
Cape Cod and Georges Bank areas, which are included in Chapter 4 (Southern New England and
the Georges Bank). As one moves south from the Maine coast into Southern New England,
larger areas of the now-submerged coastline were exposed during a time when human
occupation was possible than further north. Chapter 5 includes the New York and New Jersey
3
Figure 1.1.
Prehistoric divisions and study regions used in this study.
4
5
Figure 1.2. Chronological events associated with this study.
6
coasts, which encompasses the New York Bight, centered on the Hudson Shelf Valley. The next
study region, the Middle Atlantic (Chapter 6), includes Delaware, Maryland, Virginia, North
Carolina, and a portion of northern South Carolina. This region encompasses the Delaware and
Chesapeake bays, the Albemarle Embayment in North Carolina, and the Cape Fear Arch.
Chapter 7 examines the Georgia Bight, which extends from the vicinity of Myrtle Beach, South
Carolina, to the Georgia–Florida border. Finally, Chapter 8 includes the coast of Florida from the
Georgia border to the Dry Tortugas.
There is no consistency within the literature on the dating of certain geologic timeframes and
events of interest in this study. For example, the maximum extent of the terminal glaciations in
North America, referred to as the Last Glacial Maximum (LGM), is dated differently from
source to source, anywhere from 17,000 B.P. to 22,000 B.P. The more recent literature, however,
appears to cluster between 20,000 B.P. and 22,000 B.P. For example, Duncan et al. (2000:400)
provide a date of 22,000 B.P., Otto Bliesner et al. (2005:2526) indicate the LGM was at 21,000
B.P., Yokoyama et al. (2000) indicate the LGM was at 19,000 B.P., and McHugh et al. (2010)
use 19,000–17,000 B.P. for the LGM. Given this range, a date of 20,000 B.P. has been adopted
for the LGM in this report. LGM is an important point in time, since it represents a point of
maximum subaerial exposure of the OCS. However, given the unlikelihood of human occupation
of North America at this point (see Chapter 2), uncertainty about the precise date of the LGM has
little possibility of negatively affecting site modeling on the OCS.
1.4. PALEOSHORELINES AND SEA LEVEL DETERMINATION FRAMEWORK
The Quaternary period has been a time of extreme and rapid climatic and environmental
changes. Sea level fluctuation is a critical variable in reconstructing paleoenvironments,
prehistoric settlement patterns, and subsistence patterns in North America (Murphy 1990), and
such fluctuations represent the superposition of two independent movements, including that of
the sea surface and that of the land surface (Dix et al. 2008).
The primary factors that influence global and regional sea level include changes in the
volume of sea water; tectonics; and variations in the earth’s gravitational field (Dorsey 1997). To
a lesser extent, sea level changes reflect alterations in circulation patterns and thermal regimes.
These factors can be classified in terms of their spatial extent (e.g., global versus local
processes), their temporal extent (e.g., short term versus long term), or the medium in which they
operate (e.g., vertical movements of the sea surface versus the vertical movements of the land
surface) (Dix et al. 2008).
Over a time scale of millions of years, the primary factors that influence fluctuations in the
global sea level consist of plate-tectonic-induced changes in ocean basin geometry. The long
term movement of continental and oceanic crustal plates can result in changes of up to several
hundred meters as ocean basins are created or destroyed and expand or shrink (Dix et al. 2008).
On time scales of tens of thousands of years, the periodic exchange of mass between the
Earth’s ice sheets and oceans as a result of glacial-interglacial cycles provides the dominant
contribution. This includes both eustatic and isostatic components. Eustatic refers to changes in
ocean volume and its distribution that are linked to changes in sea and terrestrial ice volume,
7
while isostatic refers to changes linked to earth surface height, which reflect tectonism (rifting,
uplift, etc.) and/or climate-driven changes such as ice volume and crustal loading. Isostasy is
most often invoked in discussions of “isostatic rebound” after deglaciation, whereby the land
mass that was depressed by the weight of glaciers rebounds or rises in adjustment as the glaciers
melt and the weight is removed (Figure 1.3). Related to this process of isostatic rebound is
subsidence along the margins of the former glaciated area, where the weight of the glaciated land
surface would have created a forebulge. As the glaciers retreated, this forebulge would go
through a process of subsidence. Thus, as the glaciers melted, glacial isostasy would involve
uplift beneath the melted ice and subsidence along the rim of the melted ice. Figure 1.4 illustrates
the correlation between sea level rise and various rates of rise or subsidence of the land
connected with glaciation. In all cases, shorter term regional variations are superimposed on top
of the longer term, global signature of sea level (Dix et al. 2008).
The main dynamic in global (eustatic) sea level is the change in the volume of oceanic waters
in response to the cyclical growth and decay of the Earth’s ice sheets. Essentially, the growth of
the ice sheets removes water from the oceans and locks it up in glaciers, thus decreasing the
global ocean volume. However, as glaciers retreat, glacial meltwater enters the ocean, increasing
the volume (Dix et al. 2008). The mechanism for the growth and decay of glaciers throughout
time has been attributed to changes in the orbital parameters of the earth, known as the
Milankovitch cycle (Weaver 2002). While global oceanic waters were tied up in continental
glaciers, vast areas of the now-submerged continental shelves worldwide were exposed.
Three primary categories of proxy sea level data can be used in reconstruction of past
landscapes: (1) global glacio-eustatic curves; (2) glacio-isostatic adjustment models; and (3)
relative sea level curves (Dix et al. 2008). Relative sea level curves, obtained directly from past
sea level indicators, such as dated corals, foraminifera, saltwater peat, intertidal oysters, or
archaeological material, represent the most accurate way of reconstructing past coastlines for a
particular region because they reflect the local impact of eustatic, isostatic, and tectonic variables
(Dix et al. 2008).
The most recent full-glacial cycle began approximately 135,000 B.P. Global sea level and
temperatures at that time were perhaps slightly higher than present levels (Donoghue 2006).
Chappell and Shackleton (1986) derived Late Quaternary sea level curves from oxygen isotope
data that demonstrate eustatic sea level fluctuations due to the expansion and melting of
continental ice sheets throughout the past 135,000 years. Between 135,000 and 20,000 B.P.,
global sea level and temperatures fluctuated but generally fell, reaching the lowest point
approximately 20,000 B.P. during the LGM. At this time, much of the world’s water existed as
ice in extensive glaciers that covered large land areas (Clark et al. 2009; Denton and Hughes
1981), and enough water was shifted to the continental glaciers to lower the world’s oceans as
much as 120–130 m (Clark et al. 2009; Dorsey 1997; Fairbanks 1989; Peltier 2005). This period
during which sea levels were at their lowest is also referred to as the lowstand.
Fairbanks’ (1989) study of Barbados corals (Acropora palmata) yielded a detailed eustatic
sea level record for the last 20,000 years. According to this record, sea level began to rise slowly
around 20,000 B.P., following the LGM. As glaciers continued to retreat, global sea levels
increased some 20 m from the LGM to 12,500 B.P., followed by a rapid period of sea level rise
8
1) Glacial period
ICE
Subsidence
2) Ice melts
Uplift
Figure 1.3.
Illustration of isostatic rebound, whereby the land mass that was depressed by the
weight of glaciers rebounds in adjustment as the glaciers melt and the weight is
removed.
9
Figure 1.4. Factors influencing relative sea level rise.
10
known as Melt Water Pulse (MWP) 1a, which Fairbanks dates to ca. 12,500–11,500 B.P. Sea
rise slowed between 11,500–10,500 B.P., a period Fairbanks associates with the Younger Dryas,
a period of cooling temperatures. This slower period of sea level rise changed abruptly with
MWP 1b, a one thousand year period when sea levels rose approximately 28 m (Fairbanks
1989:639).
The dating of some of these events described by Fairbanks, however, has since been refined
in some cases and questioned in others. For MWP 1a, Bard et al.’s (1990a, 1990b) uraniumthorium (U-Th) dating, combined with radiocarbon dates from Fairbanks (1989), provided a date
range of (14,200–13,800 B.P.), with a corresponding sea level rise of 20 m (-94 to -74 m). Lui
and Milliman (2004:187) suggest a slightly more narrow temporal range for MWP 1a (14,300–
14,000 B.P.), representing a change in sea level of 20 m (for a mean rate of 66 mm/year), while
Stanford et al. (2006) suggest a date range of ca. 14,100–13,600 for MWP 1a.
Perhaps more salient to this study is MWP 1b, which, unlike MWP 1a, corresponds to a
period when humans were known to occupy North America. Bard et al.’s (1990a, 1990b)
research indicate that MWP 1b took place from 11,500–11,100 B.P. and represents a change in
sea level from -58 m to -43 m. Lui and Milliman (2004:187) present a similar assessment of
MWP 1b, with a temporal range of 11,500–11,200 B.P., during which sea level rose from -58 m
to -45 m (for a mean annual rate of 43 mm/year). Other scholars, however, have questioned the
extent—and even the existence of—MWP 1b (see discussion in Montaggioni and Braithwaite
2009). The precise timing and amplitude of MWP 1b are still open questions for many because
this event was originally detected as a hiatus between individual drill cores collected at different
depths off Barbados, rather than being represented in a single core sample (Bard et al. 2010). In
attempting to address the existence of MWP 1b, Bard et al. (2010) dated 47 pristine coral
samples drilled onshore of the Papeete barrier reef in Tahiti using U-Th, but found no evidence
of MWP 1b in these dated coral samples. Further research is necessary to resolve this issue,
which has implications for site archaeological preservation potential.
Likewise, the dating of the Younger Dryas event has undergone some revisions based on
more recent dating techniques. As Meltzer and Holliday (2010:8) note, “the most recent, highresolution analysis—which uses isotopic analyses of deuterium excess and oxygen isotope 18 as
indicators of past ocean surface and air temperatures, respectively—indicates that Younger
Dryas cooling began 12,900 calendar years before present, with the warming starting 11,700
calendar years before present (Steffensen et al. 2008).” Until very recently, however, the
archaeological literature has used a date range of ca. 11,000–10,000 B.P. for the Younger Dryas
(e.g., Faught 2002, 2004; Mayewski and Bender 1995; Taylor et al. 1993).
Considering the revised efforts at dating these events, after ca. 11,000 B.P.—marking the
approximate end of assumed MWP 1b—sea levels again slowed. Sometime beginning 7000–
6000 B.P., the net rate of sea level rise began to slow significantly and gradually approached its
present rate (Dunbar et al. 1992; Oldale 1992; Stanley and Warme 1994).
11
1.5. COASTAL RESPONSE TO SEA LEVEL CHANGE
Coastlines are generally high-energy environments characterized by wave and tidal processes
(Waters 1992). As a result, the preservation or destruction of sites on the coastline depends on
the position of the site relative to shoreline processes (CEI 1977; Gagliano et al. 1984).
Depending on the interplay between a variety of factors including sediment supply, subsidence,
coastal processes, and tectonic activity, shorelines have: (1) transgressed landward, (2) stabilized
and maintained a neutral configuration, (3) prograded seaward, or (4) tectonically emerged or
risen above the modern sea level (Waters 1992).
Paleoindian and Archaic period archaeological sites would have been affected by coastal
processes during the last marine transgression. Transgression is primarily a destructive process
that does not create ideal depositional sequences (Belknap and Kraft 1985). Episodes of
transgression are periods of erosion. Consequently, the process of shoreline retreat is important
to site preservation.
Transgression may occur in two ways: (1) by shoreface retreat, when the coastline slowly
advances landward; or (2) by stepwise retreat, when in-place drowning of coastal features occurs
(Waters 1992). Shoreface retreat is the erosion of previously deposited sediments by wave and
current processes as the shoreline transgresses (Waters 1992). As sea level rose during the Late
Quaternary, the beachface and shoreface erosion zones sequentially passed across those portions
of the continental shelf that had been exposed. Thus, older sediments that had been deposited in
coastal and terrestrial environments behind the shoreline were reworked, first by the swash and
backwash processes and then by the waves and currents associated with the upper shoreface and
breaker zones. Reworked terrestrial and coastal sediments are referred to as palimpsest sediments
(Swift et al. 1972b), and the erosional surface, marking the depth of maximum disturbance by
transgression, is known as the ravinement surface (Belknap and Kraft 1985). Shoreface retreat is
most common in areas where the sea level rose slowly and subsidence rates were low.
A major factor determining the severity of erosion during shoreface retreat, and as a
consequence the preservation potential of Late Quaternary sediments and any contained sites, is
the rate at which sea level rises (Belknap and Kraft 1981). If the sea level rises rapidly over the
continental shelf, erosion will be of short duration and the underlying sediments will have a
greater potential for preservation (Waters 1992). Because the shifting climate produces
meltwater pulses and shoreface retreat is not held constant for the period under consideration, the
process is not only erosional all of the time. During periods of steady sea level and even
retreating water (e.g., during the Younger Dryas), identifiable beaches and shallow water
features such as wave-cut terraces or oyster bars can be produced and occasionally preserved
(Hine 1997). More important archaeologically are those preserved terrestrial or fresh water
features indicative of the actual Pleistocene landscape inhabited by the early human arrivals.
Such features can include buried river and stream channels, karst features and more developed
sinkholes, in-place soils, peats, tree stumps, higher elevation rock outcrops, and other similar
landform features.
More important than sea level rise in the potential for site preservation is the configuration of
the topography on the continental shelf prior to transgression (Belknap and Kraft 1985). If a site
12
is located and later buried in a topographic position that will not be eroded during transgression,
it will be preserved under the ravinement surface.
Other factors that influence site preservation on the continental shelf include: (1) the energy
level of the coastal processes and depth of the wave base; (2) the cohesiveness of the sediments
comprising the site matrix; (3) the amount of subsidence prior to transgression; (4) the gradient
of the continental shelf; (5) tidal range; and (6) sediment import and export processes (Waters
1992).
1.6. ARCHAEOLOGICAL SENSITIVITY FOR PREHISTORIC SITES
This report employs two ways of characterizing the likelihood that portions of the OCS have
preserved prehistoric resources. The first method uses the term “sensitivity” and addresses
physical and culture-history constraints on site formation. Under this approach, sensitivity is
constrained by when prehistoric occupation was first possible and then likely, and is
geographically defined by reference to sea level curves used in each region that correspond to
these temporal events. This approach divides the OCS in each region into three sensitivity
categories, as described below:
1. No Sensitivity: Areas that were not subaerial at the LGM. Since such areas were always
submerged, they have no potential for containing terrestrial sites.
2. Low Sensitivity: Areas that were subaerial between the LGM and the Paleoindian period,
representing a time when it is unlikely—although possible—that human settlement of
eastern North America existed.
3. High Sensitivity: Areas that were subaerial beginning with the Paleoindian period to the
present.
The second method of characterizing the likelihood of preserved prehistoric sites uses the
concept of preservation potential. Areas of High Preservation Potential represent locations within
the High Sensitivity areas where conditions exist that provide a better likelihood that prehistoric
sites would have survived marine transgression. In some regions, there may be no specifically
known or mapped High Preservation Potential areas designated, although characteristics that
would define such areas (hence, provide the conditions that surveys would attempt to identify)
are provided. Throughout most of the OCS, detailed studies of geomorphology using seismic
sub-bottom profiling and coring in conjunction with bathymetry are needed to define the
character of submerged landforms and create more fine-grained mapping of potential intact site
settings. Over time, such information may become available for more areas—perhaps in part
through investigations prompted by offshore energy development—and mapping of High
Preservation Potential areas can be refined.
1.7. REPORT STRUCTURE
This report is divided into four sections, each containing thematically-related chapters.
Section 1 includes this introduction (Chapter 1) and an overview of the prehistory of the Atlantic
Seaboard (Chapter 2). Section 2 contains Chapters 3–8, which are contributions to current
13
research for different areas within the project area, from Maine to Florida. Drawing on Sections
1 and 2, Section 3 synthesizes the prehistoric modeling introduced in the regional chapters
(Chapter 9) and presents proposed survey methods for identifying locations for prehistoric sites
(Chapter 10). Finally, Section 4 covers work to develop the Atlantic OCS shipwreck database,
including a discussion of sources and methods for research (Chapter 11) and a historical
overview of shipping on the Atlantic Seaboard (Chapter 12).
14
PAGE LEFT INTENTIONALLY BLANK
15
2.
PREHISTORIC SETTLEMENT PATTERNS
2.1. INTRODUCTION
Despite decades of archaeological research, our understanding of the earliest humans to
occupy what is now the eastern coast of the United States remains limited. The earliest, most
broadly acknowledged human presence in the continental United States dates to approximately
12,500 B.P., during the Paleoindian period. The most well-known cultural manifestation of this
early settlement is called Clovis, which is represented archaeologically by distinctive, fluted
projectile points that have been found over a wide geographic area in the United States.
However, for decades there have been sites that indicate, if not conclusively prove, a pre-Clovis
occupation in the Americas; these include Meadowcroft Rockshelter, Pennsylvania (Adavasio et
al. 1990, 1998, 1999); Saltville, Virginia (McDonald 2000; Wisner 1996); Cactus Hill, Virginia
(McAvoy and McAvoy 1997); the Topper site in South Carolina (Goodyear and Steffy 2003);
and the Sloth Hole and Page-Ladson sites in Jefferson County, Florida (Dunbar 2002, 2006b;
Hemmings 1999, 2004). None of these sites is without controversy, but they have forced
archaeologists to revisit their models for how and when people first arrived in the Americas (e.g.,
Anderson and Gillam 2000).
Most archaeologists accept that the human occupation of North America began with a
migration of people from Asia across the Bering land bridge, which would have been exposed
from 20,000 B.P. to a time perhaps as late as 10,000 B.P. due to lower sea levels associated with
the LGM (Anderson and Gillam 2000; Dixon 1999, 2001; Fladmark 1979; Hoffecker et al.
1993:48; Meltzer 1988, 2004; Smith 1986). Once in North America, the method and timing of
migration south into the Americas remains an issue of debate. Some researchers have argued that
an ice-free corridor allowed for movement into the interior of the continent sometime after
11,000 B.P. (e.g., Haynes 1966, 1969, 1971), while others have suggested that early settlers,
once having occupied Beringia, followed a coastal route to colonize the Americas (e.g., Dixon
1999; Faught 2008; Fiedel 2000; Fladmark 1979).
Given the fact that sites that might confirm a coastal migration are almost certainly now
inundated, it may be impossible to demonstrate which route accounts for the settlement of the
continent. However, researchers have evaluated models of migration by testing them against
those data that are available. For example, Goebel et al. (2008) present a working model to
explain the origins of human occupation of the Americas that draws on both genetic and
archaeological evidence. They first summarize the results of genetic testing of contemporary
Native American populations, including nuclear gene markers, mitochondrial DNA (mtDNA),
and Y chromosomes, which demonstrate a genetic connection to contemporary, indigenous
populations of southern Siberia (Goebel et al. 2008:1497). DNA tests conducted on early skeletal
remains and human coprolites also support an Asian origin. Interestingly, analysis of genetic
variation in contemporary Native Americans, particularly certain subclades of mtDNA found in
Native American groups throughout North, Central, and South America—but not in Asian
populations—suggests common American ancestors approximately 16,600–11,200 B.P. (Goebel
et al. 2008:1498).
16
Fix (2002) has noted that the mtDNA data make it difficult to accept the notion of Clovis
people migrating south from Beringia through the ice-free corridor and spreading rapidly
throughout the Americas (cf. Martin 1973), since the timing of such a passage would have been
too late to account for the genetic variability observed. He notes that a model of coastal
migration, which presumably would have allowed for an earlier start of settlement (perhaps as
early as 16,000 years ago) is consistent with the genetic variability in contemporary Native
American groups throughout the Americas (Fix 2005). This settlement model assumes that
people moved down the Pacific coast to the narrow isthmus of Central America, crossed the
isthmus, and continuing to spread up the Gulf and Atlantic coasts, with the Mississippi River
serving as an entry to the continental interior (Fix 2005:432). Based on this model, the Eastern
Seaboard of North America could have been settled within 3,000 years (Fix 2005:432–433).
Others have suggested that Clovis derived not from Asia, but from the Upper Paleolithic
Solutrean culture of Europe, which dates to ca. 22,000–16,500 years ago (Bradley and Stanford
2004:465; Sellet 1998). Noting that there are no known pre-cursors to Clovis in Alaska or eastern
Asia, they suggest Solutrean maritime hunters entering the Atlantic coast of North America may
account for the handful of early, pre-Clovis sites on the Eastern Seaboard. They point to a
number of factors to support their hypothesis, including similarities in tool manufacturing
techniques and artifact forms between Solutrean and Clovis tools, temporal consistency, and a
plausible migration route to North America.
The logistical problems a founding population would have encountered traversing the North
Atlantic Ocean, the lack of early occupation sites above about 48 degrees north latitude, and a
gap of at least 5,000 years between Solutrean sites in Iberia and early sites in eastern North
America, suggest that any resemblance in bone and lithic tools between the two cultures is
coincidental, and not indicative of direct contact (Straus 2000; Straus et al. 2005). Furthermore,
genetic data also indicate an Asian origin for Native American populations (Fix 2005; Straus et
al. 2005:522–523). Still, should the North Atlantic migration route be shown to have been viable,
the continental shelf off the northeastern U.S. would be a logical place to search for evidence
(Stright 2004).
One study of Paleoindian settlement patterns resulted in a model to explain “routes, rates, and
reasons” (Anderson and Gillam 2000:43) for colonization of the Western Hemisphere. The study
analyzed paths at a continental scale, to determine which routes would have afforded the least
cost to traveling hunter-gatherers. Factors in the model included topographic relief, locations of
ice sheets and pluvial lakes, and the location of known Paleoindian archaeological sites. The
findings suggest that initial dispersal occurred in coastal and riverine settings and on plains, and
that founding populations probably spread and diversified rapidly. In terms of routes, Anderson
and Gillam’s model implies that now-submerged portions of the continental shelf may have been
important for early dispersal, whether by foot or by boat. In eastern North America, this is
reflected in the distribution of sites along the Atlantic Coastal Plain and the paucity of sites in the
Appalachian Mountains, which were a barrier to mobility.
One of the challenges of any of these models of population spread into the Americas is
accounting for sites that appear to predate Clovis. While none of the referenced North American
sites is universally accepted by scholars, the finds at Monte Verde, Chile, appear to have
17
convinced most skeptics (Meltzer et al. 1997). The earliest securely dated Paleoindian stratum at
Monte Verde dates as early as ca. 13,800 B.P. (Dillehay 1989), so models that cannot explain
outliers such as this site are problematic.
What is known about the early occupation of the Eastern Seaboard derives not from models,
but empirical data from field research up and down the coast. The following sections present
summaries of that early settlement, including the Paleoindian and Archaic periods—periods
when settlement of the OCS was feasible, based on what is known about Late Pleistocene/Early
Holocene sea levels along the Atlantic coast. The dates for these periods vary up and down the
coast, based on the extent to which dated contexts are available. As a general framework, the
Paleoindian period dates roughly from 13,000–10,000 B.P., while the Archaic period ranges
roughly from 10,000–3,000 B.P. Regional culture histories below refine these date ranges, as
appropriate. All dates presented here, unless noted otherwise, are given as uncalibrated years
before present (B.P.).
Drawing geographic lines to demarcate areas of culture history has always been a challenge.
Not only is one faced with cases of gradual material culture variation that must be geographically
parsed, but it is also necessary to take into account thousands of years of cultural developments
in which cultural expressions and affiliations emerge differentially across space and time.
Further complicating the task, regional similarities in prehistoric developments have been
masked to some extent by typological nomenclature influenced by where research has been
conducted and the spheres in which researchers operate, often following state boundaries. The
following geographic divisions have been defined to encompass broad areas of shared cultural
patterns through the Paleoindian and Archaic periods, tied in some measure to environmental
variables. Section 2.2 focuses on New England, including Maine, New Hampshire,
Massachusetts, Rhode Island, and Connecticut, where similar climate and environmental
conditions correspond with broadly similar cultural developments. Section 2.3 covers a larger
geography that encompasses coastal New York, New Jersey, Delaware, Maryland, and Virginia.
This area was populated by prehistoric groups whose archaeological record is more similar
internally than to regions north and south. While archaeologists would refer to most of this
region, with the possible exception of New York’s Long Island, as the Mid-Atlantic, no
geographic name effectively encompasses these states. Therefore, to avoid any confusion, the
current report refers to this area as the Mid Coast region. Section 2.4 discusses the prehistory of
the Southeast, which for purposes of this study includes North Carolina, South Carolina, and
Georgia, which generally follow the same patterns of prehistoric developments and display the
same types of material culture. Finally, Section 2.5 includes the culture history of peninsular
Florida. All of the sites mentioned in the text are plotted in Figure 2.1.
2.2. NEW ENGLAND
Archaeologists have documented over 12,000 years of human settlement in the terrestrial
terrain of New England. The archaeological record of ancient Native American habitation in the
Northeast is commonly divided into three general temporal periods: the Paleoindian, the Archaic
and the Woodland. A “Late” period is sometimes included in discussions about the Paleoindian
period, and the Archaic and Woodland periods are both further subdivided into Early, Middle
and Late categories. In addition, the Late Archaic and the Early Woodland periods are separated
18
Figure 2.1.
Locations of sites mentioned in this section.
19
by a distinct transitional period referred to as the “Transitional” or the “Terminal” Archaic. Each
division among the general periods of the ancient Native American cultural chronology is based
on the interpretations of the archaeological record. These periods are distinguishable within
thearchaeological record on the basis of observed differences in the material culture, specific
land use patterns inferred from the archaeological remains of the material culture, and,
occasionally, by other indicators, such as mortuary practices.
The ancient Native American cultures of the early pre-contact period corresponding with the
ca. 12,500–6000 B.P. period, when portions of the region were subaerially exposed and available
for human habitation include:
•
Paleoindian period (ca. 12,500–10,000 B.P.)
•
Early Archaic period (ca. 10,000–7500 B.P.)
•
Middle Archaic period (ca. 7500–5000 B.P.)
Up until a little over a decade ago, ancient Native American artifacts and/or documented
archaeological sites dating from the Paleoindian and Early and Middle Archaic periods along the
coastal plain were quite rare. This lack of archaeological data initially led archaeologists of the
1960s to conclude that the Northeast’s “closed boreal forests” of the post-Pleistocene could
support few human foragers, and that these unfavorable environmental conditions had resulted in
an apparent depopulation of the Northeast at that time (i.e., known as the “Ritchie-Fitting
hypothesis”) (Fitting 1968; Ritchie 1980). The Ritchie-Fitting hypothesis was confronted with
newer palynological data that indicated that the environment of New England, especially
southern New England, was more amenable to habitation than had been suggested previously,
and the Ritchie-Fitting hypothesis has since been rejected (Dincauze and Mulholland 1977; Jones
1998; Robinson and Petersen 1992).
While sites dating from the Paleoindian and Early and Middle Archaic periods remain rare
compared to later Woodland Period sites along the Atlantic Coastal Plain, archaeological
investigations in southern New England in the last 25 years have dramatically increased our
existing knowledge about ancient Native American settlement patterns and resource procurement
strategies (Carr 1996; Cross 1999; Doucette and Cross 1998; Dunford 1999; Forrest 1999;
Gardner 1987; Jones 1998; Jones and Forrest 2003). These archaeological data indicate that there
was a complex transition of cultures from the time of the arrival of the first Paleoindian colonists
to the florescence of the Middle Archaic populations some 3,000 years later (Jones 1998). These
studies also have brought into question the adequacy of current terrestrial archaeological survey
paradigms for locating sites from these periods in New England. Jones (1998) has opined that
limitations of archaeological testing strategies commonly used on land in the Northeast have
biased the current record of Paleoindian and Early Archaic sites towards medium to large interior
camps—sites that are probably not representative of the range of site types produced by huntergatherers of the terminal Pleistocene and early Holocene.
The rapidly changing environment that characterizes the late Pleistocene to early Holocene
time period hypothetically should have produced an archaeological record of site types that is
20
highly variable, because of the need for flexible responses in social and economic behavior to the
environmental conditions (Jones 1998). Increasingly, evidence of lowered water levels and an
emergent correlation between large wetlands and major water bodies and ancient Native
American archaeological sites suggests that water (inland and coastal) and its associated food
resources were critical factors in site selection. Hypotheses are now proposed that assert large
Early and Middle Archaic period archaeological sites in proximity to large lakes, rivers and
extensive wetlands with inlets and outlets flushing their respective systems may have been more
common on the Coastal Plain, but were submerged by the rising sea level (McWeeny and
Kellogg 2001). It is perhaps for this reason that certain site types (especially large coastal
occupations and very small interior camps and extraction locations) seem to be lacking or are
very rare in the archaeological record (Jones 1998). With very few exceptions, virtually all
documented Paleoindian and Early Archaic finds reported throughout New England lack detailed
contextual information.
2.3.1. Paleoindian Period
Following the retreat of thick glacial ice, the present terrestrial landscape, as well as the thensubaerially exposed portions of the OCS in New England, were probably inhabited by a
relatively low population of mobile hunter-gatherers employing a specialized tool kit developed
for the exploitation of large migratory game (e.g., caribou, elk, bison, and mastodon) (Dragoo
1976; Kelly and Todd 1988; Snow 1980; Waguespack and Surovell 2003). In particular,
Paleoindian people living in the region are thought to have relied mainly on caribou that
presumably were abundant in the environment of that time (Spiess et al. 1998). The presence of
these early inhabitants is recognized in the archaeological record through distinctive lithic
technologies.
The Paleoindian period in the Northeast is divided into two temporal groupings, or traditions.
Diagnostic fluted projectile points and related artifacts characterize sites of the Fluted Point
Paleoindian Tradition (12,500–10,050 B.P.) (Spiess 1990; Spiess et al. 1998). The elongate,
bifacial points have a distinctive flake scar, or flute, created by the removal of a channel flake
from one or both faces of the tool (Snow 1980). Other characteristic artifacts include unifacial
endscrapers, usually made from a single flake with a spur, or graver, on one or both ends (Snow
1980; Spiess et al. 1998). Pièces ésquillées, created from thick flakes or core fragments, are also
typical of the Paleoindian tool kit (Snow 1980). The Late Paleoindian Tradition (10,050–9,500
B.P. is distinguished by a change in bifacial technology from fluted points to parallel-flaked
lanceolate points, similar in form to Scottsbluff or Eden points of the Midwest and western U.S.
(Cox and Petersen 1997; Doyle et al. 1985; Spiess et al. 1998; Will and Moore 2002).
Throughout the Paleoindian period, artifact assemblages tend to feature non-local lithic
materials, such as chert and jasper and regionally and extra-regionally available rhyolites (e.g.,
Mount Jasper rhyolite, Lynn volcanic suite, Saugus Jasper, Munsungun Formation chert, etc.). A
marked preference for fine-grained crystalline material is reflected Paleoindian lithic technology.
Raw materials observed in artifacts recovered from sites in New England include chert,
chalcedony, jasper, quartzite, crystal quartz, and fine-grained volcanic rocks, such as rhyolite and
felsites, often found hundreds of kilometers from their primary source (Spiess et al. 1998). These
21
great distances may represent long distance travel to established source areas, trade, and/or
utilization of fluvially or glacially transported material.
Recognition of a variation in point styles has led to the establishment of a typology and
chronology for regional Paleoindian fluted points. Spiess and others (1998) developed a four-part
typology, while Dincauze’s (2007) analysis produced a slightly different division. Both schemes
use the Great Lakes region typology as a basis (see review by Ellis and Deller 1997) for New
England divisions. The sequences are based on variations in the length/width ratio of the
projectile, size and shape of the basal concavity, size of the flute scar, and the presence or
absence of “fish tail” forms at the base of the point. Chronology was established using
radiocarbon dates, when possible, and relative position on lakebeds and stepped shorelines in the
Great Lakes. Each phase is named for the site whose artifacts best represent the type’s attributes.
In New England, Spiess and others (1998) recognize the oldest phase as the Bull Brook/VailDebert, named for the fluted point types recovered from the Bull Brook site in Massachusetts,
the Vail site in Maine, and the Debert Site in Nova Scotia. This point style is parallel-sided to
lanceolate in shape with a medium to deep basal concavity, and flute scars that extend along half
the length of the point (Spiess et al. 1998). The phase is dated at a number of sites across New
England and Nova Scotia, and represents the time period from 10,800–10,500 B.P. (Spiess et al.
1998). Dincauze (2007) lists Bull Brook as the oldest period, comparable to the Gainey phase of
the Great Lakes, with Vail-Debert as stylistically different, but also representing the initial
Paleoindian occupation of the region.
The next youngest Paleoindian phase in Spiess and others’ (1998) typology is the MichaudNeponset, and it is best characterized by points from the Michaud site in Maine and the Neponset
site in Massachusetts. These points are narrow and thinner than those of the preceding phase, and
have a flute scar that extends along most of the entire length of the tool (Ellis and Deller, 1997).
The basal concavity is shallow, and the base is typified by “fishtails” or “flaring ears” (Spiess et
al. 1998). This style is similar to the Barnes points of the Parkhill Phase of the Great Lakes and
Mid-Paleoindian of the Mid-Atlantic. The Michaud-Neponset phase is associated with dates
close to 10,700–10,300 B.P. (Spiess et al. 1998). Dincauze (2007) uses essentially the same
styles, but names the phase Barnes/Parkhill/Neponset.
Crowfield Phase points occur in both the Great Lakes and New England, and are correlative
with Simpson points of the Mid-Atlantic (Spiess et al. 1998). These points are thinner and wider
than Michaud-Neponset phase artifacts, and expand from a narrow base (Ellis and Deller 1997).
They often have multiple, long flutes and a shallow, wide basal concavity (Speiss et al. 1998).
These points have been found in Vermont and eastern Massachusetts. No Crowfield Phase dates
are available for New England.
The youngest New England fluted point type is the Nicholas, named for the Nicholas site in
Maine (Spiess et al. 1998; Wilson et al. 1995). These points are similar to those from the
Holcombe phase of the Great Lakes region, and are small and thin with shallow concave bases
(Spiess et al. 1998). A Nicholas Phase site in Maine has yielded dates of 10,060 B.P., and is
thought to provide “a reasonable end-date for fluted point Paleoindian occupation in the region”
(Spiess et al. 1998:238). This conclusion is further supported by the discovery of another
Nicholas phase site in interior Maine, which yielded an Accelerator Mass Spectrometry (AMS)
22
radiocarbon age 10,110 B.P (Will et al. 2001). Dincauze (2007) groups the Crowfield and
Nicholas/Holcombe phases with lanceolate points as Terminal Pleistocene and Early Holocene.
Anderson (2001:155) suggests that the replacement of lanceolate points with notched projectile
points in the Late Pleistocene/Early Holocene coincides with a shift in hunting emphasis from
large animals to “smaller, more dispersed game animals.”
Paleoindians were once thought to be highly specialized hunters of big game, such as
mammoths, mastadons, and caribou. Archaeological data from Paleoindian sites throughout New
England (Meltzer and Smith 1986; Spiess et al. 1998) and the ecologically similar Great Lakes
(Stothers 1996) region are consistent with the hypothesis that Paleoindians subsisted on
migratory game, chiefly caribou. Paleoindian sites in New England have yielded caribou, beaver,
and bison bones, along with some charred floral remains including nuts and berries (Spiess et al.
1998). Specialized subsistence models focusing on large game derive primarily from Paleoindian
sites located in the midwestern and southwestern United States, such as the Folsom site (Figgins
1927), which clearly exhibit evidence for the exploitation of large (now extinct) animal species
by humans. However, recent evidence suggests that an emphasis on large game may be
overplayed, and that Paleoindians across North America were more likely opportunistic hunters
and gathers whose diet was largely influenced by environmental variability (Cannon and Meltzer
2004; Meltzer 1993). Similar arguments have been made by researchers in the Northeast
(Dincauze 1993; Ogden 1977). Dincauze (1990) has asserted that the Paleoindian inhabitants of
southern New England made use of the full range of readily available plant and animal species
that existed at the time. Jones and Forrest (2003) have also argued for the more generalized
subsistence model among Paleoindian peoples, citing the apparently higher occurrence in the
archaeological record of small Paleoindian encampments as compared to that of larger base
camps in the region. They assert that these smaller sites reflect a settlement system wherein small
groups of mobile foragers adapted to resource unpredictability and pursued a more generalized
subsistence regime, exploiting a variety of available floral and faunal resources present in the
resource-rich areas surrounding freshwater glacial ponds and wetlands widely distributed across
the recently deglaciated New England landscape. Archaeological evidence recovered during the
excavation of Shawnee Minisink archaeological site, situated along the upper Delaware River,
seems to support Dincauze, Jones and Forrest’s arguments for Paleoindian exploitation of a
broad subsistence base, as it includes evidence for the processing of fish, nuts, and edible plants
(e.g., Goose foot [Chenopodium sp.], ground cherry, black berry, hawthorn plum, pokeweed,
pigweed [Amaranthus sp.], smart weed [Polygonum sp.], wild lettuce, grape, hackberry, and
meadow grass) (Hanley et al. 2002).
While no evidence of sea mammal hunting by Paleoindian-period peoples has been found in
the archaeological record to date, the possible exploitation of sea mammals by Paleoindians in
the Northeast should not be ignored, as they are an important resource for many arctic and subarctic peoples today (Jones 1998). Finds of the archaeological remains of marine mammals on
the former shores of the Champlain Sea in Vermont, and the serendipitous recoveries of mammal
finds from the continental shelf by fishermen, indicate that marine mammals such as walrus,
ringed seal, harp seal, bearded seal, hooded seal, harbor seal, and gray seal all could have been
present and exploited by coastally-adapted Paleoindian settlers (Jones 1998).
23
Generally, settlement strategies during the Paleoindian are poorly understood as Paleoindian
materials and sites are, overall, quite rare in the documented archaeological record. However, it
is clear that Paleoindian site size and duration of occupation are highly variable (Jones 1998).
Identified sites include large, possibly seasonally occupied, base camps such as the Vail site in
Maine and the Bull Brook site in Massachusetts; small residential camps such as Reagan in
Vermont, Whipple and Israel River complex sites in New Hampshire and the Templeton site in
Connecticut, as well as very small task-specific loci, such as the Hidden Creek site in
Connecticut (Basa 1982; Boisvert 1998; Bouras and Bock 1997; Byers 1954; Gramly 1982;
Jones 1997; McWeeny 2002; Moeller 1984; Robinson et al. 2009; Speiss et al. 1998). In some
areas, Paleoindian sites have been found in settings removed from present day water bodies but
on landforms strategically positioned above low-lying terrain that may have been suitable habitat
for caribou and other game animals; campsites in such settings are typically indicative of shortterm habitations by small groups of people, perhaps in some cases by even a single, extended
family (Spiess et al. 1998). Elsewhere, a strong correlation has been found between Paleoindian
sites and glacial features that include well-drained sand and gravel kame deltas and outwash
terraces, suggesting a preference for high, well-drained ground, near streams or wetlands, which
also offered vantage points for observing game. In the coastal region of Maine, for example, the
pattern of small Paleoindian sites on sandy high ground near water sources is embodied in the
Hedden (Spiess and Mosher 1994; Spiess et al. 1995; Spiess et al. 1998), Spiller Farm (Hamilton
and Pollack 1996), and Neil Garrison (Douglas Kellogg, personal communication 1999) sites.
Two small, eroding sites on the coast at Boothbay contained Paleoindian artifacts eroding from
beneath Ceramic period shell middens (Spiess et al. 1998). While now at the present-day coast,
at the time of Paleoindian occupation, the site was located inland, adjacent to a stream. Work by
Kelley (2006) suggests that in inland Maine, a region dominated by lakes and wetlands during
the late Pleistocene and early Holocene, hunting and travel sites would be associated with
lake/wetland shores and thoroughfare locations between lakes. The limited size and artifact
suites of these sites has been used to identify each as a travel or hunting campsite, used by a few
people for a short amount of time. The linkage between geomorphic features and archaeological
sites has been used to suggest hunting strategies (Spiess et al. 1998), but may also suggest the
importance of locating water during drier periods. McWeeny and Kellogg (2001) suggest that
people of this time relied more heavily on upland regions where thinner glacial sediments and
bedrock control of groundwater made water availability at springs and streams more predictable.
The large Paleoindian sites known from New England and Nova Scotia are each
characterized by eight or more artifact loci, reflecting population aggregation and/or repeated site
visits (Robinson et al. 2009). For example, the Bull Brook site, located in Massachusetts, is
positioned on a large, flat-topped sandy kame or delta, and is composed of a circular pattern of
36 loci (Byers 1959; Robinson et al. 2009). Robinson et al. (2009) interpret the site as an
aggregation site for communal hunting to exploit the seasonal movement of caribou from
subaerially exposed portions of Jeffreys Ledge to the inland ca. 10,300 B.P. Likewise, Vail, in
northwest Maine and now submerged by Aziscohos Lake, is interpreted as a riverside kill site
(Gramly 1982). Large sites like these have produced the vast majority of fluted points recorded
in New England in the Paleoindian Database of the Americas (PIDBA), an online, county-bycounty (or province) database of Paleoindian period projectile points finds in North and South
America (PIDBA 2009).
24
It is probable that many Paleoindian sites were situated on the now inundated continental
shelf (Marshall 1982). The existence of submerged Paleoindian sites has been revealed by the
finds of scallop draggers. For example, the dredges on a scallop dragger off the western tip of
Black’s Island in Maine’s Blue Hill Bay recovered three Late Paleoindian lanceolate bifaces
from approximately 44 m water depth (Crock et al. 1993). Examination of bathymetric charts
from the region suggests that the artifacts may come from a site or sites located near the edge of
the now submerged Union River channel (Crock et al. 1993). These artifacts represent the oldest
recovered material from Maine waters. While indicative of occupation during the Late
Pleistocene/Early Holocene, it is unlikely that these artifacts were recovered from an intact site.
More reasonably, they represent material eroded from what was then a terrestrial site located on
the Union River banks, and redeposited into the river channel nearby. Kelley et al. (2010)
describe a submerged site in inshore waters near Bass Harbor, Maine, that is associated with
Middle Archaic period artifacts (Price and Spiess 2007). Multibeam bathymetry, side scan sonar,
seismic reflection profiling, and coring were used to identify a sheltered lake/wetland complex. It
is possible that Paleoindian inhabitants used similar settings where available.
No archaeological material has been recovered from areas further offshore, however,
draggers working to the south, offshore of Massachusetts, have recovered mammoth teeth
(Whitmore et al. 1967). The presence of these probiscidean remains suggests that these broad,
offshore areas were dominated by grassland/steppe environments, suitable to support large
grazing animals and herd animals, such as mammoth and reindeer.
Potential Paleoindian occupation of these offshore areas would most likely be situated so as
to best exploit available herd animals and coastal resources. Robinson et al.’s (2009)
identification of the Bull Brook site as an aggregation site linked to hunting of migrating caribou
requires the hunters to have knowledge of migration patterns and a potential familiarity with the
landscapes available to prey species. Shaw et al. (2006) suggest a large subaerially exposed area
with associated, adjacent islands was present to the east of Cape Cod ca. 13,000–9,000 B.P., and
may have provided hunting areas contiguous with the current mainland. Paleoindian hunters may
have followed herd movements into this area, and even across limited expanses of open water.
Use of boats by Paleoindian people is unsubstantiated, but has been suggested as a method of
colonization of the coastal portions of western North America (Dixon 1999; Erlandson 2002;
Erlandson et al. 2007; Fedje and Christensen 1999; Fladmark 1979). Paleoindian exploitation of
coastal resources has not been recognized in northern New England and the Canadian Maritimes,
primarily because any coastally focused occupation or resource-related areas of this time period
are currently submerged. In other portions of the world, where coastal zones of Terminal
Pleistocene age are preserved, Paleoindian sites are associated with marine mammal, fish, and
shellfish remains (see Sandweiss et al. [1998] for an example from coastal Peru, the Quebrada
Jaguay site).
Using available settlement models, Paleoindian period peoples are likely to have used
subaerially exposed areas of the OCS, dependent upon access across areas contiguous with the
coast or having the ability to navigate across open water. Sites may exist as: 1) small upland
hunting and travel sites associated with surface water, such as springs, 2) multi-loci aggradation
sites developed for group exploitation of resources, and 3) coastal sites positioned to access
marine resources, including marine mammals, fish, and shellfish. Site preservation in these areas
25
will be influenced largely by depth of burial and geological processes acting on the sites
following occupation.
2.3.2. Early Archaic Period
The start of the Early Archaic period (ca. 10,000–7500 B.P.) coincides with the end of the
Pleistocene and the Wisconsin glaciation and the commencement of the Holocene epoch 10,000
years ago. The early Holocene is marked by a climatic shift towards conditions that were
generally warmer and drier than those of the preceding Pleistocene epoch. This shift produced
concomitant changes in southern New England’s environmental setting to which the region’s
Native Americans inhabitants at the time adapted. By the Early Archaic, the environment had
transformed from open woodlands to more closed forests initially dominated by spruce, balsam
fir, birch, and poplar, but eventually dominated by pine (Almquist-Jacobson et al. 2001).
The Archaic period represented a time of increased familiarization and settlement within the
Eastern Woodlands. Archaeological evidence recovered to date suggests native peoples of the
Early Archaic followed a more diversified subsistence strategy relative to that of the preceding
Paleoindian period. This more diversified subsistence strategy appears to have included the
pursuit of available smaller game and fish as well as the gathering of available woodland and
wetland vegetation, and nuts (Dumont 1981; Forrest 1999, Jones 1998; Kuehn 1998; Meltzer and
Smith 1986; Nicholas 1987; Robinson et al. 1992). The archaeological record exhibits a strong
correlation between Early Archaic habitation sites and wetland locations as many of the
identified sites located around the perimeters of ponds, marshes, and wooded wetlands and at the
headwaters of major rivers. Consequently, from this distribution of sites, it may be inferred that
wetland environments became increasingly important loci for human activity during the Early
Archaic (Jones and Forrest 2003; Nicholas 1987).
Identification of Early Archaic habitation sites throughout southern New England has
typically hinged upon the recovery of corner-notched, stemmed, and bifurcate-based projectile
points during archaeological surveys. Recent documentation of Early Archaic sites in New
England has led researchers to question whether Early Archaic archaeological deposits are in fact
rare as initially believed (i.e., Sanger 1977), or have simply been overlooked because they may
be difficult to discern from later archaeological components given the widespread presence of
quartz throughout much of New England during the pre-contact past. Excavations at very deeply
buried sites along major rivers in Maine have yielded important evidence of Early Archaic
occupations. Investigations along the Saco, Kennebec, Androscoggin and Penobscot rivers have
prompted the identification of what has been called, “The Gulf of Maine Archaic Technological
tradition” (e.g., Robinson 1992; Sanger 1996; Will et al. 1996).
Non-bifacial tools that include unifacially edged tools, cores, and flakes have been proposed
as alternative diagnostic markers for the period (Robinson et al. 1992), as have “nibbled flakes”
or “denticulates” and tabular blades (Thomas 2001; Waller et al. 2010). This assemblage,
subsumed within the Gulf of Maine Archaic tradition, also includes hammerstones, milling slabs,
and notched pebble sinkers (Waller et al. 2010). Artifact types crafted using this new technology
of pecking and grinding reflect an increased focus on plant and fish resources during the Early
Archaic (Robinson 1992). Chipped stone tools were typically produced from local stone, often
collected in cobble form, suggesting more restricted territorial ranges than in Paleoindian times.
26
Archaeological investigations of the Sandy Hill Site in Ledyard, Connecticut (Forrest 1999),
record the early Holocene utilization of a distinctive quartz lithic technology focused on the
production of quartz “microliths” for use in composite tools (Forrest 1999). The preponderance
of expedient tools and nearly exclusive reliance on local or regional lithic materials are
characteristic of this tool assemblage, and suggest either a “restricted wandering” or a “centralbased wandering” settlement system (Waller et al. 2010). A restricted wandering settlement
system consists of seasonally based group movements by small, residential groups within welldefined territorial limits, while a central-based wandering settlement system describes settlement
at a place for an extended period of time by a modest population until the time arrives when the
entire community finds it necessary to move on, perhaps never to return (Ritchie 1980).
The identification of a semi-subterranean pit house associated with a LeCroy Bifurcate
complex at the Weilnau site in Ohio (Stothers 1996), a pit house dated to 8920 ± 100 B.P. from
Connecticut (Forrest 1999), and more recently two pit houses dated to 7830 ± 130 and 8110 ±
90 B.P. at the Whortleberry site in Dracut, Massachusetts (Dudek 2005), imply a previously
unknown degree of sedentism for Early Archaic populations (Waller et al. 2010). It is inferred
that these larger, longer-duration residential sites were associated with peripheral small, shortduration sites resulting from logistical forays in the Early Archaic settlement system. Jones and
Forrest (2003) assert that the Early Archaic semi-residential settlement pattern in southeastern
Connecticut is an adaptive response to predictable, readily abundant resources (Waller et al.
2010).
2.3.3. Middle Archaic Period
The environmental setting associated with the Middle Archaic (ca. 7500–5000 B.P.) is
characterized by increased precipitation relative to the preceding Early Archaic period. Forest
composition and vegetation changed in response to the increased rainfall as the pine-dominated
landscape was replaced by a deciduous forest of oak, sugar maple, elm, ash, and beech, with
smaller numbers of hemlock and white pine. Deer populations expanded and likely became a
major subsistence focus with the emergence of the “mast” forest. Bear, wolf, otter, and wild
turkey also emerged in greater numbers, while comparatively smaller populations of moose, elk,
and caribou populations persisted in spruce-fir northern hardwood forests.
An increase in the relative frequency of Middle Archaic sites in the Northeast suggests that
colonizing peoples were firmly established in New England by 7500 B.P., with a greater density
of identified Middle Archaic sites occurring in southern New England than in the north (Waller
et al. 2010). Southern New England’s resident Middle Archaic populations continued their
generalized subsistence regimes with most sites of the period discovered around ponds, lakes,
rivers, and wetlands (Bunker 1992; Dincauze 1976; Doucette 2005; Doucette and Cross 1997;
Maymon and Bolian 1992). Subsistence activities reflected at these sites included the focused
harvesting of anadromous fish, hunting and foraging, and fishing. Base camps established along
extensive wetland systems (Doucette 2005; Doucette and Cross 1997; Jones 1998) supplemented
smaller logistical camps and exploitation sites within the Middle Archaic settlement system. An
increase in the complexity of seasonal rounds is conjectured on the broad range of resources
available throughout the period (McBride 1984).
27
Middle Archaic occupations in southern New England are typically identified by the
presence of Neville, Neville-variant, Stark, and Merrimack style projectile points (Dincauze
1976; Dincauze and Mulholland 1977). The Neville type-site for Middle Archaic Native
American occupations in New England was situated along the Merrimack River in Manchester,
New Hampshire, and contained a substantial Neville projectile point tool assemblage. Many of
these points possessed slightly bifurcated or notched bases, thus providing evidence for a
possible technological evolution out of the preceding bifurcate-based Early Archaic point type.
The site was used repeatedly beginning roughly 7750 B.P., probably as a seasonal base camp
situated to take advantage of migratory fish runs. Besides fishing, other activities represented by
artifacts and features at the site include stone tool manufacture, hide working, and wood working
(Dincauze 1976). At Middle Archaic sites throughout New England, projectile points are found
in association with steep-bitted scrapers, flake knives, perforators, adzes, axes, gouges, and
choppers. Groundstone tools became more central to the material culture of the Middle Archaic.
The presence of adzes, gouges, and axes within the archaeological record suggests heavy
woodworking activities and the possible manufacture and use of dugout canoes, which is further
suggestive of the increased importance of river travel for Middle Archaic peoples (Waller et al.
2010).
A preference for regionally available lithic raw materials (quartzite and rhyolite) is reflected
in the Middle Archaic’s archaeological record. Utilization of Ossipee Mountain and Boston
Basin volcanic materials is also evident during the Middle Archaic, although quartz apparently
remained the raw material of choice (Bunker 1992). The correlation between regional lithic
material types and Middle Archaic materials has led Dincauze (1976) to theorize that Native
American band or tribal territories were established within major river drainages, and that the
scheduling of subsistence activities such as the seasonal pursuit of anadromous fish species may
have developed in response to territoriality (Dincauze and Mulholland 1977; Waller et al. 2010).
In Maine, chipped stone spear points are more abundant in the Middle Archaic
archaeological record and the first cemetery sites occur. Artifacts dating to this time period have
also been discovered submerged in places, such as Blue Hill Bay suggesting that sea level rise
has submerged sites from this time and earlier (Crock et al. 1993). The cemetery sites reveal
mortuary practices that included the sprinkling of graves with red ocher, and the offering of
grave goods, such as gouges, slate spear points, and stone rods (Moorehead 1922; Robinson
1992; Will and Cole-Will 1996; Willoughby 1898). Commonly referred to as the “Red Paint
People,” sites dating to this tradition have typically been found east of the Kennebec River with
some sites displaying a strong focus on maritime resources.
2.4. MID COAST
Surveys of regional prehistory associated with the modern states of New York, New Jersey,
Delaware, Maryland, and Virginia are provided by Cross (1941), Custer (1984, 1989), Dent
(1995), Kraft (1986), Mounier (2003), and Ritchie (1980). Archaeologists working in this region
have traditionally employed a system of three periods (Paleoindian, Archaic, and Woodland) to
divide the span of time between the first settlement of the region by Native Americans and the
arrival of the European explorers in the sixteenth century. The Paleoindian period spans roughly
12,500–10,000 B.P. It is followed by the Archaic, divided into four periods: Early Archaic
28
(10,000–8000 B.P.), the Middle Archaic (8000–6000 B.P.), the Late Archaic (6000–3000 B.P.),
and the Transitional or Terminal Archaic (3000–2700 B.P.). The Woodland period postdates any
possibility for submerged sites on the OCS.
At the Pleistocene glacial maximum, the Laurentide ice sheet extended as far south as
present-day New York City. After the retreat of the glacial ice sheet, tundra vegetation, similar to
that found today in Alaska and northern Canada, colonized the newly exposed landscape
(Gaudreau 1988). Between 19,000–11,000 B.P., a spruce dominated forest was present,
retreating northward and eventually replaced by a forest dominated by pine. Finally, by
9000 B.P. (during the Early Archaic period) hardwood forests, similar to those that characterize
the Eastern Woodlands today, were present throughout the region.
2.4.1. Paleoindian Period
Only a few sites dating to the Paleoindian period are known from this portion of the Atlantic
coast, while the presence of early peoples is implied from the occasional find (often on the
surface) of characteristic fluted projectile points that were presumably used to hunt Late
Pleistocene/Early Holocene fauna (Anderson and Faught 1998). The relative scarcity of early
sites along the modern coast is to be expected. Even if the region was well-populated prior to
10,000 B.P., most of the evidence for early human presence has been destroyed or hidden by
natural or cultural processes. Foremost among these forces is the post-glacial rise in sea level.
During the initial settlement of the region, sea level was roughly 100 m lower than today,
meaning that, for example, what is now lower New York Harbor would have been exposed land,
cut by stream channels of the Hudson and Raritan rivers.
The first reported Paleoindian habitation site in eastern North America was at Shoop,
Pennsylvania (Witthoft 1952). This discovery was soon followed by the Reagan site in Vermont
(Ritchie 1953) and the Bull Brook site in eastern Massachusetts (Byers 1954). Another early site,
Meadowcroft Rockshelter, probably has received more attention than any other early prehistoric
site excavated in eastern North America, due in large part to the very early radiocarbon dates
obtained for the lowest artifact-bearing strata. The site is located in southwestern Pennsylvania,
approximately 50 km south of the maximum extent of the last advance of the Wisconsin ice
sheet, and overlooks a small tributary of the Ohio River. The rockshelter deposits are deep,
stratified, and contain several cultural components, with an internally-consistent suite of
radiocarbon dates ranging from at least between 685±80 through 13,240±1010 B.P. (Adovasio et
al. 1990). No fluted projectile points were found in the lowest strata at the Meadowcroft
Rockshelter, leading the principal investigator to suggest that the deposits were created by people
earlier than, or at least outside of, the fluted point Paleoindian tradition (Adovasio 1993).
Another well-studied stratified Paleoindian site is the Shawnee Minisink site in the Upper
Delaware River Valley of eastern Pennsylvania. Charcoal from hearths in the Paleoindian
component of the Shawnee Minisink site has been radiocarbon dated to 10,590±300 and
10,750±600 B.P. (McNett 1985:6). More than 76 seeds from at least 10 different plant species
were recovered from Paleoindian contexts at the site (Dent and Kauffman 1985:67).
Interestingly, there are only minor differences between the Paleoindian and Early Archaic
botanical assemblages from the site. In addition to the wide variety of seeds and fruits at
29
Shawnee Minisink, fish bones were encountered in Paleoindian contexts (Dent and Kauffman
1985:73).
In general, Paleoindian settlement patterns may be described as semi-nomadic within a
defined territory. The subsistence focus was on hunting both large and small game and it is
assumed that wild plants were exploited for their food potential as well (Custer et al. 1983;
Garner 1980; Kraft 1973). Populations of Pleistocene megafauna, such as mammoth and
mastodon, were likely dwindling by this time, but pre-Clovis populations, if they existed, may
have utilized a much richer late Pleistocene ecosystem marked by extensive estuarine systems,
high order stream terraces, broad open grasslands, and wetland habitats. Sites such as Cactus Hill
(44SX202), on the Nottoway River in Virginia, show evidence of coastal plain riverine settings
being heavily utilized in Clovis and what appears to be pre-Clovis times (McAvoy and McAvoy
1997). Cactus Hill may have been one of a series of smaller sites located upstream from the
Atlantic shelf for the purpose of raw material replenishment as well as hunting and gathering.
Similar riverine settings further inland were exploited as well. The Higgins site (18AN489), for
instance, is a Clovis site on the western shore of the Chesapeake near Baltimore-Washington
International Airport in a small stream setting in a headwaters area (Curry and Ebright 1990;
Ebright 1989, 1992).
Many regional Paleoindian sites are located adjacent to what would have been fresh water
sources at the time of occupation. During the Late Pleistocene, lowered sea levels and associated
lowered ground water tables resulted in fewer fresh water resources compared to Holocene
conditions, and the few resources that were present undoubtedly attracted human foragers. Fresh
water locales would have been visited repeatedly by Paleoindians, and thus these sites are more
visible in the archaeological record than environmental niches used only sporadically. Mounier
(2003:126), for example, notes that almost all habitation sites from all prehistoric periods in New
Jersey are located near fresh water. Two Paleoindian sites, two sites with redeposited
Paleoindian period artifacts, and 12 isolated finds (mostly fluted projectile points) have been
documented on the outer coastal plain of central and southern New Jersey (Grossman-Bailey
2001:171–184). An additional 12 fluted points, most made from jasper, with others from
quartzite and argillite, have been recovered on the coastal plain between Sandy Hook and
Barnegat Bay in central New Jersey (Marshall 1982). Numerous Paleoindian isolated finds have
been identified in the Chesapeake Bay area, but relatively few intact sites are documented.
Brown’s (1979) survey of fluted points in Caroline County, Maryland shows five points—four
Clovis and one mid-Paleo. The dearth of sites is attributed, in part, to the notion that many
regional Paleoindian site locations are submerged. An upland site, Paw Paw Cove in Maryland
(Dent 1995; Lowery 1989), is located close to ancestral Susquehanna River terraces overlooking
where the Choptank and Miles rivers flow into the Susquehanna. Locations such as this may
simply have been utilized for its proximity to rich floodplain areas resources. Some evidence
also suggests that site location choices were designed to provide protection from the elements.
One common pattern consists of southern exposure sites adjacent to topographic features that
provide shelter from prevailing winds (Dent 1995:124).
Paleoindian technology is distinguished by the distinctive fluted projectile points and
specialized tool kit that included scrapers, burins, gravers, denticulates, spokeshaves, perforators,
knives, pièces ésquillées, and unifacial flake tools. Tools include highly specialized formal
30
implements, multi-purpose tools, and expedient tools. A variety of high quality cryptocrystalline
raw materials were utilized such as chert, jasper, and quartz. Kraft (1986) notes that fluted points
diminished in size after Clovis, ultimately being replaced by notched Early Archaic points such
as Palmer and Kirk.
The geography of Paleoindian settlement patterns has traditionally been interpreted based on
a presumed reliance on high-quality lithic raw material, and thus an attraction to source areas
(Gardner 1989). The use of cryptocrystalline raw material facilitates hunting and gathering
expeditions outside the usual lithic resource procurement area by allowing portable, flexible
technologies based on bifaces and blade cores with long life spans that can be reliably used,
resharpened, and recycled as Paleoindian groups moved across the landscape (Goodyear 1979).
Artifacts associated with this period in the region include high percentages of cryptocrystalline
material, although the picture of Paleoindian resource use is becoming more nuanced. Many
coastal plain sites (e.g., Paw Paw Cove) show intensive use of high quality cobble resources such
as jasper instead of more distant outcrop sources (Custer and Galasso 1980; Custer and Lowery
1994; Lowery 1989). Where high quality cryptocrystalline raw material was not available locally
as outcrops or cobble sources, one might expect a high percentage of curated tools (manifested
by resharpening) and blanks transported from source areas elsewhere. That indeed is what is seen
at the Shoop site, north of Harrisburg, Pennsylvania, where many fluted points, scrapers, and
other typical Paleoindian tool forms were made from Onondaga chert, a raw material found in
western New York north of the Finger Lakes region (Carr and Adovasio 2002; Witthoft 1952,
1954). Some Paleoindian groups in the region did, however, make use of lower quality local
lithic raw materials as well. For example, in the Chesapeake drainage and in other portions of the
coastal plain, Paleoindian assemblages include a high percentage of local, non-cryptocrystalline
material such as quartz and quartzite (Dent 1995:127). Cryptocrystalline raw materials of high
quality were more likely to be used for tools that would be curated and reused, while locally
available resources would have served well for expedient tools (Goodyear et al. 1989).
Lithic resources, like fresh water sources, may have served as focal points of human
occupation on the landscape. Stratified sites containing Paleoindian artifacts in an area rich with
good lithic raw material include the Thunderbird and Fifty sites of the Flint Run Complex in the
Shenandoah Valley of Virginia (Gardner 1977). The “Flint Run Lithic Deterministic” model of
Paleoindian settlement, where the movements of small groups of Native Americans across the
landscape were made to take advantage of this important lithic source (Anderson and Sassaman
1996), was based on finds at the complex, which included quarries, reduction sites, base camps,
and maintenance camps. On the Delmarva Peninsula, there appear to be two mechanisms
responsible for the distribution of fluted projectile points. One concentration in the north
centered on the Delaware Chalcedony Complex may reflect the Flint Run model, where
Paleoindian artifacts are associated with outcrops of high quality lithic raw material. Two other
concentrations of fluted points are along the mid-peninsular drainage divide, where the Late
Pleistocene-Early Holocene environment was riddled with swamps and wetlands attractive to
game (Custer et al. 1983), and at the mouths of the Choptank and Nanticoke rivers (Custer
1989:94, 103). Custer (1989) believes that Paleoindian sites clustered along the upper reaches of
the mid-peninsula drainage divide are base camps; however, Lowery and Phillips (1994:33)
believe that they are temporary camps.
31
Known site types for the Paleoindian period in the region include riverine base camps located
near high-quality lithic sources and smaller transient hunting camps near game-attractive areas.
For example, Werner (1964:31) describes a jasper Clovis point, hearths and associated debitage
from an alluvial terrace context at the Zierdt site in the Upper Delaware valley. Kraft (1973)
describes similar riverine contexts at Plenge in the Muscontecong River valley of western New
Jersey. Larger sites, such as Shawnee-Minisink (McNett 1985) may have been occupied for a
longer period of time or as some suggest, may represent a series of brief occupations over the
long term to exploit nearby chert outcrops (Gingerich 2007). Most of the recorded Paleoindian
sites along the Middle Atlantic coastal plain, however, are either short-term camps or isolated
finds. Isolated finds include several locations along the Delaware River described by Kinsey
(1972:328; see also Marshall 1982). Many of these projectile points are surface finds on kame
terraces on both the Pennsylvania and New Jersey sides of the Delaware River. It is likely that
Paleoindian sites in downstream areas of the Delaware, Susquehanna and Potomac, for example,
would have included larger camps, now inundated, occupied for exploiting rich estuarine
resources and the smaller, backwater swamps. From these camps, forays into upstream and
interior headwater areas (e.g., Turkey Swamp) would have been staged. Such activity would
account for the broad distribution of isolated fluted points and small sites that typify coastal plain
site settings today.
The fundamental problem with investigating Paleoindian coastal adaptations is that the
evidence is presumably underwater. However, a few sites on the West Coast (where the
continental shelf is relatively narrow, and sea level rise had much less effect than in the East)
have yielded ample evidence of early maritime adaptations. For example, a deep shell midden of
marine shellfish remains, fish bones, and lithic artifacts at the Daisy Cave site on San Miguel
Island, California, has been dated to 9,700 B.P. (Erlandson 1993, 1994). Nevertheless, it is
unclear what role aquatic resources, particularly fish and marine mammals such as seal, may
have in Paleoindian subsistence in the region. Fish bones were recovered from the Shawnee
Minisink site in the Delaware River Valley (McNett 1985), and preservation factors may explain
their absence elsewhere. The degree of hunter-gatherer dependence on maritime resources during
the Late Pleistocene and Early Holocene has been a matter of debate (Yesner 1980). Some
researchers (e.g., Perlman 1980) have postulated that because coastal environments are among
the most productive land forms (in terms of food and raw material diversity and abundance),
their occupation should coincide with their earliest development and stabilization. In contrast,
others (e.g., Bailey and Parkington 1988; Erlandson and Fitzpatrick 2006) see a trend of
expanding subsistence patterns to include specialized niches such as the coast over the course of
the Holocene. The use of marine resources as an alternative subsistence strategy in times of
seasonal nutritional stress is documented in the ethnographic literature and may serve as a model
for possible Paleoindian subsistence practices. For example, historically-known Northeastern
hunter-gatherer groups such as the Beothuk in Newfoundland relied heavily upon a variety of
aquatic resources (especially seal, salmon, cod, smelt, herring, sturgeon, and shellfish) at least
part of the year (Reynolds 1978). Furthermore, the ethnographic record of North American
hunter-gatherers suggests that coastal groups relied heavily on fish and other aquatic resources,
regardless of latitude or effective temperature (summarized in Kelly 1995:Table 3-1). Work on
submerged early prehistoric sites in eastern North America could potentially yield data to
address this problem.
32
In the southern portion of region, bathymetric research by Blanton (1996) indicates that the
Pleistocene lands now submerged in the Chesapeake Bay along the East Coast are also likely to
contain Paleoindian sites. Tidal forces on such submerged sites may explain why, within the
Lower Delmarva region, the coastline along Tangier Sound is one of the two main areas from
which Paleoindian points have been reported, the other being the interior drainage of the middle
Pocomoke River (Davidson 1981:11).
2.4.2. Archaic Period
The Archaic period is characterized by the gradual development of more-or-less modern
environmental conditions. Humans adapted to the abundant resources provided by interior
woodlands, ponds, and rivers, as well as coastal estuaries by exploiting a broad range of food
(nuts, large and small game, seed-bearing plants, fish, etc.) and industrial products (stone for
making tools and weapons, plants for baskets and textiles, bark for house construction, etc.). By
6000 B.P. the region was heavily settled, with populations on the coast and offshore islands
likely numbering in the thousands. Archaeological evidence of this apparent population
“explosion” is reflected in the large number of terrestrial archaeological sites dating to the Late
Archaic period, and by the large size of some individual settlements (Mounier 2003; Ritchie
1980). However, the Late Archaic period is roughly coincident with slowing sea level rise rates
and the establishment of the modern coastline. Late Archaic period lifeways in the Mid-Atlantic
region have a significant coastal component, characterized by the presence of shell middens,
especially towards the latter part of the period when sea levels were closest to current positions
(Braun 1974). The model of lower population during the Early and Middle Archaic periods,
followed by population growth during the Late Archaic, is based upon the terrestrial
archaeological record, and may not adequately consider the fact that numerous earlier sites
presumably are now submerged on the formerly subaerial portions of the continental shelf.
To illustrate this point, Lowery (2009b) plotted Late Woodland-age sites within the
Choptank River watershed. Included in this analysis were numerous coastal Late Woodland sites
that included large shellfish refuse middens. To understand the impact that marine transgression
has on the interpretation of the archaeological record, he induced a hypothetical 20-m sea level
rise event. As a result, all of the coastal Late Woodland midden sites in the Choptank watershed
would be drowned and the surviving interior upland sites would not provide any clues that
prehistoric human societies were interested in coastal or estuarine resources. Since the early
prehistoric archaeological record has been impacted greatly by marine transgression, previous
interpretations about this record are probably inaccurate. Some syntheses have suggested that
early prehistoric societies in the Middle Atlantic may have been only marginally interested in
coastal resources with their subsistence patterns primarily focused around interior upland
resources (Custer 1988). Thus, the early prehistoric archaeological record is biased by the fact
that these upland settings are the only landscapes that have survived marine transgression.
Regardless of the likelihood that people were living on the now-submerged coast, major
shifts in social organization and mobility strategies are not suggested by the archaeological
record at several regional sites, including Meadowcroft and Shawnee Minisink, which contain
substantial Early Archaic components underlain by Paleoindian material. At Shawnee Minisink
in particular, the archaeological record is indicative of continuity in human adaptations, with
33
gradual intensification of local resource use and broadening of diet breadth over time (McNett
1985). No decline in Early Holocene population size is indicated by a recent inventory of
prehistoric sites on the outer coastal plain of New Jersey (Grossman-Bailey 2001), where 16
Paleoindian, 19 Early Archaic, 43 Middle Archaic, and 199 Late Archaic components were
identified. Archaic toolkits expand in diversity, including the introduction of more plantprocessing tools (e.g., mortars and pestles) on the Delmarva Peninsula, but the types of locations
chosen for occupation were essentially the same between the Paleoindian and Archaic periods
(Custer 1986). Similarly, in the Chesapeake region, there does not appear to be a marked division
between the Paleoindian and Early Archaic periods. Instead, settlement and subsistence patterns
may reflect settling into an expanding mixed hardwood forest made possible by emerging
modern environmental conditions. The Early Archaic in the southern part of the region is
characterized by an increase in the number and diversity of archaeological sites (Anderson and
Sassaman 1996; Custer 1990).
Just as the fluted projectile point is regarded as representative of Paleoindian activity, a
variety of side-notched, corner-notched, and points with bifurcated bases represent the Archaic
period in the region. Stratigraphic data used as a basis for a local sequence of projectile point
styles has been derived from a variety of stratified and single component sites in the Middle
Atlantic and surrounding region (e.g., Broyles 1971; Coe 1964; Gardner 1974; Kinsey 1972;
McNett 1985; Michels and Smith 1967). Diagnostic artifacts representing Early Archaic
occupations in the region include primarily Palmer corner-notched, Kirk corner-notched and
stemmed, MacCorkle, Kanawha, Thebes, Charleston, and a variety of lesser known types. Some
researchers (e.g., Carr 1998; Custer 1996; Gardner 1987, 1989) consider the early side- and
corner-notched projectile point types, such as Palmer, Amos, and Kirk, as diagnostic of late
Paleoindian period occupations, suggesting continuity in both technology, including very
selective raw material choice, and settlement patterns.
Several sites on Staten Island have yielded Early Archaic bifurcated points, including the
large multi-component site at Ward’s Point, which yielded 21 bifurcated base points, 16 other
projectile points, and other stone tools. Charcoal from a hearth feature was radiocarbon dated to
8300±140 B.P. (Ritchie and Funk 1971). The West Creek site, on the mainland behind Little Egg
Harbor in southern Ocean County, New Jersey, had three loci of prehistoric activity likely dating
to the Early Archaic period. Jasper and chert tools included Kirk and Palmer projectile points and
scrapers. Features of calcined bone at the West Creek site, radiocarbon dated to approximately
9850 B.P., probably represent human cremation burials (Mounier 2003:198). Elsewhere in the
region, a suite of radiocarbon dates clustered around 7950 B.P. from a hearth feature at the
Turkey Swamp site on the outer coastal plain in northeastern New Jersey (near Freehold,
Monmouth County) places it within the Early Archaic period, despite the presence of several
basally-thinned triangular projectile points that are “reminiscent” of Paleoindian forms (Cavallo
1981). Early Archaic bifurcate base projectile points are found thinly scattered across the region,
perhaps representative of hunting losses or small camp sites.
Early Archaic settlements tend to be located on well drained surfaces adjacent to rivers,
ponds, and wetland terrain. For example, the Chance site (18SO5) is an Early Archaic site on the
lower Delmarva Peninsula that has produced hundreds of serrated notched and bifurcate
projectile points, all from the surface in a large swamp setting (Cresthull 1971, 1972; Dent
34
1995:171). When the site was occupied, this location may have been the headwaters of a series
of drainages overlooking the ancestral Susquehanna River (Custer 1989:107).
Hughes’ (1980:117) comprehensive study of artifact collections from Maryland’s lower
eastern shore indicates that Early Archaic sites in the region are commonly situated on welldrained ridges adjacent to freshwater streams and wetlands. Lowery (1995:23) found that in the
low coastal plain resource zone, Early Archaic sites are also found adjacent to springheads. One
Early Archaic site (18DO382) is located on a hilltop on Opossum Island, just east of Barren
Island. Lowery (2001, 2003a) has continued to focus on these drowned shoreline areas further
south in Accomack and Northampton counties, Virginia.
The Middle Archaic is the least well-represented period in the region. During this period
(roughly 8000–6000 B.P.), continued climatic warming and increased precipitation led to a nearmodern landscape. Technologically, the transition from the Early Archaic to the Middle Archaic
is characterized by the appearance of bifurcate based and stemmed rather than notched projectile
points (Custer 1989). Stanly (ca. 8000–7500 B.P.), Morrow Mountain I and II (ca. 7500–5500
B.P.), Guilford (ca. 5500–5000 B.P.), and Halifax/Vernon(ca. 5000–4000 B.P.) projectile points
mark the Middle Archaic period in the general region, following the classic Archaic sequence
first identified by Coe (1964). Most Middle Archaic sites are known through projectile point
finds on Holocene terraces and upland surfaces as well as along estuaries, swamp margins, and
near springheads. Interior streams fringed by wetlands were also common site locations, as can
be seen in the case of 18DO220 and 18DO139, located respectively at the mouth of Slaughter
Creek and on a bank of the Chicamacomico River on Maryland’s eastern shore (Lowery
1995:23, 2001, 2003a).
As with earlier Holocene sites, numerous Middle Archaic manifestations are probably
located in drowned valleys and estuaries on the outer coastal plain. Lowery and Martin (2009)
recorded an inundated Middle Archaic burial site in the Chester River, Maryland, which implies
there may be large numbers of sites along much of the submerged terrain near the present
shorelines of Chesapeake Bay as well as Atlantic shelf areas of the Middle Atlantic region. The
Middle Archaic sites may not be as far from shore, as evidenced by Lowery’s Chester River find,
indicating rising sea levels may have already inundated much of the broad, open Late Pleistocene
coastal plain.
Middle Archaic occupations represent significant changes in Early Holocene adaptations in
the region that involve exploitation of a wider range of environments and new additions to tool
kits such as drills and, later, groundstone items. The use of netsinkers indicates the more
intensive use of riverine environments for fishing (Kraft 1986). Subsistence economies became
increasingly diversified as new resources were being exploited seasonally (Custer 1989).
The earliest well-dated evidence for shellfish utilization in the region is the Middle Archaic
midden at Dogan Point, adjacent to the lower Hudson River (Claassen 1995). The site’s
radiocarbon date of 5650±200 B.P. makes it the one of the oldest shell middens on the Atlantic
coast of the United States. One of the site’s excavators noted that the early shell-bearing levels at
Dogan Point suggest “that the use of marine resources occurred prior to the stabilization of sea
35
level ca. 5000 years ago and that inundation of earlier coastal sites, not cultural retardation,
accounted for the lack of shell matrix sites before sea level stabilization” (Claassen 1995:3).
The trend toward an increased reliance on local lithic sources noticed during the Early
Archaic continued into the Middle Archaic. Raw materials commonly utilized during this time
include chert, argillite, jasper, quartz and rhyolite. Evidence from the Higgins site suggests that a
rhyolite trade system was becoming established with the import of rhyolite blanks from the north
(Ebright 1992).
During the Late Archaic period (ca. 6000–3000 B.P.), regional populations appear to have
grown markedly and, with the culture associated with broad blade technology in particular, to
have concentrated in larger base camps in riverine and estuarine settings. The proliferation of
archaeological sites dating to the Late Archaic may overplay population growth from earlier
periods when proportionately more sites are now submerged. By 5000 B.P. sea level appears to
have been relatively stable, with only minor fluctuations, but between 0.5–2 m below presentday level (Blanton 1996; Carbone 1976; Tanner 1993).
The main projectile point types believed to be diagnostic of the Late Archaic period in the
northern part of the region consist of stemmed types, such as Bare Island, Lackawaxen, Lamoka,
Poplar Island, and Rossville (Ritchie 1971). In the Chesapeake Bay area, diagnostic projectile
point types for the Late Archaic include a narrow blade series, with Vernon, Claggett, and
Piscataway types. Orient Fishtail and Dry Brook projectile types as well as the broad blade types
including Savannah River, Susquehanna, and Perkiomen and steatite pottery are commonly
associated with the later portion of the Late Archaic. In northern Virginia near the end of the
Late Archaic period, there appears a set of broad bladed lithic tools, called the Susquehanna
Complex, frequently made from rhyolite similar to that found in Maryland and Pennsylvania
(McLearen 1991). A combination of narrow bladed stemmed points (e.g., Bare Island,
Lackawaxen) and broadspears, together with steatite (soapstone) vessels, characterizes the Late
Archaic Clyde Farm-Barkers Landing Complex on the Delmarva Peninsula (Custer 1989).
Carved soapstone bowls are fairly common in Late Archaic assemblages, as are ground and
chipped axes, choppers, net-sinkers, and pestles. The proliferation of grinding implements and
cooking vessels may suggest increased use of plant resources and possibly changes in
subsistence strategies and cooking technologies. Net-sinkers could suggest greater commitment
to fishing in the subsistence economy. Alternatively, all of these more specialized implements
may indicate a more sedentary existence where people were willing to invest in the creation of a
more elaborate toolkit that would not have to be transported from place to place. Although
evidence is minimal, the first experiments with horticulture probably occurred at this time, with
the cultivation of plants such as squash, sunflower, and chenopodium (Cowan 1985; Ford 1981).
Evidence from the Higgins site and other Late Archaic sites in the region show that among the
exploited resources were deer, turkey, beaver, raccoon, opossum, berries, wild legumes, fish,
oyster, and clam (Ebright 1992).
Settlements appear to have shifted from swampy upper reaches of inland streams to the
mouths of major streams and rivers, perhaps in response to the establishment of more stable
estuarine environments relatively close to present-day shorelines (Davidson 1981:14). On the
other hand, comparable settings in earlier periods would now be submerged in most cases.
