Hein Et Al 2012

Article
Volume 13, Number 10

30 October 2012
Q10022, doi:10.1029/2012GC004286
ISSN: 1525-2027
Copper-nickel-rich, amalgamated ferromanganese

crust-nodule deposits from Shatsky Rise, NW Pacific
J. R. Hein and T. A. Conrad
U.S. Geological Survey, 400 Natural Bridges Drive, Santa Cruz, California 95060, USA
( jhein@usgs.gov)
M. Frank
GEOMAR, Helmholtz Centre for Ocean Research Kiel, Wischhofstrasse 1-3, DE-24148 Kiel,
Germany
M. Christl
Laboratory of Ion Beam Physics, ETH Zurich, Schafmattstrasse Co, CH-8093 Zurich, Switzerland
W. W. Sager
Department of Oceanography, Texas A&M University, College Station, Texas 77843, USA
[1] A unique set of ferromanganese crusts and nodules collected from Shatsky Rise (SR), NW Pacific, were
analyzed for mineralogical and chemical compositions, and dated using Be isotopes and cobalt chronometry.
The composition of these midlatitude, deep-water deposits is markedly different from northwest-equatorial
Pacific (PCZ) crusts, where most studies have been conducted. Crusts and nodules on SR formed in close
proximity and some nodule deposits were cemented and overgrown by crusts, forming amalgamated
deposits. The deep-water SR crusts are high in Cu, Li, and Th and low in Co, Te, and Tl concentrations
compared to PCZ crusts. Thorium concentrations (ppm) are especially striking with a high of 152 (mean 56),
compared to PCZ crusts (mean 11). The deep-water SR crusts show a diagenetic chemical signal, but not a
diagenetic mineralogy, which together constrain the redox conditions to early oxic diagenesis. Diagenetic
input to crusts is rare, but unequivocal in these deep-water crusts. Copper, Ni, and Li are strongly enriched
in SR deep-water deposits, but only in layers older than about 3.4 Ma. Diagenetic reactions in the sediment
and dissolution of biogenic calcite in the water column are the likely sources of these metals. The highest
concentrations of Li are in crust layers that formed near the calcite compensation depth. The onset of Ni,
Cu, and Li enrichment in the middle Miocene and cessation at about 3.4 Ma were accompanied by changes
in the deep-water environment, especially composition and flow rates of water masses, and location of the
carbonate compensation depth.
Components: 16,700 words, 8 figures, 9 tables.

Keywords: age dating; deep-water; diagenetic crusts; ferromanganese crusts and nodules; high Cu, Ni, Li, Th; mid-latitudes.
Index Terms: 0461 Biogeosciences: Metals; 1050 Geochemistry: Marine geochemistry (4835, 4845, 4850); 1065
Geochemistry: Major and trace element geochemistry.
Received 12 June 2012; Revised 10 September 2012; Accepted 26 September 2012; Published 30 October 2012.
Hein, J. R., T. A. Conrad, M. Frank, M. Christl, and W. W. Sager (2012), Copper-nickel-rich, amalgamated ferromanganese
crust-nodule deposits from Shatsky Rise, NW Pacific, Geochem. Geophys. Geosyst., 13, Q10022, doi:10.1029/
2012GC004286.
©2012. American Geophysical Union. All Rights Reserved. 1 of 23

Geochemistry 3
Geophysics
Geosystems G HEIN ET AL.: SHATSKY RISE Cu-RICH Fe-Mn CRUSTS 10.1029/2012GC004286
1. Introduction occurring at depths of 800–2500 m. The mean dry

bulk density of crusts is 1.3 g/cm3, the mean
[2] Most previous regional studies of ferromanga- porosity is 60%, and they are characterized by
nese oxide crusts (Fe-Mn crusts) utilized samples an extremely high mean surface area averaging
collected from relatively shallow water (above 325 m2/g [Hein et al., 2000]. Crusts generally grow
3000 m water depth) on seamounts and ridges at remarkably slow rates of 1–5 mm/Ma [e.g., Hein
located in the northwest-equatorial Pacific region, et al., 2000]. The extensive surface area and slow
within about 20 latitude of the equator in a corridor growth rates allow for the adsorption of many ele-
from Hawaii to the west Pacific trenches. That ments from seawater in high concentrations.
geographic focus was due to the occurrence of old [6] In general, the dominant controls for the
and large seamounts/guyots in that region and the concentration of metals in Fe-Mn crusts are the
high concentrations of cobalt, the metal presently of concentration of metals in seawater and their ratios,
greatest economic interest in those shallow-water colloid surface charge, types of complexing agents,
crusts [Hein et al., 2009]. That region is the Prime stability of surface complexes, surface oxidation
Crust Zone (PCZ) of greatest economic interest in reactions, specific-surface area, and growth rates.
the global ocean [Hein et al., 2009] and is therefore
More specifically, a first-order electrochemical
one of the areas used for comparisons with the
model proposes that positively charged species
Shatsky Rise crusts.
dissolved in seawater will adsorb onto the nega-
[3] We report here on the occurrence, composition, tively charged MnO2 surfaces and neutral and
age, and origin of a unique set of Fe-Mn crust, negatively charged species will adsorb onto the
nodule, and amalgamated crust-nodule deposits slightly positively charged surfaces of FeO(OH)
sampled from relatively deep water (mostly below [Koschinsky and Halbach, 1995; Koschinsky and
3000 m depth) at middle northern latitudes (32 – Hein, 2003] at the pH of seawater (8). Some
40 ) on Shatsky Rise (Figure 1). These Fe-Mn sorbed metals are oxidized to more stable states at
deposits from Shatsky Rise have chemical compo- the colloid surface thereby facilitating higher con-
sitions markedly different from those in the PCZ centrations than would otherwise be possible [e.g.,
and from the open-ocean Pacific and global crusts Murray and Dillard, 1979; Bidoglio et al., 1993].
in general.
3. Geomorphology and Geological
2. Background Setting
[7] Shatsky Rise is about 450 km wide and
[4] There have been four primary areas of focus in
the study of Fe-Mn crusts, their distribution, com- 1650 km long and resides on Late Jurassic and
position, and mechanisms of formation; their eco- Early Cretaceous lithosphere that becomes younger
nomic potential for Co, Mn, rare earth elements to the northeast [Nakanishi et al., 1999; Sager et al.,
(REEs), Ti, Ni, Mo, Pt, Zr, Te, Cu, W, and others; 1999]. Substantial outpourings of basalt occurred at
the use of Fe-Mn crusts as recorders of the past least three times forming three plateaus separated
70 million years (Ma) of oceanic and climatic by deep-water areas that are only a few hundred
history; and the role that the global complement meters above the regional baseline (Figure 1). These
of crusts (and nodules) play in the geochemical three plateaus were informally named TAMU, ORI,
balance of the oceans. This work predominantly and Shirshov by Sager et al. [1999]. TAMU plateau
is by far the largest in area and volume and was
focuses on the first of those four.
the first to form. A 39Ar-40Ar radiometric age of
[5] Fe-Mn crusts occur on hard-rock substrates and 144.6 0.8 Ma was determined from massive flows
nodules on sediment-covered abyssal plains and cored at IODP Site 1213 on the south flank of
plateaus throughout the global ocean. Crusts form TAMU [Mahoney et al., 2005]. Two other volcanic
at the seafloor on the flanks and summits of sea- features occur in close proximity to Shatsky Rise.
mounts and ridges where the rocks have been swept Papanin Ridge forms an apparent northeast exten-
clean of sediments for millions of years [Hein et al., sion of the Rise but is separated from it by a deep
2000]. Crusts can be up to 260-mm thick and form trough; and the Ojin Rise Seamounts, which partly
pavements on rock outcrops by way of hydro- sit on a swell and form a chain trending east-
genetic precipitation from cold ambient bottom northeast away from Shirshov plateau (Figure 1).
waters. Crusts form at water depths between 400– Both these volcanic structures may be connected
7000 m, with the thickest and most metal-rich crusts with Hess Rise to the east.
2 of 23
Geochemistry 3
Geophysics
Figure 1. Location of Shatsky Rise (inset), the three plateaus that make up the Rise, and dredge sites D2–D14 from
which Fe-Mn deposits were collected.
[8] The regional seafloor ranges in depth from that have been analyzed from Shatsky Rise have
about 5500 to 6000 m and the entire approximately MORB-like and not OIB-like geochemical and
480,000 km2 area of Shatsky Rise can be outlined petrographic characteristics [Tatsumi et al., 1998;
by the 5000 m isobath, with depths as shallow as Mahoney et al., 2005]. These new data are consis-
about 2000 m, but typically are greater than 3000 m tent with either a perisphere or meteorite-impact
[Sager et al., 1999]. Shatsky Rise is covered at its origin [Mahoney et al., 2005; Sager, 2005]. Ojin
summits by up to 1.2 km of mostly Cretaceous Rise Seamounts may have formed somewhat later
pelagic carbonates, in contrast to less than 500 m of from a different magmatic source.
sediment on the surrounding abyssal plain. Sedi-
[10] Although the principal plateaus of Shatsky
ments on the flanks of Shatsky Rise are thin with
Rise currently lie at latitudes from 30 to 40 N,
comparable thickness to surrounding abyssal sedi-
paleomagnetic data indicate that the plateau was
ments [Sager et al., 1999]. Ridges and seamounts
near the equator during its formation and subse-
on Shatsky Rise are clear of sediment in many
quently drifted northward by 35 [Sager, 2006].
places and these sediment-free areas are the loca-
Northward motion of the Pacific plate has been
tions of Fe-Mn crust formation.
28 latitude since 70 Ma ago, but only 10 since
[9] Shatsky Rise formed rapidly [Sager and Han, 30 Ma [Beaman et al., 2007]. This indicates that
1993] although episodically during three stages during the Neogene, the environmental conditions
that formed the three plateaus [Mahoney et al., related to geographic location for Shatsky Rise
2005]. Formation occurred predominantly near an dredge sites should have changed little.
oceanic ridge-ridge-ridge triple junction [Nakanishi
et al., 1999]. It was proposed that the Rise formed 4. Samples and Methods
from a mantle plume head and decompressional
melting [Sager et al., 1999], with Papanin Ridge [11] Samples were selected from dredge hauls
having formed from the mantle plume tail. How- stored at Scripps Institution of Oceanography that
ever, subsequent work showed that the few rocks
3 of 23
Geochemistry 3
Geophysics
Figure 2. Bathymetric maps with detailed location of dredge sites shown in Figure 1.
were collected in 1994 during cruise TN037 aboard of the summit plateau (Figures 1 and 2 and
the R.V. Thomas G. Thompson. Dredges D12, D13, Table 1). Dredge D11 was collected from Cooper-
and D14 were collected on TAMU plateau: D13 ation seamount, a twinned-peak basaltic volcano
and D14 are from the 1500 m-high Toronto ridge located within a basin that separates TAMU and
(900 m of which is buried beneath sediment) ORI plateaus [Sager et al., 1999]. Dredges D8–D10
located on the northwest flank of TAMU, and D12 were taken along the flanks of ORI plateau and
on a seamount called Sponge mountain, just south D6 and D7 along the flanks of Shirshov plateau
Table 1. Location of Cruise TN037 Dredges Containing Analyzed Fe-Mn Deposits

