Journal of Environmental Management 319 (2022) 115730

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

The economics of sediment quality on barrier shoreline restoration

Rex H. Caffey a, *, Daniel R. Petrolia b, Ioannis Y. Georgiou c, d, Michael D. Miner d, Hua Wang a,
Brittany Kime c
a
Center for Natural Resource Economics & Policy, Department of Agricultural Economics Louisiana State University, USA
b
Department of Agricultural Economics, Mississippi State University, USA
c
Department of Earth and Environmental Sciences and Pontchartrain Institute for Environmental Sciences, University of New Orleans, USA
d
Water Institute of the Gulf, USA

A R T I C L E I N F O A B S T R A C T

Keywords: This paper depicts a simulation-based assessment of sediment quality on the performance of dedicated dredging
Barrier system projects for barrier island restoration in coastal Louisiana, USA. The research involved the development and
Break-even value integration of two sub-models. In the first, geomorphic modeling was used to simulate sediment transport dy­
Economic tradeoff
namics within a proxy barrier island template over a 50-year trajectory. The template was assumed to be
Grain size
Mechanical placement
nourished with one of two sources of dredged material: nearshore (NS) sediments of lower quality (smaller grain
Restoration diameter, higher organic fines); or higher quality sediments from distal sources located on the Outer Continental
Shelf (OCS). In the second model, agency project records and commercial bids were used to estimate project
construction costs as a function of dredge material quantity, transport distance, and project target elevation.
These sub-models were coupled within a net present value framework from which average annual break-even
values for ecosystem services (EBEV) were derived as an efficiency metric for comparing the economic perfor­
mance of NS- and OCS-sourced projects. Results indicate that in some cases, the physical resiliency afforded by
even small increases in sand diameter (+4 μm d50) can translate to greater long-term economic viability (lower
EBEV) for OCS-sourced sediment transported over longer distances. Moreover, projects constructed with much
higher diameter OCS sediment (+44 μm d50) with low fines and transported over relatively long distances (200
μm, 5% fines, 15–20 miles) were found to be more cost-effective than all comparably-sized projects constructed
with lower quality NS sediments obtained from proximal sources (156 μm, 20% fines, 3–5 miles). For some
comparisons, this quality advantage yielded a lower EBEV for OCS-sourced projects with transport distances
exceeding 30 miles. Under storm-punctuated simulations, these quality advantages were more pronounced, with
greater physical and economic implications for earlier (Y5) versus later (Y20) occurring storms. Budgeting for
dedicated dredging projects has traditionally centered on the value of sediment as a commodity, with a focus on
material placement cost. The findings of this study, however, indicate that a more comprehensive accounting of
sediment quality and performance is required to maximize the economic efficiency of coastal restoration
spending.

1. Introduction square miles of land became open water between 1932 and 2010 (25%
of Louisiana’s coastal land area). Analyses conducted in support of
Coastal land loss processes are a significant threat to U.S. coastal Louisiana’s 2017 Coastal Master Plan found that the state could expe­
shorelines, a region that comprises less than 10 percent of the national rience annual damages from flooding coast-wide totaling $7.7 to $23.4
land area but more than 40 percent of the nation’s population (NOAA, billion over the next 50 years, depending on future coastal conditions.
2013). Such threats are especially prominent in the Gulf of Mexico re­ Due to the sediment-starved character of the Mississippi River delta
gion, and Louisiana in particular, where barrier islands and shorelines plain (MRDP), sediment suitability and availability are limiting factors
are subject to both climatic and geologic forcing (Morton, 2008). The that have historically constrained large-scale projects. However, the
United States Geological Survey (USGS) determined that about 1883 demand for addressing Louisiana’s coastal land loss crisis means that the

* Corresponding author. Department of Agricultural Economics, Louisiana State University, 179 Martin D. Woodin Hall, Baton Rouge, LA 70803, USA.
E-mail address: rcaffey@agcenter.lsu.edu (R.H. Caffey).

https://doi.org/10.1016/j.jenvman.2022.115730
Received 8 February 2022; Received in revised form 4 July 2022; Accepted 9 July 2022
Available online 20 July 2022
0301-4797/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-
nc/4.0/).
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

