Biogeochemistry of Pacific deep-sea sediments and

Biogeochemistry of Pacific deep-sea sediments

and potential impacts of deep-sea polymetallic

nodule mining

by

Sophie Anna Luise Paul

a Thesis submitted in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

in Geosciences

Approved Dissertation Committee

___________________________________ Prof. Dr. Andrea Koschinsky

Jacobs University Bremen

____________________________________________

Prof. Dr. Michael Bau

Jacobs University Bremen

____________________________________________

Prof. Dr. Sabine Kasten

Alfred Wegener Institute Helmholtz Centre for Polar and

Marine Research

____________________________________________

Dr. Thomas Kuhn

German Federal Institute for Geosciences and Natural

Resources

Date of Defense: 29th October 2018

Department of Physics & Earth Sciences

Contents

Summary ............................................................................................................................... I

List of Figures...................................................................................................................... V

List of Tables .................................................................................................................... XIII

Chapter 1 - Introduction ...................................................................................................... 1

1. Scope of the Thesis ........................................................................................................ 1

2. Outline ............................................................................................................................ 3

3. Background .................................................................................................................... 5

3.1. Sediments in the Pacific Ocean ............................................................................... 5

3.2. The redox zonation of deep-sea sediments ............................................................. 5

3.3. Trace metals in deep-sea sediments ....................................................................... 7

3.4. Rare earth elements and yttrium (REY) in deep-sea sediments .............................. 8

3.5. Deep-sea mineral resources and mining ................................................................11

3.5.1. Polymetallic nodules........................................................................................12

3.5.2. Polymetallic nodule mining and potential environmental impacts.....................13

4. Study sites .....................................................................................................................16

4.1. DISCOL area, Peru Basin ......................................................................................17

4.2. Clarion Clipperton Zone (CCZ) ...............................................................................19

5. Methods ........................................................................................................................21

5.1. Sampling ................................................................................................................21

5.2. Analytical methods .................................................................................................23

5.2.1. Acid pressure digestion ...................................................................................23

5.2.2. ICP-OES .........................................................................................................23

5.2.3. ICP-MS ...........................................................................................................24

5.2.4. Analytical quality of major and trace element analyses ....................................25

5.2.5. Dissolved organic carbon (DOC) .....................................................................31

5.2.6. Dissolved amino acids (DAA) ..........................................................................31

5.2.7. Oxygen............................................................................................................32

5.2.8. Particulate organic carbon, carbonate, and particulate organic nitrogen ..........32

5.2.9. Nitrate .............................................................................................................32

5.2.10. Porosity ...........................................................................................................32

Chapter 2 – Small-scale heterogeneity of trace metals including REY in deep-sea

sediments and pore waters of the Peru Basin, SE equatorial Pacific .............................33

Abstract ................................................................................................................................35

1. Introduction ...................................................................................................................36

1.1. Fragmentary data sets from the deep-sea ..............................................................36

1.2. Previous work in the Peru Basin .............................................................................36

1.3. Early diagenesis in the Peru Basin .........................................................................38

1.4. Fe-rich clay minerals ..............................................................................................38

1.5. Rare earth elements and yttrium (REY) ..................................................................38

1.6. Research aim .........................................................................................................39

2. Methods ........................................................................................................................40

2.1. Sampling area and methods ...................................................................................40

2.2. Sediment and pore water sampling ........................................................................41

2.3. Chemical analyses .................................................................................................42

2.4. Nitrate ....................................................................................................................42

2.5. Particulate organic carbon (POC) and CaCO3 ........................................................43

2.6. Depth correction for GCs and CaCO3 correction ....................................................43

2.7. Reporting of REY data ...........................................................................................44

3. Results ..........................................................................................................................44

3.1. Core descriptions ...................................................................................................44

3.2. Solid phase Ca, CaCO3, Ba, Al, Fe, Mn and associated metals .............................47

3.3. Pore water Mn, Co, Cu ...........................................................................................49

3.4. Redox-sensitive elements Mo, U, As, V, Cd: solid phase and pore water ...............50

3.5. REY patterns ..........................................................................................................52

4. Discussion .....................................................................................................................53

4.1. Paleoceanographic context: sedimentation history based on CaCO3 and Ba

preservation ......................................................................................................................53

4.2. Green layers...........................................................................................................55

4.3. Sedimentary Fe/Al ..................................................................................................56

4.4. REY control phases ................................................................................................56

4.5. Dissolved and solid phase Mn and associated metals ............................................60

4.6. Redox-sensitive elements Mo, U, As, and V ...........................................................61

5. Conclusions ...................................................................................................................62

Acknowledgements ..............................................................................................................63

Chapter 3 - Biogeochemical regeneration of a nodule mining disturbance site: trace

metals, DOC and amino acids in deep-sea sediments and pore waters ........................65

Abstract ................................................................................................................................67

1. Introduction ...................................................................................................................68

2. Materials and methods ..................................................................................................71

2.1. Site description .......................................................................................................71

2.2. Sediment and pore water sampling ........................................................................73

2.3. Chemical analyses .................................................................................................74

2.3.1. Solid phase .....................................................................................................74

2.3.2. Pore water .......................................................................................................75

3. Results ..........................................................................................................................77

3.1. Solid phase ............................................................................................................77

3.2. Pore water ..............................................................................................................80

4. Discussion .....................................................................................................................84

4.1. Solid phase: the Mn-oxide rich layer in the undisturbed sites .................................84

4.2. Disturbance impacts on the solid phase: sediment removal, redeposition, and

inversion ...........................................................................................................................86

4.3. Pore water natural state and impacts visible 5 weeks post-disturbance..................88

4.4. Trace metal fluxes to the ocean..............................................................................89

4.5. DOC and DAA as indicators of organic matter degradation ....................................91

5. Conclusion ....................................................................................................................91

Author contributions .............................................................................................................93

Acknowledgments ................................................................................................................93

Supplementary material ........................................................................................................94

Chapter 4 – Calcium phosphate control of REY patterns of siliceous-ooze-rich deep-sea

sediments from the central equatorial Pacific ..................................................................95

Abstract ................................................................................................................................97

1. Introduction ...................................................................................................................98

2. Samples and methods ................................................................................................. 102

2.1. Geological setting of the study site ....................................................................... 102

2.2. Sampling .............................................................................................................. 103

2.3. Analytical methods ............................................................................................... 104

2.4. Reporting ............................................................................................................. 105

2.5. Sequential extraction ............................................................................................ 105

2.6. Scanning electron microscopy .............................................................................. 106

3. Results ........................................................................................................................ 107

3.1. Bulk solid-phase major elements .......................................................................... 107

3.2. REY concentrations and shale-normalized patterns of bulk solid phase ............... 109

3.3. Sequential extraction ............................................................................................ 112

3.4. Scanning electron microscopy .............................................................................. 113

4. Discussion ................................................................................................................... 113

4.1. Controls on REY composition of bulk sediment .................................................... 113

4.2. Sequential extraction: REY in phosphate phases ................................................. 116

4.3. Alteration of biogenic Ca phosphates during early diagenesis .............................. 117

4.4. Pore-water REY pool ............................................................................................ 119

4.5. Increase of negative CeSN anomaly with depth ..................................................... 120

5. Conclusions ................................................................................................................. 123

Acknowledgements ............................................................................................................ 124

Appendix A. Supplementary Material .................................................................................. 125

Chapter 5 - Rare earth elements and yttrium in metalliferous and calcium-carbonate-rich

sediments from the central equatorial Pacific ................................................................ 127

Abstract .............................................................................................................................. 129

1. Introduction ................................................................................................................. 130

2. Methods ...................................................................................................................... 131

2.1. Sampling .............................................................................................................. 131

2.2. Chemical analyses ............................................................................................... 132

2.3. Reporting ............................................................................................................. 133

3. Results ........................................................................................................................ 134

3.1. Ca-, Mn-, and Fe-rich layers ................................................................................. 134

3.2. Rare earth elements and yttrium (REY) ................................................................ 135

4. Discussion ................................................................................................................... 137

4.1. REY control phase ............................................................................................... 137

4.2. REY concentrations with depth ............................................................................ 140

4.3. MREY enrichment ................................................................................................ 140

4.4. CeSN anomaly ....................................................................................................... 141

5. Conclusions ................................................................................................................. 141

Chapter 6 – Conclusions and Outlook ............................................................................ 143

Chapter 7 – Related scientific work ................................................................................ 149

1. Research cruises and sampling campaigns ................................................................. 149

1.1. Research cruises ................................................................................................. 149

1.2. Land-based sampling campaigns ......................................................................... 149

2. Conferences ................................................................................................................ 150

3. Co-supervised Bachelor thesis and guided research projects...................................... 151

Acknowledgements .......................................................................................................... 153

References ........................................................................................................................ 155

Appendix I ......................................................................................................................... 173

Appendix II ........................................................................................................................ 178

Appendix III ....................................................................................................................... 187

Appendix IV ...................................................................................................................... 199

I

Summary

This cumulative PhD thesis explores trace metal distributions in deep-sea sediments and pore

waters in two manganese nodule areas of the central equatorial Pacific, namely the Peru Basin

and the Clarion Clipperton Zone (CCZ). In the light of new developments in the deep-sea

mining industry and policy making, the environmental assessment of the possible impacts of

deep-sea mining on the deep seafloor and the biogeochemical processes in the sediment and

at the sediment-water interface is critical. For the analysis of mining impacts, a key area of

focused research is on the Mn-oxide-rich surface layer, which is also rich in the associated

metals Mo, Co, Ni, and Cu, and will be the layer most likely to be impacted by mining.

The deep-sea is a vast area, but at present, remains greatly understudied. The abyssal plains,

where polymetallic nodules are found, have been of less scientific interest historically than

geologically more active continental margins, upwelling areas, and spreading centers.

Because of this, the degree of heterogeneity in these areas is not well known, and before

conclusions may be drawn for large areas, further research is needed to ensure that research

samples are representative of large regions of seafloor. Baseline studies are especially

important when anthropogenic impacts such as polymetallic nodule mining are to be analyzed,

because natural variability of these remote areas needs to be understood before impacts from

mining can be clearly separated from temporal or spatial variabilities.

Small-scale heterogeneity of deep-sea sediments and contained pore water was evaluated in

the Peru Basin. The analysis of seven 10 m long cores from an approx. 13 km wide area in the

Peru Basin revealed a surprisingly variable system with respect to different sediment layers,

redox zonation (e.g., Fe(III) reduction in the clay minerals), and nodule abundance on top and

within the sediment. After a thorough analyses, small differences in bathymetry, organic matter

content, and presence of buried nodules were concluded to be the basis for measured

differences in metal concentrations and metal cycling of Mn, Fe, Co, Cu, Mo, U, As, V, and

REY. In areas with lower particulate organic carbon (POC) contents, nitrate is not consumed

and hence Fe(II) in the clay minerals is oxidized to Fe(III), whereas in cores with higher POC

contents, nitrate is consumed within the upper 2-3 m, with Fe(II) in the clay minerals not

oxidized to Fe(III), which also leads to a visible change in color of sediments from tan to green.

Slightly deeper areas of seafloor, such as troughs, show a higher abundance of buried nodules

than elevated areas as well as dark gray bands with solid phase and pore water peaks of the

redox-sensitive elements U, Mo, V, and As. These results highlight the importance for thorough

baseline studies before extrapolating scientific results to larger areas as well as before setting

up reference sites and monitoring regimes for deep-sea mining.

II

As part of this thesis project, 26-year old plow tracks in the DISCOL (DISturbance and

reCOLonization) area of the Peru Basin were revisited in 2015. The comparison of undisturbed

sites representing pre-disturbance sediment with 26-year and 5-week old disturbed sites

showed that the pore water metal distribution remained impacted after five weeks but had

regained an equilibrium after 26 years. The solid phase, however, still showed impacts within

the upper approx. 20 cm of sediments after 26 years. In the Peru Basin, this affects the surface

layer rich in Mn oxides and associated metals such as Mo, Ni, Co (and Cu). An additional key

finding of this study was that there is a high heterogeneity with respect to disturbance impacts.

Sites show sediment removal, mixing, deposition of resuspended sediment, and turnover of

surface sediment layers into deeper layers. The differences in scope and severity of these

impacts could not only be observed between time-scales of recovery but also could be

attributed to the use of different gear to create disturbances.

The focus within the CCZ was on rare earth elements and yttrium (REY), their distribution in

the sediments and the changes of the shale-normalized (SN) REY pattern with depth that

deviates from the seawater pattern and may give insights into impacts of early diagenesis on

REY fractionation. We could show that the REY in siliceous-ooze-rich central equatorial Pacific

sediments are predominantly bound in the Ca phosphate phase and are continuously

incorporated from ambient pore water without major fractionation. Both, pore water and solid

phase, show middle REY enrichment over the light and heavy REY in these sediments, as well

as no or negative CeSN anomalies. With depth, the REY concentration and middle REY

enrichment was measured to increase, as did the magnitude of the negative CeSN anomaly.

The latter due to the concentration increase of trivalent REY in the Ca phosphate phase, while

most of the Ce is bound in Mn and/or Fe phases, which is not continuously exchanged with

the pore water due to oxidative scavenging of Ce in the surface sediments. In carbonate-rich

or metalliferous sediments, the phase association is less clear, even though all sediments

showed middle REY enrichment and a negative CeSN anomaly. In these sediments, however,

the REY concentration does not increase with depth. Carbonate layers dilute the REY

concentrations, while some layers rich in Mn or Fe accumulated REY while other Mn- or Fe-

rich layers were found to be relatively depleted in REY. These variabilities in content are a

further indication of the high heterogeneity of deep-sea sediments. All REYSN patterns from the

CCZ contrast with REYSN patterns from the Peru Basin, where we see heavy REY enrichment

with negative CeSN and positive LaSN, GdSN, and YSN anomalies. In the Peru Basin, the REY

appear to be associated with Fe-rich clay or phosphates but in any case, take over the pore

water pattern without major fractionation.

III

Overall, this thesis sheds light on the small-scale heterogeneity of deep-sea sediments and

the contained pore waters in the Peru Basin, reflected in the depth distribution of POC, nitrate,

Mn, and Fe. Based on the thorough geochemical analysis, oxygen and solid phase Mn were

identified as key parameters for monitoring of potential future mining related impacts because

they are representative of the redox zonation and the behavior of other metals. The work

provides detailed baseline data for solid phase Al, Ca, Fe, Mn, Mo, Ni, Co, Cu, U, V, Cd, Pb,

Zn, and REY and pore water Mn, Co, Cu, Mo, U, V, Cd, As, and REY that increases our

knowledge of the biogeochemistry of trace metals in deep-sea sediments and provides a basis

to assess future potential anthropogenic impacts, e.g., polymetallic nodule mining. It

furthermore increases our understanding of the distribution of REY in deep-sea sediments,

and the ongoing alteration of REYSN patterns during early diagenesis in the CCZ and the Peru

Basin in a variety of redox conditions.

IV

V

List of Figures

Chapter 1 – Introduction

Figure 1: Left: Redox paradigm after Froelich et al. (1979) from Burdige (1993). Reprinted

from Earth-Science Reviews, 35/3, Burdige, D.J., The biogeochemistry of manganese and iron

reduction in marine sediments, 249-284, Copyright (1993), with permission from Elsevier.

Right: Revised redox paradigm for oxic and suboxic sediments after Madison et al. (2013),

including dissolved Mn(III) and Fe(III).

Figure 2: Visualization of the lanthanide contraction: REE show decreasing ionic radii with

increasing atomic number. Note similar size of Y and Ho, explaining why Y is often included

with the REE. Also note significant differences in ionic radii sizes of tetravalent Ce and divalent

Eu. After Merschel (2017) with data from Shannon (1976).

Figure 3: Example of REY normalization. Raw PAAS data displays concentration differences

between even and odd atomic numbers. Once normalized to chondrite, smooth patterns can

be interpreted. After Merschel (2017). PAAS data from Taylor and McLennan (1985), except

for Dy from McLennan (1989); C1 chondrite data from Anders and Grevesse (1989).

Figure 4: Schematic drawing of the mining set up consisting of system 1, the mining platform

and the transport vessel at the sea surface, system 2, the riser pipe, system 3, the nodule

collector, and system 4, the discharge of e.g., leftover sediment and nodule material after

processing on the mining platform. Reprinted from Deep Sea Research Part II: Topical Studies

in Oceanography, 48/17-18, Oebius, H.U., Becker, H.J., Rolinski, S., Jankowski, J.A.,

Parametrization and evaluation of marine environmental impacts produced by deep-sea

manganese nodule mining, p.3455, Copyright (2001), with permission from Elsevier.

Figure 5: Overview of potential geochemical impacts and processes after a disturbance in the

sediment, at the sediment surface, and in the bottom water.

Figure 6: Map showing locations of the two study sites from this PhD thesis: the Clarion

Clipperton Zone and the Peru Basin. The map was created using GeoMapApp.

Figure 7: Bathymetric map of the DISCOL area in the Peru Basin adapted from Paul et al.

(2018). The circle shows the DEA (DISCOL experimental area). Sampling locations are

marked with red dots.

VI

Figure 8: Map showing the contractor areas (BGR: Bundesanstalt für Geowissenschaften und

Rohstoffe, Germany, IOM: InterOceanMetal, eastern European consortium, GSR: Global Sea

Mineral Resources NV, Belgium, IFREMER: L’Institut Français de Recherche pour

l’Exploitation de la Mer) and the Area of Particular Environmental Interest (APEI) 3 visited

during SO239 in 2015 and the POC flux (Vanreusel et al., 2016; Creative Commons Attribution

4.0 International Public License).

Figure 9: Average reference and measured values of CRMs used for ICP-MS pore-water

analyses for Mn with standard deviation. The two SLEW-3 measured values are for results in

Chapter 2 and Chapter 3, respectively. “n” refers to the number of ICP-MS runs, in which the

CRM was usually measured multiple times.


analyses for Cu with standard deviation. The two SLEW-3 measured values are for results in




analyses for Mo with standard deviation. The two NASS-6 measured values are for results in




analyses for V with standard deviation. The two NASS-6 and SLEW-3 measured values are

for results in Chapter 2 and Chapter 3, respectively. “n” refers to the number of ICP-MS runs,

in which the CRM was usually measured multiple times.

Figure 13: Average reference and measured values of CRMs used for ICP-OES solid-phase

analyses for P with standard deviation. The two MESS-3 and BHVO-2 measured values are

for results in Chapter 2 and Chapter 3, respectively. The two NIST-2702 measured values are

for Chapter 4 and Chapter 5, respectively. “n” refers to the number of digestions and each

digested sample was usually measured multiple times during one ICP-MS run.


analyses for Fe with standard deviation. The two MESS-3 and BHVO-2 measured values are


VII



Figure 15: Average reference and measured values of CRMs used for ICP-MS solid-phase

analyses for Ce with standard deviation. The two NIST-2702 measured values are for results

in Chapter 4 and Chapter 5, respectively. The three BHVO-2 measured values are for Chapter

2, Chapter 4, and Chapter 5, respectively. “n” refers to the number of digestions and each



analyses for Nd with standard deviation. The two NIST-2702 measured values are for results




Chapter 2 - Small-scale heterogeneity of trace metals including REY in deep-sea

sediments and pore waters of the Peru Basin, SE equatorial Pacific

Figure 1: Map showing the Peru Basin and the location of the DISCOL area. The map was

created using GeoMapApp and its integrated default basemap Global Multi-Resolutional

Topography (GMRT).

Figure 2: Map of the sampling locations in the Peru Basin. Bathymetric map adapted from

Paul et al. (2018). The circle indicates the DISCOL experimental area (DEA) that was traversed

with a plow harrow.

Figure 3: Combined photos of the individual GCs with corresponding nitrate profiles. Green

layers are marked with green boxes.

Figure 4: POC profiles of the GCs.

Figure 5: Depth profiles of solid phase Ca, CaCO3, and Ba concentrations, as well as Ba/Al

ratios, highlighting two layers where carbonate was preserved.

Figure 6: Solid phase Al, Fe, Mn, P, Nd, Cu, Ni, and Co concentrations in the sediment cores

including those of the buried nodules at Reference West at 458 cm, at DEA Trough at 387 cm,

468 cm and 667 cm, and at Reference East at 290 cm depth. Nd is shown as a representative

VIII

of the REY. Fe/Al and Mn/Al ratios in the sediments (i.e. no data for the nodules is shown) are

also displayed as depth profiles, focusing on the Fe and Mn enrichment in relation to

continental sources (Al).

Figure 7: Dissolved Mn, Co, and Cu concentrations in the pore water of the sediment cores.

No pore water could be extracted from buried nodules.

Figure 8: Top: Solid phase concentrations of U, Mo, and V. Concentration peaks are visible at

229.5, 236.5 cm and 330 cm for Reference East coinciding with the gray bands in the sediment

(see pictures on the right). In this core, also a dissolving nodule was found at 290 cm (see

pictures on the right). Bottom: Dissolved concentrations of U, Mo, V, As, and Cd in the pore

water. Depths 229.5 cm and 290 cm of Reference East were not measured. Concentration

peaks are visible at 236.5 cm and 330 cm for Reference East coinciding with the gray bands

in the sediment (see pictures on the right).

Figure 9: REYSN patterns of the seven cores from this study and for the clay minerals

nontronite, illite, and kaolinite from literature for comparison.

Figure 10: Measurable REYSN pore water patterns from the Peru Basin.

Figure 11: Top: Fe-Nd plot and correlations for all cores. Pearson R coefficients show positive

correlations of REY with Fe for all cores. Middle: Al-Fe plot. Only positive correlations for the

upper parts of Reference South and DEA West are shown. Bottom: Al-Nd plot. Only positive

correlations for the upper part of Reference South, as well as for the lower part of Reference

West and the entire Small Crater core.

Figure 12: Top: P-Fe correlations for Reference South, DEA West, Reference West, and DEA

Black Patch. Middle: P-Ca correlations for samples with Ca concentrations below 1.5 wt.%

except for Reference East where P and Ca do not correlate. Samples with Ca concentrations

above 1.5 wt.% were excluded from the regression analyses because most of the Ca is then

not bound in Ca phosphates. Bottom: P-Nd correlations for all samples except Small Crater

where P and Nd do not correlate and excluding the DEA Black Patch sample with exceptionally

high P concentrations.

IX

Chapter 3 – Biogeochemical regeneration of a nodule mining disturbance site:

trace metals, DOC and amino acids in deep-sea sediments and pore waters

Figure 1: Sampling sites of sediment cores in the DISCOL area (adapted from a map by Anne

Peukert, GEOMAR, working group of Jens Greinert). The circle indicates the DISCOL

experimental area (DEA) in which the disturbance experiment had been carried out in 1989.

Figure 2: (A) Example of seafloor at a reference site, (B) example of an EBS track, (C) example

of a 26-year old plow track, indicating the four microhabitats outside track, track valley, ripple,

and white patch. Pictures copyright ROV KIEL 6000 Team, GEOMAR Helmholtz Centre for

Ocean Research Kiel, Germany.

Figure 3: Sediment major element profiles and properties of the four undisturbed and six

disturbed sites.

Figure 4: Sediment element profiles of the four undisturbed sites. Reliable Cd results only for

outside EBS track.

Figure 5: Sediment element profiles of five 26-year old disturbed sites and the 5-week old EBS

track. Reliable Cd results only for DEA West plow track.

Figure 6: Bottom water and pore water ex-situ oxygen and nitrate profiles of two undisturbed

and two disturbed sites (MUCs). In each core, four to six oxygen profiles were measured.

Figure 7: Bottom water and pore water element profiles of the undisturbed sites. The

uppermost values refer to bottom water concentrations measured in the supernatant retrieved

above the sediment surface in the MUC liner.

Figure 8: Bottom and pore water element profiles of five 26-year old disturbed sites and the

5-week old EBS track. The uppermost values refer to bottom water concentrations measured

in the supernatant retrieved above the sediment surface in the MUC liner. Mn and Co below

the LOQ for DEA West plow track.

Figure 9: Sum of dissolved amino acid (DAA) concentration profiles of three undisturbed and

two disturbed sites. The uppermost values refer to bottom water concentrations measured in

the supernatant retrieved above the sediment surface in the MUC liner.

X

Chapter 4 – Calcium phosphate control of REY patterns of siliceous-ooze-rich

deep-sea sediments from the central equatorial Pacific

Fig. 1: REYSN patterns of fish debris, fossil fish teeth, marine phosphorite, hydrogenetic Fe-Mn

crust, and seawater (PAAS from Taylor and McLennan, 1985, except for Dy from McLennan,

1989). (See above mentioned references for further information.)