Water Max Crust Av Crust
Depth Thicknessa Thicknessa
Dredge Latitude (N) Longitude (E) (m) Location (mm) (mm) Substrate/Nucleus
D2 37 37.578′ 165 27.519′ 4833 N Victoria SMb 112 85 Breccia, CFA cement
– – – – – 58 50 Limestone nucleus
D3 39 51.914′ 163 53.660′ 4227 N Arm volcano 75 65 No substrate
– – – – – 59 56 No nucleus
D4 39 50.264′ 163 54.789′ 4000 Earthwatch SM 118 90 Breccia, CFA + Mn cement
D6 37 48.292′ 163 07.521′ 3835 E flank Shirshov 116 90 Nodule fragment nucleus
D7 37 23.500′ 162 54.906′ 3759 S flank Shirshov 136 110 No substrate
– – – – – 90 44 Nodule fragment nucleus
D8 37 23.604′ 158 50.294′ 4806 N flank ORI 10 5 Sandy claystone w/barite
– – – – – 82 72 Claystone nucleus
D9 36 31.150′ 159 11.619′ 3900 E flank ORI 55 30 Altered basalt, breccia
D10 36 08.070′ 158 22.454′ 3474 Central ORI 34 29 Breccia, phosphorite
D11 35 15.700′ 158 57.050′ 3760 Cooperation SM 68 60 Altered basalt, hyaloclastite
D12 32 12.927′ 158 36.210′ 3016 Sponge Mtn 40 29 Phosphorite, breccia
D13 32 43.366′ 158 15.147′ 2357 Toronto Ridge 37 22 Phosphorite, some Mn rich
D14 32 40.226′ 158 04.866′ 2155 Toronto Ridge 110 58 Phosphorite, breccia
a
For nodules (dashes), maximum and minimum diameters are given.
b
SM = seamount.
4 of 23
Geochemistry 3
Geophysics
(Figures 1 and 2). Dredges D3 and D4 taken on (0.5 ppm), and Sn (2 ppm). Some crusts were split
seamounts (N Arm and Earthwatch) located just into two or more (up to 5) layers defined by well-
south of a trough that separates Shirshov plateau developed textural differences. Bulk crust samples
from Papanin Ridge. Dredge D2 was taken on the were also analyzed for those that were split into
lower flank of one of the approximately 80 sea- layers. All depths within Fe-Mn crusts are given in
mounts that constitute the Ojin Rise seamount millimeters from the crust surface that was in con-
province (Figures 1 and 2). tact with seawater.
[12] X-ray diffraction mineralogy was completed [15] REE plots are normalized to both chondrites
using a Philips diffractometer with Cuka radiation [Anders and Grevesse, 1989] and shale (PAAS,
and graphite monochromator. Semiquantitative Post-Archean Australian Shale [McLennan, 1989]).
mineral percentages were determined by using peak The Ce anomaly was calculated as Ce* = 2Ce/
intensities and weighting factors relative to quartz (La + Pr) for PAAS-normalized values.
set as 1 [Cook et al., 1975; Hein et al., 1988]. The
[16] Four subsamples forming a depth profile were
detection limit for each mineral falls between about
taken from each of four select Fe-Mn crusts with a
0.2 and 1.0%, except for the manganese minerals,
dentist drill to determine average growth rates and
for which the limits are greater, probably as much
ages using the ratio between cosmogenic 10Be and
as 10% for d-MnO2.
stable 9Be. The 10Be concentrations were deter-
[13] Abundances of the 10 major oxides and Ba, Cr, mined by isotopic dilution using accelerator mass
Nb, Rb, Sr, Y, Zr, Th, and U in substrate rocks were spectrometry (AMS). A precisely weighed amount
determined by X-ray fluorescence spectrometry of a 9Be carrier was added to weighed aliquots of
using a borate fused disk (XRF); Fe(II) by colori- the sample powders and the samples were dissolved
metric titration; CO2 by coulometric titration; H2O+ in 6 M HCl. The sample solutions were further
by water evolved at 925 C as determined coulo- purified and prepared for AMS analyses following
metrically by Karl-Fischer titration; H2O by a previously established method [Frank et al.,
sample weight difference after heating to 110 C for 1994]. The AMS measurements were performed
more than one hour; F and Cl by specific-ion at the 0.5 MV AMS facility “TANDY” [Müller
electrode, S and C by combustion in an induction et al., 2010], ETH Zürich, Switzerland. The mea-
furnace and analysis of the generated SO2 and CO2 sured 10Be/9Be ratios were normalized to the sec-
by infrared spectroscopy. ondary ETH Zürich S2007N with a nominal value
of 10Be/Be = 28.1 10 12 [Kubik and Christl,
[14] For Fe-Mn crusts and nodules, the concentra-
2010]. The 1s statistical uncertainties of the 10Be
tions of the 10 major elements were determined by
measurements were between 3% and 5% taking
XRF on a borate fused disk representing the aver-
into account both the counting statistics of the 10Be
age composition of the entire crust sample; B, Ba,
“events” and the reproducibility of repeated mea-
Cr, Cu, Li, Ni, Sr, V, Zn, and Zr were determined
surements, which were performed for each sample,
by inductively coupled plasma spectrometry (ICP)
standard, and blank. The newly revised half-life of
and Ag, As, Be, Bi, Br, Cd, Ce, Co, Ge, Hf, In, Mo, 10
Be of 1.387 Myr was applied [Korschinek et al.,
Nb, Pb, Rb, Sb, Sc, Se, Sn, Te, Th, Tl, U, W, and
2010; Chmeleff et al., 2010].
Y, by ICP-mass spectrometry using a 4-acid digest
(ICP-MS); REEs by ICP-MS using a Li-metaborate [17] The natural authigenic 9Be concentration of
fused disk; platinum group elements (PGEs: Ir, Pd, the Fe-Mn crust samples was acquired at the
Pt, Rh, Ru) and Au by Ni-sulfide fire assay and ICPMS-Laboratory of the Institute of Geosciences,
ICP-MS; Te and Se were also determined by University of Kiel, using an AGILENT 7500cs
hydride generation and graphite furnace atomic- ICP-MS instrument following established standard
absorption spectrometry (AAS); Hg by cold vapor procedures [Garbe-Schönberg, 1993]. The mea-
AAS; S by combustion and infrared spectroscopy; surements were performed on an aliquot of the same
and Cl by specific-ion electrode. Analytical preci- solution as used for 10Be measurement in order to
sion is better than 1.5% for all elements except Cl, guarantee that the determined natural 10Be/9Be
which is better than 5.0%, and PGEs (better than represents a true seawater ratio, which for the sur-
10%). Elements that have more than 30% of the face sample is consistent with the modern seawater
values near or below the detection limit (in paren- ratio. For ratios deeper in the crusts, a seawater
theses) include Cs (3 ppm), Se (5 ppm), Au (2 ppb), signal is indicated by a consistent decrease that
and Ir (2 ppb); elements with all values below follows an exponential decay, which means that the
detection limit are Ag (0.2 ppm), Ge (10 ppm), In likelihood of a seawater signal is very high; in
5 of 23
Geochemistry 3
Geophysics
addition, the REE patterns are typical for hydro- pavements; Figure 3), which grade with depth
genetic crusts. Repeat measurements of samples into nodule conglomerate; the cement is Fe-Mn
processed separately through the leaching proce- oxides, which also cement some breccia substrate
dure indicate a 1s reproducibility of 5%. rocks. In many dredge hauls, both crusts and nodules
were collected indicating the close proximity of rock
[18] The age of each crust below the upper 20 mm
outcrops and sediment-covered seabed. Nodules gen-
dated using Be isotopes was determined using an
erally have a small nucleus of an older nodule frag-
empirically derived cobalt chronometer method
ment, or less commonly a rock fragment (abyssal-type
(growth rate GR = 0.68/(Con)1.67, where Con = Co
nodules [Hein et al., 2000]). Deep-ocean amalgam-
(50/Fe + Mn), with metals in wt. % [Manheim and
ated deposits have not been previously described and
Lane-Bostwick, 1988]), and by extrapolation of
the close occurrence of crusts and abyssal-type nodules
the deepest layer rate determined by Be isotope;
on seamounts and ridges is not common.
the occurrence of growth hiatuses and changes
in growth rates in the parts of the crusts not covered [21] Crusts for the most part are quite thick, but
by 10Be/9Be data cannot be excluded (see Frank vary from 5 to 136 mm (Table 1); 260 mm is the
et al. [1999] for discussion). Frank et al. [1999] thickest crust collected to date from the global
adjusted the empirical Co growth-rate relationship ocean [Hein et al., 2000]. The thickest Shatsky Rise
to match the available 10Be/9Be data and then used crust was recovered from dredge haul D7 taken on
this adjusted relationship to extrapolate beyond the the flank of Shirshov plateau. Thick crusts are
time covered by 10Be/9Be data. Here, we use the layered, with columnar, massive, mottled, laminar,
two techniques independently and so differences and/or botryoidal laminae, typical of hydrogenetic
are to be expected. Fe-Mn crusts. Small rock and mineral fragments
and isolated nodules occur within some crusts
[19] The Pearson correlation coefficient was used to
attesting to the dynamic seabed environment where
calculate correlation coefficient matrices for the
the crusts formed. Surface textures of crusts are
chemical data. Q-mode factor analysis was used to
generally smooth, indicating current abrasion of
examine inter-element relationships and to identify
surfaces that were once botryoidal. Carbonate ooze
common groups of elements referred to as factors.
adhered to the surface of some nodules and red clay
On the basis of X-ray diffraction mineralogy and
to some crusts during the dredging process.
element correlations, we interpret each factor to
represent a particular mineral or phase in the Fe-Mn [22] Fe-Mn crust substrate rock types also vary
crusts and elements in that factor to be associated widely and include altered basalt, breccia, hyaloclas-
with or contained in that mineral or phase. This tite, limestone, mudstone, and phosphorite composed
links mineralogy that reflects environmental con- of carbonate fluorapatite (CFA). Pure phosphorite is
ditions to element concentrations that in part reflect represented by replaced foraminiferal limestone and
different sources. For Q-mode factor analysis, each phosphatic rocks include CFA-cemented breccia
variable percentage was scaled to the percent of and hyaloclastite and partly replaced reef limestone.
the maximum value before the values were row- These are typical substrate rocks for Fe-Mn crusts
normalized and cosine-theta coefficients calculated. collected from seamounts elsewhere in the Pacific.
Factors were derived from orthogonal rotations
of principal component eigenvectors using the 5.1. Composition of Selected Substrate
Varimax method [Klovan and Imbrie, 1971]. All Rocks
communalities, an index of the efficiency of a
reduced set of factors to account for the original [23] Six representative substrate samples of fine-
variance, are ≥0.90. Low factor scores between grained and altered rocks were selected and ana-
0.18 and +0.18 were discarded because they lyzed to determine their mineralogy and chemical
are not statistically significant. composition (Table 2). Phosphorite is common in
dredge hauls D12, D13, and D14. These are typical
seamount-type deep-water phosphorites like those
5. Results collected from many locations further south in the
PCZ [Hein et al., 1993]. These three dredge sites
[20] A wide variety of Fe-Mn morphological
are the shallowest-water sites (Table 1) and the
types were recovered during cruise TN037, includ-
southernmost of those sampled during TN037. The
ing Fe-Mn-oxide pavement, Fe-Mn-oxide-coated
other substrate rock samples analyzed are composed
cobbles, Fe-Mn nodules, and amalgamated crust-
of volcanogenic minerals and their alteration pro-
nodule deposits (Fe-Mn nodules cemented into
ducts (Table 2), which are highly altered with high
6 of 23
Geochemistry 3
Geophysics
Figure 3. Photographs of a cross-section of Fe-Mn crusts, nodules, and amalgamated crust-nodule deposits. Arrows
point out nodules that are overlain by Fe-Mn crust in two samples. (a) Fe-Mn nodule D7–1N showing the interior cut
face and the exterior texture. (b) Fe-Mn nodule D6–1N. (c) Fe-Mn crust sample D4–1, 90–118 mm thick Fe-Mn crust
with a few Fe-Mn nodules, indicated by white arrows, in the crust on a breccia substrate. (d) Sample D7–07 showing
nodules, indicated by white arrows, in the lower portion of the crusts cemented with Fe-Mn oxides and overlain by a
30 mm thick Fe-Mn crust. (e) Sample D2-03-02 is a 70–80 mm thick Fe-Mn crust with breccia substrate.
water contents and little to no reduced iron. Barite any other TN037 Fe-Mn crust. The lower layers
occurs in a sample of strongly sheared pinkish- of thick crusts from more equatorial regions are
brown claystone (D8–1A2). There is no indication commonly phosphatized [Hein et al., 1993].
of hydrothermal alteration of these rocks.
5.3. Fe-Mn Crust and Nodule
Geochemistry
5.2. Fe-Mn Crust and Nodule Mineralogy
[25] Fifty-one samples were analyzed for chemical
[24] All Fe-Mn samples are predominantly d-MnO2 composition including 21 bulk crusts, 25 crust
(vernadite), which has only two X-ray reflections layers (2 of which are nodular layers), and 5 bulk
at about 2.42 Å and 1.41 Å. One crust layer may nodules (Table 3). Chemical data are presented on
contain todorokite, but generally diagenetic and a hygroscopic water-free basis (0% H2O ). The
hydrothermal manganese minerals (7 and 10 Å mean Fe/Mn ratio is 0.9 (range 0.5–2.0; Table 3),
manganates) are below the detection limit of about within the range of most hydrogenetic Fe-Mn crusts
5%. The d-MnO2 mineralogy of the Fe-Mn nodules and only slightly higher than the 0.7 mean ratio of
indicates that they are also hydrogenetic without a PCZ crusts. Si/Al ratios (mean 4.6) indicate that Si
significant diagenetic component. The main iron is enriched relative to Al in Fe-Mn crusts and
phase is X-ray amorphous. Accessory minerals nodules compared to substrate rocks. That excess
include eolian quartz, plagioclase, and illite carried silica most likely reflects biogenic silica and eolian
by westerly winds from Asia [Leinen et al., 1994], quartz.
and detrital plagioclase and smectite from local
sources; there is no local source for quartz and [26] Ni (max 0.84%), Co (max. 0.52%), and Cu
illite. Some samples also contain phillipsite or (max. 0.50%) are the metals of greatest eco-
amphibole. The innermost layer (69–97 mm, nomic interest, along with Mn; individual nodule
adjacent to substrate) of a thick crust from dredge combined concentrations reach up to 1.33 wt. %
haul D14 contains CFA, which is not found in (D6-1N-A) and individual crust layers up to 1.57%
7 of 23
Geochemistry 3
Geophysics
Table 2. Chemical Composition of Select Fe-Mn Crust Substrate Rocksa