portfolio of rapid land building projects (dedicated dredging) will in­

crease, and large quantities (more than 90 million yd3) of sediment will
be needed for coastal restoration in the next 50 years (Khalil and Finkl,
2009). The 2017 Coastal Master Plan indicates that $22 billion (of an
estimated $50 billion) will be needed to fund restoration projects
requiring mechanical placement of sediment (CPRA, 2017).
For dedicated dredging projects, coastal managers must choose be­
tween nearshore (NS) sediment and sediment sourced from outside of
the active coastal system, such as Outer Continental Shelf (OCS) sand or
modern Mississippi River sediment load, for inputs. High-quality sand
(similar to the native beach) is required for beach and dune barrier
habitat restoration, whereas sandy muds are required to rebuild coastal
marshes (Khalil and Finkl, 2009). Availability of suitable sediment re­
sources is a vital factor in restoration efforts, with almost 80% of the
restoration budget allocated to exploration, dredging, and emplacement
of sediment (Khalil et al., 2010; Wang, 2012). Sand resources in state
waters are frequently of lower quality (smaller grain size and higher
organic fraction than OCS sand), and dredging within the littoral zone
can potentially alter wave climate, negatively affecting the landward
shoreline. Moreover, excavation of nearshore sand often occurs within
the active coastal system, compromising the long-term effectiveness of
projects and failing to address the need to supplement a deficit in the
coastal sand budget. Utilizing OCS sand resources minimizes alterations
to wave climate and introduces new sand from outside of the active Fig. 1. Conceptual trajectory of dredge-based reclamation stages (a) on a
coastal system, decreasing the coastal sand deficit and improving project coastal barrier island and (b) on a barrier island nourished with nearshore (NS)
sustainability and geomorphic function. and outer continental shelf (OCS) sediments, including a storm event.
Sediment distribution maps developed by Khalil and Freeman (2015)
estimate the total volume of Louisiana-adjacent OCS sand deposits at operational plans for sediment transport, permitting and regulatory
~100 Billion yd3 primarily from offshore shoals and Paleolithic stream compliance, and dredge vessel availability.
channels such as the Sabine Bank, Tiger and Trinity Shoal Complex, Ship Project construction (Stage II) is a relatively brief period that ac­
Shoal Complex, and St. Bernard Shoal. Approximately three-fourths of counts for the majority of FFC (85%). During this phase, an initial
this material is degradable under current technological and regulatory quantity of sediment from a designated source (Dredgeq) is mechanically
constraints. To date, there has been no comprehensive analysis transported to the project site and deposited within a bounded template
comparing the contributions of NS sediment vs. OCS sediment toward to achieve a target level of post-settlement elevation (Targetq) per
long-term project effectiveness, lifespan, cost, and contribution to sys­ sponsor agency specifications. Construction is typically completed
tem function as a whole. This study aims to provide a better under­ within a single year, although longer periods can be required depending
standing of the economic feasibility of NS and OCS sediments for coastal on various factors, including: distance between source material and
restoration projects on the basis of sediment quality and the capital project footprint; project size and design; dredge capacity limitations;
required to employ various project construction methods. We develop a weather; and, critical habitat constraints that might limit operations
conceptual framework for standardizing site- and system-level assess­ during certain seasons.
ments of dredge-based nourishment projects on Louisiana barrier Project operation and monitoring (Stage III) is the longest period and
islands. We then construct and integrate a geomorphic model of sedi­ can range from 20 to 50 years, depending on the sponsor. During this
ment dynamics with an economic model of harvest, transport, and phase, public benefits derived from an expanded barrier platform. A
deposition to simulate the benefits and costs of NS- and OCS-sourced range of benefits have been used as justification for these projects;
sand transport under a range of project scenarios. This integrated however, storm surge attenuation and provision of coastal habitat are
framework allows us to evaluate economic tradeoffs associated with two of the most frequently cited ecosystem services for coastal barrier
sediment location, quantity, quality, and meteorological forcing over systems (Petrolia and Kim, 2009; Feagin et al., 2010; Barbier et al.,
time. 2011).
While there is some potential for volumetric and surface area in­
2. Materials and methods creases of sediment due to longshore sediment transport processes at the
site and system level, in a transgressive coast, most of these materials
2.1. Overview of a restoration project (and their associated benefits) are expected to diminish over time as
restored land succumbs to physical forcing. Thus, the basic expectation
Large-scale coastal restoration projects are characterized by three is that project benefits will exceed project costs, and the nourishment
distinct stages: engineering and design, construction, and operation and will sustain a subaerial template (Projectq) that has otherwise been lost
monitoring. A graphic depiction helps to illustrate the timing, costs, and over time (Controlq). Despite accounting for the lowest portion of FFC
activities associated with these stages for the generic trajectory of a (5%), monitoring is critical for collecting the data needed to refine ex­
dedicated dredging project (Fig. 1a). Engineering and design (Stage I) is pectations and improve the design and construction of future projects.
the initial stage in which geotechnical surveys and pre-project assess­ Fig. 1b expands the basic trajectory with hypothesized responses for
ment are used to evaluate sediment availability and dynamics for a projects utilizing NS- and OCS- sourced sediments. These curves
proposed template. At this stage, feasibility decisions are based on approximate the observations of project managers and reflect two
“future-with-project” versus “future-without-project” comparisons, important tradeoffs with potential economic implications. First, while
typically expressed in terms of subaerial land surface over the project nearshore sediment sources may be less expensive to harvest and
life. This phase usually accounts for 10 percent of a project’s fully fun­ transport (given their proximity to project sites), they often contain a
ded costs (FFC) and can last 3–10 years, depending on site- and source- higher fraction of organic fines (mud) than OCS sources. Thus, for any
specific requirements for geotechnical surveying, development of

2
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

given Targetq of sand, the volume of NS sediment dredged will typically Summing expression (1) over all sites within the system, we have
exceed the volume of OCS sediment dredged, i.e., NSq > OCSq. Secondly, system quantity of sand at time t as:
managers assert that OCS-sourced projects are typically more resilient
∑
S ∑
S
over time than NS-source ones, thus OCSq’>NSq’. In other words, ΔQt = Δqst = (mst + nst ) (2)
increased resilience is attributed to the larger diameter of OCS sands, s=1 s=1
which can make them more resistant to the physical forces of coastal
Renourishment projects are typically conducted at a single site, with
transport, erosion, and storms (i.e., more energy is required for mobi­
the mechanical placement of sand at that site only. Further, project
lization and transport).
evaluation is typically based on the sand accrued at the project site only,
ignoring any changes in sand accrual at other sites in the barrier system.
2.2. Conceptual benefits framework Defining site s = 1 as the project site, and recognizing that mst = 0 for all
s∕= 1, we may rewrite (2) to separate what is typically evaluated, called
Fig. 2 presents a conceptual model of a “barrier system”, defined as a here “direct”, from what is typically ignored, called here “indirect”.
set of one or more sites that stand in relation (up-drift or down-drift) to
one another.1 Barrier systems are located within an active littoral zone ∑ ∑
S S
ΔQt = Δqst = m1t + n1t + nst (3)
characterized by subtidal transport of coastal sediments. A unit of sand s=1
⏟̅̅̅̅̅⏞⏞̅̅̅̅̅⏟
s=2
located at a position adjacent to, but external from the vertical contour ⏟̅̅̅⏞⏞̅̅̅⏟
Direct
Indirect
of a given site, is considered to be outside of the site boundary. This
designation is necessary for assessing which units of sand are to be Here we wish to distinguish benefits by whether a unit of sand is
counted as beneficial in terms of determining standing, discussed in subaerial or subaqueous; to do so, we expand (1) into:
more detail below. Δqst = mast + nast + mbst + nbst (4)
The quantity of sand in Stage III at a given site in a given period is the ⏟̅̅̅̅̅⏞⏞̅̅̅̅̅⏟ ⏟̅̅̅̅̅⏞⏞̅̅̅̅̅⏟
sum of (a) the quantity of sand at the site in the previous period; (b) any
Subaerial Subaqueous