36
Lowery (1995:23) notes that Late Archaic settlement patterns in riverine settings (e.g., the Little
Choptank River) show a preference for points of land. Sites are typically found on points of welldrained lands surrounded by broad tidal rivers, creeks, or estuaries. This type of landform is
common in the Chesapeake Bay region (as evidenced by the numerous regional locational names
that include the word “point” (e.g., Hooper Point, Poverty Point, and Holland Point), as well as
along other major drainages such as the Delaware and Potomac. During the Early-Middle
Holocene these “points” were low terraces adjacent to streams and fringed by wetlands or broad
tidal marshes, and sites at these locations were primarily established for marine resource
exploitation, probably anadromous fish.
Late Archaic sites seem to have been occupied longer than in earlier periods; as the climate
became more temperate and sea level more stable, resources became more predictably
established across the landscape. The existence of formal residential base camps occupied
seasonally or longer is inferred for riverine and estuarine locations, together with a range of
smaller, resource exploitation sites such as hunting, fishing, or plant-collecting stations (Gardner
1987). The smaller sites are scattered broadly over the landscape on every habitable surface that
is well-drained and is associated with nearby surface water. In addition to the numerous upland
and riverine terrestrial sites dating to the Late Archaic, there are likely additional long-term
camps, perhaps in great numbers, just offshore where rising sea levels have inundated sites over
the last few thousand years
By the end of the Archaic period, sea levels had risen to such an extent that later Woodland
period sites are generally not expected on the continental shelf, although sites of all ages that are
located on the modern coastline are currently witnessing submergence as sea levels continue to
rise. Human activities of the relatively recent past, notably damming streams to form mill ponds
and reservoirs, have also resulted in the creation of underwater archaeological sites.
2.5
SOUTHEAST
2.5.1. Paleoindian Period
The Paleoindian occupation of the Southeast is known predominantly from deflated surface
sites (e.g., Anderson 1990a:173; Anderson et al. 1990:44–45; McCary 1947, 1948; Perkinson
1971, 1973), and due to a general lack of radiometric dates the timing of the initial colonization
of the region is inferred from dates in the Northeast (Levine 1990) and Southwest (Haynes
1992). The most readily identifiable and accepted diagnostic Paleoindian projectile point is the
classic Clovis, which occurs ubiquitously throughout North America. Unidirectional cores used
to manufacture prismatic blades, as well as artifacts related to blade core maintenance and the
blades themselves have been increasingly seen as potentially diagnostic of the early Paleoindian
period due to their association with Clovis points in the Southwest (Collins 1999) and the
Southeast (Broster and Norton 1993; Sain 2008). Formally hafted end scrapers also constitute an
identifiable part of the Paleoindian toolkit, and these appear related to the working of hides
(Daniel 1998).
Formal variation in projectile point morphology began to emerge in regions of the Southeast
by about 11,000 B.P., probably due to restricted movement and the formation of loosely defined
social networks and habitual use areas (Anderson 1995; Anderson et al. 1992). These later forms
37
include the Dalton, Cumberland, San Patrice, Suwannee, Simpson, Beaver Lake, and Quad
types, to name some (Anderson 1990a:67–69; Anderson et al. 1990; Coe 1964; Daniel 1997;
Justice 1987:17–43; Milanich and Fairbanks 1980). Morse et al. (1996:327) point out that the
designation of multiple subdivisions of the Paleoindian period is somewhat arbitrary at present,
and at least in the Mississippi River valley there is evidence that Dalton points may overlap
Clovis temporally. These are, however, regarded generally as a terminal or transitional
Paleoindian sub-period point type (Culpepper et al. 2000).
Paleoindian tools are often manufactured of high-quality lithic materials that are recovered
archaeologically at a great distance from their source areas, suggesting a high degree of mobility
(or possibly trade) among populations. In accordance with this inference, the Paleoindian toolkit
has been interpreted as an intensively curated technology, with cores that are ideally shaped to
maximize both portability and efficiency with regard to use and potential tool manufacture
(Goodyear 1979).
A significant wood, bone, and antler technology was used as well. Organic materials such as
these do not preserve in the acidic soils that cover much of the Southeast, and they are very
rarely found. At sites where they have been preserved, primarily at wet sites in Florida, it is clear
that organic media such as wood, bone, and antler were very important. These materials were
manufactured into projectile points, foreshafts, leisters, awls, and needles, to name just a few tool
categories (Milanich and Fairbanks 1980).
Kelly and Todd (1988) have suggested that the low level of regional variation among Clovis
points and associated toolkits indicates that populations exhibited a relatively generalized
adaptation that would have been advantageous in colonizing new and unfamiliar terrain. They
conclude that Paleoindian populations entering “unmapped” terrain would have benefited from a
highly curated lithic technology since quarry locations would have been unknown, and hunting
terrestrial fauna would have required less region-specific knowledge for processing than plant
foods (Kelly and Todd 1988). While this is a cogent argument in consideration of colonizing
populations, recent analyses have challenged the traditional view of Paleoindians as highly
mobile with a subsistence strategy based on migratory (and now-extinct) large animals, such as
mastodons (e.g., Mason 1962). This characterization may have overemphasized the role of
hunting large animals due to better preservation of bones and artifacts associated with hunting
(i.e. lithics), and an early interest in kill sites found in the southwestern United States (Kornfeld
2007). Archaeologists working in the Carolinas and Georgia have yet to document a clear
association between Paleoindian tools and the remains of displaced and extinct animal species
known to have been present as late as 11,000–10,200 B.P.—mastodon, bison, giant ground sloth,
and giant armadillo, for example (Holman 1985:569–570). More recent archaeological evidence
suggests a greater dependence on plant and large and small animal food resources in the
Southeast (Hollenbach 2005, 2009; Meltzer and Smith 1986). There is very little evidence for
resource exploitation in the littoral by Paleoindian peoples living in the Southeast. This fact is
probably due to site obfuscation and destruction caused by coastal submergence during the
Holocene, and not because the resources these ecozones contained were not used (e.g., Dunbar et
al. 1988, 1991).
38
As modeled differences between Paleoindian and Archaic subsistence strategies become less
pronounced, the distinction between these broad cultural periods relies on technological
differences. Settlement models include small temporary camps and perhaps less frequent base
camps occupied by loosely organized bands. Paleoindian groups relied on high-quality
cryptocrystalline stone for tool manufacture, and many sites are associated with source areas for
rhyolite, jasper, chert, and even quartz (see Daniel 1994; Gardner 1974; Goodyear 1979).
Known Paleoindian sites in the Southeast in general represented by Clovis and closely
related variants are few in relation to other periods, and may be under-represented in the
archaeological record (eroded from upland surface sites or eroded from or deeply buried in
floodplains or offshore). Based on the relative numbers of fluted to later unfluted point styles, if
environmental conditions were a principal factor in determining tool design, then the use of
Clovis points (and their particular suitable environment) may have been relatively brief.
Solid excavation evidence for Clovis-period sites in North Carolina, South Carolina, and
Georgia is rare. The majority of Paleoindian sites in the region consist largely of diffuse lithic
scatters at open locations, with more concentrated deposits in rockshelter or cave settings. No
conclusive evidence of permanent structures or long-term encampments has been located for this
time period in the Southeast, however, limited data have been recovered from intact contexts
(Anderson and Schuldenrein 1985; Elliott and Doyon 1981; Gresham et al. 1985; O’Steen et al.
1986). Excavations at the Topper Site, on the South Carolina side of the Savannah River,
exposed a Clovis level (Goodyear and Steffy 2003). While sufficient carbon samples to date the
Clovis level were unavailable, optically stimulated luminescence (OSL) dating at the base of the
level produced a date of 13,500 ± 1000 calibrated years before present (Forman 2003). The site
also offers tantalizing evidence of a pre-Clovis occupation. Here, in stratigraphic position below
the Clovis levels, researchers uncovered a distinct lithic assemblage characterized by spatially
clustered concentrations of multifaceted flakes and chunks of chert along with several flake tools
(Goodyear 2005). These tools, referred to as “bend break tools,” are essentially thin flakes
broken to provide a chisel-like working edge, some of which exhibit use wear patterns
suggesting use as a burin or graver. This apparent pre-Clovis occupation is also distinguished by
exploitation of a separate chert source than that of later occupations. Research at the Topper site
is on-going and is subject to intense scrutiny.
Several models of early Paleoindian settlement patterning have been advanced in the past
quarter century (see Anderson et al. [1992] for an overview). Some are concerned with
Paleoindians in general (Anderson 1990b; Kelly and Todd 1988; Martin 1973), and others with
regional trends (Anderson 1995; Gardner 1983; Morse and Morse 1983). Most are mechanistic
models that portray specific economic strategies as primary reasons for how Paleoindians settled
on and utilized the landscape. Each is slightly different in its focus, with primacy placed on one
of three major influences: (1) the need to maintain access to prominent, high-quality raw material
sources (e.g., Gardner 1983); (2) a preference for exploiting specific habitual use zones and
staging areas (e.g., Anderson 1995); or (3) a nomadic or semi nomadic existence dictated to a
large degree by the movements and availability of large game (e.g., Kelly and Todd 1988).
39
2.5.2. Archaic Period
The Archaic period began around 10,000 B.P., and almost certainly was precipitated by
Holocene climatic conditions. Warmer global temperatures generally defined warmer and wetter
conditions in the Southeast. As a result of changing environmental conditions that led to shifts in
botanical communities, Pleistocene megafauna had given way to deer and smaller mammals by
the onset of the Holocene epoch. Because these changes are temporally coincident with emergent
traditions in stone tool technology they are believed to have precipitated cultural changes among
the populations.
Overall, this period is characterized by a reliance on game animals and wild plant resources,
increasing use of local lithic materials, and subsistence settlement strategies contingent on
specific environments. In much of the southeastern United States, the Archaic period is
represented predominately by stone tools and debitage due to the acidity of southeastern soils in
open sites. Subsistence data of any kind is scanty, and is generally inferred based on evidence in
other regions, especially from cave or rock shelter sites in the midcontinent. Such evidence
points to use of a wide variety of nuts, seeds, fruits, and other plant foods, with documented plant
domestication by the Late Archaic (e.g., Chapman and Watson 1993; Fritz 1990; Hollenbach
2009; Watson 1989). As early as the Middle Archaic, there appears to have been increased use of
riverine and coastal resources, as shell middens appeared along many interior rivers and shell
rings appeared along the coast (e.g., Claassen 1991, 1992; Marquardt and Watson 1989;
Parmalee and Klippel 1974; Russo 2006). Notwithstanding the substantial shell midden base
camps in the major interior river valleys and architecturally complex shell rings of the coast,
Archaic sites tend to reflect small, short-term occupations. Group organization is presumed to
have been highly mobile (as is expected for hunter-gatherers), as what were thought to be
egalitarian groups made use of seasonally available resources in different environmental settings.
However, some evidence exists for more permanent occupations, development of trade networks,
and even inter-group or interpersonal violence (Daniel 1998; Gibson 2001; Sassaman 1991,
1993; Sassaman and Anderson 1996; Smith 1991). Even mound building had its origins in the
Middle and Late Archaic in the Lower Mississippi Valley (Gibson 1994; Russo 1994; Saunders
et al. 1994; Saunders et al. 1997; Saunders 1994).
The Archaic period is noted for populations with more regionally distinct tool kits (compared
with those of the preceding Paleoindian period), with greater diversity in projectile point forms
and site sizes. Although there are broad similarities in artifact styles throughout the Southeast,
there is also sub-regional variation in biface attributes and sequences. There are no clear
boundaries between Archaic cultural periods, although each can be characterized in ways that
differ from other Archaic subperiods. It is likely that there was a great deal of cultural as well as
technological continuity between the subperiods.
The Early Archaic period is marked by the end of the glacial climate and extinction of
numerous large animals. Regional population densities on the Atlantic Slope were concentrated
along major river systems, especially the Pee Dee, but also the Savannah, Neuse, and Roanoke
rivers (Sassaman and Anderson 1994:171–175); the greatest concentrations were generally at or
near the Fall Line, rather than the coastal plain. Again in this period, low regional population
densities with a high degree of group mobility are inferred (Claggett and Cable 1982). There are
40
several distinct characteristics that have been noted for Archaic period sites throughout the
Southeast. These include a notable increase in site size and frequency; similar lithic artifact
assemblages; and tremendous variations in site size, content, and function. Ward (1983:65) has
interpreted this diversity as evidence of an ever-increasing adaptive radiation and specialization
in a varied post-Pleistocene environment. Very few Early Archaic sites have been recorded in the
Coastal Plain, which may be a result of the inundation of coastal and riverine sites during the
onset of the Holocene period (Phelps 1983).
The Early Archaic in the Coastal Plain is subdivided into the corner-notched and bifurcate
traditions, and closely follows the Piedmont sequence defined by Coe (1964). Diagnostic
artifacts of the first phase of the Early Archaic period (ca. 10,000–9000 B.P.) include Palmer,
Kirk corner notched and later stemmed points, and hafted endscrapers (Coe 1964). Kirk phase
settlement is characterized by numerous small sites in all environmental zones and suggests an
extremely mobile population and a broad spectrum adaptive strategy (Purrington 1983:113). The
later tradition (ca. 9000–8000 B.P.) includes bifurcate forms such as LeCroy, St. Albans, and
Kanawha types (Broyles 1971; Chapman 1975; Claggett and Cable 1982; Oliver 1985). Other
Early Archaic period side-notched point forms recognized in the region include Big Sandy (Tuck
1974:75) and Taylor points (Michie 1966), the latter more numerous in South Carolina and
Georgia than in North Carolina.
There are contrasting models of Early Archaic settlement, with general agreement that small
and highly mobile populations are represented (Griffin 1952:354–355). These models differ in
regard to the nature of Early Archaic group mobility and settlement pattern, depending on
theoretical perspective. The most inclusive Early Archaic settlement model is commonly referred
to as the Band-Macroband model (Anderson and Hanson 1988). Developed from the Savannah
River valley data, the model postulates drainage-wide movements in response to seasonal
changes in food resources, the need to procure mates, information exchange, and demographic
structure. Populations practice a mixed collector/forager strategy depending on the season, and
Anderson and Hanson (1988) find evidence for these patterns in the archaeological record. The
model focuses on intra-drainage adaptations, and social groups are believed to have crossed into
other major drainage valleys only on special occasions for macroband gatherings or aggregations
(Anderson and Hanson 1988:270).
Most researchers agree that Early Archaic subsistence focused on white-tailed deer, hickory
nuts, and acorns, and utilized both floodplain and inter-riverine upland locations (Gardner
1974:24; Goodyear et al. 1979:28). Subsistence is believed to have focused on more specific
resources than in later periods (Cable 1982:687; Caldwell 1958), but this argument is based
largely on Caldwell’s primary forest efficiency model, which no longer appears to adequately
characterize Archaic period developments. Ground cobbles and manos have been found in Early
Archaic contexts (Claggett and Cable 1982:37), suggesting processing of plant foods with items
that would be more difficult to transport than a biface, but such finds in Early Archaic contexts
remain rare (Daniel 1994).
The Middle Archaic, ca. 8000–5000 B.P., can be distinguished from the Early Archaic by the
more frequent recovery of groundstone artifacts and a less diverse chipped stone tool kit.
Diagnostic bifaces that were made during this period include Stanly, Morrow Mountain, and
41
Guilford types (Coe 1964; Blanton and Sassaman 1989; Phelps 1983). It is assumed that
population density increased during the Middle Archaic, but small hunting and gathering bands
probably still formed the primary social and economic units. Larger sites tend to occur near
major drainages (Coe 1964), but occupations also appear near upland watercourses (Gunn and
Foss 1992), and numerous small, dispersed upland scatters are also characteristic of this time
period. Utilizing Morrow Mountain point frequencies as a population indicator, Sassaman and
Anderson (1994:176) found that the greatest Middle Archaic concentration of population was in
the Piedmont region, while the Coastal Plain was virtually abandoned. Other researchers have
found that Middle Archaic points, and Morrow Mountain points in particular, outnumber other
Archaic types in the northern and southern parts of the coastal region (Daniel and Davis 1996;
Davis and Daniel 1990).
The Middle Archaic period is poorly understood in the Coastal Plain. Some researchers
interpret this as evidence of a general depopulation of the area. Others argue that Middle Archaic
projectile points have yet to be identified in the region (Elliott and Sassaman 1995:26–38).
It is likely that patterns of social relationships changed even in areas not characterized by
intensive occupations, especially if Middle Archaic settlement expanded into new areas. The
period is characterized by increasing territorial circumscription, even as evidence for continued
high mobility remains. It is likely that while Middle Archaic groups were moving as frequently
as or more frequently than earlier groups, their movements probably covered shorter distances
(Custer 1990:36), a trend that likely became more pronounced in the subsequent Late Archaic
period.
As is true of other cultural periods, the Late Archaic (ca. 5000–3000 B.P.) cannot be
described by a single pan-Southeast set of traits. The Late Archaic is usually summarized in
terms of its most elaborate material manifestations, but these appear to characterize only portions
of the Southeast social landscape. Despite abundant local variation, there are some broad themes
that serve to differentiate it from preceding periods. The end of the Archaic period in eastern
North America is traditionally defined by the development of mineral-tempered ceramic pottery,
in contrast to the fiber-tempered ceramics manufactured during the Late Archaic period in the
Georgia and South Carolina Coastal Plain (Sassaman 1993).
Late Archaic sites in North Carolina are as abundant in the uplands as in floodplain locations
(Spielmann 1976:85), although upland sites may be more visible archaeologically due to erosion
and plowing. Some evidence suggests that upland sites do not possess the range of artifact
classes present in river floodplain sites, meaning that activities that occurred in upland locations
were but a subset of activities that occurred in floodplain locations (this may be true of other
periods as well). There are certainly large Late Archaic sites in river floodplains, such as the
Gaston, Doerschuk, and Lowder’s Ferry sites, and some of these have characteristics of intensive
occupations, in the form of occupational middens, high feature density, and circular pit hearths
(Coe 1964:119).
In certain major river valleys, including the Savannah, Green, and middle and western
Tennessee, there is evidence for intensive shellfish exploitation at shoal areas, accompanied by
exchange of non-utilitarian objects, such as engraved bone pins, which perhaps functioned as
trade regulators facilitating exchange among culturally circumscribed groups (Ford 1974). These
42
large-scale trade networks appear to be an elaboration of trading networks established during the
preceding period (Bender 1985; Marquardt 1985; Sassaman 1995). The Late Archaic period is
often linked to higher population densities and increased sedentism (Ford 1974; Steponaitis
1986). In such a situation, mobility became less of a viable economic or social strategy, and one
would expect to see increased use of local resources, greater use of storage, and development of
formal alliances (Sassaman et al. 1988:81).
Broad, square-stemmed Savannah River points are representative of this period (Claflin
1931; Coe 1964). House and Wogaman (1978) attribute the presence of Savannah River
stemmed points in upland locations to hunting-related activities. Some suggest that Savannah
River points were more like portable cores from which tools with a variety of functional uses
could be manufactured, including spear points (Sassaman et al. 1990:320). This accompanies the
viewpoint that Late Archaic populations, being less mobile and more circumscribed by
surrounding groups, needed to extend the use lives of stone tools (Parry and Kelly 1987). These
points appear to have shown up earlier in the southern portion of the Atlantic coast and were
progressively adopted northward (Tuck 1978:38).
Other Late Archaic varieties are known by various names, such as Appalachian Stemmed,
Elora, Kiokee Creek, Ledbetter, Limestone, Otarre, and Paris Island (Bullen and Greene 1970;
Cambron and Hulse 1983; Chapman 1981; Coe 1964; Elliott et al. 1994; Harwood 1973; Keel
1976; Sassaman 1985; Whatley 1985). Except for the Ledbetter hafted biface, which appears to
have had a specialized function—it exhibits a heavily reworked, asymmetrical blade—these
latter type names are more a product of parochial terminology than actual morphological
differences; they all are characterized by triangular blades, straight or slightly contracting stems,
and straight bases.
Steatite vessels, occurring in the form of bowls or crude, shallow pans and a number of other
artifact types are also unique to this period, and began to be widely used sometime between 4000
and 3500 B.P. In the central Savannah River valley, use of steatite slabs and ceramic pottery
preceded steatite vessel use (Stanyard 2003:54). Steatite vessels were apparently used for slowly
cooking plant or animal foods over a direct heat source (McLearen 1991:108).
The most intensively occupied Late Archaic site yet discovered in Georgia is on Stallings
Island, located in the Savannah River in Columbia County (Bullen and Greene 1970; Claflin
1931; Crusoe and DePratter 1976; Fairbanks 1942; Jones 1873). One type of bone tool found at
Stallings Island is the bone “pin,” an artifact found at certain contemporary sites in the Southeast,
and representing formalized exchange networks for high-status items. These objects are
intricately decorated and highly prized by artifact collectors. Unfortunately, they were “mined”
at the site until recent measures were taken to prevent unauthorized access to the site. The
mining has devastated the site; large “potholes” and mining trenches have destroyed much of its
integrity.
The Late Archaic lithic tool kit was diverse, and included scrapers, drills, atlatl weights,
netsinkers, and grooved groundstone axes. Feature types associated with Late Archaic
occupations in North Carolina and Virginia include rock hearths (or heated rock dumps) and
small pits (Coe 1964; Idol 2009; McLearen 1991). Stallings Island ceramics were manufactured
as early as ca. 4500 B.P. in South Carolina (Anderson et al. 1982).
43
Reliance on local lithic sources continued, although small frequencies of exotic material,
such as chert, demonstrate the extensive economic and social ties of this period. Presumably,
steatite for bowl manufacture (or the finished bowls themselves) had to be obtained through trade
or direct procurement from the North or South Carolina Piedmont or mountains.
Coastal groups during the Late Archaic are thought to have been fairly sedentary (DePratter
1979; Trinkley 1980). They maintained permanent residences in the littoral zone and made
forays into estuarine and interior settings for specific needs. The permanent settlements are
recognized as shell rings, while amorphous shell mounds are thought to represent base camps.
Interior sites do not have a defining characteristic (Marrinan 1975; Simpkins 1975; Trinkley
1980; Waring and Larson 1968). Interior sites on the Coastal Plain near the project region that
are attributable to coastal groups likely served a short-term specialized function. These
occupations were generally small and ephemeral; the cultural deposits reflect the specific nature
of the occupation, such as a hunting camp.
2.6. FLORIDA
2.6.1. Paleoindian Period
Regardless of the precise timing of the first occupations of North and South America, the
current evidence suggests that Florida was not intensively inhabited by humans prior to about
12,000 B.P. Claims for an earlier occupation (e.g., Purdy 1981, 2008) are controversial. The best
evidence comes from the Sloth Hole and Page-Ladson sites in Jefferson County, Florida, where
radiocarbon dates predating 12,000 C14 years B.P. have been obtained from levels containing
lithic waste flakes, but no diagnostic tool forms (Dunbar 2002, 2006b; Hemmings 1999, 2004).
Both sites are inundated river sites, and although the contexts are thought to be intact, there is a
possibility of the downward movement of artifacts from the overlying artifact-bearing levels.
While archaeologists continue to grapple over routes to the New World, entry points, and the
timing of such events, debate also looms over the manner with which the continent was
colonized. The use of Geographic Information System (GIS) data has been employed to examine
interior routes that would have been preferred based on the ease with which people could have
passed (Anderson and Gillam 2000), while others have looked at the distribution of early sites
and lithic assemblage variability to advance hypotheses on the peopling of the New World
(Faught 2008).
Paleoindian activity is most readily recognized by the presence of the uniquely-shaped
lanceolate projectile points that were crafted during the period. Significant work has gone into
tracking the location of where these stone tools were recovered, and the PIDBA is an outstanding
source for garnering county-by-county data on these specimens (PIDBA 2009). The locational
database on this website reveals only nine Paleoindian projectile points recovered from the 13
Florida counties that abut the Atlantic Coast, including four in Brevard, three in Volusia, and one
each in Duval and St. John’s counties. It is inferred that the Atlantic Coast of Florida did not
support significant Paleoindian activity, and this is in part due to the dearth of raw material for
stone tool production in this part of the state. Counties that have yielded higher counts of
Paleoindian projectile points are within or around Florida’s karstic area, such as Gilchrist
County, along the Suwannee River, which has yielded the most with 148 specimens reported
with PIDBA.
44
The earliest radiocarbon dates firmly associated with human artifacts in unquestioned
contexts indicate people were living in north Florida by at least 11,050 B.P. (Hemmings 2004),
during the Clovis phase of the Early Paleoindian subperiod. While distinctive, fluted Clovis
lanceolate bifaces have been recovered from several north Florida rivers, only two sites have
yielded Clovis points from excavated contexts: the Silver Springs site in Marion County (Neill
1958) and the aforementioned Sloth Hole site in Jefferson County (Hemmings 1999). It is from
this latter site that the 11,050 B.P. date was obtained from a Clovis level.
Evidence for occupation of Florida during the subsequent Middle Paleoindian subperiod is
much more secure. The diagnostic Suwannee and Simpson lanceolate bifaces are relatively
common in north and central Florida, and although no radiocarbon dates have been obtained in
association with these artifacts, they are believed to date sometime around 11,000–10,500 B.P.
(Goodyear 1999). Two sites have yielded these point types in stratigraphic context: the Harney
Flats site in Hillsborough County (Daniel and Wisenbaker 1987) and the Wakulla Springs Lodge
site in Wakulla County (Tesar and Jones 2004). The final subperiod, the Late Paleoindian
(10,500–10,000 B.P.), saw the production of both fluted and unfluted forms of Dalton projectile
points elsewhere in the Southeast (Goodyear 1982), but evidence for a true Dalton phase in
Florida is limited. Dalton points appear to be transitional between the lanceolate forms of the
Early and Middle Paleoindian periods and the notched shapes of the Early Archaic period
(Ledbetter et al. 1996). Shallow-notched forms such as the Greenbriar point may represent a Late
Paleoindian manifestation in Florida.
The climate and landscape during the Paleoindian period were much different from those of
today. Not only was it cooler and drier than the present, but coastal sea levels and the inland
water table were much lower (Carbone 1983; Dunbar 2002, 2006a; Watts and Hansen 1988). The
scarcity of potable surface water sources is thought by some archaeologists to have played a
crucial role in the distribution of Paleoindian bands across the landscape (Dunbar 1991, 2006a;
Faught 2004; Milanich 1994; Neill 1964). They hypothesize that human groups frequented
sinkholes and springs to collect water and exploit the flora and fauna that were also attracted to
these “oases.” As an added bonus, many of these fresh water sources were located in areas of
exposed Tertiary-age limestone that had become silicified, providing Paleoindians with a raw
material source (chert) for tool manufacture. Thus, it is thought that permanent fresh water
sources (sinkholes, springs), along with locations of high quality chert, were primary factors
influencing Paleoindian settlement patterns in Florida.
The conventional view of Paleoindian existence in Florida has been that nomadic hunters and
gatherers wandered into an environment quite different from that of the present. Excavations at
the Harney Flats site in Hillsborough County (Daniel and Wisenbaker 1987) have altered this
view and many archaeologists now believe that Paleoindian people lived part of the year in
habitation sites that were located near critical resources such as fresh water.
2.6.2. Post-Paleoindian Period
Around 10,000 B.P., the environment and physiography of Florida underwent pronounced
changes due to climatic amelioration. These changes were interconnected and included a gradual
warming trend, a rise in sea levels, a reduction in the width of peninsular Florida, and the spread
45
of oak-dominated forests and hammocks throughout much of the state (Milanich 1994; Smith
1986).
Although sea levels rose significantly by the close of the Pleistocene, Early Archaic deposits
have been encountered in Florida waters. Some of the better known freshwater sites of this
period include Little Salt Springs, Warm Mineral Springs, and Page/Ladson (Milanich 1994;
Faught 1996). Assemblages from these sites demonstrate technological change with the
introduction of notched, concave based points (Bolen, Greenbriar or Hardaway) that suggest an
adaptation in resource procurement strategy (Faught 1996; Anderson and Hanson 1988). Sport
divers have also recovered Early Archaic artifacts in the Santa Fe, Ichetucknee and Wacissa,
Aucilla, Steinhatchee, Withlacoochee, and Oklawaha rivers (Milanich 1994).
In the marine environment, Early Archaic Dalton points have been reported in Tampa Bay
(Goodyear et al. 1983) and a Bolen point was recovered within 2–6 m of water, approximately
200 m east of the present shoreline at the Douglas Beach Midden site in St. Lucie County
(Cockrell and Murphy 1978; Pepe 2000). Later Early Archaic point types found in the Southeast,
such as Hamilton and the Kirk varieties, are surprisingly rare in Florida.
The Windover Pond site in Brevard County is a semi-permanent habitation site that has
produced a suite of radiocarbon dates indicating a minimum age of 6,980 B.P. and a maximum
age of 8,120 B.P. for burial activities (Doran 2002). Windover Pond has proven to be a crucial
site for interpreting Early Archaic lifeways, as its saturated nature and prolonged physical
stability have resulted in excellent preservation. There have been 168 human burials excavated
from the pond, 91 of which have contained human brain matter, and thus some of the oldest
DNA ever examined. These burials were generally flexed and oriented in comparable positions
to each other, signifying possible spiritual or religious significance. The ratio of interred males to
females and adults (over 20 years old) to subadults were comparable, indicating that all
community members were treated in a similar fashion. Preserved stomach contents offered
insight into diet. Wood and bone tools were preserved, and most of the burials were staked to the
base of the pond and covered with elaborately produced woven fabrics. Environmental
reconstruction was achieved through floral, faunal, palynological, and petrographic analysis, and
the dates of the semi-domesticated bottle gourd (Lagenaria siceraria) were pushed back 3,000
years earlier than what was previously accepted (Doran 2002; Rachel Wentz, personal
communication 2009). As the shoreline of 8,000 B.P. was lower than that of modern times, sites
comparable to Windover might exist in the shallow waters of the Atlantic Ocean.
The Early to Middle Archaic transition (8,000–7,000 B.P.) is marked by a shift in population
from the western side of the state to the eastern, and an increase in site size and frequency
(Milanich 1994). At this time, the tool assemblage also changed and projectile points are
characterized by convex based, stemmed point varieties, known as Florida Archaic Stemmed
points. Recorded submerged sites with Middle Archaic components are concentrated along
Florida’s eastern coast and in the St. Johns River area (Faught 1996), in particular, but also are
present at Little Salt Spring in Sarasota County and in Tampa Bay (Goodyear el al. 1983; Faught
and Ambrosino 2007). Numerous submerged Middle Archaic deposits have also been
documented in the Big Bend region north of Apalachee Bay. Sites like the Ecofina Channel, J&J
46
Hunt, and Ontolo illustrate the potential for submerged Middle Archaic remains, which are
generally found at depths of approximately 3.5 to 5 m (Faught 2002; Marks and Faught 2003).
Transgression of shorelines generally stopped between 4,000 and 5,000 B.P. As a result,
most Late Archaic sites, which post-date this period, were formed after sea levels stabilized.
However, coastal Late Archaic sites have been and continue to be susceptible to shoreline
erosion and are frequently redeposited from terrestrial to underwater contexts. Minor sea level
fluctuations are believed to have occurred around the terminal Late Archaic. Cultural materials
could have been deposited at this time when it is believed sea levels temporarily lowered. Late
Archaic components have been identified at the Douglas Beach Midden, where in addition to its
Early Archaic component, the site yielded a Newnan point and sharpened wooden stakes that
dated to Late Archaic (4,630 ± 100 years) at about 7 m depth (Cockrell and Murphy 1978; Pepe
2000). Moreover, the Apollo Beach (Warren 1968) and Venice Beach sites (Koski 1989) both
yielded water worn ceramics, in addition to shell tools and chipped stone artifacts.
The close of the Late Archaic, approximately 3,000–2,000 B.P., has also been inferred as a
period of lower sea levels. This assertion is based primarily on site distribution and the nature of
sites dating to this period (Ashley 2008; Russo 1992). Ashley (2008:126) points out that only one
site dating to this time period is documented in Duval County in the northeastern part of Florida,
attributing this minimal site occurrence to “environmental conditions related to sea level
fluctuations.” Russo’s (1992:113–114) analysis of the St. Marys region of northeast Florida and
southeast Georgia has led him to assert the potential for retreating sea levels during this interval,
noting that if this was the case, then sites of the time period might be situated in tidal flat settings
whose vegetation and drainage characteristics would have made them more hospitable to
habitation at that time. Further to the north, DePratter and Howard (1980) have recognized that
many of the Refuge phase sites (3,000–2,650 B.P.) (Thomas 2008:423) of the Georgia coast are
in tidal marshes, further substantiating this as a period of low water stand. As such, this interval
might represent a period of increased likelihood for encountering submerged pre-Columbian
cultural resources. However, this period of lower water levels was likely ephemeral, thus it is
suspected that any sites submerged during this interval would likely be close to the present
shoreline.
47
SECTION 2 – CURRENT RESEARCH
48
PAGE LEFT INTENTIONALLY BLANK
49
3.
GULF OF MAINE
This chapter reviews the Maine portion of the Atlantic OCS (Figure 3.1), which is within the
Gulf of Maine.
3.1. REGIONAL GEOLOGY
The Gulf of Maine represents the outer edge of the passive continental margin of
northeastern North America. In this region, a thick sedimentary sequence is underlain by a series
of fault-bounded basins. This series of rift basins, separated by horsts, was formed by the
extension and rifting of the Earth’s crust in the middle Triassic, creating the proto-Atlantic
Ocean. Some of the basins of the Gulf of Maine (Jordan, Crowell, Georges, and Wilkinson)
represent surficial expressions of this rift zone (Klitgord et al. 1988). These basins are filled with
early Mesozoic terrestrial clastic sediments and volcanic material, and are overlain in places by a
thick section of Cretaceous and Cenozoic sediments (Austin et al. 1980; Ballard and Uchupi
1975). The clastic sediments were derived from the weathering of the Appalachian Mountains to
the west, and represent both fluvial and shallow marine environments (Austin et al. 1980). In a
few locations near Stellwagen Bank, early Cenozoic sediments also crop out on the seafloor
(Uchupi 2004).
As the ice sheet retreated, till and glacial-marine sediment blanketed the deeper areas of the
Gulf of Maine (Schnitker et al. 2001). These deposits were later re-worked in depths less than
about 50 m and covered by Holocene mud in deep areas. Retreating ice reached the Maine coast
by about 15,000 calendar years ago (Borns et al. 2004). Near the coast and into the interior of
Maine, stratified moraines (moraine banks) were deposited in a general northeast-southwest
orientation (Hunter et al. 1996). Glacial-marine muddy sediment covers and is interfingered with
the moraines and is an abundant material in the coastal region inland to an elevation just over
100 m (Borns et al. 2004).
The inner continental shelf of the Western Gulf of Maine is divided into six different
physiographic regions based on bathymetry, relief and surficial sediments (Barnhardt et al. 2006;
Kelley et al. 1998; Kelley and Belknap 1991). The two physiographic zones most relevant to the
OCS are the Rocky Zone and Outer Basin. The Rocky Zone is the most spatially extensive area
of the Maine inner shelf, and probably the most diverse. It is a region of great bathymetric relief,
and is floored by bedrock with subordinate sand and gravel deposits. The Outer Basin is a flat,
muddy region that continues from near shore into the deeper Gulf of Maine. Rock outcrops occur
within this region, but are relatively small in area (Kelley et al. 1998). The surficial sediment of
the seafloor from the Maine coast to the 100 m isobath was mapped by Barnhardt et al. (1996a–
1996g). Most of the observational data for the maps was collected within Maine state waters
(inside the 3-mile limit). Work that extended into federal waters included the unpublished theses
by Lee (2006), Barnhardt (1996), and Shipp (1989).
3.2. RELATIVE SEA LEVEL CHANGES
North of Boston and west of Nova Scotia, the ocean accompanied the retreating glaciers
inland. Isostatic depression of the land by the weight of the ice allowed a progressively deeper
50
51
Subsidence
Figure 3.2.
Maine relative sea level curve modified from Kelley et al. (2010).
drowning north of Boston, reaching more than 100 m in central Maine (Thompson and Borns
1985). As the ice retreated from Maine, rebound of the land led to uplift of the region and local,
relative sea level fell rapidly to a lowstand of about 60 m depth at about 12,500 B.P. (Kelley et
al. 2010). The depth of the lowstand was estimated on the basis of drowned shorelines and a
drowned delta of the Kennebec River (Barnhardt et al. 1997; Kelley et al. 2003; Schnitker 1974;
Shipp et al. 1991). Dates establishing the time of the lowstand were obtained from wood
fragments and barnacle plates in cores of Kennebec River delta sediments (Barnhardt et al. 1997)
and cores containing Mya arenaria and Mytilus edulis, intertidal-shallow subtidal organisms,
from the lowstand shoreline complex off Saco Bay, Maine (Lee 2006). Figure 3.2 depicts the
relative sea level curve for the Maine coast based on these recent data.
To the south of Maine, the Merrimack River paleodelta was graded to a lowstand of about
43 m depth (Oldale et al. 1983). A date from wood fragments in a core here yielded a
12,200 B.P. radiocarbon age, which, if calibrated, was probably slightly older than the Maine
lowstand. Another lowstand shoreline was recognized on Jeffreys Ledge at about 50 m depth
(Oldale 1985a), and dated to 11,900 B.P. (Oldale et al. 1993). It should be noted that work on sea
level to the south of Maine in the Gulf of Maine contains few dated sea-level indicators.
Sea level rose rapidly from lowstand across the Maine continental shelf until about
11,500 B.P., when the rate of rise slowed dramatically (see Figure 3.2). The time between
11,500–7500 B.P. is informally termed “the slowstand” period in the western Gulf of Maine
because sea level rose only about 5 m, between 23 m and 18 m depth. Forebulge migration has
52
been invoked to explain the slowstand (Barnhardt et al. 1995), but part of this time coincided
with the catastrophic draining of a glacial lake in Canada at 8200 B.P. This release of water is
modeled to have led to uplift in coastal Maine (Kendall et al. 2008), which may have contributed
to the length of the slowstand.
Because of the slow rate of sea level rise between 11,500–7500 B.P., coastal processes had
time to cause considerable erosion of older, glacial landforms, but the eroded sand and gravel
formed numerous beaches and spits (Kelley et al. 2003). These, in turn, provided sheltered
environments in which fine-grained sediment accumulated in intertidal and shallow subtidal
settings. Many samples of Mya arenaria, Mytilus edulis, Crassostrea virginica as well as
Spartina sp. and Zostera marina, intertidal to shallow subtidal organisms, have been collected by
coring from the depth interval and radiocarbon dated to firmly establish the time and depth of
this period (Barnhardt et al. 1997; Kelley et al. 1992; Kelley et al. 2003).
It is important to note that the large quantity of datable objects from the slowstand interval
resulted from the greater amount of time available when local, relative sea level remained at
nearly the same elevation. Coastal storms also had more time to operate at nearly the same
elevation during this time, and glacial landforms (moraines, bluffs of glacial-marine mud) were
significantly eroded. It was this erosion that provided sand and gravel to build beaches and mud
to fill in estuaries and bays. Although the land surface from this time interval must be gone in
almost all locations, along with associated archaeological materials, deposits formed below mean
high water had an opportunity for preservation. Thus, lake and estuarine bottoms might have
been buried as they were drowned. Beaches would have been washed over with increasing
frequency as sea level rose across them. Beach migration must have occurred, but the paucity of
sediment and abundance of bedrock outcrops doomed the beaches of this period. Only off major
river mouths such as the Kennebec and Saco rivers is it likely that beaches migrated into their
contemporary positions. In the process of experiencing overwash, beaches may have comingled
human artifacts with washover sediments during their drowning.
After 7500 B.P., sea level initially rose very rapidly. This rapid rise may have prevented
complete destruction of constructional coastal features formed during the slowstand period and
abetted their preservation. The rise in sea level progressively slowed between 6500 B.P. and the
present. Many dates from the base of salt marshes across the State of Maine established the
progressive slowing of sea level rise and the uniform behavior of this slowing along Maine’s
coast (Gehrels et al. 1996).
3.3. MARINE TRANSGRESSION AND SITE PRESERVATION
The OCS off the coast of Maine extends seaward from the state’s 3-mile limit to the 100 m
isobath along the entire 3,478-mile long shoreline of Maine. The region is a natural continuation
of the inner continental shelf of Maine, and much of the material in Kelley et al. (1998) applies
here as well. The surficial geological maps of this region (Barnhardt et al. 1996a–g) depict this
area, although geophysical data are scarce for portions of the shelf deeper than the 60 m isobath.
Most of the seafloor seaward of the 3-mile limit is contained in the Rocky Zone and Outer
Basin physiographic zones (Kelley et al. 1998). Little research has occurred in the Outer Basin
because it generally lies beneath the late Quaternary lowstand of sea level (about 60 m). These
53
areas have remained continuously below sea level since deglaciation (Kelley et al. 1998), so
therefore are of little interest for archaeological site preservation insofar as no occupation could
have taken place.
In water depths less than the sea level lowstand depth, outcrops in the Rocky Zone were
submerged during deglaciation, emerged during the lowstand and were drowned during the ongoing transgression. The majority of any sediment covering the bedrock was eroded by waves in
exposed areas (Kelley et al. 2010). Some exceptions to that pattern may exist, however, where
outer shelf regions that were once terrestrial (shallower than 65 m) may be preserved. Such areas
may include regions seaward of the Saco and Kennebec River mouths, as well as areas within the
Wells Embayment where, because of the large volume of sediment derived from rivers during
the lowstand of sea level, some former terrestrial habitats may still remain (Figure 3.3).
Figure 3.3.
Inner Continental Shelf map of southern Maine (Barnhardt et al. 1996). The dark
blue line marks the state 3-mile boundary. Yellow represents sand deposits; blue,
mud; red, rock; and green, gravel. Location A points to the paleodelta of the
Kennebec River. A seismic line near point A is shown in Figure 3.4. B points to a
seismic line off Saco Bay and is shown in Figures 3.5 and 3.6. Location C points to
the Wells Embayment.
54
Figure 3.4.
Seismic reflection profile located near Point A in Figure 1 (from Barnhardt et al.
1997:Figure 8). BR indicates bedrock; TGL represents Thin Gravel Layer; D
indicates Delta; SG indicates Sand and Gravel. VC93-02 and VC93-03 are
vibracores that contained all sand and gravel.
Off the modern Kennebec River, a vast sand and gravel plain extends into federal waters (see
Figure 3.3). This landform is a paleodelta that was deposited as sea level fell to the lowstand
(Barnhardt et al. 1997). The most seaward parts of the delta were riverine environments that may
well have hosted early human immigrants. The rising level of the ocean eroded much of the area
(Figure 3.4), however, and acoustic reflectors interpreted as deltaic clinoforms are truncated by a
condensed Holocene section. There has been no research in this region since Barnhardt’s 1990s
work (Barnhardt et al. 1997), and there may be preserved sites that were sheltered near bedrock
outcrops that protected them from the brunt of marine transgression.
In outer Saco Bay, some sandy deposits from the depth/time of the lowstand also occur in
federal waters (Lee 2006). Early work supported by the Minerals Management Service (Figure
3.5) identified sand deposits surrounding rock outcrops in 50–70 m water depth. More detailed
seismic reflection observations coupled with vibracores found Holocene sand unconformably
overlying Pleistocene glacial-marine muddy sediment (Figure 3.6). Dates from intertidal fauna
(such as Mya arenari and Mytilus edulis) in 60 m depth led to a recognition that the lowstand of
sea level occurred about 12,500 B.P. (Lee 2006). Again, alluvial deposition may have sealed
55
Figure 3.5.
Inner Continental Shelf map off Saco Bay in southern Maine (Barnhardt et al.
1996). Location B is the same location represented as B in Figure 3.3. The red line
beside B is the location of a seismic line shown in Figure 3.6. Colors are the same
as in Figure 3.3.
archaeological deposits and protected them from subsequent erosion during transgression.
Likewise, off Wells Embayment, sand deposits are also associated with possible lowstand
shoreline positions (see Area C in Figure 3.3) (Shipp 1989; Shipp et al. 1989, 1991; Kelley et al.
2003). These features have never been studied in detail and no cores exist from this area.
In other areas near the Wells Embayment where the lowstand position of sea level lies in
federal waters, there are no observations available to evaluate the seafloor. No work has occurred
in these areas because the highly exposed nature of the region and likely paucity of sediment.
Archaeological sites are unlikely to be preserved in such settings because of this lack of
sediment.
Drowned terrestrial prehistoric sites with archaeological potential are probably focused
between the 15–25 m depth range, the depth range encompassed by the slowstand of sea level,
and the 55–60 m range, the sea level lowstand position. Because of the relatively steep, bedrockcontrolled bathymetric gradient, there are few locations along Maine’s OCS coast in the 15–25 m
range. Because of the irregularity of the bathymetric relief associated with bedrock, there are not
long, continuous stretches of OCS land at the lowstand depth.
There are particular exceptions to the model for site potential based on bathymetry.
Specifically, locations where unique landform configurations enclosed and protected areas from
wave action so that inundation took place by water spilling into the enclosed setting. The
existence of such protected environments came to light after scallop draggers discovered Middle
56
Archaic stone tools from off Mt. Desert Island, Maine, from about 20 m depth (Price and Spiess
57
Figure 3.6.
Seismic line indicated near B in Figure 3.5 (from Lee 2006).
2007), prompting a geophysical and coring study of the site (Kelley et al. 2010). The shoal from
which the artifacts were recovered is a partially eroded morainal complex with spits attached at
the ends and connecting the moraines (Figure 3.7). Seismic reflection profiles revealed multiple
acoustic reflectors within the spits that were correlated with lithologic changes in cores. The spits
unconformably overlie glacial-marine mud in seismic profiles, but cores could not penetrate past
a sandy, muddy gravel with abundant Crassostrea virginica and Mya arenaria shells. This unit
was abruptly overlain by a mud deposit with abundant Zostera marina stems lying on bedding
planes. Graded beds of sand and gravel with fragments of peat containing freshwater diatoms
58
Figure 3.7.
Multibeam image of submerged Bass Harbor morainal spit complex. Middle
Archaic artifacts have been recovered by draggers in this area. Modified from
Kelley et al. (2010).
capped the section through the spit. This overall stratigraphic section reflects medium energy
marine conditions (sandy gravel with oysters) that became suddenly very low energy (mud with
Zostera) probably as a consequence of increased shelter from waves by spit growth. Rising sea
level finally began washing gravel, shells, and freshwater peat blocks onto an accreting tidal flat
before the entire area drowned. All of the calibrated dates from the shells, peat fragments, and
Zostera fell within the slowstand interval (11,500–7500 B.P.). Where similar protected settings
exist, intact archaeological deposits are possible.
59
In the case of bedrock-sheltered embayments like Bass Harbor, various environmental
settings would have existed prior to inundation, and when more detailed data is available from
coring, it can be possible to further refine predictions about where archaeological sites might be
found. Areas that consist of mud sea floors—representing likely mud/tidal flats when subaerial,
with the mud likely the result of bluff retreat or re-suspended sediment from the erosion of tidal
flats as sea level rose—have no potential for having contained archaeological sites when the area
was subaerial. However, associated higher areas may have hosted sites because they provided
occupation areas close to a food source. Thus, what would have been higher elevations in
locations now above the 60 m isobath have some potential as site locations prior to inundation.
In other areas, extensive muddy bottoms lie near the lowstand depth and in federal waters,
but represent the retreat path of bluffs of glacial-marine sediment and tidal flats. As such, these
locations would not hold much promise for preserving drowned terrestrial sites in situ. Similarly,
off the Kennebec River paleodelta, sandy former deltaic areas are common (Barnhardt et al.
1997). However, these areas were transgressed by migrating barrier islands and spits, which
would have eroded any former archaeological sites.
It is important to note that existing bathymetry is not always capable of resolving potential
sites. The Bass Harbor site was a minor shoal on the nautical chart (Kelley et al. 2010). In other
locations, detailed surveys have revealed locations with great archaeological potential that the
nautical chart did not even hint at (Figure 3.8). The dark areas in Figure 3.8 are moraines that
were barrier islands on two occasions: once before the lowstand, around 13,000 B.P., and prior to
drowning about 8000 years ago. During those times, they may have been occupied and when
drowned, materials may have been buried and preserved. Features like this may be common in
the offshore area in water less than 60 m deep, but they cannot be recognized on nautical charts
or old bathymetric charts based on lead soundings.
All areas off the coast of Maine do not hold out equal probabilities of preserving terrestrial
environments and/or archaeological sites. The lowstand of sea level is well established at about
60 m depth, but is not as well constrained chronologically. Only a few Mya arenaria dates from
outer Saco Bay are reliable sea level indicators (Lee 2006); none of the wood fragments and
other shell dates from the Kennebec River paleodelta are related to tidal elevations. Thus, the
lowstand, a time when isostatically emerging land coincided with eustatic sea level rise, may
have lasted for hundreds of years. Such a long period of sea-level stability, as during the
slowstand, could have led to beach formation and the creation of wetland habitats attractive to
humans. No such localities have been found, however, possibly because most research in the
slowstand area has been off large river mouths (e.g., the Kennebec and Saco rivers) where
sediment deposits bury basins formed by bedrock or glacial deposits.
3.4. ARCHAEOLOGICAL SENSITIVITY AND PRESERVATION POTENTIAL
Based on the most current sea level curves for this region, archaeological sensitivity is
defined as follows:
•
No Sensitivity. Areas 60 m and greater in depth are considered to be areas with No
Sensitivity for prehistoric sites, since these areas were not subaerial during the LGM.
60
Figure 3.8.
Side scan sonar image of moraine complex approximately 5 km offshore of Wells,
Maine. Dark areas show highly reflective bottom types (till, gravel). Similar
features are likely to exist beyond Maine’s 3-mile limit in federal waters. Modified
from Kelley and Belknap (2003).
61
•
•
Low Sensitivity. This designation intends to cover areas exposed between the time of
the LGM and the earliest Clovis occupation. Given that the lowstand occurred at
approximately 12,500 B.P., which marks the beginning of the Paleoindian period,
there are no areas that fall in the Low Sensitivity designation for this region.
High Sensitivity. High Sensitivity areas include all areas within the OCS that are
shallower than 60 m.
As discussed in the previous section, submerged prehistoric archaeological sites on the
Maine OCS will be from the current coastline to a depth of 60 m, and may be found in either
inundated terrestrial or coastal environments (Figure 3.9). In the Late Pleistocene, terrestrial
environments extended to the 60 m isobath. As sea-level began to rise, coastal environments
moved landward across the previously subaerial landscape. Thus, rising sea levels created a
landward-moving mosaic of terrestrial and coastal settings. High potential terrestrial sites will be
those that offered living space associated with resource-rich environments: wetland edges (fresh
and salt), river and stream courses, and lake margins. Coastal sites will be areas that offer
protection from wind and waves (bedrock sheltered embayments) and occupation sites with
access to floral and faunal resources (beaches on spits, moraine crests, or beaches fringing
islands).
As is illustrated in Figure 3.10, site preservation potential is most likely to be best at 55–60 m
depth (lowstand) and between 15–25 m (associated with a regional “slowstand” of sea level).
Although slow sea level rise is likely to erode sites, the lowstand and slowstand periods represent
times when beach formation and wetland development would have attracted people (e.g.,
Almquist-Jacobson and Sanger 1995; Nicholas 1998), and some of these lower-situated sites
may have been preserved when spits formed below the surface and helped protect the sites from
marine transgression. During the lowstand, for example, there was possibly sufficient time to
allow burial of material to a great enough depth (thickness of deposit) that a site may have
survived transgression. Such would only have been true in areas of rapid sediment accumulation,
such as a delta, and/or moderate sediment accumulation but with shelter from large, erosive
waves.
Areas with High Preservation Potential for this region likely include regions seaward of the
Saco and Kennebec River mouths, as well as areas within the Wells Embayment where, because
of the large volume of sediment derived from rivers during the lowstand of sea level, some
former terrestrial habitats may still remain. In such areas, sites near the ocean or in deltaic
wetlands may have attracted people, and sites may have been buried deep enough to survive
transgression.
62
63
•
64
65
3) Rebound
4.
SOUTHERN NEW ENGLAND AND THE GEORGES BANK
The study area for this chapter includes the geographic region called Southern New England
and the Georges Bank. It is located within the North Atlantic Planning Area, south of the coast of
Maine (Figure 4.1).
Environmental settings, environmental conditions, and natural resources are important
factors to consider when assessing the potential for the presence of archaeological deposits that
are associated with human habitation sites inundated by eustatic or glacially-related sea level
rise. As Renfrew (1976) notes, “because archaeology recovers almost all of its basic data by
excavation, every archaeological problem starts as a problem in geoarchaeology.” The
complexity and variability of geological processes in general make every region or site unique,
and sediments comprising the massive expanse of seafloor within the Southern New England–
Georges Bank (SNE-GB) study area are no exception. Having a basic understanding of the
varied, evolving and dynamic geomorphology of the submerged landscape within this area,
approximately 40 percent of which was once shallow enough to be exposed land available for
human occupation prior to its inundation and transformation into part of the North Atlantic
continental shelf, is essential for assessing the SNE-GB study area’s pre-contact period
archaeological sensitivity.
4.1. REGIONAL GEOLOGICAL HISTORY
The geological history most relevant to the discussion and assessment of the pre-contact
period archaeological sensitivity of the SNE-GB study area is that of the Late Quaternary period
spanning the last 20,000 years and encompassing the Late Pleistocene and Early Holocene
epochs and southern New England’s nearly 12,000 years of archaeologically documented human
history. This period in time is marked by three major geological events: glaciation; ice retreat;
and sea level rise. While the basic structure of the southern New England coastline and the
adjacent continental shelf were created by glacial scouring and transport and the subsequent
erosion of sediments during glacial melting and retreat, secondary processes of relative sea level
rise, wave and tidal erosion, and subaqueous sorting and transport of sediments have further
transformed the geomorphology of the land-sea interface and the sea floor within the SNE-GB
study area.
This region is closely tied to the geological development of the Gulf of Maine to the north.
West of the Great South Channel, the shelf materials extend onto land as the Coastal Plain
(Thornbury 1965). The Georges Bank is an extension of these Coastal Plain sediments, but is
separated from the Appalachian Mountains by the almost 400-m deep waters of the Gulf of
Maine (Uchupi 1968). The removal of Coastal Plain materials from the Gulf of Maine is
presumed to be a consequence of a series of Pleistocene glaciations (Uchupi 2004).
The terminal moraine of the Laurentide Ice Sheet (LIS) formed as the most recent ice
advance reached its southernmost extent approximately 23,000 years ago (Balco et al. 2002;
Denton and Hughes 1981). The moraine is located approximately 400 km to the south of Maine
at Long Island, New York, and roughly coincides with the southernmost area of bedrock
66
exposure (Uchupi 1970; Uchupi et al. 2001). While still a prominent portion of the Long Island,
67
68
Figure 3.1.
Gulf of Maine study region.
Nantucket, and Martha’s Vineyard landscape (Balco et al. 2002; Hartshorn et al. 1991), the
moraine no longer has a significant bathymetric expression as it extends east across Georges
Bank. Reworked by waves and tides, the Georges Bank portion of the moraine is recognized
largely by the distribution of gravel (Schlee and Pratt 1970).
Charted water depth within the SNE-GB study area today ranges from a minimum of about
5.5–10 m to a maximum of approximately 2,012 m. Average depth across the entire study area is
calculated at approximately 120 m. Approximately 20,000–18,000 years ago, however, sea level
was estimated to have been approximately 90 m lower than present, and the vast majority of the
OCS was subaerial (Oldale 1985b; 1985c; Pirazzoli 1991; Uchupi et al. 2001). This was a time
when the LIS associated with the Wisconsin glaciation had advanced southward to its terminal
position, corresponding with the terminal and recessional moraine formations of poorly and wellsorted glacial till (i.e., boulders, rocks, gravel, sand, silt, and clay).
The LIS that spread across the SNE-GB study area was characterized by bulges or “lobes” in
the ice front that filled in the large basins of the existing, pre-glacial, topographic surface. The
four lobes that occupied the SNE-GB study area (from west to east) along the LIS front were the
Connecticut Valley Lobe, the Narragansett-Buzzards Bay Lobe, the Cape Cod Bay Lobe, and the
Great South Channel Lobe. The advance and retreat of these lobes led to the formation of
morainal deposits of glacial till consisting of soil, sediments, decomposed rock, and fragmentary
bedrock collected by the ice as it flowed southward across the region and formed Long Island,
Block Island, Cape Cod, Martha’s Vineyard, and Nantucket Island. Sloping away, south and east
of these morainal structures was an extensive outwash plain formed by deposits of finer materials
carried away from the ice sheet lobes in meltwater flows. North of these morainal structures and
Cape Cod, banks, basins and deep troughs cut by ice streams left the shelf more topographically
irregular and created Stellwagen and Tillies banks, Jeffreys Ledge, Race Point Channel,
Stellwagen and Scantum basins and the Massachusetts Shelf. East and south of Cape Cod,
outwash from the melting Cape Cod Bay and South Channel lobes deposited vast plains of sand
and gravel which today comprise Nantucket Shoals and Georges Bank (Oldale 2001a, 2001b).
Sloping of these sediments led to a gradation in sediment sorting and a decrease in elevation of
the plain moving away from the moraines’ topographic highs.
After reaching its apex ca. 18,000 B.P., the Wisconsin glaciation began receding because of a
climatic shift towards a cycle of global warming. Meltwater from the shrinking ice sheets was
funneled into rivers and returned to the world’s ocean basins.
Runoff from the melting ice sheet was also trapped behind the region’s terminal and
recessional moraines that acted like earthen dams, thus producing a series of proglacial lakes
covering an area 21,235 square miles in size with a combined volume of 132 cubic miles of
water, assuming an average lake depth of 10 m (Uchupi et al. 2001). Uchupi et al. (2001) have
argued that some of the depositional and erosional features on the OCS, including within the
SNE-GB area, were produced by catastrophic discharges of large volumes of water from these
proglacial lakes over about a 5,000 year period, which in some cases transported course debris
via gravity flows across hundreds of miles of the shelf and into the deep sea. They also have
argued that the visibility of these catastrophic morphologies suggests that much of the surface of
the OCS was little modified by the late Pleistocene–early Holocene marine transgression,
69
possibly because a rapid rise in sea level allowed for preservation of relict features (Uchupi et al.
2001).
By 14,000 B.P., southern New England was free of glacial ice and by 12,000 B.P. nearly all
of New England had become open to plant colonization (Jones 1998). Deglaciation heavily
reworked the landscape of New England. Moraine and other ice-contact and outwash features
left old land surfaces covered in sand and rock. Glacial meltwater deeply scoured and rapidly
filled other locations. Ice and sediment dams produced extensive proglacial lakes throughout the
region, such as Lake Hitchcock that filled the Connecticut River Valley. When these lakes
drained, extensive sandy plains and wetland systems evolved in their places.
Vegetative colonization occurred fairly rapidly depending on local soils, hydrologic and
topographic constraints. Colder and drier conditions prevailed 12,000 years ago. Overall, the
pattern was one of warm summers and severe winters in a relatively arid climate (Jones 1998).
The sequence of sub-regional plant succession at around 11,000 years ago was diverse,
reflecting local temperature gradients, soil conditions, precipitation, topography, and changes
associated with fires, floods and storm patterns. Human adaptations to the resources of a given
environment occurred at a local rather than regional level. By this time, a true forest canopy
blanketed most of southern New York and New England. This forest was unlike any currently
existing in North America, as it contained an admixture of warm-weather deciduous tree species
within an otherwise boreal forest. This situation was particularly evident along the coast in
southern New England where a pine-oak forest had established itself (Jones 1998).
An important climactic shift occurred abruptly in the Northeast, associated with the Younger
Dryas event. However, the precise timing of this event recently has been called into question. For
decades, researchers have worked under the understanding that the Younger Dryas lasted from
approximately 11,000–10,000 B.P. (e.g., Fairbanks 1989:639; Mayewski and Bender 1995;
Taylor et al. 1993). However, recent research indicates a likely range of 12,900–11,700 B.P.
(e.g., Bard et al. 2010; Carlson 2010:383; Meltzer and Holliday 2010:8). The Younger Dryas
event resulted in a shift to cooler, moister conditions and stormier weather in much of the region,
and its end appears to have been very abrupt, with climate patterns shifting to one of increased
warming over as little as three years. The onset of the Holocene is marked by an interval of rapid
global climatic warming and reduction in the ice sheets. The transition from the Younger Dryas
to the Holocene represents a period of rapid vegetation change as plant communities shifted their
ranges in response to milder growing conditions. The rapidity with which the plant community
changed indicates expansion of individual species occurred from scattered refugia where species
such as white pine, oak, and hemlock had maintained relict populations throughout the Younger
Dryas event. Estimated temperatures in southern New England ca. 9000 B.P. were comparable to
those of today. In southern New England, white pine established itself as the dominant species.
Pine forests were mixed with significant populations of oak and some birch at this time, while
spruce was all but gone. Generally speaking, vegetation change between 10,000–6000 B.P. was
more gradual and predictable than in the previous millennia (Jones 1998).
The distribution of late Pleistocene fauna in the Northeast is poorly recorded. Southern New
England’s coastal pine-oak forest probably supported most of the boreal forest animals and some
of the temperate forest animals such as mastodon, stag-moose, woodland muskox, giant ground
70
sloth, caribou, elk, moose, giant beaver, long-nosed peccary, flat-headed peccary, white tailed
deer, flying squirrel, snow-shoe hare, beaver, muskrat, red squirrel, porcupine, woodchuck, otter,
fisher, long-tailed weasel, raccoon, gray squirrel, striped skunk, and others. Long Island Sound
and Narragansett Bay must have supported rich marine resources as well. Sea level change was
less dramatic and rapid along the southern New England coastline than it was to the north. This
meant that more productive estuarine habitats could form behind barrier beaches and along
protected stretches of the coastline (Jones 1998).
Early Holocene faunal communities in the Northeast are somewhat better understood than
those of the terminal Pleistocene. Unfortunately, it is difficult to separate out which animal types
belong to the early, middle, or late Holocene. Estuarine habitats appear to have been established
along protected shorelines in southern New England during the early Holocene (Gayes and
Bokuniewicz 1991). Such settings would have supported an abundance of shellfish, as well as
sea mammals, which fed upon them. The early Holocene forests of southern New England would
have contained a limited diversity of large game mammals, but an abundance of small game
mammals. Between 10,000–6000 B.P., increasing numbers of oak, especially in southern New
England, provided an important seasonal food resource for humans as well as animals, such as
white-tailed deer, turkey, and bear. Archaeological evidence suggests that anadromous fish
species established themselves in the Northeast by this time as well. Species such as shad and
salmon would have provided rich, seasonally predictable food resources to humans in southern
New England at this time (Jones 1998).
4.2. RELATIVE SEA LEVEL CHANGES
The retreat, thinning, breakup, and final disappearance of the LIS from southern New
England by about 14,000 B.P. did not mark an end to the ice-driven morphological alterations of
the southern New England land-surface or the adjacent and exposed continental shelf within the
SNE-GB study area (Uchupi et al. 1996). Worldwide melting of the continental ice sheets led to
the return of water to the ocean basins and a concomitant rise in global sea level; however, the
sea level curves and the complex interplay between isostatic and eustatic forces was markedly
different north and south of Cape Cod.
Although it is difficult, if not impossible, to construct a model of global sea level rise,
because of local neotectonism, a sea level rise curve from Barbados has been identified by
Uchupi et al. (1996) as a close approximation of the response of sea level to Wisconsin glacial
decay. The net rate of sea level rise varied locally as differences in the landscape’s materials,
morphology, and degree of crustal depression affected the interplay between isostatic and
eustatic conditions. Local rates of sea level rise are determined through radiocarbon dating of
salt-marsh peat deposits, which are considered accurate indicators of relative sea level (Redfield
and Rubin 1962; Oldale 1992). The general trend of rapid sea level rise during this period,
however, did not follow a smooth curve, but instead fluctuated and was punctuated by episodes
of still-stand and negative sea level oscillations during times of climatic cooling and glacial
advance (Rampino and Sanders 1980). At the glacial maximum, sea level was about 100 m
below its present level. From this point it rose at a rate of about 11 m per 1,000 years as the
eustatic increase in sea level outpaced the generally slower isostatic rebound of the Earth’s crust,
formerly depressed by the weight of glacial loading. By 12,000 B.P., sea level was
71
approximately 70 m below present sea level, and was at 40 m below present by ca. 10,000 B.P.,
20 m below present by ca. 8000 B.P., and about 10 m below present by 6000 B.P. (Oldale 1992).
After about 6000 B.P., the net rate of sea level rise began to slow significantly and gradually
approached its present rate. Sea level approached its present level at about 1,000 years ago, and
continues to rise at around a 2–3 mm per year (Uchupi et al. 1996:23).
Glacio-isostatic adjustment in the Cape Cod region, however, complicates the local relative
sea level rise history of southern New England, so global isostatic curves are of limited use there.
Curves developed for New England depart significantly from the Barbados curve. As a result of
crustal depression by glacial loading, relative sea level in northeastern Massachusetts was about
30 m higher than its present level at about 14,000 B.P. This high-level stand, documented by
emerged glaciomarine sediments, raised paleoshorelines and ice-contact deltas. The waters south
of Cape Cod were not inundated during the marine transgression, as isostatic rise in the
peripheral bulge resulting from glacial unloading in adjacent areas exceeded the rate of eustatic
rise in sea level prior to about 16,000 B.P.
The high-level stand north of the Cape was short-lived, as the crust rebounded rapidly when
its glacial ice load was removed. As the crust rose, sea level dropped and by 12,000 B.P.,
reached a post-glacial low-stand -43 m below present sea level off the coast of northeastern
Massachusetts within the southwestern Gulf of Maine. Features associated with this sea-level
regression include the paleodelta off the Merrimack River, a submerged barrier beach and lagoon
on Jeffrey’s Ledge, the seaward limit of shelf valleys off of New Hampshire, submerged terraces
off of Maine, and a regressive unconformity in Penobscot Bay, Maine (Uchupi et al. 1996). A
slowing of crustal uplift, coupled with an increase in eustatic sea level rise shortly after
12,000 B.P., caused sea level to begin rising again along the coast of northeastern Massachusetts
and throughout the rest of southern New England. Between ca. 9500 and 6000 B.P., Stellwagen
Bank, most of the Billingsgate Shoal moraine, Cape Cod Bay, Nantucket Shoals, and Nantucket
Sound were all drowned (Uchupi et al. 1996).
4.3. MARINE TRANSGRESSION AND SITE PRESERVATION
Generally speaking, episodes of marine transgression are essentially periods of erosion, a
destructive process that creates less than ideal depositional sequences from an archaeological
perspective (Belknap and Kraft 1985; Goff et al. 2005; Kraft 1971, 1985; Kraft et al. 1983,
1987). Marine transgression proceeds in one of two ways: by “shore-face” retreat, when the
coastline slowly regresses inland, or by “stepwise” retreat, when in-place drowning of coastal
features occurs (Waters 1992).
Shore-face retreat describes the erosion of previously deposited sediments by wave and
current processes as the shoreline transgresses and is the dominant inundation regime during the
marine transgression process (Waters 1992). As the glaciers melted and sea level rose, beachface and shore-face erosional zones, offshore of the present southern New England coastline
within the SNE-GB study area, sequentially passed across the subaerially exposed portions of the
continental shelf outwash plain. Older sediments that had been deposited in coastal and terrestrial
environments inland of the shoreline were reworked, first by the swash and backwash processes
upon the beach face and then by the waves and currents associated with the upper shore-face
breaker and surf zones. The erosion associated with the continuous transgression of the sea
72
reworked these deposits into a thin unconformable geological unit of transgressive lag (i.e.,
gravel and coarse sand deposits) forming the top of a time-transgressive geological unit known
as a “marine unconformity” (i.e., the surface defined by the top of the buried paleosol and the
base of the overlying marine deposit). Reworked terrestrial and coastal sediments are referred to
as “palimpsest sediments” (Swift et al. 1971), and the erosional surface, marked by the depth of
the maximum disturbance by transgression, is called the “ravinement” surface. This surface often
shows up quite clearly in sub-bottom profiler data and can be a useful indicator for the presence
of relict paleolandforms (Belknap and Kraft 1985; Kraft 1971; Waters 1992). Shore-face retreat
would have probably been the prevailing marine transgressive regime in the unprotected portions
of paleoshorelines within the SNE-GB study area, especially during still-stand episodes and after
ca. 6000 B.P., when the regional rate of sea level rise appears to have slowed considerably.
Alternatively, and to a lesser extent, marine transgression also occurs by the process of
stepwise retreat, which is the sudden inundation or in-place drowning of coastal landforms and
sediments (Rampino and Sanders 1980; Sanders and Kumar 1975a, 1975b). Stepwise retreat
most commonly occurs at times and in areas of rapidly rising sea level, where the coast is
quickly subsiding and the gradient of the transgressed surface is shallow. In this case, instead of
the waves and currents of the shore-face and beach face sequentially reworking older sediments
during transgression, the breaker and surf zones jump from the active shoreline to a point farther
inland, submerging the older coastal landforms and sediments in an area seaward of the more
destructive breaker and surf zones. The surf and breaker zones then stabilize and develop a new
shoreline farther inland. Instances of in-place drowning during stepwise retreat, preserving
forested uplands, barrier-island and lagoonal sequences, and other relict shoreline features, have
been documented in a variety of places along the Atlantic coast (Rampino and Sanders 1980;
Robinson et al. 2004; Sanders and Kumar 1975a, 1975b).
Evidence of intact paleosol deposits from unprotected waters in excess of 1–2 miles from
shore in the Northeast has thus far proven exceedingly rare (John King, personal communication
2004). One documented instance of a contextually intact, stratified paleosol deposit has been
identified in a high-energy environment 8–10 miles offshore in Nantucket Sound using existing
environmental data, sub-bottom profiles, and vibracoring (Robinson et al. 2004). Sub-bottom
profiler reflectors recorded in this area were tested with coring and found to be produced by a
distinct layer of intact paleosols (i.e., a thin ravinement horizon consisting of marine sediments
with shell hash intermixed with a partially reworked organic rich A0-horizon of duff, overlying
organic A-horizon soils, oxidized B-horizon soils, and C-horizon sub-soils) buried under
approximately 2 m of reworking marine sediments.
Subsequent macro-fossil analyses of cores from several different loci within Nantucket
Sound identified several terrestrial ecozones (an upland deciduous forest floor, a shallow fresh or
brackish water marsh, and a shallow freshwater pond or swamp). Organic material (i.e., a large
piece of birch wood and a plant seed) contained in two of these cores was AMS-radiocarbon
dated by the Woods Hole Oceanographic Institution to approximately 4500 B.P. and 10,100 B.P.
(John King, personal communication 2004). The results of the coring and dating corresponded
well with the modeled general locations of these ecozones at these approximate times based on
currently available local sea level rise models that had been applied to existing bathymetry.
Utilizing this method of paleosol presence/absence detection has proven effective during other
73
investigations conducted throughout the Northeast (Herbster et al. 2004; Leveillee et al. 2002;
PAL 2003a, 2003b, 2004, 2005a, 2005b, 2005c; Robinson and Ford 2003; Robinson and Waller
2002; Robinson et al. 2004, 2005). Similar approaches have also been applied by Fehr et al.
(1996), Maymon et al. (2000), Klein et al. (1986), and Riess et al. (2003).
Although shore-face retreat is the dominant transgressive regime, it is anticipated that there
were numerous isolated occurrences of stepwise retreat also within locally favorable conditions
throughout much of the SNE-GB study area.
4.4. ARCHAEOLOGICAL SENSITIVITY AND PRESERVATION POTENTIAL
A review of the available literature revealed that although there is an extensive inventory of
ancient Native American archaeological sites spanning the entire pre-contact period on land in
Connecticut, Massachusetts, and Rhode Island, no pre-contact period archaeological deposits
have been identified to date within federally-controlled waters in the SNE-GB area. However, all
of Nantucket Sound, which includes the portion of the SNE-GB that is encompassed by the
Sound, has been determined by the Keeper of the National Register of Historic Places to be
eligible for inclusion in the National Register under all four criteria for evaluation and as a
traditional cultural property that has:
…yielded and has the potential to yield important information about the Native
American exploration and settlement of Cape Cod and the Islands, and as an
integral, contributing feature of a larger, culturally significant landscape treasured
by the Wampanoag tribes and inseparably associated with their history and
traditional cultural practices and beliefs (Advisory Council on Historic
Preservation [ACHP] 2010).
The Keeper also acknowledged the importance of the Nantucket Sound seabed as “former
aboriginal lands of the Wampanoags and the potential location for intact archaeological sites”
(ACHP 2010), based on the identification of deposits of archaeologically sensitive organic
sediments deposited in a terrestrial environment discovered in what is today a high energy
marine environment 8–11 miles offshore (Robinson et al. 2004).
The three buried terrestrial deposits were identified by a marine archaeological
reconnaissance survey conducted for the Cape Wind Offshore Energy project (Cape Wind).
These deposits were interpreted to be an intact forest floor and quiet shallow aquatic areas (e.g.,
a freshwater pond, headwaters of an estuary, or a relatively close coastal pond), with AMS
radiocarbon dates for the different samples ranging from 5490–10,100 B.P. (Robinson et al.
2004). The identification of inundated terrestrial sediments, described by the Massachusetts
SHPO as a “major scientific discovery,” was significant as the first recorded instance in southern
New England of contextually intact paleosols systematically located by archaeologists in a high
energy marine environment so far offshore (MA SHPO 2009).
Equally important was the fact that the inundated terrestrial deposits were identified through
a phased, systematic and scientific archaeological investigative approach. This approach
involved conducting an archaeological sensitivity assessment followed by a marine remote
sensing reconnaissance survey and geotechnical sampling program. The general area where the
74
paleosols were found was identified by Robinson et al. (2004) as archaeologically sensitive and
likely to contain intact inundated landforms during their 2003 archaeological sensitivity
assessment of the Cape Wind project area. The specific areas containing paleosols were
identified through a systematic survey process that involved using sub-bottom profiler data
collected during the marine archaeological reconnaissance survey to identify buried acoustic
reflectors with potential to represent relict terrestrial landforms. A select number of these
reflectors were then chosen for “ground-truthing” via a program of geotechnical sampling (i.e.,
vibracoring) to determine whether the source of the reflector was a stratified relict landform with
sensitivity for containing contextually intact ancient Native American archaeological deposits.
The implications of the discovery of intact buried paleosols representing different elements of a
partially preserved terrestrial paleolandscape are that:
1.
The locations of such deposits are predictable and may be identified fairly easily and
at a comparatively low cost within a cultural resource management context (by
combining the archaeological data needs with those of the project engineers) using
existing technologies and long-available marine archaeological survey techniques;
2.
Intact elements of the archaeologically sensitive paleolandscape did survive the early
Holocene marine transgression on a very localized level and can exist in high-energy
marine environments a significant distance offshore; and
3.
Study area-specific background research and geophysical survey and geotechnical
testing are required to identify archaeologically sensitive submerged paleosols—
broad sensitivity statements about large areas of sea floor absent of locally collected
geophysical and geotechnical data (except for characterizing areas of the sea floor
that were once exposed as having sensitivity and those that never were exposed as
having low sensitivity) is an unadvisable management strategy.
Robinson et al.’s (2004) efforts to identify archaeologically sensitive elements of the
submerged paleolandscape were comparatively unique. Up until the last 10 years, significant
efforts to identify submerged paleosols and ancient Native American archaeological deposits
were not a regular element of compliance-related marine archaeological investigations. As a
result, the absence of any identified sites within the SNE-GB study area must be considered more
a function of the negligible amount of underwater archaeological research that has been
conducted thus far in the region to identify pre-contact period submerged sites than a reliable
indicator of the potential for such sites to exist within the SNE-GB study area.
Numerous predictive models have been developed to assess archaeological sensitivity and
assist in locating pre-contact archaeological deposits on shore with great success. However, it is
not clear that such terrestrial models would logically be equally applicable to the offshore
environment. While it is easy to assume that existing bathymetry of the seafloor is a direct
correlate to the topography onshore, it is unlikely for such to be the case. Instead, bathymetry
should only be considered a vague, highly disturbed and reworked shadow of the
paleolandscape’s formerly exposed geomorphology, and survey efforts should focus on
identifying the intact relict elements of the pre-inundation paleolandscape that are preserved
buried beneath the overlying protective layer of marine sediments. Marine archaeological
75
surveys in a wide variety of offshore environments throughout southern New England in the past
10 years have repeatedly shown that the surface of nearly all of the formerly exposed
paleolandscapes have been either disturbed significantly or removed altogether and reworked by
the erosive forces associated with the marine transgression and modern wave and tidal current
regimes. These surveys have also shown that the preservation of intact paleolandsurfaces appears
to be an exceedingly rare occurrence, as the discoveries of paleosols within the Cape Wind
project area is the only example recorded to date of such deposits preserved so far offshore.
When paleosols are preserved, they are preserved on a very localized level as a result of a unique
combination of environmental circumstances that protected them from destruction.
Submerged paleolandforms potentially containing archaeological deposits found in the
originally proposed offshore Cape Wind project area were found in areas that were relatively low
on the more protected eastern flank of the Horseshoe Shoal. This is an area that was rapidly
inundated and buried, and consequently survived the erosional effects of marine transgression
and subsequent modern impacts from waves, tidal currents, and human activities. Generally
speaking, the prerequisite for preservation of inundated sites is burial in terrestrial or low-energy
marine sediments prior to the transgression of the ocean’s rising waters (Waters 1992). In these
cases, sites will be preserved if the sediments they are in remain below the depth of shore-face
erosion that occurs during and after the marine transgression process, and have not undergone
substantial sediment reworking following inundation.
Based on published rates of sea level rise above and below the Cape, the study area could
have been available for human occupation from ca. 12,000–10,000 B.P. (i.e., during the
Paleoindian and Archaic cultural periods). Progressively smaller portions of the area would have
been available thereafter up until about 1000 B.P., when local sea level had reached a point
within approximately 1 m of its current level. Prior marine archaeological survey coverage is
lacking within the study area; however, the relatively protected, shallow nature of this area
suggests that it is likely to possess conditions that would favor preservation of intact buried
paleosols.
Thus, the archaeological sensitivity of the SNE-GB study area can characterized as one of
three categories representing each area’s potential for containing pre-contact period
archaeological deposits: No Sensitivity; Low Sensitivity; or High Sensitivity (Figure 4.2).
The area designated as having No Sensitivity lies below the projected -107 m sea level
lowstand corresponding with the glacial maximum of ca. 20,000–18,000 B.P. This area
encompasses an estimated 11,857 square miles of sea floor or 46.7 percent of the overall SNEGB study area. It is presumed to have always been under water and was never subaerially
exposed in the history of a human presence in the Northeast; hence, it has no potential for
containing ancient Native American habitation sites.
The Low Sensitivity area lies between the -107 m and -70 m sea levels, the latter of which
corresponds to ca. 12,000 B.P. and the time around which archaeological evidence indicates the
first human colonists began arriving in the region. While it is unlikely that this portion of the
study area contains any archaeological sites, it cannot be ruled out entirely. The Low Sensitivity
area encompasses an estimated 3,214 square miles of sea floor or 12.7 percent of the overall
SNE-GB study area.