D8-1A2 D9-12D D10-11B D12-2B2 D13-11A2 D13-15B2
SiO2 (wt%) 49.9 47.6 47.3 3.66 1.30 10.1
Al2O3 15.7 17.4 14.9 0.66 0.39 4.53
FeO <0.1 0.1 <0.1 <0.1 0.1 0.2
Fe2O3 6.79 14.5 14.0 0.37 1.28 11.9
MgO 4.06 1.77 4.58 0.30 0.34 1.19
CaO 1.72 4.97 1.01 51.1 52.3 29.5
Na2O 1.65 2.97 1.84 0.84 0.67 1.06
K2O 3.64 2.19 5.61 0.14 0.05 0.73
TiO2 0.586 1.93 1.11 0.049 0.034 0.917
P2O5 0.88 0.25 0.53 31.3 32.1 19.1
MnO 1.87 0.25 0.79 0.02 0.77 8.49
LOI 8.52 3.60 5.45 10.7 3.30 10.3
Total 95.4 97.5 97.3 99.2 92.8 98.3
H2O+ 6.8 3.6 6.4 1.9 2.5 6.2

H2O 6.2 5.2 7.0 0.3 0.8 7.9
CO2 0.16 0.02 0.07 4.87 5.18 2.67
Ctotal – – – 1.58 – –
Stotal – – – 0.67 – –
F – – – 2.97 – –
Ba (ppm) >10000 193 812 <20 <20 2390

Cl – – – 871 – –
Cr <0.01 0.03 0.04 <0.01 <0.01 <0.01
Nb 3 8 3 <2 <2 2
Rb 115 37 71 <2 2 19
Sr 500 184 111 976 1120 1170
Y 169 36 63 233 937 822
Zr 226 108 187 27 28 186
Th – – – <2 – –
U – – – 5 – –
CaO/P2O5 1.95 19.9 1.91 1.63 1.63 1.54
Rock type claystone basalt basalt phosphorite phosphorite breccia

a
LOI = loss on ignition at 925 C. Dash = not analyzed.
(D4–1E) for those three metals, which are below the example, thin bulk crusts plus the thin outermost
2.0 wt.% commonly considered the potentially layer (<25 mm) of thicker crusts average 923 ppm
economic cut-off grade for nodules. The mean Cu Cu and 2755 ppm Ni, whereas crust layers below
content of the crusts analyzed here is about 300– 30 mm average 2281 ppm Cu and 4705 ppm Ni.
400 ppm higher than regional means elsewhere in Lithium and especially Th are also strongly enri-
the global ocean although the mean Ni content falls ched in the same older crust layers as Cu and Ni and
within the range of large regional areas (Figure 4) also compared to other crusts from open-ocean
[Hein et al., 2000; Hein and Koschinsky, 2012]. The Pacific sites (Figure 4).
mean Cu/Co ratio is 0.65 for all samples, 0.63 for
[27] Phosphorus averages about 0.54% for all the
crust layers, 0.49 for bulk crusts, and 2.0 for the five
data combined, typical of most non-phosphatized
bulk nodules (Tables 4–6). The high mean ratio for
Fe-Mn crusts. Only one crust (D14–16) is phos-
the nodules is identical to the mean ratio for the
phatized and has 3.72% P in the lowermost layer
seven crust layers with high Cu concentrations
(69–97 mm; Table 3); no nodules are phospha-
(>0.20%). The same trend is noted for Ni/Co ratios,
tized. The PGEs are generally low compared to
with bulk crust ratios averaging 1.3 and nodules 3.0.
PCZ samples; bulk crust D14–16B has the highest
The highest Cu and Ni layers in crusts are always
concentrations of PGEs, with 689 ppb Pt. Hg is
middle or lower layers, such as 46–68 mm for D3–1,
very high (69–97 ppb) in five samples as is Au
56–92 mm for D4–1, and 31–78 mm for D7–1.
(54–66 ppb) compared to PCZ samples, which
No near surface layers or thin crusts are strongly
enriched in Cu (>0.20%) or Ni (>0.45%). For
8 of 23
Geochemistry 3
Geophysics
Table 3 (Sample). Hygroscopic Water-Free (0% H2O ) Composition of Fe-Mn Crusts, Crust Layers in mm From
Crust Surface, and Nodulesa [The full Table 3 is available in the HTML version of this article]
D2-3-1A D2-3-1B D2-3-1C D2-3-1D D2-3-2A D2-3N-A D3-1A D3-1B D3-1C D3-2N-A D4-1A
Fe (wt%) 21.6 24.4 22.6 19.8 23.4 20.6 14.0 17.7 19.7 20.4 26.8
Mn 23.9 21.1 28.8 29.3 28.9 20.6 22.8 21.4 25.1 22.7 24.4
Fe/Mn 0.9 1.2 0.8 0.7 0.8 1.0 0.6 0.8 0.8 0.9 1.1
Si 9.20 10.5 7.54 8.34 8.78 11.3 15.9 15.3 11.0 11.5 8.83
Al 2.45 2.05 1.63 2.10 1.86 2.40 2.65 2.55 2.07 1.84 2.26
Na 1.75 1.72 1.87 1.76 1.88 1.84 2.17 2.32 1.88 2.04 1.61
K 0.93 0.94 0.82 0.73 0.96 1.02 1.35 1.11 1.20 0.99 1.18
Mg 1.49 1.45 1.57 1.71 1.62 1.45 2.07 1.60 2.05 1.50 1.66
Ca 2.46 2.11 2.64 2.91 2.65 1.94 1.70 1.91 1.97 1.96 2.41
Ti 1.17 0.967 1.36 1.38 1.34 0.917 0.424 0.534 0.661 0.761 1.51
P 0.41 0.42 0.41 0.39 0.42 0.32 0.17 0.26 0.27 0.28 0.46
S 0.14 0.12 0.14 0.11 0.09 0.07 0.09 0.11 0.10 0.08 0.15
Cl 1.01 1.01 1.12 1.21 0.991 0.956 0.909 1.07 1.12 0.942 0.689
As (ppm) 206 244 225 185 215 175 89 131 139 141 225
B 654 639 632 348 547 217 289 265 537 251 465
Ba 1681 1616 1877 2032 1904 1417 3848 1631 3233 1618 2352
Be 11 30 9.9 9.5 8.4 7.4 4.4 5.6 6.7 6.6 11
Bi 39 24 38 44 37 20 10 12 19 22 35
Br 24 24 25 25 21 22 22 25 23 22 17
Cd 4.3 3.5 4.0 5.2 4.0 3.9 8.8 5.9 7.1 5.4 4.2
Co 2365 1821 2872 3579 2915 2125 1263 1312 1599 1618 2121
Cr 23 34 13 42 13 7 7 61 25 62 37
Cu 1144 748 1220 1808 1188 1429 3836 2305 3257 2289 1517
Hf 4 <1 3 7 1 2 5 4 10 3 5
Li 7 6 6 9 5 15 21 18 25 18 8
Mo 537 604 662 650 662 440 442 392 499 342 591
Nb 62 47 65 77 67 44 13 20 29 30 60
Ni 2365 1749 2720 3616 2759 2695 5801 4551 5325 4358 2404
Pb 1273 1327 1537 1259 1503 1359 744 1061 1202 1445 1607
Rb 23 23 13 9 14 20 32 24 23 20 15
Sb 35 40 38 34 38 33 28 32 44 39 54
Sc 11 12 10 11 11 14 19 15 15 14 18
Se <5 19 40 <5 25 19 16 17 <5 8 13
Sr 1261 1291 1474 1372 1438 1196 895 998 1165 1237 1491
Te 22 13 22 37 26 16 4.6 6.1 8.9 11 44
Th 127 114 152 128 148 113 29 41 48 10 68
Tl 131 86 132 160 135 117 74 59 80 70 104
U 9.5 12 16 10 11 11 5.4 6.7 4.3 9.7 11
V 681 706 670 574 588 504 335 417 459 461 672
W 65 69 72 79 70 45 34 30 49 30 62
Zn 485 485 520 605 567 602 908 788 846 837 648
Zr 685 669 678 756 699 560 457 498 614 543 752
Au (ppb) 36 12 57 31 40 <2 44 58 12 6 36
Hg 14 25 14 15 12 74 29 31 29 53 10
Ir 5 2 3 5 – – – – – 2 5
Pd 5 5 8 7 – – 9 – – 7 8
Pt 150 75 136 287 – – 124 – – 127 366
Rh 13 8 14 21 – – 6 – – 8 22
Ru 12 10 13 15 – – 7 – – 9 13
Interval 0–112 0–37 37–82 82–112 0–70 0–54 0–68 0–46 46–68 0–57 0–118
Type bulk layer layer layer bulk nodule bulk layer layer nodule bulk
a
Dash = not analyzed.
may indicate a hydrothermal component in Shatsky associated elements decrease in the lower layer
Fe-Mn crusts [Hein et al., 2005]. of the two layers of crusts D3–1 and D9–12. Mn,
Mn-associated elements, and PGEs generally
[28] Two or more layers were analyzed for eight
decrease from the substrate to the crust surface for
Fe-Mn crusts. Fe, Mn, and their respective associated
crusts D2–3 (112 mm thick) and D11–2 (68 mm
elements (see below) increase, and the aluminosilicate-
9 of 23
Geochemistry 3
Geophysics
Figure 4. Comparison of Shatsky Rise mean bulk Fe-Mn crust composition for selected elements with mean crust
values for Atlantic, Indian, and open-ocean Pacific crusts.
thick; 3 layers each), whereas Fe and associated ele- 5.4. Yttrium and Rare Earth Elements
ments show the opposite trend. In contrast, most
[30] Total REEs plus Y range from very high
elements show uniform concentrations through the
(0.51%; crust D2-3-1A; 7.5% are heavy REEs) to
three layers of the relatively thin crust D10–11
quite low (0.081%; crust D3–1A; 17.6% are heavy
(25 mm thick). The other Fe-Mn crusts show more
REEs), with Ce concentrations up to 0.43% (crust
complex patterns determined by the anomalous
layer D2-3-1D; Table 7). Only Ce among the REEs
composition of one of the layers; for example, layer 4
is higher in Shatsky Rise samples compared to PCZ
of the 5 in sample D4–1 (118 mm) is high in alu-
samples. Yttrium and the REEs decrease in the
minosilicate elements and very low in Fe and layer
upper layer of the two layers of crusts D3–1 and
5 is also high in aluminosilicate elements, but very
D9–12. The REEs increase (except Ce which
low in Mn, consequently, other elements either
decreases) from the substrate to the crust surface for
increase toward or away from those layers. The
crusts D2–3 (3 layers) and D11–2 (3 layers). In
phosphatized layer (lowermost of 4 layers) in crust
contrast, the REEs increase from the bottom and
D14–16 (110 mm thick) is another example of an
surface layers to the central layer through the three
anomalous layer.
layers of sample D7–1C and sample D10–11,
[29] A ternary plot of Fe:Mn:(Co + Ni + Cu) 10 except Ce in D10, which shows the opposite trend.
shows that about half the sample data fall within The other crusts show more complex patterns
the lower part of the open-ocean hydrogenetic crust determined by the anomalous composition of one
field (field B in Figure 5). The remainder of the of the layers; for example, layer 4 of the 5 layers of
data in the ternary plot trend toward the hydro- sample D4–1 is high in aluminosilicate elements
thermal field (field C in Figure 5) or fall between and very low in Fe, consequently, the REEs increase
the hydrogenetic and diagenetic fields (field A in away from that layer and also decrease from layer 2
Figure 5). to the top layer, which has the highest concentra-
tions for that crust, except for Ce. In four-layer crust
10 of 23
Geochemistry 3
Geophysics
Table 4. Statistics for 51 Bulk Fe-Mn Crusts, Crust Table 5. Statistics for 21 Bulk Fe-Mn Crusts (Data
Layers, and Nodules (Data From Table 3) From Table 3)
N Mean Median SDa Minb Maxc N Mean Median SDa Minb Maxc
Fe (wt. %) 51 22.1 22.9 4.28 12.2 28.3 Fe (wt. %) 21 22.5 22.9 3.69 14.0 26.8
Mn 51 25.0 24.7 4.69 7.84 35.8 Mn 21 24.3 24.4 4.95 7.84 31.3
Fe/Mn 51 0.9 0.9 0.9 1.5 0.8 Fe/Mn 21 0.9 0.9 0.7 1.8 0.9
Si 51 9.51 9.00 3.08 4.55 20.1 Si 21 9.52 8.83 3.48 4.85 20.1
Na 51 1.92 1.88 0.22 1.56 2.96 Na 21 1.93 1.86 0.29 1.56 2.96
Mg 51 1.64 1.60 0.23 1.12 2.28 Mg 21 1.61 1.53 0.22 1.12 2.07
Al 51 2.05 1.88 0.87 0.72 5.20 Al 21 1.98 1.86 0.93 0.87 5.20
P 51 0.54 0.43 0.54 0.16 3.72 P 21 0.54 0.46 0.43 0.17 2.35
K 51 1.02 0.94 0.36 0.61 2.47 K 21 0.98 0.93 0.26 0.62 1.48
Ca 51 2.62 2.33 1.39 1.34 10.2 Ca 21 2.59 2.41 1.18 1.34 7.44
Ti 51 1.001 0.942 0.316 0.424 1.938 Ti 21 0.997 0.962 0.285 0.424 1.508
S 51 0.14 0.14 0.05 0.05 0.28 S 21 0.15 0.14 0.05 0.08 0.27
LOI 51 20.5 20.6 2.73 13.5 26.1 LOI 21 20.3 21.0 2.90 13.5 24.5
As (ppm) 51 200 206 66 56 334 As (ppm) 21 208 212 60 86 320
B 51 347 271 169 140 818 B 21 320 238 158 169 654
Ba 51 2469 2005 1315 1417 9214 Ba 21 2604 1955 1692 1418 9214
Be 51 7.5 7.0 3.7 2.9 29.8 Be 21 6.9 6.9 1.9 2.9 11.3
Bi 51 33.3 30.2 16.4 8.8 82.8 Bi 21 32.7 33.9 14.1 8.8 59.2
Br 51 24 23 8 13 59 Br 21 25 23 10 16 59
Cl 51 10321 10133 3059 3741 24916 Cl 21 11064 10060 3879 6179 24916
Co 51 2441 2121 1056 549 5186 Co 21 2667 2576 1095 549 4988
Cr 51 37 39 21 5 82 Cr 21 33 32 20 5 82
Cu 51 1591 1220 1098 318 4983 Cu 21 1294 1181 779 399 3836
Ga 51 14 14 5 8 35 Ga 21 13 11 4 8 21
Li 51 15 9 14 1 70 Li 21 12 8 10 3 37
Mo 51 534 537 165 67 858 Mo 21 523 537 166 67 787
Nb 51 42 40 16 13 77 Nb 21 43 43 14 13 67
Ni 51 3625 3425 1612 840 8417 Ni 21 3346 3203 1357 840 5801
Pb 51 1602 1596 486 721 2677 Pb 21 1726 1694 546 721 2677
Rb 51 19 18 8 5 44 Rb 21 19 17 9 5 39
Sb 51 43.3 44.4 9.2 10.4 65.2 Sb 21 43.1 46.0 10.7 10.4 57.0
Sc 51 12.0 11.1 4.1 4.6 27.5 Sc 21 11.7 10.9 3.8 5.7 19.3
Sr 51 1454 1472 324 705 1959 Sr 21 1526 1609 348 705 1944
Te 51 25.0 20.3 17.6 3.1 74.9 Te 21 25.1 24.2 13.2 3.1 54.7
Th 51 55.8 45.8 31.3 10.2 152 Th 21 54.8 46.9 30.2 22.3 148
Tl 51 91.3 82.8 57.9 16.7 285 Tl 21 80.0 80.6 45.2 20.8 180
U 51 10.7 10.88 3.86 2.7 21.7 U 21 10.8 10.9 3.0 4.5 16.3
V 51 539 549 130 251 771 V 21 545 559 125 251 726
W 51 57 57 19 9 94 W 21 55 54 17 9 87
Y 51 178 169 83 76 561 Y 21 176 171 82 76 468
Zn 51 699 687 141 332 999 Zn 21 680 684 148 332 973
Zr 51 598 613 101 386 850 Zr 21 603 629 97 396 752
Au (ppb) 51 23 19 18 <2 66 Au (ppb) 21 23 22 15 <2 44
Hg 51 34 31 22 6 97 Hg 21 38 38 24 10 97
Ir 13 5 5 4 2 15 Ir 7 5 5 5 2 15
Pd 18 7 8 2 4 10 Pd 10 7 8 2 4 10
Pt 18 225 159 162 75 689 Pt 10 232 165 177 97 689
Rh 18 14 12 9 6 41 Rh 10 14 12 10 6 41
Ru 18 12 12 3 7 20 Ru 10 11 12 4 7 20
Depth (m) 51 3639 3760 815 2155 4833 Thicknessd 21 48 35 40 5 136
Thicknessd 51 40 27 33 4 136 a
Standard deviation.
a b
Standard deviation. Minimum.
b c
Minimum. Maximum.
c d
Maximum. Crust thickness in millimeters.
d
Crust thickness in millimeters.
11 of 23
Geochemistry 3
Geophysics
Table 6. Statistics for 25 Fe-Mn Crust Layers (Data data show a positive Ce anomaly (Ce*) and con-
From Table 3) vex up middle REE pattern (Figure 6). Chondrite-
N Mean Median SDa Minb Maxc normalized patterns show a positive Ce anomaly,
small positive Gd anomaly, LREE enrichment, and
Fe (wt. %) 25 22.8 24.1 4.37 12.2 28.3 a progressive decrease in normalized HREE with
Mn 25 25.8 25.0 4.77 18.4 35.8
Fe/Mn 25 0.9 1.0 0.9 0.7 0.8 increasing atomic number.
Si 25 8.97 8.37 2.78 4.55 15.3 [32] The Ce* ranges from a very strong to a weak
Na 25 1.90 1.88 0.17 1.66 2.32
Mg 25 1.65 1.60 0.23 1.41 2.28 positive anomaly (11.5 to 1.7, respectively). In
Al 25 2.00 1.88 0.84 0.72 4.26 most samples, Ce* decreases in magnitude with
P 25 0.59 0.48 0.66 0.26 3.72 distance from the substrate (decreasing age) in the
K 25 1.01 0.88 0.41 0.61 2.47 crust, in other samples adjacent layers have similar
Ca 25 2.80 2.37 1.64 1.61 10.2 values. Good examples are crusts D2-3-1 where
Ti 25 1.053 0.967 0.349 0.494 1.938
S 25 0.15 0.14 0.04 0.06 0.28 Ce* decreases with time from 11.5 to 5.9 to
LOI 25 21.2 20.7 2.53 16.4 26.1 2.8 through three layers, and crust D11–2B
As (ppm) 25 212 207 61 84 334 decreases with time from 8.1 to 4.0 to 2.3 through
B 25 401 415 171 173 818 three layers. There is a reversal in layer 3 of
Ba 25 2348 2005 956 1463 5425 the oldest crust (D14–16B), where Ce* increases
Be 25 8.3 7.3 4.8 3.0 29.8
Bi 25 36.4 30.2 18.7 11.5 82.8 from 3.0 to 7.6, then decreases to 4.4 and then
Br 25 23 24 5 15 37 3.0 for the uppermost layer (Table 6).
Cl 25 10090 10608 2318 3741 15588
Co 25 2419 2046 1044 1311 5186 5.5. Inter-element Relationships
Cr 25 43 42 20 8 75
Cu 25 1522 1220 1034 318 4837 and Grouping of Elements
Ga 25 15 14 6 8 35 [33] Correlation coefficient matrices were con-
Li 25 12 8 11 1 46
Mo 25 590 615 132 373 858 structed for the chemical compositions of all data,
Nb 25 44 44 18 20 77 bulk crusts, and crust layers. In addition to chemical
Ni 25 3628 3537 1762 1387 8417 elements, all matrices included longitude, latitude,
Pb 25 1594 1596 416 749 2655 water depth, and crust thickness.
Rb 25 18 17 8 5 44
Sb 25 44.1 44.3 8.6 29.1 65.2 [34] For the 21 bulk crust samples analyzed, statis-
Sc 25 12.0 11.1 4.7 4.6 27.5 tically significant positive correlations (p = 0.01)
Sr 25 1473 1474 273 783 1959 are found among the following elements, which are
Te 25 27.4 17.6 21.4 6.1 74.9
Th 25 56.6 45.8 31.7 11.8 152 assumed to represent the Mn, Fe, aluminosilicate,
Tl 25 108 93.5 66.5 16.7 285 and CFA or biogenic phases: Mn: Mo, W, Sb, Bi,
U 25 11.5 11.5 4.1 2.9 21.7 Co, Sr, U, Te, Pb, and V (at p = 0.05 also Ni, Nb,
V 25 562 554 128 253 771
W 25 63 64 18 30 94
Y 25 192 179 87 100 561
Zn 25 691 677 125 485 961
Zr 25 609 613 108 386 850
Au (ppb) 25 26 23 21 <2 66
Hg 25 27 24 19 6 69
Ir 5 5 5 2 2 8
Pd 6 7 8 1 5 8
Pt 6 222 165 163 75 520
Rh 6 15 14 8 7 27
Ru 6 13 13 2 10 16
Thicknessd 25 25 19 17 4 83
a
Standard deviation.
b
Minimum.
c
Maximum.
d
Crust thickness in millimeters.
D14–16, the REEs increase away from layer 3,