Fig. 2: Core sampling locations of 87GC and 165GC in the CCZ and of 194GC north of the

Clarion Fracture Zone. The map was created using GeoMapApp.

Fig. 3: Top: Depth profiles of selected major elements and three representative REY (Ce, Nd,

Yb). Yb concentrations were multiplied by 10 to fit the scale of the figure. Core pictures depict

that the sediment gets darker with depth in all cores and has thin dark layers throughout.

Oxygen data from Volz et al. (2018). Bottom: Depth profiles of REY parameters HREE/LREE,

MREE/MREE*, Ce/Ce*, and Y/Ho for bulk sediment, the sequential extraction solutions (Na-

dithionite only for HREE/LREE and MREE/MREE* for 194GC 561 cm), and pore water (Y/Ho

only for 194GC-511 cm). HREE/LREE = (Ho + Er + Tm + Yb + Lu)/(La + Ce + Pr + Nd).

MREE/MREE* = (Sm + Eu + Gd + Tb + Dy)/((La + Ce + Pr + Nd + Ho + Er + Tm + Yb + Lu)*2).

Fig. 4: REYSN patterns of selected sediment layers of the three cores investigated in this study

(PAAS from Taylor and McLennan, 1985, except for Dy from McLennan, 1989). All cores and

layers show a slight enrichment of MREY and HREY and most layers display a negative CeSN

anomaly.

Fig. 5: Increase of negative CeSN anomaly with depth and with increasing P concentration. See

Eq. (1) in chapter 2.4 for the calculation of the CeSN anomaly.

Fig. 6: Left: REYSN patterns of pore waters from 194GC (PAAS from Taylor and McLennan,

1985, except for Dy from McLennan, 1989). All patterns show an enrichment of the MREY and

a pronounced negative CeSN anomaly. Right: Bulk sediment and Ca phosphate phase

normalized to pore water.

Fig. 7: REYSN patterns of sequential leaching solutions of selected sediment layers from the

three cores (PAAS from Taylor and McLennan, 1985, except for Dy from McLennan, 1989).

From each core one sample from an upper and lower part of the core was selected. Some

data points are missing for the Na-dithionite and NH4-oxalate patterns due to concentrations

below the LOQ.

XI

Fig. 8: Scanning electron microscopy (SEM) images of particles rich in phosphorus and

calcium; examples from layers 165GC-792 cm (left and middle) and 165GC-812 cm (right).

Fig. 9: Left: P vs. Ca plot. Right: P vs. Nd plot. Nd represents the REY. Linear regression lines

for the cores in both graphs and Pearson R correlation coefficients in the legend. All cores

show positive correlations of P and Ca and P and Nd. The deepest layers in 165GC (792-

912 cm) and 194GC (521-561 cm) deviate from the linear regression due to a lower Nd/P ratio

(for further discussion see text).

Fig. 10: Nd/P ratio of bulk sediment at different depths for cores 87GC, 165GC and 194GC.

Nd represents the REY. Similar values with depth suggest that the ratio of REY to P stays the

same except in the deep layers (165GC 792–912 cm and 194GC 521–561 cm) where lower

Nd/P values suggest that P is more enriched than the REY. The relative uncertainty of Nd/P

based on NIST-2702 digestions (n = 12 for P and n = 10 for Nd) and measurements is 6.27%.

Fig. 11: Ce/Ce* values for each layer. Ce/Ce* was calculated according to equation (1) in the

text. Yellow star symbols denote no CeSN anomaly. Values decrease with depth in all three

cores, starting with different Ce/Ce* values at the top of the sediment cores.

Chapter 5 - Rare earth elements and yttrium in metalliferous and calcium-

carbonate-rich sediments from the central equatorial Pacific

Figure 1: Map of sampling locations. The map was created using GeoMapApp.

Figure 2: Depth profiles of selected major elements and three representative REY (Ce, Nd,

Yb). Yb concentrations were multiplied by 10 to fit the scale of the figure. Note that ca. 1.5 m

of 117SL were lost during sampling. Therefore, two layers (7.5 cm and 23 cm) of the

corresponding MUC (116MUC) were included to represent surface sediment. The sediments

are oxic throughout (Kuhn, 2015; Volz et al., 2018).

Figure 3: REYSN patterns of the three cores from this study. All cores show MREY enrichment

and negative CeSN anomalies.

Figure 4: P-Ca plot of the three cores from this study. Pearson R coefficients for linear

regressions in the Ca-poor parts of 69SL and 117SL, as well as for the entire 122GC core,

show a positive correlation of P and Ca.

XII

Figure 5: Nd (representing the REY) vs. P plots for the three cores from this study. An outlier

was excluded from the correlation analyses in cores 117SL and 69SL each. Correlations for

122GC were conducted in two parts and the core split at 546 cm, where the Fe-rich layer starts.

XIII

List of Tables


Table 1: Previous benthic impact experiments conducted in the CCZ. Information from Jones

et al., (2017).

Chapter 2 - Small-scale heterogeneity of trace metals including REY in deep-sea

sediments and pore waters of the Peru Basin, SE equatorial Pacific

Table 1: Overview of sampled cores.

Chapter 3 – Biogeochemical regeneration of a nodule mining disturbance site:

trace metals, DOC and amino acids in deep-sea sediments and pore waters

Table 1: Overview of cores taken for sediment and pore water trace metal analyses.

Table 2: Correlation coefficients of Mn and Fe with Cu, Co, Ni, and Mo, calculated in Excel.

Table 3: Diffusive fluxes of selected dissolved metal ions across the sediment-water interface

and potential fluxes across the sediment-water interface when the Mn oxide rich layer is

removed; based on gradients across the redox-boundary in cores from this study.

Chapter 4 – Calcium phosphate control of REY patterns of siliceous-ooze-rich

deep-sea sediments from the central equatorial Pacific

Table 1: Overview of GC sampling sites.

Table 2: Leaching scheme for the sequential extraction of Mn- and Fe-(oxyhydr)oxides

(adapted from Köster, 2017).

XIV

Table 3: Pearson R correlation coefficients of Nd, representing the REY, with various major

elements of the bulk sediment digestions. Data correlated for the completely analyzed core

sections.

Chapter 5 - Rare earth elements and yttrium in metalliferous and calcium-

carbonate-rich sediments from the central equatorial Pacific

Table 1: Overview of core samples. SL= Schwerelot (German:gravity core), GC=gravity core,

MUC=multicore.


elements.


1

Chapter 1 - Introduction

1. Scope of the Thesis During the last decade, deep-sea mining has again received increasing attention; partly due

to periodically high metal prices, political interest, and technological advancements. Even

though at time of writing mining for polymetallic nodules has not yet commenced, resource

exploration by contractor states is underway in the Clarion Clipperton Zone (CCZ) in the central

equatorial Pacific Ocean and in the Indian Ocean Basin. One important component of the

deep-sea mining advances is the development of environmental regulations, and therefore

there has been a research drive to conduct environmental baseline studies, develop monitoring

strategies, and assess potential impacts. Previously, there have been various studies

investigating the potential environmental impacts of nodule mining – initially in the 1970s,

1980s, and 1990s, during the first “wave” of deep-sea mining – and recently, running 2015 to

2017, the European project Joint Programming Initiative Healthy and Productive Seas and

Oceans (JPI Oceans) Ecological Aspects of Deep-Sea Mining has integrated multidisciplinary

research from across Europe to study the deep-sea ecosystem, and most of the work carried

out in this PhD thesis has been conducted within the scope of this project. The aim of the JPI

Oceans project was to assess various environmental impacts of deep-sea mining on the

ecosystem, e.g., on fauna, food webs, and biogeochemical cycles. The aim of this thesis

project was to study the natural biogeochemical cycling and redox zonation at reference sites

in manganese nodule areas of the Pacific and to compare these with those operating within

disturbed sites, focusing on the Peru Basin, where a 26-year old disturbed site was revisited

in 2015. Besides identifying impacts on the solid phase and pore waters, another focus was

on assessing time-scales of biogeochemical regeneration.

Since the natural variability on the deep-sea was found to be unexpectedly high, thorough

analyses of the natural heterogeneity across small spatial scales within the Peru Basin, as well

as on a larger scale between the Peru Basin and the CCZ were conducted. The small-scale

comparison aimed to first understand natural background conditions of deep-sea sediments

and processes within these nodule provinces to then assess how they might be impacted by

deep-sea mining. Short- to medium-term regeneration of sediments after disturbances similar

to polymetallic nodule mining has been studied previously (Jones et al., 2017; Thiel, 2001;

Thiel and Schriever, 1990), but most investigations have focused on geology and biology

(Bluhm, 2001; Weber et al., 2000, 1995). The first (bio)geochemical work in the Peru Basin

was carried out seven years after the initial physical disturbance (Koschinsky, 2001;

Koschinsky et al., 2001b, 2001a; Marchig et al., 2001), so pre-disturbance baselines are


2

missing from the literature (Thiel and Schriever, 1990). A detailed analysis of natural

heterogeneity and variability of disturbance impacts on the sediment solid phase and pore

waters was still missing at the start of the JPI Oceans project, and filling this gap was the major

aim of this PhD thesis. The results from the environmental baseline and impact studies

conducted herein will be valuable for assessing likely future mining impacts on the deep-sea

ecosystem and therefore aid in the ongoing development of environmental regulations for

deep-sea mining of polymetallic nodules.

The analyses performed in this thesis project provide high spatial and depth resolution solid

phase and pore water data for (trace) metals important for the description of redox zonation

e.g., Mn, Fe, Co, Ni, and Cu, as well as for redox-sensitive elements, e.g., U, V, Mo, and As.

The results reveal that sediment composition and processes on the seafloor can vary on small

spatial scales and that scientists and policy makers need to exercise caution when looking for

representative sites. The analyses of a variety of disturbed sites of different ages and produced

with different gear, highlight the differences in impacts associated with particular individual

mechanical impact types, as well as improve our knowledge on the likely timescales over which

metal cycling may be impacted following disturbances.

A detailed knowledge of baseline conditions within nodule provinces is also essential for future

monitoring in case industrial nodule mining activities commence. The data acquired in the

frame of this thesis project can be used to develop monitoring guidelines, e.g., how monitoring

sites should be chosen and which key parameters would be suitable for monitoring.

A second aim of this thesis was to analyze the rare earth element and yttrium (REY) distribution

and patterns in sediments across different study sites in the Pacific Ocean. The REY behave

coherently in natural systems and anomalies or enrichments in normalized REY patterns can

be used to determine sediment provenance and alteration (e.g., Bau and Dulski, 1999; Bright

et al., 2009). Even though multiple studies of REY in sediments have previously been

conducted in the Pacific, they have tended to focus on surface sediments, sediment associated

with nodules, and ignored anomalous layers (e.g., Elderfield et al., 1981; Glasby et al., 1987;

Toyoda et al., 1990). From the Peru Basin, to the best of our knowledge, solely REY data from

hydrothermally impacted sites and shelf sediments have been thus far investigated (see e.g.,

Marchig et al., 1999; Piper et al., 1988). This PhD study provides further data on REY in deeper

sediment layers and thereby offers insights into the influence of early diagenetic processes on

REY and the REY exchange between pore water and sediments over long time intervals. We

show how shale-normalized (SN) REY patterns change with depth due to continuous

exchange of REY between the solid phase and ambient pore water and the dependence of


3

this evolution on surface water productivity and sediment composition, i.e. differences between

Ca phosphate, carbonate-rich, and metalliferous sediments.

2. Outline This cumulative PhD thesis is comprised of seven chapters. Chapter 1 is an Introduction,

explaining the scope of the thesis and its outline, presenting background information, the study

sites and state-of-the-art methods used in this thesis project. The Introduction is followed by a

manuscript introducing heterogeneity in deep-sea sediments, typical redox zonation and some

exceptions to the general rule, with the example of the Peru Basin (Chapter 2). Chapter 3 is a

published paper that discusses possible impacts of deep-sea mining on deep-sea sediments

and pore waters, again focusing on the Peru Basin. Chapters 4 and 5 focus on the second

study site covered by this PhD thesis, the CCZ. Chapter 4 is a submitted manuscript that

describes the dominating phase association of REY with Ca phosphates in deep-sea

sediments of the central equatorial Pacific, while Chapter 5 further explores exceptions to the

rule presented in Chapter 4. Each paper or manuscript consists of an abstract, an introduction,

a methods section, a results section, a discussion section and conclusions. The PhD thesis

ends with a ‘Conclusions and Outlook’ chapter (Chapter 6). This final chapter connects the

work from within the different chapters and highlights possibilities for future research foci,

followed by a brief description of related scientific work (Chapter 7). All references from this

thesis, including those from within the paper and manuscript chapters, are provided in the

combined reference list. Appendices from the published paper as well as the manuscripts are

presented after the references.

Chapter 1 introduces the objectives of the research undertaken within the framework of this

PhD project. Furthermore, background information and relevant literature on deep-sea

sediments and deep-sea mining are presented, as well as the two study sites and analytical

methods used to analyze major and trace elements in the solid phase and pore water. Further

methods to analyze parameters presented as part of the manuscripts are briefly described.

Chapter 2 focuses on the heterogeneity of deep-sea sediments in the Peru Basin. Early

diagenetic processes are analyzed in 10 m long cores. A similar redox zonation is present in

all cores but trends related to bathymetry and organic matter content are visible, even on small

scales. The data and discussion are presented in the manuscript Small-scale heterogeneity

of trace metals including REY in deep-sea sediments and pore waters of the Peru Basin,

SE equatorial Pacific which is in preparation for submission to the Special Issue Assessing

environmental impacts of deep-sea mining – revisiting decade-old benthic disturbances in

Pacific nodule areas of Biogeosciences.


4

Chapter 3 focuses on surface sediments from the Peru Basin and potential impacts of deep-

sea polymetallic nodule mining on the solid phase and pore waters of these sediments. 26-

year and 5-year old plow tracks are compared to undisturbed areas next to the tracks and

reference sites. This paper entitled Biogeochemical regeneration of a nodule mining

disturbance site: trace metals, DOC and amino acids in deep-sea sediments and pore

waters has been published in the Special Issue Anthropogenic Disturbances in the Deep Sea

of Frontiers in Marine Science (Paul et al., 2018).

Chapter 4 presents REY data from the CCZ and discusses phase association as well as

impacts of early diagenesis on the development of the REY pattern of siliceous-ooze-rich

sediments. Over a large area with varying oxygen penetration depths, REY are controlled by

Ca phosphates in the solid phase, taking over the pore water pattern during early diagenesis.

This work is presented in the paper Calcium phosphate control of REY patterns of

siliceous-ooze-rich deep-sea sediments from the central equatorial Pacific published in

Geochimica et Cosmochimica Acta (Paul et al., 2019).

Chapter 5 presents exceptions to the REY distribution in CCZ sediments. Sediments with

extensive metalliferous and carbonate-rich layers show different REY distributions and pattern

changes with depth than cores that consist primarily of siliceous-ooze-rich mud. This chapter

is presented as a draft manuscript entitled Rare earth elements and yttrium in metalliferous

and calcium-carbonate-rich sediments from the central equatorial Pacific.

Chapter 6 concludes the bulk of this PhD thesis, tying together the research outlined in the

earlier chapters. This chapter also includes an outlook for prospective research, such as larger-

scale studies of deep-sea mining impacts or further pore water analyses for REY in the CCZ

and the Peru Basin.

In Chapter 7, related scientific work such as conference presentations, co-supervised BSc

theses and guided research projects, as well as field research is briefly described.

References from all chapters are combined into an integrated list at the end of the thesis,

followed by the appendices.


5

3. Background

3.1. Sediments in the Pacific Ocean

The Pacific is the largest ocean on the planet and vast areas consist of the abyssal plains at

water depths of 4000-6000 m (e.g., Jamieson, 2015). Most of the Pacific is part of the pacific

plate and its oceanic crust forms at the East Pacific Rise (EPR) (Barckhausen et al., 2013).

Sediment thickness varies substantially between relatively thin sediment cover in areas of the

south-east Pacific (0-50 m) to up to 1000 m sediment thickness along the equator in the central

Pacific (Whittaker et al., 2013). Exceptions are seamounts, faults, and ridges with thin sediment

cover or even exposure of the basaltic crust (Kuhn et al., 2017). Marine sediments in general

are comprised of a mixture of lithogenic, biogenic, hydrogenous, and authigenic material (e.g.,

Fütterer, 2000). Deep-sea sediments in particular consist of a mixture of fine-grained red clay,

siliceous, and calcareous oozes (e.g., Fütterer, 2000).

The sediments record bioproductivity, hydrothermal signatures, and climatic events (e.g.,

glacial-interglacial cycles) as surface productivity changes or as the plate moves in and out of

areas of the aforementioned influences (Burdige, 1993; Gingele and Kasten, 1994; Piper,

1973; Weber and Pisias, 1999; Ziegler and Murray, 2007). Coastal upwelling occurs at the

North and South American coasts, where cold and nutrient-rich water cycles to the surface

(Lutz et al., 2007). Upwelling also defines the equatorial high bioproductivity zone, which

extends westwards at the equator until approx. 180°W and leads to comparatively higher

organic carbon inputs into the sediments (Pälike et al., 2012; Wyrtki, 1981), influencing the

subsequent biogeochemical processes of organic matter degradation. The equatorial

upwelling intensity diminishes beyond 5°N and S (Lutz et al., 2007; Wyrtki, 1981).

3.2. The redox zonation of deep-sea sediments

Organic matter that sinks to the seafloor is degraded by microbial activity, initially through

aerobic respiration (e.g., Arndt et al., 2013; Cai and Sayles, 1996; Jahnke and Jackson, 1992).

The oxygen penetration depth and the resulting redox zonation in marine sediments is

therefore a result of the particulate organic carbon (POC) flux to the seafloor and the microbial

activity within the sediment. The sequence in which oxidants are used depends on their ability

to act as terminal electron acceptors. In order of suitability, (1) oxygen is consumed (aerobic

respiration), (2) nitrate (denitrification), (3) manganese oxides (manganese reduction), (4) iron

oxyhydroxides or other Fe(III) bearing compounds (iron reduction), (5) sulfate (sulfate

reduction) and finally (6) methane (methanogenesis) (Figure 1) (Burdige, 1993; Froelich et al.,

1979). Oxygen respiration is an aerobic metabolism that takes place in oxic environments,

while all the others are anaerobic metabolisms, which occur in suboxic (denitrification, Mn


6

reduction, and Fe reduction) or anoxic (sulfate reduction and methanogenesis) environments

(Froelich et al., 1979). This is the prevailing Froelich paradigm but recent work has shown that

nitrate and manganese cycling are closely interlinked and that nitrate can lead to oxidation of

dissolved Mn(II) (Luther et al., 1997) as well as that anaerobic ammonium oxidation

(anammox) can occur using Mn oxides (Mn anammox) (Mogollón et al., 2016). Additionally,

the importance of Mn(III) and Fe(III) as electron acceptors and donors has been stressed in

recent years (Klewicki and Morgan, 1998; Luther, 2005; Madison et al., 2013, 2011; Oldham

et al., 2015). It has been found that Mn(III) represents up to 90% of the dissolved Mn pool in

pore waters at the oxic-suboxic boundary (Madison et al., 2013), that Mn(III) is a convenient

intermediary in electron transfer processes as it can act as electron donor and acceptor and

that one electron transfer processes are preferable to two electron transfer processes (Luther,

2005). That Mn(III) species may react with upward diffusing Fe(II) to form Fe(III) is another

new addition to a revised redox zonation paradigm (Figure 1) (Madison et al., 2013). These

recent findings suggest it may be timely to rethink the Froelich paradigm, shifting our

understanding of these processes towards a more dynamic model of redox behaviors in marine

sediments.

Figure 1: Left: Redox paradigm after Froelich et al. (1979) from Burdige (1993). Reprinted from

Earth-Science Reviews, 35/3, Burdige, D.J., The biogeochemistry of manganese and iron

reduction in marine sediments, 249-284, Copyright (1993), with permission from Elsevier.

Right: Revised redox paradigm for oxic and suboxic sediments after Madison et al. (2013),

including dissolved Mn(III) and Fe(III).

While oxygen penetration depth is shallow (< 1 cm) in sediments underlying productive surface

waters, primarily those in upwelling regions close to continents (Morford and Emerson, 1999),

penetration may be much deeper in pelagic sediments where little POC reaches the seafloor.


7

Oxygen penetration depths in these areas can be > 10 m, possibly also as a result of upward

diffusing oxygenated seawater from the basaltic basement (Kuhn et al., 2017; Mewes et al.,

2016). The redox zonation of the two study areas investigated in the framework of this PhD

thesis project will be discussed in more detail in Chapter 1 section 4 – Study Sites.

3.3. Trace metals in deep-sea sediments

The redox zonation also determines how metals are distributed between the solid phase and

the pore water. In the oxic zone, Mn and Fe are present as oxides and associated metals such

as Co and Ni are largely bound in these carrier phases (e.g., Shaw et al., 1990). These are

released to the pore waters once Mn oxides and later Fe (oxyhydr)oxides get reduced. Suboxic

pore waters therefore have elevated (nmol/L to µmol/L) dissolved metal concentrations of e.g.,

Mn, Co, Ni, and Fe compared to oxic pore waters and seawater (Klinkhammer et al., 1982;

Shaw et al., 1990). The dissolved metals diffuse upwards towards the oxic pore water and into

the seawater. They are, however, stopped by Mn oxides, which act as effective scavengers

for cations, e.g., Co, Ni, and Cu (Koschinsky, 2001; Koschinsky et al., 2001b). The oxic surface

layer is therefore enriched in Mn oxides and associated metals (Koschinsky, 2001; Paul et al.,

2018). Below the Mn oxide rich layer, Fe oxyhydroxides and clay minerals are the major carrier

phases for metals and some metals released from Mn oxides re-adsorb to the aforementioned

phases (Koschinsky, 2001).

In contrast, Mo, U, and V are highly soluble in oxic waters and become immobilized and

removed from the pore water in anoxic conditions (Beck et al., 2008; Morford et al., 2005;

Wang, 2012). Their concentration in deep-sea pore waters are therefore in the same range as

seawater (Mo ~ 111 nM (Morris, 1975), U ~13.8 nM (Ku et al., 1977), since deep-sea

sediments do not reach anoxic conditions in at least the upper 10 m investigated here. The

only exception is V, which has elevated concentrations in the pore water due to release during

organic matter degradation and stabilization by dissolved organic carbon (DOC) (Emerson and

Huested, 1991; Morford et al., 2005). Similarly, other metals, e.g., Cu, are released to the pore

water at the sediment-water interface during organic matter degradation (Heggie et al., 1986;

Kowalski et al., 2009; Sawlan and Murray, 1983; Shaw et al., 1990). Pore waters are therefore

sources of these elements to the overlying seawater.

Arsenic primarily exists as arsenite (As(III)) and arsenate (As(V)) in natural waters (e.g.,

Nicomel et al., 2015; Telfeyan et al., 2017 and references therein). Dissolved concentrations

are in the range of ~ 13-22 nM (Andreae, 1977; Cabon and Cabon, 2000) in seawater and

around 30 nM in suboxic pore water (Telfeyan et al., 2017). In the solid phase, As can be


8

bound to Mn and Fe oxides and released upon their reductive dissolution (Andreae, 1979;

Telfeyan et al., 2017).

3.4. Rare earth elements and yttrium (REY) in deep-sea sediments

The rare earth elements (REE) consist of the 15 elements of the lanthanide series (La, Ce, Pr,

Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu) with atomic numbers 57 to 71. They behave

similarly in natural environments due to their trivalent charges and similar ionic radii. The ionic

radius decreases with increasing atomic number due to stepwise filling of the 4f orbital (Figure

2) (O’Neill, 2016; Seth et al., 1995). The small differences in ionic radii are sufficient to lead to

fractionation of the REE in natural systems. Additionally, Ce and Eu can also occur in the

tetravalent and divalent state, respectively. Yttrium is also trivalent and of almost exactly the

same size as Ho (Figure 2). These elements are often referred to as “geochemical twins” and

due to exhibiting similar properties, Y is often included with the REE forming the REY.

Promethium (Pm) cannot be analyzed in natural samples as it is a radioactive element.