sand mechanically dredged from NS or OCS sources outside the system where the “a” superscript indicates “above the surface” (subaerial) and
and placed within the site during the current period; (c) the quantity of “b” indicates “below the surface” (subaqueous). At the system level,
sand “captured” from adjacent sites in the current period due to natural substituting (4) into (3), we get:
transport; and (d) the quantity of sand “lost” due to natural transport.
There is some set of functions that dictate how much sand accumulates ∑
S ∑
S ∑
S
ΔQt = Δqst = ma1t + na1t + mb1t + nb1t + nast + nbst
(or sloughs off) at the site and how much is recaptured from adjacent s=1
⏟̅̅̅̅̅⏞⏞̅̅̅̅̅⏟ ⏟̅̅̅̅̅⏞⏞̅̅̅̅̅⏟ s=2 s=2
sites. Direct Subaerial Direct Subaqueous ⏟̅̅̅⏞⏞̅̅̅⏟ ⏟̅̅̅⏞⏞̅̅̅⏟
Indirect Subaerial Indirect Subaqueous
Formally, let the change in quantity of sand at site s at time t, Δqst , be
(5)
expressed as the sum of the quantity of sand added mechanically in the
current period, mst , and the net difference between the quantity of sand where the first set of terms, “Direct Subaerial”, is what is included in a
added and lost via natural transport, nst .2 typical Stage III project performance evaluation, with all others ignored.
Δqst = mst + nst (1) Adding per-unit values to expression (4) yields the change in benefits at
site s at time t:
( ) ( )
Δbst = pa mast + nast + pb mbst + nbst (6)
1
The study team enlisted a project advisory panel consisting of coastal
geomorphologists, restoration managers, and commercial dredgers. With
where pa and pb are the respective values per unit of sand placed sub­
advisory input, 22 candidate projects were identified as the basis for a stan­ aerially and subaqueously, respectively. A simplifying but reasonable
dardized barrier island template to be used for geomorphic and economic assumption is that the value of the benefits of a unit of subaqueous sand
comparisons of NS and OCS sediment sources (Appendix A). Five key points is some fraction of that of a unit of subaerial sand. In this case, we may
emerged from the advisory process: 1) geomorphic dynamics should be write:
modeled at the particle level, as opposed to total volume approach, given that
sediment quality will be highly variable between source locations; 2) physical pb = αpa (0 ≤ α ≤ 1) (7)
simulations should focus on how sand quality affects project resiliency at the
Substituting (7) into (6), we have:
site-level and the system-level. Simulations should address both chronic and
[( ) ( )]
acute forcing (storms); 3) monetized benefits should be based on physical Δbst = pa mast + nast + α mbst + nbst (8)
outputs (volume/area) from the geomorphic model and estimated on an annual
net-basis (future-with minus future-without project); 4) monetized costs should be At the system level, the change in benefits at time t can be expressed
based on a scale-appropriate statistical models derived from extant financial as:
data (e.g. contractor bids, final project reports); and 5) economic comparisons
should go beyond the value of sediment as a commodity and consider the value
∑
S
( ) ( ) ∑ S
( a)
ΔBt = Δbst = pa ma1t + na1t + αpa mb1t + nb1t + pa nst
of ecosystem services generated by dredge material throughout the project s=1 ⏟̅̅̅̅̅̅̅̅̅⏞⏞̅̅̅̅̅̅̅̅̅⏟ ⏟̅̅̅̅̅̅̅̅̅̅̅⏞⏞̅̅̅̅̅̅̅̅̅̅̅⏟ s=2
lifetime. Direct Subaerial Direct Subaqueous ⏟̅̅̅̅̅̅̅⏞⏞̅̅̅̅̅̅̅⏟
2 Indirect Subaerial
As Fig. 2 indicates, sand added or lost via natural transport can originate
∑
S
( )
either from other sites within the barrier system (up-drift or down-drift) or + α pa nbst (9)
“vagrant” sand, i.e., sand from outside the barrier system, either from nearshore s=2
(littoral zone) or OCS sources. Thus, we may write nst = n∼s v ⏟̅̅̅̅̅̅̅̅̅⏞⏞̅̅̅̅̅̅̅̅̅⏟
st + nst , where nst
∼s
Indirect Subaqueous
represents the share originating from other sites within the barrier system (~ S
indicating “not s”) and nvst but these cannot be individually identified at the site Thus, the present value of the change in benefits at site s over time,
level; we observe only a net gain or loss of sand at a site at each period. Within a
defined barrier system, the sum of sand change via natural transport between
S
∑
all sites, is necessarily zero, i.e., n∼s
st = 0. Thus, at the barrier system level,
s=1
any net change in sand quantities via natural transport is necessarily attribut­
able to vagrant sand exchange with the larger littoral boundary and/or offshore
zone (Fig. 2).

3
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Fig. 2. Standard boundaries and processes of sand transport for dredge-based reclamation of a barrier island system comprised of individual sites.

from initial period t = 0 to terminal period t = T, can be expressed as: constructed via dedicated dredging from 1997 to 2018. Commercial bids
were used to expand these data to 93 useable observations. Ordinary
∑
T
[( ) ( )]
Δbs = pa δt mast + nast + α mbst + nbst (10) least-squares analysis was used to estimate project construction costs for
t=0 these data. The final regression model defines project construction cost
(PCC) as a function of 6 variables:
where δ = 1/(1 +r) is the discount factor, and r is the discount rate. ( )
At the system level, the present value of the change in benefits (PVB) PPCC = f Dredgeq , Distance, Distsq, Dune, Payonfill, Program (13)
over time, from initial period t = 0 to terminal period t = T, can be
expressed as: where is Dredgeq is the total quantity of material dredged in million cubic
{ } yards, Distance is the one-way distance from borrow site to the project
∑S ∑ T S ∑
∑ T
[ a ( b )] site in miles, Distsq is distance squared, Dune is the average dune
PVB = Δbst = p a t a b
δ m1t + nst + α m1t + nst (11)
s=1 t=0 s=1 t=0
elevation in feet, Payonfill is a discrete variable representing contractor
payment type, and the sponsoring agency for a project is given by Pro­
Based on geomorphic modeling of barrier island area projections, the gram. The descriptive statistics of the variables and model parameter
quantity of sand benefits can be converted to an acreage basis. In this estimates are reported in Appendix B.
case, the present value of benefits can be expressed as: As described in Fig. 1, 10% of total project costs are typically allo­
{
∑S ∑T
} cated for engineering and design and 5% is budgeted for operation and
[ a ( b )]
PVB = p a t a b
δ ma1t + nast + α ma1t + nast (12) maintenance. While specific data for these two costs are not always
s=1 t=0 readily available, they can be derived algebraically as a function of
construction costs (CC), which accounts for the largest fraction (85%) of
where maa1t represents direct subaerial net acres with mechanical a project’s fully funded cost. In turn, construction costs are estimated
placement in the current period at the project site. The expression of naast from multiple-regression analysis of cost factors for OCS and NS pro­
stands for indirect subaerial net acres via natural transport in the current jects. The corresponding cost in period t for NS and OCS projects is given
period from other sites within the system. And mab1t represents direct by the function:
subaqueous net acres with mechanical placement in the current period
FFCt = Ct (ED) + Ct (CC) + Ct (OM) = 1.18*Ct (CC) (14)
at the project site and nabst stands for indirect subaqueous net acres via
natural transport in the current period from other sites within the sys­
where FFC is the fully funded annual costs of a NS or OCS project in year
tem.
t, inclusive of engineering and design (ED), construction costs (CC), and
operation and maintenance (OM), therefore, the present value of cost
2.3. Cost data and model (PVC) function for NS and OCS projects can be expressed as:
∑
T ∑
T
A cost model was developed from 22 candidate projects identified PVC = δt *FFCt = δt *1.18*Ct (CC) (15)
during the advisory phase of the project. Model development follows t=0 t=0

previous methods for the development of cost models for coastal resto­
where t stands for the number of years of a project and range from 0 to
ration (Merino et al., 2011; Wang , 2012; Caffey et al., 2014). Fig. 3
50. PVC is the total discounted costs (in $) of a NS or OCS project during
depicts the locations of these projects and their borrow sites in relation
the project life. FFCt is the total annual costs of a NS or OCS project in
to the proxy template and the Louisiana Coastal zone. The candidate
year t and δ is the discount factor.
projects included 12 NS-sourced and 10 OCS-sourced projects