except Ce, which decreases.
Figure 5. Ternary diagram of Fe:Mn:(Co+Ni+Cu)10
[31] Normalized REE plots show patterns typical of for Shatsky Rise samples; field A is diagenetic, B hydro-
hydrogenetic crusts and nodules: PAAS-normalized genetic, and C hydrothermal.
12 of 23
Geochemistry 3
Geophysics
Table 7 (Sample). Concentrations of Yttrium and Rare Earth Elements (ppm) in Fe-Mn Crusts [The full Table 7 is
available in the HTML version of this article]
D2-3-1A D2-3-1B D2-3-1C D2-3-1D D2-3-2A D2-3N-A D3-1A D3-1B D3-1C D3-2N-A D4-1A
Y 160 168 130 103 123 135 65 88 103 105 143
La 301 308 247 185 231 233 90.3 123 200 185 250
Ce 3930 1780 3030 4340 2640 1370 468 580 878 905 1750
Pr 68.6 72.6 58.6 41.4 54.3 56.3 22.5 29.6 46.7 43.2 54.1
Nd 305 331 259 180 242 259 110 144 220 202 248
Sm 69.8 75.2 59.5 40.5 54.8 59.3 25.8 33.1 49.8 45.6 55.2
Eu 15.9 17.3 13.4 9.11 12.5 13.8 6.02 7.78 11.7 10.7 12.6
Gd 74.7 77.5 62.3 46.1 58.4 61.8 28.1 35.9 52.9 48.9 59.2
Tb 9.7 10.5 8.1 5.8 7.6 8.4 3.8 5.0 7.1 6.6 7.7
Dy 54.8 59.3 45.4 32.8 43.0 47.5 22.6 29.0 39.8 37.8 44.7
Ho 8.97 9.51 7.36 5.54 7.08 7.70 3.74 4.83 6.36 6.09 7.33
Er 26.8 27.8 22.2 17.0 20.9 22.8 11.2 14.4 18.8 18.2 22.2
Tm 3.4 3.4 2.8 2.3 2.7 2.9 1.4 1.8 2.4 2.3 2.8
Yb 23.7 23.4 19.5 15.9 18.4 19.6 9.8 12.6 16.3 15.6 19.5
Lu 3.28 3.16 2.62 2.23 2.51 2.69 1.33 1.71 2.21 2.09 2.67
∑REE 4896 2799 3838 4924 3395 2165 805 1023 1552 1529 2536
Ce*a 6.4 2.8 5.9 11.5 5.5 2.8 2.4 2.3 2.1 2.4 3.5
Intervalb B 0–112 L 0–37 L 37–82 L 83–112 B 0–70 N 0–54 B 0–68 L 0–46 L 46–68 N 0–57 B 0–118
a
Ce* = 2Ce/La + Pr from chondrite-normalized data.
b
B = bulk, L = layer, CCB = coated cobble bulk.
Zn); Fe: As, U, V, Pb, Sr, Sb, Zr, and W; Si: Al, K, identified using Q-mode analysis. In addition to the
Rb, Na, and Sc (at p = 0.05 also Mg); P: Ca, Y, S, elements listed above, the Fe factor has: REEs
Cr, Te, Bi, Se, Ir, Pt, Rh, and Ru (at p = 0.05 also (except Ce), Ti, Bi, Mo, and Te; the aluminosilicate
Tl). Copper shows a negative correlation with Fe factor has Hg, Th, and Pd; and the Mn factor
and associated elements, but does not correlate with has: Ce, Li, Cu, Mg, Ba, Ga, and K, but does not
Mn; Cu has positive correlations with Ga, Mg, Ni, have U, V, Nb, and As, all of which correlate
Cd, Ba, K, and Zn (at p = 0.05 also Au, Li, Rb). with Mn; we assume that these latter elements
Nickel also has a negative correlation with Fe and are in a phase that covaries with the Mn phase
associated elements, but unlike Cu, has a positive which explains the correlations yet exclusion
correlation with Mn as well as Cd, Zn, Cu, Mg, Ba, from the Q-mode Mn factor. If the one bulk crust
and Ga (at p = 0.05 also Li). Thorium does not with CFA in it is removed, then the elements
correlate with any of the major phases and has that correlate in that phase are distributed in either
positive correlations with Ta, B, Nb, and Be. Water the Fe (P, Ca) or Mn (PGEs) factors, otherwise
depth has positive correlations with Latitude, Lon- they are not in any of the factors.
gitude, and Th at p = 0.01 and Si, Al, B, and Sc at
p = 0.05; and negative correlations with Co, Sb, Pb, 5.6. Growth Rates and Ages
Sr, Zn, P, and Te, which indicates that Th and the
[36] Growth rates of the upper 20 mm of the four
aluminosilicate elements increase in crusts with
increasing water depth, whereas some of the ele- crusts dated using 10Be/9Be vary from 10.5 mm/Myr
ments associated with the Mn, Fe, and CFA phases to 2.1 mm/Myr (Figure 7 and Table 8). The three
decrease with increasing water depth. These same deeper-water crusts (3759–4000 m) generally have
relationships occur with increasing latitude, which higher growth rates (10.5–3.1) than the shallower-
stands to reason since water depth increases with water (2155 m) crust (2.1–3.7 mm/Ma). This is
increasing latitude. Rh, Tl, Ba, B, and Ga increase consistent with the higher oxidation potential for
with increasing crust thickness and crust thickness the deeper water compared to the oxygen mini-
increases with increasing longitude; Hg decreases mum zone (OMZ) where the shallower-water crust
with increasing crust thickness. formed.
[37] The base of the upper 20 mm of the four
[35] Q-mode factor analysis for the bulk crust data
set shows three factors that account for 96% of Be-isotope dated crusts ranges in age from 2.67
the variance and represent the Fe, aluminosilicate, to 6.57 Ma, the shallow-water crust is the oldest.
and Mn phases (57%, 15%, and 23% of the vari- The earliest initiation of crust growth was approxi-
ance respectively); a CFA/biogenic factor was not mately at 28 Ma for D14–16B based on the Co
13 of 23
Geochemistry 3
Geophysics
Figure 6. PAAS-normalized and chondrite-normalized rare earth element plots for bulk nodule and bulk crust
samples.
chronometer, which is in good agreement with the ferromanganese cement may have formed rather
30 Ma extrapolated Be isotope age. Crust D11–2 rapidly. However, the Co chronometer would put
began its growth 14.5 Ma ago based on the Co the age at 78 mm, the extent of chemical data, at
chronometer, which is in reasonable agreement with 8.2 Ma, which is very different from the extrapo-
an extrapolated Be isotope age of 12.6 Ma. Crust lated Be age of 12.5 Ma. Crust D4–1 shows signif-
D7–1 consists of a nodule conglomerate from the icantly different ages of initiation of growth based
base at 130 mm to 30 mm, which is overlain by on the Co chronometer (13.7 Ma) and extrapolated
typical parallel laminae for the upper 30 mm. The Be age (37.8 Ma), even when the surface sample is
depth at 30 mm corresponds to 3.5 Ma based on Co corrected to modern seawater Be isotope value
chronometer (4.4 Ma extrapolated Be age). Ages for (Tables 8 and 9). Why these ages for D4–1 are so
the conglomerate layer are probably meaningless different is unclear, but all of the Be-isotope ages
because the nodules may be of different ages and the are extrapolated whereas measured Co, Fe, and Mn
14 of 23
Geochemistry 3
Geophysics
Figure 7. Growth-rate curves based on 10Be/9Be ratios; depth is from the crust surface that was in contact with sea-
water. The surface isotopic ratio for TN037-D4-1 is significantly lower than the other three crusts, which reflect con-
temporary seawater, possibly because of erosion of the surface; using the mean of the surface values for the other three
crusts produces a nearly linear growth rate for D4–1. The red line for D14–16B is for a linear age model.
Table 8. Be Isotope Ages of Four Crusts From Cruise TN037