The REY can be grouped into light REY (LREY), middle REY (MREY), and heavy REY

(HREY), since the elements in each group behave with greater similarity to each other than

the REY in general. The LREY include the elements La-Nd, the MREY the elements Sm-Dy,

and the HREY the elements Ho-Lu. Yttrium is then included with the HREY.

Figure 2: Visualization of the lanthanide contraction: REE show decreasing ionic radii with

increasing atomic number. Note similar size of Y and Ho, explaining why Y is often included

with the REE. Also note significant differences in ionic radii sizes of tetravalent Ce and divalent

Eu. After Merschel (2017) with data from Shannon (1976).


9

When interpreting REY data, results are usually normalized to e.g., chondrite or shale because

naturally, elements with even atomic numbers are more abundant than the adjacent elements

from the periodic table with uneven atomic numbers (Oddo Harkins rule; Figure 3). Chondrite

is used for rock samples and to track rock origin (O’Neill, 2016). For marine samples,

normalization to shale is more suitable because marine sediments are similar to shales and

the patterns therefore highlight the significant differences between the marine samples, rather

than the differences to the material they are normalized to, which would be the case when

using chondrite (Piper, 1974a). Therefore, Post Archean Australian Shale (PAAS) from Taylor

and McLennan (1985), except for Dy from McLennan (1989), was used to normalize all

samples in this PhD thesis.

Figure 3: Example of REY normalization. Raw PAAS data displays concentration differences

between even and odd atomic numbers. Once normalized to chondrite, smooth patterns can

be interpreted. After Merschel (2017). PAAS data from Taylor and McLennan (1985), except

for Dy from McLennan (1989); C1 chondrite data from Anders and Grevesse (1989).

In the REYSN pattern, relative enrichments of LREY, MREY, and HREY can be easily identified,

as well as other anomalies of single elements. In oxic environment, Ce(III) can be removed

from the dissolved phase by oxidative scavenging (e.g., Bau and Koschinsky, 2009). Once


10

adsorbed to the particle, Ce(III) is oxidized to Ce(IV) and rendered less soluble. Oxic seawater

therefore displays negative CeSN anomalies (Alibo and Nozaki, 1999). The particles or phases

that oxidatively adsorbed Ce show positive CeSN anomalies, such as hydrogenetic nodules and

crusts (Elderfield and Greaves, 1982; Kasten et al., 1998). A negative CeSN anomaly can

already develop in rivers during terrestrial weathering and the signal is transported to the ocean

(Merschel et al., 2017; Pourret and Tuduri, 2017).

Europium shows positive anomalies in the REYSN pattern in reducing environments. Prominent

examples include hydrothermal environments and the associated sediments (German et al.,

1990; Michard, 1989; Ruhlin and Owen, 1986). Particles from hydrothermal plumes and

metalliferous sediments in the vicinity of hydrothermal systems show positive EuSN anomalies

(German et al., 1990; Michard, 1989; Ruhlin and Owen, 1986). Particles scavenge more REY

from the seawater with increasing distance from the hydrothermal vent site, taking on a more

and more seawater-like REYSN pattern (i.e. negative CeSN anomaly and HREY enrichment),

progressively obscuring the hydrothermal signature (i.e. EuSN anomaly) (German et al., 1990;

Ruhlin and Owen, 1986).

Dissolved REY concentrations in seawater increase with depth because of scavenging onto

particles in the surface ocean and release at depth during particle dissolution, except for Ce,

which shows decreasing dissolved concentrations due to oxidative scavenging on particles

(Alibo and Nozaki, 1999; Elderfield et al., 1988). In the REYSN pattern, seawater displays

enrichment of HREY over LREY, i.e. LaSN/YbSN <<1, a pronounced negative CeSN anomaly,

and positive LaSN, GdSN, and YSN anomalies (Alibo and Nozaki, 1999). The enrichment of

dissolved HREY develops because LREY are preferentially adsorbed onto (Mn oxide) particles

in the surface ocean (Elderfield et al., 1988; Sholkovitz et al., 1994). LREY are, however,

released again at depth where the particle surface coatings dissolve (Sholkovitz et al., 1994).

Others have argued, however, that Mn oxides may play a lesser role in scavenging LREY and

that Fe (oxyhydr)oxides and POC are more significant scavengers (Haley et al., 2004). REY

reach the seafloor adsorbed to organic or metal oxide particles and are released to the pore

waters during organic matter degradation (Elderfield et al., 1981; Haley et al., 2004). They are

subsequently incorporated into sedimentary phases.

Over large areas of the Pacific, REY in the sediments are controlled by Ca phosphates

(Elderfield et al., 1981; Kato et al., 2011; Toyoda et al., 1990), which are largely comprised of

biogenic material i.e. fish bones and teeth (Toyoda and Tokonami, 1990). The Ca phosphates

accumulate high REY concentrations during early diagenesis (Bright et al., 2009; Elderfield et

al., 1981). REYSN patterns of these sediments show large negative CeSN anomalies and a


11

MREY enrichment, which develops due to diagenetic overprinting (Bright et al., 2009; Toyoda

and Masuda, 1991). Sediments in the vicinity of the East Pacific Rise also display a negative

CeSN anomaly, which has been described to be a result of hydrothermal influences (Toyoda et

al., 1990), i.e. hydrothermal Fe particles scavenging REY from the seawater and therefore the

incorporation of the negative CeSN anomaly (Ruhlin and Owen, 1986). Sediments rich in

calcareous ooze also show large negative CeSN anomalies but lower REY concentrations than

deep-sea clay dominated sediments (Toyoda et al., 1990). Low REY concentrations in CaCO3-

rich sediments have been explained by quick deposition of carbonate material and therefore

little time to accumulate REY at the sediment-water interface (Kato et al., 2011) or due to

dilution of REY-carrying phases, such as clay or Mn and Fe phases, by CaCO3 (Pattan and

Higgs, 1995). Similar to Ca phosphates, unaltered Ca carbonate REYSN patterns display

seawater patterns, while diagenetically altered REY patterns of carbonates can show MREY

enrichment and elevated REY concentrations compared to skeletal material (Haley et al.,

2005; Webb and Kamber, 2000).

Various authors have published pore water REY data: from nearshore reducing sediments

(Elderfield and Sholkovitz, 1987; Sholkovitz et al., 1989), continental margin sediments (Abbott

et al., 2015; Haley et al., 2004), and oxic pelagic sediments (Deng et al., 2017). Even though

the results vary significantly, a common observation is that REY concentrations are enriched

in pore waters compared to seawater (Abbott et al., 2015; Elderfield and Sholkovitz, 1987;

Haley et al., 2004). This is likely due to mobilization during early diagenesis (Elderfield and

Sholkovitz, 1987) and release from POC or Fe oxides, the latter only in anoxic environments

(Elderfield et al., 1981; Haley et al., 2004). Dissolved REY show a concentration maximum at

the sediment water interface in vertical pore water profiles (Deng et al., 2017; Haley et al.,

2004). For the REYSN patterns, it has been proposed that a MREY enrichment develops as a

result of REY release from Fe oxides in anoxic pore waters and that a linear LREY to HREY

enrichment or a strong HREY enrichment develops from release of REY from POC in oxic and

suboxic pore waters (Haley et al., 2004).

3.5. Deep-sea mineral resources and mining

There are three types of deep-sea mineral resources that are discussed as potential

economically viable resources: polymetallic nodules on the abyssal seafloor, Co-rich Fe-Mn

crusts on seamounts, and seafloor massive sulfides (SMS), which are associated with

hydrothermal vent sites. While mining for SMS deposits has been in the public focus in recent

years due to exploration activities of Nautilus Minerals at the Solwara 1 vent field in the national

waters of Papua New Guinea (Nautilus Minerals, 2018), nodules and crusts have received less


12

attention because of relatively low metal prices and because the mining and processing

technology is less advanced. Technology is developing, however, and environmental

assessments are necessary to analyze potential impacts before mining activities commence

(Gollner et al., 2017; Halfar and Fujita, 2002). The focus in this PhD thesis is on the potential

impacts of polymetallic nodule mining.

3.5.1. Polymetallic nodules

Polymetallic (also called manganese) nodules form on the abyssal seafloor in depths between

3500 and 6500 m in flat areas where low sedimentation rates (< 10 mm/kyr) prevail and where

the bottom water is oxic (Gollner et al., 2017; Hein et al., 2013). They form through precipitation

of Mn oxides and Fe (oxyhydr)oxides around a nucleus, e.g., fish bones, broken nodule

material, volcanic rock fragments. Nodule types are differentiated based on accumulation of

oxides, from either the water column and oxic pore waters (hydrogenetic nodules) or from

suboxic pore waters (diagenetic nodules) (Halbach et al., 1981; Hein and Koschinsky, 2014;

Wegorzewski and Kuhn, 2014). Many other metals (e.g., Co, Ni, Cu, Zr, Nb, REY) get

scavenged and enriched during this process (Glasby et al., 1982; Hein et al., 2013;

Wegorzewski and Kuhn, 2014). During hydrogenetic nodule formation, Mn oxides with

negatively charged surfaces adsorb cations and Fe (oxyhydr)oxides with positively charged

surfaces adsorb anions and negatively charged molecules (Hein et al., 2013; Koschinsky and

Halbach, 1995). With time, the elements get also structurally incorporated into the minerals.

The main mineral phases in hydrogenetic nodules are Fe-bearing vernadite and amorphous

Fe (oxyhydr)oxide (Wegorzewski and Kuhn, 2014 and references therein). During diagenetic

nodule formation, coprecipitation and element substitution in the crystal lattice of the Mn

minerals (predominantly todorokite and phyllomanganates during diagenetic growth) are the

primary processes for nodule formation and metal enrichment (Wegorzewski and Kuhn, 2014).

Through these precipitation processes and the enrichment of metals, Mn nodules represent a

potential mineral resource for Cu, Ni, Co, Mn, Mo, Li, Te, Zr, Nb, W and REY but traditionally

the economic focus for resource extraction was on Ni and Cu (Hein et al., 2013; Hein and

Koschinsky, 2014). There are also mixed type nodules which are formed by a combination of

both hydrogenetic and diagenetic processes (Halbach et al., 1981; Hein and Koschinsky,

2014). In the CCZ – in the “manganese nodule belt” (Wegorzewski and Kuhn, 2014 and

references therein) – nodules form hydrogenetically because the oxygen penetration depth is

1 m to >10 m deep (Kuhn et al., 2017; Mewes et al., 2016, 2014; Rühlemann et al., 2011).

Growth from oxic pore waters results in the same characteristics as growth from the oxic

seawater (Wegorzewski and Kuhn, 2014). In the Peru Basin, mixed type nodules occur that

grow hydrogenetically on top and diagenetically at the bottom where they lie in the sediment

(Wegorzewski and Kuhn, 2014). Growth regimes for nodules may have or might vary over


13

time, however, due to fluctuations in the depth of the oxic-suboxic boundary, for example on

glacial-interglacial time scales as a result of changing surface water productivity (König et al.,

2001; Wegorzewski and Kuhn, 2014).

Nodules are not only found on the sediment surface but also buried within the sediment. This

is a common phenomenon known from the Peru Basin (Greinert, 2015), the CCZ (Heller et al.,

2018; Volz et al., 2018) and the Indian Ocean (Pattan and Parthiban, 2007). If buried in the

oxic sediment column, the nodules do not dissolve (Pattan and Parthiban, 2007). Once the

nodules are buried to sufficient depth to reach the suboxic zone, Mn and Fe (oxyhydr)oxides

dissolve (Froelich et al., 1979; Paul et al., 2018), minerals get transformed, and metals such

as Mn are released to the pore waters (Heller et al., 2018).

3.5.2. Polymetallic nodule mining and potential environmental impacts

Polymetallic nodules contain potentially economically viable concentrations of Co, Cu, Ni, Mo,

Li, Te, and REY (Hein et al., 2013). Concentrations vary between resource deposits: nodules

in the Peru Basin e.g., have lower quantities of Cu compared to CCZ nodules (Wegorzewski

and Kuhn, 2014). Many of these elements are critical high-technology metals that are needed

in our society for renewable energies such as wind farms, electrical car engines, or medical

diagnostics (Hein et al., 2013; Kulaksiz and Bau, 2013; “Mine and monitor impacts,” 2015).

Mining for polymetallic nodules will most likely start in the CCZ due to spatially extensive high

nodule coverage and high contents of Cu and Ni within individual nodules (Hein et al., 2013).

So far, the International Seabed Authority (ISA) has issued 16 exploration contracts in the CCZ

and one in the Central Indian Ocean Basin (International Seabed Authority, 2018). In The Area,

the seafloor beyond the exclusive economic zones (EEZ) of countries and therewith borders

of national jurisdiction, the ISA is responsible for issuing exploration contracts for marine

mineral resources (Lallier and Maes, 2016; Petersen et al., 2016). The industrial development

of nodule extraction technologies has stalled for many years but during the last decade

industrial involvement has grown and prototypes for nodule extraction have been built (Gollner

et al., 2017). The latest collector technologies resemble a harvester, designed to plow the

seafloor and hydraulically or mechanically collect the nodules, then “pump” the nodule material

through a riser pipe up to the mining vessel and separate these from sediment (Gollner et al.,

2017). A schematic drawing of a potential system and its main components is shown in

Figure 4.


14

Figure 4: Schematic drawing of the mining set up consisting of system 1, the mining platform

and the transport vessel at the sea surface, system 2, the riser pipe, system 3, the nodule

collector, and system 4, the discharge of e.g., leftover sediment and nodule material after

processing on the mining platform. Reprinted from Deep Sea Research Part II: Topical Studies

in Oceanography, 48/17-18, Oebius, H.U., Becker, H.J., Rolinski, S., Jankowski, J.A.,

Parametrization and evaluation of marine environmental impacts produced by deep-sea

manganese nodule mining, p.3455, Copyright (2001), with permission from Elsevier.


15

As Figure 4 shows, impacts are expected on the seafloor (sediment removal, mixing, and

compaction) as well as in the water column, where plumes are expected to occur. Plumes

could develop directly at the seafloor, where the collector whirls up sediment, but also in the

water column, if sediment is discharged into the water after separation from nodules. Potential

environmental impacts include but are not limited to: killing of sessile (and mobile) fauna by

the collector, habitat destruction, disruption of benthic (microbial) processes, disruption of

redox zonation, release of (potentially toxic) metals from the pore water, plume creation with

subsequent settling of particles, blanketing organisms in the surrounding area and inserting

unnaturally high concentrations of particles into the water column (Gollner et al., 2017; Jones

et al., 2017; Koschinsky et al., 2018; Vanreusel et al., 2016).

For this PhD project, the research focus has been to investigate the potential for disruption of

the redox zonation and the release of metals from the pore water. Both impacts are anticipated

as the upper tens of cm of sediment will likely be removed by the collector during mining

(Cronan et al., 2010; Gollner et al., 2017; Thiel and Schriever, 1990). The significance of the

impact depends, however, on the depth of the oxic zone. If the collector only disturbs the oxic

layer, metals such as Mn, Fe, Co, Ni, and Cu are unlikely to be released from the pore waters

because under those conditions they are mostly bound in the solid phase (Koschinsky, 2001;

Shaw et al., 1990). A disruption of the redox zonation and associated release of dissolved

metals from suboxic pore waters is only likely in regions where the oxic-suboxic boundary is in

the upper ca. 30 cm. Under those conditions, high concentrations (µmol/L) of released metals

e.g., Cu could be toxic for the fauna (Auguste et al., 2016; Martins et al., 2017; Mevenkamp et

al., 2017). If metals are released, however, into the oxic water column, they will be rapidly

scavenged by Mn and Fe (oxyhydr)oxide particles that are resuspended during the disturbance

as well (Koschinsky et al., 2003). An overview of the biogeochemical processes that are

impacted by and follow a mining-like disturbance is given in Figure 5.


16

Figure 5: Overview of potential geochemical impacts and processes after a disturbance in the

sediment, at the sediment surface, and in the bottom water.

To date, time-scales to reach a new equilibrium in the solid phase and pore water after a

physical disruption of the redox zonation have been unclear; though predictions based on

numerical modelling are in the range of centuries (König et al., 2001). Oxygen penetration is

unhindered if the reactive labile organic matter and Mn-oxide-rich surface layer is removed

(König et al., 2001). This would remove dissolved metals from the pore water by forming Mn

oxides, which adsorb the other positively charged metals. Oxygen penetration could possibly

be stopped by a reactive Fe(II) layer deeper in the sediments (König et al., 2001).

4. Study sites

Two study sites were chosen for the work in this PhD project: the Peru Basin in the south

eastern equatorial Pacific Ocean and the Clarion Clipperton Zone (CCZ) in the central

equatorial Pacific Ocean (Figure 6). Both sites have high nodule coverage but are

biogeochemically significantly different, as described in detail below (section 4.1. and 4.2.).

The sites were selected due to their suitability to serve as study sites for deep-sea polymetallic

nodule mining induced disturbances. All but one exploration contract for polymetallic nodules

issued by the ISA thus far is for the CCZ. Additionally, benthic impact experiments (BIE) have

been conducted by a number of organizations since the 1970s in both areas (Jones et al.,

2017). The Peru Basin has been a German research focus since the 1980s. In 1989, the

DISturbance and reCOLonization project (DISCOL) carried out a mechanical plowing action,


17

to mimic impacts of polymetallic nodule mining and to research impacts on the fauna and

sediment (Thiel and Schriever, 1990).

Figure 6: Map showing locations of the two study sites from this PhD thesis: the Clarion

Clipperton Zone and the Peru Basin. The map was created using GeoMapApp.

The DISCOL area study site spans a much smaller area than the CCZ study site: it is

approximately 13 km across from Reference West to Reference East (Figure 9). The focus in

the Peru Basin was on a small-scale heterogeneity assessment and the local impacts of deep-

sea mining. In the CCZ, the focus was on large-scale variability of REY.

4.1. DISCOL area, Peru Basin

The seafloor in the study area in the Peru Basin lies at ca. 4100-4250 m water depth.

Sediments in the Peru Basin largely consist of siliceous and calcareous muds (Weber et al.,

1995) with the carbonate compensation depth (CCD) currently located at ca. 4200-4250 m

(Weber et al., 2000) so that some carbonate material is preserved. The Peru Basin is

characterized by organic matter inputs from the southern border of the equatorial high

productivity zone and the coastal upwelling region but the DISCOL area specifically is not

impacted by these highly productive areas (Weber et al., 2000). Sedimentary surface POC

contents are between 0.5 wt.% in the DISCOL area and up to 1 wt.% in areas influenced by

the upwelling regions (Haeckel et al., 2001; Paul et al., 2018). The redox zonation in the upper

10 m is characterized by oxygen penetration depths of ca. 5-25 cm (Haeckel et al., 2001; Paul


18

et al., 2018). When the oxygen is consumed, Mn oxide reduction and denitrification take place.

Where nitrate is present in the pore water, Fe(II) in the clay mineral crystal lattice is oxidized

to Fe(III) (Drodt et al., 1997; König et al., 1999, 1997; Lyle, 1983). In locations where nitrate is

not present throughout, a higher percentage of Fe(II) remains in the clay minerals and gives

the sediments a greenish color.

In 1989, a plow harrow was used to create a disturbance similar to deep-sea mining. The

surface sediment was mixed and the nodules were plowed to deeper layers (Thiel and

Schriever, 1990). The DISCOL area has been extensively investigated following this initial

disturbance: 0.5, 3, and 7 years post-disturbance (Thiel, 2001). These investigations first

focused on the biology and geochemical analyses were only incorporated in 1996 in the course

of the ATESEPP project (Schriever et al., 1996). The research was not focused, however, on

the DISCOL area and investigations of metal geochemistry were only conducted at one station

in the DISCOL area (Koschinsky, 2001). Due to technological limitations, it was also not

possible to see where samples were taken, so that it was unclear if samples were taken inside

or outside of disturbance tracks.

In 2015, the DISCOL area was revisited and thorough sampling of plow tracks and reference

sites was conducted (Boetius, 2015; Greinert, 2015) (Figure 7). Most samples analyzed within

the framework of this PhD thesis were taken on these cruises: SO242 legs 1 and 2. Moreover,

a fresh physical disturbance was created with an epibenthic sled (EBS) during the first leg of

the cruise and sampled ca. five weeks later during leg 2. The EBS removed the surface

sediment and the nodules, rather than plowing nodules into the sediment and mixing sediment,

as had been the case during the 1989 disturbance action.


19

Figure 7: Bathymetric map of the DISCOL area in the Peru Basin adapted from Paul et al.

(2018). The circle shows the DEA (DISCOL experimental area). Sampling locations are

marked with red dots.

4.2. Clarion Clipperton Zone (CCZ)

The CCZ is bordered by the Clarion fracture to the North and the Clipperton fracture to the

South. Water depths in the CCZ are between ca. 4000 and 5000 m (Berger et al., 1976) and

the CCD is located at ca. 4600 m (Pälike et al., 2012). Sediments are primarily composed of

siliceous ooze and mud (Berger, 1974; International Seabed Authority, 2010; Kuhn, 2015). The

CCZ is geochemically different from the Peru Basin, primarily because the oxygen penetration

depth is much deeper. The oxygen penetrates approx. 1-4.5 m and the sediment is in parts


20

oxic throughout, also due to upwards diffusion of oxygen from the basaltic crust (Kuhn et al.,

2017; Mewes et al., 2016; Volz et al., 2018). This is because the CCZ receives lower POC

inputs (1-2 mg Corg/m2d) and as a result the surface sediments only have POC contents of 0.2-

0.6 wt.% (Vanreusel et al., 2016; Volz et al., 2018). The POC input and content decrease from

east to west and northwards across the CCZ, leading to a greater oxygen penetration depth in

these areas (Volz et al., 2018) (Figure 8). Below the oxic zone, Mn oxide reduction is the

dominant process in the upper 10 m (Volz et al., 2018). The zone of Fe (oxyhydr)oxide

reduction is not reached within in the upper 10 m and no dissolved Fe can be detected in the

pore waters.

Figure 8: Map showing the contractor areas (BGR: Bundesanstalt für Geowissenschaften und

Rohstoffe, Germany, IOM: InterOceanMetal, eastern European consortium, GSR: Global Sea

Mineral Resources NV, Belgium, IFREMER: L’Institut Français de Recherche pour

l’Exploitation de la Mer) and the Area of Particular Environmental Interest (APEI) 3 visited

during SO239 in 2015 and the POC flux (Vanreusel et al., 2016; Creative Commons Attribution

4.0 International Public License).

Between 1976 and 1995 several impact experiments with follow-up investigations were

conducted by different organizations in the CCZ (for an overview see Jones et al., 2017; Table

1). Several tests were conducted in the frame of the Deep Ocean Mining Environmental Study

(DOMES) (Jones et al., 2017). The CCZ was also revisited in 2015 in the frame of the JPI

Oceans project Ecological Aspects of Deep-Sea Mining (SO239) and the FLUM project (Low-


21

temperature fluid circulation at seamounts and hydrothermal pits: heat flow regime, impacts

on biogeochemical processes, and its potential influence on the occurrence and composition

of manganese nodules in the equatorial eastern Pacific) (SO240). For this PhD project,

samples were taken from the German BGR (Bundesanstalt für Geowissenschaften und

Rohstoffe) contractor area (blue in Figure 8), the IOM (InterOceanMetal) contractor area

(yellow in Figure 8), the Belgian GSR (Global Sea Mineral Resources NV)contractor area

(green in Figure 8), the French IFREMER (L’Institut Français de Recherche pour l’Exploitation

de la Mer) contractor area (red in Figure 8), and the Area of Particular Environmental Interest

(APEI) 3 (transparent red in Figure 8).

Since deep-sea mining for polymetallic nodules will most likely start in the CCZ, many

(bio)geochemical investigations have been carried out in the region in recent years (Kuhn et

al., 2017; Mewes et al., 2016, 2014; Mogollón et al., 2016; Rühlemann et al., 2011; Volz et al.,

2018; Wegorzewski and Kuhn, 2014).

Table 1: Previous benthic impact experiments conducted in the CCZ. Information from Jones

et al. (2017).