4
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

2.4. Integrated assessment model and ecosystem service values and below the surface, it yields a comparative metric for depicting
project efficiency along a broader contour. With additional economic
The present value of benefits (PVB) function (equation (12)) is valuations (or sensitivity analysis), future iterations of the model could
combined with the present value of cost (PVC) function (equation (15)) be parametrized to delineate separate values for subaqueous land (see
to yield a net present value (NPV) function: equations (7)–(9)). Equation (17) allows for the estimation of project

{ }
∑
T S ∑
∑ T
[ ( )] ∑
T
NPV = δt (Bt − Ct ) = PVB − PVC = pa δt maa1t + naast + α mab1t + nabst − δt * 1.18*Ct (CC) (16)
t=0 s=1 t=0 t=1

efficiency at two contours: direct effects (subaerial land above 0.0 m)

where Bt is the sum of benefit in time t, Ct is the sum of cost in time t, δ is and total effects (subaerial and subaqueous above − 0.5 m).
the discount factor and t is the year.
Empirically, the key inputs into the model are the estimates of
sediment quantities over time – maa1t , naast , mab1t , and nabst - which are 2.5. Sediment transport model and scenarios
provided via the sediment transport model described in the next section,
and the per-unit value of the ecosystem services associated with each Data for the development of the sub-model of project benefits were
unit of sediment, pa. The ecosystem services cited in association with obtained from extant literature (i.e. scientific manuscripts and technical
barrier islands and shorelines typically include storm surge attenuation reports), geodatabases, and federal and/or state-owned sources related
(disturbance regulation), habitat provision, and recreation, but the time to coastal sediment inventories and dynamics. Examples of such work
and effort required for the valuation of these benefits can be substantial. included citations of the location and extent of relict delta deposits, their
We adopt the method proposed by Caffey et al. (2014) that treats pa as proximity to the coastal zone, the potential availability of these deposits
endogenous, such that a break-even value for ecosystem services (EBEV) relative to the ongoing transgression of the Louisiana coast, chief drivers
can be derived by setting the BC ratio equal to 1.0 and solving for the of nearshore sediment transport processes within the delta plain, and the
average annual ecosystem service value that equates project benefits to role of coastal sediment sinks (Nairn et al., 2004; Miner et al., 2009;
costs over a given time period. This approach avoids the need to decide Georgiou et al., 2011).
upon and adopt a particular value or set of values, something that is A sediment transport model was developed to provide sediment
fraught with challenges, in favor of a process that identifies the mini­ quantity inputs into the benefits model (Georgiou et al., 2019). The
mum value of ecosystem benefits that would be required for a net pos­ sediment model was designed to proxy a barrier system based on the Isle
itive outcome.3 The calculated value for a given project/scenario Dernieres island chain with NOAA bathymetry data dating to the 1980s
combination can then be compared to that of other project/scenario (Fig. 4). The simulated system consists of a 360 ha (subaerial) central
combinations, where the lower value is interpreted as indicating rela­ barrier island with a large section (898 ha) of an up-drift barrier to the
tively higher project efficiency. In other words, a project with a lower east and a smaller section (166 ha) of a down-drift barrier to the west.
break-even value should be preferred from an efficiency standpoint Additional components include tidal inlets, spit platforms, ebb-deltas,
because it requires a lower magnitude of benefits to justify a given cost. surf zones, and nearshore depositions. The bounded area represents a
Equation (16) can be rewritten as a ratio by setting the right-hand 50 km2 domain for the application of three-dimensional modeling with
side to zero and solving for pa, and defining the value of pa that sat­ coupled waves, tidal currents and full sediment transport and
isfies this equation as the annual average break-even ecosystem service morphology. The model utilizes Delft3D modeling to simulate cumula­
value (EBEV): tive erosion and deposition, with and without project-based nourish­
ments at subaerial and subaqueous boundaries.
∑T t
δ *1.18*Ct (CC) The domain is transected by 192 × 384 grid consisting of cells of
EBEVab = ∑S ∑T t [t=0 a ] ( b ) (17)
s=1
a
t=0 δ ma1t + nast + ma1t + nast
b varying resolution (~20 m nearshore to 1 km offshore). Water is forced
at offshore and lateral boundaries (~6 h for waves, ~1 h for water level)
While equation (17) assumes no difference in the value of land above with a Neumann condition lateral using information from the Wave
Information System (WIS) of the US Army Corps of Engineers and Port
Fourchon NOAA tidal gauge. Changes in relative sea level are incorpo­
3
Richardson et al. (2015) provides an overview of the growing demand for
rated into the simulation based on forecast estimates provided by CPRA
monetized ecosystem services value (ESV) estimates and the increasing use of (2017). Sediment dynamics are depicted by a combined bed load/sus­
benefit transfer within environmental policy. Guidelines are cited for facili­ pended load transport function (van Rijn, 1984a, 1984b) using different
tating more valid transfer of values between a study site and policy site, sand classes to depict bathymetry updating (NS = 156 μm, OCS = 160
including: 1) the need for comparable scope, scale, and population; 2) recog­ μm, 165 μm, and 200 μm).
nizing differences in intermediate and final services; 3) and aggregation ap­ Morphodynamic upscaling is used to allow the model to extend bed-
proaches to avoid double counting. The authors reference web-based databases load and suspended load transport for wash-over, breaching, lateral
that have emerged as a repository for monetized ESV estimates. While these migration, and sediment bypassing. The set-up simulates sand place­
sites are increasingly utilized for benefit cost analysis, the values they contain ment in terms of direct effects (central barrier or project site) and total
often vary by orders of magnitude for a given service. For example, the Gulf of
system effects (down-drift, central, and up-drift barriers) at elevation
Mexico Ecosystem Valuation Database cites ESV estimates ranging from $2.40 -
and depth contours of 1.0, 0.0, and − 0.5 m.
$13,360 per acre/year (US $2012 dollars) as the value of disturbance regulation
from barrier islands and shorelines. Likewise, studies of the habitat provision of Four scenarios were developed for simulation using the sediment
coastal wetlands are cited with estimates ranging from $1.77 to $7854/acre/­ transport model: a baseline scenario, two storm scenarios, and a scenario
year (GecoServ, 2019). Such large ranges reflect the variety and complexity of for larger sand classes under chronic forcing (Table 1). Scenarios varied
non-market valuation methodology, which serves to compound the challenges according to sediment quantity, quality, transport distance, and other
of benefits transfer. factors related to project cost. Within each of the four scenarios,