10
Sample Mean Depth (mm) Be (at./g (1e7)) 9
Be (ppm/error) 9
Be Conc (1016 at/g) 10/9
Be (10 8) Age (Ma)
D4-1-05 0.25 0.25 327 17 0.68/0.01 4.51 0.05 7.256 0.392 a
0.91
D4-1-6 5.50 0.50 1543 47 4.06/0.04 27.12 0.30 5.692 0.183 1.39
D4-1-115 10.75 0.75 763 23 4.85/0.05 32.38 0.36 2.356 0.076 3.10
D4-1-21 20.50 0.50 183 6 5.52/0.06 36.85 0.41 0.497 0.018 6.26
D7-1-05 0.25 0.25 1030 32 1.46/0.02 9.73 0.11 10.591 0.345 0.02
D7-1-52 4.60 0.60 2008 61 3.42/0.04 22.83 0.25 8.793 0.282 0.44
D7-1-105 9.85 0.65 1371 41 3.06/0.03 20.43 0.22 6.711 0.216 0.94
D7-1-21 20.00 1.0 533 16 2.81/0.03 18.80 0.21 2.833 0.091 2.67
D11-2-05 0.25 0.25 918 29 1.23/0.01 8.20 0.09 11.193 0.378 0.03
D11-2-52 4.50 0.70 1930 58 3.33/0.04 22.22 0.24 8.688 0.279 0.54
D11-2-108 10.30 0.50 1606 48 4.55/0.05 30.39 0.33 5.286 0.170 1.64
D11-2-202 20.70 0.50 601 18 4.79/0.05 32.03 0.35 1.877 0.060 3.62
D14-16B-05 0.25 0.25 1722 57 2.33/0.03 15.56 0.17 11.068 0.389 0.09
D14-16B-65 5.75 0.75 1298 39 3.66/0.04 24.48 0.27 5.301 0.171 1.96
D14-16B-110 10.50 0.50 391 12 3.37/0.04 22.51 0.25 1.736 0.057 3.58
D14-16B-200 19.25 0.75 122 6 3.94/0.04 26.31 0.29 0.465 0.022 6.57
a
The measured surface 10/9Be ratio is too low for modern Pacific seawater and the upper 0.8 Ma of crust is likely missing; the ages given assume
11.42 10 8 to be the correct surface value (average of other 3 samples).
15 of 23
Geochemistry 3
Geophysics
Table 9. Age of Cu Enrichment in Crusts From NE to SW Along Shatsky Rise Compared to Non-enriched Crusts
(D9, D10, D11?)a
Sample and Interval Cu Ni Li Age of Boundaries Water Depth
(mm) (ppm) (ppm) (ppm) (Ma) Dating Methodb (m)
D2-3-1B 0-37 748 1749 6 0–3.7 Co 4833
D2-3-1C 37-82 1220 2720 6 3.7–11.6 Co
D2-3-1D 82-112 1808c 3616 9 11.6–19.8 Co
D3-1B 0-46 2305 4551 18 0.0–3.4 Co 4227

D3-1C 46-68 3257 5325 25 3.4–4.4 Co
10/9
D4-1B 0-26 652 1553 5 0–8.0 Be (ext) 4000
0–2.2 Co
10/9
D4-1C 26-42 699 2700 1 8.0–13.2 Be (ext)
2.2–3.4 Co
10/9
D4-1D 42-56 2270 5791 9 13.2–17.7 Be (ext)
3.4–4.7 Co
10/9
D4-1E 56-92 4837 8417 46 17.7–29.4 Be (ext)
4.7–10.8 Co
10/9
D4-1F 92-118 1735 1810 12 29.4–37.8 Be (ext)
10.8–13.7 Co
10/9
D7-1B 0-15 958 2111 7 0–1.8 Be 3759
0–1.1 Co
10/9
D7-1C 15-30 1500 3388 10 1.8–4.4 Be (ext)
1.1–3.5 Co
10/9
D7-1D 30-78 3418 7217 40 4.4–12.5 Be (ext)
3.5–8.2 Co
D9-12B 0-20 681 1387 8 0–1.2 Co 3900