76
77
78
Figure 3.9.
Archaeological Sensitivity map for the Gulf of Maine study region.
The area designated as having High Sensitivity includes the portion of the SNE-GB study
area that extends between the -70-m sea level of ca. 12,000 B.P. and the 3-mile nearshore limit
of federal waters and the SNE-GB study area. This 3-mile line also corresponds closely with sea
level at around 6000 B.P. Thus, it is within this portion of the SNE-GB study area that there was
exposed land available for habitation ca. 12,000–6000 B.P. by ancient Native Americans
associated with the Paleoindian, Early Archaic, and the early part of the Middle Archaic periods.
The High Sensitivity area encompasses an estimated 10,316 square miles of sea floor or 40.6
percent of the overall SNE-GB study area.
Archaeological research on land has repeatedly demonstrated that ancient native peoples in
the Northeast and elsewhere sought the most productive ecological zones within their cultural
landscapes, especially in those areas that offered diverse resources consistently on either
seasonal or year-round bases. Some of the richest habitats of diverse flora, fish, and wildlife are
found near the junction of land and water, both fresh and salt. Riparian corridors consisting of
rivers, streams, and estuaries, their beds, banks, and floodplains, along with the soils, plants, and
animals that exist there are among the most productive biological systems in the world. Areas
where such elements of the formerly exposed paleolandscape were preserved intact would be
most likely to contain archaeological deposits.
Archaeological site types that could be present within the High Sensitivity area would
include the full range of site types described in Chapter 2 above, as well as Paleoindian and Early
Archaic coastal site types that have yet to be encountered and identified and have no corollary in
the present terrestrial archaeological record (e.g., large, long-term coastal base camps, medium
and small special-purpose activity areas, coastal fishing sites, transportation corridors, semipermanent habitations, and burial sites distributed along formerly exposed, inundated relict river
margins, floodplains, and terraces).
Further delineation of the High Sensitivity area’s archaeological potential is not possible
without conducting area-specific geophysical survey and geotechnical sampling directed at
locating archaeologically sensitive paleosols, given the presumed discontinuous and localized
nature of their preservation throughout this offshore area (based on the experiences from the
Cape Wind project-related discoveries in Nantucket Sound and other archaeological surveys
conducted throughout southern New England waters).
79
5.
NEW YORK AND NEW JERSEY
This chapter addresses the OCS off the coast of New York (principally Long Island) and
New Jersey, which includes the curve in the shoreline in this region referred to as the New York
Bight (Figure 5.1).
5.1. REGIONAL GEOLOGY
To understand where on the OCS archaeological sites might be preserved intact, it is
necessary to understand the geomorphology of the region and the events of the Late Quaternary
period that shaped the landscape now submerged offshore. Beyond the dramatic changes taking
place at the end of Pleistocene when the continental ice sheet was in retreat, it is also critical to
understand subsequent processes in the Holocene during the course of marine transgression and
afterward that affected the potential for site preservation. The Late Pleistocene was a time in
which the landscape of the region surrounding the New York Bight was transformed by a variety
of geological events and processes, the most important of which were associated with glacial
retreat and subsequent sea level rise. The evolving landscape over the 12,000 or more years of
human presence in the region has been reconstructed through studies of the current sea floor and
dry land connected to it. A picture of the changing conditions characterizing the OCS in the New
York Bight is presented in this section.
Around 20,000 B.P., when the Wisconsin glacier had reached its maximum southern extent,
sea level was estimated to have been approximately 120 m lower than present, and the vast
majority of the OCS was subaerial (Dillon and Oldale 1978; Wright et al. 2009). It was at this
time that terminal and recessional moraine formations were created from poorly and well-sorted
glacial till (i.e., boulders, rocks, gravel, sand, silt, and clay) on Long Island and contiguous areas.
The ice front in the vicinity of Long Island created terminal and recessional moraines that in
turn created other major features in the geography of the region. The two major moraines include
the Ronkonkoma, which extends along the length of Long Island, and the northeast trending
Harbor Hill Moraine, which extends offshore from Long Island at Orient Point and continues
past Plum Island and Fishers Island to the south coast of Connecticut–Rhode Island near Point
Judith (Uchupi et al. 2001:133). The morainal deposits along these features consist of soil,
sediments, decomposed rock, and fragmentary bedrock scoured over the landscape as the ice
flowed southward across the region. Sloping away from the ridges of glacial till was an extensive
outwash plain of finer materials carried away from the ice sheet in meltwater flows. The
sediment flowing away from the ice front became size sorted with distance, resulting in a
decrease in elevation across the plain, seaward from the moraine (Oldale 2001a, 2001b).
Immediately south of the ice front, the massive weight of the glacier compressed the land
surface, creating a foredeep, or elongated depression, thought to have had relief of 200–300 m
depth and beyond that a peripheral bulge with relief of about 70 m (Peltier 1982; Uchupi et al.
2001:127). The seaward edge of the foredeep south of Long Island probably was located in the
vicinity of the present 40–50 m contours. Dillon and Oldale (1978) inferred that the peripheral
bulge covered the rest of the shelf ending southwestward of an inflection zone trending
80
northwestward to the coast from the shelf’s edge. Despite model predictions concerning the
81
82
Figure 3.10. High Preservation Potential areas for the Gulf of Maine study region.
position and relief of the peripheral bulge and foredeep (Peltier 1982), it is difficult to define its
location based on the drainage patterns observed across the shelf (Uchupi et al. 2001:127).
Quantifying the relief of the foredeep and peripheral bulge and the character of rebound during
the course of marine transgression is not a straightforward matter. For example, comparison of a
peat off Long Island to the Fairbanks (1989) sea-level curve indicates that the former foredeep
has only rebounded isostatically 27 m during the last 10,000 years such that the foredeep may
not have been as deep as Peltier (1982) suggested or else much of the rebound occurred before
the marine invasion (Uchupi et al. 2001:127).
The foredeep and peripheral bulge were not the only topographic features occupying the
exposed continental shelf of the New York Bight. This land area was also dissected with
drainages that have been observed on the current seafloor. The preservation of the drainage
patterns after transgression is not completely understood, and may be attributed to sedimentation
rates, rapid submergence (e.g., from forebulge collapse), stream capture, headland protection,
and/or bedrock controlling stream channels. The most prominent drainage channel preserved in
the New York continental shelf is the Hudson Shelf Valley (known as the Hudson Canyon),
which extends 130 km southeast from the current Hudson estuary to the outer continental shelf,
and is incised up to 100 m below the adjacent shelf surface. The central portion of the Hudson
Shelf Valley is deeper than the valley head or river mouth, as a result of being incised into the
glacial forebulge that later subsided. The mouth of the Hudson Shelf Valley is 9.5 km wide and
is adjacent to the Tiger Scarp on the southwest, a landform displaying 18 km of relief which was
the shoreline at roughly 10,000 B.P., and the Fortune Scarp to the east, a contemporary feature
that displays 10 m of relief (Freeland et al. 1981:399–402). Various buried channels have been
identified through seismic data, reflecting smaller drainages that have filled completely with
Holocene sediment (Freeland et al. 1981:406–407). More recent work has seen channels on the
New Jersey shelf that are identifiable, but affected by erosion (Goff et al. 2005:288). It is
important to note that the configuration of shelf valleys observed today represent the product of
estuary mouth scour as transgression moved the river mouths landward, rather than the
morphology of the valley prior to transgression (Freeland et al. 1981:422).
Because the OCS off New Jersey is a sediment starved environment, the seafloor bedforms
and shallow subsurface sediments preserve the effects of Pleistocene–Holocene regression and
transgression, although sand ridges and swales oriented northeast–southwest have subsequently
been formed through erosion due to bottom flow and tidal currents. Up to 10 m of sediment have
been removed from the transgressive sand sheet in some areas, and ridges in the OCS have been
winnowed, with smaller grains removed to leave consolidated sand ridges that have become
resistant to further erosion (Duncan et al. 2000; Goff et al. 1999; Goff et al. 2005:291).
Much of the New York Bight has been mapped using side scan radar (e.g., Lathrop et al.
2006; Schwab et al. 2000) and the Hudson Shelf Valley has been examined in detail (e.g.,
Butman et al. 2003). Some aspects of the Hudson Shelf Valley morphology reflect events in the
Late Pleistocene with far-reaching implications for the regional landscape and potential for site
preservation. The Hudson Shelf Valley has rectilinear sides and is oriented north-south at its
head, then northwest-southeast for its remaining length. The morphology of the valley
(particularly near its southeastern outlet) reveals a braided channel or anastomosing channels and
83
fluvial bars. Part of the valley floor resembles a straight channel with alternating bars and
sinuous thalweg and straight channel banks. At the mouth of the Hudson Shelf Valley is a deltalike feature whose surface is marked by a southwest trending distributary system draining in the
direction of Hudson Apron on the OCS southwest of Hudson Canyon (Bloom 1998; Uchupi et al.
2001:119).The Hudson Apron is one of a number of lobes on the OCS at the outlets of
paleochannels in the New York Bight as well as to the east off the coast of New England. The
lobate shelf morphology along with the wavy terrain near the mouth of the Hudson Shelf Valley,
the presence of coarse clasts on the mid-shelf and abundance of land mammal remains associated
with them, have been interpreted as evidence of episodic massive flooding of the shelf by the
draining of glacial lakes formed behind the Wisconsin terminal moraine and subsequent
flashfloods after the lakes drained. Two major episodes of catastrophic draining occurred, one at
17,000–15,000 years ago creating the features in the shelf east of the Hudson Shelf Valley and
the Hudson Apron, and another at 14,000–12,000 years ago creating the sediment lobes,
including the mid-shelf wedge, on either side of the Hudson Shelf Valley (Uchupi et al.
2001:126–127).
After ca. 20,000 B.P. when the LGM reached its southernmost extent, the ice began melting
back as the climate shifted into a cycle of global warming. Meltwater from the shrinking ice
sheets was funneled into lakes and rivers that ultimately returned it to the world’s ocean basins.
Thus the maximum exposure of the continental shelf in the New York Bight was at ca. 20,000
B.P. and the Atlantic Ocean began reclaiming that land thereafter.
Runoff from the melting ice was trapped behind the region’s terminal and recessional
moraines creating an extensive geography of proglacial lakes in eastern New Jersey, New York,
New England, and adjoining areas of Canada as the ice retreated (Figure 5.2). Seepage through
the moraine allowed the formation of small channels through what was then the exposed shelf—
channels that have been documented in mapping of the ocean floor. But more striking are aprons
of sediment and rocks forming large wedges at the mouths of paleochannels, which derive from
catastrophic floods, which took place over a 5,000 year period ending by 11,000 B.P. (Uchupi et
al. 2001). Major flooding events are hypothesized as each glacial lake was drained when erosion
weakened points along the moraine, glaciers calved off large icebergs that suddenly raised lake
levels and created high-energy waves, and/or tectonic uplift to the north and subsidence to the
south from ice unloading stressed the landform holding back the water. The sudden release of
large volumes of water, debris, and sediment is responsible for moving rocks and sediment
hundreds of miles across what was then the exposed shelf and creating episodes of extreme
turbidity at the river mouths, such that large rocks were transported well off shore into what are
now deep sea contexts. Additional splays along the paleochannels document subsequent flash
floods that originated in the drained lake beds (Uchupi et al. 2001:139–140).
A number of glacial lakes existed in the region at points in the terminal Pleistocene (see
Figure 5.2). In some cases, their geologic histories are connected. For example, Lake Passaic,
which extended across the terminal moraine, was blocked at its southwest end by till that filled
gaps at Moggy Hollow and Short Hills. As the ice retreated northward, the lake drained via the
Short Hills gap in the Watchung Mountains and the Little Falls-Paterson outlet into the region of
the future Lake Hackensack. Lake Hackensack breached its dam at Raritan Bay as a result of
rapid isostatic uplift and drained onto the shelf. Lake Hudson occupied the valley south of the
84
Figure 5.2.
Glacial lakes in the region, ca. 18,000–12,000 B.P.
85
Hudson Highlands and was dammed by the terminal moraine at the Narrows. After Lake
Hackensack drained, Hudson Lake overflowed onto the empty Lake Hackensack basin, eroded
the exposed Hackensack lake clays and drained onto the shelf via the gap previously eroded
through the terminal moraine in Raritan Bay. It was probably this drainage and the drainage of
Lake Hackensack that were responsible for the erosion of the north-south trending ancestral
Hudson River channel defined by the Reflector R unconformity on the New Jersey shelf (Dineen
et al. 1988; Newman et al. 1969; Peet 1904; Reeds 1933; Uchupi et al. 2001:134–135). Varve
counts suggest that Lake Hudson existed for approximately 3,000 years before it drained (Uchupi
et al. 2001:134). The largest catastrophic flood events occurred as a result of the ice retreat north
of the Adirondack Mountains, precipitating the draining of Lake Iroquois (in the Ontario basin)
and Lakes Vermont and Albany through the Hudson Valley (Uchupi et al. 2001). After the
draining of Lake Hudson, ca. 12,000 B.P., saltwater species are seen in dated deposits, reflecting
the establishment of estuary settings far inland (Weiss 1974; Weiss et al. 1976).
On the east side of the New York Bight, a number of features on the shelf derive from the
“Sound River,” which drained Late Wisconsin glacial meltwaters impounded in three lakes:
Lake Flushing on the west end of Long Island Sound, Lake Connecticut which was separated
from Lake Flushing by a north-south moraine, and Lake Block Island Sound. The Sound River
created a series of deltas where it breached the moraine between Long Island and Block Island,
and it ceased to flow once its channel was filled with morainal deposits (Frankel and Thomas
1966; Grim et al. 1970; Uchupi et al. 2001:128–134).
Before the draining of the proglacial lakes for a period several thousands of years, the shelf
was sediment starved. It was a vast plain swept by strong winds, dissected by a few peripheral
streams fed by waters draining through porous sections of the terminal moraine. It was only with
the catastrophic drainage of meltwater lakes and subsequent erosion of the former lakebeds by
flash floods that sediments in any significant amounts reached what is now the continental shelf.
Initially the glacial outbursts probably exceeded the capacity of what few valleys were present on
the shelf and the waters spread out over the shelf’s surface. With time, however, as the flows
dissipated they became channeled and incised the sediments of the Hudson Shelf and Block
Island valleys. These channels then served as passageways for later flows. Wetlands developed
on what had been a stark landscape. The now submerged shelf bears the imprint of these
catastrophic late Pleistocene events. Only the morphology of the inner shelf south of central
Long Island and off New Jersey appear to be have been formed by nearshore processes
associated with the last transgression (Uchupi et al. 2001:117).
In the case of the Hudson River, a series of more southerly channels once existed on the
continental shelf off New Jersey (Carey et al. 2005). The migration of the Hudson channel is
attributed to the glacier’s advancing peripheral bulge: as the ice sheet depressed the crust
immediately in front of it and created a peripheral bulge to the south, the reduction in gradient of
the Hudson may have allowed it to be captured by a smaller drainage system located within the
foredeep. The new channel, the Hudson Shelf Valley, was further incised by the drainage of
various glacial lakes by the Raritan and Hudson rivers from 17,000 to 14,000 years ago which
deepened the valley, which was subsequently reworked by marine processes during
86
transgression, becoming the modern Hudson Shelf Valley (Carey et al. 2005:169; Uchupi et al.
2001).
The Hudson River has been the principal source of sediment for the New Jersey portion of
the OCS throughout the Pleistocene (Carey et al. 2005:157; Poag and Sevon 1989). Little
sedimentation of the New Jersey portion of the OCS has taken place during the Holocene, as
sediments are trapped in lagoons and estuaries (Carey et al. 2005:158; Clarke et al. 1983; Swift
et al. 1972b). The low rates of sedimentation of the New York Bight during the Holocene have
allowed Late Pleistocene features on the shelf to remain visible. The visibility of these features
also suggests that much of the surface of the OCS was little modified by the late Pleistocene–
early Holocene marine transgression, possibly because a rapid rise in sea level allowed for
preservation of relict features (Uchupi et al. 2001:139). Such features have, however, been
reworked by bottom flow, tidal currents, and other processes (Goff et al. 2005).
The sediment starved continental shelf extends approximately 85 miles to the shelf edge, at
about the 150 m isobath, with a 0.068 gradient. The Hudson Apron, a seaward bulge marked by
later iceberg scours, dominates the modern New Jersey shelf edge along with the heads of
numerous submarine canyons (Ewing et al. 1963; Goff et al. 1999:322, 334; Pratson et al. 1994;
Pratson and Haxby 1996). In general, storm-generated, southwestward-directed currents
characterize the modern oceanographic regime on the middle and outer shelf (Butman et al.
1979; Duncan et al. 2000:398; Hopkins and Dieterle 1987).
Three seaward-stepping terraces, each bounded by a 10–15 m high scarp, have long been
interpreted as relict Quaternary stillstand shores (Dillon and Oldale 1978; Emery and Uchupi
1972; Swift et al. 1980; Veatch and Smith 1939). The seaward edge of the mid-shelf wedge
forms a prominent topographic feature on the New Jersey shelf: the Mid-Shelf Scarp or “shore,”
sometimes referred to as the Tiger Scarp or Fortune “shore” (Dillon and Oldale 1978; Knebel et
al. 1979; Milliman et al. 1990; Swift et al. 1980). Thought for decades to represent a fossil
shoreface associated with a sea-level stillstand, high-resolution seismic data has since shown it to
be depositional in origin (Duncan et al. 2000; Knebel et al. 1979; Milliman et al. 1990). Uchupi
et al. (2001) have speculated that the mid-shelf wedge could be a subaerial deposit associated
with massive outflows from breached glacial lakes along the Hudson Valley between
approximately 19,000 and 12,000 years ago during glacial retreat. However, it has been found
that the sediments forming the mid-shelf wedge date to 9,000–10,000 B.P. (based on eustatic
curves) and derive from a transgressive marine environment after the lake collapses (Goff et al.
2005:284). The outer shelf seaward of the Franklin Shore contains deltaic sediments at the mouth
of the Hudson River (Goff et al. 1999:322, 334).
5.2. RELATIVE SEA LEVEL CHANGES
Since the retreat of the late Pleistocene glaciers after approximately 20,000 B.P., the New
York and New Jersey coastline has been progressively inundated. Although sea level continues
to rise today, most shorelines attained their approximate modern positions by 1000 B.P.
(Pirazzoli 1996:102).
There has been a significant amount of work researching sea level rise for the New York
Bight. Three sea level curves for New York Harbor (summarized in Pirazzoli 1991:192–193)
87
suggest that water levels were approximately 28 m lower 10,000 years ago, 22 m lower 8,000
years ago, from 12–17 m lower 6,000 years ago, 8 m lower 4,000 years ago, and 3 m lower 2,000
years ago. More recently, Donnelly (1998), Stanley et al. (2004), Miller at al. (2009), and Wright
et al. (2009) have examined various datasets to refine a sea level curve for the region. Miller et
al. (2009) indicate that water levels may have been shallower at 10,000 B.P., suggesting a depth
of 18 m lower than today, with a depth of 13 m at 6,000 B.P. and 10 m at 4,000 B.P. Wright et
al. (2009), who focus on Late Pleistocene sea level rise, recalculated dates from Dillon and
Oldale (1978) and Duncan et al. (2000) to conclude that sea level was 120 m below present at
21,000 B.P, and 78 m lower at 14,400 B.P.
Local sea level research typically derives from core samples near shore, and may have little
relevance in reconstructing sea level rise during the late Pleistocene and early Holocene.
Recently published research by McHugh et al. (2010), however, has helped to fill in this gap for
the region. They examined 28 vibracores collected at 38–145 m depths on the New York–New
Jersey continental shelf and radiocarbon dated the shells of mollusks that characteristically lived
near the paleoshoreline and intertidal settings to document the timing and position of the
paleoshoreline as eustatic rise progressed across the shelf. Their research, which takes into
account the effects of glacio-isostatic forces, concludes that sea levels were at 120 m below
present during the LGM, confirming previous studies such as Dillon and Oldale (1978).
Currently, it appears that Wright et al.’s (2009) assessment of sea level rise for the region is
the most up-to-date and comprehensive, at least in terms of the Late Pleistocene and Early
Holocene (encompassing the LGM and Paleoindian period), and has been used most recently by
other researchers (e.g., McHugh et al. 2010; Nordfjord et al. 2009). For these earlier points in
time, it is the sea level curve adopted in this report. For the more recent Holocene, Miller et al.’s
(2009) research is used. In both cases, the sea level curves developed have taken into account the
effects of eustasy as well as glacial isostacy (Miller et al. 2009:16; Wright et al. 2009:96).
5.3. MARINE TRANSGRESSION AND SITE PRESERVATION
Recent research in the New York Bight has provided a better understanding of the process of
marine transgression. Examining data from high-resolution subbottom profiling and vibracores,
McHugh et al. (2010) have observed that the advancing paleoshoreline reworked barrier and
lagoon sediments during the period 15,000–11,000 B.P. Only a relic morphology of these
features remains on the sea floor of the OCS from 120 to 60 m of present water depth, which
they correlate with this time frame (McHugh et al. 2010:45), due to shoreface erosion and
deposition of reworked sediments both seaward and as part of the transgressive sand sheets
(Swift 1975; Swift et al. 1973). The remaining topography is characterized by ridges and swales.
They also document evidence for a slowstand occurring between 12,000–11,000 B.P., which also
disfavors preservation of what likely would have been the earliest potential prehistoric settlement
period of the subaerial OCS (McHugh et al. 2010:44). Their research also documents
dramatically decreasing sedimentation rates from ca. 11,000 B.P. to the present along the
continental margin, with little sediment having been deposited on the outer shelf during the
Holocene eustatic rise (McHugh et al. 2010:45).
The implications of these findings are that the seafloor in the region generally would have
been exposed directly to the forces associated with transgression during most of the Holocene.
88
These processes likely have erased most geologic evidence of subaerial exposure of the seafloor
developed during the last recessional sea-level cycle (Nordfjord et al. 2009:235), and the seafloor
in this area is constantly being eroded by bottom currents (Goff et al. 2005), although the sand
swales at depths greater than 50 m are considered to be generally inactive, subject to only some
localized erosion (Goff et al. 1999). The geomorphology of the New York Bight holds out little
potential for intact archaeological deposits, except in unique settings.
There are some localized exceptions to the model of an eroded seafloor across much of the
New York Bight. Sanders and Kumar (1975a) argue that evidence from the Long Island Shelf
points to in-place drowning around 7500 B.P. of a previous barrier approximately 7 km offshore
from the current position of Fire Island. They observe sediment sequences and paleontological
specimens from cores that reflect the presence of a lagoon through at least 4390 B.P. They
interpret these findings as evidence that the barrier remained in place from the point when sea
level was -24 m until it reached the top of the dunes at -16 m. During that time, the barrier
island’s dunes, thought to be 8–10 m tall like the contemporary Fire Island dunes, remained in
place, while the lagoon on the landward side widened and deepened, effectively delaying the
effects of sea level rise on the barrier. When sea level finally jumped the barrier, a new barrier
was formed in the surf zone 2 km off the modern shoreline position. The barrier that formed 2
km from the current Fire Island subsequently migrated landward as sea level continued to rise,
ultimately forming the contemporary barrier island. Previous views on transgression posited
inevitable migration of barriers when sand supply could no longer keep pace with sea level rise;
Sanders and Kumar (1975a:72), however, argue that the relationship between the rate of sea level
rise and the rate of sediment supply determines whether the barrier retreats or is drowned in
place. Bottom topography off the coast of Rhode Island suggests that barriers were drowned in
place there as well (Dillon 1970; Garrison and McMaster 1966). Although the high ground of the
submerged barrier island off Fire Island has been smoothed down with sediments deposited in
the lagoon and other low spots both seaward and landward, some relict landforms may be
preserved, in some cases buried beneath those very dune deposits that were displaced when sea
level jumped the barrier island. The lee side of the dunes at a certain elevation may have been
spared the effects of erosion, and instead could have been buried with a protective layer of
sediment from the destruction of the dunes during transgression. Some of that sediment may
have been winnowed away over time by bottom currents, storm events, and other forces, but
artifacts could remain intact in roughly the locations where they were deposited.
Looking beyond localized examples of how transgression affected the OCS in the New York
Bight, Nordfjord et al. (2009:241–242) have developed a model of the impacts of marine
transgression for the region overall, tying it to the sea level curve developed by Wright et al.
(2009). Their model is summarized as follows. During the LGM around 18,000 B.P., dendritic
patterns of drainages incised the exposed shelf (Davies et al. 1992; Duncan et al. 2000;
Nordfjord et al. 2005). During late lowstand/early transgression, shoreface retreat began and the
base of the shoreface migrated landward (Swift 1968). Given their lower elevation, the drainages
were flooded first, with the channels and valley floors scoured by wave action, creating the
ravinement surface. No archaeological sites along these drainages would have survived
transgression except where they had already been buried in deep sediments. As transgression
proceeded, erosion of nearby portions of the exposed shelf resulted in sedimentation of the now
submerged drainage channels (Duncan et al. 2000; Goff et al. 2005; Nordfjord et al. 2005, 2006;
89
Nummedal and Swift 1987). When sea level had risen to around −70 m and all known fluvial
networks on the mid- and outer shelf had been converted into estuaries, back-barrier tidal
incisions were preserved either due to (1) pauses in shoreface retreat, which allowed formation
of a substantive estuarine/back-barrier system that could not be completely destroyed by
subsequent transgressive ravinement; or (2) antecedent topography, for example preexisting
depressions that allowed the shoreline to retreat rapidly once the barrier was breached so that the
back-barrier morphology was preserved (Nordfjord et al. 2006). Due to remnant physiography of
the latest Pleistocene part of the mid-shelf wedge, landward shoreface retreat slowed down
during formation of the mid-shelf scarp around 11,400 B.P., when sea level was at approximately
−50 m. This prolonged period of shoreface effects may have focused ravinement erosion along
the seaward edge of the mid-shelf wedge, which in turn steepened its slope where that slope is
coincident with the mid-shelf scarp. Ravinement of the Pleistocene mid-shelf wedge forms the
foundation for the mid-shelf scarp. When sea level finally rose across the mid-shelf scarp, the
Holocene part of the transgressive, onlapping mid-shelf wedge was constructed from coarsegrained sediment supplied by the Hudson River and/or eroded, winnowed shoreface sediments
deposited in the near-shore zone. Powerful storm flows subsequently swept southward across
this depositional lobe, which removed sediments from parts of the mid-shelf wedge and
truncated its base (Swift and Freeland 1978). As the water column continued to deepen and the
shoreline continued to recede, shelf sand ridges evolved (Swift and Thorne, 1991; Goff et al.
1999; Snedden et al. 1999). Successive erosion of outer shelf sediments and modification of the
seafloor has continued to take place through bottom current-driven erosion, perhaps undercutting
surface armored seabeds (Goff et al. 2005) or differentially eroding preexisting topographic
lows. The transgressive ravinement surface dominates New Jersey seafloor bathymetry today,
but its morphology is complex and includes ridge and swale features which have been modified
by deeper water erosion (Goff et al. 2005).
Nordfjord et al.’s (2009) scenario for the process and effects of marine transgression in the
New York Bight has implications for the potential for archaeological site preservation. The
process of transgression has worked to scour away landforms and any archaeological deposits
they may have contained, but in some cases, it preserved particular features of the subaerial
landscape. From cases of stepwise retreat such as that documented off the south shore of Long
Island (Sanders and Kumar 1975a), and cases where existing topography allowed rapid filling
and possible sedimentation of landforms that may have supported human habitation, it is clear
that archaeological sites could be preserved in particular settings. Identifying such
microlandforms on a shelf that has been subjected to bottom currents and other forces for
thousands of years requires detailed coring studies.
5.4. ARCHAEOLOGICAL SENSITIVITY AND PRESERVATION POTENTIAL
As presented elsewhere in this study, the OCS can be divided into three general categories of site
potential (Figure 5.3). One category is No Sensitivity, applied to areas 120 m and deeper that
were subaerial prior to LGM, and are not expected to have supported human habitation. The next
category is Low Sensitivity, mapped between the 120 and 70 m isobaths based on Wright et al.’s
(2009) sea level curve, representing the exposed coastline from the LGM through the beginning
of the Paleoindian period. The remaining High Sensitivity category represents areas exposed
90
during the Paleoindian and later periods, from -70 m to more shallow areas. Within areas defined
91
92
Figure 4.1.
Southern New England–Georges Bank study region.
as Low or High Sensitivity, specific landforms will have potential for intact sites, while other
areas can be ruled out due to the effects of transgression.
During the last three to five thousand years of the prehistoric era (and possibly earlier), the
mouths of estuaries (including the Hudson River) were particularly attractive to hunter-gathererfishers, and many of the larger sites dating to the Late Holocene have been identified in these
settings. Likewise, well-drained landforms overlooking the streams that emptied into the
estuaries would also have been resource-rich, prime site settings. The question is whether such
site settings on the OCS would have survived transgression such that any archaeological remains
once present would remain intact.
Significant paleochannels of major streams which are now delineated as bathymetric features
in sonar datasets and digital elevation models include the drowned Hudson River, the Raritan
River (roughly west-east through New York Harbor into the New York Bight), a major channel
running south across the Long Island Platform (originating between the east end of Long Island
and Block Island), and numerous smaller streams. Among the landforms adjacent to the
paleochannels with higher sensitivity are terraces and places where smaller streams join the
larger channels. Similarly, irregularities along open paleoshorelines may have been more
attractive to prehistoric hunter-gatherer groups than straight and broad coasts, because such
irregular landforms could have offered better protection from the elements as well as potentially
more dense and/or diverse resources such as wetland fauna and flora adjacent to fishing grounds
(Perlman 1980). Such irregular coast features include narrow inlets, headlands, and stream
mouths (Benjamin 2010; Fischer 1995), along with protected areas like backbays and lagoons.
Unfortunately, it is difficult to identify particular small-scale landforms, some of which may be
obscured by later sediments.
Based on Nordfjord et al.’s (2009) model, only specific site settings might have been
preserved in the region. They include rapidly submerged features protected from subsequent
erosion by initial sedimentation. Candidate landforms include portions of drainage valleys that
were covered in sediment prior to transgression, the margins of sheltered tidal estuaries and
coastal ponds far enough landward to avoid the brunt of coastal shoreface erosion, and the
protected side of barrier islands that were inundated through stepwise shoreline retreat. Whereas
research in Nantucket Sound has produced evidence of an intact paleosol (see Chapter 4 above),
no such evidence has been found in the New York Bight region. However, a coring study for
dredged material placement in the New York Bight concluded “that the potential for submerged
archaeological sites is actually greater than previously recognized” based on samples that yielded
sediments indicative of a paleoshoreline environment (LaPorta et al. 1999:26). However, the
location of the paleoshoreline deposits on the continental shelf was determined to have been
susceptible to erosion during sea level rise, such that the chance of encountering an intact
prehistoric archaeological site was considered moderate rather than high (LaPorta et al. 1999:29).
Some localized settings would have been less affected by erosion, and should be prioritized
for study. For example, the flood deposits along the paleochannels such as the Hudson Shelf
Valley may very well seal archaeological deposits along river valleys dating after the period of
glacial lake draining (see Uchupi et al. 2001). The deltas created at the mouths of the
paleochannels from catastrophic flood events are probably not likely locations to find
93
archaeological remains for a number of reasons, however. The low-lying areas at a river mouth
would not have been a likely location for human settlement due to tidal inundation and restricted
access to fresh water. Even if archaeological remains were present where the delta was
subsequently formed, the high-energy flooding and turbidity from the water’s entry into the
ocean would likely displace artifacts, and compromise any site integrity. Furthermore, the large
boulders and rocks deposited during the catastrophic flooding would make identification and
investigation of any archaeological site exceedingly difficult. However, locations upstream along
paleochannels may have supported human habitation in what would have been floodplains and
terraces overlooking riverine and estuary settings. Such sites would likely only be preserved after
the catastrophic meltwater lake floods, which would have impacted any earlier sites present in
their paths. These settings could very well be preserved beneath sediment accumulated in later
flash floods, from estuarine muds deposited as marine transgression advanced upstream,
gradually inundating the valley, and from subaerial deposition governed by currents
perpendicular to the channel, which became a sediment trap (Freeland et al. 1981).
Because the seafloor has not been studied and mapped in sufficient detail to locate all the
specific landforms that existed prior to transgression, it is not possible to precisely delineate
potential site settings within high preservation potential. However, geophysical studies carried
out as part of an applicant’s feasibility and planning studies for proposed undertakings in the
OCS could support a more refined characterization of geomorphology within a lease block, and
suggest areas to target for archaeological survey. Certainly areas along relict stream channels,
estuaries, rock outcrops, and the back sides of drowned barrier islands should receive attention,
as should any other areas where paleosols may be identified. In the New York Bight, low
sedimentation rates through the Holocene have preserved a submerged landscape in which
drainage patterns have been mapped and dated, and subjected to only minimal reworking due to
bottom flow and tidal currents (Goff et al. 2005). Thus, landscape features are relatively
straightforward to identify, and if archaeological deposits were protected from transgression by
prior sedimentation or landform configurations that diverted direct shoreface impacts, they
should be recoverable from the seafloor.
94
PAGE LEFT INTENTIONALLY BLANK
95
6.
MIDDLE ATLANTIC
This chapter examines the portion of the Atlantic OCS from Delaware to Myrtle Beach,
South Carolina (Figure 6.1). This region is referred to as the Middle Atlantic, and falls mainly
within the southern portion of the Middle Atlantic Bight (which runs from Cape Cod,
Massachusetts to Cape Hatteras, North Carolina), although it includes the very northern portion
of the South Atlantic Bight (between Cape Hatteras and the Myrtle Beach, South Carolina).
Prominent geologic or geomorphic features within this region include the Delaware River,
Chesapeake Bay, Albemarle Embayment, and Cape Fear Arch.
6.1. REGIONAL GEOLOGY
In the Middle Atlantic study area, the OCS is generally characterized as sloping gradually
east to the continental slope. As a consequence of the gradual OCS slope in the study area, rivers,
sounds, and bays characterized the environment during much of the Holocene (Browder and
McNinch 2006). In conjunction with Holocene sea level rise, sediments associated with those
terrestrial features were environmentally resorted, as have the many barrier islands that once
existed off earlier coastlines (Heron et al. 1984). In the most general terms, the Middle Atlantic
OCS has been described as a “broad sand plain, characterized by a subdued ridge and swale
topography” (Shephard 1963). As the shoreline migrated westward, the wave dominated
environment of the region produced a relative equilibrium consisting of subbottom relict sounds,
bays, and channels overlain by varying depths of Holocene sand (Riggs et al. 1996; Swift et al.
1972b).
According to Belknap and Kraft (1985:238), coastal Delaware is located on a gently sloping,
low-relief coastal plain–continental shelf province that begins inland at the Fall Line and extends
seaward to the southeast to the continental shelf break, and consists of a wedge of Mesozoic and
Cenozoic sediments that stretches from New York to Virginia (Belknap and Kraft 1985:238).
Delaware is on its northwest flank in an arcing trend known as the Salisbury Embayment, which
includes parts of Virginia, Maryland, Delaware, and southern New Jersey and is bordered on the
north by the South New Jersey arch and on the south by the Norfolk arch (Ward and Powars
2004:263). A major offshore feature of this portion of the region includes the ancestral Delaware
River and its tributaries (Belknap and Kraft 1985; Swift 1973; Swift et al. 1972a; Swift and Sears
1974; Twitchell et al. 1977).
Off the coast of Maryland, researchers have used seismic profiles and vibracores to identify
likely portions of the ancestral St. Martin River and tributary system (Toscano et al. 1989; Wells
1994), and four significant paleochannels also have been documented offshore of the Delmarva
Peninsula. Those channels may have been associated with the relict channels of the
Susquehanna, Potomac, Rappahannock, and York rivers, according to a revised model for the
geologic evolution of the southern Delmarva Peninsula (Niedoroda et al. 1984). Near the edge of
the OCS, bathymetry suggestive of a broad coastal zone similar to the current coastline of the
southern Delmarva Peninsula has been identified.
96
Significant ridge and swale topographic characteristics have been identified south of Cape Henry
in the False Cape vicinity of Virginia (Swift et al. 1972a). Additional concentrations of
97
98
Figure 4.2.
Archaeological Sensitivity map for the Southern New England–Georges Bank
ridges and swales are associated with each of the North Carolina capes (Heron et al. 1984; Hunt
et al. 1977; Matteucci and Hine 1987). Bunn and McGregor (1980) have identified a regional
variation in morphology to the south and offshore of the Albemarle region of North Carolina.
There, a smooth, gently dipping, mid-slope morphology is flanked both seaward and landward
by steep dissected scarps. On the seaward scarp, associated valleys have cut into the smooth midslope region. Well-stratified sediments 500 m thick suggest an association with a significant
terrestrial drainage system, possibly associated with paleochannels of the James River (Bunn and
McGregor 1980). Additional investigation carried out in the Atlantic between Oregon Inlet and
Duck, North Carolina, identified a complex series of relict channels potentially associated with
the ancestral Roanoke/Albemarle River and likely formed during the Holocene post-glacial
transgression (Boss et al. 2002).
Exceptions to that ridge and swale model can be found to the south in Onslow Bay and south
of Cape Fear in Long Bay. Both bays are characterized as high-energy, sediment-starved shelves
with extensive exposed hard bottoms. Those hard bottoms consist of outcropping Tertiary and
Pleistocene-age stratigraphy (Gayes et al. 2002; Riggs et al. 1996). Research in Onslow Bay
indicates that the Holocene coastal lithosomes are virtually non-existent on the inner and middle
portions of that area of the OCS. Tertiary-age stratigraphy is exposed on much of the sea floor,
but a 1–3 m Pleistocene-age sequence uncomformably overlies those sequences on the inner
shelf offshore of Bogue Banks. Relict channels in the area, for the most part, represent streams
on the lower Coastal Plain that filled with fluvial and estuarine sediments during the midPleistocene. No evidence of Holocene-age barrier-related material was found within the channels
(Hine and Snyder 1985).
Near the extremity of the OCS, east-southeast of Cape Lookout and southeast of Cape Fear,
submerged terraces have been found (Matteucci and Hine 1987). Investigation of the OCS near
Cape Fear identified the remains of several large river channels and numerous smaller river
channels. The location and orientation of those relict channels suggest that the Cape Fear Terrace
represents the remains of a paleo shelf-edge delta. Although not as complex, the Cape Lookout
terrace could be a similar feature.
6.2. RELATIVE SEA LEVEL RISE
Early summaries of the Holocene marine transgression rates for the Middle Atlantic are
presented by Belknap and Kraft (1977), Edwards and Merrill (1977), and Edwards and Emery
(1977), with more recent work by Ramsey and Baxter (1996) and Nikitina et al. (2000) for the
Delaware area, Van de Plassche (1990) for Virginia, and Mallinson et al. (2005) and Horton et
al. (2009) for North Carolina.
While the more recent work employs additional samples and more refined dating methods,
there remains difficulty in assessing sea levels ca. 12,000 B.P. and earlier because of a lack of
data points. Given a lack of relative sea level data for the region at the LGM, it is assumed to
correspond to the eustatic curve for the Atlantic based on the Barbados data, placing sea level at
approximately 120 m lower at the LGM (Bard et al. 1990b; Fairbanks 1989). Assigning a
corresponding relative sea level for the period from the LGM to ca. 11,000 B.P., however, is
problematic for the region. The oldest index reported by Horton et al. (2009:1728) for North
Carolina, for example, is dated to approximately 10,800 B.P. and places relative sea level at 36
99
m below current sea level. Nikitina et al. (2000:Figure 4) have suggested that by 12,000 B.P. sea
levels were only 30 m below present. Unfortunately, there is a lack of index points for start of the
Paleoindian period (ca. 13,000–12,500 B.P.), when human occupation was certain in North
America.
Lowery (2009b) has suggested that isostatic depression of the former glacial forebulge likely
played a role in sea levels ca. 13,000 B.P. in the Middle Atlantic region, with the effects being
more prevalent in Delaware and tapering off to the south. This assumption is supported by
Reusser et al. (2004), who suggest that the observed rapid bedrock incision rates associated with
isostatic uplift during the LGM ceased along the Susquehanna and Potomac rivers ca. 14,000
B.P. By 14,000 B.P., the Laurentide forebulge that had previously impacted the unglaciated areas
south of the ice sheet’s terminus had fully collapsed, and the bulge seems to have been followed
by an isostatic trough or depression. Precisely how this depression impacted sea levels ca. 13,000
B.P. remains unclear, however.
To resolve the lack of data on sea levels ca. 13,000 B.P., the curves developed for New
Jersey are employed here. Wright et al. (2009), as developed by Nordfjord et al. (2009), indicate
that sea levels were approximately 70 m lower ca. 13,000 B.P. Sea levels likely would have been
more shallow due to isostatic depression ca. 13,000 B.P., and but until additional research is
conducted to develop additional data points, 70 m provides a conservative isobath to assume for
the beginning of the Paleoindian period for this region.
At the beginning of the Holocene, ca. 10,000 B.P., sea levels were approximately 30 m
below present (Horton et al. 2009; Mallinson et al. 2006; Nikitina et al. 2000). These studies also
put sea level at -15 to -18 m ca. 8000 B.P., and by 6,000 B.P. sea level had risen to around 10 m
below present.
6.3. MARINE TRANSGRESSION AND SITE PRESERVATION
Sea level data provide a window into understanding where to look for drowned
archaeological sites on the continental shelf. One can assume that coastal resources would be
abundant and humans would be attracted to shoreline areas during periods with slow sea level
rise rates. However, coastal archaeological sites would have a higher chance of erosion and
destruction during these same stable sea level periods. During periods of rapid sea level rise, the
productivity of coastal ecological systems would drop significantly and humans would tend to
favor subsistence areas outside of the Coastal Plain. Even so, landscapes with archaeological
sites would be rapidly inundated and preserved (Lowery and Martin 2009).
The highest rate of sea level rise during a period of known prehistoric occupation along the
Middle Atlantic is associated with MWP 1b, which best estimates currently place at 11,600–
11,100 B.P. (see discussion in Section 1.5 in Chapter 1 above). This period, which based on sea
level curves for the region corresponds to 55–42 m isobaths, experienced rapid sea level rise
averaging 200–300 cm per year (Lowery 2009b), and represents a time when intact
archaeological sites may have had a better chance of being inundated rapidly and preserved. This
period was followed by a much slower rate of sea level rise (approximately .8 cm per year) until
ca. 7000 B.P., after which the rate of sea level rise slowed even further (0.2 cm per year or less).
100
During this period, archaeological sites presumably would have a higher frequency of erosion or
destruction by the process of marine transgression.
Lowery (2001, 2003a, 2008, 2009b) has argued that areas along the Middle Atlantic that
demonstrate evidence of wide, or broad coastal zones during the prehistoric period may be
indicative of high site density, with increased possibility that sites may be preserved. A broad
coastal zone includes a coastal barrier island backed by a large shallow bay with many dendritic
tidal channels. The limited tidal marsh in these regions creates compartmentalized bays that act
as storm buffers. As such, broad coastal zones provide excellent shellfish habitat as well as
habitat for other coastal organisms. A narrow coastal zone, on the other hand, includes a coastal
beach with a confined back-barrier island area that includes extensive tidal marsh and many deep
tidal creeks. Narrow coastal zones provide limited shellfish habitat. Because of the close
proximity to the ocean, shellfish habitat areas would be greatly limited by frequent storm-related
overwash events.
Perhaps the most likely area for preserved prehistoric sites on the Middle Atlantic OCS lies
off the Delmarva Peninsula flanking the Norfolk and Washington canyons (Figure 6.2).
Bathymetry indicates that this area is an isobathic region that contains broad coastal zone
characteristics, where sites were likely to exist. The high probability area extends from south of
the Norfolk Canyon to the Virginia/Maryland border of the Middle Atlantic OCS study area. As
such, one can conclude that these possible site settings were situated along broad open-barrier
island lagoons approximately 13,000 B.P., where site burial was highly possible. Two specific
high probability locations have been isolated based on those criteria (see Figure 6.2).
In addition, the head of the Norfolk Canyon appears to be relatively unique in relation to the
other Middle Atlantic OCS canyons (the Washington Canyon and the Keller Canyon). At the
head of the Norfolk Canyon, there appears to be a large relict lagoon complex. Multi-beam data
from a NOAA survey carried out with funding from the NOAA Office of Ocean Exploration
generated the first high detail, multi-beam images of the head of the canyon and surrounding area
(Figure 6.3). That area also appears to be a high potential location for association with
Paleoindian habitation based on the nature of relict landforms identified in predictive models for
inundated Paleoindian site locations developed by Faught (2003), Faught and Latvis (2000),
Pearson et al. (1986), Lowery (2009b), and Stright (1990).
South of the Norfolk Canyon available bathymetry suggests that there are no areas similar to
those off the Delmarva Peninsula. The zone associated with possible prehistoric habitation is
narrow and not suggestive of an environment attractive to prehistoric populations. However, two
additional potential high probability sites for association with Paleoindian activity on the Middle
Atlantic OCS have been identified. The first is located southeast of Cape Lookout, North
Carolina, and is found in Figure 6.4. While not nearly so complex as the high probability area off
Virginia, bathymetry indicated that one or more islands may have existed in conjunction with a
moderate size lagoon complex that could have been attractive to Paleoindian inhabitants. A third
area southeast of Cape Fear may also merit consideration. Bathymetry in the area identified as
the Cape Fear Terrace is suggestive of an island complex (Figure 6.5). Although there is little to
suggest an association with a large lagoon, the feature represents a distinctive variation from the
paleolandscape extending north to Cape Lookout and south to the South Carolina border.
101
102
103
Figure 5.1.
New York–New Jersey study region.
104
Figure 6.3.
Head of the Norfolk Canyon imaged with multi-beam 2008 (NOAA Chart No.
12200 used as base map).
105
Figure 6.4.
High potential, terrace-associated area south-southeast of Cape Lookout, North
Carolina.
106
Figure 6.5.
High priority feature identified as the Cape Fear Terrace.
The remainder of the submerged coastline at depths associated with the Delmarva, Cape
Lookout, and Cape Fear priority areas appears to be too narrow and lack the features suggestive
of a broad coastal zone environment that would have been attractive to Paleoindian populations.
However, immediately inshore in the area rapidly inundated by MWP 1b, evidence of
Paleoindian habitation could survive. While rapid inundation might have reduced occupational
desirability of the area, pre-MWP 1b occupation sites could survive the more limited adverse
impacts associated with rapidly rising sea levels.
Inshore of the MPW 1b zone and offshore of the early Holocene sea level rise, an extensive
portion of the Middle Atlantic OCS consisting of approximately 4,700 square miles was exposed
and proportionally habitable for over 3,000 years as sea level slowly rose. Unfortunately, due to
107
the slow rise in sea level, that area was exposed to all of the destructive environmental conditions
that obscure, reduce, or destroy landforms and undermine archaeological site integrity. Offshore
of the Delmarva Peninsula, identified paleochannels reflect four iterations of the Susquehanna
River. The oldest, Exmore Channel, underlies the peninsula and extends into the OCS
approximately 30 miles north of Cape Charles. Bell Haven Channel, Eastville Channel, and Cape
Charles Channel, each progressively farther south and younger, underlie the peninsula and
appear on the OCS as well (Chen et al. 1995; Colman et al. 1990; Dame 1990; Hobbs 1997,
2004; Oertel and Foyle 1995). On the OCS these and other relict channels could have associated
prehistoric material. Although as previously discussed, channel features such as levees where
habitation might be expected could have been extensively resorted during inundation, Stright
suggests that this is not necessarily always the case (Stright 1990). Along the present foreshore,
those channels could have surviving Archaic material.
Subaerial surfaces between Cape Henry and Cape Hatteras have been extensively eroded and
redistributed during the shoreface retreat. The area also has been impacted by fluvial erosion
associated with the tributaries of the Susquehanna River and meltwater runoff (Moir 1979:44). In
the Sandbridge, Virginia area below Cape Henry, evidence of relict landforms also have been
identified. Virtually all of those consist of river and stream channels and estuary complexes
possibly associated with the paleo-James River or the paleo-Elizabeth River (Chen 1992; Dame
1990; Harrison et al. 1965; Hobbs 1990; Kimball and Dame 1989; Meisburger 1972; Swift
1975). Under certain conditions, associated archaeological evidence could survive in a relatively
undisturbed context.
Lagoonal deposits dating as early as 10,000 B.P. have been identified offshore of the North
Carolina Outer Banks, adjacent to Roanoke Island. Those deposits were associated with Platt
Shoal, a retreat massif on the north side of the ancestral Albemarle Valley (Moir 1979:44).
Nearshore evidence of paleochannels in the vicinity of Nags Head, Kitty Hawk and Duck, North
Carolina, and Sandbridge, Virginia, was identified by Browder and McNinch (2006). Farther
south, another series of paleochannels extending from Bogue Banks into the OCS have been
identified in Onslow Bay by Hine and Snyder (1985). In the vicinity of Cape Fear, relict
channels of the Cape Fear River cross the OCS in an area between Onslow Bay and Long Bay
that is not characterized by sediment starvation. Although untested, surviving landforms
associated with relict channels could contain prehistoric material.
It is difficult to reliably identify high priority upper shelf locations for OCS Paleoindian and
Archaic habitation, but the areas noted are worthy of consideration. Those areas could produce
hard data that can be used to test predictions that associate relict landforms and prehistoric sites.
In theory, relict landforms such as rivers, lakes, bays, and lagoons could be associated with
particular types of archaeological sites (Stright 1990). Those “terrestrial analogs” are based on
land-use patterns from archaeological evidence associated with terrestrial prehistoric sites
(Gardner 1980, 1982) and can be used to identify potential habitation sites on the OCS (Belknap
1983; Bonnichsen et al. 1987; Cockrell 1980; Dragoo 1976; Faught 2003; Faught and Latvis
2000; SAI 1981).
108
6.4. ARCHAEOLOGICAL SENSITIVITY AND PRESERVATION POTENTIAL
Based on the most current sea level curves for this region, archaeological sensitivity is
defined as follows, and is displayed in Figure 6.6:
•
•
•
No Sensitivity. Areas 120 m and greater in depth are considered to be areas with No
Sensitivity for prehistoric sites, since these areas were not subaerial during the LGM.
Low Sensitivity. Areas from the 120 m to 70 m isobaths. This designation covers
areas exposed between the time of the LGM and the earliest Clovis occupation, ca.
13,000 B.P.
High Sensitivity. High Sensitivity areas include all areas within the OCS that are
shallower than 70 m.
As discussed in Section 6.3 above, there are a number of locations within the High
Sensitivity areas that also have High Potential for containing sites. Such locations include areas
in the vicinity of paleochannels and relict terraces off of Cape Fear and Cape Lookout (see
Figures 6.2–6.5). Figure 6.7 illustrates the relative sea levels during the past 20,000 years,
including the high potential area associated with rapid sea level rise during MWP 1b, which
corresponds to the 55–42 m isobath range, and should be considered High Potential for site
preservation.
109
Fig
110
Figure 5.3.
Archaeological Sensitivity map for the New York–New Jersey study region.
111
Figure 6.1.
Middle Atlantic study region.
112
PAGE LEFT INTENTIONALLY BLANK
113
7.
GEORGIA BIGHT
The Georgia Bight is a sub-division of the South Atlantic Bight; for purposes of this chapter,
the Georgia Bight study region is defined as that portion of the South Atlantic Bight that lies
between Myrtle Beach, South Carolina, and Georgia–Florida border in the vicinity of St. Marys,
Georgia (Figure 7.1).
7.1. REGIONAL GEOLOGY
The Georgia Bight can be described as an embayment portion of a passive continental margin
characterized by a thin sedimentary layer overlying a Cenozoic geology made up of extensive yet
condensed sections that have resulted from paleo-oceanographic processes (Adesida 2000;
Baldwin et al. 2006; Garrison et al. 2008; Harris et al. 2005; Henry n.d.; Littman 2000; Weems
and Edwards 2001; Weems and Lewis 2002). This embayment is mildly up-warped on the north
by the Cape Fear Arch at latitude 33o 30'; it is down-warped south of this latitude until in the
south, at 30o, this margin rises again with the Florida carbonate platform. None of the change in
elevation for the Georgia Bight is dramatic, with uplifted coastal terraces rarely reaching 30 m.
Active fault systems are few, with the Charleston Fault being the most prominent. Weems and
Lewis (2002) describe a north-trending fault south of the Charleston Fault forming a discrete
hinge zone that accommodates structural movement between the Cape Fear Arch and the
Southeast Georgia embayment. Uplift, while minimal, does occur in the Georgia Bight.
The regional geologic framework for the Georgia Bight encompasses 18 stratigraphic units
from the Oligocene to the Pliocene in the coastal area of Georgia, and 11 Eocene through
Pliocene units off Charleston. Near the area of the Cape Fear Arch, Paleocene outcrops have
been identified (Weems and Edward 2001; Weems and Lewis 2002). The Georgia Bight
continental shelf is dominated by sediments, unconsolidated and consolidated, that are eroded
relicts of earlier subaerial coastal landforms characterized by dunes, wetlands, coastal rivers, and
forests much like today (see, e.g., research by Baldwin et al. 2006; Colquhoun et al. 1983; Foyle
and Henry 2004; Garrison et al. 2008; Gayes et al. 1992; Harris et al. 2005; Markewich et al.
1992; Riggs et al. 1992, 1996; Swift 1976; Swift et al. 1972b; Weems and Edwards 2001;
Weems and Lewis 2002). The proximal cause of the erosion of these coastal landforms is sea
level change, either the transgression or regression limbs of a highstand–lowstand sequence
(Johnson and Baldwin 2004:235–280). Indeed, the Georgia Bight as a geomorphic surface has
much in common with that of the Middle Atlantic Bight to its north. The Middle Atlantic Bight
sandsheet, and that of the South Atlantic Bight/Georgia Bight, is the result of sea level rise
(transgression) with landward retreat of coastal barrier and estuarine systems along with
sediment reworking by storm-, tidal- and wind-generated bottom currents (Garrison et al. in
press; Riggs et al. 1996; Swift 1976; Swift et al. 1972b).
The Georgia Bight, like most of the eastern U.S. continental shelf, is characterized as an
accommodation-dominated shelf, wherein flooding and ravinement during transgression is
dominant. While there is space or depth available for the influx of terrigenous sediments during
sea level highstands (i.e., sediments derived from erosion of the continents), the lack of large
rivers with large sediment loads precludes the buildup of sediments transported from elsewhere
in favor of the reworking of existing sediments. The Georgia Bight is a shelf dominated by
114
115
Figure 6.2.
High priority area off the Delmarva
erosional accommodation, with a sea floor in disequilibrium with erosional surfaces, irregular
grain size patterns, shoreface bypassing, and wave/wind and tidal current reworking.
Additionally, recent studies (e.g., Baldwin et al. 2006; Harris et al. 2005; Weems and Lewis
2002) have shown that this shelf is characterized by a stratigraphic heterogeneity. Indeed, in
many areas the Miocene and Pliocene units are scattered as erosional remnants. In other areas,
such as the mid-to-outer shelf, these units outcrop as ledges and erosional ramps. Riggs et al.
(1996) attribute this shelf morphology to subaerial weathering, stream erosion, and karst
formation. The latter is most notably observed on the Florida portion, which has a rugose shape
(Adams et al. 2010).
The Cenozoic aged geologic units that underlie the unconsolidated Quaternary age sediments
of the inner-to-mid shelf are well described in studies onshore along the Georgia and South
Carolina coasts. These formations span the post-Cretaceous time interval, with the most
described units being of Eocene through Quaternary age (54.5–2.588 million years ago), but with
Paleocene units (up to 65 million years ago) occurring on the northern portion at the Cape Fear
Arch. Figure 7.2 depicts those geologic units that are present to a greater or lesser degree along
the Georgia shoreline and offshore.
Nearshore, the seafloor exposures consist of Pliocene-age rocks that are members of the
Raysor Formation. Offshore, it replaces the Cypress Head Formation that pinches out or has
eroded out as one traces it onto the Georgia Bight shelf, where it has been eroded. Hard bottom
exposures of Pliocene age outcrops are common along the 20 m isobaths (Figure 7.3). The
amount of hard bottom has been estimated in the SEAMAP study (2001), and these data have
been incorporated here.
Much like Riggs et al.’s (1996) findings at Onslow Bay, North Carolina, within the Middle
Atlantic Bight, outcrops and hard bottoms are interspersed with large coextensive areas of sand.
Weaver (2002) recovered up to 3 m of unconsolidated sands in vibracores at Gray’s Reef, off the
coast of Georgia. These results are similar to those reported by Gayes and other researchers for
the Middle Atlantic Bight (Gayes et al. 1992). Sautter and her students at College of Charleston
have explored and recovered rock samples from the scarp areas of the MAB/Georgia Bight shelf
break (Stubbs et al. 2007). One of these samples has been examined recently and compared to
rock samples from mid-shelf locales. If one extrapolates the known lithostratigraphy for the
Georgia Bight to the shelf break area (e.g., Adesida 2000; Henry n.d.; Huddlestun 1988; Weems
and Edwards 2001; Woolsey 1977), those outcrops are Miocene in age.
In both hand and thin section, the rock exposures of the mid shelf and outer shelf are
lithologically calcarenites as a rule. There is variability, but the clastic nature of these rocks is
medium-to-coarse quartzitic sands cemented in a carbonate matrix of spar or micrite (Garrison et
al. 2008; Harding and Henry 1994; Hunt 1974) (Figure 7.4).
As Swift (1976; Swift et al. 1976b) initially postulated and later studies have confirmed
(Baldwin et al. 2006; Garrison et al. 2008; Harris et al. 2005; Henry n.d; Woolsey and Henry
1974), the Georgia Bight, like the Middle Atlantic Bight (Riggs et al. 1992, 1996), is incised
with Coastal Plain rivers. Their paleochannels are incised into Pliocene–Miocene strata (Foyle et
al. 2001; Foyle and Henry 2004; Henry n.d.; Woolsey and Henry 1974). One major effort in the
present study was to consolidate existing information, seismic and otherwise, to better define
116
Figure 7.2.
Generalized geological section for coastal Georgia from Oligocene through
Pleistocene (adapted from Weems and Edwards [2001] and published in Garrison
et al. [2008]).
these paleochannels. As shown in Figure 7.5, these efforts identified the Paleo-Altamaha, the
Paleo-Savannah, and the Paleo-Medway river courses off Georgia and those paleo-streams
detected for the Stono-Edisto and Pee Dee river systems in the north half of the Georgia Bight. It
is instructive to note the close correlation of these streams with those postulated by Swift (Swift
1976; Swift et al. 1972b) using bathymetric variation.
117
118
119
Figure 6.6.
Archaeological Sensitivity map for Middle Atlantic study region.
Figure 7.4.
Thin section (left) and hand section (right) of medium-to-coarse quartzitic sands
cemented in a carbonate matrix of spar or micrite from the OCS of the Georgia
Bight.
The predominant sediments in the Georgia Bight are those of Pleistocene/Holocene age.
These sediments extend onto the continental shelf to the mid shelf (-20 m). Outcrops, hard
bottoms, and paleo-valleys are primarily of Pliocene age until outer shelf/shelf break depths are
reached. Here, the strata are predominantly Miocene age. Northward along the shelf margin,
Miocene strata will outcrop shoreward onto the inner shelf because of the thinning and/or erosion
of Quaternary and Pliocene age strata. Figure 7.6 shows a north-to-south thinning-to-thickening
of Pliocene age sediments from a few meters thickness off Tybee Island to over 30 m off Sapelo
Island, based on seismic data. Foyle and Henry (2004) attribute this deepening of Pliocene
deposits in an elongate coast-parallel embayment. This embayment extends landward behind the
present day barrier system. Thinning of Quaternary age deposits near Savannah is attributed to
the presence of the Beaufort Arch, similar to the deposits near the Cape Fear Arch farther north.
Henry (n.d.), in Figure 7.6, has interpreted seismic data for the inner-to-mid shelf north of
Sapelo Island, Georgia, to Hilton Head Island, South Carolina. In this previously unpublished
figure, seismic facies are interpreted as to thickness and age. The deepest strata, south along the
strike, is the Eocene, at 50 m depth off St. Catherines island. The age for its lowest member is
54.5 million years ago (Huddlestun 1988). Henry includes upper Eocene member(s), which dates
to 37 million years ago. Only those outcrops at or near the shelf break are designated as
Miocene, 23.8–5.32 million years ago. Henry indicates middle Miocene units in light blue and
early Miocene strata in dark blue. He does not map late Miocene strata on the inner-to-mid shelf
below Sapelo Island’s latitude.
Weems and Edwards (2001) show that Quaternary/Pliocene strata below the Sapelo Island to
Cumberland Island portion of the Georgia Bight are over 40 m thick. Miocene age strata thicken
to over 60 m in cores. Not shown, but described in areas located at the shelf edge (-100 m)
beyond the Miocene age outcrops of the Savannah Scarp, are mud-bottom sediments deposited
by shelf bypass mechanisms. These were deposited during the LGM lowstand and subsequent
120
post-LGM transgression. The picture in the South Carolina portion is similar to that of the
121
122
Figure 6.7.
Sea levels for specified periods, including High Potential area corresponding to
MWP 1b i h Middl A l i
d
i
Figure 7.6.
Seismic data for the inner-to-mid shelf north of Sapelo Island, Georgia, to Hilton
Head Island, South Carolina (Henry n.d.).
Georgia section—southeast draining rivers have incised into these upper Cenozoic units creating
paleo-valleys that have subsequently backfilled during cyclic changes in sea level, with a host of
sediment types ranging from estuarine muds to clean shelly sands (Harris et al. 2005). In our site
models, a higher probability for intact prehistoric sites is postulated for these valleys and river
channels as opposed to the more readily eroded uplands and coastal plains.
Generally, the Georgia Bight is dominated by Cenozoic age strata, which represent basin
edge sediments eroded and transported from inland terrains by Piedmont and Blue Ridge
streams. However, when considering the implications of Georgia Bight geology for
archaeological site preservation, the potential for sites is directly related to the presence or
absence of more recent Quaternary age strata. Quaternary age strata are significantly eroded or
123
absent in the northern portion of the Georgia Bight as well as seaward of the Savannah River,
where University of Georgia scientists found shallow Quaternary sediments in an unconformable
relationship to Miocene strata during vibracoring in 2004 (Wendy Weaver, personal
communication, 2005). South of the Paleo-Savannah Valley, Quaternary strata thicken and
extend from beneath the coastal barrier system seaward to the middle shelf where, at Gray’s Reef
National Marine Sanctuary, they thin to a few meters thickness over Pliocene strata outcropping
along the 20 m iosbath. Northward toward the Cape Fear Arch these sediments thin rather than
thicken.
Erosion of Quaternary strata, principally the Satilla Formation off Georgia, has left
unconsolidated, time-transgressive sediment deposits that contain significant numbers of subfossil and fossil faunal remains of terrestrial mammals, notably bison, mammoth and horse.
Nearshore, such as Edisto Beach, South Carolina, extensive paleontological remains have been
found as well. Few readily identifiable prehistoric artifacts have been located in these sediments,
but those that have putatively date to the Archaic period (Garrison in preparation). The lithic
finds (2) are made from orthoquartzite, a rock type not found in local outcrops at this writing.
7.2
RELATIVE SEA LEVEL RISE
Paleoshoreline locations for the Georgia Bight are based on current relative sea level (RSL)
curves, notably those of Chappell (2002), Cutler et al. (2003), and Siddall et al. (2003, 2008). In
the Pleistocene/Quaternary, defined as 2.588 million years ago to 10,000 B.P., the Georgia Bight
experienced 10 glacio-eustatic events (Foyle and Henry 2004:73).
Relative sea level studies have benefited from extensive research in the late 20th–early 21st
centuries. The stimulus for the volume of studies has been paleoclimate research. Advances in
dating techniques that include AMS radiocarbon, OSL and U-Th methods have allowed more
accurate direct dating of sediments and sediment inclusions as well as corals. Coupled with more
refined ages for the Marine Isotope Stages, based primarily on parts per thousand (per mil)
variation in stable oxygen isotopes, well-dated proxy records have been developed for RSL for
the last glacial period (Bard et al. 1990a, 1990b, 1996; Basillie and Donoghue 2004; Blum et al.
2001; Gallup et al. 1994; Lambeck and Chappell 2001; Milliman et al. 1972; Otvos 2004;
Shackleton 1987; Sirocko 2003; Siddall et al. 2003, 2008). Perhaps most informative to this
region are the results of OSL studies of coastal deposits in North Carolina and east Florida
(Burdette et al. 2009; Mallinson et al. 2008), bracketing the Georgia Bight with well-constrained
ages for Quaternary RSL.
7.3
MARINE TRANSGRESSION AND SITE PRESERVATION
Underwater archaeological research in the Gulf of Mexico has targeted drowned river
systems and their fluvial landforms as locations for intact archaeological sites (e.g., (Dunbar
1988; Dunbar et al. 1989; Faught 2004; Faught et al. 1992; Faught and Donoghue 1997;
Gagliano et al. 1984; Pearson et al. 2008). Faught (2004) and Pearson et al. (2008) each found in
situ subaerial sediments on which the archaeological facies where located. However, unlike the
Gulf, preservation of surficial Quaternary sediments and soils on the continental shelf is made
difficult by the dynamic current regimes found there.
124
Significant portions of the mid-to-outer shelf of the Georgia Bight are covered with a
relatively thin (1–2 m) “veneer” of Quaternary aged sandy sediments (Harris et al. 2005;
Markewich et al. 1992; Pilkey et al. 1981). On the inner shelf, more evidence of cohesive mud
and rock has been found in the shallow subsurface (Pilkey et al. 1981; Riggs et al. 1996; Toscano
and York 1992), but much of this shelf is hardbottom. This general characterization can be made
for the Georgia Bight from Florida to the Cape Fear Arch at the juncture of South and North
Carolina. Indeed, one can extend the description to the Middle Atlantic Bight as well (Riggs et
al. 1996).
Paleoshorelines are geological markers of stillstands in sea level. While their broad scale use
for the Southeast (Cooke 1936, 1939) was called into question by Winker and Howard (1977),
they have continued to be used on a local and sub-regional scale (Donoghue and Tanner 1992;
Huddlestun 1988). In most instances these shorelines are raised marine terraces (Colquhuon and
Brooks 1986). In north Florida, Georgia, and South Carolina, up to seven Quaternary shoreface
systems are identified, although their identification as the same geologic unit, across these three
states, is not consistent. For instance, Adams et al. (2010) identify only three paleoshorelines in
their study of Pleistocene sea-level oscillation in northeastern Florida. In Georgia and South
Carolina the nomenclature and dating of these paleoshorelines is more consistent. Like drowned
streams and valleys, drowned shorelines can be locations of higher prehistoric site density,
although the potential for such sites to survive transgression is very limited (e.g., Driver 2004;
Stright et al. 1999).
Erosion, linked to sea level rise/fall—transgression/regression—in the Georgia Bight, has
been observed and quantified by Alexander, working at Skidaway Island and neighboring
Wassaw and Ossabaw Islands, where he has documented extensive coastal erosion of these
modern barriers (Alexander and Henry 2007; Robinson et al. 2010). Likewise, other researchers
writing of St. Catherines Island and neighboring Sapelo Island document extensive shoreline
erosion and aggradation associated with these coastal barriers of the Georgia Bight (Bishop,
Meyer et al. in press; Booth et al. 1999; Chowns et al. 2008; Rich and Booth in press). On
Skidaway Island, Ossabaw Island and St. Catherines Island, this erosion has led to the
documented loss of archaeological sites dating from the Late Archaic through Mississippian
periods (Alexander and Henry 2007; Thomas 2008).
Not all areas along the Georgia Bight are eroding, most notably the upper portion of the
Georgia coast. Alexander and Henry (2007) have dated aggrading west-to-east beach lines on
Ossabaw Island approximately 1,400–1,820 years ago (see also Robinson et al. 2010). Rich and
Booth (in press) have dated the Holocene aggradation of beach cheniers on the southern tip of St.
Catherines Island. Concomitant with that aggradation, the mid-portion of that same island is
experiencing significant erosion (Bishop, Rollings and Thomas in press; Thomas 2008). Barrier
islands along the Grand Stand (Myrtle Beach) and Folly/Kiawah Islands (Charleston-Beaufort)
show similar behavior.
Because the Georgia Bight is an active erosional geomorphic surface, particularly the Innerto-Mid shelf areas, one cannot expect significant preservation of prehistoric archaeological sites
at or near the modern seafloor. At two localities in the Georgia Bight, Garrison et al. (2008, in
press) report artifact and faunal remains that are time-transgressive, in that Holocene-aged sub125
Figure 7.7.
Profile of sediment core placed into a paleochannel in the Georgia Bight showing
estuarine/fluvial sediments below the transgressive sand sediments.
fossils are found directly in association with Late Pleistocene (ca. 40,000 B.P.) materials. In situ
mixing of seafloor sediments has been observed over a two-year study span.
Paleo-stream channels are not readily observed on the Georgia Bight except in seismic
records. Studies by Garrison et al. (2008) and others (Baldwin et al. 2006; Harris et al. 2005)
indicate that these paleo-streams are buried, albeit shallowly, under the reworked shelf
sediments. Sediment cores into one of these paleochannels show estuarine/fluvial sediments
below the transgressive sand sediments (Figure 7.7). Dating attempts on these more intact
sediments, which are more likely to preserve in situ archaeological materials, failed to obtain
ages that were prior to 27,000 B.P. (Garrison et al. 2008:Table 3). Stubbs et al. (2007) reported a
river channel on the Middle Shelf of the South Carolina portion of the South Atlantic Bight. Its
paleovalley walls showed some slight relief in contrast to that of the Paleo-Medway River, where
a sediment core was obtained in 1996 (Littman 2000). This paleostream, named the Transect
River, is part of the several paleoincisions of the paleo-North Edisto and Stono rivers that are
mapped off Folly and Kiawah islands (Harris et al. 2005).
Based on existing data from a battery of late 20th- and early 21st-century studies, a great deal
is known about sea level rise and fall and the mechanisms for its variation. Likewise, there exist
extensive data on the nature of sea floor erosion and its role in shaping the geomorphology of the
continental shelf. Modern geological models such as that of sequence stratigraphy provide a
robust conceptual framework for assessing the likelihood for the preservation of prehistoric
archaeological sites. However, there does not exist a similar wealth of data on the
presence/absence or distribution of prehistoric sites on the continental shelf of Georgia and South
126
Carolina. It is possible that the potential for preserved archaeological deposits is greater on the
nearshore or inner shelf of the Georgia Bight, where sediments are somewhat thicker than those
of the mid-shelf. However, assessing this potential will await direct archaeological evidence.
Paleochannels provide the best geological feature for investigating potentially preserved
submerged sites, since sites located adjacent to rivers have the best chance of having been buried
prior to inundation. In the absence of such pre-inundation burial, archaeological sites have little
to no potential to survive transgression. Paleochannels were plotted for this study using a
synthesis of existing data, including technical reports, unpublished data housed at the Georgia
Southern University’s Applied Coastal Research Laboratory, and personal communication with
researchers known to have unpublished data (e.g., Alexander and Robinson 2006; Foley 1981;
Foyle et al. 2001; Garrison n.d.; Hill n.d.; Henry et al. 1978; Henry and Idris 1992; Henry n.d.;
Kellam and Henry 1986; Swift et al. 1972b). These sources were used to identify and map
surface and subsurface expressions of Quaternary paleochannels on/in the continental shelf of
study region. Maps and report figures were scanned and georeferenced, and existing shapefiles
were obtained to build the database. Shapefiles in this project (provided under separate cover)
are named according to the data source used for each layer’s construction and include: research
cruise tracklines, sub-surface areas of interest (AOI), and crossings of buried paleochannels and
proposed paleochannel pathways.
Paleochannels in the Georgia Bight are known with some certainty, having been derived
from recent geophysical data. Such paleochannels include the Paleo-Savannah, the PaleoAltamaha, the Paleo-Medway rivers off the coast of the Georgia, and the Paleo-North
Edisto/Stono and the Paleo-Pee Dee systems off the South Carolina coast (Baldwin et al. 2006
Harris et al. 2005). The other paleo-channels plotted in Figure 7.5 above and Figures 7.8–7.9
below have been inferred by earlier workers such as Swift (1976), using indirect evidence such
as bathymetry. Areas adjacent to these paleochannels have the best potential for burial of sites
while the areas were subaerial, and therefore the best preservation potential. Wave action is also
a heavily destructive factor, and sites will better survive in a buried context if they exist below
the wave base. Rapid sea level rise may ameliorate the intensity of this destructive force and
allow for sites buried while the region was subaerial to survive transgression.
7.4. ARCHAEOLOGICAL SENSITIVITY AND PRESERVATION POTENTIAL
As noted in Section 7.2, there are no region-specific sea level curves defined for the Georgia
Bight, and general eustatic curves have been used, bracketed by RSL curves further north and
south of the region. Based on these data, archaeological sensitivity is defined as follows:
•
•
•
No Sensitivity. Areas 110 m and greater in depth are considered to be areas with No
Sensitivity for prehistoric sites, since these areas were not subaerial during the LGM.
Low Sensitivity. Areas from the 110 m to 70 m isobaths. This designation covers
areas exposed between the time of the LGM and the earliest Clovis occupation, ca.
13,000 B.P.
High Sensitivity. High Sensitivity areas include all areas within the OCS that are
shallower than 70 m.
127
Figure 7.8. Archaeological Sensitivity map for the Georgia Bight study region.
Figure 7.8 illustrates the sensitivity areas for the Georgia Bight study region. As indicated
above, sea levels during the LGM in the Georgia Bight regressed to the edge of the Continental
Slope at or below -110 m, so depths greater than 110 m have no potential for containing
prehistoric sites. Low Sensitivity areas encompass portions of the OCS that were subaerial
between the LGM and the beginning of the Paleoindian period, or ca. 20,000–13,000 B.P. Sea
level curves place the beginning of the Paleoindian period at approximately -70 m. High
Sensitivity areas for the Georgia Bight region range from -70 m to the 3-mile federal/state
boundary.
Other than landforms along paleochannels that may have been occupied and buried in
sediment prior to transgression, areas of High Preservation Potential are difficult to identify in
the Georgia Bight. Some areas, such as hardbottoms located in the vicinity of paleochannels,
have particular value for identifying archaeological materials, simply because of the lack of
128
Figure 7.9. Sea levels for specified periods, including High Potential area corresponding to
MWP 1b in the Georgia Bight study region.
sediments, but the seafloor surface in these locations provides little potential for intact sites,
given the scouring effects of bottom currents. As with the Middle Atlantic, it is postulated that
the rapid sea level rise associated with MWP 1b may have enhanced the preservation potential of
previously buried sites, so that region is represented in Figure 7.9 as an area with high
preservation potential.
129
8.
FLORIDA
8.1. REGIONAL GEOLOGY
The Florida Platform extends southward from the North American continent and separates
the Atlantic Ocean from the Gulf of Mexico (Figure 8.1). Since its formation, it has been
alternately flooded by shallow seas and salt lakes or has been sub-aerially exposed (Schmidt
1997). The Florida peninsula represents the exposed portion of the platform, and the remainder
extends offshore well into the Gulf of Mexico and, for a short distance, into the Atlantic Ocean.
Contemporary depths of the ocean floor around Florida are shown in Figure 8.2. The tectonic
stability, porous limestone bedrock, and thin sediment cover of the Florida Platform during the
last 15,000 years provide an ideal physical environment for the reconstruction of paleoshorelines
(Lidz and Shinn 1991). As early as 1966, Emery and Edwards (1966) identified the possibility
that prehistoric peoples inhabited the now-inundated Atlantic continental shelf of Florida.
Continental shelves around the world are remarkably dynamic, thus recreating past
submerged landscapes has been an adopted goal of archaeological, geological, and
geomorphological researchers. The intent of this chapter is to assess the past and present
geologic and physiographic changes within the study area as global sea levels rose since the
LGM, ca. 20,000 B.P. Understanding the evolution of the shelf and coastal geomorphic features
provides insight into the potential preservation and destruction of submerged prehistoric sites.
The modern Florida Atlantic continental shelf is a wave-dominated, low gradient feature with
a well-defined shelf/slope break. Tidal range and wave energy flux decrease within the shelf’s
southern portion. The modern sedimentary cover is dominated by quartz within the northern
portion, and is more carbonate-enriched to the south, while the inner shelf topography is
characterized by shoreface sand ridges on the north and relict reefs on the south (Hine 1997).
The shelf width throughout the study area varies greatly from 115 km off Jacksonville to 2 km
off southern Palm Beach County.
Generally, a thin band of Holocene sediments forms the present coastline of the state, and
these deposits consist of beach, dune, marsh, and lagoon sediments that developed in response to
the latest rise in sea level (Hine 1997). The Holocene sediments are comprised of clastic,
carbonate, and organic sediments.
During a previous study of the current project area (SAI 1981), two geomorphic provinces
(the North/Central Florida Shelf and Southern Florida Shelf north of the Keys) were identified
for the Florida Atlantic continental shelf based on wave climate, tidal range, and sediment
deposition. For the purposes of the current study, these two geomorphic provinces are
maintained.
The North/Central Florida Shelf consists of the portion of the shelf from the St. Johns River
to West Palm Beach. This geomorphic province represents a transition from the estuarine retreat
blanket of the Georgia Bight (southern South Carolina to northern Florida) and the mixed
carbonate/clastic depositional regime to the south (SAI 1981). The most common sediment in
this province is a fine to medium-grained, moderately sorted to well-sorted quartz sand having
130
131
Figure 7.1.
Georgia Bight study region.
Figure 8.2.
Offshore depths around Florida’s coastline today.
132
15 percent carbonate that consists of mostly bivalve fragments (Meisburger and Field 1975).
From Jacksonville to West Palm Beach, lithified beach deposits of the Anastasia Formation are
sporadically exposed. Major outcrops of the Anastasia Formation occur on the east coast of
Florida from St. Augustine to Ormond Beach, Cocoa to Eau Gallie, and Stuart to Boca Raton
(Lovejoy 1998).
The Anastasia Formation is a multi-cyclic deposit formed during former transgressions of the
sea (Scott 1997). Perkins (1977) recognized at least two disconformities within the Anastasia
Formation, and Osmond et al. (1970) measured two different ages for the formation, suggesting
two episodes of accumulation. At the time the Anastasia was deposited, 130,000–100,000 B.P.,
the earth was experiencing an interglacial, and sea level was approximately 6 m higher than that
of today (Lovejoy 1998). For example, Perkins (1977) points out that most of Martin and Palm
Beach counties were inundated, which would account for shells found inland as well as the low
hills that represent old barrier/beach dune deposits.