Experiment Year Main goal

OMA (Ocean Mining Association) (part of DOMES)

1976 Mining test

OMI (Ocean Management Inc.) (part of DOMES)

1978 (2x) Mining test

OMCO (Ocean Minerals Company) (part of DOMES)

1978 (unsuccessful); repeated 1979

Mining test

JET (Japan Deep-Sea Impact Experiment)

1991 Test for sediment resuspension and settling; macro- and megafauna analyses on post-disturbance cruises

BIE 1991 (unsuccessful); repeated 1992

Test to evaluate sediment resuspension

BIE-II 1993 Evaluate sediment resuspension

IOM BIE (InterOceanMetal) 1995 Benthic impact assessment

5. Methods

5.1. Sampling

Sampling for this PhD thesis was carried out during three RV SONNE cruises in 2015: SO239

as well as SO242 legs 1 and 2. Sediment sampling was conducted using gravity corer (GC),

multi corer (MUC), and a remotely operated vehicle (ROV) with push cores (PUC). These three

types of cores enabled us to sample deep sediments down to 10 m (GC) to study the sediment


22

history, and surface sediments (MUC and ROV-PUC) to study surface sediment processes

and impacts of mining disturbances. ROV-PUCs additionally allowed for sampling of specific

features because sampling by this methodology is spatially precise. This was used to sample

various disturbance feature subhabitats in the DISCOL area, such as the ripples, valleys and

white patches.

Gravity cores sampled during SO239 were brought into the cool room (4°C)and left to

equilibrate for 12 hours for oxygen measurements to be taken prior to core slicing and sampling

(Martínez Arbizu and Haeckel, 2015). Gravity cores sampled during SO242/1 were cut into 1

m pieces on deck and halved. The working half was subsequently immediately transported to

the cool room and sampled. Sediment samples were spooned into 10% nitric acid (HNO3,

technical grade) and deionized water (DI) pre-cleaned 50 mL centrifuge vials and centrifuged

at 3200 rpm for 40 minutes at 4°C to separate the pore water. This duration was determined

early in the cruise, with the largest pore water volume extracted successfully when centrifuging

for that time period. Pore water was taken up in a 0.1 M hydrochloric acid (HCl; suprapure)

and DI pre-cleaned syringe and filtered through 0.2 µm 0.1 M HCl (suprapure) and DI pre-

cleaned cellulose acetate (CA) syringe filters (Whatman). Trace metal samples were stored in

2% HNO3 and 0.2% hydrofluoric acid (HF) (both technical grade) and DI pre-cleaned

polyethylene (PE) bottles. DOC samples were stored in annealed glass vials. Samples were

immediately acidified using 1 µL 30% HCl (suprapure; Merck) for 1 mL sample. This brings the

pH to approximately 2. Filtration and acidification were performed in a glove box under a

constant flow of argon.

Prior to sediment sampling, bottom water from the MUCs and ROV-PUCs was sampled, and

the remainder drained with tubing. Sampling for MUCs and ROV-PUCs was performed in a

glove bag filled with argon. The glove bag was flushed three times before finally starting to

sample. Cores were sampled in 2 cm layers by spooning sediment into 10% HNO3 (technical

grade) and DI pre-cleaned 50 mL centrifuge vials. Pore water extraction then followed the

same procedure described above for the GCs.

Samples for dissolved amino acids (DAA) were extracted using rhizons with 0.1 µm average

pore size (Seeberg-Elverfeldt et al., 2005) in 3 to 7.5 cm intervals. Rhizons were rinsed with

DI prior to use. Samples for DAA were immediately frozen after sampling.

Samples for nitrate were taken with argon pressure squeezers at 3-5 bar and the pore water

was squeezed through 0.2 µm CA filters.


23

5.2. Analytical methods

5.2.1. Acid pressure digestion

Sediment samples were ground and homogenized and subsequently dried overnight at 105°C

to remove any remaining moisture. For the acid pressure digestion, 100 mg of dried sediment

were weighed into polytetrafluorethylene (PTFE) beakers for digestion in a Pico Trace DAS

acid pressure digestion system. Sediments were digested for 12 hours at 220°C using a

mixture of 3 mL perchloric acid (HClO4) and 3 mL HF (both of suprapure quality). This acid mix

was found to be the most efficient for completely dissolving deep-sea sediments, after testing

a mix of 3 mL HCl, 1 mL HNO3 and 1 mL HF, which was not effective, leaving residues in the

beakers after digestion. Pre-treatment with H2O2 before HClO4 and HF digestion did not

improve the dissolution outcome and was not pursued further after a test run. After digestion,

samples were evaporated to insipient dryness, re-dissolved in 5 mL HCl (30%) and evaporated

again. This procedure was repeated before the samples were finally taken up in a mix of 0.5 M

HNO3 and 0.5% HCl (v/v). Even with the HClO4-HF mix, some samples showed black particles

in the final digested sample after evaporation. These particles were probably carbon that could

not be fully digested. The samples were then filtered through 0.2 µm CA syringe filters

(Whatman).

For each digestion block of 16 beakers, two beakers were used for the certified reference

materials (CRMs) MESS-3 and BHVO-2. MESS-3 is a marine sediment from Beaufort Sea but

does not have certified values for REY and some other trace elements. Therefore, BHVO-2, a

Hawaiian basalt with reference values for REY, was added in addition in each digestion. In

addition to the CRMs, one beaker was always used as a method blank, without sample powder,

but treated the same as all beakers with samples.

5.2.2. ICP-OES

For major element analyses, digested sediment and pore water samples were measured with

an inductively coupled optical emission spectrometer (ICP-OES; SOP Spectro Ciros Vision).

Samples were introduced into the machine in liquid form: in a 0.5 M HNO3 0.5% HCl (v/v)

matrix for digested sediment samples with a total dilution factor of 1,500 and for pore water

samples in a 0.5 M HNO3 matrix with a total dilution factor of 10. The sample was nebulized

and introduced into the argon plasma, which evaporated the sample. By this process, the

molecules in the sample were dissociated into atoms, excited and partially ionized. The ions

each emitted light at a specific wavelength which was fed into the optical system and the

intensity subsequently measured by semiconductor detectors (Spectro, 2004). The elements

that were analyzed by this method were: Al, Ca, Cu, Fe, K, Li, Mg, Mn, Na, P, Sr, V, and Zn

for sediment and B, Br, Ca, K, Li, Mg, Mn, S, Si, and Sr for pore water. The samples are


24

calibrated against six standard solutions – and within the calibration, the emitted light

intensities therefore detected are proportional to the concentration in the introduced sample.

For sediment analyses, MESS-3 and BHVO-2 were used as CRMs to check the measurements

for accuracy, while IAPSO was used as a CRM for pore water analyses. To adjust for machine

drift, 10 ppm Y are added as internal standard to all calibration standards, blanks, CRMs, and

samples.

5.2.3. ICP-MS

For trace element analyses of digested sediment and pore water samples an inductively

coupled mass spectrometer (ICP-MS; Perkin Elmer Nexion 350x) was used. Samples were

introduced into the machine in liquid form: in a 0.5 M HNO3 0.5% HCl (v/v) matrix for digested

sediment samples with a total dilution factor of 50,000 and for pore water samples in a 0.5 M

HNO3 matrix with a total dilution factor of 80. For pore water analyses, all blanks and

calibration standards were prepared using a 7 mM suprapure NaCl 0.5 M HNO3 solution to

adjust the matrix to the diluted pore water matrix of the samples.

The sample was nebulized and subsequently introduced into an argon plasma, where the

molecules were dissociated and ions formed. The ions were directed through three cones into

the quadrupole mass spectrometer, which scanned the entire mass range. Only one mass-to-

charge ratio was allowed in at a given time, thereby the system only measured that one mass

at a time, with all other ions deflected. When the ions leave the mass spectrometer, they hit

the detector, an electron multiplier, where the released electrons from the ion impact are

amplified until a pulse can be measured (Perkin Elmer, n.d.).

For pore water analyses, the ICP-MS was used with the apex Q (ESI). The sample is first

introduced into the apex Q, which removes background noise and improves sensitivity. For

pore water, only Mo, Cd and U can be measured in the standard more. All other elements – V,

Mn, Co, Cu, and As – need to be measured in collision cell mode with kinetic energy

discrimination (KED) using He as the non-reactive gas to remove polyatomic spectral

interferences on their masses. In KED mode, the collision cell is flushed with He in addition to

Ar and Ar-molecules can be removed, which reduces interferences. The sensitivity of the

machine is, however, considerably reduced. Interestingly, Mo accuracy was much better in

KED mode, even though there are no Ar-complexes interfering with the Mo masses. For the

analyses of the GCs in Chapter 2, Mo data obtained with the KED mode were used.


25

A mix of Ru, Re, and Bi was added as internal standard to all samples, blanks, calibration

standards, and CRMs to be able to correct for machine drift during a run, using 1 ppb for apex

Q measurements and 2 ppb for normal measurements and apex Q KED mode measurements.

5.2.4. Analytical quality of major and trace element analyses

Replicate analyses were done for selected samples but not for entire depth profiles. Therefore,

the data shown in this thesis are usually not shown with error bars. The analytical quality was

checked by repeated measurements of CRMs.

Pore-water measurements are challenging due to the low metal concentrations and the salt

matrix, which requires sample dilution and still lowers the detection limits. For ICP-OES pore-

water measurements, IAPSO standard seawater CRM from OSIL was used. For ICP-MS pore-

water measurements, IAPSO from OSIL as well as NASS-6 and NASS-7 seawater reference

material from the National Research Council Canada were primarily used. Since metal

concentrations of the CRMs are often below the detection limit, SLEW-3, an estuarine water

CRM, as well as SLRS-6, a river water CRM, were used additionally because they have, for

example, higher Mn and Cu concentrations which are more comparable to pore waters. After

testing the adjustment of the salt matrix in the estuarine and river water CRMs, it was decided

to measure them only diluted in 0.5 M HNO3 without adjustment of the salt content, which might

nevertheless impact the accuracy as the CRM matrix is different from the blank, calibration

standard, and samples.

Difficulties were seen in the beginning for Mo, for which sometimes higher concentrations of

up to 25% were measured. This improved over time and it was found that Mo measurements

completed in KED mode were more accurate (within 2% for NASS-6 and NASS-7 and within

8% for SLEW-3). Cadmium was also difficult to measure because of low concentrations. Cobalt

concentrations in all CRMs were always below the detection limit. Pore-water Co

concentrations are, however, significantly higher in suboxic samples, wherefore Co

measurements in the samples were seen as reliable.

Figures 9-12 provide a detailed illustration of accuracy and precision for Mn, Cu, Mo, and V as

examples for pore-water data quality. For detailed information on analytical quality for each

element please refer to the methods section in each chapter and to the appendices.


26


analyses for Mn with standard deviation. The two SLEW-3 measured values are for results in




analyses for Cu with standard deviation. The two SLEW-3 measured values are for results in




27


analyses for Mo with standard deviation. The two NASS-6 measured values are for results in




analyses for V with standard deviation. The two NASS-6 and SLEW-3 measured values are

for results in Chapter 2 and Chapter 3, respectively. “n” refers to the number of ICP-MS runs,

in which the CRM was usually measured multiple times.


28

For ICP-OES and ICP-MS solid-phase measurements, MESS-3 (National Research Council

Canada) and NIST-2702 (National Institute of Standards and Technology) marine sediment

CRMs were used, as well as BHVO-2, a Hawaiian basalt from USGS, for REY reference data.

Figures 13-16 provide a detailed illustration of accuracy and precision for P, Fe, Ce, and Nd

as examples for solid-phase data quality. For detailed information on analytical quality for each

element please refer to the methods section in each chapter and to the appendices.


analyses for P with standard deviation. The two MESS-3 and BHVO-2 measured values are





29


analyses for Fe with standard deviation. The two MESS-3 and BHVO-2 measured values are





30


analyses for Ce with standard deviation. The two NIST-2702 measured values are for results





31


analyses for Nd with standard deviation. The two NIST-2702 measured values are for results




5.2.5. Dissolved organic carbon (DOC)

DOC analyses were carried out at the University of Hamburg in collaboration with the work

group of Birgit Gaye, using a high temperature combustion method (TOC-VCSH Analyzer,

Shimadzu). Prior to injection into the combustion tube, inorganic carbon was removed with 2

M HCl. In the combustion tube, C was oxidized to CO2 with a platinum catalyst at 680°C. Five

calibration standards from 0.5 mg/L to 5 mg/L DOC were used. Measurement error was <2%.

For further details see Brockmeyer and Spitzy (2013).

5.2.6. Dissolved amino acids (DAA)

Analyses of total hydrolysable DAA and hexosamines of selected pore water samples were

carried out at the University of Hamburg in collaboration with the work group of Birgit Gaye.

Samples were hydrolyzed with 6 N HCl at 110°C for 22 hours under a pure argon atmosphere

and subsequently measured with a Biochrom 30 Amino Acid Analyzer. An aliquot without


32

particles was evaporated to dryness three times to remove remaining HCl. The remainder was

taken up in a buffer of pH 2.2. Samples were then separated using a cation exchange resin

and subsequently derivatized to be detected with a Merck Hitachi L-2480 fluorescence

detector. Relative error for individual AA monomers is typically 0.2 to 3.0% after duplicate

analysis. Aspartic acid and asparagine were measured as Asp and glutamic acid and

glutamine were measured as Glu due to acid hydrolysis.

5.2.7. Oxygen

Oxygen analyses were carried out on MUCs on board RV SONNE by Matthias Haeckel. Fibre-

optic microsensors (FireStingO2 optodes) were lowered into the sediment in the liners in the

cold room of RV SONNE using a motorized micromanipulator. Every 500 µm a measurement

was taken by two optodes simultaneously. Between four and six profiles were done for each

core. The detection limit was 1 µmol/L and method precision was 1%.

5.2.8. Particulate organic carbon, carbonate, and particulate organic nitrogen

Particulate organic carbon (POC) and particulate organic nitrogen (PON) analyses were

carried out at GEOMAR, by the work group of Matthias Haeckel. Both parameters were

determined by gas-chromatography of CO2 and N2 using a Carlo Erba Element Analyzer (NA

1500). Prior to analyses, sediment samples were treated with HCl to release C from CO3.

Inorganic carbon was calculated by subtracting organic carbon from total carbon and was

subsequently converted to CaCO3.

5.2.9. Nitrate

Nitrate analyses were conducted on board RV SONNE by Matthias Haeckel’s work group from

GEOMAR directly after pore water sampling. A Hitachi UV/VIS spectrophotometer was used

and the analyses followed standard procedures described in Grasshoff et al. (1999). The

detection limit was 2 µmol/L and method precision was 3%.

5.2.10. Porosity

Porosity analyses were carried out at GEOMAR by the work group of Matthias Haeckel. The

weight difference between wet and freeze-dried sediment was used to calculate the porosity.

Chapter 2 – Small-scale heterogeneity in the Peru Basin

33

Chapter 2 – Small-scale heterogeneity of trace

metals including REY in deep-sea sediments

and pore waters of the Peru Basin, SE equatorial

Pacific

Title of publication

Small-scale heterogeneity of trace metals including REY in deep-sea sediments and pore

waters of the Peru Basin, SE equatorial Pacific

Authors

Sophie A. L. Paul, Matthias Haeckel, Michael Bau, Rajina Bajracharya, Andrea Koschinsky

Prepared for submission to

Biogeosciences


34


35

Small-scale heterogeneity of trace metals

including REY in deep-sea sediments and pore

waters of the Peru Basin, SE equatorial Pacific

Sophie A. L. Paul1, Matthias Haeckel2, Michael Bau1, Rajina Bajracharya1, Andrea Koschinsky1

1Department of Physics and Earth Sciences, Jacobs University Bremen, Germany

2GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany

Abstract

Due to its remoteness, the deep-sea floor remains an understudied ecosystem of our planet.

The patchy data sets that exist make it difficult to draw conclusions about processes that apply

to the entire area. In our study we show how different settings and processes impact sediment

heterogeneity on small spatial scales. We sampled solid phase and pore water from the upper

10 m of an approximately 13 km wide area in the Peru Basin, south east equatorial Pacific

Ocean, at 4100 m water depth. Samples were analyzed for trace metals including rare earth

elements and yttrium (REY) as well as for particulate organic carbon (POC), CaCO3, and

nitrate. The analyses revealed a surprisingly high small-scale heterogeneity of the deep-sea

sediment composition. While some cores have the typical green layer where Fe(II) in the clay

minerals is not oxidized, this layer is missing in other cores. This is due to varying organic

carbon contents: nitrate is depleted at 2-3 m depth in cores with higher total organic carbon

contents, but is present throughout in cores with lower POC contents, which then leads to

oxidation of Fe(II) to Fe(III) in the clay minerals. REY show shale-normalized (SN) patterns

similar to seawater with a relative enrichment of heavy REY over light REY, positive LaSN

anomaly, negative CeSN anomaly, as well as positive YSN anomaly and correlate with this Fe-

rich clay layer and in some cores also with P. We therefore propose Fe-rich clay minerals such

as nontronite as well as phosphates as REY controlling phases in these sediments. Variability

is also seen in dissolved Mn and Co concentrations, which might be due to dissolving nodules

in the suboxic sediment, and in concentration peaks of U, Mo, As, V, and Cu in two cores,

which might be related to deposition of different material at lower lying areas.

Keywords: deep-sea sediment, trace metals, REY, DISCOL, Peru Basin, buried polymetallic

nodules


36

1. Introduction

1.1. Fragmentary data sets from the deep-sea

The deep-sea floor below 1000 m covers approximately 60% of our planet’s solid surface

(Glover and Smith, 2003). In this large area, only a small part has been investigated so far and

the data sets are fragmentary. A recent study in the Clarion Clipperton Zone (CCZ), central

equatorial Pacific, found large-scale biogeochemical heterogeneity (Volz et al., 2018). The

analyses, however, were based on one core per work area only, hundreds of km apart. In the

past, often few, spread out samples were taken for pore water and solid phase geochemical

analyses (e.g., Drodt et al., 1997; Froelich et al., 1979; Haley et al., 2004; Klinkhammer, 1980;

Kon et al., 2014; König et al., 1999, 1997; Toyoda and Masuda, 1991). Processes might,

however, be different on small spatial scales. Mewes et al. (2014) could show this small-scale

biogeochemical pore water variability in the German contractor area for deep-sea mining in

the CCZ. It remains to determine whether studies of few samples are representative for large

areas of the deep-sea or if these results are coincidental snapshots of a largely unknown,

heterogeneous bigger picture.

1.2. Previous work in the Peru Basin

In contrast to most other deep-sea basins, the Peru Basin, located in the south east central

Pacific at approx. 4100 m water depth (Figure 1), has been well investigated including the

geochemical composition of its sedimentary solid phase, pore water, and early diagenetic

processes (Haeckel et al., 2001; König et al., 2001; Koschinsky, 2001; Koschinsky et al.,

2001a, 2001b; Marchig et al., 2001; Paul et al., 2018; Stummeyer and Marchig, 2001). This is

due to its suitability to function as a study site for impacts of polymetallic nodule mining in the

1980s and 1990s (Thiel, 2001; Thiel and Schriever, 1990). With the emerging scientific,

industrial, and political interest in deep-sea mining, the Peru Basin has recently received new

attention.


37

Figure 1: Map showing the Peru Basin and the location of the DISCOL area. The map was

created using GeoMapApp and its integrated default basemap Global Multi-Resolutional

Topography (GMRT).

In 1989, a DISturbance and reCOLonization experiment (DISCOL) investigated potential

impacts of polymetallic nodule mining in the Peru Basin (Thiel and Schriever, 1990). The

seafloor was plowed in a 11 km2 large circle, disturbing the upper cm of sediment layering and

removing the nodules from the surface (Thiel and Schriever, 1990). Geochemical

investigations of nutrients, dissolved organic carbon (DOC), amino acids, solid phase and

dissolved trace metals were conducted in the frame of the follow-up project ATESEPP in 1996

(Schriever et al., 1996). The geochemical work focused on the bioturbated surface layer, where

impacts of polymetallic nodule mining are expected (Haeckel et al., 2001; Koschinsky, 2001;

Koschinsky et al., 2001b, 2001a), whereas geochemical investigations of deeper sediment

layers down to 10 m were only performed on five cores (with only one of them located in the

DISCOL area) (Haeckel et al., 2001). Mineralogical investigations of long cores were

conducted extensively (Marchig et al., 2001; Weber et al., 2000, 1995). As part of recent work

in the MiningImpact project (https://jpio-miningimpact.geomar.de), the focus lay again on

surface sediments (Paul et al., 2018; Haffert et al., in prep). To understand biogeochemical

processes over longer time scales and to resolve more steps of the redox zonation, the

analysis of long sediment cores is crucial.


38

1.3. Early diagenesis in the Peru Basin

The Peru Basin is located at the southern border of the equatorial high-productivity zone

(Weber et al., 2000), where it receives high particulate organic matter inputs for the deep-sea.

Particulate organic carbon (POC) contents are 0.5-1 wt.% and oxygen penetrates approx. 5-

25 cm into the sediment (Haeckel et al., 2001; Paul et al., 2018). The oxic surface sediments

are rich in Mn oxides and associated elements, giving this layer its dark brown color

(Koschinsky, 2001; Paul et al., 2018). Below the oxic zone, the sediment is suboxic and Mn

oxides are reduced in the course of suboxic POC degradation, leaving the sediment with a tan

color. The Fe redox boundary is visible where the sediment color changes from tan to green.

The depth of the tan-green color change coincides with the NO3- penetration depth as the color

change indicates re-oxidation of Fe(II) to Fe(III) by NO3- (Drodt et al., 1997; König et al., 1999,

1997; Lyle, 1983). The Fe(II)/Fe(III) ratio changes from approx. 11%/89% in the tan layer to

37%/63% in the green layer (König et al., 1997). The first four steps of the typical redox

zonation for marine sediments presented by Froelich et al. (1979) – oxygen, nitrate, Mn oxide,

and Fe oxyhydroxide reduction – that develops from the order in which oxidants act as final

electron acceptors for the degradation of organic matter, are therefore visible here (König et

al., 1999; Paul et al., 2018).

1.4. Fe-rich clay minerals

Sediments of the DISCOL area are mainly composed of siliceous and calcareous oozes and

muds (Weber et al., 1995). The predominant clay minerals in the sediments are illite, kaolinite,

and chlorite (of largely detrital origin) while smectites (authigenic clay minerals) - such as

montmorillonite and nontronite - are present in smaller quantities (Fritsche et al., 2001; Marchig

et al., 2001). In the Peru Basin, the concentration of nontronite and other authigenic clay

minerals increases with increasing distance from the continent (Marchig et al., 2001).

Nontronite is the Fe(III)-rich member of the smectite group (Murnane and Clague, 1983) and

the Fe(III) in nontronite can be reversibly reduced, preferably at tetrahedral positions (Russell

et al., 1979). With reduction, the color changes from yellowish to blue-green (Drodt et al., 1997;

König et al., 1997; Lyle, 1983; Russell et al., 1979 and references therein). In contrast, Al-rich

smectites darken from off-white to gray upon reduction (Lyle, 1983 and references therein).

1.5. Rare earth elements and yttrium (REY)

Rare earth elements and yttrium (REY) are frequently used to reconstruct physico-chemical

environmental conditions and sediment provenance (e.g., Bau and Dulski, 1999; Bright et al.,

2009). Yttrium is trivalent like the rare earth elements (REE) and of similar atomic size as Ho

and therefore often included with the REE - then commonly called REY. The REE have slightly


39

decreasing atomic radii with increasing atomic number which can lead to fractionation,

resulting in distinct patterns in the shale-normalized (SN) plots, which can be used to

differentiate REY control phases and sedimentary processes. Analyses of REY in sediments

in other areas of the Pacific often found a REY association with Ca phosphates or Fe phases

(e.g., Elderfield et al., 1981; Toyoda et al., 1990), the latter especially in areas of hydrothermal

activity (German et al., 1990; Ruhlin and Owen, 1986). REY in sediments from the Peru shelf

have apatite pellets, which display heavy REY (HREY) enrichment as well as pronounced

negative CeSN anomalies and positive YSN anomalies (Piper et al., 1988). Sediments from the

Peru Basin that were interpreted to be hydrothermally influenced showed REYSN patterns

similar to seawater (HREY enrichment, negative CeSN anomaly, positive LaSN anomaly) but Eu

and Y were not reported or measured, respectively (Marchig et al., 1999).

Clay minerals such as illite and kaolinite show flat REY patterns when normalized to Post

Archean Australian Shale (PAAS), due to their detrital origin. Nontronite of hydrothermal origin

displays seawater like REYSN patterns except for a less pronounced CeSN anomaly (Alt, 1988;

Murnane and Clague, 1983) and sometimes a EuSN anomaly (Mascarenhas-Pereira and Nath,

2010). To the best of our knowledge, no nontronite REY data from non-hydrothermal origin

exists so far.