D9-12C 20-38 873 2147 5 1.2–2.7 Co
D10-11C 0-11 1360 3959 12 0–1.2 Co 3474

D10-11D 11-21 1179 2443 13 1.2–2.3 Co
D10-11E 21-25 1463 3537 13 2.3–2.9 Co
10/9
D11-2B 0-27 891 2347 6 0–4.8 Be (ext) 3760
0–2.7 Co
10/9
D11-2C 27-42 974 4374 5 4.8–7.7 Be (ext)
2.7–4.9 Co
10/9
D11-2D 42-68 1612 3849 8 7.7–12.6 Be (ext)
4.9–14.5 Co
10/9
D14-16B2 0-14 318 2474 4 0–4.6 Be 2155
0–2.8 Co
10/9
D14-16B3 14-49 651 3841 4 4.6–15.2 Be (ext)
2.8–13.3 Co
10/9
D14-16B4 49-69 1151 5199 4 15.2–21.5 Be (ext)
13.3–21.9 Co
10/9
D14-16B5 69-97 1492 4188 17 21.5–29.6 Be (ext)
21.9–27.9 Co
a
D2 is off the rise to the east.
b
Co, cobalt chronometer; ext, extrapolated.
c
Bold is highest value(s) within a crust if significantly different from other values in that crust.
data were available for the Co chronometer. Growth and erosion unconformities or periods of non-
hiatuses may also contribute to the age disparity deposition.
between the two methods. As mentioned above,
[38] The four thickest crusts analyzed, D7–1
these ages for initiation of crust growth are minima
(130 mm), D4–1 (118 mm), D2-3-1 (112 mm), and
because the Co chronometer and extrapolated Be
D14–16B (110 mm) began to grow at 9.4, 13.7,
growth rates do not take into account dissolution
19.8, and 30.8 Ma ago respectively, based on the
16 of 23
Geochemistry 3
Geophysics
growth rates of individual layers using the Co available to be acquired through adsorption by the
chronometer and the thickness of each layer. The nodules at the seabed.
three thickest crusts progressively increase in water
depth (3700, 4000, 4833 m) with increasing age of 6.2. Spatial and Temporal Enrichments
initiation of growth, and correspondingly a decrease of Cu and Ni in Fe-Mn Crust Layers
in growth rate. The other thick crust (D14–16B) is
from the shallowest water (2155 m), has the slowest [42] Shatsky Rise crusts have much higher average
growth rate, and is the oldest crust. This crust has Cu concentrations (Figure 4) and stronger Ni- and
growth rates typical of PCZ crusts (75% are ≤5 mm/ Cu-enriched layers than crusts found elsewhere in
Myr) based on earlier studies [e.g., Hein et al., the global ocean, which may be the result of a
2000]. deep-water source for those metals not available to
shallower-water crusts, or to changes in paleoen-
[39] These results indicate that the dating is inter- vironmental conditions. The strong Cu and Ni
nally consistent and as reliable as can be expected enrichments are confined to the deeper layers of the
based on combined Co and 10Be/9Be dating. The crusts rather than the surface layers, supporting the
uncertainty of the dating for the time period covered latter supposition. The Be isotope and Co chro-
by 10Be/9Be is a maximum of 0.5 Myr at 6 Myr. nometer ages indicate that this Cu-Ni enrichment
Crust sections deeper than that, where the ages were is not found in crust layers younger than about
calculated from Co-chronometry, the uncertainties 3.4 Ma at any location across Shatsky Rise
are greater but the combined 10Be/9Be and Co ages (Table 9). The exact age of the end of Cu-Ni
clearly demonstrate that these deposits are dateable enrichment at each site is not known because
and reliable archives for past ocean chemistry. chemical analyses were determined for macro-
scopically distinct layers within each crust, which
6. Discussion might not accurately constrain the compositional
changes. However, with the data available, the age
[40] Several characteristics distinguish Shatsky of initiation of Cu-Ni-enrichment for the deep-water
Rise ferromanganese deposits from those studied sites was the middle Miocene (Table 9). Five layers,
from other regions of the global ocean, including the most analyzed from a single crust (D4–1),
deposit configuration, high Cu, Ni, Th, and Li include a layer that accreted prior to enrichment of
layers in crusts, and a markedly different chemical Cu and Ni, which shows the start of enrichment at
composition compared to PCZ deposits that occur about 11 Ma and the end of enrichment at about
farther south. 3.4 Ma. However, it cannot be ruled out that
enrichment started as early as 15–20 Ma at other
6.1. Amalgamated Crust-Nodule Deposits sites. The three Shatsky Rise deep-water crusts (D3,
D4, D7) have the highest Cu concentrations, which
[41] The crusts and nodules occur in close proximity correspond to the time interval of about 12.5 to
on Shatsky Rise, and some nodule deposits were 3.4 Ma. The degree of enrichment decreases in
cemented into pavements, thereby forming amal- crusts located farther south and the D11–2 site
gamated crust-nodule deposits, which were further occurs at about the same water depth as D7, but
overgrown by Fe-Mn crust. The nodules formed by maximum Cu and Ni values are half of what they are
accretion about a small nucleus on a sediment sur- in the D7 crust. The deepest-water crust (D2;
face and the crusts grew on rock surfaces, but in 4833 m) is at about the same latitude as D7, but is far
close proximity to the nodules. The amalgamated below the calcite compensation depth (CCD) today
deposits could have formed either by winnowing of and also during the middle Miocene, which likely
the sediment on which the nodules rested and sub- had an influence on the smaller enrichment of Cu in
sequent crust formation on a rock surface populated that crust (Table 9; see below); this is also true for
with scattered nodules; or the nodule field was so crusts from dredge D8, 4806 m water depth, but not
densely packed that crusts were able to grow on that for a nodule analyzed from that dredge. For the
more-or-less continuous hard substrate. Conclusive shallower-water crust (D14–16B, 2155 m), Cu con-
evidence is not found to support one hypothesis tents are relatively low throughout the crust, but
over the other. Since the nodules in the amalgam- nonetheless decrease nearly fivefold in the four
ated deposits are enriched in Cu, Ni, and Li, the layers analyzed. This uniform decrease in Cu may
sediment was thick enough to have established the reflect a decrease in the extent and intensity of the
diagenetic gradient required to release these metals OMZ during growth of this shallow-water crust.
from their host phases where they then became
17 of 23
Geochemistry 3
Geophysics
Figure 8. Comparison of Shatsky Rise mean bulk Fe-Mn crust composition for selected elements with mean crust
values for northwest-equatorial Pacific Prime Crust Zone, the area of greatest economic interest in the global ocean.
[43] Copper concentrations are highly variable in diagenetic component results from oxic or suboxic
crusts throughout the area, which reflects water diagenesis [Dymond et al., 1984]. The most striking
depth and the associated characteristics that change contrasts between the Shatsky Rise crusts and PCZ
with water depth. For the deep-water crusts, the crusts are the high Fe, Si, Al, B, Ba, Cu, Li, and Th
dominant controls are the composition and flow rate and low Bi, Co, Te, Tl, W, and As concentrations in
of bottom water masses and the depth of the CCD, the Shatsky crusts (Figure 8). The difference in Th
which controls sedimentation rate and relative pro- concentrations is especially striking in that it is near
portions of carbonate, silicate, and organic matter the Earth’s crustal abundance (12 ppm) for PCZ
concentrations in bottom sediment, and indirectly crusts (11 ppm) yet has a mean of 56 ppm for
pore fluid composition; and for the shallow-water Shatsky Rise samples and reaches exceptionally
crusts, upwelling along the Rise and extent and high concentrations of 152 ppm for a crust layer
intensity of the OMZ. These are discussed in the from dredge D2 and 113–148 ppm for bulk crusts
next section. and a nodule also from D2 (Table 3). The mean Th
concentration of the Shatsky Rise crusts is similar
6.3. Sources and Enrichment Mechanisms to those found in crusts from the Atlantic and
of Cu, Ni, Li, and Th Indian Oceans and from seamounts off California in
the northeast Pacific (52–56 ppm) [Hein et al.,
[44] The chemical composition of crusts and 2000; Hein and Koschinsky, 2012]. This enrich-
nodules vary on regional, local, and intrasample ment relative to most other Pacific areas may relate
scales. In general, the chemical composition of to a deep-water source for Th and the location of
nodules is controlled by the ratio of the hydro- Shatsky Rise under the Asian dust plume. Dis-
genetic to diagenetic components and whether the solved Th concentrations increase systematically
18 of 23
Geochemistry 3
Geophysics
with increasing water depth and are higher than the hydrogenetic crust means are very low, ranging from
mean value for Shatsky Rise crusts in three of the about 3–7 ppm. Lithium in manganese deposits can
four deepest-water sites (D2, D4, D8; but not D3); also have a hydrothermal source [e.g., Rogers et al.,
this depth relation is similar to Indian Ocean 2001; Chan and Hein, 2007], however, the com-
crusts (J. R. Hein, USGS unpublished data, 2010). bination of high Cu, Ni, and Li together in the
High 232Th is found in the western Pacific under the Shatsky Rise crusts indicate a diagenetic source.
Asian dust plume [Nozaki et al., 1987], whereas in Such high Li concentrations have not been found
the central Pacific, 232Th is much lower and without in hydrogenetic Fe-Mn crusts and this first occur-
much variability in the water column [Roy-Barman rence requires explanation. High Li, Cu, and Ni
et al., 1996]. The source of 232Th is exclusively in the nodule layers of amalgamated crust-nodule
detrital material and is a robust tracer of dust fluxes deposits are the result of diagenetic input to the
[Hsieh et al., 2011]. In dissolved form, Th is highly nodules before they were incorporated into the
particle reactive with a residence time in seawater of crusts, which is typical for nodules [e.g., Halbach
<100 years in the North Pacific [Okubo et al., 2012]. et al., 1981; Hein and Koschinsky, 2012]. How-
Release of Th from bottom water and surface sedi- ever, other crust layers rich in those metals must
ment is currently debated and subject of an ongoing have another explanation since crusts grow on rock
GEOTRACES program. Our data indicate that in substrates and cannot have a direct diagenetic input
the deep-water parts of Shatsky Rise, Th is available [Hein and Morgan, 1999]. The most likely expla-
for scavenging and readily sorbs onto the Fe-Mn nation is that diagenetic reactions in the sediment
oxides. Such availability and bottom scavenging (biogenic silica, organic matter) and dissolution of
may be temporally and spatially variable in the deep biogenic calcite in the water column are the sour-
ocean [Okubo et al., 2012]. ces for these elements [e.g., Martin and Knauer,
1973; Boström et al., 1974; Monteiro and Orren,
[45] Another source of Th may be biogenic, which
1985; Miller and Cronan, 1994], which were
is likely the source for crusts off California where
mixed and enhanced in the deep water masses from
high coastal primary productivity occurs. The dee-
which the crusts acquired metals. Similar processes
pest-water Shatsky crusts occur below the CCD
explain Cu and Ni concentrations in the deep-water
where organic matter in the sediment is not diluted
masses of the NE Atlantic [Saager et al., 1997].
by pelagic carbonate, so degradation of organic
These enrichments would explain the concentra-
matter and dissolution of biogenic silica may be
tions in the Shatsky Rise crusts that formed in deep
seabed sources for Th [Knauss and Ku, 1983;
water and in places where deep-water upwelled to
Hunter et al., 1988; Moran and Moore, 1992; Lao
shallower depths, 2000–3000 m [Kawabe et al.,
et al., 1993]. Biogenic calcite also has high con-
2009]. The highest concentrations of Li are in
centrations of Th [Delaney and Boyle, 1983], which
crust layers that formed near the CCD while the
may provide a source of Th closer to the CCD,
enriched layers were accreting, from about 12.5 to
currently at about 4200 m [van Andel et al., 1975].
3.4 Ma (dredges D3, D4, D7); the CCD varied from
Shatsky Rise occurs below the core of the main
about 3900 m to 4300 m during that time [van Andel
midlatitude, pan-Pacific dust plume originating
et al., 1975]. All the Shatsky Rise nodules analyzed
from Asia [e.g., Rea, 1994], which is the likely
have Li concentrations (15–70 ppm) much greater
source for variable percentages of other lithogenic
than those found in most Pacific crusts (means of
elements besides Th, such as Si, Al, K, Be, etc. The
3–7 ppm), but lower than those found in East
dust undergoes dissolution in the water column and
Pacific nodules.
can enhance primary productivity.
[47] Nodules produced from oxic diagenesis have
[46] Cu, Ni, and Li are also strongly enriched in
the highest Ni and Cu contents [Halbach et al.,
Shatsky Rise samples from the deeper-water sites
1981], as well as the highest Li contents [Hein
(Table 3). High concentrations of these three ele-
and Koschinsky, 2012]. Under suboxic diagenetic
ments in nodules indicate a diagenetic (pore water)
conditions, accompanied by high contents of
source, the greater the diagenetic component the
organic carbon, abundant Fe2+ ions are generated by
higher the concentration of Li [Hein and Koschinsky,
reduction of ferric iron in sediment pore waters,
2012]. Hydrogenetic-diagenetic nodules from the
which outcompetes Cu2+, Ni2+, and Li+ for available
Clarion-Clipperton zone (CCZ) in the east Pacific
vacancy sites within the structure of the 10 Å man-
have a mean Li concentration of about 130 ppm,
ganate that characterizes nodules, thereby decreasing
whereas diagenetic nodules from the Peru Basin
the contents of metals of economic interest. The
average about 310 ppm Li; and typical Pacific
19 of 23
Geochemistry 3
Geophysics
turn-around-point at which Cu and Ni increase with potential in the CCZ and diagenetic reactions pro-
oxic diagenetic input and then decrease with sub- duce more Cu, Ni, and Li, but the redox potential is
oxic diagenetic input occurs at an Mn/Fe ratio of not so low that Fe2+ out competes the Cu, Ni,
about 5 [Halbach et al., 1981]. The high contents of and Li ions for sorption sites. This is validated
organic carbon in areas of suboxic diagenesis reflect by the maximum Mn/Fe ratio in Shatsky Rise
greater primary productivity in surface waters com- samples of 2.0, whereas that ratio is up to about 7
pared to areas where nodule growth is influenced for CCZ nodules [Halbach et al., 1981]. An Mn/Fe
only by oxic diagenetic conditions. Thus, high-grade ratio of about 2.5 divides purely hydrogenetic
nodules occur in areas of moderate primary produc- nodules from those that experienced the beginning
tivity [e.g., Cronan, 1980; Halbach et al., 1981; of early diagenetic input [Halbach et al., 1981].
International Seabed Authority, 2010]. In addition, Our data indicate that ratio is closer to 2.0 for Shatsky
the location of nodule/crust accretion relative to Rise Fe-Mn oxides.
the CCD, which is controlled by primary productiv-
[49] The “fossil” nodules in the Shatsky Rise
ity, is important because above that depth biogenic
amalgamated deposits and those currently forming
calcite increases sedimentation rates and dilutes
in the area have similar compositions, including
organic matter contents [Verlaan et al., 2004;
high Cu, Ni, and to a lesser degree Li concentra-
Cronan, 2006]. The highest-grade nodules form
tions, with the exception of a nodule from the site
near, but generally below the CCD.
of D2, which has a predominantly hydrogenetic
[48] The Shatsky Rise nodules do not fully comply signature (high Co). This relationship is not true for
with this scenario because the redox potential was the deeper-water, younger crust layers, which are
never low enough to form 10 Å manganate, which not enriched in those elements. This may be
is a mineral with tunnel structure that accom- explained by a change in the composition or flow of
modates the high Cu, Ni, and Li concentrations deep-water masses, which prior to about 3.4 Ma
found in nodules in other regions. The d-MnO2 carried the diagenetically produced Cu, Ni, and Li
mineralogy of Shatsky nodules does not have that that enriched older crust layers. This present con-
tunnel structure. Nonetheless, the data presented dition is typical of the shallower-water crusts at
here show that d-MnO2 can accommodate signifi- Shatsky Rise and for crusts elsewhere in the Pacific
cant concentrations of Cu, Ni, and Li, all of which for both younger and older layers.
correlate with Mn concentrations. The highest
[50] The key paleoenvironmental change that may
Mn/Fe ratio for Shatsky nodules is 2.0, indicating
have initiated the chemical changes seen in the
that the nodules did not reach that turn-around-
deep-water Shatsky Rise crusts during the middle
ratio of 5 where the Cu plus Ni concentrations are
Miocene was the considerable increase in Antarctic
the highest [Halbach et al., 1981]. These data
glaciation, which most likely enhanced activity of
indicate that the diagenetic production of Cu, Ni,
the deep-water masses (increasingly corrosive deep
and Li incorporated into the nodules occurred in an
water for carbonates) that accompanied major
oxic environment that was never reducing enough
global cooling. The equator-to-pole thermal gradi-
to produce 10 Å manganate, which is also true for
ent increased, which led to an increase in oceanic
the redox potential under which most seamount
mixing, and an increased supply of nutrients and
Fe-Mn crusts formed during the Cenozoic [Hein
primary productivity [e.g., Kennett and Brunner,
et al., 2000]. Today, the bottom-water dissolved
1973]. Increased continental weathering and eolian
O2 content up the flank of Shatsky Rise decreases
input from Asia also increased nutrient supply [e.g.,
progressively from about 3.8 ml/l at 4800 m to
Leinen and Heath, 1981]. There was a shift in silica
2.0 ml/l at 2200 m and is about 3.6 ml/l near the
deposition at about 15–12 Ma from the Atlantic to
CCD at 4200 m [Kawabe et al., 2009]. Even at
the Pacific [Cortese et al., 2004], accompanied by
the lowest values, which are significantly above
the beginning of the closure of the Isthmus of