When the last glacial reached its apogee about 20,000 B.P., the glaciers began to melt,
temperatures rose, and sea level began to rise. As sea level rose to its present height, outcrops of
the Anastasia Formation underwent wave erosion in the surf zone (Lovejoy 1998). The outcrops
that still remain formed offshore reefs consisting of bedrock ledges with varying amounts of
living and dead coral, and are separated by intervals of barren, sandy bottom. The outcrops are
found at successively deeper intervals down to depths of 33 m and extending 1.6 km out from the
shore near Palm Beach Inlet and up to as much as 5 km north of the inlet (Lovejoy 1998).
The Anastasia lithology varies from coarse rock to sand, and shelly marl to sandy limestone,
with mollusks. Originally the Anastasia Formation was applied to only coquina rock, but it now
includes all Pleistocene marine deposits and is estimated to be approximately 30 m thick
(Murphy 1990). The Anastasia Formation grades into Miami Limestone in southern Palm Beach
County.
Within the North/Central Florida Shelf, Holocene cover is generally thin to absent.
Moreover, the Quaternary section is seldom thicker than 5 m, except beneath the linear or cape
associated shoals such as Cape Canaveral. In addition, there are no major sources of fluvial
sediment south of the St. Johns River. As a result, all terrigenous sands, including quartz,
feldspar, and other heavy minerals are derived from the Georgia Bight by southerly long-shore
transport. Consequently, reworking of tertiary rocks is probably an important sediment source
during sea level fluctuations, as is the production of biogenic sands, particularly in the south.
Dominant topographic features within this shelf province include the Cape Canaveral cuspate
foreland and its associated offshore bathymetry. In this area, the shelf supports a complex
bathymetric variety of attached and isolated linear shoals, broad depressions and highs, as well as
one cape retreat massif off Cape Canaveral and one off False Cape (Hine 1997). According to
Swift et al. (1972a), cape retreat massifs form on shelves as cuspate forelands that migrate
landward in response to sea level rise. Maximum Holocene thickness on the shelf within this
province is 12 m and is located beneath the Cape Canaveral retreat massif. Peats obtained by
vibracoring strongly suggest that back-barrier environments are preserved within the shelf
stratigraphic section in this area (SAI 1981). Radiocarbon dates from peats show Holocene and
133
Late Pleistocene ages, which would be contemporaneous with early populations if they existed
here. As such, there is a high potential for preserved sites or artifacts in the Cape Canaveral area.
The Southern Florida Shelf north of the Keys consists of the portion of the shelf extending
from Palm Beach County to Key West. It is a narrow section consisting of four shelf
environments: sand flats and karst; sand flats and coral reefs; sand flats, hardgrounds, and coral
reefs; and tidal sand flats and ridges, hardgrounds, and coral reefs (Finkl and Andrews 2008).
Perkins (1977) traced the evolution of the southern Florida Platform from an area dominated
by quartz sands mixed with carbonate sediments during the Early Pleistocene to a carbonatedominated environment during the Late Pleistocene. The Late Pleistocene carbonates include the
Miami and Key Largo Limestone. The Miami Limestone consists of two facies: an oolitic
limestone that underlies the Atlantic Coastal Ridge south of Boca Raton and a bryozoans-rich
limestone that underlies the Everglades and Bay of Florida (Scott 1997). The Paleoreef trend in
the Florida Keys is preserved in the Key Largo Limestone.
The seaward limit of the southern province is a relict Holocene reef line that lies in
approximately 15–25 m of water (SAI 1981). The rock ridge and terrace couplets are
progressively steeper in the seaward direction and support a modern benthic community of
alcyonarians, sponges, and scattered coral heads (Lighty et al. 1978). The rock ridges were
Acropora palmate (coral)-dominated barrier reefs that flourished during the Late Holocene, and
radiocarbon dating indicates that these reef tracts terminated approximately 7000 B.P. (Lighty et
al. 1978). Lighty et al. (1978) have postulated that exposed soil horizons were eroded as sea level
rose, creating turbidity levels high enough to stress the reef community. In addition, broad
lagoons formed behind the reefs, and the flow of cold water over the reef during the winter also
contributed to reef demise.
Surface sediments on the Southern Florida Shelf north of the Keys include Halimeda,
mollusks, benthic foraminifers, bryozoans, and corals. Active sabellariid worm reefs and their
debris are common near the shoreface within water depths ranging from 3–10 m. There are also
nonskeletal carbonate components, including pellets and ooids. The southern province is
considered more tropical than that to the north, and the two provinces contain distinct
assemblages. For instance, the Halimeda and coral are not present in the northern province, and
mollusks and barnacles are not present in the southern province (Hine 1997).
In situ carbonate skeletal sands have accumulated up to 5 m in the topographic lows between
the reefs. These areas presumably have high preservation potential for cultural resources, as little
sediment exchange occurs between the mixed terrigenous/carbonate sands of the beach zone and
the offshore carbonate sands.
The shelf seaward of the Keys may be considered a subprovince of the Southern Florida
Shelf North of the Keys (SAI 1981). The Keys are a chain of Pleistocene reef and ooliticlimestone islands that extend from Miami to Key West, and which separate Florida Bay from the
Florida Straits (Randazzo and Halley 1997).
The two primary surficial stratigraphic units of the Florida Keys are Key Largo and Miami
Limestone. The Upper Keys are generally reefs of the Key Largo Limestone and are oriented
134
parallel to the shelf edge. The Lower Keys are elongated and perpendicular to the shelf and are
composed of oolite of the Miami limestone.
Seaward of the Keys, the shelf consists of intermittent living and dead linear and patch reefs,
barren rubble flats and mounds, and belts of thin, mobile sand bodies. The slope seaward of the
shelf edge reef track drops off steeply in the Florida Straits.
The ancient reefs that fossilized in the Keys and the surrounding subsurface flourished
between about 145,000–90,000 B.P. (Randazzo and Halley 1997), at which time sea level was 6–
8 m higher and the reef system extended to the Gulf Stream. Lacking Acropora palmata, the
present-day Florida Keys include a series of coalescing patch reefs. These reefs grew on the
preexisting topographic relief on an otherwise open shelf (Harrison and Coniglio 1985).
According to Shinn et al. (1989) there is a zone of older, dead reefs near the outer shelf margin.
These dead reefs are suspected to represent multiple reef tracts and may be of Pleistocene age.
Shinn et al. (1989) have outlined the control of reef distribution by eustatic sea level changes
and topography. A sea level fall that began approximately 100,000 B.P. is believed to be
responsible for the eventual distribution of a large part of the Pleistocene coral reef system and
exposure of the Florida Keys. Sea level may have fallen more than 100 m, and during this
exposure, laminated calcareous crusts formed (Mutler and Hoffmeister 1968). The exposure
surface formed the substrate colonized by more recent corals when sea level began to rise about
18,000 B.P., eventually flooding the South Florida Shelf. Peat deposits formed in bogs and 14C
dating of these materials have helped to recognize geologic events over the last 10,000 years
(Robbin and Stipp 1979). Sea level has steadily risen over the past 15,000 years, which has
renewed cross-shelf water interchanges between the Gulf of Mexico and the Atlantic Ocean. The
combination of higher salinity during the summer and colder temperatures in the winter would
greatly inhibit the ability of reef building corals to recolonize, especially Acropora palmate
(Randazzo and Halley 1997).
Localized coral reef colonization appears to be controlled by topographic conditions that
provided substrate highs, and coral debris distributed on the landward side of reefs provides the
hard substrate required for reef colonization during periods of sea level rise. Moreover, ecologic
zones have expanded upward and landward in response to sea level rise. The Caribbean reef crest
Acropora palmata forms buttresses (or trough-like grooves) on the seaward edge of the best
developed Keys reefs (Shinn 1963).
Wave and current action, along with preferential coral growth on the spurs, serve to
accentuate the configuration of the reef structure (Randazzo and Halley 1997); however,
settlement of coral larvae is prevented by the sedimentological processes in the grooves (Shinn
1988). The reef platform is comprised of fore-reef, reef tract, and back-reef zones (Randazzo and
Halley 1997), and core-boring transects across reefs (Shinn et al. 1989) show that modern reefs
have formed on the topographic relief provided by Pleistocene reefs. Reefs have not built
seaward, but have grown in place or migrated landward a short distance (about 100 m) over their
back reef sand and rubble deposits.
The potential for preservation of sites on this subaerially exposed pre-Holocene surface
would be high. As the seaward reefs build up, the back reef areas become flooded by low-energy
135
lagoonal waters. Further, carbonate-producing organisms generate sediment, which would slowly
bury man-made artifacts or sites. Since most wave energy is expended on the shelf edge reef, it
would be unlikely that the lagoonal sedimentary sequence would be disturbed to any significant
depth by high energy events.
In terms of drainage systems, the peninsula of Florida is characterized by north-to-south
orientation of rivers that reflects the nearshore environment, which contributed to its basic
landform construction (Schmidt 1997). During past sea level high stands, relict beach ridges
were constructed, and these ridges are separated by swales that were previously occupied by
shallow lagoons (Schmidt 1997). When sea level dropped, these lagoons became valleys, and
streams eroded the sands and clays, creating several coast parallel river systems (Schmidt 1997).
The only major stream or river flowing into the project area is the St. Johns. Although it is
extensive and broad, the St. Johns River has a very low discharge rate, averaging 8,300 cubic
feet per second. This discharge is related primarily to volume and less to velocity, as the river
has a wide floodplain and low gradient (0.02 m per km) (Miller 1998). For most of its length, the
St. Johns River does not exceed 1.5 m above mean sea level. The low gradient makes the river
susceptible to small changes in sea level, and even today the river is tidally influenced as far
south as the Wekiva River, approximately 120 miles from the river’s mouth. Given the lack of
major streams or rivers aside from the St. Johns River within the project area, the terrigenous
sediments along Florida’s shelf have been carried from the north by longshore transport, and the
calcareous sediments that exist have been generated in situ.
8.2
RELATIVE SEA LEVEL CHANGES
While focused on the Gulf of Mexico side of Florida, the recent work by Balsillie and
Donoghue (2004) combines all relevant classes of data in the area immediately adjacent to the
Floridian Atlantic Continental Shelf with particular reliance on directly dated features.
Representative sea levels based on their analysis are presented in Table 8.1. Balsillie and
Donoghue’s (2004) data correlate well with other late Quaternary sea-level estimates, in
particular a eustatic index based on data sets acquired from the Red Sea (Siddall et al. 2003).
They suggest that the Gulf’s relatively low-energy environment and geological stability offer a
“near-eustatic sea-level” curve with global application. The authors contend that the mid- to lateHolocene high-stands in the Gulf provide what is arguably the best example of eustatic sea level
change worldwide.
Table 8.1. Florida Sea Level Curves (based on Balsillie and Donoghue 2004).
Time Period (years B.P.)
Meters Below Present Sea Level
6000
2
8000
10
10,000
25
12,000
40
14,000
80
16,000
100
18,000
110
20,000
120
136
Using valid in-place sea level indicators (marine peats and intertidal oyster shell beneath the
sea floor) south of Cape Hatteras, Blackwelder et al. (1979) also refined the Milliman and Emery
curve. The result was a shallower curve (30 m) for the Southeast Atlantic shelf (Weaver 2002).
Their curve indicates that at 17,000 B.P. the sea stood 60 m below its present level and that by
10,000 B.P. it rose to 22 m below the present level (Dunbar et al. 1992). These figures seem in
line with the younger curve derived by Scholl et al. (1969), which is from a large, well dated set
of paired freshwater and tidal sediment samples from the Florida Keys. These data show a
gradually decreasing rate of sea level rise at about 6000 B.P. (Dunbar et al. 1992).
Ongoing research in Florida combined with recent advances in our understanding of the
Clovis time period indicate that the critical period from 13,100 to 12,800 calendar years before
present may cover the entire period in which Clovis existed (Waters and Stafford 2007). During
this brief Clovis florescence, sea level rose nearly 25 m, from roughly 75–50 m, which has
profound implications for land use, resource distribution, and resource exploitation by Early
Paleoindians (Balsillie and Donohue 2004).
Sea levels oscillate continuously, albeit some temporal periods have experienced more
drastic climatic changes than others, resulting in rapid sea level fluctuations. Worldwide sea
levels at the end of the LGM (approximately 20,000 B.P.) were 120–130 m lower than presentday levels, indicating that large portions of the continental shelf off Florida’s coast were
exposed. When Paleoindians arrived over 12,000 years ago, sea levels were approximately 40 m
lower than present-day levels (Faught 2004:277), and in Florida the climate was cooler and drier,
and water was in shorter supply at inland locations (Milanich 1994:40). A brief stadial of cooler
temperatures referred to as the Younger Dryas is believed to have occurred between ca. 11,000
and 10,000 B.P., and during this interval, sea level rise stabilized; however, by about 10,000
B.P., increased temperatures led to accelerated glacial melting, thereby increasing the rate of sea
level rise once again (Faught 2004:277). Sea levels continued to rise at a fairly rapid pace
throughout the ensuing millennia, until conditions comparable to those of modern times occurred
at about 5000 B.P. (DePratter and Howard 1977; Miller 1998:39; Randall and Sassaman
2005:17; Thomas 2008:42).
These sea level changes influenced human settlement (Faught 2004:277) and ritualistic
behavior (Sassaman 2009), led to the inundation of numerous coastal prehistoric settlements
(Marks 2006:xiii), and flooded areas along the continental shelf, greatly reducing the land mass
there (Gannon 1996:2). Once sea level changes stabilized, barrier islands began to form on the
coasts, stream gradients became reduced and stabilized, vegetative complexes and associated
fauna became essentially comparable to that of modern times, and surface waters became much
more prevalent (Miller 1998:39; Thomas 2008:42). Although sea level largely stabilized by 5000
B.P, small-scale changes are documented in later times, including the period from about 3000–
2500 B.P., when there is archaeological evidence for slightly lower sea stands (Ashley 2008;
DePratter and Howard 1980; Russo 1992).
Sea level rises over the past 2,500 years in south Florida have an average rate of about
38.1 mm per century. However, since 1932 (when tide gauge monitoring stations were installed)
south Florida has incurred a 22.86 cm relative rise in sea level. It has been postulated that this
137
accelerated rise is the result of warming (and expansion) of water in the western North Atlantic
Ocean (Broward County Climate Change Task Force 2009).
8.3
MARINE TRANSGRESSION AND SITE PRESERVATION
The rate of sea level rise is important to the development of the Florida Atlantic coast. The
spatial and temporal distribution of the archaeological record of coastally adapted cultures along
the Atlantic coast of Florida must be understood in the context of the evolving coastal landscape.
Numerous investigators have shown that Holocene sea level rise has played a key role in the
origin and development of the inner continental shelf seafloor and modern barrier island system
(Dillon 1970; Field and Duane 1976; Hoyt 1967; Pierce and Colquhoun 1970; Swift 1975;
Thomas 2008). Because shoreface retreat is the dominant transgressive process along the Florida
Atlantic coast, the great majority of Paleoindian and Archaic period archaeological sites that
were once on the continental shelf were likely destroyed during the Late Quaternary sea level
rise if they were exposed to heavy wave action or storm surges for any length of time (Waters
1992).
Generally speaking, most of the Florida Atlantic continental shelf is low and sloping, and
therefore would have been dominated by erosional transgression—particularly in the
North/Central Florida Shelf. Such conditions would afford a low potential to preserve
archaeological sites that may have existed. However, there are local settings within this region
that would have experienced a depositional transgression, such as cape-associated shoals,
nearshore linear shoals, the mainland side of lagoons, and along the banks of estuaries (SAI
1981:I-57, I-59). The carbonate-dominated Southern Florida Shelf also exhibits preservation
potential for underlying Pleistocene substrate due to its physically hard character and reef
development, resulting in a low energy wave environment. Therefore, sites may have survived in
low energy environments such as the marshes of a delta or lagoon, or tidal flats fronting the
ocean, since such sites often subside into the mud and become buried (Gagliano et al. 1982,
1984).
Native American land use is predicated on such factors as topography, access to water, soil
drainage, and resource availability. Specifically, elevated and well-drained landforms were often
preferred for habitation sites. Availability of raw materials, such as stone for the manufacture of
tools, has influenced Native American settlement, as has the proximity and direct access to
water. Within Florida, there are numerous types of water sources, including streams, rivers,
wetlands, ponds, lakes, springs, and oceans, and these sources proved important as a drinking
source for humans and the terrestrial animals they hunted, as a source of various edible aquatic
species, as well as for water travel and bathing. Today, Florida’s coasts support the state’s most
densely populated areas, and Native Americans also aggregated along the shore. Because the
inundated portion of the shelf is an extension of the Coastal Plain, it is asserted that settlement
patterning along the present terrestrial coastline should be mirrored in submerged settings of the
Coastal Plain. Native American settlements are also readily found along the state’s river and
stream banks and at the springs. In fact, many spring sites in the interior of the state contain
inundated Pleistocene-age cultural deposits with exceptional preservation, most notably at the
Page Ladson site in the Aucilla River (Dunbar 2006b).
138
Unlike the northeast Gulf of Mexico, there are no major paleochannels extending onto the
Florida continental shelf in the Atlantic; however, three offshore springs have been mapped on
the eastern part of the state. The three Atlantic Coast submarine springs include the Crescent
Beach Spring, which is a short distance off the coast of southern St. Johns County, Red Snapper
Sink, which is about 30 km off the coast, east of the boundaries of St. Johns and Flagler counties
(Rosenau et al. 1977), and Flagler Beach sinkhole (DeLoach 2000:223). All three areas represent
high probability locations for encountering submerged cultural resources (see Figure 8.4).
Examination of elevation depths along the offshore portion of Florida’s Coastal Plain permits
an estimation of the amount of available land (now submerged) during differing periods in
Florida’s past. The submerged cultural resource sensitivity zones outlined below take into
account the habitable regions of the continental shelf during periods when Florida was known to
have been inhabited.
The modern east coast of Florida is generally composed of Holocene quartz sand barrier
islands. Prior to 7000 B.P., the rate of sea level rise was too great for the prolonged stabilization
of barrier islands and would have been unlikely to have sustained prehistoric populations before
then. The stabilization of barrier islands is important because the lower topographic areas behind
these islands never infill with marsh, but instead develop into open-water lagoons. Marshes in
this environment are restricted to low-energy, narrow fringes along the lagoon/estuarine
shoreline where they are protected from wave erosion (SAI 1981). This environment produces a
highly productive backbarrier and lagoonal flats with substrates beneficial to mollusks, sea
grasses, mangroves, fish, and numerous other species. More importantly, this area would have
provided an attractive location for habitation.
The low energy level combined with the sediment trapping and sediment stabilizing effects
of salt grass and mangroves would allow for the subsequent preservation of prehistoric sites and
their protective estuarine muds. These sites and estuarine muds could lie buried beneath the
active Holocene sands on the shelf during continued sea level rise and landward barrier island
movement; recovery of peat during coring on the continental shelf could indicate potential
preservation (Brech 2004). Also, the immediate nearshore environment contains abundant post
2000 B.P. archaeological sites.
Geological considerations are important to the understanding of site preservation and/or
destruction. Factors such as wave and current action are paramount to the preservation of
submerged resources. Moreover, recognition of such off-shore geological features as springs and
sinks can aid in the detection of submerged sites.
Underwater archaeological investigations along Florida’s Gulf Coast have successfully
yielded early submerged cultural deposits (Faught 2004; Marks 2006). To the contrary, only one
offshore submerged archaeological site has been identified on Florida’s Atlantic Coast. This site
has an intact Early Archaic component and Pleistocene-era megafauna, and is referred to as the
Douglas Beach Site (8SL17); it was found 200 m off the coast in shallow waters (2–6 m in
depth) approximately 5.6 km south of Fort Pierce Inlet (Murphy 1990). The site was associated
with a peat deposit underlain by gray-green clay, and the occurrence of this underlying clay
deposit might signify the type of sediments where other submerged sites exist.
139
8.4. ARCHAEOLOGICAL SENSITIVITY AND PRESERVATION POTENTIAL
Based on the most current sea level curves for this region, archaeological sensitivity is
defined as follows and depicted in Figure 8.3.
No Sensitivity. Areas 120 m and greater in depth are considered to be areas with No
Sensitivity for prehistoric sites, since these areas were not subaerial during the LGM.
Low Sensitivity. Areas that were subaerial beginning with the LGM until approximately
13,000 B.P. fall within the 120–60 m isobaths. Such areas may, but are unlikely, to have ever
experienced human occupation while subaerial.
High Sensitivity. High Sensitivity areas include all areas within the OCS that are shallower
than 60 m. However, within the High Sensitivity area, some refinements are necessary for the
Florida region.
Although the High Sensitivity area extends to the beginning of the Paleoindian period, there
is little potential for intact Paleoindian sites in the OCS. There is very little archaeological
evidence of occupation along the Atlantic Coast during this period based on terrestrial data.
During this period, most Native American settlements in Florida were tethered to karstic areas
where springs and sinks were present and where raw material for use in stone tool production
was prevalent. Although two sinks are known to be submerged in the Atlantic Ocean off the
coast of southern St. Johns and northern Flagler counties, these two water holes lacked the
silicified Tertiary-age limestone that would have been sought as a raw material for tool
production. Thus the lack of locally available cryptocrystalline lithic raw material might have
deterred prolonged Atlantic Coast habitation during this early period in Florida prehistory;
although the presence of minimally used transitory camps during this time remains a distinct
possibility. Further substantiating the reduced potential for offshore Paleoindian period sites
along Florida’s east coast is the minimal number of Paleoindian projectile points reported for the
12 Florida counties that front the Atlantic Ocean. Specifically, only nine Paleoindian points have
been reported from these 12 counties (PIDBA 2009), and this sparse number of points suggests
that Paleoindian land migrations passing through these present coastal counties were minimal.
Even after the start of the Archaic period ca. 10,000 B.P., Florida’s Atlantic Coast is assumed
to have been minimally populated for some two thousand years; it was only shortly after this
period that the first substantial occupation of Florida’s east coast is documented. The earliest
documented site in Florida, a mortuary pond in Brevard County known as the Windover Pond
Site, dates to about 7400 B.P. Excavations from this unique site portray a sedentary society that
apparently had a restricted mobility range and ceremonious mortuary practices. Although this
site post-dates 8000 B.P., other such charnel ponds from earlier periods might exist beneath the
ocean’s waters. Depths for the time range of 8000–10,000 B.P. are -10 to -40 m below present.
The greatest potential for archaeological sites within the High Sensitivity area extends from
the outer edge of the shoreline as it was exposed at 8000 B.P. to the present shoreline, and
encompasses the Early to Middle Archaic periods (Figure 8.4). This zone is found at depths of
10 m and less. The present archaeological record shows a higher incidence of deposits of this
time period along the coastal strand than earlier sites dating to the Paleoindian period. This area
140
Figure 8.3.
Archaeological Sensitivity map for the Florida study region.
does not extend very far from the present shoreline and exhibits a high probability for yielding
deposits from seasonal camps dating to the earlier part of the Archaic period. The two known
Atlantic Coast submerged spring sites (Crescent Beach Spring and Red Snapper Sink) are
included within this High Sensitivity area.
141
Figure 8.4.
Site potential areas for archaeological sites within the High Sensitivity area in the
Florida study region.
142
143
PAGE LEFT INTENTIONALLY BLANK
144
SECTION 3 – DISCUSSION
145
9.
DISCUSSION–SYNTHESIS OF MODEL
Any approach for identifying where prehistoric sites may exist on the Atlantic OCS must
address a few of basic questions. First, when were people here? Second, what portions of the
Atlantic OCS were subaerial at a time when human occupation was present? And, third, where
would such sites, assuming they once existed, have survived marine transgression and be
preserved? Finally, assuming they have been preserved, what techniques should one employ to
find them?
This chapter summarizes the research presented in Chapters 3–8 and presents considerations
and an approach for locating prehistoric sites on the OCS. Modeling site potential and
preservation potential depends on a reconstruction of sea levels and geomorphology. Regionally
distinct geological and environmental processes have been at work since the Late Pleistocene,
and thus the discussion in Chapters 3–8 will be more pertinent to planning for any particular
project. The current chapter outlines the general principles on which the regional models are
based, and notes certain weaknesses and gaps in current scholarly work on issues related to sea
level and geomorphology that can be refined through further research and that should be taken
into account when planning investigations for a particular undertaking.
9.1. TIMING OF HUMAN SETTLEMENT
Chapter 2 discussed the current knowledge of the prehistoric occupation of the East Coast.
The best data currently indicates that Paleoindians were the first to occupy the Eastern Seaboard,
but there still remain hints of possible pre-Paleoindian settlement. From the best evidence
available, the earliest Paleoindian component, Clovis, dates no earlier than 13,000 B.P., although
the precise dating of this Clovis period remains largely uncertain due to the paucity of sites.
Assuming that a pre-Clovis occupation is possible, this date might be pushed back another 500–
1,000 years, depending on which possibilities one might be willing to entertain for an earlier
entrada into North America. The support for a European-based settlement of the Atlantic
Seaboard is exceedingly limited, and remains little more than informed speculation. If one
assumes an early, coastal migration model originating from Beringia that can account for an
occupation at Monte Verde, Chile, at ca. 13,800 B.P., a date of 13,000 B.P. also seems
reasonable for a likely early date of human occupation in the study area. Therefore, while not to
discount the possibility of an earlier occupation of the East Coast, there are no data to indicate
that significant resources should be invested searching for submerged prehistoric sites that may
date earlier than 13,000 B.P. It is this cutoff in time that defines the High Sensitivity areas for the
model.
The timing of human settlement in the region is important because it helps to set some
parameters on the depths at which prehistoric sites are possible. Effective exploration of OCS for
surviving prehistoric sites will, however, closely depend on correlating when sites may have
been formed while portions of the OCS was subaerial with contemporary ocean depths.
9.2. USE OF SEA LEVEL CURVES
Defining what portions of the Atlantic OCS would have been subaerial 13,000 years ago is
not a straightforward undertaking. Sea levels during the Late Pleistocene and early Holocene
146
were influenced by a wide range of factors, and involved not only the surface level of the sea but
the surface level of the land mass as well (Dix et al. 2008:13). A number of scholars have
prepared sea level curves, using a combination of sea level indicators such as shells, tree stumps,
bulk saltmarsh-peat, plant remains, foraminifera, and many other formerly-living plants and
animals. These materials are dated by various methods and plotted based on their elevation to
create sea-level curves for various regions. However, there is a great deal of regional variability
in sea levels, with the impacts of post-glacial continental rebound (due to relief from the weight
of glaciers on the land surface) and hydro isostatic rebound (uplift in coastal areas resulting from
the increased weight of water associated with rising sea levels) experienced differentially.
Despite the many efforts at refining regional sea level curves, to a certain extent this work is
as much art as science. As Pirazzoli (1991:21) notes, “most sea-level curves are not deduced
mathematically from the data, but interpreted by their author in order to follow more or less
closely certain index points considered more reliable and to express trend variations deduced
from other information and from interpretation of observations which are not always strictly
quantifiable. From this point of view, as in many scientific fields depending mainly on
observation, a linear sea-level curve is constructed not only from data, but also from subjective
ideas, and in some cases preconceived theories of their author.” The difficulty in understanding
relative sea level changes during earlier periods of the Holocene, for example, is compounded by
the lack of data points, so researchers extrapolate curves, often without detailing the methods or
assumptions underpinning them.
This study has endeavored to summarize the most recent literature on sea level rise, and
illustrates both the regional differences and gaps in knowledge (Table 9.1). In the Gulf of Maine,
for example, isostatic depression of the land by the weight of the ice impacted sea levels in ways
not experienced to the south. Likewise, as the glaciers melted and worldwide ocean levels
increased, isostatic rebound in that region slowed the pace of RSL rise between 11,500 B.P. and
7500 B.P. Assuming the existence of a rapid period of eustatic sea level rise during MWP 1b (ca.
11,500–11,100 B.P.), RSL changes in the Gulf of Maine differed dramatically from regions
further south.
Table 9.1. Sea Level Curves for Atlantic OCS.
References
Maine
S. New
England
New YorkNew Jersey
Middle
Atlantic
Kelley et al. 2010
Oldale 1992
60
107
Wright et al. 2009
120
70
30
19 (7.5k)
12 (5k)
Wright et al. 2009; Nikitina et al.
2000; Horton et al. 2009; Mallinson et
al. 2005
No local research; inferred from
global eustatic curves and constrained
by research in NC and FL
Balsillie and Donoghue 2004
120
70
32
15
12
110
70
58
25
10
120
60
40
25
10
Georgia
Bight
Florida
20,000
LGM
Sea Level (Meters) & Years B.P.
13,000–
10,000
8000
12,500
Paleoindian
60
20
19
70
40
20
Region
147
6000
N/A
10
The sea level curve for the Southern New England region relies on the research undertaken
by Oldale (1992). Based on his calculations, sea level during the LGM was 107 m below its
present level. During the beginning of the Paleoindian period in the region, sea levels
approximately 70 m below present sea level, and were at 40 m below present by ca. 10,000 B.P.,
the start of the Archaic period. By 6000 B.P., sea levels were about 10 m below present. In the
New York–New Jersey region, the most recent research by Wright et al. (2009) indicate that the
LGM sea levels were approximately 120 m below today’s, a trend that is similar down the coast
to Florida (although researchers in the Georgia Bight have assumed a depth of -110 m in that
region at the LGM). Seventy meters below present currently is a generally accepted depth for the
the start of the Paleoindian period (ca. 13,000 B.P.) south of the Gulf of Maine, with more
variability region to region as one moves closer to the present. These differences may be caused
by a number of factors, ranging from actual local variations in RSL from isostasy to dating
errors. As noted above, some of the differences also may reflect the fact that the curves are
drawn to connect sometimes sparse data points further out in time.
Reconciling the many variables and inconsistencies among sea level curves is beyond the
scope of this project. For future undertakings within the Atlantic OCS, applicants should ensure
that the most current research on sea level rise is used to determine which portions, if any, of
their project area would have been subaerial as far back as 13,000 B.P.
Once the appropriate sea level curve has been selected, the paleoshoreline from ca.
13,000 B.P. should be reconstructed by using gridded bathymetric data. A starting point should
be the bathymetry available from NOAA’s National Geophysical Data Center, U.S. Coastal
Relief Model (NOAA 2010). These data, which incorporate 3 arc-second data from sources such
as the U.S. National Ocean Service Hydrographic Database, the U.S. Geological Survey, the
Monterey Bay Aquarium Research Institute, and the U.S. Army Corps of Engineers, can be
imported into a Geographic Information System such as ESRI’s ArcGIS and modified into
classified fields, allowing selection of appropriate depths. This process provides only a gross
overview of a paleoshoreline location at a given point of time. Accordingly, it is recommended
that the data be refined locally within a project area by means of bathymetric survey, which will
provide more accuracy and enable a better understanding of the seafloor topography within the
project area.
Bathymetry also is a good starting place, but it is important to recognize that conceptually
draining the ocean to the current bathymetric depths does not provide an accurate view of what
the project area would have looked like when subaerially exposed. The impacts of marine
transgression need to be considered, since areas of the seafloor that now appear to be flat may
represent former valleys or paleochannels or paleosols now covered in sediment. Likewise, the
configuration of valleys on the seafloor reflects channel erosion during transgression and
directional reworking due to bottom currents, tidal actions, and storm patterns. Methods for
reconstructing the geomorphology of the subaerial topography include subbottom profilers
paired with coring, and are discussed in more detail in Chapter 10 below.
9.3. SITE LOCATION AND SURVIVAL
There are myriad factors that can influence—or in some cases dictate—the potential for an
archaeological site to survive transgression intact, including disturbances while the site was
148
subaerial, forces associated with transgression, and post-inundation impacts. At a fundamental
level, however, there is a key prerequisite for preservation of an inundated site: it must be buried
in terrestrial or low-energy marine sediments prior to the transgression of the ocean’s rising
waters (Waters 1992). In the absence of such pre-inundation burial, the re-leveling process of
marine transgression will simply destroy the site. Pre-inundation burial is no guarantee of site
preservation, but simply a necessary condition.
Settings that have the most potential for site burial, then, should include well developed
marsh-lagoon-barrier systems and flood plain-marsh-estuary systems—areas that would have
been attractive to human occupation, but also conducive to low-energy sedimentation either prior
to or during the early stages of transgression (Moir 1979:I-206). Stright (1990:457) identified
“landforms such as river valleys, bays, estuaries, lagoons, lakes, ponds, sinkholes and subsiding
deltas” as low-energy environments where sites could be buried deep enough to survive the
impacts of marine transgression. Conversely, contexts where sites are exposed to the direct
impact of breaking waves, as well as the subsequent oscillatory water motion along the seabed
when inundated a few meters, would eliminate any potential for site survival (Bailey and
Flemming 2008:2159). Where favorable environments do not exist, the impact of waves and
currents—particularly during storm events—could be highly destructive, leaving little more than
resorted artifacts. Therefore, beaches, dunes, washover flats, and major headlands are areas
where preservation is unlikely to exist (Swift 1976). Exceptions exist where paleo deltas are
present with deeply buried artifacts, or where spits have developed, or other local conditions
allowed stepwise retreat of the coastline as sea level rose, such that sites on the lee side of dunes
or on sheltered beaches are protected from direct shoreface erosion and instead are rapidly
submerged.
The rate of sea level rise also is a major factor in both site formation and site preservation
throughout the study area. During periods of slow sea level rise, stable productive estuarine
settings were created that were attractive for human settlement. Lower energy environments tend
to yield the highest returns in terms of biological productivity, since slow currents allow the
accumulation of a sediment base for aquatic plants, which will in turn support invertebrate life
and protective areas for fish hatcheries. In addition, higher net productivity tends to coincide
with broad, shallow bathymetry, particularly at fresh and saline water interchanges (Perlman
1980). Lagoons and estuaries are among the most productive coastal features, and thus are likely
to have attracted prehistoric hunter-gatherers. Thus, slow sea level rise likely facilitated
archaeological site formation. However, as a general rule, slow rates of sea level rise also have a
greater probability of negatively impacting the integrity of coastal archaeological sites, as tidal
scouring of adjacent upland or drowned upland archaeological sites would be greater during
periods of relatively stable sea levels than it would be during periods experiencing rapid marine
transgression.
There are, of course, exceptions to the general rule. In certain areas and settings, relatively
slow rates of sea level rise can result in the preservation of drowned archaeological sites with
intact features. As such, there are conditions where the integrity of drowned prehistoric
archaeological features and deposits would not be negatively impacted by stable sea levels. The
upper reaches of slow moving tidal creeks, for example, could provide such a setting. The
regular input of organic detritus and a lack of open water, which would limit the degree of fetch149
related erosion, could provide a situation where anaerobic and anoxic conditions could be
attained over a very short period of time. If so, the anaerobic and anoxic conditions would limit
the degree of bioturbation disturbance to the archaeological features and deposits. In these
settings, the drowned archaeological remains would be preserved.
Rates of marine transgression have varied in the Atlantic OCS over the past 13,000 years,
which defines the period of High Sensitivity for the presence of archaeological sites. It is
assumed that the sea level rise rates during this period would have greatly impacted the longterm preservation of archaeological deposits that had only been submerged recently. Fetchrelated wave erosion, littoral drift, tidal scouring, and bioturbation represent some of the natural
processes that would have impacted archaeological sites during the transition from an upland site
to the offshore swash and berm zone. However, it is assumed that during periods in the past that
experienced rapid rates of marine transgression, the integrity of archaeological landscapes,
deposits, and features would have had a greater chance of long-term preservation.
Rapid rates of sea level rise would not have eliminated the natural processes impacting
archaeological deposits. Rapid marine transgression would, however, have limited an
archaeological site’s duration of exposure to these natural destructive forces. When a former
upland archaeological site is situated offshore in water depths below wave base, it is assumed
that the archaeological remains would be preserved in situ or near the location of original
deposition.
The rapid burial of an archaeological site also has the benefit of possibly limiting some
aspects of post burial processes that would cause disturbance to the site’s integrity. The quicker a
drowned archaeological site reaches a setting with anaerobic or anoxic conditions, for example,
the more likely the integrity of the site will be preserved. As Lowery (2009b) notes for sites in
and around the coastal areas of the Delmarva Peninsula in the Middle Atlantic region, tidal
marshes are teeming with numerous burrowing crab species. If sea level in a region is stable or
rising at very slow rates, bioturbation via tidal marsh and inter-tidal organisms would negatively
impact the integrity of coastal or drowned upland archaeological sites. Along the lower
Chesapeake Bay and the Atlantic coastline, tidal marshes with drowned upland archaeological
sites can, during the summer months, contain literally millions of burrowing fiddler crabs.
Burrows extend a foot or more below the surface and vary in diameter depending on the size of
the individual crab. As a result, old sediments and archaeological remains can be brought to the
surface and modern surface organics and non-archaeological sediments can be introduced into
the deeper strata. Rapid submergence is no guarantee of protection from such bioturbation, as
Ferrari and Adams (1990) have noted in their discussion of the impact that burrowing by various
fish and crustaceans can have on marine sediments. It may, however, limit some of the intensive
disturbances documented in tidal and inter-tidal settings.
As noted in Chapters 3–8, archaeological site preservation potential on the OCS is tied to
unique, protected settings in different regions. Site preservation may also be related to regional
topography, resistance of sediments to erosion, sediment supply, depth of erosion and wave
energy, and tidal range (Belknap and Kraft 1981; Waters 1992). Site protection may be provided
by bedrock formations. Morphology of the Coastal Plain prior to transgression is important,
since sites that are lower in topography (such as sites in river valleys or adjacent to low-lying
150
lagoons) will withstand erosion comparatively well, as they are below the height of most of the
impact energy resulting from waves, tides, and currents. The cohesive strength of sediments will
determine their relative ability to resist erosion, while sediment deposition can compensate for
any instability in the original matrix. Recently, Leach et al. (2009) reported on a preserved
submerged landform dating to 6300 B.P. in approximately 13 m of water off the coast of Maine,
although not within the OCS. They postulate that preservation at this locale was likely due to
armoring of the sediment package by bedrock and a relict oyster bed, as well as rapid sea level
rise.
In the Gulf of Maine, sheltered areas that developed during a slowstand may have been
protected from marine transgression because of spit formation. Whereas one would normally
assume that coastal processes would cause considerable erosion during periods of slow sea level
rise, these forces also created conditions that may have preserved certain contexts that would
have been attractive for human occupation. Thus, some of these sites may have been preserved
because spits offered some protection from marine transgression. In the New England and New
York–New Jersey regions, there is evidence for intact, relict surface features or paleosols that
may contain preserved sites. The presence of such features (e.g., Rampino and Sanders 1980;
Robinson et al. 2004; Sanders and Kumar 1975a, 1975b) suggest that stepwise retreat may have
aided burial of archaeological sites as well in these regions, although no actual intact sites have
yet been reported.
Seismic evidence in the New York Bight has indicated the potential for preserved lagoonal
settings that may have been attractive settings for prehistoric occupation and offer the hope of
site preseravation. Nordfjord et al. (2009) identified two examines in their seismic profiling of
the seafloor with apparent preserved remnants of back-barrier morphology exist at depths of 50–
60 m, which would be associated with the Paleoindian period. Likewise, Sanders and Kumar
(1975a) have identified similar seismic evidence they interpret as a preserved back-barrier,
which was noted at a depth of 24 m off the coast of Fire Island, New York. Nordfjord et al.
(2009) offer two explanations for how such features might have survived erosion during
shoreface retreat. One possibility they suggest is that these features represent pauses in the retreat
of the shoreline during the last transgression not yet documented in regional sea level curves,
which may have provided “time for barrier systems and related deposits to develop that were
substantial enough not to be erased by the subsequent passage of the shoreface” (Nordfjord et al.
2009:239). Another possible explanation is that there may have been more topography seaward
of these features that protected them initially from transgression, but when this topography was
eventually flattened, the localized shoreface retreat would have been rapid, enhancing
preservation (Nordfjord et al. 2009:239).
In the Middle Atlantic region, research identified the presence of likely paleo shelf-edge
deltas off the coast of North Carolina, which would have been preserved during a period of
relatively rapid sea level rise. Such relict features suggest the potential for preserved
archaeological sites as well. Further south, in the Georgia Bight, geological conditions indicate
that site preservation is most likely along paleochannels, where archaeological sites may have
been buried prior to inundation. Because of intensive scouring of the seafloor in the region, any
archaeological sites preserved would likely be deeply covered in sediment—making discovery
challenging. On the Florida coast, there is an absence of any major paleochannels extending into
151
the OCS. Therefore, the most likely settings for intact prehistoric sites are more limited, and
settings such as offshore springs represent the most likely location for both prehistoric
occupation and survival from transgression.
While sites dating to the early prehistoric period may have existed on the OCS prior to
inundation, they did not necessarily survive marine transgression. Many variables are at play
with respect to site preservation or destruction, including the type of landform, the site’s position
vis-à-vis a shoreline (i.e., would the site have been in a backshore area that would have
experienced sedimentation rather than direct, erosional wave impacts), and the rate of marine
transgression (with slower rates likely impacting areas more thoroughly). However, even with
faster rates of marine transgression, shoreface locations would have experienced significant
impacts through many years or decades of natural forces, so it is unclear if the rate of marine
transgression would have allowed sites in such locations to survive. Accordingly, it is important
to evaluate locations where sites may have survived on a case-by-case basis, focusing on the
unique characteristics of the local geography and stratigraphy. Current mapping of relict
landforms across the OCS is inadequate for the purpose of identifying potential locations of
preserved sites. Detailed studies of geomorphology using seismic sub-bottom profiling and
coring in conjunction with bathymetry are needed to characterize settings that could hold intact
sites.
9.4. SUMMARY
The process for exploring the potential for submerged prehistoric sites, then, can be
summarized as follows. The first step in launching such an investigation is to obtain information
on the regional and local cultural context and environmental setting. This is accomplished by
consulting with the SHPO, local Tribes, and other interested parties, and conducting a
comprehensive literature search to document the archaeological record on the nearest land
adjacent to the offshore study area and reviewing the available geological literature to obtain an
understanding of the area’s geomorphological history and the current environmental conditions.
The second step is to get a sense of the “lay of the land” underwater by examining bathymetric
charts, or, if available, existing multibeam bathymetric and subbottom records to look for
evidence of potential vestigial elements of the pre-inundated landscape within the study area.
The third step is to apply local relative sea level rise models to the study area’s bathymetry to
attain a general sense of where shorelines may have been located and when various parts of the
inundated landscape could have been subaerially exposed. The fourth step is to develop and
execute a combined program of geophysical survey and geotechnical sampling to identify
archaeologically sensitive landforms and paleosols, which once located, may then be subjected to
different forms of sub-surface archaeological testing to locate archaeological deposits.
As the authors of the Research Planning, Inc. et al. report note (2004:60), the hypothesis that
key relict landforms like stream channels and estuary complexes served as attractors for
prehistoric utilization, and thus have high archaeological sensitivity, remains largely untested
along the Atlantic coast. Intensive subsurface testing in advance of construction and/or
monitoring by an archaeologist during offshore work (particularly dredging operations) would
serve to test the site patterning model. In the case of monitoring, if artifacts were encountered
during construction, work would stop or move to another location until the find could be mapped
152
and verified (i.e., following the protocol discussed in Research Planning, Inc. et al. [2004:59–
60]).
Until findings accumulate to disprove a model for prehistoric settlement along coastal
streams and estuaries, survey efforts should focus on these portions of the submerged landscape.
The methods discussed in the next chapter provide cost-effective and reliable tools for
characterizing the landforms and prehistoric site potential within a given project area.
Archaeological exploration for submerged prehistoric sites on the Atlantic OCS takes on
some urgency in light of the accelerating rate of potential site destruction resulting from offshore
dredging, drilling, and construction activities. Given the significant archaeological questions that
can only be answered through investigation of early sites along the relict coast, it is important to
protect and study these unique components of the North American archaeological record.
There are a number of methods available to search for the appropriate contexts for
submerged sites. The objective of these methods is first to identify areas with reasonable
potential for containing prehistoric sites and, second, to sample targeted areas with high potential
to determine if preserved sites exist. The current state-of-the-art methods for identification and
investigation of submerged prehistoric sites are discussed in Chapter 10. How such methods can
be applied in the context of federally permitted undertakings is covered as well.
153
PAGE LEFT INTENTIONALLY BLANK
154
10. RECOMMENDED FIELD SURVEY METHODS
10.1 INTRODUCTION
Early prehistoric archaeological resources are virtually invisible to remote sensing equipment
available today. However, the association of Paleoindian and Archaic sites with relic landforms
appears to be the key to locating and identifying areas of high potential. There have been few
systematic surveys conducted specifically to locate submerged prehistoric sites in the Atlantic to
date. A notable exception is Robinson et al.’s recent study in Nantucket Sound for an offshore
wind power project (Robinson et al. 2004). Studies carried out elsewhere have illustrated the
value of correlating potential site locations with submerged landscape features. The Sabine River
study carried out by Pearson et al. (1986) over two decades ago and current research carried out
by Faught (2003, 2004) off the Gulf coast of northern Florida provide the most convincing
evidence of the value of that correlation. Likewise, a team from Parks Canada has explored the
continental shelf in the Hecate Strait off British Columbia, where ancient human occupation sites
may rest in as much as 150 m of water. The Canadian team has employed high-resolution
multibeam sonar, remotely operated vehicles (ROVs), and manned submersibles to image the sea
floor, and coring and grab methods to sample it (Carper 2007). In conducting surveys designed
to identify relic landforms and prehistoric archaeological sites, acoustic instruments appear to be
the most effective (Faught 2003; Hoyt et al. 1990; Research Planning et al. 2004).
The three instruments that generate the most useful data are multibeam echo sounders, side
scan sonar, and subbottom profilers. The side scan sonar and multibeam echo sounders generate
high-resolution data that can be used to reconstruct and map surface geological features that
reflect paleotopography. Used in conjunction with highly sophisticated terrain modeling
programs, acoustic data from those instruments can be turned into highly detailed bottom surface
maps that cover broad areas. Characteristics of the bottom surface can be associated with buried
geomorphological features using high-resolution subbottom profilers. With sufficient data,
sophisticated computer modeling programs can be used to develop three-dimensional, georeferenced models of relic landforms that could be associated with areas that have prehistoric
archaeological site potential. Using GIS software to store, analyze, and project the data,
archaeologists and submerged cultural resource managers can identify high priority areas for
research or protection. Areas of high potential where sea floor disturbances are proposed can
then be surveyed using higher resolution geophysical techniques (like seismic reflection profiling
studies), coring, and direct observation of the sea floor using remotely operated vehicles (ROV)
or direct submersible investigation. Intensive studies of submerged cultural resources will be
expensive, and developers may choose to avoid areas of high potential, rather than carry out
costly investigations.
Each of the methods available to characterize the sea floor and identify areas of high
potential for cultural resources are described below, along with methods for sampling and
investigating such areas. There is also a brief discussion of planning considerations related to the
cost and logistics of conducting such studies.
155
10.2. UNDERWATER SURVEY METHODS
10.2.1 Multibeam Bathymetry and Backscatter Intensity Data
One remote sensing method relevant to detecting areas of high sensitivity for prehistoric sites
is high-resolution multibeam swath bathymetry (where the data set consists of both depth and
backscatter/reflectivity information) to image surficial features on the sea floor. This method
allows the identification of relict landscape features such as stream channels along which
prehistoric sites would have been concentrated. Multibeam bathymetry and backscatter intensity
data provide information on water depth, sea floor morphology, and sediment types. Multibeam
systems are so-named because they consist of a group of sonar beams directed at and reflected
back from the ocean floor, as opposed to earlier, single beam systems. Bathymetric data and sea
floor composition are interpreted from the speed and intensity of the reflection of the acoustic
signals, which are collected simultaneously and then processed. Multibeam systems collect data
in a swath that typically extends beyond either side of the host vessel along the ship’s track to a
distance of five to seven times the depth. Ship tracks are designed to overlap and provide 150
percent coverage of the study area. These tracks are then combined to form a seamless image of
the morphology of the ocean floor, as well as detailed bathymetric data. Because wider swaths
are gathered in deeper water, surveys are much faster in greater depths.
Multibeam bathymetry and backscatter intensity data is the first information that should be
collected during a survey for submerged cultural resources. The bathymetric data provides a
detailed image of sea floor morphology, allowing identification of landforms and an accurate
assessment of depths within the study area. Backscatter data can provide generalized information
on sea floor bottom types, based on the intensity of acoustic returns. When combined, these two
data sets establish the basis for more detailed studies of the sea floor and underlying stratigraphy.
10.2.2. Side Scan Sonar
Side scan sonar is also an acoustic technique, but is focused on a detailed image of sea bed
characteristics rather than bathymetry. This technique also can be used to identify shipwrecks,
but in the context of prehistoric site survey, it can serve to characterize the sea floor with greater
resolution than multibeam bathymetry. Side scan sonar is accomplished using a towfish that both
sends and receives acoustic signals and reflections from the sea floor. As in multibeam surveys,
side scan sonar surveys image swaths of the sea floor several times the water depth. Ship tracks
are designed to overlap and provide 150 percent coverage of the study area, allowing production
of maps showing sea floor characteristics. When combined with multibeam bathymetric data, a
great deal of information on the morphology and composition of the sea floor is obtained. This
information is critical to identifying geomorphological settings of high archaeological potential.
10.2.3. Seismic Reflection Profiling
Seismic profiling is a geophysical technique used to gather information about sea floor
subsurface data. This technique also employs acoustic energy, but rather than receiving and
processing returns strictly from the ocean floor, the signals are designed to penetrate subsurface
sediments. Reflections from interfaces between layers of varying acoustic properties are recorded
and used to create a seismic-stratigraphic profile of the material beneath the ocean floor. The
156
depth of penetration into seafloor sediments is determined by the frequency of the acoustic signal
and the sediment characteristics. Higher frequency (CHIRP) systems provide greater resolution,
but less depth penetration, and provide excellent results in settings with fine-grained sediments.
Lower frequency (Boomer) systems produce greater penetration of thick sediment sequences, but
generally with less resolution.
Seismic reflection data is produced as a series of 2-dimensional profiles along the research
vessel’s tracks, unlike the 100 percent coverage that can be achieved with multibeam
bathymetric studies and side scan sonar investigation. Thus, the spacing of seismic reflection
profiles is important if the study area’s stratigraphy is to be adequately investigated. Seismic
reflection profiles are frequently collected using gridded cruise tracks (lines oriented at right
angles), with the spacing between lines determined by the approximate size of landforms or
buried features to be imaged. Data from multibeam bathymetric studies, as well as any previous
work in the study area can be used to guide this decision. More closely spaced data collection,
with a maximum lane spacing of 15 m, may be used to further refine interpretations in areas
identified as having a high potential for cultural resources. Prominent acoustic reflections that
occur throughout a study area can be selected in some processing systems and a surface of that
reflector can be interpolated and the thickness of overlying sediment mapped.
The complementary properties of these two seismic reflection techniques indicate that both
should be used in a survey for submerged prehistoric cultural resources. The higher frequency
data will provide higher resolution data of near bottom stratigraphy, while the lower frequency
technique will investigate more of the subsurface stratigraphic package. While most culturally
sensitive areas may be concentrated in the upper portion of the subsurface sediments, it is
difficult to understand the geologic history and setting of the study area without seeing as much
of the section as possible. In addition, this information is routinely collected for engineering
studies for offshore projects. With advance planning, survey for culturally sensitive areas can be
accomplished at the same time geotechnical and engineering information is collected, reducing
costs.
10.2.4. Vibracoring
Vibracoring may be required for the analysis of high potential geomorphic settings, to allow
further analysis of the seabed subsurface geology. While it is highly unlikely that artifacts will be
recovered by vibracoring, the sediments and faunal and floral remains obtained provide
information about the physical setting and age of the area. A geotechnical program of
vibracoring also can determine the presence or absence of paleosols likely associated with
prehistoric occupation. This information can then be used to further assess a study area’s cultural
resource potential. Vibracores previously taken in portions of the Atlantic sea floor suggest that
the top 1 m (and sometimes deeper) of sediments are recent and/or reworked (LaPorta et al.
1999; Schuldenrein et al. 2000). However, it is possible that intact former land surfaces that may
contain prehistoric archaeological deposits are buried beneath the sea floor. If proposed seafloor
impacts will disturb more than the top meter of sediment, it is recommended that vibracoring (or
similar method of coring) be undertaken in areas of moderate to high potential for the presence
of prehistoric sites. The goal of vibracoring would be to determine if there are intact Late
Pleistocene and Holocene strata in areas slated for impact. Analysis of the vibracore samples
157
would consist of lithostratigraphic evaluation, dating of any organic material, and identification
of any pollen, macrofloral, and/or foraminiferal samples recovered. If intact strata are identified,
then it is recommended that those areas be avoided. If avoidance is not possible, then more
subsurface testing and/or monitoring to determine if prehistoric materials are present may be
recommended.
10.2.5. Remotely Operated Vehicles (ROVs), Autonomous Underwater Vehicles
(AUVs), Video Surveys and Submersibles
Ground-truthing of high sensitivity areas identified by remote sensing that lie within an area
of proposed impact is typically done by vibracoring, although in cases where surficial deposits
are suspected (e.g., around rock outcrops), then it may be accomplished by direct visualization
by scuba divers or by ROVs, depending on the bottom conditions (e.g., depth, currents,
visibility). These methods are also used to investigate areas once cultural resources have been
identified at the seabed surface. ROVs act as the eyes, and sometimes hands, of the investigators.
They are, however, limited to material exposed at the seafloor. The equipment is operated
tethered from a vessel. A ROV will allow investigation of seabed conditions, visual analysis of
features (like rock outcrops, shipwrecks, etc.), and inspection of exposed artifacts. Use of ROVs
is restricted by water clarity. Fine-grained bottom sediments can create turbid conditions that
greatly reduce visibility. AUVs are programmed to “fly” over the bottom and can be equipped
with cameras and a variety of geophysical sensors. In locations like the Gulf of Maine, with a
large lobstering industry, lobster buoys may preclude use of AUVs.
Video surveys with a towed camera can provide detailed color images of the seabed capable
of imaging artifacts and seafloor sediment. These surveys acquire a series of overlapping images
along a transect of the seabed. Since it is difficult to know the precise position of the camera for
every frame, transects are often short.
Submersible vehicles provide a way for scientists to make direct observations at the seafloor,
and in some situations, collect samples. As with ROV’s, water clarity can create visibility issues
for studies employing submersibles. Submersible vehicles are expensive to build, maintain, and
operate, so costs associated with this type of investigation are high.
10.2.6. Geophysical Survey Planning
Initial survey to identify high potential areas for submerged cultural resources requires some
of the same information and employs many of the same techniques as those used by the offshore
development applicant. Thus, the multibeam bathymetry and backscatter intensity data, side scan
sonar, and high resolution (CHIRP) and deep penetration (Boomer) seismic reflection profiling,
as well as precision mapping carried out for other aspects of project planning can also serve the
needs of cultural resource assessment, with data collected simultaneously that will serve a variety
of needs. Depending on the size of the research vessel and project budget, seismic reflection,
multibeam and side scan sonar profiles can usually be collected simultaneously. Generally,
multibeam data can be gathered at a higher vessel speed than the other techniques, and if such a
system is leased, it is sometimes more cost effective to collect bathymetric data first and use it to
plan seismic and side scan sonar lines. Interferometric side scan sonar methods additionally
provide good quality side scan images and bathymetric data, especially in shallow water. More
158
detailed, higher resolution surveys should be reserved for examination of identified submerged
cultural resources, such as shipwrecks and areas of high prehistoric archaeological potential.
Even when investigations are carried out in cooperation with project engineers, the work
should be performed under the supervision of a marine archaeologist, with marine archaeological
staff on board the survey vessel for the duration of the survey to monitor data as it is acquired.
This arrangement should allow the archaeologist to generate a preliminary “real-time” inventory
of acoustic reflectors with moderate to high potential for representing archaeologically sensitive
inundated paleosols. Upon completion of the field investigation and post-processing and plotting
of the survey data, acoustic reflectors identified by the field archaeologist as having moderate to
high potential for representing archaeologically sensitive areas should be reevaluated by the
archaeologist using the post-processed data in combination with core logs and photographs from
any geotechnical coring/boring performed as part of the project. The results of these combined
analyses should then be used to generate a final list of archaeologically sensitive areas
recommended for avoidance or further investigation and National Register evaluation.
Specific guidelines for remote sensing surveys updating current BOEMRE protocols are
provided in Research Planning, Inc. et al. (2004:35–39, 53). They recommend the use of submeter differential global positioning systems for navigational accuracy, acoustic positioning
systems that track towed sensor position, a track line spacing no greater than 30 m, and lines for
anomaly definition spaced 10 m on either side of initial contact.
Following these updated guidelines is likely to result in the discovery of more archaeological
sites (both prehistoric and historic period) than would have been identified under the old
standards, thus possibly preventing future incidents of accidental site disturbance during
construction.
10.3. SUMMARY
Investigation of the Douglass Beach Site (8FL17) in Florida state waters illustrates the types
of analyses possible in the context of underwater prehistoric sites, analyses that are commonly
employed at terrestrial sites (Murphy 1990). In addition to radiocarbon dating of organic
materials recovered, sedimentary and geochemical analyses can be employed to understand
taphonomy and identify the signatures of human occupation in sea floor sediments (to help refine
expectations about evidence of archaeological deposits elsewhere), palynological analysis can be
conducted to assist in environmental reconstruction, ethnobotanical and faunal analyses can be
carried out on materials whose preservation state may be enhanced by submersion, and artifacts
and their provenience can be analyzed as is done for terrestrial sites, although stratigraphic
recovery is limited to approximate strata through propeller wash deflector modifications, and
small samples obtained through coring. The information potential of submerged sites is
comparable to those on land, and could be key to our understanding of the peopling of North
America and coastal adaptations in the early millennia of human occupation. The Douglass
Beach Site was preserved in a back barrier setting, where it was buried by overwash sediments
during transgression, protecting it from high-energy shoreface erosion (Murphy 1990:52). Sites
in comparable settings likely exist throughout the Atlantic OCS, and await discovery through the
survey methods discussed here.
159
160
PAGE LEFT INTENTIONALLY BLANK
161
SECTION 4 – HISTORIC SHIPPING AND SHIPWRECKS
162
PAGE LEFT INTENTIONALLY BLANK
163
11. ATLANTIC OCS SHIPWRECK DATABASE
11.1. INTRODUCTION
One of the goals of this project was to compile a database of shipwrecks within the Atlantic
OCS, using existing databases, sources, and archival research. This chapter describes the efforts
that went into developing the Atlantic OCS Shipwreck Database (ASD). The collected data, both
existing and new, were incorporated into a Microsoft Access database modeled on the one
generated for shipwrecks in the Gulf of Mexico in 2003 by Panamerican Consultants and Coastal
Environments, Inc. (Pearson et al. 2003). Although the original intent of the database was to
cover only those wrecks found within the boundaries of the Atlantic OCS, wreck locations that
are within state waters were left in the database rather than purged, since it seemed more sensible
to keep the data once collected.
The remainder of this chapter discusses the methods and sources used to compile the ASD. In
all, there were 33,851 entries placed into the database through these efforts. Appendix 1 lists the
sources used in data collection, all of which are also referenced in the ASD for the appropriate
entries.
11.2. METHODS AND SOURCES
For each source consulted, a Data Source Form (DSF) was completed. The form notes the
location of the source, the type of source (published, online, or archival), and other descriptive
information. A Data Source Form was completed even if the source did not provide any relevant
information, which will help future researchers concentrate on the most useful sources. For each
shipwreck noted in the sources, an Archaeological Resource Information Form (ARIF) was
created to record the data. The forms included spaces for all of the variables in the 2003
database, including a space for notes. The forms were numbered sequentially as they were filled
out to provide an identification number. Previously assigned MMS numbers are reported under
the field “Previous Survey Number.” Identification numbers assigned in other databases were not
retained, but the source from which the entry derived is noted in the Source and/or Comment
field of the ASD.
The sources can be divided into three major types: primary sources, secondary sources, and
existing database entries. There is considerable overlap among these sources, since secondary
sources derived information largely from primary documents, and existing commercial and
government databases used primary and secondary documents as their source of information,
along with actual reports from mariners, divers, and surveyors. Some sources were checked
against the larger databases to determine if the source had already been inventoried. It was
expected that some sources would not need to be recorded, since they already would have been
entered into an existing database. This was rarely the case, however, as previously compiled
databases did not always record all entries or all of the information provided in the sources. New
ARIFs were completed for these missing or incomplete entries.
Conversely, information on shipwrecks that were already listed in existing databases was not
recorded, unless the new source differed significantly from the existing entry. Slight variations in
164
location were not considered significant enough to add to the existing data, but large deviations
in location, or inconsistencies regarding ship size, type, or cargo were recorded on new ARIFs
and the source of the information noted. It should not be assumed that all information from all
sources was recorded for each vessel. Time constraints sometimes limited the data that could be
effectively recorded.
In general, photographs could not be collected from most sources because of copyright
issues, copy fees, or time retraints. Instead, an effort was made to note the existence of photos in
the record, by marking it in the vessel photo field of the ARIF. Photographs and images also
exist for many vessels in separate collections at museums and archives. The Mariner’s Museum
in Newport News has an extensive photograph and image collection, but even creating a list of
sunken vessels with images on file was determined to be too time consuming for the current
inventory. A partial search of the index was conducted for steamships recorded in the ASD from
primary sources, and a note was made on the forms of those with images on file. Based on this
search, however, it was determined that checking all of the ASD would require 30–40 hours,
without even viewing the images. Obtaining catalog and/or citation information for available
images would likely double that time.
For the most part, if a source cited a shipwreck that appeared to be located in the Atlantic
OCS project area, it was entered into the database with the information provided in the source.
Rather than combine data from apparent duplicates, all entries were retained to minimize the
chance of losing data that might prove useful. For example, for sources reporting different
location information or details regarding the vessel, it is difficult to determine without specific
research which source is the more accurate. In addition, certain vessel names were common, and
thus combining entries risked eliminating a unique entry from the database or assigning
erroneous information to another vessel. Combining entries and referencing all sources would
have resulted in a loss of information and decreased utility in the sort function of the database. In
fields that only allow one value to be entered, alternate values would have to be dropped, and in
text fields, entries could only be sorted by the first value. Thus with the possibility of multiple
entries, many vessels are listed more than once under different sources. This approach preserves
all of the source information and allows the researcher to parse the data based on its source using
the ship name or other identifying information.
The existing databases were supplemented with shipwreck locations from secondary sources,
diving books, and websites. Many of these provided Loran numbers that were converted to
Lat/Long coordinates so that they would be GIS compatible. Andren LoranGPS© software
(Version 7.3) was used to make the conversion. The major secondary sources, such as Berman’s
Encyclopedia of American Shipwrecks and Marx’s Shipwrecks of the Western Hemisphere,
were generally already included in existing databases and provided only the vaguest location
information. However, a number of these types of published inventories were consulted and the
information transferred to the database.
The information from primary sources varied considerably depending on the time period of
the shipwreck. For the earliest period of exploration and settlement, roughly defined as the 16th
and 17th centuries, primary sources such as colonial records, ships’ logs, and first-hand accounts
are widely scattered and difficult to use, particularly with the large geographic area encompassed
165
by the Atlantic OCS. For the most part, these sources provide only the broadest location
information, such as “off Cape Cod,” or “Hurricane Shoals.” Many are even less specific than
that, stating only that a ship was lost “off the coast of North Carolina,” “at sea,” or “hasn’t been
heard from.” If information on origin and destination is provided, it gives some indication of
where a wreck might be, but searching for the wreck based on that information alone is
impossible. Over time, sources provided more specific location information. Government
reports, which begin just before the Civil War with the formation of the U.S. Life-Saving
Service, generally give location as a distance and heading from a known landmark. Still, even
these reports are often vague, in many cases because the exact location was not documented at
the time, or there were no witnesses or survivors to provide the information.
The accuracy of location information is quantified in the ASD by the “Location Reliability”
field. This ranking from 1 to 4 is based on the system devised for the 2003 Gulf of Mexico
shipwreck inventory (Pearson et al. 2003). Shipwrecks that have been positively located through
recent survey are given a location reliability rank of 1. Those shipwrecks with specific locations
provided by informants, reported in literature, or marked on a map are considered a 2. A location
reliability of 3 indicates that the location is given generally rather than specifically by an
informant, in the literature, or on a map. Those locations that are unreliable or vague, such as
“off the coast of North Carolina” or “at sea” are ranked at 4.
The data fields developed for the 2003 inventory provided a well-developed framework for
the current inventory, and in general the coding conventions used in that list were used for the
ASD. Some additional codes for “Cause of Loss,” “Vessel Type,” “Where Built,” and
“Nationality” were added, as necessary. These followed the two- and three-letter coding
conventions used by Pearson et al. (2003
The sources used to compile the ASD are discussed below by category and tabulated in
Appendix A with the number of entries gleaned from each.
11.1.1. Existing Shipwreck Databases
Existing governmental databases formed the core of the data for the current BOEMRE
Shipwreck Database. The National Oceanic and Atmospheric Administration (NOAA) maintains
the Automated Wreck and Obstructions Information System (AWOIS), a database of wrecks and
obstructions compiled from hydrographic surveys and field reports. The database is updated
when new surveys are conducted (about 30–50 per year) and new wrecks are only added after
they have been surveyed. Because the information is based on survey, contains detailed location
information, and is updated regularly, it is one of the more reliable comprehensive databases,
although older entries are often inaccurate. Unfortunately, AWOIS often lacks identifying and
descriptive information on its listings.
The U.S. Navy created the Non-Submarine Contact List (NSC) for military use in
distinguishing shipwrecks from submarines hiding on the ocean floor. The inventory contains a
large number of objects that are not shipwrecks, but debris, seafloor pinnacles, and other
features. The list is maintained by the National Geospatial-Intelligence Agency (NGA), a
Department of Defense Agency that supports the National Intelligence Community through
imagery and map-based intelligence for national defense, homeland security and safety of
166
navigation. The NSC data obtained for this project included shape files for ons of objects; they
were added to the project GIS, after entries outside the OCS had been culled.
The U.S. Navy also maintains a database entitled Partial List of Foundered U.S. Navy Craft.
Ships from this source were added to the database as well.
Three commercial databases were also obtained as part of the current effort: The Global
Maritime Wrecks Database, the International Registry of Sunken Ships, and the Northern
Shipwrecks database. The information from these inventories is copyrighted, and thus was not
included in the ASD. However, the publications are now part of the BOEMRE’s holdings, and
the information is available to qualified agency researchers. The Global Maritime Wrecks
Database (GMWD) is published by General Dynamics Advanced Information Systems and
includes more than 250,000 shipwreck locations worldwide. The International Registry of
Sunken Ships (IRSS) is a commercial database compiled by Hugh Brown of Saskatchewan,
Canada (Brown 2008). It uses many of the sources cited here, and proved to be reasonably
complete and accurate. The data is organized by state and includes numerous wrecks not in
federal waters. A list of wrecks that appear to be within the OCS boundary was generated from
the master list as part of the current research effort. For these two databases, all wrecks with
coordinate information that falls within the Atlantic OCS were compiled, and their locations
were plotted using GIS software as part of the location analysis. There were 1,364 wrecks in the
IRSS and 3,505 wrecks in the GMWD that had coordinates within the project area. The Northern
Shipwrecks Database, available from Northern Maritime Research of British Columbia proved
cumbersome, since the data could not be copied or organized into reports by variables such as
location; therefore it was not used in the location analysis.
All entries from existing databases for which locational information was available were
projected in ArcGIS and those that were located within the Atlantic OCS were incorporated into
the database. To create shape files, X-Y coordinates in Lat/Long decimal degree format were
used to generate points for each shipwreck. Any coordinates provided in Lat/Long decimal
degree-minutes or degrees-minutes-seconds formats were converted into Lat/Long decimal
degrees using the U.S. Army’s Corpscon software (Version 6.0). Where locational information
from an original source was descriptive rather than offering actual coordinates (in the vein of “5
miles east of Cape Hatteras”) approximate coordinates were obtained from the GIS by measuring
to the distance and direction described.
11.2. U.S. GOVERNMENT DOCUMENTS
11.2.1. U.S. Coast Guard, Record Group 26
The U.S. Coast Guard had its origins in the Revenue Marine Service, later the Revenue
Cutter Service, a branch of the Treasury, which was established in 1790 to enforce tariffs and
trade laws and prevent smuggling. Other agencies were created during the 19th century to
improve navigation, provide assistance to mariners, and enforce maritime regulations, including
the U.S. Lighthouse Service, U.S. Life-Saving Service, and Steamboat Inspection Service (later
the Bureau of Marine Navigation and Inspection). In 1915, the Revenue Cutter Service and LifeSaving Service were merged to form the U.S. Coast Guard. In 1938, the Lighthouse Service also
became a part of the Coast Guard (U.S. Coast Guard Historian’s Office 2008a). In 1946, the
167
Coast Guard assumed most of the duties of the Bureau of Marine Inspection and Navigation. All
of these agencies generated records related to shipwrecks, most of which are housed in National
Archives repositories in Washington and at regional facilities.
Records of the U.S. Coast Guard and its two predecessors, the Life-Saving Service and
Revenue Cutter Service, are part of Record Group 26 in the National Archives. Records of the
U.S. Life-Saving Service include Life-Saving Station logbooks from various stations on the
Eastern Seaboard dating to the 1870s and 1880s, as well as Reports of Assistance Rendered from
the 1880s until the creation of the Coast Guard in 1915. These reports are held at the regional
archives branches for stations in that region. Because they cover incidents handled by shorebased Life-Saving Stations, these reports largely cover wrecks that occurred within 3 miles of
shore. Nevertheless, extant reports located at the branch archives in Boston (Waltham,
Massachusetts), New York City, and Atlanta (Morrow, Georgia), were examined and
approximately 391 entries were added from these sources. It was determined from a preliminary
search in the Life-Saving Station records at the archives branch in Philadelphia that the time
required to review the records would not be productive because most of the entries were for nearshore wrecks and wrecks in inland waters such as Delaware Bay.
For the period 1913–1939, the wreck reports provided to the Coast Guard are indexed on
microfilm as U.S. Coast Guard Casualty and Wreck Reports (National Archives Microfilm
Publication T926). The card index was compiled by the Works Progress Administration and
includes enough information that an examination of the actual report is unnecessary. Time
constraints precluded a complete review of the index; approximately 32 hours were spent
reviewing the first five-and-a-half years of the index for shipwrecks on the OCS by using the
stamps on the card that noted general location (“Atlantic”) and loss to vessel (“Total Loss” for
sunken or destroyed vessels). Wrecks with specific location information that indicated near shore
waters, such as “New York Harbor,” “Pamlico Sound,” or “Chesapeake Bay,” were not recorded.
For all potential shipwrecks on the OCS, all of the information from the cards was recorded.
From that review, 128 entries were added to the database. This is a good source of information
on wrecks for the early twentieth century, and likely would yield a significant number of
additional entries for the period between World War I and World War II, including war losses of
merchant vessels. The wreck reports appear to be the source for the annual list of American
Vessels Lost, published from 1903 to the present in Merchant Vessels of the United States,
compiled by the Coast Guard and its predescessors. Those listings are included in the ASD
through 1923, after which the location information included in the published volumes becomes
so vague that it cannot be determined whether wrecks were in off-shore waters.
The Life-Saving Service Annual Reports for the years 1874–1914 include tables of casualties
that provide their location in relation to the nearest station. The locations in the annual reports
are not as specific as those in the wreck reports themselves, so the wreck reports were consulted
wherever possible.
The Coast Guard also maintains Disaster Files at its headquarters in Washington, D.C.,
which include incident reports, communications transcripts, newspaper clippings, and
photographs of incidents involving the Coast Guard. These files mostly concern search and
rescue operations during the mid to late 20th century, with some information on early 20th century
168
shipwrecks, particularly high-profile losses. In a number of cases, the rescue operations involved
a sunken or sinking vessel, and the communications records often document the latitude and
longitude of where the ship was lost. These files were examined in the Coast Guard Historian’s
Office to identify shipwreck locations.
11.2.2. U.S. Customs Service, Record Group 36
In 1789, the Congress authorized the creation of the U.S. Customs Service to collect tariffs
on imports as a source of revenue for the fledgling federal government. The Customs Service
was also charged with registering and licensing American vessels and enforcing maritime laws
and regulations, including the entry of seamen and passengers to U.S. ports. Customs offices
were established in 59 locations in 11 states (Stein 1992). In 1790, Secretary of the Treasury
Alexander Hamilton approved the establishment of a fleet of 10 cutter ships to assist in the
enforcement of customs regulations. The spartan ships were deployed to the federal customs
offices on the eastern seaboard (Ross 1886).
Beginning in 1874 as a result of Congressional legislation, customs districts were required by
law to report any instance of shipwrecks that resulted in property damage over $300, loss of life,
or injuries. File copies of these reports were maintained at the Custom Houses, and copies were
sent to various agencies and officials at different times depending on administrative
responsibilities. The agencies included the Life-Saving Service and the Coast Guard, and these
copies constitute the wreck reports in the records of the U.S. Coast Guard (RG 26). The National
Archives region branches have the reports from the Customs Districts in their region, and these
are part of RG 36. The U.S. Customs wreck reports were examined for all of the reporting
customs houses on the eastern seaboard.
11.2.3.
Bureau of Marine Inspection and Navigation, Record Group 41
The Steamboat Inspection Service maintained reports of Casualties and Violations of Law,
many of which contain information on vessels lost. Summaries of these reports were compiled
into the annual “Proceedings of the Board of Supervising Inspectors of Steam Vessels” from
1852–1899, and “Annual Report of the Supervising Inspector-General, Steamboat Inspection
Service” from 1895–1931. Incidents involving vessels that sank and that appeared to have
occurred in open water or more than 3 miles from shore were recorded and added to the
database.
Beginning in 1906, the Bureau of Navigation’s yearly publication Merchant Vessels of the
United States listed American vessels lost, with information on the size and type of vessel,
number of lives lost, the nature of the accident, the date, and the general location. The loss lists
for the years 1903–1915 were compiled in volumes at the NARA, and the books are available in
the NARA Finding Aids Room for the years 1917–1960. The location information in these lists
was generally not specific enough to determine if the vessel was in federal waters, and since
many of the losses are reported in more detail in other sources, these were not entered into the
database separately.
169
11.3. STATE AND FEDERAL AGENCIES
11.3.1. U.S. Army Corps of Engineers
United States Army Corps of Engineers (USACE) records in the National Archives include
Wreck and Obstruction Files documenting reports of obstructions affecting navigation or other
maritime activity. These files document the identification, monitoring, and disposal of the
obstruction by the USACE.
Corps of Engineers district offices were contacted to determine if reports or surveys related
to shipwrecks were available at those facilities. The New England District Office in Concord,
Massachusetts, has Harbor Commission reports for the 19th century and at least one report that
pertains to a shipwreck, but the contact there indicated that these would concern near shore
resources. Offices in New York, Baltimore, Norfolk, Wilmington, Charleston, Savannah, and
Jacksonville reported that they had no information that would be of use to the current project,
since their reports dealt almost exclusively with waters within 3 miles of shore.
11.3.2. State Historic Preservation Offices
State Historic Preservation Offices in all of the states on the Atlantic Seaboard were
contacted for information they might have on shipwrecks in or near federal waters (Pennsylvania
and Connecticut were not included since their state waters do not abut federal jurisdiction).
Although most states reported keeping a database of shipwrecks of some sort, most include few
if any listings for vessels in federal waters.
Officials in Maine, Virginia, North Carolina, and Florida provided database information on
shipwrecks in or believed to be in federal waters off the coasts of their states. Massachusetts,
Maryland, South Carolina, and Georgia did not maintain a database, but provided previous
inventories and reports that included shipwreck data and bibliographic information. Where this
information was sufficiently specific and did not duplicate other sources, the data was entered
into the inventory.
New Jersey and New York reported having no information on wrecks in federal waters that
would be of use to the project. The Delaware SHPO did not respond in time for inclusion in the
report.
Christopher Amer of the South Carolina Institute of Archaeology and Anthropology provided
information on USS Hector, a navy collier of over 11,000 tons that sank 10 miles off Cape
Romaine in a violent gale in July 1916. He also provided data on U.S. Navy vessels believed to
be in South Carolina waters from a report compiled by his office.
Florida’s Department of Historical Resources, Bureau of Archaeological Research maintains
a site file documenting archaeological sites in state waters. A review by Vincent Birdsong,
Database Administrator of the Florida Master Site File found seven shipwreck sites that lie
outside of the 3-mile state waters boundary. Those seven sites are included in the inventory.
170
11.4. PUBLISHED SOURCES AND CONTEMPORARY DOCUMENTS
11.4.1. Newspapers and Magazines
Trade papers for the shipping and trade industries for the 19th century are an excellent source
of information for lost vessels in the period prior to systematic record keeping by government
agencies. The Library of Congress has a fairly complete collection of the New York Shipping and
Commercial List (NYSCL) for the years 1815–1832. The earliest editions do not list disasters
separately, so these were scanned for entries related to lost ships. Beginning in 1818, a special
section lists disasters, including losses of life, damages to vessels, and other incidents that did not
result in a vessel being lost. Those entries that appeared to be relevant to potential shipwrecks
within the project area were recorded and added to the database for the period 1818–1830. John
L. Lochhead abstracted wreck information from the NYSCL, which is published for the MidAtlantic states in Joan Charles’ (2003, 2004a, 2004b) shipwreck account books for the period
1821–1846. Although most of the entries in the NYSCL have only the vaguest location
information, it may prove useful for matching ship names with existing but unidentified wrecks
in the database.
The Shipping and Mercantile Gazette is a daily newspaper published in London in the 18th
and 19th centuries that contains information on maritime disasters. The Library of Congress has a
significant run of the Gazette, but because it was published daily and covers shipping worldwide,
it was determined that it was not a productive source of shipwreck information for the Atlantic
Seaboard. About 7 months worth of issues were examined but only 9 wrecks were identified in 7
hours of work.
Besides these two sources, no other newspapers or magazines were systematically examined.
The process was considered to be too time consuming relative to the results. However, a number
of shipwrecks were identified from periodical sources through online searches, from clippings
files in manuscript libraries, and from secondary sources. The Mariner’s Museum in Newport
News and the Boston Public Library and Massachusetts Historical Society in Boston all had
collected material on shipwrecks from contemporary periodicals. For the Mid-Atlantic states,
Joan Charles’ books of shipwreck accounts mine a number of newspapers for information on lost
vessels for the antebellum period (Charles 1997, 1999, 2003, 2004a, 2004b).
11.4.2. Lloyd’s Lists
Lloyd’s of London published a weekly newspaper of maritime information beginning in 1734
that included information on losses. There is an index to losses prior to 1838, as well as yearly
indexes from 1838 by ship name, but these indexes are not readily available in the United States
and are of limited use for locating wrecks by geographic location. The Guildhall Library in
London holds a number of Lloyd’s publications and manuscripts that report losses, including
Weekly Shipping Index (1880–1920), Weekly Casualty Reports (1920–present), Loss and
Casualty Books (1837–1998), Wreck Returns (1900–1990), and War Losses (1914–1918, 1939–
1945). Reviewing these sources was impractical under the scope of work, however, since it
would require a trip to England.