1.6. Research aim

The sampling campaign of the Joint Programming Initiative Oceans pilot action Ecological

aspects of deep-sea mining (MiningImpact) conducted with RV SONNE in 2015 found that the

sediments in the upper 10 mbsf are surprisingly heterogenous in the approx. 13 km wide study

area. Therefore, we aim to address the question which parameters show heterogeneity with

respect to sediment composition and sedimentation input? To shed more light onto this small-

scale regional variability, we investigated trace metal distributions in the solid phase and

corresponding pore water to distinguish patterns and exceptions with respect to sediment

layers, impacts of bathymetry, and early diagenetic processes. We consider such information

on small-scale variability important for interpreting the representativeness of individual

sediment cores on which previous studies were often based. Here we focus on parameters

relevant for the description of the redox zonation (POC, NO3-, Mn, Fe, and the Mn associated

metals Co, Ni, and Cu), REY and indicators for their control phases (P, Al, Fe), CaCO3 and Ba

for paleo-reconstructions, and redox-sensitive elements such as U, Mo, As, V, and Cd.


40

2. Methods

2.1. Sampling area and methods

Samples were taken during RV SONNE cruise SO242/1 in 2015 with a gravity corer (GC)

(Greinert, 2015). The sampling campaign was part of a project on deep-sea mining (Joint

Programming Initiative Healthy and Productive Seas and Oceans (JPI Oceans) pilot action

Ecological Aspects of Deep-Sea Mining (MiningImpact); https://jpio-miningimpact.geomar.de).

Therefore, areas connected to the deep-sea mining experimental sites are also listed in Table

1. A disturbance experiment mimicking nodule mining was conducted in this area in 1989

(DISCOL project), during which a circular area of approximately 11 km2 was traversed with a

plow harrow (Thiel and Schriever, 1990). The affected area is called the DISCOL experimental

area (DEA), while undisturbed sites around this area are reference areas. Within the DEA, we

sampled one slightly low-lying area (trough) as well as an area without nodules at the surface,

which did not show significant acoustic backscatter intensity in the side-scan sonar images

(black patch). In addition, one GC was taken inside an inactive small volcanic crater in close

proximity to the DEA (Figure 2).

The plowing affected approximately the upper 20 cm of sediment (Paul et al., 2018), which are

often removed or disturbed during GC sampling anyways so that the disturbance experiment

should not affect the comparison of the GCs, regardless whether they were sampled in

disturbed or undisturbed sites.

Table 1: Overview of sampled cores.

Sample ID SO242/1

Area Location Water depth [m]

Core length [cm]

No. of samples

Nodule on top

Buried nodules

38GC1 Reference South

7°07.537‘ S 88°27.047‘ W

4160.9 917 13 yes no

51GC2 DEA West 7°04.411‘ S 88°27.836‘ W

4147.7 978 16 no no

84GC3 DEA Black Patch

7°03.951‘ S 88°27.093‘ W

4146 947 17 no no

89GC4 Reference West

7°04.562‘ S 88°31.577‘ W

4125.4 958 11 yes 1

100GC5 DEA Trough 7°04.342‘ S 88°27.442‘ W

4150.9 878 14 no 3

123GC6 Reference East

7°06.045‘ S 88°24.848‘ W

4216.8 921 16 no 4

132GC7 Small Crater 7°03.369‘ S 88°26.031‘ W

4151.7 936 12 no 2


41

Figure 2: Map of the sampling locations in the Peru Basin. Bathymetric map adapted from Paul

et al. (2018). The circle indicates the DISCOL experimental area (DEA) that was traversed with

a plow harrow.

2.2. Sediment and pore water sampling

Once on deck, GCs were cut into 1 m pieces and then cut into a working and an archive half.

Working halves were immediately transported to the cool room (approx. 4°C), while the

counterparts were stored as archive halves. After visual inspection, samples were taken in

layers of different color, roughly one to two per meter, and transferred with plastic spoons into

50 mL acid pre-cleaned centrifuge tubes. Samples were centrifuged at 3200 rpm for 40

minutes at 4°C to separate pore water from the solid phase. In a glove box with a steady stream


42

of argon, pore water was then filtered through 0.2 µm cellulose acetate (CA) syringe filters,

which had been cleaned previously with 0.1 M suprapure hydrochloric acid (HCl) and

deionized water. Pore water samples were acidified with suprapure HCl (30%) using 1 µL for

1 mL of sample and kept cool until further analysis.

2.3. Chemical analyses

To determine bulk sediment metal concentrations, 100 mg of ground and oven-dried (105°C)

sample was acid pressure digested in a Pico Trace DAS system at 220°C for 12 hours using

3 mL of perchloric acid (HClO4, 70%, suprapure) and 3 mL hydrofluoric acid (HF, 38-40%,

suprapure). Samples were evaporated and taken up in HCl (20-30%, suprapure) two times

and at the end in 0.5 M nitric acid (HNO3, suprapure) and 0.47M HCl (suprapure). Some

digested samples had small black particles left after the digestion and were filtered with 0.2 µm

CA filters prior to analyses. For major elements, solutions of digested solid phase samples

were measured with ICP-OES (SpectroCiros SOP instrument) and for trace elements,

including REY, with ICP-MS (Perkin Elmer Nexion 350x). For pore water analyses, the sample

was first passed through an apex Q (ESI), which was connected to the ICP-MS, to decrease

background noise and to improve sensitivity. Additionally, dissolved V, Mn, Co, Cu, As, and

Mo were measured in kinetic energy discrimination mode using He gas to remove polyatomic

interferences. The certified reference materials (CRM) MESS-3 and BHVO-2 were used for

sediment and NASS-6, NASS-7, and SLEW-3 for pore water samples. Accuracy and precision

were determined based on averages of the CRMs from ICP-OES and ICP-MS runs. Accuracy

for Al in MESS-3 during ICP-OES measurements (n=13) was within 20% but has been known

for too low Al values for some digestion methods (Roje, 2010). Data below the limit of

quantification (LOQ) were excluded except for pore water As values for 84GC and 132GC that

were below the LOQ but were still included due to good agreement of NASS-7 As data, which

is in the same range as the sample concentrations. For detailed information about LOQ,

accuracy, and method precision see Supplementary Material 1.

Detailed tables with data for major and trace elements are available online at PANGAEA:

https://doi.org/10.1594/PANGAEA.903019.

2.4. Nitrate

Nitrate was measured directly after sampling on board RV SONNE. Analyses followed

standard procedures described by Grasshoff et al. (1999), using Cd for reduction to NO2- and

determining it as sulphanile-naphthylamide with a Hitachi UV/VIS spectrophotometer. Method

precision was 3% and the limit of detection 2 µmol/L.

https://doi.org/10.1594/PANGAEA.903019


43

Data is available online at PANGAEA:

38GC: https://doi.org/10.1594/PANGAEA.884953







2.5. Particulate organic carbon (POC) and CaCO3

Total carbon of freeze-dried, ground sediment was measured with a Carlo-Erba NA-1500

Elemental Analyzer, analyzing CO2 that was produced by flash combustion, at GEOMAR

laboratories in Kiel. To determine total organic carbon and CaCO3, carbonate-bound carbon

was removed with HCl from the sample prior to organic carbon measurement and the total

inorganic carbon content was calculated from the difference between total carbon and organic

carbon. It was then converted to CaCO3 wt.%.

Data is available online at PANGAEA:








2.6. Depth correction for GCs and CaCO3 correction

Part of the semi-liquid surface sediments of the DISCOL area is typically lost from the GCs

when placing the gear horizontally on deck. Hence, the thickness of the lost sediment was

estimated by comparison of various geochemical data (i.e. POC, CaCO3, porosity, dissolved

silicate) and core photos of the GCs with multicorer (MUC) cores to derive true sediment

depths of the samples. Between 10-30 cm were lost during sampling.

Solid phase data (except Ca) is presented on a carbonate-free basis and was corrected for

CaCO3 due to high carbonate concentrations in some layers. Different sediment aliquots were

taken for CaCO3 and metal analyses and therefore the corrections were calculated using

https://doi.pangaea.de/10.1594/PANGAEA.884953















44

CaCO3 data within an up to 15 cm range of mean metal sediment sample depth. Buried nodule

data was not corrected for CaCO3.

2.7. Reporting of REY data

All REY patterns are normalized to PAAS, using REY data from Taylor and McLennan (1985),

except for Dy from McLennan (1989). Anomalies of REY in the SN patterns were calculated

as described in equation (1). These equations calculate the ratio of e.g., CeSN/CeSN* which

results in the value of the anomaly and helps to discern the extent of the respective anomalies.

Calculation of CeSN anomaly after Bau and Dulski, 1996a:

(1) 𝐶𝑒

𝐶𝑒∗=

𝐶𝑒(𝑆𝑁)

(0.5∗𝐿𝑎(𝑆𝑁)+0.5∗Pr(𝑆𝑁))

3. Results

3.1. Core descriptions

The Mn-oxide-rich dark brown top layer was largely lost during GC sampling in all cores, except

for the core from Small Crater where 10 cm remained and Reference West, where it was

absent. Below, all cores have a light brown to gray-brown color (2.5 Y 5/2 or 6/2 on the Munsell

color chart; de Stigter, 2015) down to approx. 2-2.5 m, followed in four cores (of Reference

South, DEA Black Patch, DEA Trough, and Reference East) by a greenish-gray color (5Y 5/2,

5Y 6/2; 5GY 5/1 on the Munsell color chart; de Stigter, 2015)) to approx. 5-7 m depth. The

cores of Reference West, DEA West, and Small Crater showed no extensive greenish layers,

but olive (2.5Y 5/3 at around 1 m in the DEA West core and 2.5Y 5/4 at around 1-2 m in the

Small Crater core) colors were found. In the bottom 2-2.5 m of all GCs mottled dark brown

sediment (10YR 4/3, 4/4 and 5/4 on the Munsell color chart; de Stigter, 2015) was found

(Figure 3).


45

Figure 3: Combined photos of the individual GCs with corresponding nitrate profiles. Green layers are marked with green boxes.


46

Figure 4: POC profiles of the GCs.

The GCs of Reference South and Reference West recovered a nodule from the sediment

surface, whereas buried nodules were found in the cores of Reference West at 458 cm, at

DEA Trough at 387 cm, 468 cm and 667 cm, and at Reference East at 290 cm. Additionally,

buried nodules occurred at DEA Trough at 564 cm, at Reference East at 346 cm, 747 cm and

870 cm, and at Small Crater at 719 cm and 792 cm but were not analyzed as part of this study.

Consequently, buried nodules exist below 290 cm in the DISCOL area. The dissolving nodules

at DEA Trough at 468 cm, 564 cm and 667 cm, and at Reference East at 290 cm and 747 cm

have brownish ‘halos’ around them in the green sediment. At DEA Black Patch at 487 cm and

at DEA Trough at 575 cm, there are brown patches within the green sediment without a buried

nodule being visible anymore.

At Reference East, diffuse dark gray bands of approximately 1 cm thickness are found at

depths of 229.5 cm, 236.5 cm and 330 cm. The dark gray bands are present again between

324 cm and 358 cm, from 386 cm to 402 cm and 510 cm to 518 cm depth (de Stigter, 2015).

Between 476 cm and 500 cm, the gray bands extend vertically (de Stigter, 2015).


47

POC and nitrate are presented because they are important parameters when analyzing the

redox zonation of marine sediments. POC contents in the sediment vary between 0.5 and

0.8 wt.% in the upper layers and decrease with depth to 0.1 to 0.3 wt.% (Figure 4). Nitrate

concentrations are 50-70 µmol/L in surface sediments and are depleted (<10 µmol/L) below

the upper 2-3 m except in cores Reference South, where NO3- is depleted below 6 m, and

Reference West and Small Crater, where NO3- remains at approx. 25 µmol/L throughout

(Figure 3).

3.2. Solid phase Ca, CaCO3, Ba, Al, Fe, Mn and associated metals

Calcium concentrations are around 1 wt.% throughout most of the sediment cores with

increased concentrations of up to 15 wt.% between 150 and 500 cm as well as between 800

and 1000 cm (Figure 5). Calcium carbonate concentrations are therefore also elevated in these

depth ranges, with concentrations of up to 35 wt.%. Barium concentrations are between 0.5

and 1 wt.% in the upper 400 cm and increase downcore, except at DEA Trough, where

concentrations are below 0.1 wt.% and only significantly increase below approx. 7.5 m and at

Small Crater, where concentrations are relatively constant (Figure 5).

Figure 5: Depth profiles of solid phase Ca, CaCO3, and Ba concentrations, as well as Ba/Al

ratios, highlighting two layers where carbonate was preserved.


48

Figure 6: Solid phase Al, Fe, Mn, P, Nd, Cu, Ni, and Co concentrations in the sediment cores

including those of the buried nodules at Reference West at 458 cm, at DEA Trough at 387 cm,

468 cm and 667 cm, and at Reference East at 290 cm depth. Nd is shown as a representative

of the REY. Fe/Al and Mn/Al ratios in the sediment (i.e. no data for the nodules is shown) are

also displayed as depth profiles, focusing on the Fe and Mn enrichment in relation to

continental sources (Al).


49

Aluminum concentrations decrease below 400 cm depth, most strongly at the western sites

Reference West and DEA West. In these cores, concentrations of P, Nd, Mn, as well as metals

associated with Mn, such as Cu, Ni, and Co, increase below 400 cm (Figure 6). The sum of

REY concentrations varies between approx. 180 ppm and 550 ppm (not shown). The buried

nodules at Reference West, DEA Trough, and Reference East show similar to slightly lower

REY concentrations than the sedimentary REY (see Nd in Figure 6). Iron displays a constant

concentration of 3-4 wt.% down to 3-4 m. Below, Fe concentrations increase up to 7.5 wt.% at

the bottom of all cores (Figure 6). Consequently, the Fe/Al ratio, which eliminates effects from

CaCO3 and opal dilution and allows the interpretation of Fe depletion or enrichment relative to

detrital sources (Lyons et al., 2003), is stable in the upper approx. 400 cm at around 0.65-0.75

and increases to 1.2-1.5 at depth. The increase pointing to an Fe enrichment is much more

pronounced in the westerly cores Reference West and DEA West, while the easterly cores

show no substantial increase (Small Crater) or only to around an Fe/Al ratio of 1.10 (Reference

East). Mn/Al displays similar profiles, with higher ratios in Reference West and DEA West (0.3-

1.3), while the other cores have similar ratios between 0.02 and 0.2 except for a few single

layer outliers.

3.3. Pore water Mn, Co, Cu

Figure 7: Dissolved Mn, Co, and Cu concentrations in the pore water of the sediment cores.

No pore water could be extracted from buried nodules.

Manganese concentrations in the pore water increase with depth in varying gradients,

asymptotically reaching maximum concentrations of 40-130 µmol/L at depths below 5-8 m

(Figure 7). Concentrations are lower in the western areas and the Small Crater where nitrate

does not get depleted (Figure 3). Such a distinct difference among the sites can also be

observed in dissolved Co concentrations. However, dissolved Co concentration profiles display

elevated concentrations compared to bottom water already between 2 and 3 m, and show


50

further increase below 6 m. The Reference West core exhibits the lowest concentrations.

Dissolved Cu concentrations remain rather low and show no downcore trend in contrast to Mn

and Co.

3.4. Redox-sensitive elements Mo, U, As, V, Cd: solid phase and pore

water

Figure 8: Top: Solid phase concentrations of U, Mo, and V. Concentration peaks are visible at

229.5, 236.5 cm and 330 cm for Reference East coinciding with the gray bands in the sediment

(see pictures on the right). In this core, also a dissolving nodule was found at 290 cm (see

pictures on the right). Bottom: Dissolved concentrations of U, Mo, V, As, and Cd in the pore

water. Depths 229.5 cm and 290 cm of Reference East were not measured. Concentration

peaks are visible at 236.5 cm and 330 cm for Reference East coinciding with the gray bands

in the sediment (see pictures on the right).

Pore water concentrations of the redox-sensitive elements U, Mo, and As as well as Cd are

constant with depth in suboxic sediments, and U and Mo also show straight profiles in the solid

phase (Figure 8). Arsenic and Cd could not be determined in the solid phase due to the acid

digestion procedure using HF and therefore formation of gaseous AsF5 as well as unreliable


51

Cd measurements with the ICP-MS. Considerable peaks in the solid phase and pore water

concentrations of U, Mo, and As are, however, visible for Reference East at depths 229.5 cm,

236.5 cm and 330 cm, where diffuse dark gray bands of approximately 1 cm thickness exist in

the sediment (de Stigter, 2015). Vanadium concentrations peak at 229.5 cm in the solid phase

(235 ppm) and the concentration is still elevated at 236.5 cm (190 ppm), which is again

reflected in the pore water profiles. There is an additional peak in the solid phase concentration

at 290 cm, where the buried nodule was sampled, but no pore water data exists for this exact

layer. At DEA Black Patch, dissolved U, V, and Cu peaks coincide at 251 cm and U and Cd at

318 cm (Figure 8). Solid phase concentrations of U and V are also elevated in these layers

(Figure 8).


52

3.5. REY patterns

Figure 9: REYSN patterns of the seven cores from this study and for the clay minerals

nontronite, illite, and kaolinite from literature for comparison.


53

All solid phase REYSN patterns show an enrichment of HREY over light REY (LREY) with

LaSN/YbSN ratios of 0.20-0.50, a negative CeSN anomaly, and positive LaSN, EuSN, GdSN, and

YSN anomalies (Figure 9). The negative CeSN anomaly increases with depth (CeSN/CeSN*=0.6-

0.3), the only exception being Small Crater, where the CeSN/CeSN* ratio remains at around 0.6

throughout the core. Y/Ho ratios range between 29 and 42, i.e. representing chondritic to

super-chondritic values, and Eu/Eu* ratios are between 1.2 and 1.4. REYSN patterns of the

buried nodules show LaSN/YbSN ratios of 0.40-0.44 similar to the sediment solid phase REY,

with negative CeSN anomalies, slightly positive LaSN, EuSN, and GdSN anomalies, and Y/Ho

ratios of 27-30 (Figure 9). Pore water REYSN also show a HREY enrichment, a negative CeSN

anomaly and a positive YSN anomaly (Figure 10), similar to the sedimentary solid phase REYSN

patterns.

Figure 10: Measurable REYSN pore water patterns from the Peru Basin.

4. Discussion

4.1. Paleoceanographic context: sedimentation history based on

CaCO3 and Ba preservation

Sediments in the Peru Basin consist of clays and siliceous mud with some layers rich in CaCO3

(Marchig et al., 2001; Weber et al., 1995) and CaCO3 as well as Ca concentrations can be

used to deduce carbonate layers (Figure 5). During times when the seafloor lies above the

carbonate compensation depth (CCD), calcareous ooze can be preserved in the sediments


54

upon burial. The present CCD is located approximately between 4200 and 4250 m water depth

(Weber et al., 2000). Carbonate layers are present in the DISCOL area between 150-500 cm,

concentrations and depths of CaCO3 peaks vary slightly among the cores. Concentrations are

lowest in the western cores Reference West and DEA West, which could be a sampling artefact

due to sparse sampling, but both cores as well as DEA Black Patch have a second carbonate-

rich layer at the base of the cores at approx. 800-1000 cm (Figure 5). Carbonate dilutes other

mineral phases, such as clay and Mn and Fe oxides, which is why concentrations of various

(trace) elements in the solid phase, e.g., Al, Fe, Cu, Mn, Co, Ni, Zn, and REY are lower in

carbonate-rich layers, while few are enriched, e.g., Sr.

Approximately 400 ka ago at the Mid-Brunhes event, major carbonate dissolution occurred in

the Pacific and less carbonate has been preserved since then (Weber et al., 1995; Weber and

Pisias, 1999). The uppermost carbonate peak at approx. 150-200 cm can therefore be

correlated to an approximate age of 400 ka. The end of the upper CaCO3-rich region

at 500 cm, potentially correlates with the beginning of the deepening of the CCD 1.1 Ma ago,

which remained until the Mid-Brunhes event 400 ka ago (Weber et al., 1995). The bottom

carbonate layer is absent in some cores and based on our data set it is not possible to date.

With 10-35 wt.% CaCO3, the carbonate layers in our cores have similar concentrations as

carbonate-rich layers described in the literature for the DISCOL area (Weber et al., 2000,

1995). Weber et al. (2000) distinguished areas of higher bioproductivity and hence higher

CaCO3 input into the sediments in the northwestern and northeastern Peru Basin from less

productive areas in the western and southern Peru Basin, including the DISCOL area.

Barium concentrations in marine sediments are often used as a marker for paleoproductivity

but its use depends on the reliability of the Ba record and that it was not subjected to alteration

after burial of marine barite (Dymond et al., 1992; Gingele et al., 1999; McManus et al., 1998).

In highly productive settings, authigenic barite formation can occur during diagenesis, while in

most other settings under oxic and suboxic conditions, pore waters are saturated with respect

to barite and solid phase barite is preserved (Reitz et al., 2004). Additionally, the biogenic

barium concentration needs to be distinguished from the detrital barium concentration before

it can be used as a paleoproductivity indicator (Gingele et al., 1999). We are therefore using

Ba/Al ratios to only focus on biogenic Ba (Figure 5).

Ba/Al ratios in the DISCOL sediments analyzed here show elevated concentrations below

approx. 350 to 450 cm, depending on the core, except for the cores from DEA Trough and

Small Crater, which display elevated concentrations only below 8 m or low concentrations


55

throughout, respectively (Figure 5). The layers with elevated Ba/Al ratios suggest a higher

primary productivity and increased sedimentation rates at the time of deposition. These depths

correspond with the occurrence of buried nodules in the sediment, suggesting nodule burial

due to increased sedimentation rates. Sedimentation rates between 0.4 and 2.0 cm/ka have

been reported for Peru Basin surface sediments (Haeckel et al., 2001).

4.2. Green layers

Considering the small sampling area, the cores show a high heterogeneity of different layers

and thickness of these layers. The color change from tan to green, visible in four cores (Figure

3), represents the NO3- penetration depth and the green color results from increased Fe(II)

content in the nontronite (Drodt et al., 1997; König et al., 1999, 1997; Lyle, 1983). Nitrate is

present throughout the cores of Reference West and Small Crater (Figure 3) and consequently,

no green layers are observed, as Fe(III) dominates considerably in the nontronite. Nitrate is

depleted at approx. 3 m depth at DEA West but no green layer is visible. Dissolved Mn

concentrations are also lowest in these three cores (Figure 7). This may be due to the lower

POC contents of only 0.1-0.2 wt.% at depth and less in these cores compared to 0.2-0.4 wt.%

at depth at the other sites without green layers (Figure 4), which only allows for NO3- and

Mn(IV) reduction, but does not reach Fe(III) reduction in the electron acceptor sequence for

POC degradation.

The cores with extensive green layers were located in depressions (DEA Trough and

Reference East) and had few or no nodules on the surface (DEA Black Patch, DEA Trough,

Reference East). Mewes et al. (2014) discovered that microbial respiration was higher at sites

without nodules in the CCZ. This fits the scenario in the Peru Basin, where fewer nodules occur

in areas with more POC and therewith probably higher microbial activity. Most buried nodules,

however, were found in depressions (Table 1) so that their distribution and burial might be

related to bathymetric effects, i.e. sediment deposition centers. The dissolving nodules were

found in the suboxic parts of the cores, as well as the brown patches inside the green sediment

layers (e.g., DEA Black Patch-497 cm and DEA Trough-585 cm). The latter might be remnants

of dissolving nodules because dissolving nodules impact their surrounding sediment, which is

also visible in the ‘halos’ around the larger buried nodules. Green sediment gets oxidized ‘back’

and is tan colored again, as Fe(II) in nontronite is oxidized to Fe(III) (König et al., 1997; Russell

et al., 1979), due to the provision of oxides by the nodules.