0.1 ml/l characteristic for the suboxic zone, both the
Panama to deep-water exchange at about 12 Ma
nodules and crusts are composed of d-MnO2. The
[e.g., Keigwin, 1982]. Chemical changes in central
abyssal bottom water O2 content in the CCZ in the
Pacific Fe-Mn crusts, including Cu and Ni, occurred
eastern Pacific is about the same, 3.8 ml/l [Mantyla
at about that time [Hein et al., 1992; Frank et al.,
and Reid, 1983], but the nodules there are composed
1999]. The Shatsky Rise enrichments of Cu and
predominantly of 10 Å manganate and lesser
Ni in crusts ended about 3.4 Ma, which broadly
amounts of d-MnO2. The difference is that the pore
corresponds to initiation of northern hemisphere
water chemistry is dominated by a lower redox
glaciation, final closing of the Isthmus of Panama,
20 of 23
Geochemistry 3
Geophysics
decreasing activity of AABW, and an increase in latitude location, which commonly has high seas.
eolian dust flux [Keigwin, 1982; Rea, 1994; Burton The higher Cu, Th, and Li concentrations in Shatsky
et al., 1999]. Rise crusts than found elsewhere in Pacific crusts
are, however, not high enough to offset the
7. Summary and Conclusions decreased value related to the lower Co, Te, W, Bi,
etc. concentrations (Figure 8). On the other hand,
[51] A unique set of Fe-Mn crust and nodule sam- the total REE concentrations in dredge D2 crusts are
ples collected from Shatsky Rise, located at middle very high, varying from 0.3 to 0.5%, and therefore
latitudes in the NW Pacific, show characteristics warrant further studies in the Ojin Rise Seamounts.
unlike samples from the PCZ region. One distinc-
tion is that the deep-water Shatsky Rise crusts show Acknowledgments
a diagenetic chemical signal, but not a diagenetic
mineralogical signal; that constrains the diagenetic [53] Onboard sample descriptions were by Maria Tejada,
redox conditions to early oxic diagenesis. Diage- Glenn Brown, and Atsushi Horio. Warren Smith at SIO facili-
netic input of metals to Fe-Mn crusts is rare, tated locating and collecting samples. Jutta Heinze carried out
although it is unequivocal in these deep-water the chemical preparation of the samples for Be isotope measure-
crusts. Thorium concentrations in PCZ crusts aver- ments and Ulrike Westernströer and Dieter Garbe-Schönberg
age about 12 ppm, whereas the deep-water Shatsky are thanked for the 9Be measurements. Renee Takesue and
crusts average 56 ppm, with an exceptionally high two anonymous reviewers provided helpful comments that
concentration of 152 ppm. Thorium enrichments in improved this manuscript.
the older crust layers (pre 3.4 Ma) reflect move-
ment of Shatsky Rise into the core of the Asian References
dust plume and additionally deep-water diagenetic
sources that enhanced the Cu, Ni, and Li con- Anders, E., and N. Grevesse (1989), Abundances of the
centrations in the same older crust layers. Crusts elements: Meteoritic and solar, Geochim. Cosmochim. Acta,
and nodules occur in close proximity and some 53, 197–214, doi:10.1016/0016-7037(89)90286-X.
Beaman, M., W. W. Sager, G. D. Acton, L. Lanci, and J. Pares
nodule deposits were cemented by and overgrown (2007), Improved Late Cretaceous and early Cenozoic paleo-
by crusts forming amalgamated deposits. The non- magnetic apparent polar wander path for the Pacific Plate,
amalgamated Shatsky Rise deep-water crusts show Earth Planet. Sci. Lett., 262, 1–20, doi:10.1016/j.epsl.
a diagenetic signal characterized by especially high 2007.05.036.
Cu, Ni, and Li concentrations, which has not been Bidoglio, G., P. N. Gibson, M. O’Gorman, and K. J. Roberts
(1993), X-ray absorption spectroscopy investigation of sur-
seen in crusts studied elsewhere from the global face redox transformations of thallium and chromium on
ocean. This diagenetic input is confined to crust colloidal mineral oxides, Geochim. Cosmochim. Acta, 57,
layers older than about 3.4 Ma and began during 2389–2394, doi:10.1016/0016-7037(93)90576-I.
the middle Miocene. The sources of these metals is Boström, K., O. Joensuu, and I. Brohm (1974), Plankton: Its
most likely from diagenetic reactions that took chemical composition and its significance as a source of
pelagic sediments, Chem. Geol., 14, 255–271, doi:10.1016/
place in the deep-water sediment (biogenic silica, 0009-2541(74)90063-1.
organic matter) and dissolution of biogenic calcite Burton, K. W., D.-C. Lee, J. N. Christensen, A. N. Halliday,
in the water column, which were mixed and and J. R. Hein (1999), Actual timing of neodymium isotopic
enhanced in the deep water masses from which the variations recorded by Fe-Mn crusts in the western North
crusts acquired metals. The highest concentrations Atlantic, Earth Planet. Sci. Lett., 171, 149–156, doi:10.1016/
S0012-821X(99)00138-7.
of Li are in crust layers that formed near the CCD. Chan, L.-H., and J. R. Hein (2007), Lithium contents and
The onset of Ni, Cu, and Li enrichment during the isotopic compositions of ferromanganese deposits from the
middle Miocene and cessation at about 3.4 Ma global ocean, Deep Sea Res., Part II, 54, 1147–1162,
were accompanied by changes in the deep-water doi:10.1016/j.dsr2.2007.04.003.
environment, especially composition and flow Chmeleff, J., F. von Blanckenburg, K. Kossert, and D. Jakob
(2010), Determination of the 10Be half-life by multicollector
rates of water masses; and location of the CCD, ICP-MS and liquid scintillation counting, Nucl. Instrum.
which responded to surface water productivity. Methods Phys. Res., Sect. B, 268, 192–199, doi:10.1016/
j.nimb.2009.09.012.
[52] The shallower-water crusts (<2500 m) in the Cook, H. E., P. D. Johnson, J. C. Matti, and I. Zemmels (1975),
PCZ offer a resource potential for Co, Ni, Mn, and Methods of sample preparation and X-ray diffraction data
perhaps also REEs, Te, Zr, W, and other rare analysis (X-ray mineralogy laboratory, Deep Sea Drilling
metals, but the Shatsky Rise crusts are less pro- Project, Univ. of California Riverside), Initial Rep. Deep
spective because of the relatively low Co con- Sea Drill. Proj., 28, 999–1007.
Cortese, G., R. Gersonde, C.-D. Hillenbrand, and G. Kuhn
centrations, deep-water environment, and middle (2004), Opal sedimentation shifts in the World Ocean over
21 of 23
Geochemistry 3
Geophysics
the last 15 Myr, Earth Planet. Sci. Lett., 224, 509–527, Hein, J. R., T. A. Conrad, and R. E. Dunham (2009), Seamount
doi:10.1016/j.epsl.2004.05.035. characteristics and mine-site model applied to exploration-
Cronan, D. S. (1980), Underwater Minerals, 362 pp., Academic, and mining-lease-block selection for cobalt-rich ferromanga-
London. nese crusts, Mar. Georesour. Geotechnol., 27, 160–176,
Cronan, D. S. (2006), Processes in the formation of central doi:10.1080/10641190902852485.
Pacific manganese nodule deposits, J. Mar. Sci. Environ., Hsieh, Y.-T., G. M. Henderson, and A. L. Thomas (2011),
C4, 41–48. Combining seawater 232Th and 230Th concentrations to deter-
Delaney, M. L., and E. A. Boyle (1983), Uranium and thorium mine dust fluxes to the surface ocean, Earth Planet. Sci. Lett.,
isotope concentrations in foraminiferal calcite, Earth Planet. 312, 280–290, doi:10.1016/j.epsl.2011.10.022.
Sci. Lett., 62, 258–262, doi:10.1016/0012-821X(83)90088-2. Hunter, K. A., D. J. Hawke, and L. K. Choo (1988), Equilib-
Dymond, J., M. Lyle, B. Finny, D. Z. Piper, K. Murphy, rium adsorption of thorium by metal oxides in marine electro-
R. Conard, and N. Pisias (1984), Ferromanganese nodules lytes, Geochim. Cosmochim. Acta, 52, 627–636, doi:10.1016/
from MANOP Sites H, S, and R—Control of mineralogi- 0016-7037(88)90324-9.
cal and chemical composition by multiple accretionary International Seabed Authority (2010), A geological model
processes, Geochim. Cosmochim. Acta, 48, 931–949, of polymetallic nodule deposits in the Clarion-Clipperton
doi:10.1016/0016-7037(84)90186-8. fracture zone, ISA Tech. Study, 6, Kingston, Jamaica.
Frank, M., J.-D. Eckhardt, A. Eisenhauer, P. W. Kubik, Kawabe, M., S. Fujio, D. Yanagimoto, and K. Tanaka (2009),
B. Dittrich-Hannen, M. Segl, and A. Mangini (1994), Water masses and currents of deep-circulation southwest of
Beryllium 10, thorium 230, and protactinium 231 in Galapagos the Shatsky Rise in the western North Pacific, Deep Sea
Microplate sediments: Implications of hydrothermal activity Res., Part I, 56, 1675–1687, doi:10.1016/j.dsr.2009.06.003.
and paleoproductivity changes during the last 100,000 years, Keigwin, L. (1982), Isotopic paleoceanography of the Caribbean
Paleoceanography, 9, 559–578, doi:10.1029/94PA01132. and East Pacific: Role of Panama uplift in late Neogene time,
Frank, M., R. K. O’Nions, J. R. Hein, and V. K. Banakar Science, 217, 350–353, doi:10.1126/science.217.4557.350.
(1999), 60 Myr records of major elements and Pb-Nd isotopes Kennett, J. P., and C. A. Brunner (1973), Antarctic Late
from hydrogenous ferromanganese crusts: Reconstruction of Cenozoic glaciation: Evidence for initiation of ice rafting
seawater paleochemistry, Geochim. Cosmochim. Acta, 63, and inferred increased bottom-water activity, Geol. Soc.
1689–1708, doi:10.1016/S0016-7037(99)00079-4. Am. Bull., 84, 2043–2052, doi:10.1130/0016-7606(1973)
Garbe-Schönberg, C. D. (1993), Simultaneous determination 84<2043:ALCGEF>2.0.CO;2.
of 37 trace elements in 28 international rock standards by Klovan, J. E., and J. Imbrie (1971), An algorithm and
ICP-MS, Geostand. Newsl., 17, 81–97, doi:10.1111/j.1751- FORTRAN-IV program for large-scale Q-mode factor analysis
908X.1993.tb00122.x. and calculation of factor scores, Math. Geol., 3, 61–77,
Halbach, P., C. Scherhag, U. Hebisch, and V. Marchig (1981), doi:10.1007/BF02047433.
Geochemical and mineralogical control of different genetic Knauss, K., and T.-L. Ku (1983), The elemental composition
types of deep-sea nodules from the Pacific Ocean, Miner. and decay-series radio-nuclide content of plankton from the
Deposita, 16, 59–84, doi:10.1007/BF00206455. east Pacific, Chem. Geol., 39, 125–145, doi:10.1016/0009-
Hein, J. R., and A. Koschinsky (2012), Deep-ocean ferroman- 2541(83)90076-1.
ganese crusts and nodules, in Treatise on Geochemistry, Korschinek, G., et al. (2010), A new value for the half-life
vol. 12, edited by S. Scott, Elsevier, New York, in press. of 10Be by heavy-ion elastic recoil detection and liquid
Hein, J. R., and C. L. Morgan (1999), Influence of substrate scintillation counting, Nucl. Instrum. Methods Phys. Res.,
rocks on Fe-Mn crust composition, Deep Sea Res., Part I, Sect. B, 268, 187–191, doi:10.1016/j.nimb.2009.09.020.
46, 855–875, doi:10.1016/S0967-0637(98)00097-1. Koschinsky, A., and P. Halbach (1995), Sequential leaching of
Hein, J. R., W. C. Schwab, and A. S. Davis (1988), Cobalt and marine ferromanganese precipitates: Genetic implications,
platinum-rich ferromanganese crusts and associated substrate Geochim. Cosmochim. Acta, 59, 5113–5132, doi:10.1016/
rocks from the Marshall Islands, Mar. Geol., 78, 255–283, 0016-7037(95)00358-4.
doi:10.1016/0025-3227(88)90113-2. Koschinsky, A., and J. R. Hein (2003), Uptake of elements
Hein, J. R., W. A. Bohrson, M. S. Schulz, M. Noble, and D. A. from seawater by ferromanganese crusts: Solid phase associ-
Clague (1992), Variations in the fine-scale composition of a ation and seawater speciation, Mar. Geol., 198, 331–351,
central Pacific ferromanganese crust: Paleoceanographic impli- doi:10.1016/S0025-3227(03)00122-1.
cations, Paleoceanography, 7, 63–77, doi:10.1029/91PA02936. Kubik, P. W., and M. Christl (2010), 10Be and 26Al measure-
Hein, J. R., H.-W. Yeh, S. H. Gunn, W. V. Sliter, L. M. ments at the Zurich 6 MV Tandem AMS facility, Nucl.
Benninger, and C.-H. Wang (1993), Two major Cenozoic Instrum. Methods Phys. Res., Sect. B, 268, 880–883,
episodes of phosphogenesis recorded in equatorial Pacific sea- doi:10.1016/j.nimb.2009.10.054.
mount deposits, Paleoceanography, 8, 293–311, doi:10.1029/ Lao, Y., R. F. Anderson, W. S. Broecker, H. J. Hofmann, and
93PA00320. W. Wolfli (1993), Particulate fluxes of 230Th, 231Pa, and 10Be
Hein, J. R., A. Koschinsky, M. Bau, F. T. Manheim, J.-K. in the northeastern Pacific Ocean, Geochim. Cosmochim.
Kang, and L. Roberts (2000), Cobalt-rich ferromanganese Acta, 57, 205–217, doi:10.1016/0016-7037(93)90479-G.
crusts in the Pacific, in Handbook of Marine Mineral Deposits, Leinen, M., and G. R. Heath (1981), Sedimentary indicators of
edited by D. S. Cronan, pp. 239–279, CRC Press, Boca Raton, atmospheric activity in the northern hemisphere during the
Fla. Cenozoic, Palaeogeogr. Palaeoclimatol. Palaeoecol., 36,
Hein, J. R., A. Koschinsky, and B. R. McIntyre (2005), 1–21, doi:10.1016/0031-0182(81)90046-8.
Mercury- and silver-rich ferromanganese-oxides, Southern Leinen, M., J. M. Prospero, E. Arnold, and M. Blank (1994),
California Borderland: Deposit model and environmental Mineralogy of aeolian dust reaching the North Pacific
implications, Econ. Geol., 100(6), 1151–1168, doi:10.2113/ Ocean: 1. Sampling and analysis, J. Geophys. Res., 99,
gsecongeo.100.6.1151. 21,017–21,023, doi:10.1029/94JD01735.
22 of 23
Geochemistry 3
Geophysics
Mahoney, J. J., R. A. Duncan, M. L. G. Tejada, W. W. Sager, Ocean: Effect of bottom scavenging, Earth Planet. Sci. Lett.,
and T. J. Bralower (2005), Jurassic-Cretaceous boundary age 339–340, 139–150, doi:10.1016/j.epsl.2012.05.012.
and mid-ocean-ridge-type mantle source for Shatsky Rise, Rea, D. K. (1994), The paleoclimatic record provided by eolian
Geology, 33, 185–188, doi:10.1130/G21378.1. deposition in the deep sea: The geologic history of wind, Rev.
Manheim, F. T., and C. M. Lane-Bostwick (1988), Cobalt Geophys., 32, 159–195, doi:10.1029/93RG03257.
in ferromanganese crusts as a monitor of hydrothermal Rogers, T. D. S., R. A. Hodkinson, and D. S. Cronan (2001),
discharge on the sea floor, Nature, 335, 59–62, doi:10.1038/ Hydrothermal manganese deposits form the Tonga-Kermadec
335059a0. Ridge and Lau Basin Region, Southwest Pacific, Mar. Georesour.
Mantyla, A. W., and J. L. Reid (1983), Abyssal characteristics Geotechnol., 19, 245–268, doi:10.1080/106411901753335326.
of the world ocean waters, Deep Sea Res., 30, 805–833, Roy-Barman, M., J. H. Chen, and G. J. Wasserburg (1996),
230
doi:10.1016/0198-0149(83)90002-X. Th -232Th systematics in the central Pacific Ocean: The
Martin, J. H., and G. A. Knauer (1973), The elemental com- sources and the fates of thorium, Earth Planet. Sci. Lett.,
position of plankton, Geochim. Cosmochim. Acta, 37, 139, 351–363, doi:10.1016/0012-821X(96)00017-9.
1639–1653, doi:10.1016/0016-7037(73)90154-3. Saager, P. M., H. J. W. de Baar, J. T. M. de Jong, R. F. Nolting,
McLennan, S. M. (1989), Rare earth elements in sedimentary and J. Schijf (1997), Hydrography and local sources of dis-
rocks: Influence of provenance and sedimentary processes, in solved trace metals Mn, Ni, Cu, and Cd in the northeast
Geochemistry and Mineralogy of Rare Earth Elements, Rev. Atlantic Ocean, Mar. Chem., 57, 195–216, doi:10.1016/
Mineral., vol. 21, edited by B. R. Lipin and G. A. McKay, S0304-4203(97)00038-8.
pp. 169–200, Mineral. Soc. of Am., Washington, D. C. Sager, W. W. (2005), What built Shatsky Rise, a mantle plume
Miller, S., and D. S. Cronan (1994), Element supply to sur- or ridge tectonics?, in Plates, Plumes, and Paradigms, edited
face sediments and interrelationships with nodules along by G. R. Foulger et al., Spec. Pap. Geol. Soc. Am., 388,
the Aitutaki-Jarvis Transect, South Pacific, J. Geol. Soc., 721–733, doi:10.1130/0-8137-2388-4.721.
151, 403–412, doi:10.1144/gsjgs.151.2.0403. Sager, W. W. (2006), Cretaceous paleomagnetic apparent polar
Monteiro, P. M. S., and M. J. Orren (1985), Trace metals in the wander path for the Pacific plate calculated from Deep Sea
southern ocean: On the geochemistry of copper, Mar. Chem., Drilling Project and Ocean Drilling Program basalt cores,
15, 345–355, doi:10.1016/0304-4203(85)90045-3. Phys. Earth Planet. Inter., 156, 329–349, doi:10.1016/j.pepi.
Moran, S. B., and R. M. Moore (1992), Kinetics of the removal 2005.09.014.
of dissolved aluminum by diatoms in seawater: A comparison Sager, W. W., and H.-C. Han (1993), Rapid formation of the
with thorium, Geochim. Cosmochim. Acta, 56, 3365–3374, Shatsky Rise oceanic plateau inferred from its magnetic
doi:10.1016/0016-7037(92)90384-U. anomaly, Nature, 364, 610–613, doi:10.1038/364610a0.
Müller, A. M., M. Christl, J. Lachner, M. Suter, and H.-A. Sager, W. W., J. Kim, A. Klaus, M. Nakanishi, and L. M.
Synal (2010), Competitive 10Be measurements below 1 MeV Khankishieva (1999), Bathymetry of Shatsky Rise, northwest
with the upgraded ETH–TANDY AMS facility, Nucl. Instrum. Pacific Ocean: Implications for ocean plateau development
Methods Phys. Res., Sect. B, 268(17–18), 2801–2807, at a triple junction, J. Geophys. Res., 104, 7557–7576,
doi:10.1016/j.nimb.2010.05.104. doi:10.1029/1998JB900009.
Murray, J. W., and J. G. Dillard (1979), The oxidation of cobalt(II) Tatsumi, Y., H. Shinjoe, H. Ishizuka, W. W. Sager, and A. Klaus
adsorbed on manganese dioxide, Geochim. Cosmochim. (1998), Geochemical evidence for a mid-Cretaceous super-
Acta, 43, 781–787, doi:10.1016/0016-7037(79)90261-8. plume, Geology, 26, 151–154, doi:10.1130/0091-7613(1998)
Nakanishi, M., W. W. Sager, and A. Klaus (1999), Magnetic 026<0151:GEFAMC>2.3.CO;2.
lineations within Shatsky Rise, northwest Pacific Ocean: van Andel, T. H., G. R. Heath, and T. C. Moore Jr. (1975),
Implications for hot spot-triple junction interaction and oce- Cenozoic History and Paleoceanography of the Central
anic plateau formation, J. Geophys. Res., 104, 7539–7556, Equatorial Pacific Ocean, Mem. Geol. Soc. Am., 143, 134 pp.
doi:10.1029/1999JB900002. Verlaan, P. A., D. S. Cronan, and C. L. Morgan (2004), A com-
Nozaki, Y., H.-S. Yang, and M. Yamada (1987), Scavenging parative analysis of compositional variations in and between
of thorium in the ocean, J. Geophys. Res., 92, 772–778, marine ferromanganese nodules and crusts in the South
doi:10.1029/JC092iC01p00772. Pacific and their environmental controls, Prog. Oceanogr.,
Okubo, A., H. Obata, T. Gamo, and M. Yamada (2012), 230Th 63, 125–158, doi:10.1016/j.pocean.2004.11.001.
and 232Th distributions in mid-latitudes of the North Pacific
23 of 23