5
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Fig. 3. Geographic locations of candidate projects (NS- and OC-sourced) for developing a dedicated dredging cost model for barrier shoreline and barrier island
restoration in Louisiana.

quality), and 1.5 (low quality); and, for OCS sources: 1.05 (high quality),
1.1 (average quality), and 1.2 (low quality) (Table 1). These ratios
translate to initial dredge volumes of 12.8–16.2 million m3 for NS
sources and 11–13 million m3 for OCS sources. This material is deposited
on the central barrier to yield a target restoration template of 726 ha
(1794 acres) of subaerial land measured at and above the 0.0 m contour
and a template of 942 ha (2327 acres) of subaqueous land measured at
and above the − 0.5 m contour at Year 0. Sediment transport distance is
assumed to be 3–5 miles for NS and 15–20 miles for OCS sources.
Maximum project life is modeled at 50 years, for a 4% discount rate on
projects costs and benefits. Contractors are assumed to be paid "on the
fill" for target quantities. Dune elevation is set at 6.4 feet and the effects
of sponsoring program on estimated costs are averaged across the six
federal programs that sponsor such projects in the region. A more
detailed explanation of the economic model set-up is provided in Caffey
et al. (2020).
Fig. 4. Model domain and system components used in geophysical simulations.

3. Results
simulation outputs are presented within four boundaries: central barrier
subaerial, central barrier subaqueous, barrier system subaerial, and bar­ The baseline scenario (Appendix C1) is revisited below in Fig. 5 to
rier system subaqueous. Within these boundaries, a total of 44 unique examine how small differences in sediment grain size (NS = 156 μm,
response trajectories are simulated for treatments and controls. All tra­ OCS = 160 μm) interact with sediment quantity, distance, and other
jectories are assumed to begin post-construction, immediately after variables to influence economic performance (Table 1). The order of
required post-settlement elevation is achieved (Stage III @ y0). Results EBEV curves is similar in all four panels (5a,5b,5c,5d), indicating that
are reported in acres to facilitate integration with cost modeling (Section increases in distance tend to increase cost, and increases in depth
2.3). A more detailed explanation of the geophysical model set-up is (contour) typically serve to decrease unit costs. The effects of changes in
provided in Caffey et al. (2020). boundary level (from central to system) are less notable, given that
All model scenarios are based on variations of a single project in EBEV are calculated on a net-acre basis, and the underlying system is
which either NS- or OCS-sourced sediment is mechanically transported deteriorating. Therefore, the brunt of net change at all boundary levels is
into the proxy barrier template. The baseline scenario assumes a target primarily driven by the project.
placement of 10, 700, 000 m3 (~14 million yd3) of sand. Given the At all common transport distances, NS-sourced projects with medium
differences in sediment quality by source and the associated losses due to high CTF ratios (1.3–1.5) are less efficient than OCS projects. Inset
to handling and fines, additional sediment would need to be cut values are provided to facilitate EBEV comparisons at more relevant
(Dredgeq) to fill the desired restoration template (Targetq). Using ranges of sediment transport. At a 3–5 mile range, NS projects with a
regional geotechnical surveys as a basis, the following cut-to-fill (CTF) CTF of 1.2–1.3 are more cost-effective than OCS projects with 10–15
ratios were applied for NS sources: 1.2 (high quality), 1.3 (average

6
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Table 1
Scenarios and assumptions for simulations.
Assumptions Baseline Scenario Early Storm Scenario Late Storm Scenario Large Sand Scenario
3
Targetq (million yd ) 14 14 14 14
Dredgeq Cut-To-Fill (ratio)
NS 1.2, 1.3, 1.5 1.2, 1.3, 1.5 1.2, 1.3, 1.5 1.2, 1.3, 1.5
OCS 1.05, 1.1, 1.2 1.05, 1.1, 1.2 1.05, 1.1, 1.2 1.05, 1.1, 1.2
Sand size (μm)
NS 156 156 156 156
OCS 160 160 160 200
Boundaries: central, central, central, central,
Direct, Indirect system system system system
(Fig. 2 areas) (qst, Qt) (qst, Qt)) (qst, Qt) (qst, Qt)
Contours:
Subaerial, Subaqueous (m) 0.0, − 0.5 0.0, − 0.5 0.0, − 0.5 0.0, − 0.5
Sediment transport (miles)
NS 3–5 3–5 3–5 3–5
OCS 15–20 15–20 15–20 15–20
Project life (years) 50 50 50 50
Discount rate (%) 4 4 4 4
Payonfill (fill = 1) 1 1 1 1
Dune elevation (feet) 6.4 6.4 6.4 6.4
Program (1–6) averaged averaged averaged averaged

Note; Central refers to the project site.

miles transport and CTF of 1.1–1.2 (panel a). However, OCS projects tides, salinity), which helps in the protection of leeward shorelines and
with the highest quality borrow sites (CTF = 1.05) yield an EBEV range wetlands. Under more acute conditions, their role is often described as
of $6659-$6982 at 15-20 miles, which is more efficient than the $7527- “sacrificial”.
$7843 range at 3–5 miles from NS projects using the lowest quality The panels of Fig. 6 depict the economic aspects of simulations
borrow sites (CTF = 1.50). described in Appendix C2. Recall that under the “early storm scenario”,
The disturbance regulation function of barrier islands is a primary the project site is assumed to take a direct hit from a Category 2 hurri­
focus of restoration managers. Under typical meteorological conditions, cane at y5. This impact reduces the amount of remnant subaerial land at
these landforms serve as a buffer to chronic physical forcing (waves, y50 by 60% for the NS-sourced project and 54% for the OCS-sourced

Fig. 5. Ecosystem break-even values (EBEV) for NS- and OCS-soured projects at various boundaries, distances and cut-to-fill ratios (baseline scenario).