171
The Mariner’s Museum in Newport News, Virginia, holds microfilm copies of Lloyd’s List
from 1740–1854, but they are not indexed. Editions for 1762 were examined at the University of
Georgia library, which turned up very limited information. A number of entries from Lloyd’s
List are included in Charles’ shipwreck account books for the Mid-Atlantic states mentioned
above (Charles 1997, 1999, 2003, 2004a, 2004b), and these are included in the inventory.
11.4.3. Business Records
Modern marine insurance dates to the early 17th century when English law established
separate Courts of Assurance and developed standardized legal language for insurance contracts.
A search of manuscript collections specializing in maritime records located a number of records
from marine insurance companies, but these generally did not include individual “claims” on
insured properties that would provide data on shipwrecks.
Records of merchants and ship owners were also examined, with only a few returning
information on shipping losses. Because the records were not concerned specifically with losses,
the time required to sort through the records was prohibitive, and the information was likely to
be available in other sources. They did not contain helpful information on locations of lost ships.
In general, business and insurance records were not found to be an efficient source of
information on shipwrecks. However, in the case of individual wrecks, these sources could be
very significant, providing information on the date of construction, size, and other features of the
vessel, its cargo, crew, origin and destination, and other data.
11.4.4. Published and Manuscript Accounts
Prior to the 19th century when regular newspaper accounts are available and government
records were kept for statistical purposes, data on shipwrecks come largely from written accounts
in diaries, journals, and collected writings. Tales of tragic disasters and survival at sea have been
popular since the beginning of sailing. The Boston Public Library and Massachusetts Historical
Society have a significant number of published accounts of shipwrecks from the 17th, 18th and
19th centuries. These were reviewed for information on wrecks that might not have been included
in other databases.
One of the earliest of these types of accounts found is a volume by Increase Mather, son of
famed minister Cotton Mather, who used the tales of survival to illustrate both God’s terrible
power and His gracious providence. Mather’s An Essay for the Recording of Illustrious
Providences reports Anthony Thatcher’s narrative of a shipwreck off the New England coast that
took the lives of much of his family, as edited by Mather. Many other similar accounts were
written during the period, with varying degrees of accuracy, since the sources for many of the
tales is not known or is taken from secondhand recounting (Sievers 2006). Sievers illustrates
how the ends of the writer could influence the accuracy of the account, which is a significant
drawback to these types of sources. However, given the task of finding isolated references to
shipwrecks in historic newspapers that are not indexed, the compilations proved useful. The
information that they provide should be considered unreliable and only a starting point for
further research.
172
In the first half of the 19th century there seems to have been a surge in the number of
collected volumes of shipwreck accounts, sometimes accompanied by tables listing incidents
within a certain geographic region during a particular time period. These included The Mariner’s
Chronicle by Durrie and Peck (1834), two volumes by Charles Ellms (1836, 1840), and one by
the anonymous “Friend of Mariners” (1823). All of these were consulted during the present
research effort.
In the late 19th century, the Procter Brothers published two volumes of vessel losses from
Gloucester, Massachusetts, and surrounding communities dating back to 1830 (Procter 1873;
Procter Brothers 1882). These losses were primarily to fishing vessels that plied the fishing
banks of the North Atlantic. Although many of the losses were in Canadian waters off the coast
of Nova Scotia and Newfoundland or in the open ocean, a significant number were lost on
George’s Bank, which is located about 135 miles east of Cape Cod and within the Atlantic OCS
and the project area. Although location information is naturally vague, the books are an
important source of information on losses outside of the merchant vessels that were tracked by
insurance companies and bankers. They also provide extensive information on the history of the
fishing industry in New England, statistics on the types of fish caught at different times and
locations, and details of the boats, rigging, and equipment used.
11.4.5. Published Shipwreck Inventories
Interest in shipwreck diving and salvage increased dramatically in the second half of the 20th
century as diving technology became widely available to recreational divers and treasure hunters.
Consequently, compiled volumes of shipwrecks appeared during the period covering the entire
East Coast, as well as specialized volumes for portions of the coast. Those concentrating on
documentary sources for their lists provide a reference for connecting shipwrecks located
through survey with known losses during the historic period. A large number of these were
consulted for the current inventory (Berman 1972; Gardner 1954; Gray 2003; Kimball 2005;
Lonsdale and Kaplan 1964; Marx 1981; Quinn 1979; Rattray 1973; Shomette 2007; Snow 1944).
Diving guides reference known shipwrecks, many unidentified, that can be visited by divers.
In addition to published guides covering different parts of the Atlantic Seaboard, a number of
dive organizations maintain websites that contain information on the location of shipwrecks.
These are valuable sources that often include well-researched background information on the
history of the vessel and the circumstances of its demise. In some cases, precise location
information is withheld, either to guard the site from too much traffic, or to preserve the thrill of
the hunt for subsequent recreational divers. Dive books covering all the states of the Atlantic
Seaboard were consulted in compiling the current inventory, concentrating on those that provide
detailed location information (Aqua Explorers, Inc. 2009; Association of Underwater Explorers
[AUE] 2009; Barnette 2003; Berg and Berg 1991; BFDC 2007; Freitag 1997; Galiano 2009;
Gentile 1990, 1992, 2002, 2003). Locations in these volumes are typically given in LORAN, but
Lat/Long in various formats is also used, depending on the source of the information. All
locations were converted to NAD 83 for the current effort to make them compatible with GIS
and each other.
173
12. HISTORIC SHIPPING AND SHIPWRECKS ON THE ATLANTIC
SEABOARD
This chapter provides a brief overview of historic shipping and shipwrecks on the Atlantic
Seaboard. The purpose of this overview is to provide the historic context for the thousands of
historic shipwrecks present within the Atlantic OCS.
12.1. SEAFARING DURING THE AGE OF EXPLORATION (1000–1600 A.D.)
The earliest European exploration of the coast of North America was made by Norse peoples
as early as 3000 B.P. Viking tales dating to some 250 years after the fact tell of expeditions by
explorer Leif Eriksson that discovered Helluland, Markland, and Vinland. Eriksson and others
led colonizing expeditions to the area he called Vinland, generally believed to be Newfoundland.
The discovery of the Viking settlement at L’Anse aux Meadows in Newfoundland supports this
theory (Nydal 1989; Wallace 2003). None of the colonies lasted more than a few years, and no
further attempts were made to settle the area (McGhee 1984). It is possible that Norse explorers
ventured as far south as New England during this period, but as yet, no unequivocal evidence
exists to support the idea. The discovery of a Viking ship in U.S. waters would be a find of major
significance.
The period of early exploration by European nations beginning with Columbus’ first voyage
in 1492 and continuing until the first permanent English settlement of the North American coast
in the early 17th century was characterized by irregular forays by military parties seeking
exploitable resources, primarily in the Caribbean, Gulf Coast, and Central and South America.
From the first voyage, the process of exploration, exploitation, and colonization resulted in
shipwrecks. Santa Maria grounded on a bank off the north side of Hispaniola on Christmas Eve
1492, becoming the first known European shipwreck in the New World. The majority of the
activity during this period of exploration was undertaken by the Spanish. Englishman Henry
Cabot explored the Canadian coast in 1497, but did not return from a subsequent voyage the
following year, discouraging the English from significant colonization efforts until the early 17th
century. There is evidence that Portuguese and French cod fishermen were aware of the fisheries
off the coast of Newfoundland, possibly before Columbus’ voyages, but did little to document
their experiences (Keith 1988:47).
The Line of Demarcation, established by the Treaty of Tordesillas in 1494, gave Spain the
upper hand in the exploitation of the Americas, granting everything west of the line to them and
everything east of the line to Portugal, leaving only Brazil in the western hemisphere within the
Portuguese colonial sphere. Columbus led three more voyages to the Caribbean and South
America, the last in 1502–1503. He had been granted exclusive rights to the Americas, but after
his third voyage, the Spanish government began to authorize other explorers. Four expeditions
were sanctioned in 1499, with at least seven more over the next five years. These exploratory
voyages were comprised of fleets of as many as 20 ships. In many cases the actual number of
ships was not documented, since only the principal ships were discussed in the records. From
1499 to 1520, at least 50 ships were lost in the Americas (Keith 1988:50, 66–67). According to
Keith, besides the four ships of John Cabot that did not return from an expedition to the coast of
New England, it is not believed that any of these vessels were lost on the Atlantic Seaboard. The
174
earliest known loss on the Atlantic Seaboard was an unidentified ship in the fleet of Spanish
explorer Vasquez de Ayllon that was lost off Cape Romain, South Carolina, in 1520. Ayllon also
lost a caravel off of Cape St. Helen, believed to be around the mouth of the Savannah River, in
1525 (Marx 1981:195).
In 1513, Ponce de Leon found his way to Florida, bringing attention to the Gulf Stream
flowing from the Gulf of Mexico northward between the Bahamas and Florida and providing
swift passage back to Spain. Following Cortes’ conquest of Mexico in 1519, the systematic
plundering of its riches began, and Spanish fleets began to regularly use the Straits of Florida for
the voyage back to Spain, a practice that would continue for nearly 300 years. Loaded with gold
and silver, the ships would depart from Spanish outposts in Central and South America and
rendezvous at the colonial settlement of Havana on the north coast of Cuba, then continue on to
Spain via the Gulf Stream. Their course took them past reefs and shoals along the Florida Keys
and the North Carolina coast, where navigational errors or hurricane winds could send them
crashing onto shore or cause them to founder and sink. Using primitive navigation tools and
driven by greed to overload their boats and push their luck, Spain lost an estimated 5–12 percent
of its fleet yearly during the period from 1500 to 1700 (SAI 1981:III-26).
The Spanish merchant vessel San Anton, a 100-ton caravel under the command of Gonzalo
Rodriguez was lost in the Florida Keys in 1521, perhaps the first of many Spanish “treasure
ships” to be lost off the Atlantic coast while sailing from Cuba to Spain. The lure of these
treasure wrecks has contributed to extensive research on shipwrecks of the period, not all of
which is made public by the investigators. The research has resulted in the discovery of a number
of these ships, including Nuestra Senora de Atocha, located by Mel Fischer in 1985, and the
ships of the 1715 and 1733 disasters that sent dozens of ships down in the Florida Keys and near
Cape Canaveral respectively (Barnette 2003; Marx 1987). These discoveries have further fueled
treasure hunting divers, who often focus on recovery at the expense of scientific investigation.
Spain dominated trade in the New World during the 16th and 17th centuries, although the
wealth arriving each year in Madrid prompted its rivals England and France to begin exploring
the New World as well, in an effort to break the Spanish stranglehold on trade. In addition, the
vast treasures crossing the Atlantic made alluring targets for naval ships and privateers seeking to
disrupt trade and capture bounties. Each new war in Europe led to increased piracy and threats to
ships sailing in the Western Hemisphere. Beginning in the late 16th century, English captains of
private merchant vessels were granted letters of marquee that permitted them to plunder Spanish
ships as privateers. Second cousins John Hawkins and Sir Francis Drake were among the British
sea captains who engaged in piracy against the Spanish fleet in the second half of the 16th
century. To counter these threats, the Spanish began to ship their treasure in convoys or flotas,
with the main cargo being heavily guarded by Spanish war galleons. There were two main fleets:
the Nueva España Flota, which picked up its cargo at Vera Cruz, Mexico, and the Tierra Firma
Flota, which stopped at Porto Bello in Panama and Cartagena in Colombia. The flotas left Spain
together yearly for the colonies, then split up to call at their respective ports. Ideally, the fleets
would reconvene at Havana, Cuba, to prepare for the Atlantic crossing and return together
(Smith 1988:85–86).
175
While the flotillas provided protection from attack by other vessels, they could not protect
the Spanish treasure from dangerous reefs and shoals, or from the threat of hurricanes, which
swept through the Caribbean and the east coast of North America with regularity during the late
summer and fall hurricane season. With the boats traveling together, unlucky timing of a voyage
could mean the loss of dozens of ships in a single storm. Such disasters befell the flotas of 1554,
1622 (of which Atocha was a part), 1623, and 1632. Santa Margarita was lost off the coast of
Florida in 1595 with a cargo estimated at $3 million in gold and silver. The loss of over a dozen
Spanish ships off the Florida Keys in a 1733 hurricane it was one of the last disasters to befall
the flotas, capping over 200 years of massive losses (Lonsdale and Kaplan 1964:58).
The Spanish established the first permanent European settlement in the U.S. at St. Augustine,
Florida, in 1565, expanding northward in the late 16th century into Georgia and the Carolinas.
Between 1562 and 1588, settlements were established at San Pedro (Cumberland Island) and
Santa Catalina (St. Catherines Island) in Georgia, and Fort San Marcos in Port Royal, South
Carolina. Although the goal of the settlements was to exploit local resources, missions were also
established to convert the local Native Americans to Catholicism. Eventually, Franciscan monks
established more than 100 missions in the Southeast. The east coast of Florida south of St.
Augustine remained largely uninhabited by Europeans, with the exception of a fortified
watchtower at Matanzas Inlet, until the 17th century (SAI 1981:III-26–28).
Increasingly, Spanish ships carried civilians, trade goods, and slaves to supply its outposts.
The Spanish conquistadors had routinely enslaved the Native Americans of Central and South
America to work in the mines and perform other labor, but a decline in population from warfare
and disease resulted in a shortage of labor in the mid-16th century. The Portuguese had been
securing slaves from Africa and transporting them to Portugal and its possessions in the Eastern
Atlantic since the 1440s, and Spain soon followed in the practice in competition with its rival.
The introduction of slaves to the New World was gradual and sporadic, however, as the
importation of slaves was seen as potentially undermining the use of Native Americans for labor.
The earliest African slaves arrived in the New World as servants to the explorers or as laborers
on ships (Thomas 1997:54–76, 90–92).
The native groups were considered inferior slaves to Africans, and as their populations
declined, King Ferdinand began to authorize small numbers of slaves to be sent to Hispaniola to
work in the gold mines, beginning in 1510. The importation was closely regulated, with perhaps
50 per year being permitted. The Spanish colonists were soon clamoring for more to work on the
sugar plantations that were being established on Hispaniola, and in 1518, the recently crowned
King Charles I (later Charles V, Holy Roman Emperor) issued a license to import 4,000 or more
slaves to the Spanish colonies. Sugar cane would prove to be the life blood of the Caribbean and
the principal driver of the Atlantic slave trade. By 1537, when Hernando DeSoto was granted
permission to bring 50 slaves on his expedition to the mainland of North America, thousands of
Africans had already been transported from Europe and Africa to the New World (Thomas
1997:92–103).
The Portuguese, British, French, and Dutch all eventually developed sugar plantations on
their claims in the Caribbean and South America, and all were involved in the slave trade to
varying degrees. Over a period of more than 300 years, approximately 10 million Africans were
176
transported from Africa as slaves. The vast majority of this number went to Brazil and the West
Indies, but about 645,000 were brought into the British colonies in North America and later the
U.S. to labor on rice, indigo, and cotton plantations in the southern colonies. Many of these
slaves were transferred there from the West Indies. To maximize profits, slave traders packed
their human cargo into ships under deplorable conditions. It is estimated that one in eight slaves
died en route and about one-third died in the first three years, ensuring a steady demand (Preston
and Natkiel 1986:54–55). Slaves were also imported into the Mid-Atlantic and New England
colonies, although in smaller numbers, to serve primarily as household servants and general
laborers. Nevertheless, large dairy and cattle farms in Rhode Island and Connecticut employed
slave labor, and by the mid 18th century, the black populations of these colonies numbered over
5,000 each (Russell 1976:196).
In the late 16th century, France began an effort to colonize North America to provide a haven
for French Protestant Huguenots. The colonies would also provide a base from which they could
launch attacks against the Spanish flotas. Two French settlements actually predated St.
Augustine, which the Spanish had established in response to the threat of the French colonies. In
1562, French explorer Jean Ribault attempted a settlement at Port Royal, South Carolina, called
Charlesfort, but the effort was a failure. Some of the settlers from that venture relocated
southward under René Laudonnière and established Fort Caroline, at the mouth of the St. John’s
River, Florida, in 1563. Most of the inhabitants of this colony were slaughtered by Pedro
Menéndez de Aviles, who had established St. Augustine to protect the Spanish claim to Florida.
After that time, the French concentrated their efforts on the Canadian provinces and the lucrative
fur trade there (SAI 1981:III-26).
12.2. SHIPPING AND SEAFARING
SEABOARD (1600–1884)
IN THE
ENGLISH
AND
DUTCH COLONIES
OF THE
ATLANTIC
Soon after the abortive French colonies on the Southeast coast, English and Dutch explorers
began to scout the Eastern Seaboard for favorable locations for colonial settlements. The vast
coastline north of the Spanish missions was first explored by John Cabot in 1497 and by Italian
Giovanni da Verrazano, sailing under a French flag, in 1524; but it was not until Sir Walter
Raleigh founded the Roanoke Colony in 1585 that any effort was made to establish a permanent
settlement. Raleigh returned to England for supplies for the struggling colony, only to return to
find that all of the inhabitants had disappeared without a trace. James Smith established the first
successful English settlement in the New World at Jamestown, Virginia, in 1607. The survival of
the settlement was by no means assured, however, and colonists nearly perished in the winter of
1609–1610. The cultivation of tobacco in the Chesapeake proved to be a lucrative enterprise, and
by the mid-17th century the region was thriving (Steffy 1988).
The purchase of 20 “negroes” from a Dutch privateer in 1619 by the Jamestown colony is the
first mention of the slave trade in the English colonies. In 1624, a census of the settlement noted
22 blacks, some likely personal servants of arriving colonists. By the late 1620s, reports of large
numbers of slaves arriving in Virginia can be found. The owner of Benediction complained in
1629 of the seizure of his ship, which was engaged in its “accustomed trade” with 180 slaves by
a French vessel (Thomas 1997:174–176).
177
Large plantations were established in the estuaries and tidal rivers of the Chesapeake Bay,
and a regular flow of English ships brought settlers and manufactured goods, while returning to
England with tobacco and furs. A regular trade in slaves and rum from the West Indies built up
the plantation landings, which became mini-entrepôts that enriched the aristocratic families of
Maryland and Virginia (Isaac 1982:16, 32–33).
Soon after the establishment of Jamestown, Englishman Henry Hudson, sailing for the Dutch
East India Company, explored the river that now bears his name. In 1610, the Dutch established
a trading post on the island of Manhattan from which to engage the Native Americans in the fur
trade. However, it was not until 1624 that the permanent colony of New Amsterdam was
established on Manhattan Island under the oversight of the newly established Dutch West India
Company. The Dutch soon established settlements up the Hudson River to Albany (called Fort
Orange at that time), and engaged in extensive trade with Europe and the English colonies,
which took advantage of political turmoil in England to skirt trade regulations. The Netherlands
led the world in global trade by the 1650s (Steffy 1988:107, 116). Among the cargo of Dutch
ships were African slaves, many of them taken from Portuguese ships seized in the ongoing
conflict between the two competing shipping powerhouses in the first half of the 17th century.
Slaves were sold in New Amsterdam as early as 1625, and the Dutch West India Company in
1629 promised to bring as many slaves as possible to the colonists (Thomas 1997:170).
Tensions between the English and the Dutch escalated after the restoration of Charles II to
the throne, and in 1664, a British fleet forced the surrender of New Amsterdam without a shot
being fired. The Dutch briefly regained control of New Amsterdam in 1674 during the Third
Anglo-Dutch War, but it was recaptured by the English the following year.
About the same time that New Amsterdam was established, refugee Protestant groups from
England were arriving in New England. Plymouth in 1620, Salem in 1626, Massachusetts Bay in
1628, Providence in 1630, and Hartford in 1635 were among the earliest settlements there. A
regular supply of provisions, livestock, and manufactured goods from England was needed to
support these colonies in the early years, with little produced for the return voyage. By the last
quarter of the seventeenth century, however, the New England colonies were producing corn,
flax, potatoes, salted meat and fish, livestock, and a wide variety of forest products (lumber,
ships’ masts, staves, pitch, and tar) for export to Europe, the Caribbean, and even Africa. A
strong coastal trade also developed (Russell 1976).
An adequate inventory of sailing ships was needed to facilitate this trade. Although the
colonies were rich with materials for shipbuilding, skilled workers and tools were lacking in the
first half of the sixteenth century. However, an array of small, coastal vessels was constructed in
New England and to some extent in the Chesapeake Bay. These included flats, skiffs, sloops,
cutters, shallops, and pinnaces. Fishing fleets were also needed to harvest the bounties of the
North Atlantic and Chesapeake Bay. Fishermen established winter camps along the coast of
Maine in the 1600s, many of them located on islands for protection from hostile native groups.
Ships were often constructed or repaired in these camps during the winter. Most ocean-going
vessels supplying the English colonies during the early 17th century were manufactured in the
motherland, with local shipyards dedicated primarily to maintenance and repair. By the mid
1600s, with the supply unable to meet demand, however, Massachusetts Governor Winthrop
178
instituted a number of measures to encourage shipbuilding. Between 1641 and 1646, at least six
ships of more than 200 tons and several smaller ones were built. Restrictions on trade enacted in
England only encouraged smuggling, which in turn benefitted the shipyards. By the late 17th
century, Boston had become the leading ship-builder, turning out an average of 15 ships a year
(Bauer 1988:26, 31–32; Steffy 1988:116).
With the New England and Chesapeake Bay colonies achieving stability and returning profits
to their investors and Spanish power waning on the South Atlantic coast, a number of new
ventures were attempted in the Mid-Atlantic and Southeast in the second half of the 17th century.
New colonies were founded in North Carolina (1653), New Jersey (1664), South Carolina
(1670), and Delaware (1681). In 1686, Spain withdrew from all of its settlements and missions
north of the St. Mary’s River. Overland travel along the Atlantic Seaboard was difficult in these
regions because of low-lying floodplains and large estuarial waters. Consequently, coastal and
inland water travel was critical to development. The Carolina settlement of Charles Town served
as the shipping and administrative center for the river plantations growing rice, indigo, and later
cotton. Charleston traders also ventured far into the interior to trade with Native American
groups for deerskins. The demand for deerskins was so high in England that the deer population
in the Southeast was nearly wiped out. Beginning in 1698 and continuing until the American
Revolution, 50,000–100,000 skins were shipped annually from Charleston alone. Charleston’s
pre-eminence as the leading southern port was eventually challenged by Savannah, which was
matching the South Carolina port’s exports in deerskins by 1768, and also shipped large
quantities of rice and naval stores (Braund 1993:97–98).
During the 18th century, Spanish colonial power in the Western Hemisphere continued to
wane, although Spain clung stubbornly to Florida, Mexico, and three of the four largest islands in
the Caribbean (Cuba, Hispaniola, and Puerto Rico) during the first half of the century.
Meanwhile, the British gained control of Jamaica and many of the islands of the Lesser Antilles,
while strengthening its position in North America with the establishment of the Georgia colony
in the 1730s. The ongoing colonial feuds encouraged piracy on the high seas, and the early 18th
century is regarded as the heyday of Caribbean piracy and illegal trade. Piracy was not, however,
confined to the Caribbean. Pirates were active on the Atlantic coast as well, using the isolated
inland waters of the barrier islands as hideouts (SAI 1981:III-36). In 1984, the wreck of the
pirate ship Whydah was discovered 500 feet off Marconi Beach, Cape Cod, Massachusetts. The
flagship of the pirate “Black Sam” Bellamy wrecked in a storm in 1717. To date, it is the only
confirmed pirate shipwreck that has been discovered. The supposed wreck of Queen Anne’s
Revenge, the flagship of famed pirate Blackbeard (née Edward Teach) lost in 1718, has been
recently located at the mouth of Beaufort Inlet in North Carolina. Both ships were former slave
vessels that had been captured, and shackles used during its slave days (and likely its pirate days,
as well), have been recovered from the Whydah (Expedition Whydah ca. 2008; Queen Anne’s
Revenge n.d.).
British privateers weakened Spain’s sea power, even as Spanish flotillas continued to ship
gold and other goods from Mexico through the Straits of Florida. In 1714, all but one of a fleet of
11 ships were smashed in the shallows off Sebastian Inlet south of Cape Canaveral. In 1733 a
convoy of 21 ships, including three armed galleons and 18 merchant ships were struck by a
hurricane off the Florida Keys while on a return voyage to Spain. Only one of the ships was able
179
to return safely to Havana. Portions of these ships and their cargoes have been recovered, but
likely many still remain. The Spanish conducted their own salvage operations after these losses,
employing drag nets and Native American and African-American free divers to recover the lost
cargo (Preston and Natkiel 1986:69; Smith 1988:95–103).
The success of colonialism and its attendant mercantile economic policy during the 18th
century contributed to a golden age of seafaring among European nations. Ship design became
more sophisticated and advances in engineering, navigation, and meteorology made voyages
increasingly successful, lining the pockets of those men who could control the lucrative contracts
issued by the crown heads of Europe (Schlesinger 1917). By the early 18th century, nearly 400
ships were clearing Boston yearly for Britain, and the number continued to increase. New York,
Philadelphia, Newport, and Providence would follow in the mid-18th century, with 300 or so
ships a year arriving from foreign ports (Bauer 1988:38–42).
England and France struggled for control of the North American continent during the 18th
century, sparking a series of wars between the two powers and their Native American allies.
Control of the seas was critical in these colonial conflicts, since the transport of troops and
supplies by water was more efficient than overland travel in the undeveloped backcountry.
Queen Anne’s War (1702–1713) was fought primarily in the border area between the English
and French colonies, but the hostilities did affect shipping on the Atlantic seaboard, as the
French fortification at Louisbourg on Cape Breton Island provided a base for the French navy
and privateers to harass English merchant vessels and fishing fleets. The Treaty of Utrecht,
which ended the war, provided the English with a permit to import slaves to the Spanish West
Indies, catapulting the country into the slave trade (Crisman 1988).
In King George’s War (1744–1748), a group of New England colonists organized an
expedition against Louisbourg, utilizing merchant vessels and fishing boats outfitted with guns to
bombard the fortress. With the Royal Navy blockade preventing the resupply of the fort, the
French eventually were forced to surrender. England was also at war with Spain during this
period, a conflict known as the War of Jenkins Ear in the colonies. The main theater in that war
was the coast of Georgia, where England had recently established a colony. Led by Georgia
colony founder and accomplished sea captain James Oglethorpe, the English raids on Spanish
positions south of Savannah were successful in providing breathing room for the Georgia colony
to prosper (Crisman 1988; Spalding 1977:30–33).
The uneasy truce following King George’s War would collapse soon after, drawing the
world’s colonial powers back into conflict for control of world trade. The French and Indian War
(1755–1763) was the North American component of the Seven Years War, which saw superior
British sea power eventually crush the French and their Spanish allies in the Gulf and North
America. The fall of Quebec in 1760 effectively ended the French resistance in North America,
and Canada was transferred to Britain under the terms of the Treaty of Paris in 1763. The
Spanish relinquished Florida under the terms of the treaty as well, and it remained a British
possession until after the American Revolution, when it was returned to Spain (Crisman 1988).
Great Britain became a major player in the slave trade in the 18th century, in large part
because of a provision in the Treaty of Utrecht in 1713 that granted the country a license to ship
slaves to the Spanish colonies in the West Indies. In the 1720s, Great Britain brought about
180
100,000 slaves to the New World, mostly to Panama, Buenos Aires, and Cartagena, as well as to
the British colony of Jamaica. About 10,000 slaves were brought to mainland North American
colonies such as Virginia and South Carolina. Another 40,000 arrived in the 1730s. Slaves were
also brought into Boston, New York, and Philadelphia, although in smaller numbers. Eltis (2001)
estimates that 361,000 slaves were brought to British North America before 1868. Colonial
merchants and shipping concerns gradually became involved in the slave trade during the 18th
century. The majority of these were New Englanders, being more heavily invested in
shipbuilding and overseas trade. Rhode Island was a major center for the slave trade, having
good harbors and a strong shipbuilding industry. Newport and Providence were important stops
in a triangular trade that brought slaves to the West Indies, sugar to New England, and rum to
Africa (Thomas 1997:244, 246, 259–261).
It is difficult to estimate the number of slave ships that may have been lost along the Atlantic
seaboard, since even during the 17th and 18th centuries, much of the slave trade was clandestine.
Two lost slave ships, Henrietta Marie and Guerrero, have been identified off the Florida Keys.
Henrietta Marie was wrecked on New Ground Reef in 1700 after dropping off a cargo of 190
slaves in Port Royal, Jamaica. Guerrero was engaged in illegal slave trading and became
stranded on Carysfort Reef while trying to elude a British naval schooner, H.M.S. Nimble.
Guerrero did not sink immediately, but 41 of the 561 captives aboard were drowned when the
holds filled with water (Barnette 2003).
Nurtured by mercantile laws intended to spur trade, the triangular trade in the Atlantic
reached its peak in the 18th century. With the waning of Dutch and Spanish power in the Western
Hemisphere, the British rose to dominance in the global shipping industry. The British
government sought to control trade with the colonies to the benefit of the mother country by
requiring that ships calling at English ports be owned and manned by Englishmen and that
shipments between the colonies be routed through England, but these regulations were largely
ignored. Although the majority of ships in colonial ports were British-owned, hundreds of
American built and owned vessels were engaged in local and transatlantic trade. American
merchant families and plantation owners rose to political and social prominence during the
period, forming the cultural foundation for the American Revolution (Steffy 1988:119).
Each region of the colonies developed its own export specialties and import needs. The New
England colonies exported beef, forest products, grain, fish, and rum while importing
manufactured goods, sugar, molasses, and metals. The Mid-Atlantic colonies produced fish,
tobacco, and furs for export, while importing manufactured goods and slaves. In the South,
tobacco, indigo, rice, furs, naval stores, and deerskins were the chief exports (with cotton
becoming important in the latter part of the period). The South imported manufactured goods
from England, sugar from the West Indies, and slaves from the West Indies and Africa (Preston
and Natkiel 1986:68). In addition to these major commodities, all manner of goods were shipped
across the Atlantic and between the colonies, including coffee, tea, spices, salt, produce, tableand glassware, fabrics, tools, tin- and brassware, ivory, and precious stones.
12.2.1. The Development of Shipping and Shipbuilding in Colonial America
The vigorous colonial trade gave rise to the great seaports of the Atlantic Seaboard. On the
eve of the American Revolution, Philadelphia was the largest city, with a population of
181
approximately 40,000. New York was a bit more than half that size, with a population of 25,000.
Boston, Charleston, and Newport all boasted more than 10,000 residents. For each of these cities,
maritime trade was its lifeblood, and each developed shipbuilding industries. The British
government encouraged shipbuilding in the colonies, and in the New England states the large
number of artisans, the high demand for sailing vessels, and greater emphasis on industry led to a
burgeoning shipbuilding industry. Philadelphia and Boston dominated in the construction of
larger vessels, while New England ports like Essex and Gloucester, Massachusetts, Providence,
Rhode Island, and Mystic, Connecticut, produced smaller schooners and sloops for fishing or
coastal and Caribbean trade. By the start of the American Revolution, 30 percent of the British
merchant fleet was from colonial shipyards and there were an estimated 2,000 American ships
engaged in trade. It was these merchant vessels that would be pressed into service in the
Revolution and form the basis of the U.S. Merchant Marine (Bogart 1912:56; Butler 1997:8–9;
Steffy 1988:128).
Despite the vigorous trade and the vast supply of raw materials for their construction, few
ships were built in the South until about the mid-18th century. With an adequate supply of
merchant ships from England, along with an increasing supply from New England, southern
shipyards concentrated on making repairs rather than fabrication. Vessels that were built were
generally sloops and schooners involved in the coastal trade. Of the 229 vessels trading in North
Carolina in the period from 1710–1739, less than 17 percent were built there (SAI 1981:III-39).
The shipping trade carried the potential for great profits for ship owners, but was a risky
business during a period of widespread piracy, primitive navigation, and sailing ships that were
at the mercy of Atlantic storms, Caribbean hurricanes, and poorly-charted hazards. In addition to
these catastrophic threats, there were inadequate crews, bureaucratic delays in overseas ports,
uncertain markets, and impatient investors. The potential for huge loses was ever present, and as
many fortunes were lost as were won. The cost of constructing a large ocean-going vessel with a
capacity of 100 tons was a significant investment, one that could result in a total loss in the event
of a cabin fire, a navigational miscalculation, or a fierce storm. During the 18th century, the
practice of insuring vessels and cargo became more widespread and organized. In the 1690s,
Edward Lloyd began to publish a weekly newsletter in London reporting on the arrival and
departure of ships to attract mariners, merchants, and bankers to his coffeehouse on Lombard
Street. The newsletter also reported on shipping losses such as ships not heard from or coming
into port damaged. The paper came to be called Lloyd’s List and has been in continuous
publication since. Starting in 1735, the list was published twice weekly. Lloyd’s Register of
Shipping, a separate yearly inventory of British and foreign merchant ships, is a good source of
information on the size, construction, and condition of vessels, as well as the owner, captain,
home port, and other valuable data (SAI 1981:III-40). Unfortunately, however, until recently the
registers did not include a record of losses, which would have been of great use for documenting
early American shipwrecks.
To mitigate potential losses, 18th century merchants and shipping interests began to invest in
aids to navigation and protection for mariners. The first lighthouses on the Atlantic Seaboard
appeared in the early 18th century. The Boston Light on Little Brewster Island in Boston Harbor
was the first of these, built in 1716, with others built in Massachusetts, New York, and Maine.
These were largely funded by states and cities that hoped to make their harbors more attractive to
182
mariners. The Sandy Hook lighthouse protecting the entrance to New York harbor was
completed in 1764 and is the oldest lighthouse in the U.S. still in operation. In 1789, the U.S.
Lighthouse Service was created as a part of the Department of Treasury, the first public works
agency of the federal government. The construction, maintenance, and operation of all
lighthouses in the U.S. fell to the new agency. In 1791, the Portland Head lighthouse in Maine,
originally funded by the state of Massachusetts, became the first lighthouse completed by the
federal government. A dozen more were constructed over the next decade (Lighthouse Depot
n.d.; Smithsonian Institution n.d.).
Later, in the 19th century, lightships were deployed at dangerous shoals, guiding ships along
the coast and into port where lighthouses could not be built. Between 1820 and 1952, 179
lightships were built for use in American waters. These vessels, both manned and unmanned,
often became casualties of storms themselves, and were thus added to the inventory of
submerged historic vessels (Delgado 1989).
Working on ships was extremely dangerous, and in 1742 the Boston Marine Society was
formed to support the families of captains disabled or lost at sea. The society was the first
organization of its kind in the world. The Humane Society of Massachusetts was established in
1786 and constructed rescue stations along the New England coast, as well as shelters for
stranded mariners (Boston Marine Society 2001; Humane Society of the Commonwealth of
Massachusetts 2009).
In addition to physical protections for their ships, colonial merchants sought to maximize
their profits by ensuring full-capacity loads for each leg of a voyage, securing reliable markets,
and minimizing fees and duties. They soon chafed under the restrictions on trade imposed by the
British Crown, which sought to wring as much benefit from the colonies as possible. It was, in
part, the growing power of these merchants, and Great Britain’s futile efforts to control them and
their profits, that brought on the American Revolution (Bauer 1988:44–49; SAI 1981:III-42).
12.2.2. Naval Action in the Revolutionary War
At the outbreak of the American Revolution, the British had the most powerful navy in the
world, while the colonies had none. Yet sea power was to play a decisive role in the conflict. The
British wasted considerable time in attempting to subdue the revolt on land, and it was not until
the war was in its fifth year that the strategy shifted toward blockading the coast to prevent
supplies from coming in and the exports that were sold to fund the war effort from going out.
Against the mighty British ships-of-the-line, commonly fitted with 74 guns, and a vast fleet of
supply and support ships, the Continental Congress cobbled together a collection of refitted
merchant ships of 10–20 guns, eight new frigates of 24–32 guns (13 were originally authorized),
and varying quality sloops and schooners purchased from merchant shippers. The new frigates
were generally well-made and fast, as evidenced by the admiration bestowed on them by the
British navy, but all were scuttled, destroyed, or captured, in large part because of incompetent
and inexperienced captains and poor strategic planning. The Continental Congress ordered
several more ships in 1776, including three ships-of-the-line of 72 guns, but only one of these
was built, and it was not launched until the war was effectively over, in 1782 (Chapelle 1949;
Sands 1988). None of the wrecks of these original frigates are known to be in Federal waters.
183
Although the frigates ordered by the Continental Congress had little impact on the war, the
sloops, schooners, brigs, and brigantines that were purchased, pressed into service, or operated as
privateers proved to be of use to the Continental cause. Fast, maneuverable schooners, a ship
design indigenous to the American colonies, were able to harass British supply boats, slip past
blockades, and serve as packets for re-supplying the Continental army. In addition, smaller
vessels such as galleys, gunboats, and barges were used on inland waterways (Chapelle 1949;
SAI 1981; Sands 1988).
The individual colonies also contributed vessels to the Patriot cause. Much of the naval
action during the war took place south of the Mason-Dixon Line, as the British shifted their
emphasis to stifling trade and restricting the Continental Navy’s ability to resupply troops on
land. The South Carolina State Navy was small and undermanned, but was very active in the
defense of Charleston and the coast of South Carolina. Between 1776 and 1779, South Carolina’s
navy was able to capture 35 ships in the Carolinas, Florida, and the West Indies. Its fleet
included at least four vessels acquired from France. Three of these French-made ships, as well as
four Continental Navy frigates, and four other vessels were scuttled at the mouth of the Cooper
River and their masts joined by a boom to impede the British squadron, during the siege of
Charleston in 1780 (SAI 1981; Watts 1995:19).
In North Carolina, the state navy was primarily concerned with protecting Ocracoke Inlet,
which was the only reliable access to Pamlico Sound and important supply lines for General
Washington. The dangerous waters of the Outer Banks took their toll, however, as the ship
Caswell and the sloop Independence were lost during these patrols (SAI 1981:III-44).
Perhaps the most potent maritime weapons of the Revolutionary cause were the privateers,
private ship-owners granted letters of marque permitting them to outfit their boats with guns and
seize the ships and cargoes of the enemy. In all, some 600 British ships were seized during the
course of the war, although in many cases the privateers simply sold the captured goods back to
the British (Chapelle 1949; SAI 1981; Sands 1988).
Although the Continental Navy, aided by privateers, was able to secure isolated victories
against the Royal Navy, their efforts were ineffective in breaking the British hold on the
coastline. The state navies were generally unable to protect the coastal ports, which fell one-byone to the British forces, beginning with New York in 1776, where a massive British fleet
deposited 12,000 British regulars and 9,000 Hessian mercenaries to take control of the city and
its harbor. The Penobscot Bay expedition in 1779 was an ill-fated American effort led by
political appointee Dudley Saltonstall that maneuvered itself into a trap, with the result that all of
its ships were scuttled, run aground, destroyed, or captured. None of these wrecks are known to
be in federal waters (Sands 1988).
One of the largest naval battles of the war, and its most decisive, was the Battle of the
Chesapeake, also known as the Battle of the Virginia Capes, in September 1781. A British fleet
under Sir Thomas Graves sailed from New York to deliver supplies to General Cornwallis, who
was hemmed in at Yorktown, Virginia, by General Washington’s forces. A French squadron,
under the command of the Comte de Grasse had arrived to protect the entrance to Chesapeake
Bay. The British fleet caught the French at anchor behind Cape Henry, but did not attempt to
enter the bay, instead drawing up in a line of battle off the coast. Only the leading lines of the
184
squadrons were engaged in the fight, which ended in a draw, and after several days of
maneuvering, Graves was unable to break the French blockade. Cornwallis was consequently
forced to surrender at Yorktown after a brief siege (Preston and Natkiel 1986:80).
12.3. THE RISE OF THE UNITED STATES AS A MARITIME POWER (1790–1865)
Although the defeat of Cornwallis at Yorktown marked the effective end of the American
Revolution, the settlement would not be completed until the European powers negotiated the
Treaty of Paris in 1783. The importance of naval power to the Patriot victory was not regarded
well enough to induce the newly organized U.S. Confederation of States to authorize its own
navy after the war. There was little money in the Treasury, and even after the strengthening of
the federal government under the new Constitution, there was little effort to develop a navy.
Consequently, ships flying the American flag were subject to seizure by pirates and privateers.
Particularly troublesome were the Barbary Coast pirates of North Africa, but the continuing
European wars meant that an American ship might be boarded by British, French, or Dutch
ships, its crew impressed into service on meager pretenses, or its cargo seized as contraband.
12.3.1. The Birth of the U.S. Navy
By the late 1790s, the need for some sort of navy had become obvious, and in 1794 Congress
authorized the construction of six frigates. However, subsequent peace treaties slowed the
progress on the ships, and only three were launched in 1797. Among these was the 44-gun USS
Constitution, the only surviving ship from the first U.S. Navy. While the construction of these
vessels was going on, the wars in Europe escalated the threats to U.S. merchant ships. The
French Revolution increased the number of detentions and confiscations, and by 1798, the issue
had reached the point of undeclared war. Consequently, Congress authorized the construction of
more warships, as well as funds to complete previously contracted vessels that had languished in
the yards. Five new frigates of 28–36 guns were launched in 1799—Boston, New York,
Philadelphia, Essex, and John Adams. These frigates were constructed by subscription at various
shipyards on the east coast, the first three in their namesake ports, Essex in Salem, and John
Adams in Charleston (Chapelle 1949:126–135).
Although the ships were strong and of sound design, they were subject to frequent
“improvements” by the Navy and their individual captains that compromised their effectiveness.
Some were fitted with more guns than they were rated for initially, adversely affecting speed and
maneuverability. To compensate for the added weight of the guns, captains often called for
longer spars to carry a greater amount of sail, hoping to increase speed, but often exceeding the
capacity of the hull design. A number of these ships were subsequently modified to improve their
versatility, and the feedback of officers led to a shift toward smaller ships in the first part of the
19th century (Chapelle 1949:150, 168–169).
The Navy also acquired a number of cutters on loan from the Revenue Department, which
operated as U.S. Customs Service cutters, the precursors of the Coast Guard. The Customs
Service, created in 1790, had received nine small schooners and one sloop. These ships proved to
be too small, and 11 double topsail schooners were built by subscription in 1797. The Navy took
nine of these into service, returning six to the Revenue Service in 1799 (Chapelle 1949:146–
147).
185
The Navy was scaled back in 1801 after a treaty was established with the French in which
U.S. neutrality rights were accepted and obligations originating during the American Revolution
were terminated, but the issue of freedom of the seas was far from settled. After disposing of a
number of useful schooners and other small vessels, the Navy was left with just 13 large frigates,
only six of which were maintained for active duty. The Barbary pirates continued to harass
American merchant vessels, but the Jefferson administration hoped that privateers could provide
assistance to the meager force that was sent to Tripoli. It soon became clear, however, that
cruiser class vessels were necessary, and in 1802 Congress appropriated funds for four vessels of
no more than 16 guns, as well as 15 small gunboats. The use of single-cannon gunboats of less
than 100 feet in length for harbor and river patrol was developed by the French and proved
attractive to the U.S. because of the relative economy of constructing these vessels. An
additional 25 gunboats were ordered in 1805, 50 more in 1806, and 188 in 1807, although not all
of these were built (Chapelle 1949:179–219).
12.3.2. American Shipbuilding Comes of Age
While the Navy was shifting to smaller vessels and showed little interest in innovation during
the early 19th century, commercial shipbuilders were developing increasingly sophisticated
designs. Brigs and schooners were made larger, with greater speed and carrying capacity.
Improvements were also made to rigging and spars that made them lighter and stronger. The
demand for American-made schooners was high, and many were purchased by foreign traders, to
the dismay of those concerned with protecting American interests. The Jefferson Embargo,
instituted in 1807 following the British firing on USS Chesapeake, worked to depress
shipbuilding in the North, however, while encouraging illegal trading. Southern shipyards fared
better, turning out fast ships for the contraband trade that took advantage of the less wellregulated southern ports (Chapelle 1949:243).
With its navy neglected and its merchant marine struggling against the embargo, the U.S.
was in a poor position to assert its maritime rights. The limited scope of the U.S. Navy at the
outbreak of the War of 1812 can be attributed to a conservative approach that emphasized coastal
defense using gunboats, as well as a resigned acceptance of British naval dominance. The
suggestion that the fledgling country could dictate the maritime policy of the indomitable Royal
Navy was met with derision. The cost of matching British sea power was considered prohibitive.
However, following a series of attacks on American merchant vessels, the cost of inaction
became evident. Had the U.S. deployed a sufficient navy before the war, it might have avoided
the conflict altogether. Instead, it would cobble together a fleet from its merchant marine, the
Revenue Service, and a series of hastily-built cruisers, while getting early support from
privateers (Chapelle 1949:150, 235, 240–244, 254).
The main theater of the War of 1812 was on the Great Lakes, where the U.S. planned an
ambitious invasion of Canada. Although the invasion failed, the U.S. was eventually able to
control the lakes by launching a formidable fleet against a poorly-supplied British command. On
the Atlantic side, American privateers scored a number of victories at the beginning of the war,
but the British stifled these attacks by blockading the most active U.S. ports to prevent the fast
ships from getting to sea. A flotilla system was instituted to provide safety in numbers. With only
a few frigates and nothing to match the large British ships, the U.S. Navy was unable to break
186
the blockade. With trade being strangled, the public began to demand a more effective naval
force. Congress finally acted in 1813, authorizing three 44-gun frigates and six 18-gun sloops-ofwar. Later appropriations followed, but many of the ships constructed were launched too late to
be effective in the war. By 1814, the British were able to attack coastal cities, striking a
demoralizing blow on Washington, D.C., in August 1814 (Cassavoy and Crisman 1988:171–173,
178, 188; Chapelle 1949:255).
Despite the advantage held by the British, and with the U.S. on the verge of bankruptcy and
anti-war sentiment running high, the British signed the Treaty of Ghent in 1814, ending the war.
Prime Minister Liverpool saw little to be gained from prolonging the conflict, which had thus far
been a stalemate, and from which both countries were suffering from a lack of trade.
Although the War of 1812 was largely a naval war, the majority of the engagements took
place on the Great Lakes and in inland waters, particularly Chesapeake Bay. Only a few
shipwrecks associated with the conflict are located on the Atlantic OCS. USS Wasp, a sloop-ofwar constructed at Newburyport, Massachusetts, by Cross & Merrill, is reported (Brown 2008) to
have been lost off the coast of South Carolina, but no reliable location information is known.
12.3.3. The Transatlantic Trade
Many of the issues that had brought on the War of 1812 were not settled by the Treaty of
Ghent, but the war boosted American shipbuilding and instilled confidence in commercial
interests that the U.S. was capable of defending its maritime rights. The period from the end of
the War of 1812 to 1880 was consequently one of rapid growth in the U.S. shipping industry and
is often considered the “Golden Age” of American maritime trade. The period also coincided
with the peak development of the sailing vessel, epitomized by the clipper ships, as shipbuilders
fought the futile battle to maintain sails in the age of steam (SAI 1981:III-3).
In the years immediately following the War of 1812, New York City emerged as the preeminent port of the U.S. Strategically located midway on the coast with a good harbor and quick
access to the open ocean, her position was further enhanced by the completion of the Erie Canal
in 1825, providing access to the burgeoning Great Lakes and Ohio Valley regions (Laing
1961:183). Great Britain and the U.S. quickly reinstated trade between the two countries after the
war to the benefit of both. American vessels could soon be found in ports all over the world. The
extent of this trade is evident from shipping trade papers published at New York that
documented the arrival and departure of cargo and passenger ships, as well as current market
prices for a variety of goods. The New York Shipping and Commercial List began publication in
1815 and included information on ships lost at sea, foundered, or run aground, as well as
accidents aboard ships (Mystic Seaport 1999). Although location information was generally
vague, these reports of “disasters, etc.” provide one of the best sources for documenting
American shipping losses for the antebellum period.
Transoceanic trade during the antebellum period was carried out primarily by three-masted,
square-rigged ships of 350 tons or more known as packet ships. They were among the most wellconstructed sailing vessels to ever be built, with full bows and decks for both passengers and
cargo. Packets had their origin in mail ships and passenger vessels of the 18th century. Thomas
Jefferson had suggested in 1785 that a regular packet service be instituted between the U.S. and
187
France, with ships scheduled for regular departures and arrivals, an indication of the increasing
reliability of ocean travel. However, it was the resumption of trade with England after the War of
1812 that led to the establishment of packet service between New York and Liverpool. The
Black Ball Line was the earliest and one of the most successful of the many packet lines
established during this period. Beginning with four ships, it eventually expanded to 39 vessels by
1855. The ships increased in size to exceed 1,000 tons. The Black Ball Line operated until 1878,
while its two major competitors, the Red Star and Blue Swallowtail lines, both ceased operations
in 1867 (Johnston 1988:235; Laing 1961:221–225).
The packets operated as common carriers and could charge higher rates because of their
speed and dependability. By carrying mail, perishable goods, and passengers, they could offset
the losses caused by sailing without full cargo holds. The New York to Liverpool packet service
was so successful that service to France was instituted in the 1820s, and similar lines were
established in other U.S. ports, including Boston, Philadelphia, Baltimore, Charleston and
Savannah. Coastal packets called at cotton ports in the South, carrying the fiber to textile mills in
England, then returning to New York with manufactured goods. By the 1830s, four packet lines
employing nearly 100 ships were operating at New Orleans (Butler 1997:35–38; Laing
1961:226–227).
The success of the transatlantic packet lines put a new emphasis on speed and gave rise to a
new era of ship design. Baltimore shipbuilders were noted for their fast privateers, which became
the model for the famed clipper ships that emerged in the mid-19th century. Ann McKim, built by
the Baltimore yard of Kennard & Williamson in 1833, is widely regarded as the first American
clipper ship. These tall, square-rigged ships were designed with raked masts, wedge-shaped
bows and sweeping sterns that sacrificed storage space for clean, fast lines. The ships generated
excitement with the public, which followed their exploits with the intensity of sports fans. The
California Gold Rush of 1849 raised the stakes in the sailing ship speed wars. In one year, 800
ships entered the port of San Francisco carrying 90,000 men seeking their fortunes. Most arrived
by clippers plying the Cape Horn route around South America, although Cornelius Vanderbilt
had success with a combination land and sea route that crossed from Atlantic to Pacific via a 12mile macadam road through Nicaragua (Butler 1997:65).
In 1850, Samuel Pook, only 23 years old, designed the first clipper built at Boston, the 1,261ton Surprise, which bested the previous record to San Francisco by one day on its maiden
voyage. Donald McKay, the industrious Baltimore ship builder, laid out plans for his first
clipper, Stag Hound, in October of 1850 and launched it just 60 days later. He went on to design
and build many of the fastest sailing ships of the day, each progressively larger in size,
culminating in Great Republic in 1856, a 4,455-ton vessel that carried an acre-and-a-half of sail
cloth and a 210-foot main mast. Unfortunately, the vessel burned in a dock fire in New York
after sailing from Boston to New York to load its first cargo and was sold and rebuilt on a
smaller scale. McKay scaled back his ships when it became clear the demand for such vehicles
was not enough to justify the costs of their maintenance and operation (Butler 1997:73–75).
12.3.4. The American Schooner
In addition to overseas trade, coastal trade prospered as Yankee shipyards turned out
hundreds of top-sail schooners that could carry a sizeable load with a small crew and were
188
capable of travel on the open ocean. These ships were fore-and-aft rigged vessels with squarerigged sails atop one or both of its masts. The Dutch developed the schooner rig in the 17th
century, and this was further refined in America during the 18th century. In the 19th century, the
two-masted design was often expanded to three. In the quest for greater capacity, by the end of
the century shipbuilders were experimenting with four, five, and six masts, all of equal size and
rigging that required only a modest increase in crew size. These multi-mast vessels benefitted
from the development of the steam donkey to assist in hauling lines. However, they proved to be
difficult to handle and were often saddled with inexperienced crews to reduce costs. As a result,
many met their demise along the Atlantic seaboard. The fleet of schooners amassed by William
F. Palmer in the early 20th century serves as an example; of 15 ships, 13 were wrecked, burned or
abandoned at sea, and one was torpedoed during World War I. Only one was retired intact.
Thomas W. Lawson, launched in 1902 with seven masts, was the final desperate effort to extend
the schooner principle to a vessel of a size that could compete with the tankers then being built,
but it proved unwieldy and sank in a storm off the coast of England in 1907 (Johnston 1988:249;
Laing 1961:203–206).
The dependable American schooner could be found all along the eastern seaboard during the
19th century, hauling cotton from Savannah northward, lumber from Maine southward, and all
manner of raw materials and finished goods in between. Massachusetts began shipping ice to
New York and the Caribbean; lime and granite from Maine were used for concrete and street
paving in New York City; turpentine and naval stores from Georgia were major exports to
shipbuilding centers in the north; and textiles and leather goods from New England were needed
to clothe the growing slave population in the Deep South (Bauer 1988:273–275).
12.3.5. The Slave Trade in the United States
Shipyards in New England benefitted from the ongoing slave trade, despite the region’s
political opposition to slavery. Many of the states had made the importation of slaves illegal by
1790. However, American ships were frequently used, and many New Englanders were
complicit in the trade, which continued to find markets in the Caribbean and Central and South
America. From 1796 to 1807, 44 slave ships entered Havana, of which 35 were American
registered, though some were owned by Britons. Representative John Brown of Rhode Island
spoke out against a federal law in 1800 that would make it illegal for a U.S. citizen to hold a
stake in a slave ship, noting that New England rum, much of it made in Providence, was in great
demand in Africa. Although no slaves were imported to Rhode Island, both its shipbuilding and
rum industries were inextricably tied to the slave trade. From 1803, when South Carolina lifted
its temporary ban on the importation of slaves, until 1807, when the trade was prohibited by
federal law, 88 ships from Rhode Island landed 8,000 slaves at Charleston. In 1804, the
abolitionist collector of customs at Bristol, Rhode Island, was replaced by Charles Collins,
whose brother-in-law, James de Wolf, was a noted slave trader, bringing an end to any
prosecutions of slave traders there (Thomas 1997:543, 545).
The slave trade (although not slavery) was banned by both Great Britain and the U.S. in
1807; however, American interests continued to be involved in the business and a coastwise
trade in slaves was active through the Civil War. Rhode Island and Baltimore were major
suppliers of ships and captains, while Charleston was headquarters for well-heeled aristocrats
189
who arranged for clandestine deliveries along the coast of Georgia and South Carolina. It is
estimated that approximately 50,000 slaves were introduced to the U.S. between 1807 and the
start of the Civil War, primarily before 1840. Most of these slaves were smuggled into the
southern states through the Gulf coasts of what are now Texas (Spanish until 1821, part of
Mexico until 1835, and independent until 1846), Louisiana, and Florida (Spanish until 1818). On
the Atlantic coast, however, Amelia Island, Florida, St. Mary’s, Georgia, and any number of
barrier islands and isolated estuaries of the Sea Islands of Georgia and South Carolina provided
protection for slave traders (Thomas 1997:568–570, 614–617).
The Department of Treasury, through the Customs service, was charged with enforcing the
ban, but had no real police force. The scandal of Amelia Island, a well-known slave-trading
haven, prompted President Monroe to introduce a more comprehensive Antislaving Act in 1818,
and assigned enforcement to a fleet of naval vessels that were dispatched to the coast of Africa.
In all, 11 ships were captured with nearly 600 slaves on board. Still, considering the
contemporary estimate of one captain that there were 300 vessels of all nations engaged in the
trade, the effort had little impact. The U.S. was steadfast in its opposition to British naval ships
stopping U.S. flagged ships, and as a result, many slaving ships owned and registered elsewhere
began flying the American flag. In 1839, British naval ships brought five slavers flying
American flags into New York. Two were Spanish-owned, but the other three were owned by
Baltimore merchants and the ships were confiscated. Although American merchants were less
involved in the actual trade in the 1840s and 1850s, they continued to turn out ships that were
destined for the Brazilian slave trade. In addition to lax enforcement of the slave trade law, there
was no prohibition on the trade in slaves between states, and the entrepreneurial minded carried
out a profitable business transporting slaves from the depleted tobacco fields of Virginia to the
burgeoning cotton plantations of the Mississippi Valley and Texas (Thomas 1997:615–616 660661, 677–680).
12.3.6. The Rise of Steam
While ocean-going sailing ships were reaching their design peak, steam-powered vessels
were making their mark on America’s waterways. The earliest steam-powered vessels operated
on rivers, lakes and other inland waterways. Robert Fulton’s sidewheel steamer North River
Steamboat began service between New York City and Albany, New York in 1807, and was the
first commercially successful steamboat operation. After the War of 1812, sidewheel steamers
began to make regular runs on navigable rivers throughout the country, most famously on the
Ohio and Mississippi rivers between Pittsburgh and New Orleans.
Paddlewheel boats were not ideally suited for ocean travel, however. Ocean crosswinds lifted
windward paddles out of the water while submerging the leeward paddles. As the large quantities
of coal required for long voyages was used up during the trip, the vessels floated higher in the
water, decreasing the paddles’ effectiveness. Also, the engine and fuel meant less room for
cargo. Sailing ships, meanwhile, were capable of nearly continuous travel, as long as food and
water for the crew could be maintained (Butler 1997:24–28). Despite these drawbacks, the
promise of regular service offered by the steamship ensured its continued development and use
for ocean voyages (Haws 1976:119). In 1838, two paddlewheel steamers, Sirius and Great
190
Western, arrived in New York within hours of each other, becoming the first vessels to cross the
Atlantic using only steam power (Haws 1976:119, 127–128).
Early steamers of the 19th century were generally hybrid ships equipped with sailing rigs that
often provided the majority of the power. SS Savannah, the first steam-powered vessel to cross
the Atlantic in 1819, was in fact a full-rigged sailing ship equipped with a 90-h.p. single cylinder
steam engine and collapsible paddlewheels that were stored on deck when not in use. Steam
power was used for only about 80 hours of the nearly 28-day voyage. Gradually, as
improvements in design and performance were made to the engines, the sails became the backup
power, and eventually were abandoned altogether. They remained a fixture on these ships into
the 20th century, however. Tugs and other coastal steam-powered vessels were generally not
equipped with sails.
Steamships began to ply the Atlantic seaboard in the 1820s, beginning with service between
New York and various New England ports through Long Island Sound. In 1824, Captain Seward
Porter of Portland, Maine, announced plans to operate the 300-ton steamship Patent, “elegantly
appointed for passengers,” between Portland and Boston (Preble 1883:114), although regular
service was not established until the 1830s (Bauer 1988:108). A number of lines operating out of
New York were incorporated by J.P. Morgan into the New England Steamship Company, which
became famous as the carrier of New York’s industrial and financial elite between the city and
their summer homes in Rhode Island. In 1846, Richard Borden opened a railroad between
Boston and Fall River, Massachusetts, where he operated steamships to Providence. In 1847, he
began regular service to New York. Passenger steamers were soon operating between New York,
Baltimore, Norfolk, Charleston, and Savannah.
The construction of steamboats required considerable use of iron, which eventually found its
way into the frames, fittings, and hulls of sailing vessels as well. As quality timber became
scarcer, the importance of iron increased. The need for iron foundries and iron workers around
shipyards gave an advantage to Northern yards, where capital and labor were more plentiful. The
more complex designs and need for expensive mechanical tools favored larger facilities over
smaller traditional ones, and shipbuilding declined in places like Salem and Bangor, as wellfinanced yards in Boston, Philadelphia, Baltimore, and Hampton Roads came to dominate the
industry (Butler 1997:37–38).
Iron provided the material necessary for another development in steamboat technology, the
use of screw propulsion. Initially, steam engines were not able to produce sufficient power to
drive a large ship with a propeller, but the addition of second and third cylinders to the engines
and the use of hot burning anthracite coal increased horsepower and made screw propulsion
more practical. Swedish engineer John Ericsson introduced a number of improvements to
propeller-driven vessels and is credited with the first American commercial steamship with
screw propulsion, Robert F. Stockton, in 1839. His USS Princeton was the first steam-powered
warship, and he was instrumental in designing USS Monitor in the Civil War (Butler 1997:40).
In 1845, HMS Rattler proved the superiority of screw propulsion in a tug-of-war with an
identical ship fitted with paddles. Still, paddlewheel steamships continued to be built, in part
because of mail contracts requiring their use (Haws 1976:133–134).
191
Perhaps of more concern for the paying passenger than the mode of propulsion was the care
and operation of the boilers. Early steam vessels were exceedingly dangerous. The steam engines
were notoriously temperamental and required constant attention. Corroded pipes, inattentive
operators, dirty stacks, and faulty valves often led to disastrous explosions and fires. Scalding
steam, consuming flames, and rapid sinking cost thousands of lives. Maginnis (1892) reported
more than 6,300 lives lost on passenger steamers of the Atlantic between 1840 and 1892. The
dangers of steamship travel prompted Congress in 1838 to pass a steamboat safety law that
required inspection of passenger vessels powered in whole or in part by steam. This piece of
legislation was one of the first to regulate private industry for the benefit of the American public
(Voulgaris 2009). The law was administered by the Justice Department, but did not provide for
an enforcement agency. Between 1847 and 1852, a series of disastrous boiler accidents, fires,
and collisions highlighted the need for codified regulations and stricter enforcement. In 1852,
Congress passed the Steamboat Act, which placed the enforcement under the Treasury
Department and created nine districts with inspectors assigned to each. The regulations did not
apply to non-passenger vessels, however, and crews were continually at risk from poorly
maintained machinery. In 1871, the regulations were extended to all steam-powered vessels, and
a comprehensive Marine Safety Code was established. The 1871 Act also provided for the
creation of the Steamboat Inspection Service as a formal entity (Voulgaris 2009).
At about the same time that the first steamboat act was passed, it had come to the attention of
Congress that the nation’s lighthouses were inadequate. Administration of these “aids to
navigation” had been carried out with a lack of oversight by a number of governmental
departments with little maritime expertise. Lights were poorly placed, poorly maintained, and
lacking improved equipment that would make them more effective. After more than a decade of
investigations, reports, and recommendations, Congress overhauled the Lighthouse Service
beginning in 1852. Twelve districts were created with an inspector for each. Operators were
required to be licensed and to submit frequent reports that were published as “Notices to
Mariners” (Butler 1997:50–51, 75–76).
In 1838, the British ship Great Western, a wooden, paddlewheel steamer, crossed the
Atlantic in 15 days, launching a new era of transatlantic travel using steam-powered vessels. The
ships owners, the Great Western Steam Ship Company, established the first regular transatlantic
service by steamship between England and New York. Great Western Company foundered,
however, when it lost out to Samuel Cunard for the British mail contract. Cunard was able to put
four paddlewheel steamers into service between Liverpool and Boston starting in 1840, while
Great Western Company had only one vessel. In 1845, after extensive delays, Great Western
Company launched the massive Great Britain, the first iron-hulled, screw-driven steamer, which
was by far the largest vessel then afloat. Great Britain was plagued with troubles, however, and
eventually bankrupted the company when it ran aground in 1846. Another British line running
the New York route, the British and American Steam Navigation Company, also was bankrupted
by the loss of one of its two ships, President, in 1841 (Corlett 1975; Gibbs 1963).
Great Britain represented the future of transatlantic travel, but further developments in
design and construction would be necessary before such ships were profitable. Meanwhile, the
Cunard line was highly successful by stressing safety and using ships that were similar in design
to the SS Great Western. In 1848, the Cunard line received a subsidy from the British
192
government to increase the frequency of its departures in order to compete with new lines
launched by American competitors. The mail packet’s terminus was switched to New York at
that time.
Concerned about the Cunard monopoly, in 1845 Congress authorized subsidies to American
shipping lines that could carry the mail in ships capable of military service in case of war.
Several routes were established under this authorization, including service to France, Central
America, and the Pacific Coast. It was not until 1847 that Edward Collins proposed a line to
compete with Cunard, launching the United States Mail Steamship Company in 1850 with four
ships departing bi-weekly from Liverpool and New York. At nearly 3,000 tons, the Collins
Line’s ships were twice as large as Cunard’s vessels, and Atlantic established a new record for
the west-to-east crossing on its maiden voyage. Unfortunately, two other ships in the line were
lost to accidents with significant loss of life, eventually leading to the demise of the company
(Johnston 1988:241–242).
12.3.7. The Civil War at Sea
The effectiveness and dependability of steamships would be tested in the crucible of war.
The U.S. Civil War would mark the first significant use of steamships in naval warfare, along
with a number of other maritime technological developments, such as iron hulls, submarines, and
shell artillery (Watts 1988:207).
A naval blockade of the Confederate States was a crucial part of the U.S. strategy, but the
Navy was in poor condition. Of 90 ships in the fleet, only 41 were in active service. Many of the
officers were Southerners who joined the Confederacy. However, the Union had the resources to
build up the navy, with most of the major shipyards and industrial facilities located in the
Northern states. The Northern ports were also home to the vast merchant marine fleet that had
been engaged in commercial trade, but was now called upon to contribute to the cause. Secretary
of the Navy Gideon Welles oversaw the naval buildup, purchasing existing vessels and outfitting
them for service, refurbishing existing vessels, and ordering the construction of new ones.
Among the hundreds of ships armored for blockade duty were many paddlewheel and screwdriven steamers (Preston and Natkiel 1986:117–118).
The Confederate naval vessels were superior to this patchwork fleet, but the Union had a
significant numerical advantage. Although Confederate warships scored some significant naval
victories against the Union blockade, with its superior numbers the Union was able to effectively
impact the export of cotton to England and the import of food and other vital supplies to the
South. The Confederacy had more success with its privateers and commerce raiders, which
disrupted Northern commerce by capturing commercial vessels on the high seas and keeping
Union warships preoccupied with protecting its merchant fleet. The steamer CSS Alabama
captured or destroyed 76 vessels from the North Atlantic to the Indian Ocean.
In an effort to break the blockade, the Confederate Navy constructed a fleet of ironclad
gunships, but the Union was able to draw on its foundries and shipyards to counter the effort
with its own fleet of armored ships. The Confederates’ first ironclad was Merrimack, a U.S.
frigate salvaged after the capture of the Norfolk Navy Yard. The Union was aware of the
conversion of Merrimack, and called on Swedish engineer John Ericsson to develop a similar
193
gunboat for the federal navy. Ericsson responded with USS Monitor, a superior vessel that was
more maneuverable and was equipped with a revolving gun turret. Merrimack, rechristened CSS
Virginia had already dispatched with two Union blockade vessels when Monitor appeared on the
scene at the mouth of Chesapeake Bay. Neither ship was able to penetrate the armor of the other,
and the two vessels withdrew in stalemate. The ships engaged each other cautiously over the next
two months, with each captain reluctant to risk the loss of his ship. Virginia was scuttled in the
Elizabeth River, Virginia, in May 1862 when it was unable to escape the Union invasion of the
Peninsula. Monitor was lost in a squall off of Cape Hatteras on New Year’s Eve, 1862. Her
location was unknown until 1973, when magnetometer and side scan sonar identified the remains
in 230 feet of water. Artifacts and large portions of the ship, including the propeller and gun
turret have been recovered from the ship over the last four decades (Preston and Natkiel
1986:119–120; Watts 1988:210–211).
The engagements between Virginia and Monitor demonstrated to both the Union and
Confederacy that the ironclads were indispensible components of an effective naval force.
Approximately 25 armored warships were put into service by the South and served effectively in
river and harbor defense. John Ericsson continued to develop ironclad ships for the Union navy,
and contracts were let for other classes of “monitors,” as they became known, from other
designers, as well. Twenty-seven such vessels saw service for the U.S. Navy in the Civil War.
Three Union monitors—Weehawken, Keokuk, and Patapsco—were lost off Charleston, South
Carolina. In addition to these full-sized armored vessels, many armored gunboats were produced
by both sides for river and harbor defense. These were steam vessels of approximately 100-foot
length, outfitted with one or two guns. Because they were not intended for extended travel in the
open ocean, few of these are known to have wrecked in the federal waters of the OCS. However,
the CSS/USS Atlanta, which was sold to Haiti after the war and renamed Twilight, later sank off
Cape Hatteras and has not been found (Watts 1988:214–217).
While ironclad gunships battled each other for control of the U.S. coast, another class of
vessel, the blockade runner, was being developed to try to slip through the Union patrols and
deliver needed goods to the Confederacy. All manner of vessels were involved in running the
blockade, but the most successful examples were sleek, shallow-draft, steam-powered vessels
designed for speed. British merchants transported goods to neutral ports in the Caribbean and
Bermuda, then loaded the cargo on these smaller, faster vessels. As the Union blockade tightened
around Charleston and Savannah, the favored port for these runs became Wilmington, North
Carolina. At the outset of the war, blockade-running steamers were ships originally designed for
river service, but as these were captured or lost, they were replaced by vessels specifically
designed for blockade running, many of them built in England. Banshee, built in Mersey for this
purpose, was a light, steel-hulled ship, the first of its type to cross the Atlantic. The iron-hulled
Flora had telescoping masts and stacks that could be lowered to make the vessel less visible.
Sailing on moonless nights and in bad weather to avoid capture, as many as 30 steam-powered
blockade runners were lost around the entrance to the Cape Fear River alone (Watts 1988:216–
219).
Yet another innovative vessel put to use during the Civil War was the submarine.
Submersible vessels had been developed by a number of scientists and engineers during the 19th
century, but considerable improvements were necessary to make them useful for military
194
purposes. French engineer Brutus de Villeroi accepted a contract with the U.S. Navy to construct
a submarine for service in Hampton Roads. De Villeroi’s design, which used duck foot-like
paddles, proved impractical. Meanwhile, Confederate financiers, led by Horace L. Hunley,
sponsored the construction of three different submarines, the last of which, named for its chief
patron, became the first submarine to sink an enemy ship when it planted an explosive device in
the hull of USS Housatonic in Charleston Harbor, before itself sinking (Watts 1988:225–230).
The remains of H.L. Hunley were positively identified in 1995, and the vessel, with its doomed
crew of eight still inside, was recovered with much fanfare in 2000. Although submarines had no
significant impact on the outcome of the war, the stage was set for their future development, and
they would play a major role in World War I.
12.4 DECLINE OF U.S. MERCHANT SHIPPING
The Civil War had a significant effect on American shipping. Over 100,000 tons of shipping
capacity had been lost to Confederate commerce raiders and other causes during the war, and
nearly 800,000 tons had been removed from American registry to avoid the same fate. In
addition, the vast forests of the Eastern Seaboard were nearly depleted of the timber needed for
large wooden vessels. Iron and steel were replacing wood, and England was leading the way in
the transition. Shipyards in New York and Boston had given up on wooden ships by 1870.
Startup costs were high for converting to metal hulls, and much of the available raw material and
capital was controlled by the railroads. American labor costs were high, as well. As a result,
American-built ships were considerably more expensive than their British counterparts.
Although large, ocean-going ships were no longer built in American yards, a variety of smaller
vessels such as coastal schooners, fishing boats, work boats, and recreational craft were turned
out by American builders (Bauer 1988:241–243, 289).
During the late 19th and early 20th centuries, U.S. merchant ships concentrated on specialized
trade routes with Central and South America and the Caribbean, while maintaining a steady
coastal trade in iron ore, coal, and other commodities. U.S. shipping lines built their fortunes on
coffee from Brazil, guano from Chile and Peru, and bananas and other tropical fruits from the
Caribbean basin. These trade routes provided service for aging sailing vessels that were being
eclipsed by steam packets in the trade in manufactured goods (Bauer 1988:244–246).
Many sailing vessels were also pressed into service as barges towed by steamships and tugs
along the Atlantic seaboard after the Civil War. Railroads were carrying much of the cotton that
had once been transported by water from the South, and cargo ships increasingly carried bulk
materials for which speed of delivery was not critical. This included coal, lumber, sand, stone,
and lime. By 1907, these products accounted for about 65 percent of the Gulf and Atlantic
coastal trade. All of the coastal trade was less than 10 percent of the tonnage carried by the
railroads. Manufactured goods were still regularly shipped from New England factories to New
York merchant houses by sea, however. Shoes, clothing, and paper were generally handled by
New York concerns, which shipped the products by railroad to the rest of the nation. Cargo ships
were able to transport these goods to New York as quickly, and more cheaply, than the railroads
(Bauer 1988:261–262).
Of the coastal trade products, coal was the most vital, having become the most common fuel
for steamships, railroad locomotives, heating, manufacturing, and electricity. Much of the coal
195
for the Eastern Seaboard came from West Virginia and was shipped by railroad to Newport
News and Norfolk, Virginia, and loaded on to schooners and barges. Multi-masted schooners
remained the most common type of vessel to haul coal long after steamships had supplanted sail
for most commercial shipping. These ships could be built with a large capacity, yet operated by a
small crew. In 1880, Goss, Sawyer, and Packard in Bath, Maine, launched the four-masted
William L. White that was nearly 1,000 tons. It was so popular that the company built 67 more
over the next 10 years. Beginning in 1889, they turned out 55 five-masted schooners that were
even larger still. By the early 20th century, however, these vessels had reached their maximum
practical size, while steel-hulled steam- and gas-powered vessels were nearly boundless in
capacity and were becoming cheaper to operate (Bauer 1988:270–273).
The importance of the coastal trade is reflected in the attention given to safety and navigation
during the late 19th century. The large number of wrecks along the Atlantic coast prompted the
Treasury Department to overhaul the life-saving stations that had been funded by the U.S.
Revenue Marine since 1848, but which had suffered from neglect and mismanagement.
Preventing the loss of lives and property at sea was seen as beneficial to commerce by reducing
insurance rates and encouraging shipping. In 1871, Sumner Kimball, the newly appointed chief
of the Revenue Marine Division of the Treasury Department, ordered a complete inspection of
the life-saving stations, and subsequently secured funding for updating equipment, mandated
adequate training and pay for recruits and superintendents, and ordered more extensive record
keeping (Noble 1988:5–10). The result was not only an increase in the number of ships and
crews saved from total disaster, but a thorough record of accidents along the Atlantic Seaboard.
A number of American steamship lines continued to provide transatlantic service in the late
19th century, despite the obstacles cited above. Many of these lines relied on the less lucrative
immigrant market, while maintaining mail service contracts with the federal government to
supplement receipts. The Red Star Line was the most successful of these, which operated under
the Belgian flag. Another American-owned line that operated under a foreign flag was the Guion
Line, which used British-built and flagged vessels for the Liverpool to New York route. The
company moved toward luxury liners in the late 1870s, but struggled to operate profitably and
went out of business in 1894. American financier J.P. Morgan organized a number of these
American-owned lines, as well as several major foreign-owned ones into the International
Mercantile Marine, which attempted to challenge the Cunard Line’s dominance of the
transatlantic trade. The company overestimated the value of some of the acquisitions, however,
and the loss of the White Star Line’s Titanic in 1912 pushed the company into receivership
(Bauer 1988:246–249).
Coastal steamship lines offered passenger service along the Atlantic Seaboard well into the
20 century, although railroad travel was gradually becoming more convenient for such trips.
The Chesapeake Bay relied on maritime traffic for longer than other regions because of the
difficulties of building railroads over the many bodies of water. The Baltimore Steam Packet
Company, better known as the Old Bay Line, was the principal player in that market, and
maintained its dominance by switching to iron hulls soon after the Civil War. New Orleans,
Houston, and Galveston were key ports on the long-distance coastal routes. Charles Morgan’s
Louisiana and Texas Railroad and Steamship Company was the major player in Houston, and
was later absorbed by Southern Pacific (Bauer 1988:267–269).
th
196
In 1888, the first American-built oil tanker, Standard, was produced by American steel ship
pioneer John Roach for John D. Rockefeller’s Standard Oil Company. Standard Oil gradually
built up a fleet of 60 tankers to transport its crude oil to coastal refineries. The Shell Transport
and Trading Company, with 15 ships, was its only real competitor. However, the breakup of the
Standard Oil monopoly in 1911, coupled with the development of the internal combustion engine
and its widespread use during World War I, created an explosion in the demand for tankers, and
over 300 were built between 1916 and 1921 by the U.S. (Devanney 2006:14–18).
Internal combustion engines also began to be used to power other boats and ships,
particularly smaller vessels, in the early 20th century. This freed the owners from federal
steamboat regulations, which required a licensed engineer to operate the engine, a considerable
expense for modest commercial vessels. Diesel engines were in use in submarines and other
applications during World War I and began to appear in civilian vessels soon after. By the start
of World War II, diesel had largely replaced steam for large vessels then under construction.
Steam continued to be used in turbine engines that powered electrical motors, however. Turboelectric drives developed by Westinghouse were used in U.S. battleships and aircraft carriers.
The turbo-electrics became less popular after the gearing problems that had previously plagued
turbine engines were solved (Bauer 1988:291–293).
The shipbuilding industry revived in the last decade of the 19th century as the Navy built up
its steel-hulled “White Fleet,” and commercial shippers began to replace outmoded vessels that
could no longer be made useful. The Newport News Shipbuilding and Dry Dock Company, a
subsidiary of the Chesapeake and Ohio Railroad, became one of the largest and most efficient
shipyards in the U.S. It continues today as a part of Northrup Grumman and builds nuclear
submarines and super carriers (Bauer 1988:293–296).
The U.S. merchant fleet at the turn of the 20th century was woefully inadequate for the
growing U.S. presence as a world power, despite the growth of the shipbuilding industry in the
1890s. Woodrow Wilson was a strong advocate for the merchant marine and in 1916 Congress
approved his proposal for a U.S. Shipping Board. The board created the Emergency Fleet
Corporation (EFC) that built up the merchant marine with a fleet of utilitarian, contract-built,
wooden- and steel-hulled vessels that proved seaworthy if not glamorous. The goal was to build
a “bridge of ships” to France. Attacks on merchant and passenger ships by German submarines
had drawn strong rebuke from the U.S., which was finally drawn into World War I in early 1917.
Nearly two dozen ships were sunk by German U-boats before the end of the war.
The EFC building program led to the construction of dozens of new shipyards and fabrication
plants, although much of the proposed work was not completed before the end of the war. Still,
the program introduced the approach of separating the fabrication and assembly facilities to
shorten production time, which spread the benefits of the program along the coasts of the entire
country. In two short years, the number of “ways” (the racks upon which ships were built) in the
nation’s shipyards had increased more than four-fold. By 1922, the U.S. had the largest merchant
marine fleet in the world and controlled 22 percent of the available tonnage. Unfortunately, it did
not have the shipping industry or trade connections to fill the holds (Bauer 1988:297–301).
The glut of ships brought shipbuilding to a halt and left the federal government with
hundreds of vessels that it could not afford to operate. An effort to sell or lease these ships and
197
their routes to private enterprise met with little success. The EFC fleet also delayed the use of
diesel engines in large craft, since the vessels were built before improvements were made in the
engines during the war. Diesels were quickly adopted by work boats, fishing vessels, and
recreational boats, however. Government subsidies for overseas mail service spurred some
companies to start or revive shipping lines, and funds from the Shipping Board for new
construction resulted in modest production of coastal vessels to replace aging fleets. The value of
the board was further brought into doubt by cases of corruption that emerged in the 1930s
regarding these federal subsidies and contracts (Bauer 1988:302–306).