When clay minerals become concurrently enriched in Fe(III), they can transform into other clay

minerals, such as glauconite or nontronite (Pedro et al., 1978). Nontronite can form in three


56

ways at the seafloor: (1) “precipitation from hydrothermal fluids”, (2) “alteration of volcanic

rocks”, and (3) “low-temperature combination of biogenic silica and” Fe (oxyhydr)oxides (Cole

and Shaw, 1983, p. 239). Hydrothermally derived nontronite has been found in Pliocene

sediments of the Peru Basin and the adjacent Bauer Basin, but volcanic activity in the DISCOL

area ended about 6 Ma ago (Marchig et al., 1999) and this age is not covered by the GCs

presented here. Therefore, it is most likely that Fe (oxyhydr)oxides and (biogenic) silica form

Fe(III)-Si complexes, which then develop into nontronite (pathway 3) (Cole and Shaw, 1983;

Pedro et al., 1978). This Fe(III) is provided by the buried nodules.

4.3. Sedimentary Fe/Al

Fe/Al ratios of 0.6-0.75 persist in the upper meters of all cores and throughout the core of the

Small Crater (Figure 6). This is in line with Fe/Al ratios of 0.6-0.7 of Pacific deep-sea sediments

from other locations (Bischoff et al., 1979; Paul et al., 2019). Elevated Fe/Al ratios of up to 1.3

or even above 3 in certain layers of our cores coincide with Fe/Al ratios of metalliferous layers

in the central equatorial Pacific below approx. 5.5 or 8 m (Fe/Al: 1.3-1.7; Paul et al., 2019).

Dissolving nodules analyzed in this study have Fe/Al ratios between 1.2 and 5.3, suggesting

that the enrichment in the sediment could result from the dissolving nodules.

4.4. REY control phases

Like Fe and P, REY concentrations increase with depth, especially at Reference West and

DEA West (Figure 6), and except at Small Crater. All cores except Small Crater can be divided

into two parts based on the REY concentration increase, increase in Fe/Al ratios, and a

decrease of CeSN/CeSN* ratios: Reference West and DEA West at 4.5 m, Reference South,

DEA Black Patch and DEA Trough at 6 m, and Reference East at 8 m (Figure 9). The Fe/Al

ratios remains steady in the Small Crater core, as well as the negative CeSN anomaly. The first

three above mentioned cores (Reference West, DEA West, Reference South) also have higher

Y/Ho ratios in their lower parts. The concentration increase is associated with the bottom of

the green layer in cores Reference South, DEA Black Patch, DEA Trough, and Reference

East. In Reference West and DEA West, where no green layer exists, the concentration

increase correlates with the color change from tan to dark brown at approx. 4.5 m and the

increasing Fe and P concentrations at the corresponding depth. REY are most abundant,

where a higher percentage of Fe(II) in the clay minerals prevails (Reference West and DEA

West).

Neodymium (Nd) is used in the correlations to represent the REY. Correlations of solid phase

Nd and major elements such as Al as an indicator for detrital inputs, Mn as an indicator for Mn


57

oxides, Fe as an indicator for Fe phases (Fe (oxyhydr)oxides or Fe-rich clay minerals), and P

as an indicator for phosphates – showed that Fe, Al, and P correlate positively with Nd (Figures

11 and 12) while Mn shows no correlation.

Iron-Nd correlations are positive in all cores (Figure 11) and show the highest Pearson R

coefficients overall. At Reference South and DEA West, Fe also correlates with Al in the upper

part of the cores (Figure 11). The Fe-Al correlation points to the occurrence of an Fe-rich clay

mineral. The carrier phase for the REY could therefore be a Fe-rich clay such as nontronite.

The REY also correlate with Al at Small Crater and at DEA West until approx. 450 cm and at

Reference West below approx. 450 cm, correlating with the color change from tan to dark

brown sediment in the latter two cores. Even though it is unclear why only part of the core

shows a correlation of Al with Nd and Fe and especially why this is once the upper and once

the lower part, it corroborates the association of REY with Fe-rich clay minerals. Additionally,

REYSN patterns of detrital clay minerals, such as illite or kaolinite, are flat due to their detrital

origin and can therefore be excluded here due to HREY enrichment and the pronounced

negative CeSN anomaly (Figure 9). The REYSN patterns with La/Yb << 1, negative CeSN

anomaly, and positive LaSN, GdSN and YSN anomalies are similar to REYSN patterns from

nontronites (Figure 9), which occur in these sediments due to the tan-green color change and

the high Fe/Al ratio. While the nontronite REYSN patterns in literature are from hydrothermally

derived nontronite, the nontronite in cores from this study are not hydrothermally affected but

rather derived from altered clay minerals or Fe (oxyhydr)oxides. To the best of our knowledge,

no REY data of nontronite that evolved from the combination of Fe (oxyhydr)oxides and

biogenic silica exists.


58

Figure 11: Top: Fe-Nd plot and correlations for all cores. Pearson R coefficients show positive

correlations of REY with Fe for all cores. Middle: Al-Fe plot. Only positive correlations for the

upper parts of Reference South and DEA West are shown. Bottom: Al-Nd plot. Only positive

correlations for the upper part of Reference South, as well as for the lower part of Reference

West and the entire Small Crater core.


59

Figure 12: Top: P-Fe correlations for Reference South, DEA West, Reference West, and DEA

Black Patch. Middle: P-Ca correlations for samples with Ca concentrations below 1.5 wt.%

except for Reference East where P and Ca do not correlate. Samples with Ca concentrations

above 1.5 wt.% were excluded from the regression analyses because most of the Ca is then

not bound in Ca phosphates. Bottom: P-Nd correlations for all samples except Small Crater

where P and Nd do not correlate and excluding the DEA Black Patch sample with exceptionally

high P concentrations.


60

Phosphorus correlates with Fe in cores from Reference South, DEA West, Reference West,

and DEA Black Patch, which could be a sign of P bound to Fe phases. But P-Ca correlations

in the Ca-poor parts of all cores except Reference East are positive as well (Figure 12),

indicating a Ca phosphate phase. Since P and Nd also correlate in all cores except Small

Crater (Figure 12), phosphates might play a role as a REY controlling phase. The correlation

of P and Nd in some cores is similar to results from large areas of the central equatorial Pacific,

where REY are bound to Ca phosphates (e.g., Elderfield et al., 1981; Paul et al., 2019; Toyoda

et al., 1990; Toyoda and Masuda, 1991; Toyoda and Tokonami, 1990). There, Ca phosphates

show MREY enriched patterns with no or negative Ce anomalies ( Paul et al., 2019; Toyoda

et al., 1990; Toyoda and Masuda, 1991). Apatite pellets with similar REY patterns as presented

here (Figure 9) were found on the Peru shelf (Piper et al., 1988), supporting the possibility of

Ca phosphate control on REY in the sediment. The sedimentary REYSN pattern is also similar

to the pore water REYSN pattern (Figure 10), suggesting that REY are continuously

incorporated into the Ca phosphates from ambient pore water. This is the same process as in

the central equatorial Pacific (see e.g., Paul et al., 2019, Chapter 4), but the pore water REYSN

pattern is different in the Peru Basin, leading to different patterns in the solid phase.

In conclusion, we cannot ascertain a definite REY carrier phase in the sediments of the Peru

Basin based on element correlation. Calcium phosphates and Fe-rich clays are potential

control phases, yet both phases seem to incorporate REY from the ambient pore water without

major fractionation because the pore water and solid phase REYSN patterns match (compare

Figures 9 and 10). Similar REYSN patterns have been found in sediments in the DISCOL area

and were explained to result from hydrothermal inputs and scavenging of REY from seawater

(Marchig et al., 1999). Since hydrothermal inputs do not play a role in the sediments we

investigate here, it is unlikely that hydrothermal activity is the main reason for the REYSN

patterns in the GCs from this study. We believe that the incorporation of REY from ambient

pore water is the dominant process resulting in the found REYSN patterns.

4.5. Dissolved and solid phase Mn and associated metals

Dissolved Mn concentrations increase with depth and from west to east (except for Small

Crater), which depict a mirror image to solid phase Mn in these cores as well as in the surface

sediments, where concentrations are higher in the west than in the east (Paul et al., 2018).

Similarly, dissolved Co concentrations at depth are higher in the east than in the west and vice

versa in the solid phase except for Small Crater (Figure 6). Both western cores and Small

Crater have lowest POC concentrations and deepest NO3- penetrations depths (Figure 3).


61

Manganese oxides are therefore less utilized as electron acceptors during the degradation of

organic matter in these cores and less Mn is released to the pore water.

The marked increase of dissolved Mn and Co concentrations at depth might also be related to

a release of trace metals from buried nodules. Reference South, DEA Black Patch, DEA

Trough, and Reference East show highest dissolved Mn and Co concentrations at depth and

show green layers, in which nodules are dissolving.

Copper does not display the west-east-trend in the pore water profiles and does also not show

an increase at depths where Mn and Co are enriched in the suboxic zone. A deviation of Cu

from the behavior of Mn, Co, Ni etc. has already been found in our previous study (Paul et al.,

2018). While Mn, Co, and Ni are largely controlled by Mn oxides and reduction thereof (Heggie

and Lewis, 1984; Klinkhammer, 1980; Shaw et al., 1990), Cu is largely controlled by release

from organic matter during early diagenesis and only partially due to association with Mn

oxides (Klinkhammer, 1980; Shaw et al., 1990).

4.6. Redox-sensitive elements Mo, U, As, and V

The redox-sensitive metals Mo, U, As, and V are soluble under oxic conditions and are bound

to the solid phase under anoxic conditions in the sediment (Beck et al., 2008; Elbaz-Poulichet

et al., 1997; Wang, 2012). They display conservative type profiles in oxic waters (Beck et al.,

2008). In the gray bands in Reference East, where U, Mo, V, and As concentrations peak in

the solid phase and pore water, dissolved Co concentrations are low (even below the LOQ at

0.13 mg/kg) and dissolved Mn concentrations are slightly lower than in the surrounding

sediment above and below (Figure 7). This might be a sign of locally oxic conditions releasing

U, Mo, As, V, and Cd into the pore water but removing Co and Mn. Elevated concentrations of

U, M, V, and As in the pore water are also possible due to the chemical equilibrium between

the high concentrations in the solid phase and the pore water, so that oxic conditions might not

necessarily be required. Total dissolved S in the pore water is not elevated in these layers,

while at 238 cm, where another gray band was sampled for solid phase S analyses, elevated

concentrations of 0.53 wt.% S were measured compared to ~0.3-0.4 wt.% S in the remaining

layers of the core, possibly a sign of anoxic-sulfidic deposition of material, but this cannot be

said with certainty.

Pore water data for these redox-sensitive elements needs to be viewed with caution due to

potential artifacts from sampling carbonate-rich sediment in a depressurized environment

(Haley et al., 2004). This usually leads to lower e.g., U concentrations in the pore water


62

because U is removed from the pore water by precipitating CaCO3 (Toole et al., 1984). The

elevated concentrations we find here should therefore be reliable, as the sampling artefacts

would not increase the dissolved concentrations of the respective elements.

Both cores, DEA Black Patch and Reference East, are located in areas with few or no nodules

at the seafloor surface. Possibly, the gray bands impact upward metal cycling by removal of

e.g., Mn and Co from the pore water due to changed redox conditions. In addition, Reference

East is located at greater water depth. Deposition of different material – also more organic

material that might lead to periods of anoxic conditions – is the standard explanation for

enrichment of U, Mo, As, and V in other settings, but the observations here can most likely not

be explained by anoxic conditions because of low POC contents (~0.3-0.5 wt.% in the

Reference East core).

The solid phase and dissolved U, V, and Cu concentration peaks in DEA Black Patch suggest

the presence of a Cu-rich uranium-vanadium phase. This is known from oxidation fronts in

turbidites in North Atlantic clays, where U, V, and Cu are enriched in the solid phase (Colley

et al., 1984; Colley and Thomson, 1985). The metals are mobilized during organic oxidation of

the turbidite material, migrate downwards, and are immobilized at depth (Colley et al., 1984).

They get preserved by burial of other material on top (Colley and Thomson, 1985) which might

have also happened at Reference East.

5. Conclusions

The analyses of seven GCs from the DISCOL area showed that a deep-sea basin can be

highly heterogeneous on small spatial scales. The variability is visible in organic matter content

(POC) and related differences in NO3-, Mn, Fe (and REY) concentrations as well as for rare

layers where redox-sensitive elements such as U, Mo, V, and As are enriched. Especially

Small Crater is different in the measured parameters from the other cores: no green layer and

generally more layers with dark brown sediment, Fe/Al ratios remain constant, and REY

correlate with Fe and Al throughout the cored sediment. Since these exceptions correspond to

special locations, such as lower lying areas without/less nodules where redox-sensitive metals

are enriched and the Small Crater where a different deposition environment might prevail, the

importance of small topographical changes becomes clear. Variability, however, could be

higher at DISCOL than in areas further away from continents, because the DISCOL area might

be more impacted by continental inputs and higher primary productivity than e.g., the CCZ.


63

The results call for caution when extrapolating findings from a small set of samples to larger

ocean areas. With respect to deep-sea mining, the results show how variable the deep-sea

floor can be and that extensive baseline studies are necessary before the onset of mining and

impact analyses. This has been stressed by various advocates for the preservation of the

deep-sea ecosystem (Glover and Smith, 2003; Mengerink et al., 2014; Schindler and Hilborn,

2015; Van Dover et al., 2014). Since the geochemical composition of the sediment, including

POC content and redox conditions, has a major impact on microbial processes in the sediment

and associated biological life, this small-scale heterogeneity may also be relevant for biological

productivity and diversity in these deep-sea areas.

Another interesting finding of this study is the influence of dissolving nodules on the

surrounding sediment and geochemical cycling, e.g., in the form of visible “halos” in the

sediment or increased Fe/Al ratios and dissolved Mn and Co concentrations in the pore water.

These dissolving nodules can also lead to significant small-scale differences in the

mineralogical and chemical composition of sediment cores and care should be taken that such

signatures are not misinterpreted as e.g., hydrothermal influence.

Acknowledgements

Thanks to the crew of RV SONNE and chief scientist Jens Greinert on cruise SO242/1, who

enabled our sampling. Our great appreciation goes to Katja Schmidt, Annika Moje, Inken

Preuss, Tim Jesper Suhrhoff and Laura Ulrich for help with sampling and laboratory analyses

at the geochemistry lab, Jacobs University Bremen. We thank Meike Dibbern, Bettina

Domeyer, Anke Bleyer, and Regina Surberg for analytical work during the RV SONNE cruise

and at GEOMAR. Thanks also go to Anne Peukert, GEOMAR, for providing the original

bathymetry map, Laura Haffert from GEOMAR for providing the depth correction for the GCs,

and Charlotte Kleint, Jacobs University Bremen, for helpful comments during the writing

process. The work was funded by the German Federal Ministry of Education and Research in

the framework of the JPI Oceans MiningImpact project (grant no. 03F0707A+G).


64

Chapter 3 – Impacts of nodule mining disturbances on deep-sea sediments

65

Chapter 3 - Biogeochemical regeneration of a

nodule mining disturbance site: trace metals,

DOC and amino acids in deep-sea sediments

and pore waters

Title of publication

Biogeochemical regeneration of a nodule mining disturbance site: trace metals, DOC and

amino acids in deep-sea sediments and pore waters

Authors

Sophie A. L. Paul, Birgit Gaye, Matthias Haeckel, Sabine Kasten, Andrea Koschinsky

Published in

Frontiers in Marine Science (2018) 5:17

doi: 10.3389/fmars.2018.00117

Originally published by Frontiers

https://doi.org/10.3389/fmars.2018.00117


66


67

Biogeochemical regeneration of a nodule

mining disturbance site: trace metals, DOC and

amino acids in deep-sea sediments and pore

waters

Sophie A. L. Paul 1*, Birgit Gaye 2, Matthias Haeckel 3, Sabine Kasten 4,5 and Andrea

Koschinsky 1

1Department of Physics and Earth Sciences, Jacobs University Bremen, Bremen, Germany

2Institute of Geology, University of Hamburg, Hamburg, Germany


4Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Bremerhaven,

Germany

5Faculty of Geosciences, University of Bremen, Bremen, Germany

Abstract

Increasing interest in deep-sea mineral resources, such as polymetallic nodules, calls for

environmental research about possible impacts of mineral exploitation on the deep-sea

ecosystem. So far, little geochemical comparisons of deep-sea sediments before and after

mining induced disturbances have been made, and thus long-term environmental effects of

deep-sea mining are unknown. Here we present geochemical data from sediment cores from

an experimental disturbance area at 4,100 m water depth in the Peru Basin. The site was

revisited in 2015, 26 years after a disturbance experiment mimicking nodule mining was carried

out and compared to sites outside the experimental zone which served as a pre-disturbance

reference. We investigated if signs of the disturbance are still visible in the solid phase and the

pore water after 26 years or if pre-disturbance conditions have been re-established.

Additionally, a new disturbance was created during the cruise and sampled 5 weeks later to

compare short- and longer-term impacts. The particulate fraction and pore water were

analyzed for major and trace elements to study element distribution and processes in the

ORIGINALRESEARC H published:06 April 201 8

7 doi:10.3389/fmars.2018.0011

https://www.frontiersin.org/articles/10.3389/fmars.2018.00117/full





http://loop.frontiersin.org/people/413206/overview




https://www.frontiersin.org/journals/marine-science#editorial-board









68

surface sediment. Pore water and bottom water samples were also analyzed for oxygen,

nitrate, dissolved organic carbon, and dissolved amino acids, to examine organic matter

degradation processes. The study area of about 11 km2 was found to be naturally more

heterogeneous than expected, requiring an analysis of spatial variability before the disturbed

and undisturbed sites can be compared. The disturbed sites exhibit various disturbance

features: some surface sediments were mixed through, others had the top layer removed and

some had additional material deposited on top. Pore water constituents have largely regained

pre-disturbance gradients after 26 years. The solid phase, however, shows clear differences

between disturbed and undisturbed sites in the top 20 cm so that the impact is still visible in

the plowed tracks after 26 years. Especially the upper layer, usually rich in manganese-oxide

and associated metals, such as Mo, Ni, Co, and Cu, shows substantial differences in metal

distribution. Hence, it can be expected that disturbances from polymetallic nodule mining will

have manifold and long-lasting impacts on the geochemistry of the underlying sediment.

Keywords: heavy metals, deep-sea mining, ecosystem disturbance, long-term changes,

DISCOL

1. Introduction

Deep-sea mining has been featured prominently on the political and scientific agenda for the

past years and experiences a new phase of interest after first major exploration activities in the

1970s and 1980s. Recent advances in deep-sea mining technology, such as the building of

collector prototypes, (Gollner et al., 2017) and an increasing number of exploration contracts

issued by the International Seabed Authority (ISA): seven of 16 in the last 5 years (International

Seabed Authority, 2014) show the need for environmental baseline data and knowledge about

the response of deep-sea sediments to impacts by polymetallic nodule mining. The number of

articles discussing the need for mining regulation to protect the deep-sea ecosystem and its

fauna has skyrocketed in recent years (Barbier et al., 2014; Davies et al., 2007; Glover and

Smith, 2003; Mengerink et al., 2014; Ramirez-Llodra et al., 2011; Van Dover, 2011; Van Dover

et al., 2014). Yet, most research on trace metals in deep-sea sediments is approximately 30

years old (Heggie et al., 1986; Heggie and Lewis, 1984; Klinkhammer et al., 1982;

Klinkhammer, 1980; Müller et al., 1988; Sawlan and Murray, 1983; Shaw et al., 1990) and

recent trace metal research mostly does not focus on the deep-sea (Beck et al., 2008; Kowalski

et al., 2009; Morford et al., 2005; Morford and Emerson, 1999). Therefore, more state-of-the-

art deep-sea sediment studies focusing on the current issue of mining impacts are needed.


69

Exploration for polymetallic nodules has been carried out in many different areas of the oceans,

with a focus on the Clarion-Clipperton Fractures Zone (CCZ) in the central Pacific (Hein et al.,

2013). There, nodules have a high percentage of economically interesting metals—Ni, Cu, and

Co (Hein et al., 2013). All ISA exploration contract areas except one are located in the CCZ

(International Seabed Authority, 2014); however, some long-term research projects on the

impacts of nodule mining on the deep-sea ecosystem have also been carried out in the Peru

Basin. Polymetallic nodules in the Peru Basin are characterized by higher growth rates and

larger size in comparison to the CCZ nodules (Marchig et al., 2001). Cu contents are generally

lower than in CCZ nodules (Wegorzewski and Kuhn, 2014). The overlying waters are more

productive than in the CCZ and hence, sediments are characterized by higher organic carbon

concentrations of around 0.5 wt% up to rarely 1 wt%, low sedimentation rates (0.4–2.0 cm/ka)

and an oxygen penetration depth around 10–15 cm (Haeckel et al., 2001).

In 1989, a DISturbance and reCOLonization (DISCOL) experiment mimicking polymetallic

nodule mining was carried out in the Peru Basin. The deep-sea floor was plowed in an area of

approximately 11 km2 (Thiel, 2001). Environmental assessments were carried out 0.5, 3, and

7 years after the disturbance (Thiel, 2001). The assessments, however, mainly focused on

fauna (Thiel and Schriever, 1990) and the first geochemical studies in the wider DISCOL area

were conducted in 1996 (cruise SO106, ATESEPP project), unfortunately only after the

disturbance so that no baseline data from prior to the experiment exists. The six sampling sites

of SO106 were spread out across the Peru Basin and only one site was located in the DISCOL

experimental area (DEA) (Schriever et al., 1996). It is not known, however, if the 1996 DISCOL

sample is from within or outside a plow track because the multi-corer (MUC) sampling then

was not TV-guided and the geochemical data (Koschinsky, 2001) does not give a clear picture

to draw conclusions about a disturbance.

Mining operations to recover nodules will likely remove or disturb the upper 10–50 cm of

sediment and create a sediment plume (Cronan et al., 2010; Gollner et al., 2017; Oebius et al.,

2001; Thiel and Schriever, 1990). Ex-situ experiments with sediment cores from the Peru Basin

showed that pore water metals, dissolved organic carbon (DOC) and nutrients are released

when the sediment is stirred up (Koschinsky et al., 2001b). It has been assumed that such a

disturbance would also be caused by polymetallic nodule mining (Thiel and

Forschungsverbund Tiefsee-Umweltschutz, 2001). Depending on the redox zonation of the

area and depth of sediment removal, a change in redox zonation can occur. The redox zonation

develops as a result of organic matter degradation, following a roughly set sequence in which

oxidants are used according to their potential to produce energy: oxygen, nitrate, Mn-oxide,

Fe-oxide, and sulfate (Froelich et al., 1979). The redox zonation in marine sediments


70

determines how metals are distributed between the solid phase and pore water: metals are

either bound in the solid phase or dissolved in the pore water (König et al., 2001; Koschinsky,

2001). Elements soluble in oxic water (Mo, U, possibly V, As) are released from oxic pore

water, but they have similar concentrations in the oxic bottom water (Koschinsky, 2001). Mn,

Fe, Co, Ni, Cu, Zn, Cd, and Pb have higher concentrations in the sediment pore water than in

the bottom water, especially in suboxic pore water, which are up to two orders of magnitude

higher than in the bottom water at the sediment-water interface (Koschinsky, 2001). If the oxic

layer is thick enough (few cm), most pore water metals diffusing upwards from the suboxic

layer will be scavenged and bound to e.g. Mn-oxides because Mn-oxides are effective

scavengers and positively charged metal species such as Co, Ni, Cu, Zn, Pb, and Cd are

associated with Mn (Koschinsky, 2001). Similarly, if the disturbance is limited to the oxic layer,

Mn-oxides would bind most of the released heavy metals in a relatively short period of time

(Koschinsky et al., 2001b) which makes them immobile and they do not diffuse into the bottom

water (Koschinsky, 2001; Morford et al., 2005). The depth of the oxic layer hence is a decisive

factor for the heavy metal budget (Koschinsky, 2001). Besides the Mn-oxide rich surface layer,

polymetallic nodules also act as metal scavengers (Koschinsky et al., 2003). If these nodules

are mined, this option of metal scavenging is removed. If, however, the oxic layer is removed

or contracted, metals dissolved in the pore water from the suboxic layer can discharge into the

oxygenated bottom water, causing the release of dissolved heavy metals and an increase of

seawater heavy metal concentrations (König et al., 2001; Koschinsky et al., 2003). Since some

heavy metals could potentially reach toxic concentrations with detrimental effects for the fauna,

sediment disturbance and potential metal release is a serious issue to be considered with

respect to deep-sea nodule mining. Ecotoxicological experiments showed that LC50 values for

animals subjected to colder temperatures and higher pressures, to simulate deep-sea

environmental conditions, for dissolved Cu ranged between 8.85 and 29.4 µmol/L for

nematodes (Mevenkamp et al., 2017) and 380–420 µmol/L for shrimp (Brown et al., 2017a).