7
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Fig. 6. Ecosystem break-even values (EBEV) for NS- and OCS-sourced projects at various boundaries, distances and cut-to-fill ratios (early storm scenario – Year 5).

project. than observed in the baseline scenario. Second, the tradeoff between
As a result, the EBEV curves for subaerial land at the project site quantity and distance has narrowed. At 15–20 miles, OCS projects with a
(panel 6a) increase by a range of $3000 to $4000 compared to the CTF of 1.05 and 1.10 are now either more competitive or somewhat
baseline scenario. As expected, efficiency losses are less pronounced at equal in efficiency to NS-sourced projects at 3–5 miles with CTF of
broader and deeper boundaries, given the amount of sand remaining 1.50–1.30, respectively. The most pronounced differences are evident in
beneath the surface. As a result, the subaqueous contour of the barrier panel 8d, in which the much larger 200 μm sand at 15–20 miles out­
system (panel 6d) depicts a smaller efficiency loss, an increase of only performs nearly all of the EBEVs for NS projects at 3–5 miles. In some
$2400-$3000 in EBEV in panels 6b and 6d. comparisons, the OCS advantage holds up for distances beyond 30 miles.
Under the late storm scenario (Appendix C3), a Category 2 hurricane
is assumed to make landfall on the central barrier in the year 20. Similar 4. Discussion
to the early storm scenario, the storm reduces the amount of remnant
subaerial land at y50, though to a lesser extent (48% NS and 38% OCS). Geomorphic simulations of sand transport within the proxy barrier
The main difference seen is in terms of economic performance. Because template were developed to examine the performance of a 14 million
the storm occurs later in the trajectory, there is more time for pre-impact yd3 renourishment project using NS sand of 156 μm and OCS sands of
benefit accrual. 160–200 μm. Under baseline conditions, the trajectory of subaerial land
In Fig. 7, the EBEV curves for subaerial land at the project site (panel for the central barrier (project site) indicates a small advantage in
7a) increase by a range of only $1000 to $1300 compared to the baseline resilience (increased volume and area) for the OCS sand. This divergence
scenario. The associated efficiency loss at the subaqueous contour of the appears evident at year 10 and slowly expands through year 50. The
barrier system (panel d) is even less, with increases in EBEV of only advantage is less evident when measured at the system level and at
$700-$1000 (panels 7b and 7d). These results clearly illustrate the time subaqueous contours, primarily because of a dilution effect and the net
value of benefits in terms of economic performance, but the estimates export of vagrant sand from the proxy system over time.
depicted in these panels are driven primarily by source material quality In terms of final subaerial land, the central barrier at y50 ends with a
(CTF) at various distances. net quantity of 489 acres for the OCS-sourced project, compared to 325
The large sand class scenario is based on the simulations of Appendix net acres for the nearshore-sourced project. These remnant areas are
C4, which compares the resiliency of a project sourced with much larger reduced by 60% and 54%, respectively, under an early storm scenario
diameter sand (OCS = 200 μm) to the smaller diameter sediment from (y5); and by 48% and 38% under a late storm scenario (y20). Perfor­
the nearshore-sourced project (NS = 156 μm). As mentioned in Ap­ mance advantages are clearly evident for larger OCS sands, ranging from
pendix C4, projects sourced with the OCS sand yielded 70% more marginal improvements at 165 μm to substantial improvements at 200
remnant subaerial land in the large sand simulations. From an economic μm. Projects sourced with 200 μm sand depict the largest divergence
standpoint, this physical advantage is manifest in two distinct ways from the baseline, with 825 net acres of remnant subaerial land
(Fig. 8). remaining in y50 on the central barrier, a near 70% increase over the
First, the range of EBEV for OCS sand is approximately 10% lower performance compared to 160 μm sand. In all simulations, the project

8
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Fig. 7. Ecosystem break-even values (EBEV) for NS- and OCS-sourced projects at various boundaries, distances and cut-to-fill ratios (late storm scenario – Year 20).

trajectories successfully maintain subaerial land, compared to the no- amount of remnant, subaqueous sand remains at subsurface contours
action (control) in which the year of disappearance (YOD) ranged after a storm. In each case (early and late storm), the OCS-sourced
from 30 to 40 years. projects continued to outperform the NS-source projects in terms of
Cost modeling for the baseline project yielded estimates ranging physical resilience. This result appears to confirm manager assertions
from $115 to $148 million for the target placement of 14 million y3 of that OCS-sourced projects perform better under both chronic forcing and
material in y0 (2016 dollars). This cost range reflects a combination of acute conditions.
two types of sediment quality (NS at 156 μm and OCS at 160 μm), 6 CTF Isolating the effects of small differences in sediment quality (grain
ratios (1.05, 1.1, 1.2, 1.3, 1.4, and 1.5), and two transport distance size) on project performance requires holding sediment quantity con­
ranges (3–5 miles for NS-sourced projects and 15–20 miles for OCS stant. With both NS- and OCS-sourced projects modeled at a common
sourced projects). At 3–5 miles of transport, NS-sourced projects with a CTF of 1.20, any economic effects of a 4 μm (2.5% larger) advantage
CTF of 1.2–1.5 were 15–30% less expensive than the OCS projects with a from OCS sand are offset by the shorter transport distance (lower costs)
CTF of 1.05–1.2 and 15–20 miles distances. On a per unit basis, these for NS-sourced projects. This NS advantage narrows, however, under
costs equate to sediment placement costs of $8.20 - $10.60/yd3. storm-punctuated scenarios, given the increased resilience of larger
While placement cost is a metric that often dominates project se­ diameter sand.
lection, it reflects the value of sand as a commodity and fails to fully The economic implications of larger sands are more pronounced. The
account for the services generated throughout a project’s lifetime. An increased resilience of 200 μm sand (28% larger) results in two distinct
examination of project performance yields different results. Integration advantages: (1) EBEVs are 10% lower than OCS-sourced projects at 160
of benefit and cost models within an EBEV framework indicated that, μm; and, (2) a substantial narrowing of the tradeoff between source
despite having higher construction costs, the OCS-sourced projects material quality (CTF) and distance, with superior efficiency for all OCS
outperform NS-sourced projects in many cases, some of which involve projects at a moderate CTF of 1.10, and in some cases at a CTF of 1.20.
large transport distances. The simulated performance of projects with OCS-sourced sand (200 μm,
The minimum ecosystem service value required for project justifi­ 15–20 miles) outperformed nearly all NS projects at 3–5 miles. This OCS
cation (EBEV) increased under storm-punctuated scenarios. This efficiency advantage holds for distances exceeding 30 miles in some
reduction in efficiency is due to the net export of sand for all boundaries comparisons.
and contours and was found to be 20% greater for earlier (Y5) versus
later occurring storms (Y20). In short, the earlier a storm occurs in the 5. Limitations and additional research
trajectory, the less time there is for benefit accrual and the manifestation
of any performance advantages due to sand quality (grain size). As ex­ Most of this analysis examines the physical and economic perfor­
pected, storm-induced efficiency losses were less noticeable beneath the mance of a relatively small difference (4 μm) in sand diameter. It is
surface, with subaqueous EBEVs averaging 10% less than subaerial important to note, however, that considerable time is required for
EBEVs, compared to 5% lower in the baseline scenario. This outcome quality differences to manifest at this range. Fifty-year trajectories are at
reflects an important finding from the geomorphic model – a substantial the outer limit of coastal restoration planning, and many programs set