With the rise of fascism in Europe, President Franklin Roosevelt suspected that the U.S.
would be drawn into war and was convinced of the need for a strong merchant marine fleet. He
overhauled the Shipping Board with the creation of the U.S. Maritime Commission. The
Commission proposed the construction of 500 vessels by 1947 under its Long Range
Shipbuilding Program. By 1940 about 150 well-designed and versatile cargo ships had been
built, which served well as transport ships during World War II. However, the heavy losses
incurred by the British at the hands of the German U-boats showed that a more urgent effort was
required, and in 1940 the Long Range program was scrapped for the Emergency Shipbuilding
Program. The emergency building program produced over 2,600 “Liberty Ships,” the hastilybuilt, but serviceable cargo steamers that were not intended for post-war service. In addition, 534
“Victory Ships,” an improved version of the Liberty Ships, were completed as well. A large
number of these were sold after the war to foreign countries, helping to re-establish a worldwide
shipping industry (Bauer 1988:306–311; Butler 1997).
12.4.1. Losses in U.S. Waters in World War II
War activities resulted in a large number of shipwrecks in the Atlantic OCS area, not only
from German submarine activity, but from training exercises and accidents involving all manner
of vessels and aircraft.
The U.S. staunchly maintained its neutrality through the 1930s, but with the fall of France
and Italy’s entry into the conflict on the side of the Axis powers in 1940, Britain’s ability to
control the seas and supply its Allied forces was seriously imperiled. The U.S. came to the
Allies’ aid with the transfer of 50 U.S. World War I destroyers to Great Britain in September
1940 and the passage of the Lend-Lease Act in March 1941 that provided for a steady supply of
materiel in exchange for various in-kind payments. Roosevelt’s promise to provide all aid short
of war provoked Germany to ramp up its submarine warfare effort against U.S. shipping. In
May, a German U-boat sank an American merchant ship carrying no military supplies off the
west coast of Africa, making it clear that U.S. shipping would require military support to remain
unmolested at sea. In early 1942, Germany launched Operation Drumbeat, sending U-boats to
the U.S. east coast, refueling them from tanker ships. With no blackouts in effect, the submarines
could track ships at night by spotting their silhouettes against the lights of shore. Unarmed, the
merchant vessels were sitting ducks. Nearly 400 ships were destroyed and 5,000 seamen killed in
a six-month period before the U.S. implemented the necessary changes to curtail the losses and
began assigning full convoy protection to its merchant ships. With these protections in place, the
Germans shifted their focus back to massed attacks in the North Atlantic against the heavily
guarded convoys (Butler 1997:175–177).
198
One of the first vessels sunk by German submarines off the Atlantic Coast was SS Norness, a
German-built, Panamanian-flagged oil tanker of approximately 9,600 tons that was torpedoed by
U-123 about 60 miles off Montauk, Long Island. One of the largest vessels taken out by the Uboats was MV Amerikaland, a German-built twin-screw diesel cargo ship of over 15,000 tons
built in 1925. It was sunk about 75 miles off of False Cape, North Carolina. Another large vessel
lost off the coast of North Carolina, Ulysses, was a 14,647-ton, steel-hulled British merchant ship
that was torpedoed about 445 miles south of Cape Hatteras in April 1942 by U-160 (uboat.net
2011).
James E. Longstreet, lying in 14 feet of water off East Brewster, Cape Cod, is an example of
an intentional shipwreck used for military training. The 7,176-ton Liberty Ship owned by the
U.S. Maritime Commission, was used for target practice by Avenger aircraft and other small
bombers (IRSS). USS Bass, a 2,500-ton SS-164 battleship was used for target practice 7.5 miles
south of Block Island Southeast Light. The ship had earlier burned off the coast of Panama
(Brown 2008).
The Battle of the Atlantic continued into 1943, by which time the Allies had developed
improved tracking technologies and weapons that were finally successful in blunting the Uboats’ attacks. Also of consequence was the sheer number of ships that the Allies were able to
bring to the convoys, most of them American built, which gradually eroded the German effort to
isolate Great Britain and force its surrender.
Many of the vessels sunk during Operation Drumbeat, including lost German U-boats, have
been identified or recovered by modern salvors and diving enthusiasts. NOAA has conducted
search expeditions over the last two years under its Battle of the Atlantic program, an effort
conducted in consultation with the British and German governments, with technical expertise
and logistical support from various federal agencies, including the BOEMRE, and numerous
colleges and universities. In 2008, the Battle of the Atlantic expedition identified the remains of
three German U-boats. In the summer of 2009, the remains of the retrofitted fishing trawler YP389, which was sunk off Cape Hatteras in June 1942, were identified (NOAA 2009).
12.4.2. U.S. Shipping and Maritime Activity since World War II
World War II established the U.S. as a world power and put the U.S. merchant marine back
on the world stage. The Marshall Plan, launched in 1948, put billions of dollars worth of cargo
into American ships destined for Europe and other areas affected by the war. In that year, there
were 3,644 U.S.-flagged ships operating in international trade, the most ever for the country.
However, postwar aid also benefited the other shipping powers of the world, and over the next
50 years the U.S. gradually fell into the background.
The U.S. Marine Commission launched a new type of vessel in 1950 to replace the Liberty
and Victory ships, which were in need of updating. The Mariner class, a 14,000-ton, steam
turbine vessel with updated electronics and rigging, was capable of a remarkable 20 knots, and
the ships served well in the Korean War. Two dozen were put into service by three different
shipping lines in the first five years and over 50 were eventually built. However, diesel engines
were proving more economical if somewhat less powerful than steam. In addition, container
199
ships, with their large undivided cargo holds were replacing the “stick ships,” of the early 20th
century (Butler 1997:196).
Labor cost, government regulation, and a shift in the U.S. from manufacturing and trade to a
service economy resulted in the decline of U.S. owned and operated shipping lines. Vast
container ships now fly flags of convenience and carry small international crews, few of whom
are American.
Passenger liners, long considered the most fashionable and luxurious way to travel, could not
compete with the speed and convenience of airlines. The giant luxury liners America and United
States of the United States Lines no longer provided transatlantic passenger service by the end of
the 1960s. However, the U.S. retained a strong presence in the tanker industry, carrying 21
percent of the world’s oil in 1954 (Butler 1997:199; Sloan 2004).
The second half of the 20th century saw significant improvements to navigation technology,
with the widespread availability of radar, sonar, and Loran positioning systems. The Coast
Geodetic Survey and the Hydrographic Office compiled modern charts based on data gathered
during the war effort, and the satellite mapping and monitoring beginning in the 1960s made it
possible to update data more rapidly (Butler 1997:197).
Increased leisure time and disposable income led to a dramatic increase in the number of
recreational vessels, both sail- and motor-powered, in U.S. waters after World War II. Chartered
fishing and diving boats also were increasingly common. While increased regulation and
technological advancements continued to make commercial vessels safer, recreational and
charter vessels were more likely to be involved in accidents as a result of their small size,
inadequate safety precautions, poor maintenance, and inexperienced or impaired operators. Coast
Guard Disaster Files at the U.S. Coast Guard Headquarters in Washington, D.C., document
accidents involving private vessels for which they were involved in the rescue efforts.
12.5 DISCUSSION OF VESSEL TYPES
Mariners are known for their extensive jargon, and identifying vessels and their components
is an integral part of the maritime vocabulary. As with any typology, vessels could be
categorized in a variety of ways, and naming conventions are notoriously inconsistent (U.S.
Bureau of Marine Inpsection and Navigation 1886). Sailing vessel types were based on size, hull
shape, and rigging, as well as function. With the advent of engine-driven vessels, the type of
engine and type of propulsion came to define the type of vessel. Increasingly specialized vessels
also contributed to an expansion in the number of vessel types referred to in contemporary
nomenclature.
For the purposes of the current shipwreck database, vessel type is based on that given by the
source of information on the wreck, using the codes and types developed by Pearson et al. (2003)
for the Gulf of Mexico shipwreck database. Many of the types listed by Pearson et al. were not
encountered in the current investigation, and a number of new types were added that could not be
easily fit into the existing typology. These include the ship of the line, man of war, pilot boat ,
scow, and the ship’s boat, which includes the long boat, shore boat, and tender.
200
Although there is considerable variation within types, the characteristics of major vessel
types can help to identify wrecks in the field. Descriptions from secondary and contemporary
sources are discussed below. They are drawn from Gibbons (2001), László and Woodman
(1999), Culver (1992), Kemp (1980), and U.S. Bureau of Marine Inspection and Navigation
(1886). Figure 12.1 identifies some of the major components of a sailing vessels’ rigging, by
which many vessels were categorized.
Figure 12.1. Components of a historic sailing vessel.
12.5.1 Sailing Vessels of the Age of Exploration
The earliest European vessels to be found along the Atlantic Seaboard of the United States
were the sailing vessels of the Spanish, English, Dutch, and French explorers of the 15th, 16th,
and 17th centuries. These vessels developed from a merging of Northern and Southern European
traditions during the late Middle Ages to create a sailing ship that was adaptable to long periods
at sea and could further the ambitions of European powers seeking to extend their influence. The
Mediterranean caravel was built on a frame that allowed for larger hulls than the cogs of
Northern Europe, while the square-rigging of the cog proved better suited to the demands of
ocean travel than the lateen sails used in the Mediterranean. The resulting vessel was called the
201
caravela redonda, which was one of the vessel types used by Columbus and the Portuguese
explorers of the 15th century.
Caravel. The fore-and-aft rigged, carvel-built vessels developed by the Spanish and
Portuguese in the 15th century were the standard exploration ship. The butted and caulked planks
on a frame provided a stronger longitudinal hull, which could carry greater amounts of cargo and
handle the stresses created by the increasingly elaborate sailing rigs being developed. The larger
caravels were typically fitted with four masts, and were capable of long days at sea. However,
the fore-and-aft rig on the front proved difficult to handle in severe weather and was replaced by
square-rigging in the caravela redonda. A caravel fitted with armaments was called a caravela
de armada. The traditional, lateen-sailed caravel was known as the caravela latina. Ocean-going
caravels were often referred to simply as naus, a term for a large merchant vessel, usually with
three masts.
Carrack. The carrack was a large
merchant ship and warship that carried a rig
similar to the caravela redonda, that is, a
mixture of square sails in the front and lateen
sails on the mizzen mast. They typically
carried a large mainsail, and were
characterized by prominent fore- and
aftercastles, which made them cumbersome in
a crosswind. Their robust construction and
large size made them intimidating war
vessels. Guns mounted in the castles made the
carrack top-heavy, and were eventually
moved below decks with the development of
the gunport. As the size of the carracks
increased, the masts were increased in number
and height, and some vessels approached
2,000 tons. In the 17th century, however, the
carrack was gradually replaced by the sleeker
and more maneuverable galleon. A replica of
Columbus’s carrack Santa Maria is shown in
Figure 12.2.
Figure 12.2. A replica of Santa Maria, Columbus’s flagship (Creative Commons AttributionShare Alike 2.0 Generic license).
Galleon. Originally a larger carrack refined for ocean voyages, the galleon developed into a
sleek, versatile craft that was well-suited to the needs of national fleets guarding convoys of trade
vessels. The prow was lowered and the large mainsail incorpated into a more balanced sail plan.
At the end of the 16th century, English ship designer Sir John Hawkins created the race-built
galleon, a version which further improved its speed and maneuverability by eliminating the high
forecastle, lowering the aftercastle and sides of the ship, and replacing the round stern with a
202
square one. The galleon was soon adopted by all the European navies and served as the basis of
fighting vessels through the end of the age of sail.
12.5.2 Naval Vessels of the Age of Empire
The success of the galleon led to the development of a number of specialized vessels that
provided protection to the fleets of the Spanish, Portuguese, French, English, and Dutch as they
sought to control the world’s trade during the 17th and 18th centuries. Increasingly large gunboats
were also constructed to engage the enemy’s armadas.
Frigate. This term came to be applied to warships fitted with broadside guns. Frigates were
generally single-decked, 3-masted vessels carrying between 28 and 36 guns. They served as
scouts, interceptors, and escorts. The hull design was low, emphasizing speed and sea-worthiness
in all conditions. The earliest examples were used as merchantmen as well a gunships, but by the
middle of the 18th century, the true frigate emerged as a dedicated fighting ship. In the second
half of the century, Britain and France built increasingly large frigates for use in their on-going
war of the seas. These 40- to 50-gun vessels proved as effective as double-decked ships with
equal numbers of guns. The first six vessels constructed for the U.S. Navy at the end of the 18th
century were frigates, with the 44-gun Constitution being the most well-known. In the 19th
century, what were essentially two-decked frigates were constructed by connecting the forecastle
and quarter deck in a continuous line to form the spar deck, where a second row of guns was
mounted. The frigate USS Boston, constructed in 1799, is shown in Figure 12.3.
Figure 12.3. The frigate USS Boston in the Mediterranean in 1802 (Public Domain).
203
Ship of the Line. Multiple-decked vessels with 40 or more guns were called ships of the line,
or line-of-battle ships. In the standard battle formation that developed in the 17th century, these
ships took up position in a line, engaging the enemy ships in a parallel line as they sailed with the
wind. The largest of these ships, such as the British Soveriegn of the Seas built in 1637, were
three-decked, 100-gun vessels. Eventually, 120- and 140-gun ships were built, but these were
rare because of the cost and restrictions on size imposed by wooden construction.
Man-of-War. This term was applied to any number of armed sailing vessels, but generally to
those of larger size.
East Indiaman. These Dutch ships of the colonial era were primarily trading vessels
outfitted with guns to fight off pirates and privateers. They were deep-draught, three-masted,
square-rigged vessels of 500 to 1,200 tons. Their large capacity made them slow, but they were
well-armed and presented a low profile to an attacking vessel. The vessels were well-appointed
and often richly decorated, and were the pride of the Low Country. The flute was a smaller,
three-masted Dutch vessel of the 16th and 17th century that was confined primarily to European
waters.
12.5.3 Coastal vessels
Cutter. Like the frigate, the term cutter was applied to a large number of small, fast vessels
used as patrol boats, escorts, and privateers. The sailing cutter typically had one mast with a gaffrigged mainsail and a number of foresails set on a long bowsprit. The gaff-rigged sail is a foursided, fore-and-aft sail that is attached to the mast by one edge rather than in the middle like a
square sail. Cutters sometimes carried square topsails, as well. They were distinguished as much
by their sharp lines as their rig. Their speed and ease of handling made them popular with
smugglers as well as revenue officers. They were also employed as pilot boats using a simpler
rig.
In modern usage, the Customs Service and the Coast Guard continue to refer to all of its
patrol boats, including motor vessels, as cutters. The term cutter is also used for a type of yacht
with a Bermuda-rig (triangular mainsail) or gaff-rig.
Lugger. Originally used as a coastal fishing vessel, the lugger is a small ship with with one
or more masts rigged with lug sails. Lug sails are a type of fore-and-aft rigged sail that have four
sides and are secured to a yard that is attached to the mast off-center, allowing one side of the
sail to be higher than the other. With lugger topsails added, the vessel could carry a vast amount
of sail. The French lugger, or chasse-marée, was used for fishing, as well as shore patrols and
privateering. They served a similar role to the British cutter and American schooner. In the
United States, a lugger generally referred to a single-masted coastal sailer that could be operated
by a single pilot.
Ketch. The ketch is a two-masted sailing vessel used for coastal trading and fishing.
Originally rigged with square sails and used as gunboats, they evolved into fore-and-aft rigged
vessels in the 19th century that were useful as a traders, trawlers, and packets. They differ from
schooners in that the forward mast is taller than the mizzen.
204
Scow. A flat-bottomed, blunt-nosed cargo vessel designed for trade in coastal and inland
waters. They used retractable centerboards for stability and could be beached to unload cargo.
They typically had a main and a foremast with a fore-and-aft rig on both.
12.5.4 Sailing Vessels of the Mercantile Era
Ship. The term ship was used for all decked sailing vessels, but also referred to a specific
type of large sailing vessel equipped with 3 or 4 masts and a bowsprit. The forward masts were
all rigged with square sails and included main, top, topgallant, and royal (uppermost) sails. The
sails on the aft mast were sometimes fore-and-aft rigged. Ship-rigged vessels are often referred
to as “tall ships” today and were historically known as “lofty ships.” Figure 12.4 shows a fullyrigged, “double topsail” ship.
Clipper. The famous clipper ships originated in the Baltimore shipyards and were named for
their ability to clip through the water, as well as clip time off of ocean crossings. They were built
for the China Tea Trade, as well as other routes and jobs where speed was of the essence.
Clippers were fully rigged ships of two or three raked masts and could be schooner rigged or
square-rigged. The distinguishing feature of the clipper ship was its streamlined shape, with a
narrow beam and a concave bow that cut through waves like a wedge. Clippers reached their
peak in the 1840s but declined after the Civil War, with the opening of the Suez Canal and the
completion of the transcontinental railroad in the United States in in 1869.
Schooner. The workhorse of the American coastal trade in the nineteenth century, the schooner
is a fore-and-aft rig ship of two or more masts (Figure 12.5). It often was fitted with small,
square topsails. Schooner rigs were used on smaller vessels in the early nineteenth century, but
soon were employed on larger and larger vessels, with increasing numbers of masts. By the
1880s, schooners of 800–1,000 tons with 3–4 masts were not uncommon.
Bark (or barque). A bark has three masts, with square-rigged sails on the fore- and main
masts and fore-and-aft sails on the mizzen mast (Figure 12.6). It is typically smaller than a fully
square-rigged vessel, although barks sometimes exceeded 1,000 tons.
Barkantine. The barkantine is a smaller version of the bark, with square sails only on the
foremast and fore-and-aft sails on the main and mizzen. It was designed with a long, narrow
body for easy maneuvering in inland waters.
Brig. A 2-masted, square-rigged vessel, designed for speed and manueverablility, the brig
was a versatile ship. It was employed both as a fighting vessel and a merchant ship, and was
most popular during the 18th and early 19th centuries. The mainsail was sometimes a fore-and-aftrigged, but with a square-rigged topsail and topgallant sail (Figure 12.7). On a hermaphrodite, or
half-brig, the foremast was square-rigged and the mainmast was schooner-rigged (all fore-and-aft
sails). The brig was similar to the snow, which differed only in having a square mainsail, with a
spanker, or driver sail, attached by hoops to an auxillary spar abaft (toward the stern of) the
mainmast. The brig’s spanker was larger and boomed, serving as the vessel’s mainsail.
205
Figure 12.4. Four-masted, double topsail ship, commonly refered to as a “tall ship.”
Figure 12.5. A typical schooner-rigged sailing vessel of the 19th century (Public Domain).
206
Figure 12.6. Bark-rigged vessel (Public Domain).
Figure 12.7. A typical brig of the late 19th century (Public Domain).
207
Brigantine. A smaller version of a brig that was rigged similarly to the half brig, but with a
small topsail over the fore-and-aft mainsail. It actually falls between the half brig and brig in
terms of the amount of fore-and-aft rigging on the main mast.
Sloop. Generally used for coastwise trade, but occasionally making ocean voyages, a sloop
has one mast rigged with fore-and-aft sails, including a stay foresail rigged to the bowsprit. The
sloop is easily handled by a small crew and is reliable in a variety of conditions.
Collier. The collier is a vessel designed to carry coal. Until the advent of steamships, coal
was carried by brigs and other sailing vessels equipped with large holds. Only 300 to 400 tons of
coal could be carried on these vessels. The steam collier emerged in the 19th century to transport
coal to depots along trade routes where coal-fired steamships could refuel. These could carry as
much as 6,000 tons.
12.5.5 Unpowered Vessels
Lighter. A lighter is a flat-bottomed boat used to transport goods between a cargo vessel and
shore. They are towed or pushed by tugs. They are used where deep-draught vessels cannot reach
the docks.
Barge. Flat-bottomed vessels used for transporting goods, barges can be powered, but more
typically on the Atlantic Seaboard were towed. Towed barges became common during the
second half of the nineteenth century carrying bulk items that could be transported more
inexpensively by boat than by railroad, including coal, sand, wood, and lime. Scores of these
barges were lost in storms while transporting goods along the coast. Having no power, the barges
were an encumberance to the towing vessel in a storm and were frequently cut loose, leaving
their small crews and the vessels at the mercy of the waves.
Schooner barge. As steam vessels became more reliable in the late 19th century, old coastal
schooners were converted to barges and towed by steam ships. In some cases the rigging was
removed to create more usable space, but other barges retained their masts and might have used
their sails in an emergency.
13.5.6 Steamships of the 19th and 20th Centuries
Paddlewheels. The earliest steam vessels were paddle-driven, with the paddles mounted on
the stern or on the sides of the vessel. Advances in engine and propeller design led to a decline in
new paddlewheel boats starting around the mid 19th century, although the requirement of certain
mail contracts for wooden-hulled, paddle-driven vessels kept them in service until after the Civil
War. The design of these vessels varied greatly, but ocean-going steamers of the mid 19th century
generally had a full sailing rig (bark or brig rigging), with the paddles and smokestack amidships
(most commonly between the foremast and the mainmast). Figure 12.8 shows SS Sirius, a
paddlewheel steamship with a full sailing rig. They ranged from 500 tons up to the 5,888 tons of
the Collins Line’s Adriatic, built in 1856. The massive, 18,915-ton Great Eastern, built in 1858,
was equipped with two 56-foot paddlewheels, as well as a single screw and six masts carrying
15,000 square feet of sail.
208
Figure 12.8. SS Sirius, which crossed the Atlantic in 1838 in 18 days (Public Domain).
Screw-driven Steamers. Experiments with propeller-, or screw-driven vessels date back to
the late 18th century, but it was not until the 1830s that the first successful screw steamers were
put into service in England. In 1840, S.S. New Jersey, built as Robert F. Stockton in England in
1848, became the first screw steamer to operate in U.S. waters. A variety of propeller designs
have been used, including fully rotated screws and multi-bladed designs. Large ships were
typically fitted with two propellers rotating in opposite directions to counter the effect of heeling
torque, which pulls the bow of the boat in the opposite direction of the prop rotation.
Screw steamers were similar in appearance to their paddlewheel counterparts, with the
exception of the lack of the paddlewheels and their casings amidships. In some cases, the engine
and stack on screw steamers were located farther toward the stern, abaft the mainmast.
Passenger steamers. The earliest steamships to carry passengers operated on rivers and
inland waterways, including the famous Fall River Line that operated in Long Island Sound
between New York and Fall River, Massachusetts. By 1850, steamships were carrying
passengers between the major cities of the Eastern Seaboard. Passenger steamships often carried
mail and other goods at a premium freight rate because of their reliability and regularly
scheduled service. The ships were designed with multiple decks to increase the number of berths
available for passengers.
Steam Engines. The type of engine used on a vessel might be classified by its cylinder type
or its driving mechanism. The earliest steam vessels used vacuum engines, which relied on the
compression resulting from the condensation of cooling steam to move pistons. This was an
inefficient method, since the amount of pressure generated was limited by the atmospheric
209
pressure. Expansion engines were soon developed to drive locomotives and ships and were more
commonly found on steam vessels than the vacuum engine. A simple expansion engine had only
one stage of expansion, so that all the cylinders operated under the same pressure.
The invention of the compound engine in 1855 resulted in greater efficiency in the engine by
expanding the gas in steps. The first compound engines used two stages, with later version
expanding to three and four stages. These engines would have compartments for the pistons and
chambers that might be evident in the remains of a wrecked vessel.
Prior to the invention of the surface condenser, which allowed water to circulate through the
engine in a loop, boilers were fed with salt water. This made it necessary to clean the boilers
frequently. Samuel Hall developed the condenser in 1834, allowing the engine to have a looping
system that brought the condensed steam back to the boiler. Prior to that date, the engines on
steamers would not have a closed loop setup.
Most 19th century vessels used some variety of reciprocating drive, with turbine drives
coming in after 1884. In a reciprocating drive, a piston moving in and out of a cylinder drove a
mechanism linked to the drive shaft. The link on most paddlewheel steamers was a beam or
sidearm linkage. The pistons drove an iron arm up and down on either end of a fulcrum, much
like a see-saw. The cylinders were vertical and the heavy arm was located at the base, making
the setup ideal for ships, which needed a low center of gravity.
13.5.7 Modern Motor Vessels
The development of the reciprocating marine diesel engine around the turn of the 20th
century provided the last significant development for marine propulsion outside of experimental
and nuclear craft. The diesel engine operates more cheaply and efficiently than a steam engine,
occupies less space in the hold, and is easier to maintain, making it the ideal choice for vessels
formerly powered by steam. The gasoline engine, with its greater thermal output, is more
commonly used for smaller vessels with higher cruising speeds.
Cabin Cruiser. Ocean-going cabin cruisers are mid-sized recreational boats with an
enclosed cabin providing accommodations for a small crew. They range in size from 25 to 40
feet, with larger versions generally referred to as yachts. They typically have a deck in front and
an open well in the rear, with the cockpit on top of the cabin amidships. Early 20th century
models were constructed of wood or metal, while fiberglass was introduced about 1950.
Yacht. The modern yacht is a large pleasure craft fitted with accommodations for passengers
and crew. Motor yachts are vessels of over 40 feet in length. Luxury yachts can reach over 200
feet in length, but these are more in a class of Superyachts, or even ships. During the late 19th
and early 20th centuries, a number of European royal families engaged in yachting, acquiring
ever larger and more luxuriously appointed vessels to demonstrate their great wealth and style.
Such conspicuous consumption was also practiced by international businessmen and American
industrialists and financiers. Open water sport fishing vessels with enclosed cabins are often
referred to as yachts and are typically 30–60 feet in length. During the first half of the 20th
century, yachts were made primarily of wood or metal, but since about 1960, many are made of
fiberglass.
210
Trawler/Seiner/Fishing Boat. Generally a screw-driven vessel with booms and winches for
deploying trawling and seining nets, a working deck, storage holds below decks for the catch,
and a superstructure for the pilot house and crew accommodations. These boats have a variety of
different rigs and deck layouts depending on the type of task for which they are outfitted.
Smaller boats typically have the working deck in the rear and the superstructure in the forward
part of the ship. In trawlers, the booms are attached to a center mast and are lowered as
outriggers. This layout is common in shrimp trawlers. Seining boats have a boom that is parallel
to the keel for reeling in the net after it is closed.
Large factory fishing vessels may have a working deck midship and an aft superstructure.
They use a type of outrigger called a beam trawl. Other trawling rigs include otter trawlers, pair
trawlers (two boats pulling a single trawl), side trawlers, and stern trawlers. Some factory fishing
boats use longlines (baited hooks), jigs, or gillnets. Most larger fishing boats are constructed of
wood or metal, with fiberglass used for some smaller boats.
Pleasure Craft. A variety of undecked motor vessels travel in the open waters of the ocean,
including runabouts, pontoon boats, and skiffs, though most do not venture far beyond the 3-mile
line marking federal waters.
Dive/Exploratory/Research vessel. These custom-rigged vessels might have a variety of
hull and engine configurations, and are equipped for open water diving or research. Dive vessels
can be used for day trips or overnight and carry only a few passengers or dozens. Their decks are
designed to accommodate scuba divers and their gear, and their sterns are often fitted with
lowered decks and ladders to facilitate entry and exit.
Research vessels carry data collection equipment that might include remote sensing devices,
sample collection equipment, and climate and water monitoring devices. They can range in size
from 40-foot cabin cruisers with a handful of crew to 200-foot multi-deecked ships with dozens
of crew members. Research vessels are operated by private foundations, universities, government
agencies, and military organizations. Major types of research vessels include hydrographic
survey, oceanographic research, polar exploration, and fishing research vessels. Vessels with
unusual hull types include icebreakers and SWATH (Small Waterline Area Twin Hull)
catamarans with submerged hulls like RV Kilo Moana, an oceanographic research vessel of the
University of Hawaii.
Cutters. Originally a term used for a small, fast coastal sailing vessel, the term came to be
applied to a broad class of vessels that includes patrol boats, buoy tenders, and rescue boats.
Cutters generally refer to vessels operated by the U.S. Coast Guard (Figure 12.9). These types of
working shore boats can vary greatly in terms of hull shape, deck configuration, and propulsion.
Pilot Boats. Pilot boats fall within a range of 20–75 feet and are designed to be fast and
durable to withstand heavy seas and frequent bumping with large vessels when transporting a
pilot to a ship to be brought into harbor.
211
Figure 12.9. Coast Guard cutter Hamilton, a typical coastal patrol boat, in the Navy Yard in
Norfolk in 1898 (courtesy of U.S. Coast Guard Historian’s Office).
13.5.8 Modern Sailing Vessels
Sailboat. Personal, recreational sailing vessels without enclosed cabins are commonly
referred to simply as sailboats. They have simple sail plans that can be easily operated by a crew
of one or two.
Sailing Yacht. A sailing yacht might be any size vessel that includes accommodations for
the crew, usually 25–40 feet in length. Sailing yachts are typically single-mast, fore-and-aft
rigged vessels with foresails and a spinnaker, the large, triangular sail deployed in front of the
vessel when sailing downwind. Larger luxury yachts might have more complex rigs.
12.5.9 Modern Work Vessels
Freighter. Modern, ocean going freighters can reach the length of three football fields and
can reach nearly 100,000 gross tons. Container ships are designed to carry standard-sized
containers. The containers are typically the size of a truck trailer and can be loaded onto a flatbed
without unloading the contents. Freighters are often equipped with cranes for loading and
unloading cargo, although container ships are usually unloaded with gantry cranes in port.
212
Tanker. Vessels designed to carry liquid in large holds have existed since at least the 19th
century and have carried wine, molasses, and oil. Tankers have evolved to include a variety of
vessel sizes, with some supertankers in the Ultra Large Crude Carrier (ULCC) class exceeding
200,000 gross tons and over 1,000 feet in length. These mammoth vessels cannot fit through the
Suez or Panama canals and must travel around the capes of Africa and South America to travel
between oceans. Crude tankers are designed to carry raw crude from their extraction point to
refineries. Smaller vessels, known as product tankers, are designed to carry refined products
from processing facilities to market. They range from 10,000–60,000 DWT.
Tug. These vessels are specifically designed to tow or push barges or other vessels. They
typically have high bows and a short forward superstructure. The engine occupies a large part of
the vessel’s usable space, but modern tugs are often equipped with firefighting equipment as
well. Tugs today have diesel engines, but the earliest examples had steam engines. Power and
maneuverability are critical on a tug, so the drives often have multiple screws or directable
thrust. Harbor tugs might range up to 500 tons and have 2,500-h.p. engines, while ocean going
tugs might be 2,000 tons and have 15,000-h.p. motors. The larger tugs are used for rescue and
salvage of distressed ships, as well as for towing off-shore drilling rigs and work platforms.
Drilling Rig/Oil Platform. Not technically vessels, these large work stations might
nevertheless contain living quarters, scientific equipment, and marine-related vessels. They can
be fixed to the ocean floor or floating and have collapsed, toppled, and capsized. Radio towers
and other communications towers have also become underwater wreckage. The Texas Tower, a
dive site off Long Island, is an Air Force radar platform built in 1955 that collapsed in a storm in
1960.
12.6 SHIPWRECKS
12.6.1 Number of Shipwrecks in the Atlantic OCS Survey Area and Perservation
Conditions
The number of shipping losses during the historical period in the Atlantic is staggering.
Earlier estimates for the number of shipwrecks off the Atlantic coast ranged from 15,000–20,000
(SAI 1981:III-19), but the totals can vary widely based on the criteria used in compiling the list.
Over 10,000 entries are included in the ASD, which contains duplicates, but which excludes
wrecks that appear to be within state waters. The IRSS database includes approximately 12,400
listings and includes wrecks in both near-shore and open water wrecks. According to Bascom
(1976), only 10–20 percent of historic losses were in open water, with the remainder occurring
on shore or on submerged rocks, reefs, or bars. This would mean that for a database such as the
IRSS, only 1,500–2,500 wrecks are off-shore. However, some rocks, reefs, and bars are more
than 3 miles from shore. Furthermore, primary source research for the current investigation
produced hundreds of wrecks that had not been listed in previous databases, and certainly others
remain that have not been cataloged. Among the earliest compiled statistics are those of the
volunteer life-saving service established on the New York and New Jersey shores, where the
service reported approximately 500 shipwrecks between 1839 and 1848 (U.S. Life-Saving
Service 1894). Although this figure may represent a peak period for shipwrecks, falling in the
golden age of maritime trade and predating a number of modern developments in navigation and
shipbuilding, extrapolating from these statistics over a coastline seven times longer and a period
213
of 50 decades, it is not hard to conclude that the number of ships lost would be in the tens of
thousands. Therefore, a figure of 10,000 or more does not seem unreasonable for wrecks within
the OCS project area.
The number of reported shipwrecks is considerably greater than the number of extant wrecks
on the ocean floor, however. Ships that went to pieces on bars or in severe storms did not leave
intact remains on the bottom. Although parts of these ships might be recovered at some time, it is
unlikely that they will ever be identified or yield significant historical information. Shallow
wrecks were often recovered by salvors, sometimes years after the disaster. Although ships that
sank in open water are more likely to remain intact and are less susceptible to damage after
sinking from storms, currents, and wave action, many of the ships lost in the Atlantic OCS have
been covered by shifting sands, eroded by water and abrasion, corroded by chemical and organic
processes, or carried away by currents and tides, and are unlikely to ever be found.
Shipwreck preservation on the OCS between the 3 mile territorial waters of the Atlantic
seaboard states and the shelf break is related to a broad spectrum of environmental factors that
vary from Maine to Florida. Over that area water temperatures vary considerably. Overall, colder
waters of the Labrador Current that parallel the coastline north of Cape Hatteras contribute to
better preservation than waters warmed by the Gulf Stream south of Cape Hatteras. Differences
in temperature also have an effect on various types of biological activity that impact organic and
inorganic preservation at shipwreck sites. Water depths are also a significant consideration in
shipwreck site preservation. In the most general terms, the deeper a site is found, the better it is
likely to be preserved. Depth can be related to additional factors that play a role in preservation
such as temperature, salinity, oxygen, sunlight, and motion dynamics. The nature of sediments
also vary from north to south on the OCS. Rock and gravel bottoms more prevalent in the north
do not necessarily provide the level of protection afforded by sand, silt and mud, which are more
often found along the Middle and South Atlantic seaboard. Bottom conditions are also regionally
specific. In the northeastern OCS, extensive sandbanks can be found offshore while to the south,
areas of hardbottom can be found. Where wreck remains are covered by accreting sediment,
preservation is generally much better. The more rapidly a shipwreck is buried, the more likely it
will be well preserved.
Shipwreck preservation also depends on the type of vessel and the nature of its loss. In
wooden vessels, preservation can depend on the type of wood used. Some woods are more
resistant to deterioration than others. Cedar and teak often have a greater resistance to decay than
pine and and spruce. Oak, a much preferred shipbuilding material due to its strength, is also
subject to rot and unlike cedar and teak, and is attractive to various marine organisms such as
teredo navalis, a mollusk in a group of organisms known as shipworms. Fastners can also have
an impact on structural preservation. Brass and bronze having greater resistance to salt water
than iron. Wooden vessels can also be more heavily impacted by extreme surface conditions that
can break up the structure. Unless weighted down by ballast, cargo or ordnance, the fragmentary
remains of a wreck can float, distributing structural elements over a broad area. Once deposited
on the bottom in water depths as deep as 200 feet, surface conditions can continue to impact
structural remains.
214
Iron hull vessels may have a higher potential for structural integrity, but they are not
necessarily more resistant to the forces of nature than wood ships. Dissimilar iron used in plates
and rivets can be highly electrolytic. Electrolysis is one of the major causes for iron and steel
vessel deterioration. When iron and steel hulls begin to break up they have no natural buoyancy
as wood does. That can serve to localize the distribution of structural material. Once on the
bottom, electrolytic reduction, inorganic processes, and transferred surface motion contribute to
the destruction of exposed structure. Like wooden vessel remains, iron and steel shipwreck
materials can be better preserved by burial. Although cold water appears to contribute more to
metal hull preservation than warm water, the combination of cold, depth, and zero sunlight
appears to create ideal conditions for inorganic deterioration.
Vessel structure and the elements are not the only factors that impact preservation on the
OCS. Human activity can be equally destructive. Perhaps the most destructive is trawling and
dredging. Serious damage to shipwreck sites has been attributed to trawling. The effects of nets
fouling an exposed wreck can be highly destructive. Dredging for shellfish such as scallops can
be equally destructive. Although today surveys are required in advance of construction activity,
burying cables and pipelines has been determined to impact shipwreck remains. As dredging
activity associated with seafloor minerals and aggregate increases, the potential for shipwreck
disturbance is going to increase. Finally, diving and even submersible access to the OCS has
placed shipwrecks within the grasp of recreational wreck divers and commercial salvors willing
to recover both shipwreck remains and the artifacts associated with those vessels.
In spite of the numerous factors that can compromise the integrity and long-term preservation
prospects of shipwrecks, thousands of wrecks have been documented and remain on the OCS. In
few cases have actual surveys been conducted to assess the condition of the remains.
12.6.2 Shipwrecks by Vessel Type and Period
A total of 3,659 vessels from the ASD were identified by vessel type (this does not include
airplanes, obstructions, or other non-vessel sites). Of the 58 vessel types recorded, 15 types
accounted for 86 percent of the total. Schooners, steamers, freighters, barges, and brigs were the
five most common types. The vessel types were grouped by the date of the wreck in 25-year
increments to illustrate the changes in vessel types over the years. The distribution of wrecks by
time period is shown in Table 12.1. Of the 3,244 wrecks for which a date of loss is given, about
half (49.4 percent) wrecked in the 20th century. Better reporting of shipwrecks, the large number
of vessels, and the loss of large numbers of freighters, steamers, tankers, and barges to German
submarines in World War I and World War II, seem to account for this.
Some vessel types occur over long periods of time, such as the schooner, sloop, and brig,
which were adapted to different uses while retaining the same name. Steamers first appear in the
wreck record in the second quarter of the 19th century and continue into the third quarter of the
20th century. Although most vessels built after 1910 were equipped with diesel engines,
steamships still played an important role in maritime trade in the 20th century. The peak of losses
for steamships occurred in the period from 1850–1874 when 115 wrecks were reported, many of
them Civil War losses. The 80 losses in the first quarter of the twentieth century are largely from
U-boat attacks.
215
1925-1949
1950-1974
1975-1999
2000-2010
unknown
Total
1
90
3
2
5
7
37
4
30
3
1
32
13
194
17
71
7
5
148
45
1
5
3
1
5
16
49
1
4
31
6
117
224
24
10
45
35
3
1800-1824
3
1
58
3
5
7
5
1
1
9
34
36
5
25
1775-1799
1900-1924
9
8
1875-1899
4
1750-1774
1725-1749
2
1850-1874
1
1825-1849
barge
barkentine
barque
battleship
boat
brig
brigantine
buoy tender
cabin cruiser
collier
corvette
cruiser
cutter
destroyer
dive tender
dredge
escort
ferry
fishing vessel
freighter/cargo
frigate
galleon
gunboat
landing craft
landing ship
landing ship,
tanks
Liberty ship
lighhter
lightship
merchant ship
mine sweeper
monitor/ironclad
motor vessel
man-o-war
paddlewheel boat
passenger
patrol
pilot boat
pleasure craft
sailboat
schooner
schooner barge
scow
ship
ship of the line
shrimp trawler
sidewheel
steamer
1700-1724
<1700
Table 12.1. Vessel Types in the ASD by Historical Periods.
2
20
5
5
1
2
2
4
1
1
1
2
1
4
4
2
2
4
6
31
2
27
1
41
1
10
146
1
15
14
1
3
1
6
1
1
1
2
8
1
7
2
4
1
1
3
7
1
1
3
4
2
2
1
31
6
4
1
1
1
1
5
3
1
1
2
4
37
8
1
51
13
3
3
34
1
1
10
2
14
1
1
4
11
1
1
2
8
1
13
32
10
3
12
1
5
22
7
19
5
8
7
12
17
2
1
6
81
21
3
21
1,231
103
10
97
6
6
2
9
4
1
2
3
12
1
1
2
10
1
1
3
4
43
1
1
31
399
16
1
1
2
329
75
3
3
4
1
1
1
2
7
2
7
1
2
10
16
81
2
102
5
5
6
1
28
8
222
20
4
4
1
2
1
35
2
1
1
2
2
1
7
216
1800-1824
1825-1849
1850-1874
1875-1899
1900-1924
1925-1949
1950-1974
8
19
30
17
4
11
115
3
13
47
15
80
21
2
30
44
2
100
6
1
33
15
2
1
17
1
6
1
2
6
3
1
15
16
15
45
70
3
24
13
4
12
3
8
17
520
570
4
768
1
2
218
171
6
1
2
6
15
11
574
157
1
103
1
Total
1775-1799
4
unknown
1750-1774
7
2000-2010
1725-1749
1
1975-1999
1700-1724
sloop
steamer
submarine
supply vessel
tanker
torpedo boat
transport
trawler
tug or tow boat
whaler
yacht
Totals
<1700
Table 12.1. Vessel Types in the ASD by Historical Periods, continued.
2
11
4
131
288
95
4
120
20
15
66
118
24
17
3,659
7
6
13
52
2
1
415
Other peak periods of loss for vessels also seem to have been influenced by war losses.
Losses of brigs, sloops, ship-rigged vessels, and gunboats all peaked during the first quarter of
the 19th century, likely as a result of the War of 1812 and ongoing maritime conflicts with
European nations. The greatest number of losses for freighters and cargo vessels was during the
25-year period that included World War II, when German submarines stalked U.S. merchant
ships along the coast.
Because of their specialized nature, the wrecks of many vessel types were confined almost
entirely to the 20th century, including the tanker, fishing vessel, submarine, trawler, tug boat, and
dredge. The sailboat, yacht, pleasure craft, and motor vessel were primarily privately-owned
vessels, and were generally limited to the 20th century. At the other end of the timeline, relatively
few early vessels (prior to 1700) are referred to by ship type. Only a few brigs, sloops, ships, and
galleons are found in the ASD for that period.
12.6.3 Analysis of Shipwreck Locations
The shipwreck inventory for the Gulf of Mexico (Pearson et al. 2003) included a model of
spatial distribution for shipwreck locations in the GOMR based on data collected for that report.
The report points out the many difficulties of working with shipwreck data, including inaccurate
and incomplete reporting, under-reporting of smaller vessels lost, historical variations in ship and
place names, and confusion regarding similar names for ships and geographic locations (Pearson
et al. 2003:4-2–4-6). Previous shipwreck location models by Pierson et al. (1987), Science
Applications, Inc. (1981), and Garrison et al. (1989) are cited that produced mixed results.
Pierson et al. and SAI (which examined the southern Atlantic Seaboard for an earlier MMS
study), found shipping routes, port locations, and natural hazards to represent causal factors in
the location of shipwrecks. Garrison et al.’s study of shipwrecks in the Gulf of Mexico
conducted factor analysis on temporal and areal distribution. They found an association between
shipwreck location and the development of port locations over time, as well as between
shipwreck locations and the location of shipping routes, ports, and hazards (Pearson et al.
(2003:4-42–4-43).
217
In general, however, these associations were weak as a result of inaccurate location
information for a large number of reported wrecks. For this reason, Pearson et al. did not
emphasize such factor analysis. In their study of shipwreck distribution in the Gulf, they found
broad patterns of wreck distribution, with the vast majority of wrecks located near shore inside
the 60-meter contour. There is also evidence of increased occurrence in the vicinity of ports, in
areas of heavy traffic, such as the Straits of Florida, and near hazards, such as the reefs around
the Dry Tortugas. They also found a greater concentration of wrecks in the eastern part of the
Gulf, most likely because vessel traffic has been greater in that area for a longer period.
The results of the current investigation are consistent with these findings and suggest that
certain geographic variables are the best predictors of shipwreck location. These include
proximity to shore, proximity to ports or confined navigational routes, and the presence of
navigational hazards. Because different methods were used in compiling existing databases and
the current ASD, shipwreck locations from these inventories were plotted separately as well as in
combination to see if patterns of shipwreck location were consistent across different lists. Figures
12.10–12.14 illustrate the distribution of shipwrecks in the OCS study area from various sources.
Only those entries that provided coordinate locational data were included.
Figure 12.10 shows the location of wrecks from NOAA’s AWOIS database. This list is based
on actual features identified from survey and does not include wrecks reported but not located.
However, it does include some features identified from survey that may not be shipwrecks but
some other type of feature on the ocean floor. It is evident from the distribution on this map that
the vast majority of shipwrecks are located within 50 miles of shore, with the greatest localized
concentration off the coast of New Jersey, Delaware and Maryland. Significant clusters of
wrecks are also located around Cape Cod, the entrance to Chesapeake Bay, the North Carolina
capes (Hatteras, Lookout, and Fear), and the entrances to the Savannah and St. John’s rivers.
Farther from shore, wrecks are more widely distributed, with noticeable concentrations around
the Georges Bank fishing grounds and off of Cape Hatteras. The Global Wrecks database, a
commercial shipwreck inventory compiled by General Dynamics, has more entries but shows a
nearly identical distribution to the AWOIS list (see Figure 12.11).
Figure 12.12 shows the distribution of shipwrecks from primary and secondary sources
consulted during the current research effort (excluding AWOIS). Because a large number of
these wrecks come from regionally specific dive guide books, the distribution is heavily skewed
toward wrecks in the Mid-Atlantic region, from Long Island, New York to Chesapeake Bay,
Virginia. There are also a large number of wrecks reported for Cape Hatteras and Cape Lookout.
Wrecks are apparently under-reported for New England (from which no specific dive book was
included). Offshore wrecks appear to be fairly randomly scattered between New York and North
Carolina.
The map of shipwrecks from the existing BOEMRE shipwreck database shows clustering
around the coast of Massachusetts, around Cape Hatteras, and at the mouth of the St. Johns River
in Florida. (Figure 12.13). This is likely due to the extensive surveys conducted in these areas,
where shipwreck locations have been assigned locational data. The small sample size results in
less evidence of clustering around port locations. Even with this small sample, however, smaller
218
Figure 12.10. Location of shipwrecks with coordinate data in AWOIS database.
219
Figure 12.11. Location of shipwrecks with coordinate data in the Global Wrecks database.
220
Figure 12.12. Location of shipwrecks with coordinate data from primary and secondary sources.
221
Figure 12.13. Location of shipwrecks with coordinate data from the existing BOEMRE database.
222
clusters of sites are evident south of New York City, off Cape Lookout, Cape Fear, and Cape
Canaveral, and along the Florida Keys
Figure 12.14 shows all of the wrecks with location information in the ASD, which includes
governmental databases and primary and secondary sources. Combining the inventories
mitigates some of the biases in the individual lists and reinforces the broad patterns evident on
the other maps. Although these sources have significant overlap, the repeated sites should not
significantly affect the distribution since at the scale of the map, many of the duplicate locations
are not differentiated. The duplicate entries may skew the distribution closer to shore, however,
due to the increased likelihood of coordinate data being available for these wrecks.
The distribution maps presented above demonstrate that patterns of wreck location are
similar using a number of different data sources. Vessels travelled to and from myrid locations
along the Atlantic Seaboard during the historic period, and even though ships followed certain
routes, they often sacrificed safety to take a shorter route, made navigational errors, or were
blown off course. As a result, shipwrecks can be found in almost any part of the project area.
Nevertheless, the distribution of shipwrecks on the Atlantic Seaboard is clearly not random.
Plotting shipwrecks with known locations reveals that shipwrecks are concentrated in certain
geographic areas that can be used to establish zones of low, medium, and high probability for
encountering recorded or unknown wrecks as part of cultural resources investigations. Using the
observed distribution of shipwrecks to model the likely distribution of additional wrecks may be
considered somewhat reflexive, but is consistent with the evidence of traffic volume and
navigation hazards as factors in vessel loss. Shipwreck density was measured within 2,304hectare BOEMRE lease blocks in the OCS from the AWOIS database, the existing MMS
Atlantic Shipwreck database, and primary and secondary data collected during the current effort;
the results are shown on the map in Figure 12.15. Density is defined in three categories: no
wrecks, 1–4 wrecks (low frequency), and 5 or more wrecks (high frequency). Beyond
approximately 100 miles from shore, lease blocks with one or more wrecks are sparse and widely
scattered, with the exception of the Georges Bank fishing grounds, located in an area 100–200
miles off the coast of Massachusetts. Only three high frequency lease blocks are farther than 100
miles from shore. One of these is in the Georges Bank area. The other two are off the coast of
Cape Hatteras and are apparent outliers. Within the approximate 100-mile boundary, the density
of low frequency lease blocks increases noticeably, and high frequency lease blocks cluster along
certain portions of the coastline, especially near ports, estuary entrances, and navigational
hazards. Between the Chesapeake Bay and New York City, the high frequency blocks run along
the length of the coast and extend farther from shore along high traffic routes.
The distribution map in Figure 12.15 was used to draw low, medium, and high probability
zones based on patterns of shipwreck density evident in the distribution (Figure 12.16). Low
probability zones are defined by scattered low frequency blocks, with large areas between with
no wrecks. Medium probability areas are characterized by a more uniform distribution of low
frequency blocks, with occasional occurrences of high frequency blocks. High probability zones
contain dense areas of low frequency blocks, with significant clusters of high frequency blocks.
Further refinement of these zones is possible using the existing data, but these broad categories
of shipwreck density provide a predictive model for the existence of both known and potential
223
Figure 12.14. Location of shipwrecks with coordinate data from all sources in the ASD.
224
Figure 12.15. Shipwreck density map for BOEMRE lease blocks in the Atlantic OCS study area.
225
Figure 12.16. High, medium, and low probability zones for shipwrecks in the Atlantic OCS
based on shipwreck density.
226
shipwrecks in Atlantic OCS lease blocks based on geographic factors that are proven to influence
shipwreck occurrences. The implications of the predictive model for cultural resources survey
strategies are discussed further in the next section.
Wreck locations were also compiled for each state based on the reported “Nearest State”
entry to explore regional distribution of shipwrecks (Table 12.2). Of 10,434 wrecks in the ASD
for which the nearest state was reported, the greatest number of wrecks was found off the coast
of Virginia (2,306). New Jersey had the second largest number of wrecks with 1,806. North
Carolina was third with 1,801. Florida and Maryland both have over 1,000 wrecks.
Table 12.2. Distribution of Shipwrecks in the ASD by State and Region.
Number
of
Wrecks
Miles of
Shoreline
in State
Sites Per
Linear Mile
Region
Number of
Wrecks
Miles of
Shoreline in
Region
Sites Per
Linear Mile
ME
133
240
0.55
NH
13
14
0.93
MA
650
230
2.83
Northeast
954
539
1.77
RI
144
40
3.60
CT
14
15
0.93
NY
373
120
3.11
NJ
1,911
130
14.70
DE
363
25
14.70
Middle
6,127
308
19.89
MD
1,139
33
34.52
Atlantic
VA
NC
2,341
112
20.90
1,954
320
6.11
SC
576
185
3.11
4,217
749
5.63
GA
193
97
1.99
FL
1,494
635
2.35
Totals
11,298
2,196
5.14
Nearest
State
Southeast
Because of the great variation in the ocean frontage of each state, the number of shipwreck
sites per mile was calculated based on an estimate of the length of the federal waters boundary
for each state (see Table 12.2). By far the greatest concentration of sites per mile is found in the
Mid Atlantic states. Maryland has the highest ratio of shipwrecks with over 34 per mile of
coastline. Despite a relatively long coast of 112 miles, Virginia’s 2,306 shipwrecks place it
second with 20.90 sites per mile. Delaware and New Jersey also have a very high ratio of
shipwrecks per mile. It was anticipated that the New England states would have a high ratio of
shipwrecks per mile due to the high volume of ship traffic around Boston and the presence of
navigational hazards around Cape Cod. However, the low volume of traffic along Maine’s 240mile coast brings down that region’s average. The Southeast states have a surprising high density
of sites with 5.63 per mile.
It appears that the distribution of wrecks within the Atlantic OCS project area is closely
correlated to vessel traffic, especially in the vicinity of port approaches and navigational hazards.
227
Consistent data on vessel traffic from ports on the Atlantic seaboard over long periods of time is
difficult to find and widely scattered in the sources, and most economic statistics of the Atlantic
trade are concerned with the value of imports and exports rather than numbers of vessels or even
tonnage (Evans 1976; Huebner 1922). Historical studies have also favored the Colonial period
over the 19th and 20th centuries (Crothers 2004; French 1987; Matson 1998; Morgan 1989; Smith
2003). However, a reasonable understanding of vessel traffic in the Atlantic OCS is possible
from the literature.
The British customs houses prior to the Revolution collected data on the total tonnage of
vessels arriving and departing from each of the colonies in 1769, broken down by origin and
destination (Johnson 1922:92). These figures provide a broad picture of the most popular trade
routes. For the New England colonies, New York, and North Carolina, the greatest tonnage of
vessels arrived from other colonies and from the Bahamas. In New Hampshire, Pennsylvania,
and Georgia, the largest tonnage of vessels arrived from the West Indies. For Maryland, Virginia,
and South Carolina, the greatest amount of tonnage of vessels was from England and Ireland.
The tonnage of exports showed a similar pattern: New England exported primarily to the rest of
the colonies, the Bahamas, and the West Indies, while the Mid-Atlantic and Southern states
exported predominantly to Great Britain. New Jersey, Pennsylvania, and Georgia also exported
mostly to the West Indies. These figures indicated that there was significant traffic along the
Atlantic Seaboard in both directions, with the greater portion of the traffic in Mid-Atlantic and
Southern ports arriving from and traveling to England and Europe or the Caribbean, while in
New England, there was more traffic coming from England and Europe, but more departing for
the southern colonies or the Caribbean. Fishing and whaling vessels sailing from New England
ports and Long Island would have traveled primarily on east-west and northeast-southwest routes
to access Georges Bank, the Grand Banks of Newfoundland, and the North Atlantic. In general,
the majority of all ship traffic was southwest-northeast along the Atlantic Seaboard following the
Gulf Stream northward and the coastal currents southward. The major colonial trade routes are
illustrated in Figure 12.17.
These major trade routes did not change significantly in the 19th and 20th centuries. Rather,
the coastwise trade increased, while foreign trade expanded to include Central and South
America, the west coast of the United States, and Asia. The Northeast continued to dominate the
import business, while Southern commodities, like cotton and timber products, remained the
primary export. Midwest grain also became a staple export via New Orleans. Manufactured
goods from Europe and the Northeast were shipped to the South and Midwest from New
England, New York, and Philadelphia.
The effect on vessel traffic of these expansions was primarily to increase the number of
vessels entering and exiting U.S. ports and traveling along the Atlantic Seaboard. Ship traffic
was greater over a longer period of time in the Northeast, thus the number of shipwrecks was
greater in that region. This is consistent with Pearson et al.’s findings of a greater concentration
of wrecks in the eastern part of the Gulf of Mexico due to the greater amount of traffic there over
time. The number of vessels entering ports was greatest in the northern ports during the 17th and
18th centuries, but southern ports became increasingly important during the 19th and 20th
centuries following the spread of cotton culture across the region. The rise of New Orleans as a
shipping point for the Midwest catapulted Louisiana ahead of all states but New York in exports
228
Figure 12.17. Major trade routes of colonial North America, 1769 (from Johnson 1922).
229
by 1818. Despite the rise of southern ports, Boston was the nation’s busiest port until it was
passed by New York in 1840, 15 years after the completion of the Erie Canal. In the late 19th
century, Philadelphia became the busiest port on the Atlantic Seaboard, thanks to its booming
shipbuilding industry (Huebner 1922; Johnson 1922; New York State Canal Corporation n.d.;
Shepherd and Walton 1972).
The effect of vessel traffic being funneled into a harbor entrances can be seen in the
concentration of wrecks around the approaches to New York, Delaware and Chesapeake bays,
Charleston Harbor, and the Savannah River (see Figures 12.10–12.14). The number of wrecks
recorded increases as one approaches the entrance, just as the number of vessels within a given
area increased as they funneled into the port or bay. Since the Chesapeake and its estuaries
covered such a broad area, the traffic into that bay was considerable. Many vessels were lost as
they tried to make their way between the Virginia Capes: Cape Charles on the north and Cape
Henry on the south. Charleston and Savannah served as the entry point for nearly all goods
coming into their colonies and were the largest ports south of Norfolk, meaning that hundreds of
vessels a year approached Charleston Harbor and the Savannah River from the ocean.
All other factors being equal, the greater number of vessels that traveled through a particular
area, the more wrecks are likely to occur. The increased likelihood of collision and the increased
hazards in shallow water contribute to this risk. The frequency of shipwrecks around the
entrances to ports and harbors is also affected by the increased presence of hazards as a ship
approaches shore. Shoals are more common in shallow waters, and bars extend from capes and
form at sea. Waves become steeper and less predictable closer to shore and in shallow areas
farther from shore. Distressed ships often make their way toward harbors, as well, with many not
making it safely before sinking or stranding. The many wrecks around Nantucket and Cape Cod,
where rocky shoals and ledges are common, attest to the twin factors of heavy traffic and
dangerous hazards in contributing to shipwrecks. Although shipwrecks in shallow water are
more likely to be salvaged, the large number of recorded wrecks around the entrances to bays
and harbors shows the effect of increased traffic within a defined area. The coastwise trade
created traffic along the entire Atlantic seaboard, but the number of vessels was greatest between
the busiest ports, resulting in the highest density between Boston and Philadelphia. As shipping
volume migrated southward in the nineteenth century, the Mid-Atlantic coast became the most
frequently traveled route. As a result, the coast of New Jersey, Delaware, Maryland, and Virginia
appear to have the greatest concentration of shipwrecks in the OCS project area.
In addition to vessels arriving from other ports, there was a significant volume of traffic in New
England ports from fishing vessels leaving from and returning to their home ports. The New
England cod fleet alone numbered over 600 vessels carrying 4,400 crew members during the
decade from 1765 to 1775. The vast majority sailed from Massachusetts ports for Georges Bank,
about 150 miles due east of Cape Cod, or for the Newfoundland Banks farther to the northeast.
Marblehead, Gloucester, Salem, and Plymouth, were the leading fishing communities. Whaling
ships added further to the traffic out of Massachusetts, with smaller fleets in the various ports of
New York, Connecticut and Rhode Island (Johnson 1922:154–161). The large number of wrecks
shown between 100 and 200 miles off of Cape Cod in Figure 13.14 illustrate the large number of
vessels plying those waters during the historical period. Commerial fishing was not as significant
230
in the southern colonies, although the Chesapeake developed a modest trade in shad and herring
during the 18th century, along with crabs and oysters. These hauls came mostly from the bay,
however, so ocean-going fishing vessels were less common than in New England (Casey n.d.).
Two noteworthy areas have experienced a large number of shipwrecks due principally to
navigation hazards: Cape Hatteras and the Straits of Florida (between the Florida Keys and
Cuba). Although the areas do not have busy port approaches, both saw considerable traffic
because of their location along major shipping routes. Cape Hatteras projects far into the Atlantic
and the turbulence created by the currents around the cape, as well as the shifting sand bars that
form off its point, have put thousands of ships in peril, despite the construction of lighthouses
and life-saving stations in the late 19th century (Figure 12.18). Vessels traveling along the coast
must pass the cape. This includes not only those going between ports on the northern and
southern seaboard, but also many coming from the Caribbean, the Gulf, or Central and South
America, which utilize the strong Gulf Stream current to return to Europe from the Americas.
Storms originating in the Caribbean frequently rake the Outer Banks, driving vessels north and
west into the shore. Captains concerned with making the shortest passage sometimes tried to cut
the distance between ports by sailing closer to the cape than advised, often resulting in disaster.
The large number of wrecks around Cape Hatteras has earned it the nickname, “The Graveyard
of the Atlantic.”
Figure 12.18. The steamer Metropolis was lost off Currituck Beach in 1878, 3 years after the
lighthouse was put in operation there (Library of Congress).
231
The 90-mile wide passage between the Florida Keys and Cuba has also proven hazardous.
Spanish fleets passed through the straits regularly with cargos of treasure from its possessions
during the period of colonial domination of Central and South America. The numerous reefs and
shoals coupled with the frequency of hurricanes during the late summer and early fall when the
flotillas typically sailed led to numerous disasters over a 200-year period in the 16th and 17th
centuries. The route continued to be used from 18th century on by vessels carrying goods from
New Orleans and the Gulf Coast to the east coast of the United States and Europe.
12.7 RECOMMENDATIONS FOR CULTURAL RESOURCES MANAGEMENT OF SHIPWRECKS IN THE
ATLANTIC OCS
12.7.1 Implications of ASD for Survey Approaches
The purpose of compiling the Atlantic Shipwreck Database was to gather existing data,
update previous inventories, and identify potential sources of new information that can be used
to inform cultural resources management decisions regarding off-shore leases in the Atlantic
OCS. In addition to providing information on specific shipwrecks that might be located in a
defined project area, the ASD is useful for planning purposes, providing a guide for site selection
and budget considerations based on the likelihood of encountering known or unknown
shipwrecks in a given area. The ASD can also be used to design survey methodology that is most
appropriate for a particular project.
The probability zones defined in the previous section and shown in Figure 12.16 can be used
to define survey strategies in low, medium and high probability zones that can be incorporated
into cultural resources management planning, providing a guide for site selection, budget
considerations, and survey strategy. Lease applicants should factor whether potential cultural
resource impacts could be reduced by changing site location and, if not, what costs might be
expected for cultural resource mitigation within high probability or medium probability zones.
Since significant unknown shipwrecks could be found anywhere within the Atlantic OCS,
cultural resources survey cannot be ruled out in low probability zones; however, the survey
approach could be scaled to the likelihood of findings. For example, lease areas in high
probability zones should be surveyed well in advance to identify all areas of concern and to
develop mitigation strategies, while low probability areas could incorporate cultural resource
surveys into geophysical testing, so that in the event of encountering cultural materials, a strategy
could be developed for mitigation if necessary.
BOEMRE survey requirements under the 2005 Notice to Lessees and Operatorsfor the Gulf
of Mexico (NTL 2005-G07) call for magnetometer survey transects of 50 meters or less in areas
of high shipwreck probability, an interval that has proven capable of identifying 19th century
wooden sailing vessels under controlled conditions. The same survey interval is appropriate in
the Atlantic OCS, where broadly similar expectations pertain to shipwreck frequency and
potential to find sites. Pearson et al. (2003) recommend a 30-meter interval for high probability
areas in order to insure identification of smaller and older vessels.
Areas of medium probability could warrant additional survey coverage beyond that afforded
by concurrent monitoring of the applicant’s geophysical studies. The multibeam bathymetry and
backscatter intensity data, side scan sonar, and magnetometer surveys conducted as part of the
232
applicant’s project planning could be supplemented with targeted sampling of areas off grid that
have slightly higher potential for shipwrecks due to relative proximity to a harbor entrance or
shoal, for example. Alternatively, if the lease block in question occurs in an area where potential
for submerged prehistoric sites is prompting additional survey coverage, such investigations
would also serve to identify shipwrecks. Guidance for survey coverage of medium probability
areas should remain flexible such that project specific factors can be considered in crafting a
survey strategy.
12.7.2 Suggestions for Supplemental Historic Research and Database Management
When potential or identified shipwrecks are encountered in a lease block, additional
historical research will need to be carried out. Research in primary and secondary sources of
information on shipwrecks revealed the high degree of inaccuracy in the data, and underscores
the importance of using as many sources as possible in conducting cultural resources
investigations. Shipwreck locations from actual survey, such as AWOIS and cultural resource
management reports are useful for identifying wrecks and other anomalies on the ocean floor,
but provide little information on the name or history of these features. Both primary and
secondary sources generally focus on the features of the vessel and other historical information,
but usually have limited locational data. Dive books and websites often include both historical
and locational information, but cover a limited number of wrecks, with an emphasis on those
closer to shore, in shallow waters, and with good exposure. Depending on the nature of
archaeological investigations and findings, contextual information on vessel type, construction,
materials, cargo, and other attributes could assist the researcher in identifying the wreck using
available historical information. The location of the wreck and the type of ship uncovered will
determine which sources will be most valuable in piecing together the story of the ill-fated vessel
as well as its historical context. Historians, underwater archaeologists, and other consultants
conducting investigations for BOEMRE permitted projects will be in the best position to
ascertain the sources of potential relevance to their project.
Beyond the specific archival research conducted in the context of particular projects,
BOEMRE may contemplate sponsoring additional historical research to augment the contents of
the ASD. To that end, the following observations may prove valuable in allocating resources to
such an effort. The relative value of the primary sources consulted in compiling the ASD is
discussed in the methods section (Chapter 11), and for the most part, these sources have been
sufficiently mined as part of this effort and previous research. A few sources could be covered
more thoroughly, including a more systematic inventory of shipwrecks listed on internet dive
guides and shipwreck related sites such as The Wreck Site (www.wrecksite.eu) and The U-boat
Project (www.uboat.net). Also, the index of U.S. Coast Guard Reports of Casualty, 1913–1939,
on file at the National Archives in Washington, was only searched through 1918. Although many
of the wrecks reported in the index appear in other databases, the wreck cards provide a wealth
of information.
Less information is available for wrecks prior to the 1850s, before government data began to
be collected. Further research for the period up until the mid 19th century will likely continue to
turn up wrecks previously unreported in existing shipwreck inventories, but a systematic search
of sources from this period for the purpose of identifying new wrecks would require considerable
233
time and effort for the number of wrecks that would be added. In addition, these sources rarely
provide more than the vaguest location information. Of more use would be repositories dedicated
to seafaring such as the Mariner’s Museum in Newport News, Virginia, the Museum of Ships
and the Sea in Mystic Seaport, Connecticut, and the Independence Seaport Museum in
Philadelphia. These facilities contain libraries specializing in records related to ships and
seafaring, with staffs who can guide research. On-going research in these records yields an ever
expanding knowledge base that is shared among researchers and staff. During the current
research effort, a total of 5 days was spent at the Mariner’s Museum Library, but considerable
resources remain. The Independence Seaport Museum library was unavailable for research
because of a vacancy in the staff, and time constraints precluded a visit to the Museum of Ships
and the Sea. Further research at these institutions is recommended.
The ASD represents a powerful tool for cultural resources management, pulling information
from a wide variety of sources on shipwrecks together in a compatible GIS database. The format
preserves source data, allowing researchers access to as much information as possible from
which to draw conclusions. This will aid both in locating vessels that have not yet been found
and identifying those that are discovered. Ideally, as the database is put to use, entries that appear
to be duplicates can be cross-referenced, while still maintaining the original source data. This is
preferable to merging sources and risking perpetuating an erroneous conclusion. The database is
infinitely expandable as new data is collected, and further analysis of the data will likely yield
valuable information on vessel types, wreck locations, site preservation, and more.
234
PAGE LEFT INTENTIONALLY BLANK
235
13. CONCLUSION
The goal of this study was two-fold. First, through the contributions of various scholars doing
research on underwater sites, it has revised a model for identifying areas of the Atlantic OCS
where there is potential for preserved prehistoric sites. In doing so, it has addressed the following
questions:
1. When were people living along the Eastern Seaboard?
2. What portions of the now-submerged Atlantic OCS were subaerial when people likely
occupied the region?
3. What types of landforms were likely locations for settlement, such that sites would have
been formed and left behind?
4. What is the likelihood that such sites would have survived marine transgression?
5. What methods should researchers employ to identify areas that may contain sites, and to
look for sites once such areas have been identified?
The second component of this study was to conduct research and assemble a database of
known and likely historic shipwrecks along the Atlantic OCS, and provide a written historic
context for these wrecks. The database, which is provided under separate cover, contains over
33,000 entries and incorporates extant databases, literature on shipwrecks, state and federal data
on shipwrecks, and archival research at a number of institutions along the east coast.
236
PAGE LEFT INTENTIONALLY BLANK
237
REFERENCES CITED
Adams, P.N., N.D. Opdyke, and J.M. Jaeger. 2010. Isostatic uplift driven by karstification and
sea-level oscillation: Modeling landscape evolution in North Florida. Geology 38(6):531–
534.
Adesida, A.O. 2000. The stratigraphic framework of Eocene through Miocene sequences beneath
Sapelo Island, Georgia: A shallow seismic reflection study. M.S. Thesis, University of
Georgia, Athens, GA.
Adovasio, J.M. 1993. The ones that will not go away: A broad view of pre-Clovis populations in
the New World. In: Soffer, O. and N.D. Praslov, eds. From Kostenski to Clovis: Upper
Paleolithic Paleoindian Adaptations. New York: Plenum Press. Pp. 199–218.
Adovasio, J.M., J. Donahue, and R. Stuckenrath. 1990. The Meadowcroft Rockshelter
radiocarbon chronology 1975–1990. American Antiquity 55:348–354.
Adovasio, J. M., D. Pedler, J. Donahue, and R. Stuckenrath. 1998. Two decades of debate on
Meadowcroft Rockshelter. North American Archaeologist 19:317–41.
Adovasio, J. M., D. Pedler, J. Donahue, and R. Stuckenrath. 1999. No vestige of a beginning nor
prospect for an end: Two decades of debate on Meadowcroft Rockshelter. In: Bonnichsen,
R. and K. L. Turmire, eds. Ice age peoples of North America: Environments, origins, and
adaptations of the First Americans. Corvallis: Oregon State University Press. Pp. 416–431.
Advisory Council on Historic Preservation. 2010. Comments of the Advisory Council on
Historic Preservation on the proposed authorization by the Minerals Management Service
for Cape Wind Associates, LLC to construct the Cape Wind Energy project on Horseshoe
Shoal in Nantucket Sound, Massachusetts, April 2, 2010. Advisory Council on Historic
Preservation, Washington, D.C.
Alexander, C.R. and V.J. Henry. 2007. Wassaw and Tybee Islands—Comparing undeveloped
and developed barrier islands. In: Rich, F. and C.R. Alexander, eds. Field Trip Guidebook
for the 56th Annual Meeting of the Southeastern Section. Geological Society of America.
Pp. 187–198
Alexander, C.R. and M. Robinson. 2006. Identification and review of existing seabed
information for potential windfarm site locations. Prepared for: Strategic Energy Initiative,
Georgia Institute of Technology, Atlanta, Georgia.
Almquist-Jacobson, H., A.C. Dieffenbacher-Krall, R. Flanagan-Brown, and D. Sanger. 2001.
The Holocene record of lake levels of Mansell Pond, central Maine, USA. The Holocene
11(2):189–201.
Almquist-Jacobson, H. and D. Sanger. 1995. Holocene climate and vegetation in the Milford
drainage basin, Maine, USA, and their implications for human history. Vegetation History
and Archaeobotany 4(4):211–222.
238
Anderson, D.G. 1985. Middle Woodland societies in the lower South Atlantic Slope: A view
from Georgia and South Carolina. Early Georgia 13:1–2:29–66.
Anderson, D.G. 1990a. The Paleoindian colonization of Eastern North America: A view from the
southeastern United States. In: Issac, B. and Kenneth Tankersley, eds. Early PaleoIndian
economies of eastern North America. Journal of Economic Anthropology Supplement 5. Pp.
163–216.
Anderson, D.G. 1990b. Political evolution in chiefdom societies: Cycling in the late prehistoric
southeastern United States. Ph.D. dissertation, University of Michigan.
Anderson, D.G. 1995. Paleoindian interaction networks in the eastern woodlands. In: Nassaney,
M.S. and K.E. Sassaman, eds. Native American interactions: Multiscalar analyses and
interpretations in the eastern woodlands. Knoxville: University of Tennessee Press. Pp. 1–
26.
Anderson, D.G. 2001. Climate and culture change in prehistoric and early historic eastern North
America. Archaeology of Eastern North America 29:143–186.
Anderson, D.G., C.E. Cantley, and A.L. Novick, comps. 1982. The Mattassee Lake sites:
Archeological investigations along the lower Santee River in the Coastal Plain of South
Carolina. Special Bulletin 1. Interagency Archeological Services Division, National Park
Service, Atlanta.
Anderson, D. G. and M.K. Faught. 1998. The distribution of fluted Paleoindian projectile points:
Update 1998. Archaeology of Eastern North America 26:163–188.
Anderson, D.G. and C. Gillam. 2000. Paleoindian colonization of the Americas: Implications
from an examination of phsiography, demography, and artifact distribution. American
Antiquity 65(1):43–66.
Anderson, D.G. and G.T. Hanson. 1988. Early Archaic settlement in the southeastern United
States: A case study from the Savannah River Valley. American Antiquity 53:262–286.
Anderson, D.G., R.J. Ledbetter, and L.D. O’Steen. 1990. Paleoindian period archaeology of
Georgia. Georgia Archaeological Research Design Paper No. 6. University of Georgia,
Athens.
Anderson, D.G. and K. Sassaman. 1996. Modeling Paleoindian and Early Archaic settlement in
the Southeast: A historical perspective. In: D.G. Anderson and K.E. Sassaman, eds. The
Paleoindian and Early Archaic Southeast. Tuscaloosa: University of Alabama Press. Pp.16–
28.
Anderson, D.G., K.E. Sassaman, C. Judge, eds. 1992. Paleoindian and Early Archaic period
research in the lower Southeast: A South Carolina perspective. Columbia: Council of South
Carolina Professional Archaeologists in conjunction with Savannah River Archaeological
239
Research Program, South Carolina Institute of Archaeology and Anthropology, University
of South Carolina.
Anderson, D.G. and J. Schuldenrein, eds. 1985. Prehistoric human ecology along the upper
Savannah River: Excavations at the Rucker’s Bottom, Abbeville and Bullard site groups.
Atlanta: Interagency Archeological Services, National Park Service.
Aqua Explorers, Inc. 2009. Aqua Explorers, Inc. and Capt. Dan Berg’s Wreck Valley Shipwreck
Charter Boat. Internet website: http://www.aquaexplorers.com.
Ashley, K.H. 2008. Refining the ceramic chronology of northeastern Florida. The Florida
Anthropologist 61(3-4):123–131.
Association
of
Underwater
Explorers.
http://uwex.us/wreckinfo.html.
2009.
Shipwrecks.
Internet
website:
Austin, D., B. Carriker, T. McGuire, J. Pratt, T. Priest, and A. G. Pulsipher. 2004. History of the
offshore oil and gas industry in southern Louisiana: Interim report; Volume I: Papers on the
evolving offshore industry. U.S. Dept. of the Interior, Minerals Management Service, Gulf
of Mexico OCS Region, New Orleans, LA. OCS Study MMS 2004-049. 98 pp.
Austin, J.A, Jr., E. Uchupi, D.R. Shaughnessy, and R.D. Ballard. 1980. Geology of New England
passive margin. The American Association of Petroleum Geologists Bulletin 64:501–526.
Bailey, G.N. and N.C. Flemming. 2008. Archaeology of the Continental Shelf: Marine resources,
submerged landscapes, and underwater archaeology. Quaternary Science Reviews 27:2153–
2165.
Bailey, G. and J. Parkington. 1988. The archaeology of prehistoric coastlines. Cambridge:
Cambridge University Press.
Balco, G., J.O.H. Stone, S.C. Porter, and M.W. Caffee. 2002. Cosmogenic-nuclide ages for New
England coastal moraines, Martha’s Vineyard and Cape Cod, Massachusetts, USA.
Quaternary Science Reviews 21:2127–2135.
Baldwin, W.E., R.A. Morton, T.R. Putney, M.P. Katuna, M.S. Harris, P.T. Gayes, N.W. Driscoll,
J.F. Denny, and W.C. Schwab. 2006. Migration of the Pee Dee River system inferred from
ancestral paleochannels underlying the South Carolina grand strand and Long Bay inner
shelf. Geological Society of America Bulletin 118(5/6):533–549.
Ballard, R.D. and E. Uchupi. 1975. Triassic rift structure in Gulf of Maine. The American
Association of Petroleum Geologists Bulletin 59:1041–1072.
Balsillie, J.H. and J.F. Donoghue. 2004. High resolution sea-level history for the Gulf of Mexico
since the last glacial maximum. Report of Investigations 103. Florida Geological Survey.
Tallahassee.
240
Bard, E., B. Hamelin, M. Arnold, L. Montaggioni, G. Cabioch, G. Faure, and F. Rougerie. 1996.
Sea level record from Tahiti corals and the timing of deglacial meltwater discharge. Nature
382:241–244.
Bard, E., B. Hamelin, and D. Delanghe-Sabatier. 2010. Deglacial meltwater pulse 1B and
Younger Dryas sea levels revisited with boreholes at Tahiti. Science 327:1235–1237.
Bard, E., B. Hamelin, and R. Fairbanks. 1990a. U-Th ages obtained by mass spectrometry in
corals from Barbados: Sea level during the past 130,000 years. Nature 346:456–458.
Bard, E., B. Hamelin, R.G. Fairbanks, and A. Zindler. 1990b. Calibration of the 14C timescale
over the past 30,000 years using mass spectrographic U-Th ages from Barbados corals.
Nature 345:405–410.
Barnette, M.C. 2003. Shipwrecks of the Sunshine State. Association of Underwater Explorers.
Barnhardt, W.A. 1994. Late Quaternary relative sea-level change and evolution of the Maine
inner continental shelf 12-7 ka B.P. Unpublished Ph. D. thesis, University of Maine, Orono,
ME. 196 p.
Barnhardt, W.A., B. Andrews, and B. Butman. 2006. High resolution geological mapping of the
inner continental shelf: Nahant to Gloucester, Massachusetts. U.S. Geological Survey OpenFile Report 2005-1293.
Barnhardt, W.A., D.F. Belknap, and J.T. Kelley. 1997. Sequence stratigraphy of submerged
river-mouth deposits in the northwestern Gulf of Maine: Responses to relative sea-level
changes. Geological Society of America Bulletin 109:612–630.
Barnhardt, W.A., W.R. Gehrels, and J.T. Kelley. 1995. Late Quaternary relative sea-level change
in the western Gulf of Maine: Evidence for a migrating glacial forebulge. Geology 23:317–
320.
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996a. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Piscataqua River
to Biddeford Pool. Maine Geological Survey Geologic Map 96-6, 1:100,000.
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996b. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Ogunquit to the
Kennebec River. Maine Geological Survey Geologic Map 96-7, 1:100,000.
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996c. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Cape Elizabeth to
Pemaquid Point. Maine Geological Survey Geologic Map 96-8, 1:100,000.
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996d. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Boothbay Harbor
to North Haven. Maine Geological Survey Geologic Map 96-9, 1:100,000.
241
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996e. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Rockland to Bar
Harbor. Maine Geological Survey Geologic Map 96-10, 1:100,000.
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996f. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Mt. Desert Island
to Jonesport. Maine Geological Survey Geologic Map 96-11, 1:100,000.
Barnhardt, W.A., J.T. Kelley, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1996g. Surficial
geology of the inner Continental Shelf of the northwestern Gulf of Maine: Petit Manan Point
to West Quoddy Head. Maine Geological Survey Geologic Map 96-12, 1:100,000.
Basa, L. 1982. Fluted points in Vermont. Archaeology in Eastern North America 10:42-43.
Bascom, W. 1976. Deep water, ancient ships: The treasure vault of the Mediterranean. Garden
City, NY: Doubleday.
Bauer, K.J. 1988. A maritime history of the United States: The role of America's seas and
waterways. Columbia: University of South Carolina Press.