LC50 values for Cd ranged between 521 and 548 µmol/L for shrimp (Brown et al., 2017a).

Experiments are usually carried out with spiked Cu and Cd concentrations in the µmol/L range

(Auguste et al., 2016; Martins et al., 2017; Mevenkamp et al., 2017). Trace metals are an

important part of the biogeochemical cycle of the surface sediment and should be well

understood before mining commences.

These preliminary experiments from the past thus have shed some light on the geochemical

behavior of heavy metals in deep-sea sediments. Yet, no detailed in-situ studies or long-term

monitoring of the mining impacts have been carried out in the DISCOL area (as mentioned

above, only one station was sampled there in 1996) to determine: (1) degrees of disturbance

at different sites to obtain a comprehensive picture of the geographic extent and degree of the


71

disturbance and (2) the processes in the sediment and new equilibrium establishment after a

disturbance. This is essential as most geochemical processes in the deep-sea are slow and

therefore environmental recovery rates can also be expected to be slow.

As part of the European JPI Oceans MiningImpact project (“Ecological Aspects of Deep-Sea

Mining”) (GEOMAR, 2017), we revisited the DISCOL area in 2015 during the SO242 cruise

with RV SONNE, to study the geochemical long-term development of the site. We aim to

answer the following research questions: (1) Are there differences between the reference sites

demonstrating natural variability in the particulate fraction and pore water? After 5 weeks and

26 years, (2) are signs of the disturbance still visible in the solid phase and pore water or has

a new equilibrium been reached? (3) Are there differences between the disturbed sites across

the DEA and between the microhabitats within a disturbed track? Answering these questions

will help to understand natural variability and how deep-sea mining could affect the deep-sea

floor geochemistry. Since little is still known about sediment recovery after a disturbance, these

background studies are extremely valuable—especially if carried out on relatively long time-

scales. Decision-makers can also draw from our results as a basis for defining baseline data

and threshold values because if the disturbance sites vary considerably, this needs to be taken

into account for monitoring in future mining scenarios.

From a research perspective, the DISCOL area provides a good comparison to the well-

researched and industrially more pertinent CCZ, where similar studies have been and are

being carried out, but the geochemistry is quite different. The DISCOL area has an oxic layer

of approximately 10–15 cm (Haeckel et al., 2001), whereas the CCZ sediment is oxic down to

ca. 200 cm and deeper (Kuhn et al., 2017; Mewes et al., 2016, 2014). A comparison of these

sites will help to assess the possible range of changes in the trace metal cycle during deep-

sea mining in relation to the different environmental conditions.

2. Materials and methods

2.1. Site description

The Peru Basin is located in the south eastern tropical Pacific (Marchig et al., 2001).

Predominantly, siliceous and calcareous muds and oozes make up the sediments in this region

(Weber et al., 1995). For detailed site description of the DISCOL area and the disturbance

experiment carried out in 1989 please refer to Thiel and Schriever (1990); Boetius (2015);

Greinert (2015).


72

During RV SONNE cruise legs SO242/1 and 2 in 2015, the reference sites outside the DEA

(Figure 2A), as well as undisturbed and disturbed sites inside the DEA were sampled

(Figure 1). The 26-year old plowed tracks exhibit various disturbance features. Due to the plow

harrow, grooves traverse the sediment and form ripples and valleys. Throughout the DEA

“white patches” of lighter sediment occur in the disturbed sites. These three features are

microhabitats of the disturbed sites, which were sampled to study the disturbance variety

(Figure 2C).

Figure 1: Sampling sites of sediment cores in the DISCOL area (adapted from a map by Anne

Peukert, GEOMAR, working group of Jens Greinert). The circle indicates the DISCOL

experimental area (DEA) in which the disturbance experiment had been carried out in 1989.

During leg SO242/1 (Greinert, 2015), a new disturbance in addition to the plow tracks from

1989 was created using an epibenthic sled (EBS), and the affected sites were sampled

approximately 5 weeks later during leg SO242/2. The sediment surface layer was visibly


73

removed so that the lighter sediment below the Mn-oxide rich layer became exposed

(Figure 2B). The EBS track sediment disturbance was created to simulate nodule mining and

to be able to take samples for geochemical analyses shortly after the disturbance. This had

not been done in the frame of the original DISCOL project and geochemical data from shortly

after the impact is missing. Therefore, the EBS track samples add a point in time between

undisturbed samples and samples from the 26-year old disturbance. It is important to note,

however, that the tracks created 26 years ago were created using a plow harrow, which affects

the sediment in a different way than the EBS. The general disturbance impact is comparable,

but variations due to the gear used are probable.

Figure 2: (A) Example of seafloor at a reference site, (B) example of an EBS track, (C) example

of a 26-year old plow track, indicating the four microhabitats outside track, track valley, ripple,

and white patch. Pictures copyright ROV KIEL 6000 Team, GEOMAR Helmholtz Centre for

Ocean Research Kiel, Germany.

2.2. Sediment and pore water sampling

Sediment was collected using MUC and ROV push cores (ROV-PUC). TV-guided MUCs

allowed for exact sampling of the tracks, while the precision with the ROV was even higher and

microhabitats within the plow tracks could be sampled. The cores were immediately brought

into the 4°C cold room of RV SONNE and for trace element and DOC analyses, sliced into

2 cm layers in a glove bag filled with argon. The sediment slices were transferred into 50 mL

acid-cleaned centrifuge tubes in the glove bag and centrifuged at 3,200 rpm for 40 min. The


74

supernatant was filtered through 0.2 µm cellulose acetate syringe filters, pre-cleaned with

0.1 M hydrochloric (HCl) acid and deionized water, again using a glove box. The water

overlying the particulate fraction within the MUC liner was sampled as well to get bottom water

values for each core. The pore water samples were acidified to pH 2 with concentrated,

suprapure HCl and stored at 4°C. Pore water samples for amino acid analyses were taken with

rhizons according to the procedure described by Seeberg-Elverfeldt et al. (2005) and frozen.

Pore water samples for nitrate analyses were extracted with a low pressure (argon at 3–5 bar)

squeezer using 0.2 µm cellulose acetate filters. An overview of the cores taken at different

locations in the working area is given in Table 1.

Table 1: Overview of cores taken for sediment and pore water trace metal analyses.

Sample ID Area Location Water

depth (m)

SO242/1_34MUC Reference South 7◦07.5244′S 88◦27.031′W 4162

SO242/1_56MUC DEA West plow track 7◦04.414′S 88◦27.760′W 4149

SO242/1_80MUC Reference West 7◦04.542′S 88◦31.581′W 4130

SO242/1_108MUC DEA East plow track 7◦04.483′S 88◦26.919′W 4169

SO242/2_163ROV-PUC38 DEA East ripple 7◦04.493′S 88◦26.933′W 4143

SO242/2_166ROV-PUC63 DEA East valley 7◦04.478′S 88◦26.918′W 4143

SO242/2_166ROV-PUC64 DEA East outside track 7◦04.459′S 88◦26.924′W 4143

SO242/2_169ROV-PUC10 DEA East white patch 7◦04.481′S 88◦26.913′W 4144

SO242/2_202ROV-PUC63 DEA EBS track 7◦04.953′S 88◦28.198′W 4150

SO242/2_211ROV-PUC57 DEA EBS outside track 7◦04.967′S 88◦28.193′W 4150


All acids used were of suprapure quality (HCl and HF by Merck, HClO4 and HNO3 by Roth). All

PE containers were acid cleaned prior to use to avoid any trace element contamination.

2.3.1. Solid phase

2.3.1.1. Major and trace elements

For bulk chemical analyses, centrifuged sediment samples were crushed and dried at 105◦C

to remove moisture. 100 mg of sediment were then digested with a PicoTrace DAS acid

digestion system using 3 mL perchloric acid (HClO4 70%) and 3 mL hydrofluoric acid (HF

38-40%) at 220◦C for 12 h. Samples were evaporated, taken up in 5 mL HCl and evaporated

again. This step was repeated before the samples were taken up in a mix of 0.5 M nitric (HNO3)

acid and 0.5% HCl (v/v). Samples were analyzed with ICP-OES (SpectroCiros SOP


75

instrument) for major elements and ICP-MS (Perkin Elmer Nexion 350x) for trace elements.

For ICP-OES measurements, the certified reference material (CRM) MESS-3 was within 5%

accuracy of certified values for Cu, K, Mg, and Mn, and within 10% for Ca, Fe, P, and Zn.

Accuracy for Al was −13% but too low values for Al in MESS3 have been reported before (Roje,

2010). Method precision was within 8% for all elements except P (10%), Mg (13%), and Al

(21%). Accuracy of BHVO-2 reference material was within 5% for all elements except P and

Zn (12%) and method precision was within 4% for all elements except P (13%). Accuracy for

MESS-3 and BHVO-2 for ICP-MS measurements were within 3% except for Pb (−8% MESS-

3) and Ni (6% BHVO-2). Method precision was within 3% except for Pb (26%). Cd could not

be measured reliably in the reference material due to high discrepancies of the two measured

isotopes. For detailed information on limit of quantification1 (LOQ), accuracy, and method

precision refer to Supplementary Material 1.

2.3.1.2. POC and PON

Particulate organic carbon (POC) and particulate organic nitrogen (PON) of the sediment were

determined through gas-chromatography of CO2 and N2, produced by flash combustion using

a Carlo Erba Element Analyzer (NA 1500). Samples were treated with HCl to release carbon

bound to carbonates prior to analysis.

2.3.1.3. Porosity

Porosity was calculated from the weight difference of wet and freeze-dried sediment. For

further details see Haeckel et al. (2001).

2.3.2. Pore water

2.3.2.1. Major and trace elements

Pore water major elements were measured with ICP-OES (SpectroCiros SOP instrument).

Overall accuracy for IAPSO seawater reference material was within 5% for all measured

elements except Mg (+11%). Method precision was 2–3% except for Si (17%). For trace

elements (As, Cd, Co, Cu, Mn, Mo, U, V), a ICP-MS (Perkin Elmer Nexion 350x) coupled with

an apex Q (ESI) introduction system to increase sensitivity and decrease background was

used. As, Co, Cu, Mn, and V were measured in a reaction cell in collision cell mode (KED

mode) with helium gas, to eliminate interferences. Cd and Co values of all CRMs are below

the LOQ and could not be verified. Mo and U were verified with IAPSO and NASS-6 seawater

reference material, V, Mn, Cu, and As were checked in NASS-6, SLEW-3, and SLRS-6.

Accuracy and method precision vary for each CRM, possibly due to varying salt contents but

1 10* standard deviation of acid blanks for each run.


76

generally agree with the reference materials. Only Mo values are slightly too high. For detailed

information on LOQ, accuracy, and method precision refer to Supplementary Material 1. Ni,

Zn, Cr, and Pb concentrations in the pore water could not be reliably quantified because of

poor accuracy (Ni and Zn) or too low concentrations (Pb LOQ = 0.01–2.29 µg/kg; Cr LOQ =

0.32–1.42 µg/kg).

For ICP-OES measurements, 10 ppm Y were used as internal standard and for ICP-MS, a

mixed internal standard containing Ru, Re, and Bi was used: 2 ppb without APEX and in KED

mode and 1 ppb with APEX.

2.3.2.2. DOC

DOC concentrations [mg/L] were determined via a high temperature combustion method (TOC-

VCSH Analyzer, Shimadzu). Inorganic carbon was removed by 2 M HCl prior to injection into the

combustion tube where organic carbon is oxidized to CO2 at 680◦C with a platinum catalyst. A

5-point calibration from 0.5 to 5 mg DOC/L was used. The error of measurement is less than

2% (for further analytical details see Brockmeyer and Spitzy (2013).

2.3.2.3. Amino acids

Total hydrolysable dissolved amino acids (DAA) and hexosamines (HA) of selected samples

were analyzed with a Biochrom 30 Amino Acid Analyzer after hydrolysis of ca. 3 ml of filtrate

with 6 N HCl for 22 h at 110◦C under a pure argon atmosphere. A particle free aliquot was

evaporated three times to dryness in order to remove the unreacted HCl; the residue was taken

up in an acidic buffer (pH 2.2). After injection and subsequent separation with a cation

exchange resin, the individual AA monomers were post-column derivatized with

o-phthaldialdehyde in the presence of 2-mercaptoethanol and detected with a Merck Hitachi

L-2480 fluorescence detector. Duplicate analysis of a standard solution according to this

method results in a relative error of 0.1 to 1.3% for the concentrations of individual AA

monomers and 0.2 to 3.0% for individual AA monomers of water samples. Due to acid

hydrolysis, aspartic acid and asparagine are both measured as Asp and glutamic acid and

glutamine are both measured as Glu.

2.3.2.4. Oxygen

Oxygen was measured ex-situ using fiber-optic microsensors (FireStingO2 optodes from

Pyroscience GmbH, Aachen, Germany) which were lowered into the MUC sediment with a

motorized micromanipulator (MU1, Pyroscience GmbH, Aachen, Germany). Measurements

were taken in 500 µm steps, with two optodes at the same time. In total, four to six


77

concentration profiles were completed for each MUC core. Method precision was 1% and the

detection limit 1 µmol/L. For further details see Haeckel et al. in Greinert (2015).

2.3.2.5. Nitrate

Nitrate was analyzed on-board RV SONNE directly after sampling using a Hitachi UV/VIS

spectrophotometer. The analysis followed standard analytical procedures, measuring nitrate

as sulphanile-naphthylamide after reduction with Cd (Grasshoff et al., 1999). The detection

limit was 2 µmol/L and analytical precision was 3%.

Detailed tables with results for trace elements, DOC and DAA can be found online at

PANGAEA: https://doi.pangaea.de/10. 1594/PANGAEA.880596.

3. Results

3.1. Solid phase

Two reference sites outside the DEA and two sites inside the DEA—next to an old plow mark

(DEA East) and the freshly disturbed EBS track—not directly impacted by plowing, were

analyzed for a range of background values and to determine natural spatial variability. Based

on the natural background conditions we will compare the 26-year old and 5-week old disturbed

sites.

The major elements in the sediments of the undisturbed and disturbed sites are Al, Fe, and Ca

(Figures 3–5). The latter increases with depth, usually more strongly below 10 cm or displays

pronounced peaks between 15 and 20 cm. The major element concentrations in general, as

well as the Ca variability, are comparable in the disturbed sites and do not vary substantially

from the undisturbed sites (Figure 3 and Supplementary Material 2). Si could not be measured

due to the acid digestion procedure using HF. Porosity decreases with depth from

approximately 0.93 at the surface to 0.86 at around 15 cm and is slightly higher in the surface

sediments of undisturbed sites compared to disturbed sites. Especially the EBS track has a

lower porosity at the surface. The slope of decreasing porosity is steeper in the disturbed sites

than in the undisturbed sites (Figure 3).




78

Figure 3: Sediment major element profiles and properties of the four undisturbed and six

disturbed sites.

The surface sediments of the undisturbed sites in the DEA have lower Mn concentrations than

those of the reference sites but display the same curved profile shape as the reference sites

(Figure 4). Additionally, the Mn-oxide rich layer is thicker in the reference sites. POC content

is higher within the DEA and lower in Reference South and West. Mo, Ni, Co, Cu, and Cd have

similarly curved patterns as Mn. We were only able to get reliable results for Cd in the

undisturbed site next to the EBS track. Similar to Mn, the above mentioned Mn-associated

metals show slightly higher concentrations at the reference sites compared to the undisturbed

sites within the DEA, outside DEA East plow track being the undisturbed site with overall lowest

concentrations.


79

Figure 4: Sediment element profiles of the four undisturbed sites. Reliable Cd results only for

outside EBS track.

All disturbed cores have Mn solid phase concentrations below 1.5 wt.%, with DEA West plow

track having the highest concentration (up to 1.26 wt%) and the EBS track having by far the

lowest concentration (<0.5 wt%) (Figure 5). Mo, Ni, Co, and Cu have similar concentrations at

the disturbed sites as at undisturbed sites, only the white patch and the EBS track have

considerably lower concentrations of Mo, Ni, and Co. These concentrations are in the same

range as the concentrations at depth below the Mn-oxide rich layer in the other cores. The

profiles of Mn and associated metals do not show the typical curves as it was the case for the

undisturbed sites (Figure 4). They are rather straight (DEA East ripple and valley) or have

peaks at depth (DEA East plow track at 23 cm).


80

Figure 5: Sediment element profiles of five 26-year old disturbed sites and the 5-week old EBS

track. Reliable Cd results only for DEA West plow track.

3.2. Pore water

Pore water ex-situ oxygen profiles (Figure 6) show that the oxygen penetration depth is

between 12 and more than 20 cm, decreasing from west to east. It is lower in the DEA plow

tracks but this is partly due to the natural gradient. Nevertheless, oxygen profiles from plow

track cores are more linear than in undisturbed sites. Oxygen measurements from Vonnahme

et al. (in prep.) show that oxygen penetration in the microhabitats is between 11 and 14 cm

deep. Nitrate is relatively stable throughout the upper 25 cm of sediment with concentrations

between 40 and 60 µmol/L. Below, concentrations slightly decrease as visible in the DEA East

plow track profile (Figure 6).


81

Figure 6: Bottom water and pore water ex-situ oxygen and nitrate profiles of two undisturbed

and two disturbed sites (MUCs). In each core, four to six oxygen profiles were measured.

The measured pore water major element concentrations are in the same range for each

element across all 10 sites—undisturbed and disturbed—and are mostly in the same range as

bottom water concentrations (Supplementary Material 3). Only Si shows the typical increase

with depth, which is steepest in the EBS track. Trace element concentrations in the pore waters

are in the same range for each element across all undisturbed sites as well (Figure 7). Mn, Mo,

U, and As concentrations are generally in the range of bottom water values. Co, Cu, V, Cd,

and DOC pore water concentrations are usually twice as high as bottom water concentrations,

at least in the upper centimeters. Co is rarely above the LOQ (0.08–0.22 µg/kg) in pore waters

of the undisturbed sites. Based on selected best data, we assume the background

concentration to be approximately 0.5 nmol/L. Overall, bottom water and pore water trace

element concentrations and profiles at the 26-year old plow tracks (Figure 8) are similar to

those at undisturbed sites. The surface layer DOC peaks are less pronounced in the disturbed

cores but some cores have peaks at greater depths.


82

Figure 7: Bottom water and pore water element profiles of the undisturbed sites. The

uppermost values refer to bottom water concentrations measured in the supernatant retrieved

above the sediment surface in the MUC liner.


83

Figure 8: Bottom and pore water element profiles of five 26-year old disturbed sites and the

5-week old EBS track. The uppermost values refer to bottom water concentrations measured

in the supernatant retrieved above the sediment surface in the MUC liner. Mn and Co below

the LOQ for DEA West plow track.

In some cores—undisturbed and disturbed—local peaks in certain elements occur which could

be sampling artifacts from filtration or nanoparticles. Another explanation would be local redox

signals, where Mn and associated metals are released into the pore water while Mo and U get

removed due to a reducing environment. Diagenetically, however, these peaks cannot be

sustained long. These peaks were already found in prior pore water studies in the Peru Basin

(see Koschinsky, 2001). Here, they will not be considered further.


84

Pore water DAA usually have their concentration maxima in the upper 10 cm, best visible for

Reference South (Figure 9). At Reference South, the DAA peak roughly coincides with the

DOC peak. The pore water DAA concentrations in the 26-year old plow tracks are generally in

the same range as in the undisturbed cores with DAA concentrations between 2.2 and

11.1 µmol/L. Peaks in the upper 10 cm are not pronounced (Figure 9). Especially in the DEA

East plow track the pore water DAA spectra differ from the undisturbed samples with relatively

higher contents of nonprotein amino acids β-Ala, γ-Aba as well as Lys, Val, and Met

(Supplementary Material 4).

Figure 9: Sum of dissolved amino acid (DAA) concentration profiles of three undisturbed and

two disturbed sites. The uppermost values refer to bottom water concentrations measured in

the supernatant retrieved above the sediment surface in the MUC liner.

4. Discussion

4.1. Solid phase: the Mn-oxide rich layer in the undisturbed sites

Mn-oxides are the main host phase for heavy metals in the upper 20 cm in the DISCOL area

(Koschinsky et al., 2001a; Marchig et al., 2001). The solid phase Mn content decreases steeply

around the oxygen penetration depth due to Mn-oxide utilization in organic matter degradation

(Figure 4). Oxygen is utilized first because it is energetically the most favorable pathway

(Froelich et al., 1979). Once oxygen is consumed, NO3− and MnO2 function as electron

acceptors. The processes can run in parallel, even though nitrate is the energetically favorable

option. Additionally, it has been suggested that the Mn and N cycles are linked and that MnO2

can provide O2 to oxidize N, leading to nitrate formation and Mn-oxide reduction (Mogollón et


85

al., 2016). Since the MnO2 concentration is declining while NO3− is still present in the pore water

(Figure 6), this might be a relevant process here. The low Mn content (∼0.5 wt%) below the

Mn reduction zone is likely bound in detrital minerals or Mn-carbonates and represents the

constant level of solid phase Mn for sediment below the Mn oxide rich layer (Gingele and

Kasten, 1994; Koschinsky, 2001). The sediments change color from dark brown in the oxic

zone to light brown in the suboxic zone, the tan color is due to Fe(III) in the clay minerals (König

et al., 1997).

The natural variability of Mn in the oxic layer is 1.1 to 1.7 wt%. There is an increase in Mn

content with an increase in oxygen penetration depth, from east to west. Additionally, the

reference sites have higher Mn concentrations than the undisturbed sites within the DEA. It

remains unclear, if solid phase element concentrations of the undisturbed sites within the DEA

are lower solely due to natural variability or because they were impacted by the disturbance as

well. Since the undisturbed sites within the DEA are adjacent to plow tracks, they have likely

been impacted by resettling suspended sediment from the plowing. Mo, Ni, Co, and Cu are

associated with Mn-oxides which is well known from other studies (Klinkhammer et al., 1982;

Heggie and Lewis, 1984; Shaw et al., 1990; Koschinsky, 2001; Morford et al., 2005). Mo and

Ni show a similar increase from east to west and from the DEA to the reference sites. Co and

Cu do not show such variability, though; the concentrations are in the same range for the four

undisturbed sites. Correlation coefficients show that of the four metals Cu is least associated

with Mn (Table 2). Shaw et al. (1990) only name Mo, Ni, and Co as being associated with Mn-

oxides. Cu seems to be neither controlled by the Mn-oxide phase nor by the Fe-oxyhydroxide

phase, as indicated by only weak correlation with Mn and Fe (Table 2). Since Cu is generally

known to show a high affinity to organic matter, we assume that binding to organic functional

groups may play a role in controlling Cu distribution in the surface sediment.


86

Table 2: Correlation coefficients of Mn and Fe with Cu, Co, Ni, and Mo, calculated in Excel.

Mn Mn Mn Mn Mn Mn Mn Mn Mn Mn

Mn 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

Cu 0.88 0.96 0.90 0.78 0.84 −0.18 −0.51 0.71 0.01 −0.12

Co 0.94 0.87 0.97 0.95 0.97 0.92 0.84 0.98 0.92 0.71

Ni 0.98 0.96 0.99 0.99 0.97 0.94 0.78 0.94 0.93 0.78

Mo 0.99 0.98 0.99 0.98 0.98 0.90 0.98 1.00 0.97 0.97

Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe

Fe 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

Mn 0.76 0.26 0.57 0.44 0.85 0.88 0.88 0.94 0.00 0.73

Cu 0.55 0.23 0.39 −0.18 0.99 −0.11 −0.57 0.58 −0.03 0.46

Co 0.88 0.23 0.67 0.61 0.84 0.78 0.64 0.94 0.21 0.21

Ni 0.84 0.43 0.63 0.44 0.94 0.78 0.48 0.92 −0.19 0.73

Mo 0.75 0.21 0.57 0.32 0.83 0.81 0.88 0.96 0.15 0.83

Reference

South

Reference

West

Outside

DEA East

plow track

Outside

EBS

track

DEA

West

plow

track

DEA

East

plow

track

DEA

East

ripple

DEA

East

valley

DEA

East

white

patch

EBS

track

The entire profiles were correlated.