9
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Fig. 8. Ecosystem break-even values (EBEV) for NS- and OCS-sourced projects at various boundaries, distances and cut-to-fill ratios (larger sand scenario).

the useful life of projects at 20 years. And while clear advantages are derived estimates, EBEVs do not indicate whether a project should be
evident for OCS sands on the larger end of the d50 spectrum (200 μm), built (or not). Instead, they serve as metrics of relative efficiency be­
less is known about the performance of projects sourced with tween and within project construction alternatives.
intermediate-sized sands (170–180 μm range). As a result, the current Finally, the integrated framework developed in this study could be
analysis constitutes a lower and upper bound of the economics of sand replicated for the examination of important challenges facing state and
performance. federal restoration programs. With updating, the model could be used to
Data for the development of the cost model were limited and highly address a number of pertinent questions, including: What are the eco­
variable. The use of contractor bids expanded a dataset of 22 candidate nomic tradeoffs of more regular maintenance versus a sacrificial approach to
projects into 93 useable observations. Though some might question the barrier island renourishment? How do large-scale dedicated dredging projects
potential redundancies introduced by this process, no viable alternatives compare (economically) to more frequent renourishment efforts via smaller
exist for this analysis. State and federal restoration managers face dredges? and What feasibility thresholds for restoration might exist given
similar limitations in the budgeting and allocation of funds for large- YOD projections for specific coastal barriers?
scale ecosystem restoration projects. Despite these limitations, the
management of large-scale restoration projects requires a systematic 6. Conclusions and policy implications
analysis of available costs and benefits.
As currently structured, this analysis does not attempt to assign Forty percent of the U.S. population resides in a contiguous band of
different ESVs for land above and beneath the water’s surface. The counties at increasing risk from coastal land loss and storms. This threat
mechanics for this delineation, however, are described within the is especially prominent in the MRDP, where nearly 2000 square miles of
mathematical framework of sections 2.2 and 2.3. The EBEV at subaerial coastal land have been lost in the past century, primarily due to hy­
and subaqueous boundaries (as used in this analysis) are derived values, drologic modification, navigation canals, sediment starvation, subsi­
intended to capture any areal effects resulting from the net transport of dence, sea-level rise, and climatic and geologic forcing. The outer
sand into or out of the proxy barrier system. Moreover, no attempt was boundary of this plain is bordered by a thin network of remnant delta
made to assign (or derive) ESVs based on functional differences associ­ lobes, the region’s barrier shorelines and islands.
ated with disturbance regulation (surge attenuation) or habitat. Since 1995, the Bureau of Ocean Energy Management has conveyed
Readers may question how the range of EBEVs estimated in the four access to 162 million yd3 of sand for 56 projects in eight states. These
scenarios compare to published ESV estimates for disturbance regula­ projects have restored 343 miles of U.S. shoreline and protected billions
tion and beach habitat. The insets highlighted in Figs. 5–8 feature EBEVs of dollars of coastal infrastructure and habitat. However, the availability
estimates from a low of $5459 (OCS, 200 μm, 1.05 CTF, 15 miles) to a and suitability of coastal sediments for dedicated dredging is a growing
high of $12,004 (NS, 156 μm, 5 miles, storm at y5). While this is a wide concern. This concern is especially prominent in the Gulf of Mexico
range, it is within the bounds of published estimates of ecosystem ser­ region and in Louisiana in particular, where an estimated 90 million yd3
vice value for barrier islands and shorelines.. It is important to reiterate, will be required in the next 50 years to address the state’s coastal land
however, the difference between EBEV estimates and NPV estimates. As loss crisis. This demand is reflected in Louisiana’s 2017 Coastal Master