Belknap, D.F. 1983. Potentials of discovery of human occupation sites on the continental shelf
and near shore coastal zone. In: Proceedings of the Fourth Annual Gulf of Mexico
Information Transfer Meeting. Minerals Management Service, Gulf of Mexico Regional
Office, New Orleans. Pp. 382–387.
Belknap, D.F. and J.C. Kraft. 1977. Holocene sea level changes and coastal stratigraphic units on
the northwest flank of the Baltimore Canyon trough geosyncline. Journal of Sedimentary
Petrology 47(2):610–629.
Belknap, D.F. and J.C. Kraft. 1981. Preservation potential of transgressive coastal lithosomes on
the U.S. Atlantic Shelf marine. Geology 42:429–442.
Belknap, D.F. and J.C. Kraft. 1985. Influence of antecedent geology on stratigraphic
preservation potential and evolution of Delaware’s barrier systems. In: G.F. Oertel and S.P.
Leatherman, eds. Barrier islands. Special issue of Marine Geology, Vol. 63. Pp. 235–262.
Benjamin, J. 2010. Submerged prehistoric landscapes and underwater site discovery:
reevaluating the ‘Danish Model’ for international practice. Journal of Island & Coastal
Archaeology 5:253–270.
Bender, B. 1985. Emergent tribal formations in the American midcontinent. American Antiquity
50(1):52–62.
Berg, D. and D. Berg. 1991. Florida shipwrecks: The diver’s guide to shipwrecks around the
state of Florida and the Florida Keys. East Rockaway, NY: Aqua Explorers, Inc.
Berman, B.D. 1972. Encyclopedia of American shipwrecks. Boston: The Mariners Press, Inc.
242
Bernabo, J.C. and T. Webb III. 1977. Changing patterns in the Holocene pollen record of
northeastern North America: A mapped summary. Quaternary Research 8:64–96.
BFDC. 2007. North Carolina wreck diving. Internet website: http://www.nc-wreckdiving.com.
Bishop, G.A., B.K. Meyer, K. Vance, and F.J. Rich. In press. Geoarchaeological research at St.
Catherines Island: Defining the geological foundation. In: G.A. Bishop, H.B. Rollins and
D.H. Thomas, eds. Proceedings of the Caldwell Conference IV, Geoarchaeology of St.
Catherines and the Georgia Bight. Anthropological Papers of the American Museum of
Natural History.
Bishop, G.A., H.B. Rollings, and D.H. Thomas, eds. In press. Proceedings of the Caldwell
Conference IV, Geoarchaeology of St. Catherines and the Georgia Bight. Anthropological
Papers of the American Museum of Natural History.
Blackwelder, B.W., O.H. Pilkey, and J.D. Howard. 1979. Late Wisconsin sea levels on the
southeast U.S. Atlantic Shelf based on in-place shoreline indicators. Science 204:618–620.
Blanton, D.B. 1996. Accounting for submerged mid-Holocene archaeological sites in the
Southeast: A case study from the Chesapeake Bay estuary, Virginia. In: K.E. Sassaman and
D.G. Anderson, eds. Archaeology of the mid-Holocene Southeast. Gainesville: University
Press of Florida. Pp. 200–217
Blanton, D.B. and K.E. Sassaman. 1989. Pattern and process in the Middle Archaic period of
South Carolina. In: A. Goodyear and G. Hanson, eds. Studies in South Carolina
archaeology: Essays in honor of Robert L. Stephenson. Anthropology Studies No. 9.
Occasional Papers of the South Carolina Institute of Archaeology and Anthropology,
Columbia. Pp. 53–71
Bloom, A.L., 1998. Geomorphology. A systematic analysis of Late Cenozoic landforms. 3rd ed.
Prentice Hall, Upper Saddle River, NJ.
Blum, M.D., T.J. Misner, E.S. Collins, D.B. Scott, R.A. Morton, and A. Aslan. 2001. Middle
Holocene sea-level rise and highstand at +2m, center Texas coast. Journal of Sedimentary
Research 71(4):581–588.
Blumler, M.A. 1996. Ecology, evolutionary theory and agricultural origins. In: D.R. Harris, ed.
Origins and spread of agriculture and pastoralism in Eurasia. Washington, D.C.:
Smithsonian Institution Press. Pp. 25–49.
Bogart, E.L. 1912. The economic history of the United States. New York: Longman, Green, and
Co.
Boisvert, R.A. 1998. The Israel River complex: A Paleoindian manifestation in Jefferson, New
Hampshire. Archaeology of Eastern North America 26:97–106.
243
Boldurian, A.T. 2006. A Clovis fluted point from coastal waters in southern New Jersey. North
American Archaeologist 27:245–270.
Bonnichsen, R., D. Stanford and J.L. Fasthook. 1987. Environmental change and developmental
history of human adaptive patterns, the Paleo Indian case. In: W.F. Ruddiman and H.E.
Hright, eds. North America and adjacent oceans during the last deglaciation, Vol. K-3.
Geological Society of America, Boulder, CO. Pp. 403–424.
Bonnichsen, R. and K.L. Turnmire. 1999. Ice age people of North America. Corvalis: Oregon
State University Press.
Booth, R.K. 1998. Palynology and depositional history of late Pleistocene and Holocene coastal
sediments from St. Catherines Island, Georgia, U.S.A. Palynology 22:236–237.
Booth, R.K., G.A. Bishop, and F.J. Rich. 1999. Palynology and depositional history of late
Pleistocene and Holocene coastal sediments from St. Catherines Island, Georgia, U.S.A.
Palynology 23:67–86.
Borns, H.W., L.A. Doner, C.C. Dorion, G.L. Jacobson, M.R. Kaplan, K.J. Kreutz, T.V. Lowell,
W.B. Thompson, and T.K. Weddle. 2004. The deglaciation of Maine, U.S.A. In: J. Ehlers
and P.L. Gibbard, eds. Quaternary glaciations—extent and chronology, Part II. Amsterdam:
Elsevier. Pp. 89–109.
Boss, S.K., C.W. Hoffman, and B. Cooper. 2002. Influence of fluvial processes on the
Quaternary geologic framework of the Continental Shelf, North Carolina, USA. Marine
Geology 183(1-4):45–65.
Boston
Marine
Society.
2001.
Boston
http://www.bostonmarinesociety.org/.
Marine
Society.
Internet
website:
Bouras, E.F. and P.M. Bock. 1997. Recent Paleoinidan discovery: The first people in the White
Mountain region of New Hampshire. The New Hampshire Archeologist 37:48–58.
Bourn, R.Q. Jr. 1972. New submerged sites of coastal Connecticut. Bulletin of the
Archaeological Society of Connecticut 37:5–16.
Bourn, R.Q. Jr. 1977. The Hammonasset Beach site (6 NH 36). Bulletin of the Archaeological
Society of Connecticut 40:14–39.
Bradley, B. and D. Stanford. 2004. The North Atlantic ice-edge corridor: A possible Palaeolithic
route to the New World. World Archaeology 36(4):459–478.
Braun, D.P. 1974. Explanatory models for the evolution of coastal adaptation in prehistoric
eastern New England. American Antiquity 39:582–596.
Braund, K.E. Holland. 1993. Deerskins & duffels: Creek Indian trade with Anglo-America,
1685–1815. Lincoln: University of Nebraska Press.
244
Brech, A. 2004. Neither ocean nor continent: Correlating the archaeology and geomorphology of
the barrier islands of east central Florida. M.A. Thesis, University of Florida, Gainesville,
FL.
Broster, J.B. and M.R. Norton. 1993. The Carson-Conn-Short site: An extensive Clovis
habitation in Benton County, Tennessee. Current Anthropology 20:661–684.
Broward County Climate Change Task Force (Science and Technical Subcommittee). 2009.
Recommendations
on
Sea
Level
Rise.
http://www.broward.org/climatechange/05062009_materials_sealevelrise.pdf.
Browder, A.G. and J.E. McNinch. 2006. Linking framework geology and nearshore morphology:
Correlation of paleo-channels with shore-oblique sandbars and gravel outcrops. Marine
Geology 231(1-4):141–162.
Brown, H., comp. 2008. International registry of sunken ships. Electronic file, Published by the
compiler: Regina, Saskatchewan, Canada.
Brown, L. 1979. Fluted projectile points in Maryland. Unpublished report to Dr. William
Gardner, Catholic University of America, Washington, D.C. Ms. on file, Maryland
Historical Trust.
Broyles, B.J. 1971. Second preliminary report: The St. Albans site, Kanawha Valley, West
Virginia. Report of Archeological Investigations No. 3. West Virginia Geological and
Economic Survey, Charleston.
Bullen, R.P. and H.B. Greene. 1970. Stratigraphic tests at Stallings Island, Georgia. Florida
Anthropologist 23:8–28.
Bunker, V. 1992. Stratified components of the Gulf of Maine Archaic tradition at the Eddy site,
Amoskeag Falls. In: B. Robinson, J. Peterson, and A. Robinson, eds. Early Holocene
occupation in northern New England. Occasional Publications in Maine Archaeology, No. 9,
Maine Historic Preservation Commission, Augusta. Pp. 135–148.
Bunn, A.R. and B.A. McGregor. 1980. Morphology of the North Carolina continental slope,
western North Atlantic, shaped by deltaic sedimentation and slumping. Marine Geology
37(3-4):253–266.
Burdette, K.E., W.J. Rink, G.H. Means, and R.W. Portell. 2009. Optical dating of the Anastasia
Formation, northeastern Florida, USA. Southeastern Geology 46(4):173–185.
Butler, J.A. 1997. Sailing on Friday: The perilous voyage of America’s Merchant Marine.
Dulles, VA: Potomac Books.
Butman, B., M. Noble, and D.W. Folger. 1979. Long-term observations of bottom current and
bottom sediment movement on the Mid-Atlantic continental shelf. Journal of Geophysical
Research 84:1187–1205.
245
Butman, B., P.S. Alexander, C.K. Harris, P.A. Traykovski, M.B. Tenbrink, F.S. Lightsom, and
M.A. Martini. 2003. Oceanographic observations in the Hudson Shelf Valley, December
1999–April 2000: data report. U.S. Geological Survey Open-File Report 02-217.
Byers, D.S. 1954. Bull Brook – A fluted point site in Ipswich, Massachusetts. American
Antiquity 19:343–351.
Byers, D.S. 1959. Radiocarbon dates for the Bull Brook site, Massachusetts. American Antiquity
24:427–429.
Cable, J.S. 1982. Lithic assemblage analysis: 31CH29. In: S.R. Claggett and J.S. Cable, comps.
The Haw River sites: Archaeological investigation at two stratified sites in the North
Carolina Piedmont. Jackson, MI: Commonwealth Associates. Pp. 637–688.
Caldwell, J.R. 1958. Trend and tradition in the prehistory of the United States. Memoirs of the
American Anthropological Association Number 88.
Cambron, J.W. and D.C. Hulse. 1983. Handbook of Alabama archaeology: Part 1, point types.
Alabama Archaeological Society, Huntsville.
Cannon, M.D. and D.J. Meltzer. 2004. Early Paleoindian foraging: Examining the faunal
evidence for large mammal specialization and regional variability in prey choice.
Quaternary Science Reviews 23:1955–1987.
Carbone, V.A. 1976. Environment and prehistory in the Shenandoah Valley. Ph.D. Dissertation,
Catholic University of America, Washington, D.C.
Carbone, V.A. 1983. Late Quaternary environments in Florida and the Southeast. The Florida
Anthropologist 36:3–17.
Carlson, A.E. 2010. What caused the Younger Dryas cold event? Geology 38:383–384.
Carper, R.G. 2007. Current applications: Underwater archaeology, exploring the submerged
continental shelf. Current Anthropology 48:779.
Carr, K.W. 1996. Environmental change and the early Middle Archaic in Pennsylvania. Paper
presented at the Annual Meeting of the Society for American Archaeology, New Orleans.
Carr, K.W. 1998. The Early Archaic period in Pennsylvania. Pennsylvania Archaeologist
68:2:42–69.
Carr, K.W. and J.M. Adovasio. 2002. Paleoindians in Pennsylvania. In: K.W. Carr and J.M.
Adovasio, eds. Ice Age peoples of Pennsylvania. Recent Research in Pennsylvania Number
2, Pennsylvania Historical and Museum Commission, Harrisburg. Pp. 1–50.
Carey, J.S., R.E. Sheridan, G.M. Ashley, and J. Uptegrove. 2005. Glacially-influenced late
Pleistocene stratigraphy of a passive margin: New Jersey’s record of the North American ice
sheet. Marine Geology 218:155–173.
246
Casey, J. n.d. A short history of commercial fishing in the Chesapeake Region. Fisheries Service,
Annapolis, MD: Maryland Department of Natural Resources.
Cassavoy, K.A. and K.J. Crisman. 1988. The War of 1812: Battle for the Great Lakes. In: G.F.
Bass, ed. Ships and shipwrecks of the Americas: A history based on underwater
archaeology. New York: Thames and Hudson. Pp. 169–188.
Cavallo, J. 1981. Turkey Swamp: A late Paleo-Indian site in New Jersey’s Coastal Plain.
Archaeology of Eastern North America 9:1–18.
Chapelle, H.I. 1949. The history of the American sailing Navy: The ships and their development.
New York: Bonanza Books.
Chappell, J. and N.J. Schackleton. 1986. Oxygen isotopes and sea level. Nature 324:137–140.
Chappell, J. 2002. Sea level changes forced ice breakouts in the last glacial cycle: New results
from coral terraces. Quaternary Science Reviews 21(10):1229–1240.
Chapman, J. 1975. The Rose Island site and the bifurcate point tradition. University of Tennessee
Department of Anthropology Report of Investigations No. 14.
Chapman, J. 1981. The Bacon Bend and Iddins sites: The Late Archaic period in the lower
Tennessee River Valley. University of Tennessee, Department of Anthropology, Report of
Investigations 31. Knoxville.
Chapman, J. and P.J. Watson. 1993. The Archaic period and the flotation revolution. In: M.C.
Scarry, ed. Foraging and Farming in the Eastern Woodlands. Gainesville: University Press
of Florida. Pp. 27–38.
Charles, J. 1997. Mid-Atlantic shipwreck accounts to 1899: Over 1400 entries for New Jersey,
Pennsylvania, Delaware, Delaware Bay, Maryland, Virginia, Chesapeake Bay, North
Carolina. Published by the author, Hampton, VA.
Charles, J. 1999. Mid-Atlantic shipwreck accounts to 1914: Over 1700 entries for New Jersey,
Pennsylvania, Delaware, Delaware Bay, Maryland, Virginia, Chesapeake Bay, North
Carolina. Published by the author, Hampton, VA.
Charles, J. 2003. New Jersey, Delaware, Pennsylvania, shipwreck accounts, 1705 to 1950,
including Delaware Bay and Delaware River. Published by the author, Hampton, VA.
Charles, J. 2004a. North Carolina shipwreck accounts, 1709 to 1950, including over 1100 named
wrecks. Published by the author, Hampton, VA.
Charles, J. 2004b. Virginia and Maryland shipwreck accounts, 1623 to 1950, including
Chesapeake Bay: Over 1200 named wrecks. Published by the author, Hampton, VA.
247
Chen, Z.Q. 1992. An investigation of the late Pleistocene paleo-channel systems in the
continental shelf, south of Chesapeake Bay mouth. M.S. Thesis, Virginia Institute of Marine
Science, College of William & Mary, Gloucester Point, VA.
Chen, Z., C.H. Hobbs III, J.F. Wehmiller and S.M. Kimball. 1995. Late Quaternary paleochannel
systems on the Continental Shelf, south of the Chesapeake Bay entrance. Journal of Coastal
Research 11(3):605–614
Cherryman, A. 1994. An underwater archaeological survey of Belmont Bay: Method study and
site report. M.A. Thesis, George Mason University, Fairfax, VA.
Chowns, T.M., B.S. Schultz, J.R. Griffin, and M.R. Crook Jr. 2008. Relocation of Brunswick
River and other estuaries on the Georgia, USA coast as a consequence of Holocene
transgression. Southeastern Geology 45(3):143–159.
Claassen, C. 1991. Gender, shellfishing, and the Shell Mound Archaic. In: J.M. Gero and M.W.
Conkey, eds. Engendering archaeology: Women and prehistory. Oxford: Basil Blackwell.
Pp. 276–300.
Claassen, C. 1992. Shell mounds as burial mounds: A revision of the Shell Mound Archaic. In:
D. Pollack and A.G. Henderson, eds. Current archaeological research in Kentucky, Vol. 2.
Frankfort: Kentucky Heritage Council. Pp. 1-11.
Claassen, C. 1995. Dogan Point: A shell matrix site in the lower Hudson Valley. Occasional
Publications in Northeastern Anthropology, No. 14. Archaeological Services, Bethlehem,
CT.
Claflin, W.H. Jr. 1931. The Stallings Island mound, Columbia County, Georgia. Papers of the
Peabody Museum of American Archaeology and Ethnology 14(1).
Claggett, S.R. and J.S. Cable. 1982. The Haw River sites: Archeological investigations at two
stratified sites in the North Carolina Piedmont. Commonwealth Associates, Inc., Jackson,
MI. Submitted to U.S. Army Corps of Engineers, Wilmington District.
Clark, P.U., A.S. Dyke, J.D. Shakun, A.S. Carlson, J. Clark, B. Wohlfarth, J.X. Mitrovica, S.W.
Hostetler, and A.M. McCabe. 2009. The last glacial maximum. Science 325:710–714.
Clarke, T.L., Swift, D.J.P., Young, R.A., 1983. A stochastic modeling approach to the fine
sediment budget of the New York Bight. Journal of Geophysical Research 88:9653–9660.
Coastal Environments, Inc. 1977. Cultural resource evaluation of the northern Gulf of Mexico
Continental Shelf. 3 vols. Prepared for Interagency Archaeological Services, Office of
Archaeology and Historic Preservation, National Park Service, U.S. Department of the
Interior.
248
Cockrell, W.A. 1980. Drowned sites in North America. In: K. Muckleroy, ed. Archaeology
under water: An atlas of the world’s submerged sites. New York: McGraw-Hill. Pp. 138–
145.
Cockrell, W.A. and L. Murphy. 1978. Pleistocene man in Florida. Archaeology of eastern North
America. Vol. 6, Eastern States Archaeological Federation, Newark, DE.
Coe, J. 1964. The formative cultures of the Carolina Piedmont. American Philosophical Society,
Transactions 54. Philadelphia, Pennsylvania.
Coles, J.M. 1988. A wetland perspective. In: B.A. Purdy, ed. Wet site archaeology. Caldwell,
NJ: The Telford Press. Pp. 1–14.
Collins, M.B. 1999. Clovis blade technology: A comparative study of the Kevin Davis cache,
Texas. Texas Archaeology and Ethnohistory Series, University of Texas Press, Austin.
Colman, S.M., Halka, J.P., Hobbs III, C.H., Mixon, R.B., Foster, D.S. 1990. Ancient channels of
the Susquehanna River beneath Chesapeake Bay and the Delmarva Peninsula. Geological
Society American Bulletin 102:1268–1279.
Colquhuon, D.J. and M.J. Brooks. 1986. New evidence from the southeastern U.S. for eustatic
components in the Late Holocene sea levels. Geoarchaeology 1(3):275–291.
Colquhoun, D.J., I.D. Woolen, D.S. Van Nieuwenhuise, GG. Padgett, R.W. Oldham, D.C.
Boylan, J.W. Bishop, and P.D. Howell. 1983. Surface and subsurface stratigraphy structure
and aquifers of the South Carolina coastal plain. Report to the Department of Health and
Environmental Control, Ground Water Protection Division, published through the Office of
the Governor, State of South Carolina, Columbia.
Cooke, C.W. 1936. Geology of the Coastal Plain of South Carolina. U.S. Geological Survey
Bulletin 876.
Cooke, C.W. 1939. Scenery of Florida interpreted by a geologist. Florida Geological Survey
Bulletin 17.
Corlett, E. 1975. The iron ship: The history and significance of Brunel’s Great Britain. Bradfordon-Avon: Moonraker Press.
Cowan, C.W. 1985. Understanding the evolution of plant husbandry in eastern North America:
Lessons from botany, ethnography, and archaeology. In: R.I. Ford, ed. Prehistoric food
production in North America. Anthropological Papers 75, Museum of Anthropology.
University of Michigan, Ann Arbor. Pp. 205–243.
Cox, B.L. and J.B. Petersen. 1997. The Varney Farm (ME 36-57): A late Paleoindian
encampment in western Maine. Bulletin of the Maine Archaeological Society 37(2):25–48.
Cresthull, P. 1971. Chance 18SO5: A major Early Archaic site. Maryland Archeology 7:51–52.
249
Cresthull, P. 1972. Chance 18SO5: A major Early Archaic site, Part 2. Maryland Archeology
8:40–53.
Crisman, K.J. 1988. Struggle for a continent: Naval battles of the French and Indian Wars. In:
G.F. Bass, ed. Ships and shipwrecks of the Americas: A history based on underwater
archaeology. New York: Thames and Hudson. Pp. 129–148.
Crock, J.G., J.B. Petersen, and R.M. Anderson. 1993. Scalloping for artifacts: A biface and
plummet from eastern Blue Hill Bay, Maine. Archaeology of Eastern North America
21:179–192.
Cronin, T.M., ed. 2000. Initial report on IMAGES V cruise of the Marion-Dufresne to the
Chesapeake Bay June 20–22, 1999. U.S. Geological Survey Open-File Report 00-306. U.S.
Department of the Interior, U.S. Geological Survey. Reston, VA.
Cronin, T.M., P.R. Vogt, and D.A. Willard. 2002. Rapid sea-level rise at 8.0–7.5 ka and
Antarctic Ice Sheet thinning. Abstracts of the American Geophysical Union Fall Meeting,
(#OS71D-0324).
Cross, D. 1941. Archaeology of New Jersey, Vol. 1. Archaeological Society of New Jersey and
New Jersey State Museum, Trenton.
Cross, J.R. 1999. “By any other name…”: A reconsideration of Middle Archaic lithic technology
and typology in the Northeast. In: M.A. Levine, K. Sassaman, and M. Nassaney, eds. The
archaeological Northeast. Westport, CT: Brigin & Gravey. Pp. 57–74.
Crothers, A.G. 2004. Commercial risk and capital formation in early America: Virginia
merchants and the rise of marine insurance, 1750–1815. The Business History Review
78(4):607–633.
Crusoe, D.L. and C.B. DePratter. 1976. A new look at the Georgia coastal shell mound Archaic.
Florida Anthropologist 29:1–23.
Culpepper, W.S., C.L. Heath, J.D. Irwin, and J.M. Herbert. 2000. From Drowning Creek to
Sicily: Archaeological investigations at Fort Bragg, North Carolina. Fort Bragg Cultural
Resources Program. Cultural Resources Management Series Research Report No. 2. Report
submitted to the North Carolina State Historic Preservation Office, Raleigh.
Culver, H.B. 1992. The book of old ships: From Egyptian galleys to clipper ships. Republished.
New York: Dover Publication, Inc. Originally published in 1924 as The book of old ships:
And something of their evolution and romance, etc. Garden City, NY: Doubleday, Page and
Company.
Curry, D.C. and C.A. Ebright. 1990. Buried Archaic sites in ridgetop settings on the Middle
Atlantic Coastal Plain. Paper presented at the Joint Archaeological Congress, Baltimore,
MD.
250
Custer, J.F. 1984. Delaware prehistoric archaeology: An ecological approach. Cranbury, NJ:
Associated University Presses.
Custer, J.F. 1986. Analysis of early Holocene projectile points and site locations from the
Delmarva Peninsula. Archaeology of Eastern North America 14:45–64.
Custer, J.F. 1988. Coastal adaptations in the Middle Atlantic region. Archaeology of Eastern
North America 16:121–136.
Custer, J.F. 1989. Prehistoric cultures of the Delmarva Peninsula: An archaeological study.
Newark: University of Delaware Press.
Custer, J.F. 1990. Chronology of Virginia’s Early and Middle Archaic periods. In: T.R. Reinhart
and M.E.N. Hodges, eds. Early and Middle Archaic research in Virginia: A synthesis.
Special Publication No. 22 of the Archeological Society of Virginia, Richmond. Pp. 1–60.
Custer, J.F. 1996. Prehistoric cultures of eastern Pennsylvania. Anthropological Series 7,
Pennsylvania Historical and Museum Commission, Harrisburg.
Custer, J.F., J. Cavallo, and R.M. Stewart. 1983. Paleo-Indian adaptations of the Coastal Plain of
Delaware and New Jersey. North American Archaeologist 4:263–276.
Custer, J.F. and G. Galasso. 1980. Lithic resources of the Delmarva Penninsula. Maryland
Archaeology 16(2):1–13.
Custer, J.F. and D. Lowery. 1994. Recent excavations at the Paw Paw Cove Paleoindian site,
Talbot County, Maryland. Current Research in the Pleistocene 11:28–30.
Cutler, K.B., R.L. Edwards, F.W. Taylor, H. Cheng, J. Adkins, C.D. Gallup, P.M. Cutler, G.S.
Burr, and A.L. Bloom. 2003. Rapid sea-level fall and deep-ocean temperature change since
the last interglacial period. Earth and Planetary Science Letters, 206(3-4):253–271.
Daiber, F.C. 1976. An atlas of Delaware’s wetlands and estuarine resources. Delaware Coastal
Zone Management Program, Technical Report 2. Newark, Delaware.
Daley, B. 2005. Sunken treasure: Scientists find evidence of ancient forest buried under the
seabed of Nantucket Sound. The Boston Globe, December 4, page 30.
Dame, J.K. 1990. Origin of a solitary sand shoal offshore of Sandbridge Beach, Virginia. M.A.
thesis, Virginia Institute of Marine Science, College of William & Mary, Gloucester Point,
VA.
Daniel, I.R. Jr. 1994. Raw material availability and Early Archaic settlement in the Southeast. In:
D.G. Anderson and K.E. Sassaman, eds. The Paleoindian and Early Archaic Southeast.
Tuscaloosa: University of Alabama Press. Pp.84–91.
Daniel, I.R. Jr. 1997. North Carolina Paleoindian point survey. Paper presented at the 54th
Southeastern Archaeological Conference, Baton Rouge, Louisiana.
251
Daniel, I.R. Jr. 1998. Hardaway revisited: Early Archaic settlement in the Southeast. Tuscaloosa:
University of Alabama Press.
Daniel, I.R. Jr. 2000. Paleoindian points in North Carolina. Current Research in the Pleistocene
17:14–16.
Daniel, I.R. and R.P.S. Davis Jr. 1996. Projectile point classification project: The classification
of projectile points in existing archaeological collections from North Carolina (Phase II).
Report submitted to Division of Archives and History, Raleigh.
Daniel, I.R. Jr. and M. Wisenbaker. 1987. Harney Flats: A Florida Paleo-Indian site.
Farmingdale, NY: Baywood Publishing Co.
Davidson, T.E. 1981. Cultural resource management plan for the lower Delmarva region of
Maryland. Maryland Historical Trust Monograph No. 2, Annapolis, MD.
Davis, R.P.S. and I.R. Daniel Jr. 1990. Projectile point classification project: The classification
of projectile points in existing archaeological collections from North Carolina. Report on
file, Office of State Archaeology, Raleigh.
Delabarre, E.B. 1928. A prehistoric skeleton from Grassy Island. American Anthropologist
30:476–480.
Delgado, J.P. 1989. The maritime heritage of the United States National Historic Landmark
theme study, Part One: Large preserved historic vessels – lightships. National Park Service.
http://www.nps.gov/history/maritime/ltshipnhltheme.htm.
DeLoach, N. 2000. Ned DeLoach’s diving guide to underwater Florida. Jacksonville, FL: New
World Publications.
Dent, R.J. 1995. Chesapeake prehistory: Old traditions, new directions. New York: Plenum
Press.
Dent, R.J. and B.E. Kauffman. 1985. Aboriginal subsistence and site ecology as interpreted from
microfloral and faunal remains. In: C.W. McNett, ed. Shawnee Minisink: A stratified
Paleoindian–Archaic site in the upper Delaware Valley of Pennsylvania. Orlando: Academic
Press. Pp. 55–79.
Denton, G.H., and T.J. Hughes. 1981. The last great ice sheets. New York: Wiley.
DePratter, C.B. 1979. Ceramics. In: D.H. Thomas and C.S. Larsen, eds. The anthropology of St.
Catherines Island, The Refuge-Deptford mortuary complex. Anthropological Papers of the
American Museum of Natural History 56(1). Pp. 109–132.
DePratter, C.B., and J.D. Howard. 1977. History of shoreline changes determined by
archaeological dating: Georgia coast, United States. University of Georgia Marine Institute
337.
252
DePratter, C.B. and J.D. Howard. 1980. Indian occupation and geological history of the Georgia
coast: A 5000 year summary. In: J.D. Howard, C.B. DePratter, and R.W. Frey, eds.
Excursions in southeastern geology: The archaeology-geology of the Georgia coast. The
Geologic Society of America, Guidebook 20. Atlanta, GA. Pp. 1–65.
Devanney, J. 2006. The tankship Tromedy: The impending disasters in tankers. 2nd edition.
Tavernier, FL: CTX Press.
Dillehay, T.D. 1989. Monte Verde: A late Pleistocene settlement in Chile. Washington, D.C.:
Smithsonian Press.
Dillon, W.P. 1970. Submerged effects on a Rhode Island barrier and lagoon and inference on
migration of barrier. Geology 78:94–106.
Dillon, W.P. and R.N. Oldale. 1978. Late Quarternary sea-level curve: Reinterpretation based on
glaciotectonic influence. Geology 6:56–60.
Dincauze, D.F. 1972. The Atlantic Phase: A Late Archaic culture in Massachusetts. Man in the
Northeast 4:40–61.
Dincauze, D.F. 1976. The Neville site: 8,000 years at Amoskeag. Peabody Museum Monographs
No. 4. Cambridge, MA: Harvard University.
Dincauze, D.F. 1990. A capsule prehistory of southern New England. In: L.M. Hauptman and
J.D. Wherry, eds. The Pequots in southern New England: The fall and rise of an American
nation. Norman: University of Oklahoma Press. Pp. 19–32.
Dincauze, D.F. 1993. Fluted points in the eastern forest. In: O. Soffer and N.D. Praslov, eds.
From Kostenki to Clovis: Upper Paleolithic Paleo-Indian adaptations. New York: Plenum
Press. Pp. 279–292.
Dincauze, D.F. 2007. Earliest Americans: Northeast regional context. National Park Service,
available at http://www.nps.gov/archeology/pubs/nhleam/E-Northeast.htm.
Dincauze, D.F. and M. Mulholland. 1977. Early and Middle Archaic site distributions and
habitats in southern New England. Annals of the New York Academy of Sciences 288:439–
456.
Dineen, R.J., DeSimone, D.J., Hanson, E.L., 1988. Glacial Lake Albany and its successors in the
Hudson Lowlands. In: Brigham-Grette, J., ed. Field guidebook AMQUA 1899. Department
of Geology, University of Massachusetts, Amherst, MA, pp. 1–40.
Dix, J., R. Quinn, and K. Westley. 2008. Re-assessment of the archaeological potential of
continental
shelves.
ALSF
Project
Number
3362.
http://ads.ahds.ac.uk/catalogue/resources.html?continentshelves_eh_2008.
Dixon, E.J. 1999. Boats, bones, and bison: Archeology and the first colonization of western
North America. Albuquerque: University of New Mexico Press.
253
Dixon, E.J. 2001. Human colonization of the Americas: Timing, technology and process.
Quaternary Science Reviews 20:277–299.
Donnelly, J.P. 1998. Evidence of late Holocene post-glacial isostatic adjustment in coastal
wetland deposits of eastern North America. Georesearch Forum (3-4):393–400.
Donoghue, J.F. 2006. Geography and geomorphology of the Aucilla River region. In: S.D.
Webb, ed. First Floridians and last mastodons: The Page-Ladson site in the Aucilla River.
Dordrecht: Springer. Pp. 31–48.
Donoghue J.F. and W.F. Tanner. 1992. Quaternary terraces and shorelines of the panhandle
Florida region. In: J.F. Wehmiller and C.H. Fletcher, eds. Quaternary coasts of the United
States: Marine and lacustrine systems. SEPM Special Publication 48. Pp. 233–241.
Doran, G.H. 2002. The Windover radiocarbon chronology. In: G.H. Doran, ed. Windover:
Multidisciplinary investigations of an Early Archaic Florida cemetery. Gainesville:
University Press of Florida. Pp. 59–72.
Dorsey, R.A. 1997. Predicted sea level changes, Florida Gulf Coast over the past 21,000 years.
Gulf Archaeology Research Institute, Crystal River, FL.
Doucette, D. 2005. Reflections of the Middle Archaic: A view from Annasnappet Pond. Bulletin
of the Massachusetts Archaeological Society 66(1):22–33.
Doucette, D. and J.R. Cross. 1997. Annasnappet Pond archaeological district, North Carver
Massachusetts: An archaeological data recovery program. The Public Archaeology
Laboratory, Inc. Report No. 580. Prepared for U.S. Department of Transportation, Federal
Highway Administration and Massachusetts Highway Department, Boston.
Doucette, D. and J.R. Cross. 1998. Archaeological investigations within the Annasnappet Pond
archaeological district: Data recovery program for Loci 1, 2, 8 and 9, Carver, Massachusetts.
The Public Archaeology Laboratory, Inc., Report No. 580. Prepared for the Massachusetts
Highway Department.
Doyle, R. Jr., N. Hamilton, J. Petersen, and D. Sanger. 1985. Late Paleo-Indian remains from
Maine and their correlations in northeastern prehistory. Archaeology of Eastern North
America 13:1–34.
Dragoo, D.W. 1976. Some aspects of eastern North American prehistory: A review 1975.
American Antiquity 41:3–27.
Driver, D. 2004. 30-acre wetland creation project, McFaddin National Wildlife Refuge
archaeological reconnaissance and assessment, Jeferson County, Texas (MAC PN 04-85).
Letter report, Moore Archaeological Consulting, Inc.
Dudek, M.G. 2005. The Whortleberry Hill site: An early Holocene camp in Dracut, MA. Bulletin
of the Massachusetts Archaeological Society 66:12–21.
254
Dumont, J. 1981. The PaleoIndian–Early Archaic continuum: An environmental approach.
Archaeology of Eastern North America 9:18–37.
Dunbar, J.S. 1988. Archaeological sites in the drowned Tertiary karst region of the eastern Gulf
of Mexico. Florida Anthropologist. 41(1):177–181.
Dunbar, J.S. 2002. Chronostratigraphy and paleoclimate of late Pleistocene Florida and the
implications of changing Paleoindian land use. M.S. Thesis, Florida State University,
Tallahassee.
Dunbar, J.S. 2006a. Pleistocene–early Holocene climate change: Chronostratigraphy and
geoclimate of the southeast U.S. In: S.D. Webb, ed. First Floridians and last mastodons: The
Page-Ladson site in the Aucilla River. Dordrecht: Springer. Pp. 103–158.
Dunbar, J.S. 2006b. Paleoindian archaeology. In: S.D. Webb, ed. First Floridians and last
mastodons: The Page-Ladson site in the Aucilla River. Dordrecht: Springer. Pp. 403–435.
Dunbar, J.S., M.K. Faught, and S.D. Webb. 1988. Page/Ladson (8JE591): An underwater PaleoIndian site in northwestern Florida. Florida Anthropologist 36:18–30.
Dunbar, J.S., M.K. Faught, and S.D. Webb. 1992. Inundated prehistoric sites in Apalachee Bay,
Florida, and the search for the Clovis shoreline. In: L. Johnson and M. Stright, eds.
Paleoshores and prehistory: An invetigation of method. Boca Raton, FL: CRC Press. Pp.
117–146.
Dunbar, J.S., S.D. Webb and D. Cring. 1989. Culturally and naturally modified bones from a
Paleo-Indian site in the Aucilla River, north Florida. In: R. Bonnichsen and M. Sorg, eds.
Bone modification. Center for the Study of the First Americans, University of Maine,
Orono. Pp. 473–97.
Duncan, C.S., J.A. Goff, J.A. Austin Jr., C.S. Fulthorpe. 2000. Tracking the last sea-level cycle:
seafloor morphology and shallow stratigraphy of the latest Quaternary New Jersey middle
continental shelf. Marine Geology 170:395–421.
Dunford, F.J. 1999. Paleoenvironmental context for the Middle Archaic occupation of Cape Cod,
Massachusetts. In: M.A. Levine, K. Sassaman, and M. Nassaney, eds. The archaeological
Northeast. Westport, CT: Brigin & Gravey. Pp. 39–54.
Durrie and Peck. 1834. The mariner’s chronicle: Containing narratives of the most remarkable
disasters at sea, such as shipwrecks, storms, fires, and famines; also, naval engagements,
piratical adventures, incidents of discovery, and other extraordinary and interesting
occurrences. New Haven, CT: Durrie and Peck. On file, Boston Public Library, Boston.
Ebright, C. 1989. Maryland’s oldest residents: Archaeological investigations at the Higgins site.
The Maryland Naturalist 33(1-2):1–29.
255
Ebright, C. 1992. Early Native American prehistory on the Maryland Western Shore:
Archeological investigations at the Higgins site. Maryland State Highway Administration,
Project Planning Division, Environmental Evaluation Section Archeological Report Number
1 (Maryland Geological Survey, Division of Archeology File Report Number 250),
Baltimore.
Edwards, R.L. and K.O. Emery. 1977. Man on the Continental Shelf. In: W.S. Newman and B.
Salwen, eds. Amerinds and their paleoenvironments in northeastern North America. Annals
of the New York Academy of Sciences 228. Pp. 245–256.
Edwards, R.L. and A.L. Merrill. 1977. Reconstruction of the Continental Shelf areas of eastern
North America for the times 9,500 B.P. and 12,500 B.P. Archaeology of Eastern North
America 5:1–43.
Elliott, D. and R. Doyon. 1981. Archaeology and historical Geography of the Savannah River
floodplain near Augusta, Georgia. University of Georgia Laboratory of Archaeology Series
Report No. 22. Athens.
Elliott, D.T., R.J. Ledbetter, and E.A. Gordon. 1994. Data recovery at Lovers Lane, Phinizy
Swamp, and the Old Dike sites, Bobby Jones Expressway extension corridor, Augusta,
Georgia. Occasional Papers in Cultural Resource Management No. 7. Georgia Department
of Transportation Office of Environmental Analysis, Atlanta, GA.
Elliott, D.T. and K.E. Sassaman. 1995. Archaic period archaeology of the Georgia Coastal Plain
and Coastal Zone. University of Georgia Laboratory of Archaeology Series Report No. 35.
Georgia Archaeological Research Design Paper No. 11. University of Georgia, Athens.
Ellis, C.J. and D.B. Deller. 1997. Variability in the archaeological record of northeastern early
Paleoindians: A view from southern Ontario. Archaeology of Eastern North America 25:1–
30.
Ellms, C., comp. 1836. Shipwrecks and disasters at sea, or historical narratives of the most noted
calamities, and providential deliveries from fire and famine on the ocean. Russell, Shattuck,
and Company, Boston.
Ellms, C., comp. 1840. Awful calamities: or, The shipwrecks of December, 1839; Being a full
account of the dreadful hurricanes of December 15, 21 & 27 on the coast of Massachusetts;
in which were lost more than 90 vessels and nearly 200 dismasted, driven ashore or
otherwise damaged and more than 150 lives destroyed, of which full statistics are given;
comprising also a particular relation of the shipwreck of the following vessels: Barque
Lloyd, Brigs Pocahontas, Rideout and J. Palmeri, and Schooners Deposite, Catherine
Nichols and Miller, and also of the deadly disasters at Gloucester. 3rd edition. Boston: J.
Howe.
Eltis, D. 2001. The volume and structure of the transatlantic slave trade: A reassessment. The
William and Mary Quarterly, Third Series 59(1):17–46.
256
Emery, K.O. and R.L. Edwards. 1966. Archaeological potential of the Atlantic Continental Shelf.
American Antiquity 31:733–737.
Emery, K.O. and E. Uchupi. 1972. Western North Atlantic Ocean:topography, rocks, structure,
water, life and sediments:American Association of Petroleum Geologists Memiors 17.
Environmental Law Institute. 2009. Marine spatial planning in U.S. waters: an assessment and
analysis of existing legal mechanisms, anticipated barriers, and future opportunities.
Available at http://www.policyarchive.org/handle/10207/bitstreams/22069.pdf.
Erlandson, J. 1993. Evidence for a terminal Pleistocene human occupation of Daisy Cave, San
Miguel Island, California. Current Research in the Pleistocene 10:17–21.
Erlandson, J. 1994. Early hunter-gatherers of the California coast. New York: Plenum Press.
Erlandson, J. 2002. Anatomically modern humans, maritime migrations, and the peopling of the
New World. In: N. Jablonski, ed. The first Americans: The Pleistocene colonization of the
New World. Memoirs of the California Academy of Sciences. San Francisco. Pp. 59–92.
Erlandson, J M., M.H. Graham, B.J. Bourque, D. Corbett, J.A. Estes, and R.S. Steneck. 2007.
The kelp highway hypothesis: Marine ecology, the coastal migration theory, and the
peopling of the Americas. Journal of Island and Coastal Archaeology 2:161–174.
Erlandson, J.M. and S.M. Fitzpatrick. 2006. Oceans, islands, and coasts: Current perspectives on
the role of the sea in human prehistory. Journal of Island & Coastal Archaeology 1:5–32.
Evans, C.H. 1976. Exports, domestic and foreign from the American colonies to Great Britain
from 1697 to 1789 inclusive; Exports, domestic from the United States to all countries from
1789 to 1883. Reprinted. New York: Arno Press. Originally published 1884, Washington,
D.C.: Government Printing Office.
Ewing, J., Pichon, X.L., Ewing, M., 1963. Upper stratification of Hudson apron region. Journal
of Geophysical Research 68:6303–6316.
Expedition Wydah. ca. 2008. Wydah pirate shipwreck.
http://www.whydah.com/whydah.html.
Historic Shipwrecks,
Inc.
Fairbanks, C. 1942. The taxonomic position of Stallings Island Georgia. American Antiquity
7:223–231.
Fairbanks, R.G. 1989. A 17,000-year glacoi-eustatic sea level record: Influence of glacial
melting rates on the Younger Dryas event and deep ocean circulation. Nature 342:637–642.
Fairbanks, R.G., R.A. Mortlock, T.C. Chiu, L. Cao, A. Kaplan, T.P. Guilderson, T.W. Fairbanks,
A.L. Bloom, P.M. Grootes, and M.J. Nadeau. 2005. Radiocarbon calibration curve spanning
0 to 50,000 years BP based on paired 230Th/234U/238U and 14C dates on pristine corals.
Quaternary Science Review 24:1781–1796.
257
Faught, M.K. 1996.Clovis origins and underwater prehistoric archaeology in northwestern
Florida. Ph.D. Dissertation, University of Arizona, Tucson.
Faught, M.K. 2002. Submerged Paleoindian and Archaic sites of the Big Bend, Florida. Journal
of Field Archaeology 29:273–290.
Faught, M.K. 2003. Geophysical remote sensing and underwater cultural resource management
of submerged prehistoric sites in Apalachee Bay: A deep-water example, site predictive
models, and site discoveries. In: M. McKay, ed. Proceedings: Twenty Second Annual Gulf
of Mexico Information Transfer Meetings, 2002. Minerals Management Service, New
Orleans. Pp. 115–129.
Faught, M.K. 2004. The underwater archaeology of paleolandscapes, Apalachee Bay, Florida.
American Antiquity 69(2):275–289.
Faught, M.K. 2008. Archaeological roots of human diversity in the New World: A compilation
of accurate and precise radiocarbon ages from earliest sites. American Antiquity 73(4):670–
698.
Faught, M.K. and J. Ambrosino. 2007. Archaeological monitoring for berth 218 dredging, Port
Sutton, Hillsborough County, Florida. Technical report on file at the Florida Master Site
File.
Faught, M.K. and J.F. Donoghue. 1997. Marine inundated archaeological sites and paleofluvial
systems: examples from a karst-controlled continental shelf setting in Apalachee Bay.
Geoarchaeology 12(5):417–458.
Faught, M.K., J.S. Dunbar and S.D. Webb. 1992. New evidence for paleoindians on the
continental shelf of northwestern Florida. Current Research in the Pleistocene 9:11–12.
Faught, M.K. and J.M. Latvis. 2000. PaleoAucilla prehistory project: Underwater prehistoric
archaeology in Apalachee Bay 1999 interim report. Program in Underwater Archaeology
Research Reports # 1. Florida State University, Tallahasee, FL.
Fedje, D.W. and T. Christensen. 1999. Modeling paleoshorelines and locating early Holocene
coastal sites in Haida Gwaii. American Antiquity 64(4):635–652.
Fehr, A., D.S. Robinson, M. Williams, J.L. Seidel, J. Irion, and D. Maher. 1996. Phase I
terrestrial and marine archaeological surveys for the Poplar Island Reclamation Project and
Phase II investigations of site 18TA237 and six marine anomalies, Talbot County,
Maryland. Final Report submitted to GBA-M&N, A Joint Venture, and R. Christopher
Goodwin & Associates, Inc., Frederick, MD.
Ferrari, B. and J. Adams. 1990. Biogenic modification of marine sediments and their influence
on archaeological material. International Journal of Nautical Archaeology and Underwater
Exploration 19:139–151.
258
Fiedel, S. 2000. The peopling of the New World: present evidence, new theories, and future
directions. Journal of Archaeological Research 8:39–103.
Field, M.E. and D.B. Duane. 1976. Post-Pleistocene history of the Unites States Inner
Continental Shelf: Significance to origin of barrier islands. Geologic Society of America
Bulletin 87:691–702.
Figgins, J.D. 1927. The antiquity of man in America. Natural History 27(3):229–239.
Finkl, C.W. and J.L. Andrews. 2008. Shelf geomorphology along the southeast Florida Atlantic
Continental Platform: Barrier coral reefs, nearshore bedrock, and morphosedimentary
features. Journal of Coastal Research 24(4):823–849.
Fischer, A. 1995. Man and sea in the Mesolithic: coastal settlement above and below present sea
level. Oxbow Monograph Number 53. Oxbow Books, Oxford.
Fitting, J.E. 1968. Environmental potential and the postglacial readaptation in eastern North
America. American Antiquity 33:441–445.
Fix, A.G. 2002. Colonization models and initial genetic diversity in the Americas. Human
Biology 74(1):1–10.
Fix, A.G. 2005. Rapid deployment of the five founding Amerind mtDNA haplogroups via
coastal and riverine colonization. American Journal of Physical Anthropology 128:430–436.
Fladmark, K. 1979. Routes: Alternate migration corridors for Early Man in North America.
American Antiquity 44:55–69.
Foley, F.D. 1981. Neogene seismic stratigraphy and depositional history of the lower Georgia
coast and continental shelf. M.S. Thesis, University of Georgia.
Foyle, A.M., V.J. Henry, and C. Alexander. 2001. Seawater intrusion threat to the upper
Floridian aquifer, coastal Georgia and South Carolina. Georgia Geologic Survey Publication
Series, Report 39.
Ford, R.I. 1974. Northeastern archaeology: Past and future developments. Annual Review of
Anthropology 3:385–413.
Ford, R.I. 1981. Gardening and farming before A.D. 1000: Patterns of prehistoric cultivation
north of Mexico. Journal of Ethnobiology 1:1:6–27.
Forman, S. 2003. Luminescence dating of Quaternary sediments. The Mammoth Trumpet 18
(3):10–13. Center for the Study of the First Americans, Texas A&M University, College
Station.
Forrest, D.T. 1999. Beyond presence or absence: Establishing diversity in Connecticut’s early
Holocene archaeological record. Bulletin of the Archaeological Society of Connecticut
67:79–99.
259
Foyle, A.M. and V.J. Henry. 2004. Groundwater study reveals new details on the Cenozoic
stratigraphic framework and Quaternary paleodrainage networks at the apex of the Georgia
Bight. Unpublished manuscript in possession of authors.
Foyle, A.M., V.J. Henry, and C.R. Alexander. 2001. The Miocene Aquitard and the Floridian
Aquifer of the Georgia / South Carolina coast: Geophysical mapping of potential seawater
intrusion sites. Georgia Geologic Survey Bulletin 132, Georgia Department of Natural
Resources, Atlanta.
Fraakel, L. and H.F. Thomas. 1966. Evidence of freshwater lake deposits in Block Island Sound.
Journal of Geology 74:240–242.
Freeland, G.L., D.J. Stanley, D.J.P. Swift, and D.N. Lambert. 1981. The Hudson Shelf Valley: its
role in shelf sediment transport. Marine Geology 42:399–427.
Freitag, N. 1997. Shipwrecks unforgotten: From New Jersey to the Gulf of Florida. Island Park,
NY: Finley-Greene Publications.
Freitag, N. 1998. Shipwrecks unforgotten: From Maine to the Hudson River. Island Park, NY:
Finley-Greene Publications.
French, C.J. 1987. Productivity in the Atlantic shipping industry: A quantitative study. Journal of
Interdisciplinary History. 17(3):613–638.
Friend of Mariners, comp. 1823. Accounts of shipwrecks and other disasters at sea: Designed to
be interesting and useful to mariners, with an appendix containing Dr. Payson’s address to
seamen; and a few prayers for their use. Brunswick, ME: Joseph Griffin.
Fritz, G.J. 1990. Multiple pathways to farming in precontact eastern North America. Journal of
World Prehistory 4:387–435.
Gagliano, S.M., C.E. Pearson, R.A. Weinstein, D.E. Wiseman, and C.M. McClendon. 1982.
Sedimentary studies of prehistoric archaeological sites: Criteria for the identification of
submerged archaeological sites of the northern Gulf of Mexico Continental Shelf. Coastal
Environments, Inc., Baton Rouge, LA.
Gagliano, S.M., C.E. Pearson, R.A. Weinstein, D.E. Wiseman, and C.M. McClendon. 1984.
Geoarchaeology of the northern Gulf Shore. In: D. Davis, ed. Perspectives on Gulf Coast
prehistory. Ripley P. Bullen Monographs in Anthropology and History, No. 5. University of
Florida Press and the Florida State Museum, Gainesville. Pp. 1–40.
Galiano, R. 2009. Scuba diving—New
http://njscuba.net/sites/index/html.
Jersey
and
Long
Island,
New
York.
Gallagher, W.B., D.C. Parris, B.S. Grandstaff, and C. DeTample. 1989. Quaternary mammals
from the Continental Shelf off New Jersey. The Mosasaur, The Journal of the Delaware
Valley Paleontological Society 4:101–110.
260
Gallup, C.D., R.L. Edwards, and R.G. Johnson. 1994. The timing of high sea levels over the past
200,000 years. Science 263(5148):796–800. DOI:10.1126/science.263.5148.796.
Gannon, M. 1996. First European contacts. In: M. Gannon, ed. The new history of Florida.
Gainesville: University Press of Florida. Pp. 16–39.
Gardner, A.H., comp. 1954. Wrecks around Nantucket since the settlement of the island, and the
incident connected therewith, embracing over seven hundred vessels, with additional
records to 1954. Reprinted. Originally published in 1877 as, A list of wrecks around
Nantucket. On file, Massachusetts Historical Society, Boston.
Gardner, W.M. 1974. The Flint Run complex: Pattern and process during the Paleo-Indian to
Early Archaic. In: W.M. Gardner, ed. The Flint Run Paleo-Indian complex: A preliminary
report, 1971–1973 seasons. Catholic University of America, Department of Anthropology
Occasional Paper No. 1. Pp. 5–47.
Gardner, W.M. 1977. The Flint Run Paleo-Indian complex and its implications for eastern North
American prehistory. Annals of the New York Academy of Sciences 288:257–263.
Gardner, W.M. 1980. Settlement-subsistence strategies in the Middle and South Atlantic portions
of the eastern United States during the late Pleistocene and Early Holocene. Paper presented
at the 1980 American Anthropological Association, Washington, DC.
Gardner, W.M. 1982. Early and Middle Woodland in the Middle Atlantic: An overview. In:
R.W. Moeller, ed. Practicing environmental archaeology: Methods and interpretations.
Occasional Paper No. 3, American Indian Archaeological Institute, Washington, D.C. Pp.
53–86.
Gardner, W.M. 1983. Stop me if you’ve heard this one before: The Flint Run PaleoIndian
complex revisited. Archaeology of Eastern North America 11:49–59.
Gardner, W.M. 1987. Comparison of Ridge and Valley, Blue Ridge, Piedmont, and Coastal Plain
Archaic period site distribution: An idealized transect. Journal of Middle Atlantic
Archaeology 3:49–80.
Gardner, W.M. 1989. An examination of cultural change in the late Pleistocene and Early
Holocene (circa 9200 to 6800 B.C.). In: J.M. Wittkofski and T.R. Reinhart, eds. Paleoindian
research in Virginia: A synthesis. Special Publication 19. Archaeological Society of
Virginia, Courtland. Pp. 5–51.
Garrison, E.G. n.d. Unpublished data. Department of Anthropology, University of Georgia,
Athens, GA.
Garrison, E.G. In preparation. Prehistoric artifacts from the southeastern U.S. Continental Shelf,
Gray’s Reef National Marine Sanctuary, Georgia.
261
Garrison, E.G., C.P. Giammona, F.J. Kelly, A.R. Tripp and G.A. Wolff. 1989. Historic
shipwrecks and magnetic anomalies of the northern Gulf of Mexico: Reevaluation of
archaeological resource management zone 1. 3 volumes. Prepared for the Gulf of Mexico
OCS Region, Minerals Management Service, U.S. Department of the Interior by the Texas
A&M Research Foundation, College Station, Texas. OCS Study MMS 89-0024 to 0025.
Garrison, E.G., G. McFall, and S. E. Noakes. 2008. Shallow marine margin sediments, modern
marine erosion and the fate of sequence boundaries–Georgia Bight (USA). Southeastern
Geology 45:127–142.
Garrison, E.G., S.E. Noakes, J.C. Hale and G.A. McFall. In press. Scour and sorting in late
Pleistocene shell beds, South Atlantic-Georgia Bight, U.S.A. PALAOIS.
Garrison, L.W. and R.L. McMaster. 1966. Sediments and geomorphology of the continental
shelf off southern New England. Marine Geology 4:273–289.
Gaudreau, D.C. 1988. The distribution of late Quaternary forest regions in the Northeast: Pollen
data, physiography, and the prehistoric record. In: G.P. Nicholas, ed. Holocene human
ecology in northeastern North America. New York: Plenum Press. Pp. 215–256.
Gayes, P.T. and H.J. Bokuniewicz. 1991. Estuarine paleoshorelines in Long Island Sound, New
York. In: Gayes, P.T., Lewis, R.S., and Bokuniewicz, H.J., eds., Quaternary geology of
Long Island Sound and adjacent coastal areas. Journal of Coastal Research, Special Issue
No. 11. Pp. 39–54.
Gayes, P.T., W.C. Schwab, N.W. Driscoll, R.A. Morton, W.E. Baldwin, J.F. Denny, O.Y.
German, J.Y. Park. 2002. Transgressive shoreface architecture within a sediment starved
arcuate strand: Long Bay, South Carolina. Paper presented at the Annual Meeting of the
American Geophysical Union, San Fransisco, CA.
Gayes, P.T., D.B. Scott, E.S. Collins, and D.D. Nelson. 1992. A late Holocene sea level
fluctuation in South Carolina. In: C.H. Fletcher III and J.F. Wehmiller, eds. Quaternary
coasts of the United States: Marine and lacustrine systems. SEPM Special Publication No.
48. Society for Sedimentary Geology, Tulsa. Pp. 155–160.
Gehrels, R., D. Belknap, and J. Kelley. 1996. Integrated high-precision analyses of Holocene
relative sea-level changes: Lessons from the coast of Maine. Geological Society of America
Bulletin 108(9):1073–1088.
Gentile, G. 1990. Shipwrecks of Delaware and Maryland. Philadelphia: Gary Gentile
Productions.
Gentile, G. 1992. Shipwrecks of Virginia. Philadelphia: Gary Gentile Productions.
Gentile, G. 2002. Shipwrecks of Delaware and Maryland. Philadelphia: Gary Gentile
Productions.
262
Gentile, G. 2003. Shipwrecks of South Carolina and Georgia. Philadelphia: Gary Gentile
Productions.
Gibbons, T., ed. 2001. The encyclopedia of ships. San Diego, CA: Thunder Bay Press.
Gibbs, C.V. 1963. Passenger liners of the western ocean: A record of Atlantic steam and motor
passenger vessels from 1838 to the present day. London: Putnam.
Gibson, J.L. 1994. Before their time? Early mounds in the lower Mississippi Valley.
Southeastern Archaeology 13:162–181.
Gibson, J.L. 2001. The ancient mounds of Poverty Point: Place of rings. Gainesville: University
Press of Florida.
Gingerich, J.A.M. 2007. Picking up the pieces: New Paleo-Indian research in the upper Delaware
Valley. Archaeology of Eastern North America 35:117–124.
Glynn, F. 1953. The Pilot’s Point submerged sites. Bulletin of the Archaeological Society of
Connecticut 27:11–29.
Goebel, T., M.R.Waters, and D.H. O’Rourke. 2008. The late Pleistocene dispersal of modern
humans in the Americas. Science 319:1497–1502.
Goff, J.A., D.J.P. Swift, C.S. Duncan, L.A. Mayer, and J. Hughes-Clarke. 1999. High-resolution
swath sonar investigation of sand ridge, dune and ribbon morphology in the offshore
environment of the New Jersey margin. Marine Geology 161:307–337.
Goff, J.A., J.A. Austin, Jr., S. Gulick, S. Nordfjord, B. Christensen, C. Sommerfield, H. Olson,
and C. Alexander. 2005. Recent and modern marine erosion on the New Jersey outer shelf.
Marine Geology 216:275–296.
Goodyear, A.C. 1979. A hypothesis for the use of cryptocrystalline raw materials among PaleoIndian groups of North America. South Carolina Institute of Archaeology and
Anthropology, Research Manuscript Series 156, Columbia.
Goodyear, A.C. 1982. The chronological position of the Dalton horizon in the southeastern
United States. American Antiquity 47:382–395.
Goodyear, A.C. 1999. The Early Holocene occupation of the southeastern United States: A
geoarcheological summary. In: R. Bonnichsen and K.L. Turnmire, eds. Ice Age people of
North America: Environments, origins, adaptations. Oregon State University Press for the
Center for the Study of the First Americans, Corvallis, OR. Pp. 432–481.
Goodyear, A.C. 2000. Topper site: Results of the 2000 Allendale Paleoindian expedition. Legacy
5(2):18–26.
Goodyear, A.C. 2005. Clovis in the Southeast Conference: Technology, time and space. October
26–29. Columbia Metropolitan Convention Center, Columbia.
263
Goodyear, A.C., J.H. House, and N.W. Ackerly. 1979. Laurens-Anderson: An archaeological
study of the inter-riverine Piedmont. Anthropological Studies 4. Institute of Archaeology
and Anthropology, University of South Carolina, Columbia.
Goodyear, A.C., J.L. Michie, and T. Charles. 1989. The earliest South Carolinians. In: A.C.
Goodyear and G.T. Hanson, eds. Studies in South Carolina archaeology: Essays in honor of
Robert L. Stephenson. Anthropological Studies 9: Occasional Papers of the South Carolina
Institute of Archaeology and Anthropology. University of South Carolina, Columbia. Pp.
19–52.
Goodyear, A.C. and K. Steffy. 2003. Evidence for a Clovis occupation at the Topper Site,
38AL23, Allendale County, South Carolina. Current Research in the Pleistocene 20:23–25.
Goodyear, A.C. III, S.B. Upchurch, M.J. Brooks, and N.N. Goodyear. 1983. Paleo-indian
manifestations in the Tampa Bay region, Florida. Florida Anthropologist 36(1): 40–66.
Gramly, M.R. 1982. The Vail site: A Paleo-indian encampment in Maine. Bulletin No. 30,
Buffalo Society of Natural Sciences, Buffalo.
Gray, E. 2003. Disasters of the deep: A comprehensive survey of submarine accidents and
disasters. Annapolis, MD: Naval Institute Press.
Gresham, T.H., W. D. Wood, C.O. Braley, and K.G. Wood. 1985. The Carmouche site:
archeology in Georgia’s western Fall Line Hills. Southeastern Archeological Services, Inc.,
Athens. Submitted to the U.S. Army Corps of Engineers.
Griffin, J.B. 1952. Culture periods in eastern United States prehistory. In: J.B. Griffin, ed.
Archaeology of the Eastern United States. Chicago: University of Chicago Press. Pp. 325–
364.
Grim, M.S., C.L. Drake, and J.R. Heirtzler. 1970 Sub-bottom study of Long Island Sound.
Geological Society of America Bulletin 81:649–666.
Grossman-Bailey, I. 2001. The people who lived by the ocean: Native American resource use
and settlement in the outer Coastal Plain of New Jersey. Ph.D. Dissertation, Temple
University, Philadelphia.
Gunn, J.D. and J. Foss. 1992. Copperhead Hollow (38CT58): Middle Holocene upland
conditions on the Piedmont-Coastal Plain margin. South Carolina Antiquities 24(1&2):1–17.
Hamilton, N.D. and S.G. Pollack. 1996. The Munsungan chert utilization and Paleoindians in
southwestern Maine. Current Research in the Pleistocene 13:117–119.
Hanley, R.J., M.A. Cinquino, S.L. Weber, A. Pickman, D.M. Cadzow, and R. Emans. 2002.
Phase II cultural resources investigation of the Sayre & Fisher Brick Company site (28
Mi163) for the Neptune Regional Transmission System project, Borough of Sayreville,
Middlesex County, New Jersey. Report prepared for Ecology and Environment, Inc.,
264
Lancaster, NY, by Panamerican Consultants, Inc., Buffalo, NY. Report on file at the New
Jersey Historic Preservation Office, Trenton.
Harding, J.L. and V.J. Henry. 1994. Geological history of Gray’s Reef National Marine
Sanctuary. Report to the National Oceanic and Atmospheric Administration. 10 Ocean
Science Circle. Savannah, GA.
Harper, D. 2003. Slavery in the North. www.slavenorth.com/slavenorth.htm.
Harris, M.S., P. T. Gayes, J. L. Kindinger, J. G. Flocks, D.E. Krantz, and P. Donovan. 2005.
Quaternary geomorphology and modern coastal development in response to an inherent
geologic framework: an example from Charleston, South Carolina. Journal of Coastal
Research 21(1):42–64
Harrison, R.S. and M. Coniglio. 1985. Origin of the Key Largo limestone, Florida Keys. Bulletin
of Canadian Petroleum Geology 33:350–58.
Harrison, W., Malloy, R.J., Rusnak, G.A., Terasmae, J. 1965. Possible late Pleistocene uplift
Chesapeake Bay entrance. The Journal of Geology 73(2):201–229.
Hartshorn, J.H., W.B. Thompson, W.F. Chapman, R.F. Black, G.M. Richmond, D.R. Grant, and
D.S. Fullerton. 1991. Quaternary geologic map of the Boston 4 x 6 degree quadrangle,
United States and Canada. U.S. Geological Survey Miscellaneous Series Map I-1420,
1:1,000,000.
Harwood, C.R. 1973. Quartzite points from the Appalachian Highlands. Tennessee
Archaeologist 15:89–95.
Haws, D. 1976. Ships and the sea: A chronological review. New York: Crescent Books.
Haynes, C.V. Jr. 1966. Elephant hunting in North America. Scientific American 214:104–112.
Haynes, C.V. Jr. 1969. The earliest Americans. Science 166:709–715.
Haynes, C.V. Jr. 1971. Time, environment and early man. Arctic Anthropology 8(2):3–14.
Haynes, C.V. Jr. 1992. Contributions of radiocarbon dating to the geochronology of the peopling
of the New World. In: R.E. Taylor, A. Long, and R.S. Kra, eds. Radiocarbon after four
decades. New York: Springer-Verlag. Pp. 355–374.
Haynes, G. 2007. A review of some attacks on the overkill hypothesis, with special attention to
misrepresentations and doubletalk. Quaternary International 170:84–94.
Hemmings, C.A. 1999. The Paleoindian and Early Archaic tools of Sloth Hole (8JE121): An
inundated site in the lower Aucilla River, Jefferson County, Florida. M.A. Thesis,
University of Florida, Gainesville.
265
Hemmings, C.A. 2004. The organic Clovis: A single continent-wide cultural adaptation. Ph.D.
Dissertation, University of Florida, Gainesville.
Henry, V.J., Jr. n.d. Unpublished data for the Georgia Continental Shelf. Data and manuscripts
on deposit at the Applied Coastal Research Laboratory, Georgia Southern University.
Savannah, Georgia.
Henry, V.J., R.T. Giles, and J.L. Harding. 1978. Geologic evaluation of potential pipeline
corridor sites along the Georgia coast: final report, phase I. Georgia Office of Planning and
Budget, Atlanta, GA. 101 p.
Henry, V.J. and F.M. Idris. 1992. Offshore minerals assessment studies on the Georgia
Continental Shelf – Phase 2: seismic stratigraphy of the TACTS area and evaluation of
selected sites for economic hard mineral potential. Georgia Department of Natural
Resources, Environmental Protection Division, Georgia Geologic Survey. Project Report
No. 18.
Herbster, H., M.L. Rainey, and D. Robinson. 2004. Archaeological reconnaissance and intensive
(locational) surveys Nantucket Cable terrestrial route Nantucket and Barnstable,
Massachusetts and marine archaeological reconnaissance Nantucket Cable underwater route.
Report prepared for National Grid USA, Northborough, MA. Report on file at PAL,
Pawtucket, RI.
Heron, D.S., Jr., T.F. Moslow, W.M. Berelson, J.R. Herbert, G.A. Steele III, and K.R. Susman.
1984. Holocene sedimentation of a wave-dominated barrier-island shoreline: Cape Lookout,
North Carolina. Marine Geology 60(1-4):413–434.
Huebner, G.G. 1922. Foreign trade of the United States since 1789. In: E.R. Johnson, T.W. Van
Metre, G.G. Huebner, and D.S. Hanchett, contrib. History of domestic and foreign
commerce of the United States. Vol. 2. Reprinted. Washington, D.C.: Columbia Planograph
Company. Originally published in 1915, Washington, D.C.: Carnegie Institution.
Heusser, L.E. 1985. Quaternary palynology of marine sediments in the northeast Pacific,
northwest Atlantic, and Gulf of Mexico. In: V.M. Bryant Jr., and R.G. Holloway, eds.
Pollen records of Late-Quaternary North American sediments. Dallas: American
Association of Stratigraphic Palynologists Foundation. Pp. 385–403.
Heusser, L.E. and N.J. Shackleton. 1979. Direct marine-continental correlation: 150,000 year
oxygen isotope-pollen record from the North Pacific. Science 204:837–839.
Hill, J. n.d. Unpublished data. Burroughs & Chapin Center for Marine and Wetland Studies,
Coastal Carolina University, Conway, SC.
Hine, A.C. 1997. Structural and paleoceaonographic evolution of the margins of the Florida
Platform. In: A.F. Randazzo and D.S. Jones, eds. The geology of Florida. Gainesville:
University Press of Florida. Pp. 169–194.
266
Hine, A.C., and S.W. Snyder. 1985. Coastal lithosome preservation: Evidence from the shoreface
and Inner Continental Shelf off Bogue Banks, North Carolina. Marine Geology 63(14):307–330.
Hobbs, C.H. III. 1990. Acoustic geology of a portion of Virginia’s innermost continental shelf.
In: C.R. Berquist Jr., ed. Heavy-mineral studies-VA Inner Continental Shelf. Virginia
Division of Mineral Resources Publication, Vol. 103:1–12.
Hobbs, C.H. III. 1997. Sediments and shallow stratigraphy of a portion of the Continental Shelf
of southeastern Virginia. Cooperative Agreement 14-35-0001-30740, Contract Report for
U.S. Department of the Interior, Mineral Management Service, Herndon, VA.
Hoffecker, J.F., W.R. Powers, and T. Goebel. 1993. The colonization of Beringia and the
peopling of the New World. Science 259:46–53.
Hollenbach, K.D. 2005. Gathering in the late Paleoindian and Early Archaic periods on the
Middle Tennessee River Valley, northwest Alabama. Ph.D. Dissertation. University of
North Carolina, Chapel Hill.
Hollenbach, K.D. 2009. Foraging in the Tennessee River Valley, 12,500 to 8,000 years ago.
Tuscaloosa: University of Alabama Press.
Holman, J.A. 1985. Ladd’s Quarry in Bartow County. National Geographic Research 1(3):423–
436.
Hopkins, D.M. 1973. Sea level history in Beringia during the past 250,000 years. Quaternary
Research 3:520–540.
Hopkins, T.S., and D.A. Dieterle. 1987. Analysis of baroclinic circulation of the New York Bight
with a 3-D diagnostic model. Continental Shelf Research 7:237–265.
Horton, B.P., W.R. Peltier, S.J. Culver, R. Drummond, S.E. Engelhart, A.C. Kemp, D.
Mallinson, E.R. Thieler, S.R. Riggs, D.V. Ames, and K.H. Thomson. 2009. Holocene sealevel changes along the North Carolina coastline and their implications for glacial isostatic
adjustment models. Quaternary Science Reviews 28:1725–1736.
House, J.H. and R.W. Wogaman. 1978. Windy Ridge, a prehistoric site in the interriverine
Piedmont in South Carolina. Anthropological Studies 3, South Carolina Institute of
Archaeology and Anthropology, University of South Carolina, Columbia.
Hoyt, J.H. 1967. Barrier island formation. Geological Society of America.Bulletin 78:112–1136.
Hoyt, W.H., J.C. Kraft, and M.J. Chrzastowski. 1990. Prospecting for submerged archaeological
sites on the Continental Shelf: southern Mid-Atlantic Bight of North America. In: N.P.
Lasca and J. Donahue, eds. Archaeological geology of North America. Geological Society
of America, Centennial Special Volume 4, Boulder, CO. Pp. 147–160.
267
Huddlestun, P.F. 1988. A revision of the lithostratigraphic units of the Coastal Plain of Georgia:
the Miocene through Holocene. Georgia Geologic Survey Bulletin 104, Georgia Department
of Natural Resources, Atlanta, GA.
Hughes, R. 1980. A cultural and environmental overview of the prehistory of Maryland’s lower
Eastern Shore based upon a study of selected artifact collections. Maryland Historical Trust
Manuscript Series Number 26. Prepared for the Maryland Historical Trust and the Tidewater
Administration, Maryland Department of Natural Resources, Coastal Resources Division,
Annapolis.
Humane Society of the Commonwealth of Massachusetts. 2009. Humane Society of the
Commonwealth of Massachusetts History. http://www.masslifesavingawards.com/history/.
Hunt, J.L. Jr. 1974. The geology and origin of Gray’s Reef, Georgia Continental Shelf. M.S.
Thesis, University of Georgia, Athens.
Hunt, R.E., D.J.P. Swift, and H. Palmer. 1977. Construction shelf topography, Diamond Shoals,
North Carolina. Geological Society of America Bulletin 88:299–311.
Hunter, L.E., R.D. Powell, and G.W. Smith. 1996. Facies architecture and grounding line
processes of morainal banks during the deglaciation of coastal Maine. Geological Society of
America Bulletin 108:1022–1038.
Idol, B. 2009. Archaeological excavations at the Island Creek site (31ST49) and site 31ST56,
Stanly County, North Carolina. TRC, Chapel Hill, NC. Report prepared for Progress
Energy, Inc., New Hill, North Carolina.
Isaac, R. 1982. The transformation of Virginia, 1740–1790. Chapel Hill: University of North
Carolina Press.
Isachsen, Y.W., E. Landing, J.M. Lauber, L.V. Rickard, and W.B. Rogers. 2000. Geology of
New York: A simplified account. Second edition. New York State Museum Educational
Leaflet 28. State University of New York, Albany.
Institute for Conservation Archaeology (ICA). 1979. Summary and analysis of cultural resource
information on the Continental Shelf from the Bay of Fundy to Cape Hatteras. Final Report
(Vols 1–4). Prepared by the Institute for Conservation Archaeology, Peabody Museum,
Harvard University, Cambridge, MA, for the Bureau of Land Management under contract
number AA 551-CT8-18.
Jacobson, G.L., T. Webb, and E.C. Grimm. 1987. Patterns and rates of vegetation change during
the deglaciation of eastern North America. In: W.F. Ruddiman and H.E. Wright, Jr., eds.
America during deglaciation: The geology of North America, Vol. K-3. Geological Society
of America, Boulder, CO. Pp. 277–288.
268
Jenkins, N.J. and C.B. Curren. 1976. Archaeological investigations on the central Tombigbee
River, Alabama: Chronology, subsistence and settlement patterns (a preliminary report).
Southeastern Archaeological Conference No. 32, Gainesville, Florida.
Jepsen, G.L. 1964. A New Jersey mastodon. Revised edition. New Jersey State Museum Bulletin
6, Trenton.
Jochim, M.A. 1981. Strategies for survival: Cultural behavior in an ecological context. New
York: Academic Press.
Johnson, E.R. 1922. American commerce to 1789. In: E.R. Johnson, T.W. Van Metre, G.G.
Huebner, and D.S. Hanchett, contrib. History of domestic and foreign commerce of the
United States. Vol. 1. Reprinted. Columbia Planograph Company, Washington, D.C.
Originally published in 1915, Washington, D.C.: Carnegie Institution.
Johnson, F. and H.M. Raup. 1947. Grassy Island: Archaeological and botanical investigations of
an Indian site in the Taunton River, Massachusetts. Papers of the Robert S. Peabody
Foundation for Archaeology, Volume 1, Number 2. Andover, MA.
Johnson, H.D. and C.T. Baldwin. 2004. Shallow clastic seas. In: H.G. Reading, ed. Sedimentary
environments: Processes, facies and stratigraphy, 3rd ed. Malden, MA: Blackwell Science.
Pp. 232–280.
Johnston, P.F. 1988. The end of the age of sail: Merchant shipping in the nineteenth century. In:
G.F. Bass, ed. Ships and shipwrecks of the Americas: A history based on underwater
archaeology. New York: Thames and Hudson. Pp. 231–250.
Jones, B.D. 1997. The late Paleoindian Hidden Creek site in southeastern Connecticut.
Archaeology of Eastern North America 25:45–80.
Jones, B.D. 1998. Human adaptation to the changing northeastern environment at the end of the
Pleistocene: Implications for the archaeological record. Ph.D. Dissertation, University of
Connecticut, Storrs.
Jones, B.D. and D.T. Forrest. 2003. Life in a postglacial landscape: Settlement-subsistence
change during the Pleistocene-Holocene transition in southern New England. In: D.L.
Cremeens and J.P. Hart, eds. Geoarchaeology of landscapes in the glacieated Northeast.
Albany: New York State Museum. Pp. 75–89.
Jones, C.C. 1873. Antiquities of the southern Indians, particularly of the Georgia tribes. New
York: Appleton.
Justice, N.D. 1987. Stone age spear and arrow points of the midcontinental and eastern United
States. Bloomington: Indiana University Press.
Keel, B.C. 1976. Cherokee archaeology: A study of the Appalachian Summit. Knoxville:
University of Tennessee Press.
269
Keith, D.H. 1988. Shipwrecks of the explorers. In: G.F. Bass, ed. Ships and shipwrecks of the
Americas: A history based on underwater archaeology. New York: Thames and Hudson. Pp.
45–68.
Kellam, J.A. and V.J. Henry. 1986. Interpretation of the seismic stratigraphy of the phosphatic
middle Miocene on the Georgia Continental Shelf. Georgia Department of Natural
Resources, Environmental Protection Division.
Kelley, A.R. 2006. Archaeological geology and postglacial development of the central Penobscot
River Valley, Maine, USA. Ph.D. Dissertation, University of Maine, Orono.
Kelley, J.T., W.A. Barnhardt, D.F. Belknap, S.M. Dickson, and A.R. Kelley. 1998. The seafloor
revealed: The geology of Maine’s Inner Continental Shelf. Report submitted to the Regional
Marine Research Program, Maine Geological Survey Open-file Report Maine Geological
Survey
Open-File
Report
98-6.
http://www.state.me.us/doc/nrimc/mgs/explore/marine/seafloor/contents.htm
Kelley, J.T., and Belknap, D.F., 1991, Physiography, Surficial Sediments and Quaternary
Stratigraphy of the Inner Continental Shelf and Nearshore Region of Central Maine, United
States of America. Continental Shelf Research V. 11: 1265-1283.
Kelley, J.T. and D.F. Belknap. 2003. Evaluation of sand and gravel resources of the Wells, ME
inner shelf embayment. Report submitted to the U.S. Minerals Management Service.
Kelley, J.T., D.F. Belknap, and S. Claesson. 2010. Drowned coastal deposits with associated
archaeological remains from a sea-level “slowstand”: northwestern Gulf of Maine, USA.
Geology 38(8):695–698.
Kelley, J.T., S.M. Dickson, D.F. Belknap, W.A. Barnhardt, and D.C. Barber. 2003. Distribution
and volume of sand bodies on the rocky, glaciated Inner Continental Shelf of the
northwestern Gulf of Maine. Journal of Coastal Research 19:41–56.
Kelley, J.T., S.M. Dickson, D.F. Belknap, and R. Stuckenrath. 1992. Sea-level change and the
introduction of late Quaternary sediment to the southern Maine Inner Continental Shelf. In:
J.F. Wehmiller and C.H. Fletcher, III, eds. Quaternary coasts of the United States. Society of
Economic Paleontologists and Mineralogists Special Paper 48. Pp. 23–34.
Kellogg, D. 1999. Personal communication.
Kelly, R.L. 1995. The foraging spectrum. Washington D.C.: Smithsonian Institution Press.
Kelly, R.L., and L.C. Todd. 1988. Coming into the country: Early PaleoIndian hunting and
mobility. American Antiquity 53:231–244.
Kemp, P., ed. 1980. Encyclopedia of ships and seafaring. New York: Crown Publishers, Inc.
Kendall, R.A., J.X. Mitrovica, G.A. Milne, T.E. Törnqvist, and Y. Le. 2008. The sea-level
fingerprint of the 8.2 ka climate event. Geology 36:423–426.
270
Kerwin, M., J.T. Overpeck, R.S. Webb, A. DeVernal, D.H. Rind, and R.J. Healy. 1999. The role
of oceanic forcing in mid-Holocene northern hemisphere climatic change.
Paleoceanography 14:200–210.
Kimball, J.R.H. 2005. Disaster, etc.: The maritime world of Marblehead, 1815–1865.
Portsmouth, NH: Peter E. Randall Publisher, LLC.
Kimball, S.M. and J.K. Dame. 1989. Geotechnical evaluation of sand resources on the Inner
Shelf of southern Virginia. Virginia Institute of Marine Science, Gloucester.
King, J. 2004. Personal communication.
Kinsey, W.F. III, ed. 1972. Archaeology in the upper Delaware Valley. Anthropological Series
No. 2, Pennsylvania Historical and Museum Commission, Harrisburg.