Hein Et Al 2012

Uploaded by

Copyright:

Available Formats

Hein Et Al 2012

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hein Et Al 2012

Uploaded by

Copyright:

Available Formats

Article

Volume 13, Number 10

Copper-nickel-rich, amalgamated ferromanganese

Components: 16,700 words, 8 figures, 9 tables.

©2012. American Geophysical Union. All Rights Reserved. 1 of 23

1. Introduction occurring at depths of 800–2500 m. The mean dry

Table 1. Location of Cruise TN037 Dredges Containing Analyzed Fe-Mn Deposits

Table 2. Chemical Composition of Select Fe-Mn Crust Substrate Rocksa

H2O+ 6.8 3.6 6.4 1.9 2.5 6.2

Ba (ppm) >10000 193 812 <20 <20 2390

CaO/P2O5 1.95 19.9 1.91 1.63 1.63 1.54

Rock type claystone basalt basalt phosphorite phosphorite breccia

D14–16, the REEs increase away from layer 3,

Table 8. Be Isotope Ages of Four Crusts From Cruise TN037

D3-1B 0-46 2305 4551 18 0.0–3.4 Co 4227

D9-12B 0-20 681 1387 8 0–1.2 Co 3900

D10-11C 0-11 1360 3959 12 0–1.2 Co 3474

You might also like

Hein Et Al 2012

Uploaded by

Copyright:

Available Formats

Hein Et Al 2012

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hein Et Al 2012

Uploaded by

Copyright:

Available Formats

Article

Volume 13, Number 10

Copper-nickel-rich, amalgamated ferromanganese

Components: 16,700 words, 8 figures, 9 tables.

©2012. American Geophysical Union. All Rights Reserved. 1 of 23

1. Introduction occurring at depths of 800–2500 m. The mean dry

Table 1. Location of Cruise TN037 Dredges Containing Analyzed Fe-Mn Deposits

Table 2. Chemical Composition of Select Fe-Mn Crust Substrate Rocksa

H2O+ 6.8 3.6 6.4 1.9 2.5 6.2

Ba (ppm) >10000 193 812 <20 <20 2390

CaO/P2O5 1.95 19.9 1.91 1.63 1.63 1.54

Rock type claystone basalt basalt phosphorite phosphorite breccia

D14–16, the REEs increase away from layer 3,

Table 8. Be Isotope Ages of Four Crusts From Cruise TN037

D3-1B 0-46 2305 4551 18 0.0–3.4 Co 4227

D9-12B 0-20 681 1387 8 0–1.2 Co 3900

D10-11C 0-11 1360 3959 12 0–1.2 Co 3474

You might also like

1. Introduction occurring at depths of 800–2500 m. The mean dry