10
R.H. Caffey et al. Journal of Environmental Management 319 (2022) 115730

Plan, which calls for more than $22 billion in expenditures for dedicated the Louisiana Coastal Protection and Restoration Authority for provision
dredging projects during this same time period. of the detailed project data required for model development.
Sediment availability and suitability for renourishment projects have
emerged as a major focus of restoration managers. Sediment for these Appendix A. Supplementary data
projects has historically come from one of two primary sources. Near­
shore (NS) sediment offers the economic advantage of proximity, but at Supplementary data to this article can be found online at https://doi.
the performance cost of smaller diameter sands with more organic fines. org/10.1016/j.jenvman.2022.115730.
Outer continental shelf (OCS) sediment offers better performing, larger
diameter sands with lower fines, but at higher transport costs due to References
distance. Previous economic comparisons of these sources have been
piecemeal, focusing almost entirely on terminal quantities of sediment Barbier, E.B., Hacker, S.D., Kennedy, C., Koch, E.W., Stier, A.C., Silliman, B.R., 2011. The
value of estuarine and coastal ecosystem services. Ecol. Monogr. 81, 169–193.
over a limited range of cost factors. https://doi.org/10.1890/10-1510.1.
Sediment resource maps estimate the total volume of Louisiana- Caffey, R.H., Wang, H., Petrolia, D.R., 2014. Trajectory economics: assessing the flow of
adjacent offshore surficial sand at nearly two billion cubic meters, ecosystem services from coastal restoration. Ecol. Econ. 100, 74–84. https://doi.org/
10.1016/j.ecolecon.2014.01.011.
with 60% of this material considered recoverable under current tech­ Caffey, R.H., Petrolia, D., Georgious, I.Y., Miner, M., Wang, H., Kime, B., 2020. Economic
nological and regulatory constraints. Until recently, access to these OCS and Geomorphic Comparison of Outer Continental Shelf Sand and Nearshore Sand
deposits was considered economically infeasible in comparison to lower for Coastal Restoration Projects. US Department of the Interior, Bureau of Ocean
Energy Management, New Orleans (LA), p. 54. Contract No.: M15AC00013. Report
quality, proximal sources. Traditional approaches to project evaluation No.: OCS Study BOEM 2020-035.
have centered on the value of sediment as a commodity, with a focus on Coastal Protection and Restoration Authority (CPRA) of Louisiana, 2017. Louisiana’s
placement cost. The findings of this study, however, provide economic Comprehensive Master Plan for a Sustainable Coast. Coastal Protection and
Restoration Authority of Louisiana, Baton Rouge, LA.
confirmation to the notion that grain-size matters - and that a more
Feagin, R.A., Smith, W.K., Psuty, N.P., Young, D.R., Martínez, M.L., Carter, G.A.,
comprehensive accounting of project performance (beyond the filled Lucas, K.L., Gibeaut, J.C., Gemma, J.N., Koske, R.E., 2010. Barrier islands: coupling
template) is required to maximize the return on coastal restoration anthropogenic stability with ecological sustainability. J. Coast Res. 26, 987–992.
spending. https://doi.org/10.2112/09-1185.1.
GecoServ, 2019. Gulf of Mexico Ecosystem Services Database. http://www.gecoserv.
This study has provided a decision support tool for managers seeking org/. (Accessed 25 August 2019).
more in-depth information on the economic tradeoffs of various alter­ Georgiou, I.Y., Weathers, H.D., Kulp, M.A., Miner, M., Reed, D.J., 2011. Interpretation of
natives for dedicated dredging. The extent to which the framework is Regional Sediment Transport Pathways Using Subsurface Geologic Data. Final
Report submitted to US Army Corps of Engineers. CESU Contract # W912HX-09-2-
utilized will ultimately depend on its utility - which in turn, hinges on 0027.
the availability (and validity) of performance data from existing pro­ Georgiou, I.Y., Kime, B.L., Miner, M.D., 2019. The effect of sediment properties on
jects. Long-term monitoring of project performance, however, has his­ restored Barrier Island morphodynamics: implications for using nearshore vs
offshore sediments. In: Coastal Sediments 2019: Proceedings of the 9th International
torically received less than 5% of restoration budgeting. Project Conference, pp. 230–234. https://doi.org/10.1142/9789811204487_0021.
monitoring must be prioritized to ensure adaptive management and Khalil, S.M., Finkl, C.W., 2009. Regional sediment management strategies for coastal
improved efficiency of restoration programs in Louisiana, the Gulf of restoration in Louisiana, USA. J. Coast Res. 1320–1324. Available at: https://www.
jstor.org/stable/25738003.
Mexico region, and the coastlines of the United States. Khalil, S.M., Freeman, A.M., 2015. Challenges of ecosystem restoration in Louisiana-
availability of sediment and its management. Proc. Int. Assoc. Hydrol. Sci. 367,
Credit author statement 455–462. https://doi.org/10.5194/piahs-367-455-2015.
Khalil, S.M., Finkl, C.W., Roberts, H.H., Raynie, R.C., 2010. New approaches to sediment
management on the inner continental shelf offshore coastal Louisiana. J. Coast Res.
Rex Caffey: Conceptualization, Methodology, Writing – original 26, 591–604. https://doi.org/10.2112/10A-00004.1.
draft, Writing – review & editing, Visualization. Daniel Petrolia: Merino, J., Aust, C., Caffey, R.H., 2011. Cost efficacy of wetland restoration projects in
Methodology, Formal analysis, Software, Writing – review & editing. coastal Louisiana. Wetlands 31, 367–375.
Miner, Michael D., Kulp, M.A., FitzGerald, D.M., Flocks, J.G., Weathers, H.D., 2009.
Ioannis Georgiou: Methodology, Investigation. Michael Miner: Delta lobe degradation and hurricane impacts governing large-scale coastal
Conceptualization, Supervision, Project administration. Hua Wang: behavior, South-central Louisiana, USA. Geo Mar. Lett. 29 (6), 441–453.
Data curation, Formal analysis, Investigation, Writing Reviewing and Morton, R.A., 2008. National Assessment of Shoreline Change: Part 1: Historical
Shoreline Changes and Associated Coastal Land Loss along the US Gulf of Mexico.
Editing. Brittany Kime: Data curation. Diane Publishing.
Nairn, R.B., Langendyk, S.K., Michel, J., 2004. Preliminary Infrastructure Stability Study,
Funding Offshore Louisiana. MMS 2005-043. U.S. Department of Interior, Minerals
Management Service, Gulf of Mexico Region, New Orleans, LA, OCS Study, p. 179 (p.
with appendices).
This work was supported by the US Department of the Interior, Bu­ NOAA, 2013. National Coastal Population Report: Population Trends from 1970 to 2020.
reau of Ocean Energy Management [grant number M15AC0013]. National Oceanic and Atmospheric Administration. Department of Commerce and U.
S. Census Bureau. Available at: https://aambpublicoceanservice.blob.core.windows.
net/oceanserviceprod/facts/coastal-population-report.pdf.
Declaration of competing interest Petrolia, D.R., Kim, T.G., 2009. What are barrier islands worth? Estimates of willingness
to pay for restoration. Mar. Resour. Econ. 24, 131–146. https://doi.org/10.1086/
mre.24.2.42731376.
The authors declare that they have no known competing financial
Richardson, L., Loomis, J., Kroeger, T., Casey, F., 2015. The role of benefit transfer in
interests or personal relationships that could have appeared to influence ecosystem service valuation. Ecol. Econ. 115, 51–58. https://doi.org/10.1016/j.
the work reported in this paper. ecolecon.2014.02.018.
van Rijn, L.C., 1984a. Sediment transport, part I: bed load transport. J. Hydraul. Eng.
110, 1431–1456. https://doi.org/10.1061/(ASCE)0733-9429, 1984)110:10(1431.
Acknowledgments van Rijn, L.C., 1984b. Sediment transport, part II: suspended load transport. J. Hydraul.
Eng. 110, 1613–1641. https://doi.org/10.1061/(ASCE)0733-9429, 1984)110:11
This research was sponsored by the U.S. Bureau of Ocean Energy (1613.
Wang, H., 2012 [Master’s Thesis. In: An Economic Assessment of Competing
Management under Cooperative Agreement M15AC00013 and admis­ Technologies for Coastal Restoration. Louisiana State University. https://digital
tered by the Louisiana State University Coastal Marine Institute. The commons.lsu.edu/gradschool_theses/3669.
authors would like to thank members of the project advisory panel and

11