Klein, J.I., S.B. Marshall, J.L. Nolan, W.S. Newman, G.P. Watts, Jr., W.N. Still, M.B. Dierickx,
and R.B. Eidt. 1986. Sound Cable project cultural resources report. Final Report prepared
for the New York Power Authority by Ebasco Services, Inc., New York.
Klitgord, K.D., D.R. Hutchinson, and H. Schouten. 1988. U.S. Atlantic continental margin:
Structural and tectonic framework. In: R.E. Sheridan and J.A. Grow, eds. The Atlantic
continental margin of the U.S.: The geology of North America, Vol. 2. Geological Society
of America, Boulder, CO. Pp. 19–55.
Knebel, H.J., S.A. Wood, and E.C. Spiker. 1979, Hudson River: evidence for extensive
migration on the exposed continental shelf during Pleistocene time. Geology 7:254–258.
Kornfeld, M. 2007. Are Paleoindians of the Great Plains and Rockies subsistence specialists? In:
R.B. Walker and B.N. Driskell, eds. Foragers of the Terminal Pleistocene in North America.
Lincoln: University of Nebraska Press. Pp. 32–58.
Koski, S.H. 1989. The Archaeological Investigations of a Prehistoric Site in the Near-Shore
Zone off the West Coast of Venice, Florida: 8SO26. In: J.P. Delgado, ed. The 1988
Underwater Archaeology Proceedings from the Society for Historical Archaeology
Conference. Reno, NV: Society for Historical Archaeology. Pp. 25–28.
Kraft, H.C. 1973. The Plenge site: A Paleo-Indian occupation site in New Jersey. Archaeology of
Eastern North America 1(1):56–117.
Kraft, H.C. 1986. The Lenape: Archaeology, history, and ethnography. Collections of the New
Jersey Historical Society Volume 21. Newark, NJ.
Kraft, J.C. 1971. Sedimentary facies patterns and geologic history of a Holocene marine
transgression. Geological Society of America Bulletin 82:2131–2158.
Kraft, J.C. 1977. Late Quaternary paleogeographic changes in the coastal environment of
Delaware, Middle Atlantic Bight, related to archaeological settings. In: W.S. Newman and
271
B. Salwen, eds. Amerinds and their paleoenvironments in northeastern North America.
Annals of the New York Academy of Sciences 228. Pp. 35–69.
Kraft, J.C. 1985. Marine environments: Paleogeographic reconstructions in the littoral region. In:
J.K. Stein and W.R. Farrand, eds. Archaeological sediments in context. Peopling of the
Americas Edited Volume Series (Vol. 1), Center for the Study of Early Man, Institute for
Quaternary Studies, University of Maine at Orono. Pp. 111–125.
Kraft, J.C., D.F. Belknap, and J.M. Demarest. 1987. Prediction of effects of sea-level change
from paralic and inner shelf stratigraphic sequences. In: M.R. Rampino, J.E. Sanders, W.S.
Newman, and L.K. Konigsson, eds. Climate: History, periodicity, and predictability. New
York: Van Nostrand Reinhold. Pp. 166–192.
Kraft, J.C., D.F. Belknap, and I. Kayan. 1983. Potentials of discovery of human occupation sites
on the continental shelves and nearshore coastal zone. In: P.M. Masters and N.C. Flemming,
eds. Quaternary coastlines and marine archaeology: Towards the prehistory of land bridges
and continental shelves. London: Academic Press. Pp. 87–120.
Kuehn, S.R. 1998. New evidence for late PaleoIndian-Early Archaic subsistence behavior in the
western Great Lakes. American Antiquity 63(3):457–476.
Laing, A. 1961. American sail: A pictorial history. New York: E. P. Dutton & Company, Inc.
Lambeck, K. and J. Chappell. 2001. Sea level change through the last glacial cycle. Science
292:679–686.
LaPorta, P.C., L.E. Sohl, and M.C. Brewer. 1999. A geomorphological and archaeological
analysis of potential dredged material management alternative sites in the New York Harbor
Apex region. Report prepared by LaPorta and Associates, Warwick, NY for U.S. Army
Corps of Engineers, New York District.
László, V. and R. Woodman. 1999. The story of sail: Illustrated with 1,000 scale drawings.
Annapolis, MD: Naval Institute Press.
Lathrop, R.G., M. Cole, N. Senyk, and B. Butman. 2006. Seafloor habitat mapping of the New
York Bight incorporating sidescan sonar data. Estuarine, Coastal and Shelf Science 68:221–
230.
Leach, P.A., W.J. Chadwick, and D.F. Belknap. 2009. Multiple phases in the marine
transgression of coastal archaeological sites in Maine and Delaware. Paper presented at the
Annual Meeting of the Geological Society of America, Portland, Oregon.
Ledbetter, R.J., D.G. Anderson, L.D. O’Steen, and D.T. Elliott. 1996. Paleoindian and Early
Archaic research in Georgia. In: D.G. Anderson and K.E. Sassaman, eds. The Paleoindian
and Early Archaic Southeast. Tuscaloosa: University of Alabama Press. Pp. 270–287.
272
Lee, K.M. 2006. Late Quaternary sea-level lowstand environments and chronology of outer Saco
Bay, Maine. M.S. Thesis, University of Maine, Orono.
Leigh, D.S. 2008. Late Quaternary climates and river channels of the Atlantic coastal plain.
Geomorphology 101:90–108.
Leveillee, A., D.S. Robinson, and B. Ford. 2002. Terrestrial and marine archaeological
sensitivity survey along the proposed Narragansett Electric Cable route, Narragansett Bay
West Passage, Narragansett and Jamestown, Rhode Island. Report prepared for Vanasse
Hagen Brustlin, Inc., Providence, Rhode Island. Report on file at PAL, Pawtucket, RI.
Levine, M.A. 1990. Accomodating age: Radiocarbon results and fluted point sites in northeastern
North America. Archaeology of Eastern North America 18:33–63.
Lidz, B.H. and E.A. Shinn. 1991. Paleoshorelines, reefs, and a rising sea: South Florida, U.S.A.
Journal of Coastal Research 7(1):203–229.
Lighthouse Depot. n.d. Lighthouse history. http://www.lighthousedepot.com/lighthousehistory/a/111/.
Lighty, R.G., I.G. Macintyre, and R. Stuckenrath. 1978. Submerged early Holocene barrier reef,
southeast Florida shelf. Nature 276:59–60.
Littman, S. 2000. Pleistocene/Holocene sea level change in the Georgia Bight: A
paleonenvironmental reconstruction of Gray’s Reef National Marine Sanctuary and J-Reef.
M.S. Thesis, University of Georgia, Athens.
Liu J.P. and J. D. Milliman. 2004. Reconsidering melt-water pulses 1A and 1B: Global impacts
of rapid sea-level rise. Journal of Ocean University of China 3(2):183–190.
Lonsdale, A.L. and H.R. Kaplan. 1964. A guide to sunken ships in American waters. Arlington,
VA: Compass Publications, Inc.
Lovejoy, D.W. 1998. Classic exposures of the Anastasia Formation in Martin and Palm Beach
counties, Florida. Miami Geological Society Field Trip. Miami Geological Society, South
Miami, FL.
Lowery, D. 1989. The Paw Paw Cove Paleoindian site complex, Talbot County, Maryland.
Archaeology of Eastern North America 17:143–164.
Lowery, D. 1995. An archaeological survey of the Little Choptank River watershed: Dorchester
County, Maryland. Submitted to Maryland Historical Trust, Crownsville, MD.
Lowery, D. 2001. Archaeological survey of the Atlantic Coast shorelines associated with
Accomack County and Northampton County, Virginia. Survey and Planning Report Series
No. 6, Virginia Department of Historic Resources, Richmond, VA.
273
Lowery, D. 2003a. Archaeological survey of the Atlantic Coast shorelines associated with
Accomack County and Northampton County, Virginia. Survey and Planning Report Series
No. 7, Virginia Department of Historic Resources, Richmond, VA.
Lowery, D. 2003b. Investigating Clovis on the Delmarva Peninsula. Mammoth Trumpet 19(1).
Lowery, D. 2003c. Archaeological survey of the Atlantic coast shorelines associated with
Accomack County and Northampton County, Virginia. Survey and Planning Report Series
No. 7. Virginia Department of Historic Resources, Richmond, Virginia.
Lowery, D. 2007. The archaeology of sea level rise: Geoarchaeological and geomorphological
investigations along Messongo Creek in Accomack County, Virginia. Manuscript prepared
for the Virginia Department of Historic Resources, Richmond, VA.
Lowery, D. 2008. Archaeological survey of the Chesapeake Bay shorelines associated with
Mathews County, Virginia: An erosion threat study. Survey and Planning Report Series.
Virginia Department of Historic Resources, Richmond, VA.
Lowery, D. 2009a. The challenge of conducting prehistoric underwater archaeology. Notes on
Virginia 2009:52–55.
Lowery, D. 2009b. Geoarchaeological investigations at selected coastal archaeological sites
along the Delmarva Peninsula: the long-term interrelationship between climate, geology,
and culture. Unpublished Ph.D. Dissertation, Department of Geology. University of
Delaware, Newark, Delaware.
Lowery, D. and R. Martin. 2009. Archaeology of marine transgression: An inundated Middle
Archaic burial in Chesapeake Bay, Maryland. Archaeology of Eastern North America
37:159–173.
Lowery, D. and T. Phillips. 1994. The Meekins Neck Paleoindian site complex, Dorchester
County, Maryland: A development of a Paleoindian settlement model for the Delmarva
Peninsula. Maryland Archeology 30(2):29–36.
Lynch, D. 2001. Submerged cultural inventory of Rhode Island: A method for recording
submerged archaeological sites. Paper presented at the 41st Annual Meeting of the
Northeastern Anthropological Association, Hartford, CT.
Maginnis, A.J. 1892. The Atlantic ferry: Its ships, men, and working. London: Whitaker and
Company.
Mallinson, D., K. Burdette, S. Mahan, and G. Brook. 2008. Optically stimulated luminescence
age controls on late Pleistocene and Holocene coastal lLithosomes, North Carolina, USA.
Quaternary Research 69:97–108.
Mallinson, D., S. Riggs, E. R. Thieler, S. Culver, K. Farrell, D. S. Foster, D. R. Corbett, B. P.
Horton, and J. Wehmiller. 2005. Late Neogene and Quaternary evolution of the northern
274
Albemarle embayment (Mid-Atlantic continental margin, USA). Marine Geology 217:97–
117.
Markewich, H.W., C.M. Hacke, and P.F. Huddlestun. 1992. Emergent Pliocene and Pleistocene
sediments of southeastern Georgia: An anomalous, fossil-poor, clastic section. In: C.H.
Fletcher III and J.F. Wehmiller, eds. Quaternary coasts of the United States: Marine and
estuarine systems. SEPM Special Publication No. 48. Society for Sedimentary Geology,
Tulsa. Pp. 173–189.
Marks, B.S. 2006. Site formation processes and activity areas at Ontolo (8JE1577): A submerged
Prehistoric human occupation site in Apalachee Bay, Florida. Ph.D. Dissertation, Florida
State University, Tallahassee.
Marks, B.S. and M.K. Faught. 2003. Ontolo (8JE1577): Another early prehistoric site submerged
on the Continental Shelf of NW Florida. Current Research in the Pleistocene 20:49–51.
Marquardt, W.H. 1985. Complexity and scale in the study of fisher-gatherer-hunters: An
example from the eastern United States. In: T.D. Price and J.A. Brown, eds. Prehistoric
hunter-gatherers: The emergence of cultural complexity. Orlando: Academic Press. Pp. 59–
98
Marquardt, W.H. and P.J. Watson. 1983. The Shell Mound Archaic of western Kentucky. In: J.L.
Phillips and J.A. Brown, eds. Archaic hunters and gatherers in the American Midwest. New
York: Academic Press. Pp. 323–339.
Marrinan, R.A. 1975. Ceramics, molluscs, and sedentism: The Late Archaic period on the
Georgia coast. Ph.D. dissertation, University of Florida, Gainesville.
Marshall, S. 1982. Aboriginal settlement in New Jersey during the Paleo-Indian cultural period:
circa 10,000 B.C. – 6000 B.C. In: O. Chesler, ed. New Jersey’s archeological resources: A
review of research problems and survey priorities. Trenton: Office of New Jersey Heritage.
Pp. 10–51.
Martin, L.D., R.A. Rogers, and A.M. Neuner. 1985. The effect of the end of the Pleistocene on
man in North America. In: J.I. Mead and D.J. Meltzer, eds. Environments and extinctions:
Man in late glacial North America. Center for the Study of Early Man, Orono, Maine. Pp.
15–30.
Martin, P.S. 1973. The discovery of America. Science 179:969–974.
Marx, R.F. 1981. Shipwrecks of the western hemisphere, 1492–1825. New York: The World
Publishing Company.
Marx, R.F. 1987. Shipwrecks in the Americas. Mineola, NY: Dover.
Mason, R.J. 1962. The Paleoindian tradition in eastern North America. Current Anthropology
3:227–283.
275
Massachusetts State Historic Preservation Officer. 2009. Opinion: Eligibility for National
Register of Historic Places Nantucket Sound Wampanoag Traditional Cultural Property,
November 5, 2009. Letter on file at Massachusetts Historical Commission, Boston.
Matson, C. 1998. Merchants and empire: Trading in colonial New York. Baltimore: The Johns
Hopkins Press.
Matteucci, T.D. and A.C. Hine. 1987. Evolution of the Cape Fear terrace: A complex interaction
between the Gulf Stream and a paleo-shelf edge delta. Marine Geology 77(3-4):185–205.
Mayewski, P.A. and M. Bender. 1995. The GISP2 ice core record: Paleoclimate highlights.
Reviews of Geophysics (Supplement) 33:1287–1296.
Maymon, J. and C. Bolian. 1992. The Wadleigh Falls site: An Early and Middle Archaic period
site in southeastern New Hampshire. Occasional Publications in Maine Archaeology.
9:117–134.
Maymon, J.H., J.B. Pelletier, R. Vidutis, M. Williams, P. Godwin, W.P. Giglio, S. Milstead, L.
Post, B. Stone, K. Bastis, K. Grandine, and C.R. Polglase. 2000. Interim report on cultural
resource survey for the proposed Eastchester Marine Pipeline, Suffolk and Bronx Counties,
New York. Prepared for ENSR Northborough, MA, by R. Christopher Goodwin &
Associates, Inc., Frederick, MD. Report on file at the New York State Historic Preservation
Office, Peebles Island Resource Center, Cohoes, NY.
McAvoy, J.M. and L.D. McAvoy. 1997. Archaeological investigations of site 44SX202, Cactus
Hill, Sussex County, Virginia. Virginia Department of Historic Resources Research Report
Series No. 8, Richmond.
McBride, K.A. 1984. Prehistory of the lower Connecticut River Valley. Ph.D. Dissertation,
University of Connecticut, Storrs.
McCary, B.C. 1947. A survey and study of Folsom-like points found in Virginia. Quarterly
Bulletin of the Archaeological Society of Virginia 2(1):1–131.
McCary, B.C. 1948. A report on the Folsom-like points found in Granville County, North
Carolina. Archeological Society of Virginia Quarterly Bulletin 3(1).
McDonald, J.M. 2000. An outline of the pre-Clovis archaeology of SV-2, Saltville, Virginia,
with special attention to a bone tool dated 14,510 yr B.P. Jeffersoniana 9:1–59.
Contributions from the Virginia Museum of Natural History, Martinsville.
McGhee, R. 1984. Contact between native North Americans and the Medieval Norse: A review
of the evidence. American Antiquity 49(1):4–26.
McHugh, C.M., C.A. Hartin, G.S. Mountain, and H.M. Gould. 2010. The role of glacio-eustasy
in sequence formation: Mid-Atlantic continental margin, USA. Marine Geology 277:31–47.
276
McLearen, D.C. 1991. Late Archaic and Early Woodland material culture in Virginia. In: T.R.
Reinhart and M.E.N. Hodges, eds. Late Archaic and Early Woodland research in Virginia:
A synthesis. Special Publication No. 23 of the Archeological Society of Virginia,
Richmond. Pp. 89–137.
McNett, C. 1985. Shawnee Minisink: A stratified Paleoindian-Archaic site in the upper Delaware
Valley of Pennsylvania. Orlando: Academic Press.
McWeeney, L.J. 1994. Archaeological settlement patterns and vegetation dynamics in southern
New England in the late Quaternary. Ph.D. Dissertation, Yale University.
McWeeney, L.J. 1999. A review of late Pleistocene and Holocene climate changes in southern
New England. The Bulletin of the Archaeological Society of Connecticut 62:3–18.
McWeeney, L.J. 2002. Archaeological charcoal used for environmental reconstruction: The
Templeton Paleoindian site, Connecticut, USA. In: S.L.R. Mason and J.G. Hather, eds. The
archaeobotany of temperate zone hunter-gatherers. London: Institute of Archaeology
Publications. Pp. 117–127.
McWeeny, L. and D.C. Kellogg. 2001. Early and middle Holocene climate changes and
settlement patterns along the eastern coast of North America. Archaeology of Eastern North
America 29:187–212.
Meisburger, E.P. 1972. Geomorphology and sediments of the Chesapeake Bay entrance.
Technical Memorandum, TM 38. U.S. Army Engineer Waterways Experiment Station,
Coastal Engineering Research Center, Vicksburg, MS.
Meisburger, E.P. and M.E. Field. 1975. Geomorphology, shallow structure, and sediments of the
Florida inner continental shelf, Cape Canaveral to Georgia. U.S. Army Corps of Engineers
Coastal Engineering Research Center Technical Memorandum No. 54.
Meltzer, D.J. 1988. Late Pleistocene human adaptations in eastern North America. Journal of
World Prehistory 2:1–52.
Meltzer, D.J. 1993. Pleistocene peopling of the Americas. Evolutionary Anthropology 1:157–
169.
Meltzer, D.J. 2004. Peopling of North America. In: A.R. Gillespie, S.C. Porter, and B.F.
Atwater, eds. The Quaternary period in the United States, Vol. 1. New York: Elsevier
Science. Pp. 539–563.
Meltzer, D., D. Grayson, G. Ardila, A. Barker, D. Dincauze, C.V. Haynes, Jr., F. Mena, L.
Nunez, and D. Stanford. 1997. On the Pleistocene antiquity of Monte Verde, southern Chile.
American Antiquity 62:659–663.
Meltzer, D.J. and V.T. Holliday. 2010. Would North American Paleoindians have noticed
Younger Dryas age climate changes? Journal of World Prehistory 23:1–41.
277
Meltzer, D.J. and B.D. Smith. 1986. Paleo-Indian and Early Archaic subsistence strategies in
eastern North America. In: S.W. Neusius, ed. Foraging, collecting, and harvesting: Archaic
period subsistence and settlement in the Eastern Woodlands. Occasional Paper No. 6, Center
for Archeological Investigations, Southern Illinois University, Carbondale. Pp. 1–30.
Merwin, D.E. 2003. The potential for submerged prehistoric archaeological sites off Sandy
Hook. Bulletin of the Archaeological Society of New Jersey 57:1–10.
Merwin, D.E. 2010. Submerged evidence of early human occupation in the New York Bight.
Ph.D. dissertation, Anthropological Sciences, Stony Brook University.
Merwin, D.E., D.P. Lynch, and D.S. Robinson. 2003. Submerged prehistoric sites in southern
New England: Past research and future directions. Bulletin of the Archaeological Society of
Connecticut 65:41–56.
Michels, J.W. and I. Smith. 1967. Archaeological investigations of Sheep Rock Shelter,
Huntingdon County, Pennsylvania. 3 vols. Pennsylvania State University, University Park.
Michie, J.L. 1966. The Taylor point. The Chesopiean 3(5):123–124.
Milanich, J.T. 1994. Archaeology of precolumbian Florida. Gainesville: University Press of
Florida.
Milanich, J.T. and C.H. Fairbanks. 1980. Florida archaeology. New York: Academic Press.
Miller, J.J. 1998. An environmental history of northeast Florida. Gainesville: University Press of
Florida.
Miller K.G., P.J. Sugarman, J.V. Browning, B.P. Horton, A. Stanley, A. Kahn, J. Uptegrove, and
M. Aucott. 2009. Sea-level rise in New Jersey over the past 5000 years: implications to
anthropogenic changes. Global and Planetary Change 66:10–18.
Milliman, J.D. and K.O. Emery. 1968. Sea levels during the past 35,000 years. Science
162:1121–1123.
Milliman, J.D., O.H. Pilkey and D.A. Ross. 1972. Sediments of the continental margin off the
eastern United States. Geology Society of America Bulletin 83:1315–1334.
Milliman J.D., Z. Jiezao, L. Anchun, and J.I. Ewing. 1990. Late Quaternary sedimentation on the
outer and middle New Jersey continental shelf: result of two local deglaciations? Journal of
Geology 98:966–976.
Moeller, R.W. 1984. Regional implications of the Templeton site for Paleo-Indian lithic
procurement and utilization. North American Archaeologist 5(3):235–245.
Moir, R. 1979. A summary and analysis of cultural resource information on the continental shelf
from the Bay of Fundy to Cape Hatteras, Vol. I—Physical environment. Institute for
Conservation Archaeology, Peabody Museum, Boston.
278
Montaggioni, L.F., C.J.R. Braithwaite. 2009. Quaternary coral reef systems: history,
development processes and controlling factors. Developments in Marine Geology 5:419–
420.
Moorehead, W.K. 1922. A report on the archaeology of Maine. Andover, MA: The Andover
Press.
Morgan, K. 1989. Shipping patterns and the Atlantic trade of Bristol, 1749–1770. The William
and Mary Quarterly. 3rd Series, 46(3):506–538.
Morse, D.F., D.G. Anderson, and A.C. Goodyear. 1996. The Pleistocene-Holocene transition in
the eastern United States. In: L.G. Strauss, B.V. Eriksen, J. Erlandson, and D.R. Yesner, eds.
Humans at the end of the ice age: The archaeology of the Pleistocene-Holocene transition.
New York: Plenum Press. Pp. 319–338.
Morse, D.F. and P.A. Morse. 1983. The archaeology of the central Mississippi Valley. New
York: Academic Press.
Mounier, R.A. 2003. Looking beneath the surface: The story of archaeology in New Jersey.
Piscataway, NJ: Rutgers University Press.
Murphy, L.E. 1990. 8SL17: Natural site-formation processes of a multiple-component
underwater site in Florida. Southwest Cultural Resources Center Professional Papers No. 39,
National Park Service, Santa Fe, NM.
Mutler, H.G. and J.E. Hoffmeister. 1968. Subaerial laminated crusts of the Florida Keys.
Geological Society of America, Bulletin No. 79:183–192.
Mystic Seaport. 1999. Conducting vessel research. G. W. Blunt White Library Bulletin 67-4.
Mystic Seaport Museum of America and the Sea, Mystic, Connecticut.
National Oceanographic and Atmospheric Administration (NOAA). 2009. NOAA locates U.S.
Navy ship sunk in World War II battle. Press release, September 9.
http://www.noaanews.noaa.gov/stories2009/20090909_battleofatlantic.html.
NOAA. 2010. NOAA National Geophysical Data Center, U.S. Coastal Relief Model,
http://www.ngdc.noaa.gov/mgg/coastal/coastal.html.
Neill, W.T. 1958. A stratified early site at Silver Springs, Florida. The Florida Anthropologist
11:33–52.
Neill, W.T. 1964. The association of Suwannee points and extinct animals. The Florida
Anthropologist 17:17–32.
New World Encyclopedia. 2008. Leif Ericson.
http://www.newworldencyclopedia.org/entry/Leif_Ericson?oldid=681083.
279
New York Shipping and Commercial List. 1815–1830. Marine List. On file, Library of
Congress, Washington, D.C.
New York State Canal Corporation. n.d. The Erie Canal: A brief history. Albany, NY: New York
State Canal Corporation. Published online, http://www.canals.ny.gov/cculture/history/eriecanal-history.pdf.
Newman, W.S., Thurber, D.H., Zeiss, H.S., Rokach, A., Musich, L. 1969. Late Quaternary
geology of the Hudson River estuary: a preliminary report. N.Y. Acad. Sci. Trans. 31, 548–
570.
Newby, P., B. Shuman, P. Leduc, and T. Webb III. 2001. Integrating scales and data sets:
Environment and culture in northeastern North America 14,000 to 10,000 CAL YR B.P.
Paper presented at the Society for American Archaeology 66th Annual Meeting, New
Orleans, Louisiana.
Nicholas, G.P. 1987. Rethinking the Early Archaic. Archaeology of Eastern North America
15:99–123.
Nicholas, G.P. 1998. Wetlands and Hunter-Gatherer Land Use in North America. In Hidden
Dimensions: The Cultural Significance of Wetland Archaeology, edited by Kathryn.
Bernick, pp. 31-46. UBC Press, Toronto.
Niedoroda, A.W., D. J.P. Swift, A.G. Figueiredo, and G.L. Freeland. 1984. Shoreface
morphodynamics on wave-dominated coasts. Marine Geology 60:331–354.
Nikitina, D., J. Pizzuto, R. Schwimmer, and K. Ramsey. 2000. An updated Holocene sea level
curve for the Delaware coast. Marine Geology 171:7–20.
Noble, D.L. 1988. A legacy: The United States Life-Saving Service. U.S. Coast Guard,
Washington, D.C.
Nydal, R. 1989. A critical review of radiocarbon dating of a Norse settlement at L'Anse aux
Meadows, Newfoundland, Canada. Radiocarbon 31:976–985.
O’Shea, S. 2000. Personal communication.
Oertel, G.F. and A.M. Foyle. 1995. Drainage displacement by sea-level fluctuation at the outer
margin of the Chesapeake seaway. Journal of Coastal Research 11(3):583–604.
Ogden, J.G. 1977. The late Quaternary paleoenvironmental record of northeastern North
America. Annals of the New York Academy of Sciences 288:16–34.
Oldale, R.N. 1985a. A drowned Holocene barrier spit off Cape Ann, Massachusetts. Geology
13:375–377.
Oldale, R.N. 1985b. Late Quaternary sea-level history of New England: A review of the
published sea-level data. Northeastern Geology 7:192–200.
280
Oldale, R.N. 1985c. Rapid postglacial shoreline changes in the western Gulf of Maine and the
PaleoIndian environment. American Antiquity (50)1:145–150.
Oldale, R.N. 1986. Late-glacial and postglacial sea-level history of New England: A review of
available sea-level curves. Archaeology of Eastern North America 14:89–99.
Oldale, R.N. 1992. Cape Cod and the Islands: The geologic story. East Orleans, MA: Parnassus
Imprints.
Oldale, R.N. 2001a. Cape Cod, Martha’s Vinyard and Nantucket, the Geologic Story. Yarmouth
Port, MA: On Cape Publications.
Oldale, R.N. 2001b. The geologic history of Cape Cod, Massachusetts. USGS, Woods Hole
Field Center, Massachusetts. http://pubs.usgs.gov/gip/capecod/index.html.
Oldale, R.N., S.M. Colman, and G.A. Jones. 1993. Radiocarbon ages from two submerged
strandline features in the western Gulf of Maine and a sea level curve for the northeastern
Massachusetts coastal region. Quaternary Research 40:38–45.
Oldale, R.N., L.E. Wommack, and A.B. Whitney. 1983. Evidence for a postglacial low relative
sea-level stand in the drowned delta of the Merrimack River, western Gulf of Maine.
Quaternary Research 19:325–336.
Oliver, B.L. 1985. Tradition and typology: Basic elements of the Carolina projectile point
sequence. In: R.S. Dickens Jr. and H.T. Ward, eds. Structure and process in southeastern
archaeology. Tuscaloosa: University of Alabama Press. Pp. 195–211.
Osmond, J.K., J.P. May, and W.F. Tanner. 1970. Age of the Cape Kennedy barrier and lagoon
complex. Journal of Geophysial Research. (75):469–479.
O’Steen, L.D., R.J. Ledbetter, D.T. Elliott, and W.W. Barker. 1986. Paleo-Indian sites of the
inner Piedmont of Georgia: Observations of Settlement in the Oconee Watershed. Early
Georgia 14(1, 2):1–63.
Otto, B.A. 1999. The strange emergence of a deep sea plummet off Plymouth’s Gurnet Head.
Bulletin of the Massachusetts Archaeological Society 60(1):35–36.
Otto-Bliesner, B.L., E.C. Brady, G.Clauzet, R. Tomas, S. Levis, and Z. Kothavala. 2005. Last
Glacial Maximum and Holocene Climate in CCSM3. Journal of Climate 19:2526–2544.
Otvos, E.G. 2004. Holocene gulf levels: Recognition issues and updated sea level curve. Journal
of Coastal Research 60:680–699.
Overpeck, J.T., R.S. Webb and T. Webb, III. 1992. Mapping eastern North American vegetation
change of the past 18 ka: No-analogs and the future. Geology 20:10710–1074.
PAL (Public Archaeology Laboratory, Inc.). 2002. Islander East Pipeline Project, offshore
cultural resources anchor spread survey: Connecticut and New York, supplemental report to
281
June 2001. Resource Report 4 Filing. Report prepared for the Islander East Pipeline
Company, Branford, CT. On file at PAL, Pawtucket, RI.
PAL (Public Archaeology Laboratory, Inc.). 2003a. Everett Extension Project Resource Report
4 - cultural resources. Report prepared for Duke Energy- Algonquin Gas Transmission,
LLC, Boston, MA. Report on file at PAL, Pawtucket, RI.
PAL (Public Archaeology Laboratory, Inc.). 2003b. Island End River marine archaeological
reconnaissance survey, Everett and Chelsea, Massachusetts. Summary Memorandum
prepared for Epsilon Associates, Inc., Lawrence, MA. Report on file at PAL, Pawtucket, RI.
PAL (Public Archaeology Laboratory, Inc.). 2004. Phase I marine archaeological survey of
RIDOT nearshore reefs - Sheep Point and Gooseberry Island areas, Newport, Rhode Island.
Summary Memorandum prepared for the Rhode Island Department of Transportation,
Providence, Rhode Island. Report on file at PAL, Pawtucket, RI
PAL (Public Archaeology Laboratory, Inc.). 2005a. Appendix 4F: Northeast Gateway Pipeline
Lateral marine archaeological reconnaissance survey. Report prepared for Duke EnergyAlgonquin Gas Transmission, LLC, Waltham, MA. Report on file at PAL, Pawtucket, RI.
PAL (Public Archaeology Laboratory, Inc.). 2005b. Northeast Gateway Energy Bridge, LLC
deepwater port application, Section 4.1.1 cultural resources. Report prepared for Excelerate
Energy, LLC, The Woodlands, TX. Report on file at PAL, Pawtucket, RI.
PAL (Public Archaeology Laboratory, Inc.). 2005c. Summary report (PAL No. 1710) Neptune
Regional Transmission Project, New York and New Jersey, supplemental underwater
archaeological investigation. Prepared for Neptune Regional Electric Transmission System,
LLC, Pittsfield, ME. Report on file at the New Jersey Historic Preservation Office, Trenton,
NJ.
Paleoindian Database of the Americas (PIDBA). 2009. PIDBA: Florida. University of Tennessee
Department of Anthropology, electronic document, http://pidba.org/flimages.html, accessed
January 2010.
Parmalee, P.W. and W.E. Klippel. 1974. Freshwater mussels as a prehistoric food source.
American Antiquity 39(3):421–434.
Parris, D.C. 1983. New and revised records of Pleistocene mammals of New Jersey. The
Mosasaur, The Journal of the Delaware Valley Paleontological Society I:1–21.
Parry, W.J. and R.L. Kelly. 1987. Expedient core technology and sedentism. In: J.K. Johnson
and C.A. Morrow, eds. The organization of core technology. Boulder, CO: Westview
PressPp. 285–304.
Pearson, C.E., S.R. James, Jr., M.C. Krivor, S.D. El Darragi, L. Cunningham. 2003. Refining and
revising the Gulf of Mexico Outer Continental Shelf region high-probability model for
shipwrecks: final report. Vol. II: Technical narrative. Panamerican Consultants, Inc.,
282
Memphis, TN and Coastal Environments, Inc., Baton Rouge, LA. Submitted to Minerals
Management Service, U.S. Department of the Interior.
Pearson, C.E., D.B. Kelley, R.A. Weinstein and S.M. Gagliano. 1986. Archaeological
investigations on the Outer Continental Shelf: A study within the Sabine River Valley,
offshore Louisiana and Texas. Coastal Environments, Inc. Prepared by Coastal
Environments Inc. Published by Minerals Management Service, US Department of Interior,
Reston, VA.
Pearson, C.E., R.A. Weinstein, and D.B. Kelley. 2008. Prehistoric site discovery on the Outer
Continental Shelf, United States of America. Paper presented at the 6th World
Archaeological Conference, Dublin, Ireland, June 28–July 4.
Peet, C.E. 1904. Glacial and postglacial history of the Hudson and Champlain valleys. Journal of
Geology 12:415–469, 617–660.
Pelletier, B.G. and B.S. Robinson. 2005. Tundra, ice, and a Pleistocene cape on the Gulf of
Maine: A case of Paleoindian transhumance. Archaeology of Eastern North America
33:163–176.
Peltier, W.R. 2002. On eustatic sea level history: Last glacial maximum to Holocene. Quaternary
Science Reviews 21:377–396.
Peltier, W.R. 2005. On the hemispheric origins of meltwater pulse 1a. Quaternary Science
Reviews 24:1655–1671.
Pepe, J. 2000. An archaeological survey of St. Lucie County, Florida. Report on file, Florida
Division of Historical Resources, Tallahassee.
Perkins, R.D. 1977. Depositional framework of Pleistocene rocks in south Florida. In: P. Enos
and R.D. Perkins, eds. Quaternary sedimentation in south Florida. Geological Society of
America Memoir no. 147. Pp. 131–98.
Perkinson, P. 1971. North Carolina fluted points: Survey report number one. Southern Indian
Studies 23:3–40.
Perkinson, P. 1973. North Carolina fluted points: Survey report number two. Southern Indian
Studies 25:3–60.
Perlman, S.M. 1980. An optimum diet model, coastal variability, and hunter-gatherer behavior.
In: M.J. Schiffer, ed. Advances in archaeological method and theory, Vol. 3. New York:
Academic Press. Pp. 257–310.
Phelps, D.S. 1983. Archaeology of the North Carolina Coast and Coastal Plain: Problems and
hypotheses. In: M.A. Mathis and J.J. Crow, eds. The prehistory of North Carolina: An
archeological symposium. North Carolina Division of Archives and History, Raleigh. Pp. 1–
51.
283
Pierce, J.W. and D.J. Colquhoun. 1970. Holocene evolution of a portion of the North Carolina
coast. Geological Society of America Bulletin 81:3697–3714.
Pierson, L.J., G.I. Shiller, and R.A. Slater. 1987. Archaeological resource study: Morro Bay to
the Mexican Border. Prepared for the Pacific OCS Region, Minerals Management Service,
U.S. Department of the Interior by PS Associates, Cardiff, California. OCS Study MMS 870025.
Pilkey, O.H., B.W. Blackwelder, H.J. Knebel, and M.W. Mayers. 1981. The Georgia
Embayment continental shelf: stratigraphy of a submergence. Geological Society of
America Bulletin 92:52–63
Pirazzoli, P.A. 1991. World atlas of Holocene sea level changes. Amsterdam: Elsevier
Oceanography Series.
Poag, C.W., and W.D. Sevon. 1989. A record of Appalachian denudation in postrift Mesozoic
and Cenozoic sedimentary deposits of the U.S. middle Atlantic continental margin.
Geomorphology 2:119–157.
Powell, B.W. 1965. Spruce Swamp: A partially drowned coastal shell midden in Connecticut.
American Antiquity 30:460–469.
Pratson, L.F., W.B.F. Ryan, G.S. Mountain, and D.C. Twichell. 1994. Submarine canyon
initiation by downslope-eroding sediment flows: evidence in Late Cenozoic strata on the
New Jersey continental slope. Geological Society of America Bulletin 106:395–412.
Pratson, L.F., and W. Haxby. 1996. What is the slope of the US continental slope? Geology
24:3–6.
Preble, G.H. 1883. A chronological history of the origin and development of steam navigation,
1543–1882. Philadelphia: L. R. Hamersly & Company.
Preston, A. and R.Natkiel. 1986. Atlas of maritime history. New York: Facts on File, Inc.
Price, F.H. and A.E. Spiess. 2007. A submerged prehistoric sSite and other fishermen’s reports
near Mount Desert Island. The Maine Archeological Society Bulletin 47:21–35.
Procter Brothers. 1882. The Fishermen’s own book: Comprising the list of men and vessels lost
from the port of Gloucester, Mass., from 1874 to April 1, 1882, and a table of losses from
1830, together with valuable statistics of the fisheries, also notable fares, narrow escapes,
startling adventures, fishermen’s off-hand sketches, ballads, descriptions of fishing trips and
other interesting facts and incidents connected with this branch of maritime industry.
Gloucester, MA: Procter Brothers.
Procter, G.H. 1873. The fisherman’s memorial and record book, containing a list of vessels and
their crews lost from the Port of Gloucester from the year 1830 to October 1, 1873.
Gloucester, MA: Procter Brothers.
284
Purdy, B.A. 1981. Florida’s prehistoric stone technology. Gainesville: University Press of
Florida.
Purdy, B.A. 2008. Florida’s people during the last ice age. Gainesville: University Press of
Florida.
Purrington, B.L. 1983. Ancient mountaineers: An overview of the prehistoric archaeology of
North Carolina’s western mountain region. In: M.A. Mathis and J.J. Crow, eds. The
prehistory of North Carolina: An archaeological symposium. North Carolina Department of
Cultural Resources, Division of Archives and History, Raleigh. Pp. 83–160.
Queen Anne’s Revenge. n.d. Queen Anne’s Revenge History. Maritime Research Institute and
North Carolina Underwater Archaeology Branch.
http://www.qaronline.org/history/historyintro.htm.
Quinn, W.P. 1979. Shipwrecks around New England: A chronology of marine accidents and
disasters from Grand Manan to Sandy Hook. Lower Cape Publishing Company, Orleans,
MA.
Rampino, M.R. and J.E. Sanders. 1980. Holocene transgression in south-central Long Island,
New York. Journal of Sedimentary Petrology 50:1063–1080.
Ramsey, K.W. and S.J. Baxter. 1996. Radiocarbon dates from Delaware: A compilation.
Delaware Geologic Survey Report 54, Newark, DE.
Randall, A.R. and K.E. Sassaman. 2005. St. Johns archaeological field school 2003–2004:
Hontoon Island State Park. Technical Report 6, Laboratory of Southeastern Archaeology,
Department of Anthropology, University of Florida.
Randazzo, A.F. and R.B. Halley. 1997. Geology of the Florida Keys. In: A.F. Randazzo and D.S.
Jones, eds. The geology of Florida. Gainesville: University Press of Florida. Pp. 251–259.
Rattray, J.E. 1973. Perils of the port of New York: Maritime disasters from Sandy Hook to
Execution Rocks. New York: Dodd, Mead & Company.
Redfield, A.C. and M. Rubin. 1962. The age of salt marsh peat and its relation to recent changes
in sea level at Barnstable, Massachusetts. National Academy of Science Proceedings
48:1728–1735.
Reeds, C.A. 1933. The varved clays and other glacial features in the vicinity of New York City.
XVI International Geol. Congress Guidebook 9. Pp. 52–63.
Renfrew, C. 1976. Archaeology and the earth sciences. In: D.A. Davidson and M.L. Shackley,
eds. Geoarchaeology: Earth science and the past. London: Duckworth. Pp. 1–5.
Research Planning, Inc., Tidewater Atlantic Research, Inc., and W.F. Baird & Associates, Ltd.
2004. Archaeological damage from offshore dredging: Recommendations for preoperational surveys and mitigation during dredging to avoid adverse impacts. Report
285
prepared for the Leasing Division, Sand and Gravel Unit, Mineral Management Service,
U.S. Department of the Interior, Herndon, VA. OCS Report MMS 2004–005.
Reusser, L.J., P.R. Bierman, M.J. Pavich, E. Zen, J. Larsen, and R. Finkel. 2004. Rapid late
Pleistocene incision of the Atlantic passive-margin river gorges. Science 305:499–502.
Reynolds, B. 1978. Beothuk. In: B.G. Trigger, ed. Handbook of North American Indians, Vol.
15: Northeast. Washington, D.C.: Smithsonian Institution. Pp. 101–108.
Rich, F.J. 1996. Palynology and depositional history of Pleistocene sediments from Ulmers
Pond, Valdosta, Georgia. Palynology 20:251.
Rich, F.S. and R.K. Booth. In press. Quaternary vegetation and depositional history of St.
Catherines Island. In: G.A. Bishop, H.B. Rollins and D.H. Thomas, eds. Proceedings of the
Fourth Caldwell Conference. Anthropological Papers of the American Museum of Natural
History.
Riess, W., M. Mulholland, J. Donnelly, P. Newby, and I. Buynevich. 2003. Underwater
prehistoric and historic archaeological study, Neptune Regional Transmission Project, New
York and New Jersey. Prepared for Ecology & Environment, Inc., by WRCM, Inc., Bristol,
ME. Report on file at the New Jersey Historic Preservation Office, Trenton.
Riggs, S.R., S.W. Snyder, A.C. Hine, and D.L. Mearns. 1996. Hardbottom morphology and
relationship to the geologic framework: Mid-Atlantic Continental Shelf. Journal of
Sedimentary Research 66(4):830–846.
Riggs, S.R., L.L. York, J.F. Wehmiller, and S.W. Snyder. 1992. Depositional patterns resulting
from high-frequency Quaternary sea-level fluctuations in northeastern North Carolina. In:
C.H. Fletcher III and J.F. Wehmiller, eds. Quaternary Coasts of the United States: Marine
and Estuarine Systems. SEPM Special Publication No. 48. Society for Sedimentary
Geology, Tulsa. Pp. 141–153.
Ritchie, W.A. 1953. A probable Paleo-Indian site in Vermont. American Antiquity 18:249–258.
Ritchie, W.A. 1971. A typology and nomenclature for New York projectile points. Revised
edition. New York State Museum Bulletin 384. The University of the State of New York,
Albany.
Ritchie, W.A. 1980. The archaeology of New York State. Revised edition. Garden City, NY:
Natural History Press.
Ritchie, W.A. and R.E. Funk. 1971. Evidence for Early Archaic occupation on Staten Island.
Pennsylvania Archaeologist 41:45–49.
Robbin, D.M. and J.J. Stipp. 1979. Depositional rate of laminated soilstone crusts, Florida Keys.
Journal of Sedimentary Petrology 49:175–180.
286
Robinson, B.S. 1992. Early and Middle Archaic period occupation in the Gulf of Maine region:
Mortuary and technological patterning. In: B.S. Robinson, J.B. Petersen, and A.K.
Robinson, eds. Early Holocene occupation in northern New England. Occasional
Publications in Maine Archaeology, No. 9, Maine Historic Preservation Commission,
Augusta. Pp. 63–116.
Robinson, B.S., J.C. Ort, W.A. Eldridge, A.L. Burke., and B.G. Pelletier. 2009. Paleoindian
aggregation and social context at Bull Brook. American Antiquity 24:427–429.
Robinson, B.S. and J.B. Petersen. 1992. Introduction: Archaeological visibility and patterning in
northern New England. In: B.S. Robinson, J.B. Petersen, and A.K. Robinson, eds. Early
Holocene Occupation in Northern New England. Occasional Publications in Maine
Archaeology No. 9, Maine Historic Preservation Commission, Augusta. Pp. 1–12.
Robinson, B.S., J.B. Peterson, and A.K. Robinson, eds. 1992. Early Holocene occupation in
northern New England. Occasional Publications in Maine Archaeology No. 9, Augusta.
Robinson, D.S. 2003. Prehistoric underwater CRM in the northeastern United States: Recent
projects and approaches. Paper presented at the Society for Historical Archaeology 36th
Annual Conference on Historical and Underwater Archaeology, Providence, RI.
Robinson, M., C. Alexander, C. McCabe, C. Jackson, and D. Crass. 2010. Threatened
archaeological, historic and cultural resources of the Georgia Coast: Identification,
prioritization and management using GIS technology. Geoarchaeology 25(3):312–326.
Robinson, D.S. and B. Ford. 2003. Marine archaeological reconnaissance survey Weaver’s Cove
LNG Project (offshore) FERC Docket No. PF-03-04-000, Fall River and Somerset,
Massachusetts. Report prepared for Weaver’s Cove Energy, Fall River, MA. Report on file
at PAL, Pawtucket, RI.
Robinson, D.S., B. Ford, H. Herbster, and J.N. Waller, Jr. 2004. Marine archaeological
reconnaissance survey, Cape Wind Energy Project, Nantucket Sound, Massachusetts. Report
prepared for Cape Wind Associates, LLC, Boston, MA. Report on file at PAL, Pawtucket,
RI.
Robinson, D.S., A.D. Leveillee, and J.N. Waller, Jr. 2005. Cedar Tree Beach underwater
archaeology project, Warwick, Rhode Island. Informational poster summarizing the interim
results of an on-going underwater archaeological investigation of a submerged pre-contact
settlement site in northwestern Narragansett Bay. Poster prepared for Rhode Island
Archaeology Day 2005. Poster on file at the Rhode Island Historical Preservation and
Heritage Commission, Providence, and PAL, Pawtucket, RI.
Robinson, D.S. and J.N. Waller, Jr. 2002. Phase I underwater archaeological reconnaissance
survey for submerged prehistoric cultural resources, HubLine Mainline and Deer Island
Lateral Offshore Gas Transmission Pipeline Project, Boston, Massachusetts. Prepared for
Algonquin Gas Transmission Company, Boston, MA. PAL Report No. 1098. On file at the
Public Archaeology Laboratory, Inc., Pawtucket, RI.
287
Rosenau, J.C., G.L. Faulkner, C.W. Hendry, Jr., and R.W. Hull. 1977. Springs of Florida.
Bulletin No. 31. Bureau of Water Resources Management Florida Department of
Environmental Regulation. Tallahassee, FL.
Ross, W.G. 1886. Our Coast Guard: A brief history of the United States Revenue Marine
Service. Harper’s New Monthly Magazine 73:438.
Russell, H.S. 1976. A long deep furrow: Three centuries of farming in New England. Hanover,
NH: University Press of New England.
Russo, M. 1992. Chronologies and cultures of the St. Marys region of northeast Florida and
southeast Georgia. Florida Anthropologist 45:2:107–126.
Russo, M. 1994. A brief introduction to the study of Archaic mounds in the Southeast.
Southeastern Archaeology 13:82–92.
Russo, M. 2006. Archaic shell rings of the southeast U.S.: National Historic Landmarks historic
context. Southeast Archeological Center, National Park Service, Tallahassee.
Sain, D. 2008. The role of Clovis blade technology at the Topper Site (38AL23). Paper presented
at the 65th meeting of the Southeastern Archaeological Conference, Charlotte, NC.
Sanders, J.E. and N. Kumar. 1975a. Evidence of shoreface retreat and in-place “drowning”
during Holocene submergence of barriers, shelf off Fire Island, New York. Geological
Society of America Bulletin 86:65–76.
Sanders, J.E. and N. Kumar. 1975b. Holocene shoestring sand on inner continental shelf off
Long Island, New York. American Association of Petroleum Geologists Bulletin 59:997–
1009.
Sands, J.O. 1988. Gunboats and warships of the American Revolution. In: G.F. Bass, ed. Ships
and shipwrecks of the Americas: A history based on underwater archaeology. New York:
Thames and Hudson. Pp. 149–168.
Sandweiss, D.H., H. McInnis, R.L. Burger, A. Cano, B. Ojeda, R. Paredes, M.C. Sandweiss, and
M.D. Glascock. 1998. Quebrada Jaguay: Early South American maritime adaptations.
Science 281:1830–1832.
Sanger, D. 1996. Gilman Falls site: Implications for the Early and Middle Archaic of the
maritime peninsula. Canadian Journal of Archaeology 20:7–28.
Sassaman, K.E. 1985. A preliminary typological assessment of MALA hafted bifaces from the
Pen Point site, Barnwell County, South Carolina. South Carolina Antiquities 17:1–17.
Sassaman, K.E. 1991. Economic and social contexts of early ceramic vessel technology in the
American Southeast. Ph.D. Dissertation, University of Massachusetts, Amherst.
288
Sassaman, K.E. 1993. Early pottery in the Southeast: Tradition and innovation in cooking
technology. Tuscaloosa: University of Alabama Press.
Sassaman, K.E. 1995. The cultural diversity of interactions among mid-Holocene societies of the
American Southeast. In: M.S. Nassaney and K.E. Sassaman, eds. Native American
interactions: Multiscalar analyses and interpretations in the eastern woodlands. Knoxville:
University of Tennessee Press. Pp. 174–204.
Sassaman, K.E. 2009. Outsmarting rising water for 8000 years: An archaeology of ancient
cultures of the St. Johns River Valley. Lecture at the “Evenings at Whitney,” Whitney Lab’s
Center for Marine Studies, Marineland, FL.
Sassaman, K.E. and D.G. Anderson. 1994. Middle and Late Archaic archaeological records of
South Carolina: A synthesis for research and resource management. Council of South
Carolina Professional Archaeologists, Columbia.
Sassaman, K.E. and D.G. Anderson, eds. 1996. Archaeology of the Mid-Holocene Southeast.
Tallahassee: University of Florida Press.
Sassaman, K.E., M.J. Brooks, G.T. Hanson, and D.G. Anderson. 1990. Native American
prehistory of the middle Savannah River Valley: A synthesis of archaeological
investigations on the Savannah River Site, Aiken and Barnwell Counties, South Carolina.
Savannah River Archaeological Research Papers 1, Occasional Papers of the Savannah
River Archaeological Research Program, South Carolina Institute of Archaeology and
Anthropology, Columbia.
Sassaman, K.E., G.T. Hanson, and T. Charles. 1988. Raw material procurement and the
reduction of hunter-gatherer range in the Savannah River Valley. Southeastern Archaeology
7(2):79–94.
Saunders, J.W., T. Allen, and R.T. Saucier. 1994. Four Archaic mound complexes in northeast
Louisiana. Southeastern Archaeology 13:134–153.
Saunders, J., R.D. Mandel, R.T. Saucier, E.T. Allen, C.T. Hallmark, J.K. Johnson, E.H. Jackson,
C.M. Allen, G.L. Stringer, D.S. Frink, J.K. Feathers, S. Williams, K.J. Gremillion, M.F.
Vidrine, and R. Jones. 1997. A Mound Complex in Louisiana at 5400–5000 Years B.P.
Science 277:1796–1799.
Saunders, R. 1994. The case for Archaic period mounds in southeastern Louisiana. Southeastern
Archaeology 13:118–134.
Schlee, J. and R.M. Pratt. 1970. Atlantic Continental Shelf and the slope of the United States –
gravels of the northeastern part. U.S. Geological Survey Professional Paper 529-H. U.S.
Government Printing Office, Washington, D.C.
Schlesinger, A.M. 1917. The colonial merchants and the American Revolution, 1763–1776.
Ph.D. Dissertation, Columbia University.
289
Schmidt, W. 1997. Geomorphology and physiography of Florida. In: A.F. Randazzo and D.S.
Jones, eds. Geology of Florida. Gainesville: University Press of Florida. Pp. 1–12.
Schnitker, D. 1974. Postglacial emergence of the Gulf of Maine. Geological Society of America
Bulletin 85:491–494.
Schnitker, D., D.F. Belknap, T.S. Baccus, J. Friex, B. Lusardi, and D.M. Popek. 2001.
Deglaciation of the Gulf of Maine. In: T.K. Weddle and M.J. Retelle, eds. Deglacial history
and relative sea-level changes, northern New England and adjacent Canada. Geological
Society of America Special Papers 351. Pp. 9–34.
Scholl, D.W., F.C. Craighead, and M. Stuiver. 1969. Florida submergence curve revised: Its
relation to coastal sedimentation rates. Science 163:562
Schuldenrein, J., D.M. Thieme, T. Epperson, M. Smith, E. Thomas, and L. McWeeney. 2000. A
geomorphological and archaeological study in connection with the New York and New
Jersey Harbor Navigation Study, Upper and Lower Bay, Port of New York and New Jersey,
Hudson, Essex, and Union Counties, New Jersey, Kings, Richmond, and New York
Counties, New York. Report prepared by Geoarchaeology Research Associates, Riverdale,
NY for U.S. Army Corps of Engineers, New York District.
Schwab, W.C., Denny, J.F., Butman, B., Danforth, W.W., Foster, D.S., Swift, B.A., Lotto, L.L.,
Allison, M.A., and Thieler, E.R.. 2000. Seafloor Characterization Offshore of the NewYorkNew Jersey Metropolitan Area using Sidescan-Sonar: U.S. Geological Survey Open-File
Report 00-295.
Science Applications, Inc. (SAI). 1981. A cultural resource survey of the continental shelf from
Cape Hatteras to Key West. 4 vols. Report submitted to the U.S. Dept. of the Interior,
Bureau of Land Management, New Orleans, LA.
Scott, T.M. 1997. Pleistocene to Holocene history of Florida. In: A.F. Randazzo and D.S. Jones,
eds. The Geology of Florida. Gainesville: University Press of Florida. Pp. 57–68.
Sellet, F. 1998. The French connection: Investigating a possible Clovis-Solutrean link. Current
Research in the Pleistocene 15:67–68.
Shackleton, N.J. 1987. Oxygen isotopes, ice volume and sea level. Quaternary Science Reviews
6:183–190.
Shaw, J., D.J.W. Piper, G.B.J. Fader, E.L. King, B.J. Todd, T. Bell, M.J. Batterson, and D.G.E.
Liverman. 2006. A conceptual model of the deglaciation of Atlantic Canada. Quaternary
Science Reviews 25:2059–2081.
Shepard, F.P. 1963. Submarine geology. Original published 1948. New York: Harper & Row.
Shepherd, J.F. and G.M. Walton. 1972. Shipping, maritime trade, and the economic development
of colonial North America. London: Cambridge University Press.
290
Shinn, E.A. 1963. Spur and groove formation on the Florida Reef Tract. Journal of Sedimentary
Petrology 33:291–303.
Shinn, E.A. 1988. The geology of the Florida Keys. Oceanus 31:46–53.
Shinn, E.A., B.H. Lidz, J.L. Kindinger, J.H.Hudson, and R.B. Halley. 1989. A field guide: Reefs
of Florida and the Dry Tortugas, a field trip to the modern carbonate environments of the
Florida Keys and the Dry Tortugas. International Geological Congress. IGC Field Trip T1
76. American Geophysical Union, Washington, D.C.
Shipp, R.C. 1989. Late Quaternary sea-level fluctuations and geologic evolution of four
embayments and adjacent inner shelf along the northwestern Gulf of Maine. Unpublished
Ph. D. thesis, University of Maine, 2 volumes.
Shipp, R.C., Belknap, D.B., Kelley, J.T. 1989. A submerged shoreline on the inner continental
shelf of the western Gulf of Maine. In R.D. Tucker and R.G. Marvinney, eds. Studies in
Maine Geology, vol. 5. Maine Geological Survey, Augusta, ME. Pp.11–28.
Shipp, R.C., D.F. Belknap, and J.T. Kelley. 1991. Seismic-stratigraphic and geomorphic
evidence for a post-glacial sea-level lowstand in the northern Gulf of Maine. Journal of
Coastal Research 7:341–364.
Shipping and Mercantile Gazette. 1856. Disasters. On file, Library of Congress, Washington,
D.C.
Shomette, D.G. 2007. Shipwrecks, sea raiders, and maritime disasters along the Delmarva coast.
Baltimore: Johns Hopkins University Press.
Siddall, M., E.J. Rohling, A. Almogi-Labin, C. Hemleben, D. Meischner, I. Schmelzer, and D.A.
Smeed. 2003. Sea-level fluctuations during the last glacial cycle. Nature 423:853–858.
Siddall, M., E.J. Rohling, W.G. Thompson, and C. Waelbroeck. 2008. Marine isotope stage 3 sea
level fluctuations: Data synthesis and new outlook. Reviews of Geophysics 46:1–29
Sievers, J. 2006. Drowned pens and shaking hands: Sea providence narratives in seventeenthcentury New England. William and Mary Quarterly 63(4).
Simpkins, D.L. 1975. A preliminary report on test excavations at the Sapelo Island shell ring,
1975. Early Georgia 3(2):15–37.
Singewald, J.T. and T.H. Slaughter. 1949. Shore erosion in Tidewater Maryland. Maryland
Department of Geology, Mines, and Mineral Resources, Bulletin 6. Baltimore, MD.
Sirkin, L. 1977. Late Pleistocene vegetation and environments in the Middle Atlantic region.
Annals of the New York Academy of Sciences 288:206–217.
Sirocko, F. 2003. Ups and downs of the Red Sea. Nature 423:813–814.
291
Sloan, G. 2004. Aboard the new grande dame of cruise ships. USA Today January 16.
http://www.usatoday.com/travel/destinations/2004-01-16-queen-mary-2_x.htm.
Smith, B.D. 1986. The Archaeology of the southeastern United States: From Dalton to deSoto
10,500–500 B.P. Advances in World Archaeology Vol. 5. New York: Academic Press, Inc.
Smith, B.W. 1991. The Late Archaic-Poverty Point trade network. In: K.M. Byrd, ed. The
Poverty Point culture: Local manifestations, subsistence practices, and trade networks.
Geoscience and Man 29:173–180.
Smith, R.C. 1988. Treasure ships of the Spanish Main: The Iberian-American maritime empires.
In: G.F. Bass, ed. Ships and shipwrecks of the Americas: A history based on underwater
archaeology. New York: Thames and Hudson. Pp. 85–106.
Smith, S.D. 2003. Review: Reckoning with the Atlantic economy. The Historical Journal
46(3):749–764.
Smithsonian Institution. n.d. A Brief History of Lighthouses. National Museum of American
History,
Behring
Center,
Smithsonian
Institution,
Washington,
D.C.
http://americanhistory.si.edu/collections/lighthouses/history.htm.
Snow, D.R. 1980. Archaeology of New England. New York: Academic Press.
Snow, E.R. 1944. Great storms and famous shipwrecks of the New England coast. Boston: The
Yankee Publishing Company.
Southeast Area Monitoring and Assessment Program–South Atlantic (SEAMAP–SA). 2001.
Distribution of bottom habitats on the continental shelf from North Carolina through the
Florida Keys. SEAMAP–SA Bottom Mapping Workgroup, Atlantic States Marine Fisheries
Commission, Washington, D.C.
Spalding, P. 1977. The vision of Georgia and the challenge of Spain. In: K. Coleman, ed. A
history of Georgia. Athens: University of Georgia Press. Pp. 25–33.
Spielmann, K.A. 1976. A consideration of the Archaic period in Forsyth County, North Carolina.
B.A. Honors Thesis, Harvard University, Cambridge, MA.
Spiess, A.E. 1990. Maine’s unwritten past: State plan for prehistoric archaeology. Maine Historic
Preservation Commission, Augusta.
Spiess, A.E. and J. Mosher. 1994. Hedden: A Paleoindian site on the Kennebunk Plains. Maine
Archaeological Society Bulletin 34:25–54.
Spiess, A.E., J. Mosher, K. Callum, and N.A. Sidell. 1995. Fire on the plains:
paleoenvironmental data from the Hedden site. Maine Archaeological Society Bulletin
35:13–52.
292
Spiess, A.E., D. Wilson, and J.W. Bradley. 1998. Paleoindian occupation in the New Englandmaritimes region: Beyond cultural ecology. Archaeology of Eastern North America 26:201–
264.
Stanford, J.D., E.J. Rohling, S.E. Hunter, A.P. Roberts, S.O. Rasmussen, E. Bard, J. McManus,
and R.G. Fairbanks. 2006. Timing of meltwater pulse 1a and climate responses to meltwater
injections. Paleoceanography 21:1029–1039.
Stanley, A., K.G. Miller, and P.J. Sugarman. 2004. Holocene sea-level rise in New Jersey: An
interim report. Report submitted to the New Jersey Department of Environmental Protection
Division of Science, Research, and Technology, Trenton.
Stanley, D.J. and A.G. Warme. 1994. Worldwide initiation of Holocene marine deltas by
deceleration of sea-level rise. Science 265:228–231.
Stanyard, W.F. 2003. Archaic period archaeology of northern Georgia. University of Georgia
Laboratory of Archaeology Series Report No. 38 and Georgia Archaeological Research
Design Paper No. 13, Atlanta.
Steffensen, J., K. Andersen, M. Bigler, H. Clausen, D. Dahl-Jensen, H. Fischer et al. 2008.
Highresolution Greenland ice core data show abrupt climate change happens in few years.
Science 321:680–684.
Steffy, J.R. 1988. The thirteen colonies: English settlers and seafarers. In: G.F. Bass, ed. Ships
and shipwrecks of the Americas: A history based on underwater archaeology. New York:
Thames and Hudson. Pp. 107–128.
Stein, D.L. 1992. American maritime documents, 1776–1860. Mystic, CT: Mystic Seaport
Museum Publications.
Steponaitis, V. 1986. Prehistoric archaeology in the southeastern United States. Annual Review
of Anthropology 15:363–404.
Stothers, D.M. 1996. Resource procurement and band territories: A model for lower Great Lakes
PaleoIndian and Early Archaic settlement systems. Archaeology of Eastern North America
24:173–216.
Straus, L.G. 2000. Solutrean settlement of North America? A review of reality. American
Antiquity 65:219–226.
Straus, L.G., D.J. Meltzer, and T. Goebel. 2005. Ice age Atlantis? Exploring the Solutrean-Clovis
“connection.” World Archaeology 37(4):507–532.
Stright, M.J. 1990. Archaeological sites on the North American Continental Shelf. In: N.P. Lasca
and J. Donahue, eds. Archaeological geology of North America. Centennial Special Volume
4, Geological Society of America, Boulder. Pp. 439–465.
293
Stright, M.J. 2004. Research strategy for locating evidence of pre-Clovis sites. Paper presented at
the Society for American Archaeology 69th Annual Meeting, Montreal, Canada.
Stright, M.J., E.M. Lear, and J.F. Bennett. 1999. Spatial data analysis of artifacts redeposited by
coastal erosion: A Case study of McFaddin Beach, Texas. OCS Study, MMS 99-0068. U.S.
Department of Interior, Minerals Management Service, Herndon, VA.
Stubbs, C.C.A., L.R. Sautter, and H.M. Scott. 2007. The Transect River channel: Sonar and
scuba exploration of an ancient river. Poster presentation, Southeastern Section Meeting of
the Geological Society of America, Savannah, GA.
Swift, D.J.P. 1973. Delaware shelf valley: estuary retreat path, not drowned river valley.
Geological Society of America Bulletin 84:2743–27488.
Swift, D.J.P. 1975. Barrier island genesis: Evidence from the middle Atlantic shelf of N.
America. Sedimentary Geology 145:1–43.
Swift, D.J.P. 1976. Coastal sedimentation. In: D.J. Stanley and D.J.P. Swift, eds. Marine
sediment transport and environmental management. New York: John Wiley and Sons. Pp.
255–310.
Swift, D.J.P., B. Holliday, N. Avignone, and G. Shideler. 1972a. Anatomy of a shore face ridge
system, False Cape, Virginia. Marine Geology 12(1):59–84.
Swift, D.J.P., J.W. Koefoed, F.P. Saulsbury, and P. Sears. 1972b. Holocene evolution of the shelf
surface, central and southern Atlantic coast of Florida. In: D.J.P. Swift, D.B. Duane, and
O.H. Pilkey, eds. Shelf Sediment Transport: Process and Pattern. Stroudsburg, PA: Dowden,
Hutchinson, and Ross. Pp. 499–574.
Swift, D.J.P., D.J. Stanley, and J.R. Curray. 1971. Relict sediments on continental shelves: A
reconsideration. Journal of Geology 79:322–346.
Swift, D.J.P., D.B Duane, T.F. McKinney. 1973. Ridge and swale topography of the middle
Atlantic Bight, North America: secular response to the Holocene hydraulic regime. Marine
Geology 15:227–247.
Swift, D.J.P., R. Moir, and G.L. Freeman. 1980. Quaternary rivers on the New Jersey shelf:
relation of seafloor to buried valleys: Geology 8:276–280.
Swift, D.J.P. and P. Sears. 1974. Estuarine and littoral depositional patterns in the surficial sand
sheet central and southern Atlantic of North America. Memoires de l'Institut du Bassin
d'Aquitaine 7:171–189.
Tanner, W.F. 1993. An 8000-year record of sea level change from grain-size parameters: Data
from beach ridges in Denmark. The Holocene 3:220–231.
294
Taylor, K.C., G.W. Lamorey, G.A. Doyle, R.B. Alley, P.M. Grootes, P.A. Mayewski, J.W.C.
White, and L.K. Barlow. 1993. The ‘flickering switch’ of late Pleistocene climate change.
Nature 366:549–552.
Tesar, L.D. and B.C. Jones. 2004. Wakulla Springs Lodge Site (8WA329) in Edward Ball
Wakulla Springs State Park, Wakulla County, Florida: A summary of eleven projects and
management recommendations. Bureau of Archaeological Research, Florida Division of
Historical Resources, Tallahassee.
Thomas, D.H. 2008. The Native American landscapes of St. Catherines Island, Georgia.
Anthropological Papers of the American Museum of Natural History 88. 3 vols.
Thomas, H. 1997. The slave trade: The story of the Atlantic slave trade, 1440–1870. New York:
Simon & Schuster.
Thomas, P. 2001. Contributions to understanding Vermont prehistory: The Chittenden County
Circumferential Highway archaeological studies, 1983–2000. Report No. 204. Submitted to
Vermont Agency of Transportation, Montpelier.
Thompson, W.B. and H. Borns. 1985. Surficial Geologic Map of Maine. Maine Geological
Survey, 1: 500,000.
Thornbury, W.D. 1965. Regional geomorphology of the United States. New York: John Wiley
and Sons, Inc.
Thornbury, W.D. 1980. Investigation of the Woodland period along the South Carolina coast.
Ph.D. Dissertation, University of North Carolina, Chapel Hill.
Toscano, M.A., R.T. Kerhin, L.L. York, T.M. Cronin, and S.J. Williams. 1989. Quaternary
stratigraphy of the inner continental shelf of Maryland. Maryland Geological Survey Report
of Investigation 50.
Toscano, M.A. and L.L. York. 1992. Quaternary stratigraphy and sea level history of the U.S.
Middle Atlantic coastal plain. Quaternary Science Reviews 11:301–328.
Trinkley, M.B. 1980. Investigation of the Woodland period along the South Carolina coast.
Chapel Hill: University of North Carolina.
Tuck, J.A. 1974. Early Archaic horizons in eastern North America. Archaeology of Eastern
North America 2(1):72–80.
Tuck, J.A. 1978. Regional cultural development, 3000 to 300 B.C. In: B.G. Trigger, ed.
Handbook of North American Indians, Vol. 15: Northeast. Washington, D.C.: Smithsonian
Institution Press. Pp. 28–43.
Twichell, D., H. Knebel, and D.W. Folger. 1977. Delaware River: evidence for its former
extension to Wilmington Submarine Canyon. Science 195:483–485.
295
Uboat.net. 2011. Ships hit by U-boats: Ulysses, British
ship. http://www.uboat.net/allies/merchants/ships/1524.html.
steam
passenger
Uchupi, E. 1968. Atlantic Continental Shelf and slope of the United States-physiography. U.S.
Geological Survey Professional Paper 529-C. U.S. Government Printing Office,
Washington, D.C.
Uchupi, E. 1970. Atlantic Continental Shelf and slope of the United States-shallow structure.
U.S. Geological Survey Professional Paper 529-I. U.S. Government Printing Office,
Washington, D.C.
Uchupi, E. 2004. The Stellwagen Bank region of eastern Massachusetts: A Wisconsin glaciated
Cenozoic sand banks/delta? Marine Geology 204:325–347.
Uchupi, E., N. Driscoll, R.D. Ballard, and S.T. Bolmer. 2001. Drainage of late Wisconsin glacial
lakes and the morphology and late Quaternary stratigraphy of the New Jersey-southern New
England Continental Shelf and slope. Marine Geology 172:117–145.
Uchupi, E., G.S. Giese, D.G. Aubrey, and D.J. Kim. 1996. The late Quaternary construction of
Cape Cod, Massachusetts: A reconsideration of the W. M. Davis model. Geological Society
of America Special Paper 309. Geological Society of America, Boulder.
U.S. Bureau of Marine Inspection and Navigation. 1886. Merchant vessels of the United States,
with the official numbers and signal letters and list of vessels belonging to the United States
Government with distinguishing signals. 46th edition. Washington, D.C.: Government
Printing Office.
U.S.
Coast
Guard
Historian’s
Office.
http://www.uscg.mil/history/default.asp.
2008a.
U.S.
Coast
Guard
Historian’s
Office.
2008b.
http://www.uscg.mil/History/webcutters/First_Cutters.asp.
Coast
The
first
Guard
history.
ten
cutters.
U.S. Life-Saving Service. 1894. Organization and methods of the United States Life-Saving
Service: Report read before the Committee on Lifesaving Systems and Devices,
International Marine Conference, November 22, 1889. Government Printing Office,
Washington, D.C.
Van de Plassche, O. 1990. Mid-Holocene sea-level change on the Eastern Shore of Virginia.
Marine Geology 91:149–154.
Veatch, A.C., and P.A. Smith. 1939. Atlantic submarine valleys of the United States and the
Congo submarine valley. Geological Society of America Special Paper 7.
Volmar, M. 2003. Riverine dynamics at a submerged Early Woodland site in the Connecticut
River. Paper presented at the Society for Historical Archaeology 36th Annual Conference on
Historical and Underwater Archaeology, Providence, RI.
296
Voulgaris, B. 2009. From Steamboat Inspection Service to U.S. Coast Guard: Marine safety in
the
United
States
from
1838–1946.
http://www.uscg.mil/History/articles/VoulgarisMarineSafety2009.pdf.
Waguespack, N.M. and T.A. Surovell. 2003. Clovis hunting strategies, or how to make out on
plentiful resources. American Antiquity 68:333–352.
Walker, R.B. and B.N. Driskell, eds. 2007. Foragers of the terminal Pleistocene in North
America. Lincoln: University of Nebraska Press.
Wallace, B.L. 2003. L’Anse aux Meadows and Vinland: An abandoned experiment. In: J.H.
Barrett, ed. Contact, continuity, and collapse: The Norse colonization of the North Atlantic.
Trunhout, Belgium: Brepols Publishers. Pp. 207–238.
Waller, J.N., Jr., S. Cherau, A.P. Mair, and M. Hubbard. 2010. Cultural resources review and
assessment Silvio O. Conte National Wildlife Refuge, Connecticut, Massachusetts, New
Hampshire and Vermont. PAL Report No. 2418. Submitted to the U.S. Fish and Wildlife
Service, Hadley, Massachusetts. On file at the PAL, Pawtucket, RI.
Ward, L.W., and D.S. Powars. 2004. Tertiary lithology and paleontology, Chesapeake Bay
region. In: S. Southworth and W. Burton, eds. Geology of the National Capital Region—
field trip guidebook. Reston, VA: U.S. Geological Survey Circular 1264. Pp. 263–281.
Ward, T.H. 1983. A review of archaeology in the North Carolina Piedmont: A study of change.
In: M.A. Mathis and J.J. Crow, eds. The prehistory of North Carolina: An archeological
symposium. North Carolina Department of Cultural Resources, Division of Archives and
History, Raleigh. Pp. 53–80.
Waring, A.J. Jr. and L.H. Larson Jr. 1968. The shell ring on Sapelo Island (originally 1955–
1960). In: S. Williams, ed. The Waring Papers: The collected works of Antonio J. Waring,
Jr. Papers of the Peabody Museum of Archaeology and Ethnology Vol. 58. Cambridge, MA:
Harvard University. Pp. 263–278.
Warren, L.O. 1968. The Apollo Beach site, Hillsboourgh County. Florida Anthropologist 21: 92–
94.
Waters, M.R. 1992. Principles of geoarchaeology: A North American perspective. Tucson: The
University of Arizona Press.
Waters, M.R. and T.W. Stafford Jr. 2007. Redefining the age of Clovis: Implications for the
peopling of the Americas. Science 315:1122–1126.
Watson, P.J. 1989. Early plant cultivation in the eastern woodlands of North America. In: D.R.
Harris and G.C. Hillman, eds. Foraging and farming: The evolution of plant exploitation.
London: Unwin Hyman. Pp. 555–571.
297
Watts, G.P. 1988. The Civil War at sea: Dawn of an age of iron and engineering. In: G.F. Bass,
ed. Ships and shipwrecks of the Americas: A history based on underwater archaeology. New
York: Thames and Hudson. Pp. 207–230.
Watts, G.P. 1995. A submerged cultural resource management document and GIS database for
the Charleston Harbor Project study area, Charleston, South Carolina. Tidewater Atlantic
Research, Inc., Washington, NC. South Carolina Department of Health and Environmental
Control. http://www.scdhec.gov/environment/ocrm/pubs/docs/tech_docs/cult-hist6.pdf.
Watts, W.A. 1975. A late-Quaternary record of vegetation from Lake Annie, south-central
Florida. Geology 3:344–346.
Watts, W.A. and B.C.S. Hansen. 1988. Environments of Florida in the late Wisconsin and
Holocene. In: B. Purdy, ed. Wet site archaeology. West Caldwell, NJ: Telford Press. Pp.
307–323.
Watts, W.A., B.C. S. Hansen, and E.C. Grimm. 1991. Camel Lake: A 40,000 year record of
vegetational and forest history from northwest Florida. Ecology 73 (3): 1056–1066.
Watts, W.A. and M. Stuiver. 1980. Late Wisconsin climate of northern Florida and the origin of
species-rich deciduous forest. Science 210:325–327.
Weaver, A., O. Saenko, P. Clark, and J. Mitrovica. 2003. Meltwater pulse 1A from Antarctica as
a trigger of the Bolling-Allerod warm interval. Science 299:1709–1713.
Weaver, W. 2002. Paleoecology and prehistory: Fossil pollen at Gray’s Reef National Marine
Sanctuary. M.S. Thesis, University of Georgia, Athens.
Weems, R.E. and L.E. Edwards. 2001. Geology of Oligocene, Miocene and younger deposits in
coastal Georgia. Georgia Geologic Survey Bulletin 131. Department of Natural Resources,
Environmental Protection Division, Atlanta.
Weems, R.E. and W.A. Lewis. 2002. Structural and tectonic setting of the Charleston, South
Carolina, region: Evidence from the Tertiary stratigraphic record. Geological Society of
America Bulletin 114(1):24–42.
Weiss, D. 1974 Late Pleistocene stratigraphy and paleoecology of the Lower Hudson River
estuary. Geological Society of America Bulletin 85(10):1561–1570.
Weiss, D., K. Geitzenauer, F. Shaw. 1976 Foraminifera, diatoms and mollusks as potential
Holocene paleoecologic indicators in the Hudson Estuary. Abstract. Geological Society of
America Abstract Programs (USA) (GAAPBC) 8(2):296–297.
Wells, D.V. 1994. Non-energy resources and shallow geological framework of the inner
continental margin off Ocean City, Maryland. Maryland Geological Survey, Coastal and
Estuarine Geology Open File Report No. 16.
Wentz, R. 2009. Personal communication.
298
Werner, D. 1964. Vestiges of Paleo-Indian occupation near Port Jervis, New York. New World
Antiquity 11:30–52.
Whatley, J. Jr. 1985. The possible extension of the Paris Island stemmed point to the Georgia
Coastal Plain. The Profile 47:11–13.
Whitmore, F.C., Jr., K.O. Emery, H.B.S. Cooke, and D.J.P. Swift. 1967. Elephant teeth from the
Atlantic Continental Shelf. Science 156:1477–1481.
Will, R., J.A. Clark, E. Moore. 1996. Phase III archaeological data recovery at the Little Ossipee
North site (7.7), Bonny Eagle Project (FERC #2529), Cumberland County, Maine. Draft
report on file with the Maine Historic Preservation Commission, Augusta.
Will, R. and R. Cole-Will 1996. A probable Middle Archaic cemetery: The Richmond-Castle site
in Surry, Maine. Archaeology of Eastern North America 24:149–158.
Will, R. and E. Moore. 2002. Recent late Paleoindian finds in Maine. Bulletin of the Maine
Archaeological Society 42(1):1–14.
Will, R., E. Moore, and C. Dorian. 2001. The Esker site (84.12): A 14C dated Paleoindian
campsite along the Kennebec River in Caratunk, Maine. Report on file with the Maine
Historic Preservation Commission, Augusta.
Willoughby, C.C. 1898. Prehistoric burial places in Maine. Archaeological and Ethnological
Papers of the Peabody Museum I(6). Cambridge, MA: Harvard University.
Wilson, D., R. Will, and J. Cormier. 1995. The Nicholas site: A late Paleoindian campsite in
southern Oxford County, Maine. Report on file with the Maine Historic Preservation
Commission, Augusta.
Winker, C.D. and J.D. Howard. 1977. Correlation of tectonically deformed shorelines of the
southern Atlantic coastal plain. Geology 5:123–127.
Winterhalder, B. 1981. Optimal foraging strategies and hunter-gatherer research in anthropology:
theory and models. In: B. Winterhalder and E.A. Smith, eds. Hunter-gatherer foraging
strategies. Chicago: University of Chicago Press. Pp. 13–35.
Wisner, G. 1996. Saltville site has evidence of 14,000-year-old feasts. The Mammoth Trumpet
1(4):18–20. Center for the Study of the First Americans, Corvallis, OR.
Witthoft, J. 1952. A Paleo-Indian site in eastern Pennsylvania: An early hunting culture.
Proceedings of the American Philosophical Society 96:464–495.
Witthoft, J. 1954. Notes on the Archaic of the Appalachian region. American Antiquity 25:79–
85.
Woolsey, J.R. Jr. 1977. Neogene stratigraphy of the Georgia Coast and Inner Continental Shelf.
Ph.D. Dissertation, University of Georgia, Athens.
299
Woolsey, J.R. and V.J. Henry. 1974. Shallow, high rResolution seismic investigations of the
Georgia Coast and Inner Continental Shelf. Georgia Geological Survey Bulletin 87:167–
187.
Wright, J.D., R.E. Sheridan, K.G. Miller, J. Uptegrove, B.S. Cramer, and J.V. Browning. 2009.
Late Pleistocene sea level on the New Jersey margin: implications to eustasy and deep-sea
temperature. Global and Planetary Change 66:93–99.
Yesner, D.R. 1980. Maritime hunter-gatherers: Ecology and prehistory. Current Anthropology
21:727–750.
Yokoyama, Y., K. Lambeck, P. De Deckker, P. Johnston, and L.K. Fifield. 2000. Timing of the
Last Glacial Maximum from observed sea-level minima. Nature 406:713–716.
300
PAGE LEFT INTENTIONALLY BLANK
301
APPENDIX A – SOURCES USED FOR SHIPWRECK DATABASE