4.2. Disturbance impacts on the solid phase: sediment removal,

redeposition, and inversion

Disturbed sediments have lower solid phase Mn concentrations than the undisturbed

sediments in the upper 15 cm, suggesting that the top Mn-oxide rich layer has been removed

or mixed. This is explicit in the 1 month old disturbance—the EBS track— but also still visible

in the profiles of the 26-year old plow tracks (compare Mn in Figures 4, 5). Even though the

highest concentrations of the disturbed sites (DEA West plow track) are in the range of the Mn

concentrations of the undisturbed sites within the DEA, a comparison of the overall ranges

shows that the disturbed sites on average show lower Mn concentrations. Comparing average

Mn concentrations of undisturbed and disturbed sites in the individually sampled layers, the

26-year old disturbed sites have 20 to 47% less Mn in the upper 16 cm. Similarly, there is 17

to 48% less Mo in disturbed surface sediments, while Co, Ni and Cu contents are successively

less impacted: Co 7 to 25%, Ni 5 to 18% and only Cu −7 to 7%.


87

The plowing removed most of the Mn-oxide rich layer in the plow tracks and disturbed

microhabitats. Only a thin layer is left, the thickness varying between the disturbed sites but

being clearly reduced compared to the undisturbed sites. Specifically the white patch does not

have a Mn-oxide rich layer left at the sediment surface. The same is true for the EBS track and

the tan sediment layer that is usually beneath the dark brown Mn-oxide layer lies at the surface.

Since the EBS track sediment is now exposed to the bottom water, trace metals diffuse from

the suboxic pore water until a sufficiently thick Mn-oxide layer has formed that scavenges the

trace metals. The higher the Mn-oxide content in the surface layer, the lower the diffusive flux

of heavy metals into the bottom water and the higher the sorption capacity (Fritsche et al.,

2001). Mn-oxides should form with time and the Mn, Mo, Ni, Co, and Cu concentrations in the

particulate fraction should slowly increase to establish the typical layering but the time scale of

these processes is unknown. The particulate fraction has not yet recovered in these parts but

at DEA West and East plow tracks the Mn-oxide rich layer is building up again and it is thicker

than in the microhabitats. In conclusion, both, Mn layer thickness and Mn content, are lower in

the disturbed sites.

In addition to sediment removal, suspended sediment was deposited on the plow tracks. The

sedimentation of suspended sediment was most measurable in the track valley as indicated

by increased concentrations of Mn, Mo, Ni, Co, and Fe in the DEA East valley in the upper 2

to 4 cm (Figure 5). The sediment seems to be composed of different material than the sediment

further downcore, possibly resettled particles from the Mn-oxide rich layer that were suspended

during the plowing. Moreover, the porosity of the valley’s surface layer is higher compared to

other disturbed sites, proving that loose material was deposited on top. Both impacts have also

been found in other biogeochemical investigations (Vonnahme et al., in prep.).

A third impact is sediment inversion, where Mn-oxide rich surface sediment got plowed to

greater depth. This is visible in the white patch and the DEA East plow track profiles. At the

white patch, the porosity also increases again below this depth, supporting the assumption that

surface sediment got turned. At both sites, Mn, Mo, Ni, Co, and Cu show elevated

concentrations at 11 and 23 cm, respectively. In other areas of the Peru Basin, a Mn peak at

the redox boundary was discovered (Koschinsky, 2001). Internal redox cycling of Mn around

the redox boundary can lead to such pronounced solid phase Mn peaks (Burdige, 1993). This

is comparable to the marked Mn peak at DEA East plow track (Figure 5) and could be an

explanation for the peak because at 25 cm depth the dissolved Mn concentration increases

drastically, a clear sign that Mn-oxides are reduced. Since the upper part of the core shows

disturbance impacts, we assume that the peak at 23 cm is another sign of the disturbance.


88

4.3. Pore water natural state and impacts visible 5 weeks post-

disturbance

The degradation of organic matter during early diagenesis at the sediment-water interface

releases various elements to the pore water—e.g., V, Cu, Mo, and DOC (Sawlan and Murray,

1983; Heggie et al., 1986; Shaw et al., 1990; Koschinsky, 2001; Kowalski et al., 2009). In this

study, we clearly see this release in form of a marked peak in the top 2 cm of the concentration

profiles for V, Cu, and DOC, and to a lesser extent for As (Figures 7, 8). Extensive V peaks at

the sediment-water interface can be sustained due to complexation by DOC (Emerson and

Huested, 1991; Morford et al., 2005) and because it is not significantly adsorbed to Mn-oxides

(Koschinsky, 2001). There are no major differences in the pore water profiles of undisturbed

and 26-year post-disturbance sites; the pore water is recovered. Only at DEA West plow track

the usual trace metal peak in the top 2 cm is less pronounced, which is interesting considering

that DEA West plow track has the least impacted solid phase. The typical peaks of V, Cu, and

As in the top 2 cm are not clearly developed yet in the EBS track profile (Figure 8; teal filled

diamonds). Even though this feature is already visible in the V profile, the extent of the peak is

less than half the concentration of that in the undisturbed and 26-year old disturbed sites from

DEA East. This could also be a sign of a lower microbial activity and hence lower rates of

organic matter degradation, which usually releases metals to the pore water at the sediment-

water interface (Sawlan and Murray, 1983; Heggie et al., 1986; Shaw et al., 1990).

Typically, pore water concentrations of Mn and some associated metals (e.g., Mo, Ni, Co, Cu)

increase in the Mn reduction zone as the solid phase concentrations decrease (Froelich et al.,

1979; Heggie and Lewis, 1984; Koschinsky, 2001; Morford et al., 2005), approximately below

20 cm in the Peru Basin. This phenomenon is rarely visible in our data except for Cu (Figures

7, 8), as the cores are usually too short to cover the entire Mn reduction zone or the Mn and

Co pore water concentrations are below the detection limit when they first increase. The

increase is only visible at DEA East plow track (Figure 8) because it is a 35 cm long MUC and

we were not able to retrieve those sediment depths with any other core. The Mn and Co

concentration increase is natural since it occurs below the redox boundary where such

increase is expected, even though the profile is from a disturbed site. Samples taken at the

DISCOL area in 1996 also show increasing pore water Mn below 25 cm (Koschinsky, 2001).

Correlating with the Mn and Co release, DOC, Mo, and V concentrations increase at the depth

of Mn release in the DEA East plow track core. The EBS example indicates, however, that 5

weeks post-disturbance the pore water shows signals of the impact and Mn and Co are already

detectable in the pore water at ca. 8 cm depth. This is markedly closer to the sediment-water

interface than for the undisturbed and 26year old disturbed sites. Also, Mo is more variable

below 8 cm compared to the largely conservative profiles in the undisturbed sites. Therefore,


89

this difference can be clearly attributed to the disturbance. The sediment profiles show that the

entire Mn-oxide rich layer was removed with the EBS so that suboxic pore water with dissolved

Mn and Co must have been at the sediment-water interface. After 5 weeks, Mn and Co have

already been removed from the pore water due to diffusion of oxygen into the sediment and

concentrations in the top 8 cm are within the natural background range. According to Einstein-

Smoluchowski calculations the diffusional length for 5 weeks is 7–8 cm. König et al. (2001)

predicted diffusion of oxygen into the sediment after a disturbance at short time-scales; the

establishment of the original redox zonation might, however, well take a few centuries. The

natural redox zonation has, however, not been established yet and the pore water, as well as

the particulate fraction, is in the process of approaching a new equilibrium.

4.4. Trace metal fluxes to the ocean

The increased concentration of V and Cu in the surface pore water suggests diffusion to the

bottom water to some degree (Table 3) (also see Koschinsky, 2001). Fluxes of the trace metals

which have concentrations in the range of bottom water (Mo) are negligible or metals diffuse

from the bottom water into the pore water (Mn) (Table 3). Trace metal input would, however,

be considerably enhanced when metals would diffuse from the suboxic pore water after

removal of the Mn-oxide rich layer due to deep-sea mining (Table 3). Nevertheless, metals do

not reach concentrations potentially toxic to animals, e.g., Cu release 0.3 µmol∗m−2∗month−1

compared to lowest LC50 values of 8.85 µmol/L (Mevenkamp et al., 2017). The trace metal

concentrations and fluxes in the upper cm are already reduced after 5 weeks, as the data from

the EBS track shows (Figure 8). The diffusion of oxygen into the sediment leads to oxidation

of the dissolved metal ions and removal from the pore water. Similarly, the metal ions released

into the bottom water will be quickly scavenged by particles in the oxic bottom water and are

not expected to greatly impact the trace metal budget of the ocean. Therefore, the numbers for

diffusive fluxes after the disturbance presented in Table 3 are a worst-case scenario and will

probably be lower, even within the first month after the disturbance, because they decline every

day. Further non-steady state modeling would be needed to portray a realistic post-disturbance

scenario.


90

Table 3: Diffusive fluxes of selected dissolved metal ions across the sediment-water interface

and potential fluxes across the sediment-water interface when the Mn oxide rich layer is

removed; based on gradients across the redox-boundary in cores from this study.

Diffusive flux across the

sediment-water interface

Potential diffusive flux across

the sediment-water interface

when oxic layer removed

[µmol*m-2*month*-1] [µmol*m-2*month*-1]

Site Mn Cu V Mo Mn Cu Co

Reference South 3 −4 −15 −3

Reference West −1 −5 −20 −1

Outside DEA East plow track −5 −13 −1

Outside EBS track 1 −9 −28 −3

Average undisturbed 1 −6 −19 −2

DEA West plow track −2 −7 −3

DEA East plow track −6 −14 −2 −79 −0.3 −0.07

DEA East ripple 1 −4 −15 1

DEA East valley 1 −6 −11 −2

DEA East white patch 5 5 −4 −1

Average disturbed 26

years old

2 −3 −10 −1

EBS track −2 −2 −4 3 −21 −0.2 −0.03

Negative fluxes indicate diffusion from the pore water to the bottom water and positive fluxes indicate diffusion

from bottom water to the pore water. The assumptions are that the system is steady state and diffusion the

only process. F=−øDsed average porosity of the sediment in each core for the used depth; Dsed ; DSW

Mn = 3.02E-10 m2s−1; DSW Cu = 3.22E-10 m2s−1; DSW Co = 3.15E-10 m2s−1; DSW V = 5E-10 m2s−1; DSW Mo =

5E-10 m2s−1); DSW at temperature 0°C from Schulz (2006) after Boudreau (1997), closest diffusion coefficient

values considering deep-sea temperatures of 1.85°C (Brown et al., 2017b), except for V and Mo, where only

general diffusion coefficients published in Emerson and Huested (1991) and Scholz et al. (2011) were used.

For a more detailed table including the pore water and bottom water metal concentrations see Supplementary

Material 5.


91

4.5. DOC and DAA as indicators of organic matter degradation

DOC and DAA can be intermediates of sedimentary organic matter degradation. DOC and

DAA concentrations are elevated in pore waters compared to bottom water so that the

sediments are sources of DOC and DAA to the water column (Lahajnar et al., 2005; Burdige

and Komada, 2015). In the pore water, DOC and especially the more reactive DAA, may be

further degraded to inorganic nutrients or reintegrated into the sediment by bacterial uptake or

sorption processes (Burdige and Martens, 1990; Ding and Henrichs, 2002). DAA

concentrations in nearshore pore water are elevated in the upper 25 cm and drop to values of

2-5 µmol/L at depth (Burdige and Martens, 1990), which also fits well with our deep-sea pore

water data (Figure 9). DAA bottom water and pore water spectra are dominated by

Ser>Gly>Ala>His>Orn>Asp (Supplementary Material 4), irrespective of disturbance and redox-

zonation, which is quite different from sediment and suspended matter spectra that are

dominated by Gly, Asp, Glu, and Ala or, respectively, by Gly, Glu, Asp, and Ser (Gaye et al.,

2013) but similar to DAA spectra from the water column (Ittekkot and Degens, 1984). Pore

waters tend to accumulate those amino acids which are preferably removed from the

particulate phase, including Ser, Gly, and Glu (Seifert et al., 1990) as well as degradable amino

acids (e.g., Met) and basic amino acids (e.g., Lys) preferentially sorbed to mineral surfaces

(Ding and Henrichs, 2002). The latter indicate the degradation or desorption of amino acids of

the particulate pool. In addition, pore waters also accumulate the non-protein amino acids

β-Ala, γ-Aba, and Orn which are either degradation products of proteinaceous amino acids or

are not taken up by bacteria ( Seifert et al., 1990; Davis et al., 2009). It has been shown

experimentally with cores from the Peru Basin that a few hours after a disturbance, particulate

AA concentrations in the sediment and DOC in pore waters sharply increased. The increase in

particulate AA was attributed to enhanced bacterial activity which could be related to spreading

of fresh organic matter from deeper layers (Koschinsky et al., 2001b) and augmented by the

oxygen availability in the upper sediments which may reinforce organic matter degradation

(Lee, 1992). After 26 years, however, the concentration differences between disturbed and

undisturbed sites are not so visible anymore so that degradation possibly slowed down due to

decreasing quality of organic matter (Vonnahme et al., in prep.). The high DOC concentration

in the DEA East plow track surface layer should therefore not be a remnant of the 1989

disturbance but rather due to a recent incident, such as a local input of organic material or

bioturbation.

5. Conclusion

The solid phase results of the undisturbed sites show natural variability with respect to element

concentrations, yet the profile shapes agree. The pore water profiles do not show major


92

differences between the undisturbed and the 26-year old disturbed sites. Five weeks post-

disturbance, the impacts were still visible in the pore water profiles but signs of regeneration in

the upper centimeters were already visible and an elevated metal flux to the ocean seems to

prevail on even shorter time scales. Differences in DOC and DAA concentrations as well as

spectra are not visible or cannot be attributed to the disturbance after 26 years. In general, the

re-establishment of a new steady-state in the solid phase takes longer than in the pore water.

The differences between undisturbed and 26-year old disturbed sites, especially the loss or

redistribution of Mn-oxide rich sediment, are clearly visible in the profiles. An important finding

of our study is that degrees and types of disturbance differ strongly among the disturbed sites.

The EBS track is quite distinct due to its recency but even the other five 26-year old disturbed

sites vary with respect to concentrations of the metals—especially Mn, Mo, Ni, and Co—as

well as profile shapes in the solid phase. As discussed above, these can be results of different

disturbance impacts such as removal, mixing, redeposition of suspended sediment, and

inversion or most often a unique combination of several impacts.

The geochemical variability which was discovered in the undisturbed as well as disturbed sites

elucidates that the deep-sea is a highly complex system that is still poorly understood as has

also been recently shown for the CCZ (Kuhn et al., 2017; Mewes et al., 2016, 2014; Mogollón

et al., 2016; Volz et al., in review). With respect to polymetallic nodule mining, it will be

necessary to carry out baseline studies on the geochemistry of the potentially impacted sites

and reference sites for quite a high number of locations to assess the heterogeneity of both,

the natural area and the types of impact. The difficulty of gaining representative baselines and

ranges of disturbance impacts is a general challenge for deep-sea mining related research and

has been discussed elsewhere, too (see for example Jones et al., 2017).

Metal concentrations in pore water are not suitable for monitoring purposes because their

concentrations quickly reach a new steady-state after a disturbance, probably on time scales

of months. Therefore, they could imply that the system has recovered which truly is not the

case for other components. In the DISCOL area, the disturbance impact was most pronounced

in the Mn-oxide rich top layer. Since it was shown in this study and previous work (Shaw et al.,

1990; Koschinsky, 2001; Morford et al., 2005) that many other metals – such as Mo, Ni, Co,

and Cu, are associated with Mn-oxides in this layer, Mn is a key parameter for monitoring, if

not all parameters can be measured due to financial, technical, and time constraints in an

industrial mining scenario. In addition to the oxygen penetration depth, knowing the Mn

concentration in the solid phase and pore water gives a lot of insights into the geochemical

system at this site, including potential release of other trace metals, and would be a useful

parameter to measure pre-mining and postmining for monitoring purposes. Mn is a good


93

indicator for disturbance in sediments with a relatively shallow oxic layer, such as the Peru

Basin. Areas with a different redox-zonation might have different key parameters because a

largely oxic system (such as the CCZ where deep-sea polymetallic nodule mining is most likely

to start) will be differently impacted (also see Cronan et al., 2010; Rühlemann et al., 2011;

Mewes et al., 2014, 2016; Mogollón et al., 2016; Kuhn et al., 2017; Volz et al., in review). We

are only able to draw this conclusion about Mn as a key parameter, however, because this

small site has been extensively studied over a long period of time. Baseline studies are vital to

(1) understand the system, (2) select key parameters, and (3) define thresholds. More

extensive research in different geochemical seafloor systems and on a larger scale needs to

be carried out before it can be determined what a negative impact on the environment may be

and which thresholds should therefore not be exceeded.

Author contributions

SP: research design, data collection, trace metal analyses and data interpretation, article

drafting and revision. MH, AK, and SK: research design. MH: sampling, oxygen, POC, PON,

porosity data collection, analyses and interpretation. BG: DOC and DAA analyses and data

interpretation. AK, MH, BG, and SK: article revision. SP, MH, BG, SK, and AK: final approval

of the version to be published.

Acknowledgments

We are deeply grateful to the crew of RV SONNE, the ROV KIEL 6000 team and the chief

scientists Jens Greinert and Antje Boetius on cruise SO242/ 1 and 2 who made the sampling

possible. Our great appreciation goes to Katja Schmidt, Annika Moje, Inken Preuss, Rajina

Bajracharya, Tim Jesper Suhrhoff, Seinab Bohsung and Laura Ulrich for their help with

laboratory work in the geochemistry laboratory at Jacobs University Bremen and sampling

onboard RV SONNE. We thank Peggy Bartsch for DOC analyses and Niko Lahajnar for amino

acid analyses carried out at the University of Hamburg as well as Meike Dibbern, Bettina

Domeyer, Anke Bleyer and Regina Surberg for analytical work onboard RV SONNE and at

GEOMAR. Thanks also go to Anne Peukert from GEOMAR, for providing the map. This work

was funded by the German Federal Ministry of Education and Research in the framework of

the JPI Oceans project MiningImpact (grant no. 03F0707A+G) and the Post-Grant-Fund (grant

no. 16PGF0058). We also thank two reviewers for their helpful comments that improved this

manuscript.


94

Supplementary material

The Supplementary Material for this article can be found online at:

https://www.frontiersin.org/articles/10.3389/fmars. 2018.00117/full#supplementary-material

https://www.frontiersin.org/articles/10.3389/fmars.2018.00117/full#supplementary-material

Chapter 4 – Ca phosphate control of REY in the central Pacific

95

Chapter 4 – Calcium phosphate control of REY

patterns of siliceous-ooze-rich deep-sea

sediments from the central equatorial Pacific

Title

Calcium phosphate control of REY patterns of siliceous-ooze-rich deep-sea sediments from

the central equatorial Pacific

Authors

Sophie A. L. Paul, Jessica B. Volz, Michael Bau, Male Köster, Sabine Kasten, Andrea

Koschinsky

Published in

Geochimica et Cosmochimica Acta (2019) 251:56-72

doi: 10.1016/j.gca.2019.02.019

Originally published by Elsevier


96


97

Calcium phosphate control of REY patterns of

siliceous-ooze-rich deep-sea sediments from


Sophie A. L. Paul1*, Jessica B. Volz2, Michael Bau1, Male Köster2, Sabine Kasten2,3, Andrea

Koschinsky1


2Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research, Bremerhaven,

Germany

3University of Bremen, Faculty of Geosciences, Bremen, Germany

*corresponding author (email: [email protected])

Abstract

Rare earth elements and yttrium (REY) are often used as proxies for (paleo)environmental

conditions and for the reconstruction of element sources and transport pathways. Many

geological systems are well described with respect to the behavior of REY but deep-sea

sediments with their manifold processes impacting the sediment during early diagenesis leave

some questions about the origin and development of the shale-normalized REY (REYSN)

patterns unanswered. Here we report REY data for sediment solid phase and pore water from

the upper 10 m of deep-sea sediments from the Clarion Clipperton Zone (CCZ) in the central

equatorial Pacific. The solid-phase REY profiles show highest concentrations at depth below

5–8 m. The REYSN patterns show an enrichment in middle REY (MREY) (LaSN/GdSN between

0.35 and 0.60; GdSN/YbSN between 1.19 and 1.47) and either no or negative CeSN and YSN

anomalies (i.e. chondritic to sub-chondritic Y/Ho ratios between 24.7 and 28.7). Based on

correlation analyses of bulk sediment element concentrations and sequential extractions, we

suggest that a Ca phosphate phase controls the distribution and the patterns of REY in these

silty clay pelagic sediments rich in siliceous ooze. The MREY enrichment develops at the

sediment-water interface and intensifies systematically with depth. The negative CeSN anomaly

mailto:[email protected]


98

intensifies with depth possibly because Ce is mostly bound to Mn- and Fe-(oxyhydr)oxides.

Therefore, Ce concentrations remain relatively constant throughout the sediment core, while

its trivalent REY neighbors are mostly hosted by the Ca phosphate phase that continuously

incorporates REY from ambient pore waters. The non-redox-sensitive trivalent REY

concentrations increase with depth, producing or enhancing a negative CeSN anomaly through

coupled substitution of REY3+ and Na+ for Ca2+. The solid-phase REYSN pattern is therefore

determined by the pore-water REYSN pattern and not suitable for paleoceanographic

interpretation. The similarity of the pore-water and solid-phase REYSN patterns suggests,

however, that only minor fractionation occurs during REY incorporation into the Ca phosphate

crystal structure.

Keywords: rare earth elements and yttrium, calcium phosphates, early diagenetic alteration,

coupled substitution, CCZ

1. Introduction

Rare earth elements (REE) encompass the lanthanide series from La to Lu in the periodic

table, 15 elements with atomic numbers 57–71. They are useful proxies for environmental

conditions and processes because of their similar physico-chemical properties and

geochemical behavior (e.g., Cantrell and Byrne, 1987; Elderfield and Greaves, 1982; Toyoda

et al., 1990). The REE are exclusively trivalent in the natural environment, with the exception

of Ce and Eu which also occur in the tetra- and divalent state, respectively. Yttrium, which is

exclusively trivalent and of almost the same ionic size as its geochemical twin Ho, is often

combined with the REE as the REY. The trivalent REE’s ionic radii decrease with increasing

atomic number which is referred to as the ‘‘lanthanide contraction” (Seitz et al., 2007 and

references therein). These minimal changes in ionic radii are sufficient to lead to fractionation

of the REY during incorporation into crystal lattices or during chemical complexation.

Depending on the ionic radii, either the light REY (LREY; La-Nd), middle REY (MREY; Sm-

Dy) or heavy REY (HREY; Ho-Lu) may be preferentially scavenged or mobilized, which then

leads to fractionation and, therefore, distinct signatures in shale-normalized (SN) REY patterns

(e.g., Cantrell and Byrne, 1987; Elderfield, 1988). From these patterns, past and present

environmental conditions, such as ancient seawater composition, the influence of

hydrothermal activities on sediments, and sediment provenance can be inferred (Bau and

Dulski, 1999; Bright et al., 2009; German et al., 1990; Kashiwabara et al., 2018; Viehmann et

al., 2015). Several studies have targeted the distribution of REY in Pacific pelagic sediments

(e.g., Elderfield et al., 1981; Glasby et al., 1987; Kon et al., 2014; Toyoda et al., 1990; Toyoda

and Masuda, 1991; Toyoda and Tokonami, 1990); mostly focusing on surface sediments and


99

interactions of surface sediments with manganese nodules. To our knowledge, less attention

has been paid so far to the impact of early diagenesis in deeper subsurface sediments on REY

(e.g., Soyol-Erdene and Huh, 2013; Zhang et al., 2016). Understanding the behavior of REY

during early diagenesis is important, however, when aiming to use them as proxies for

paleoenvironmental reconstructions (Bright et al., 2009; Trotter et al., 2016).

In modern seawater, dissolved HREY show an enrichment in the REYSN patterns (Fig. 1)

because the LREY are preferentially scavenged by particles in the water column (e.g., Chen

et al., 2015; Elderfield, 1988). Except for Ce, REY concentrations increase with water depth

(e.g., Alibo and Nozaki, 1999; Chen et al., 2015), because the REY are bound to particulate

matter in surface waters and released at greater depth when particles dissolve or are

microbially degraded. In contrast, dissolved Ce concentrations decrease with depth (e.g., Chen

et al., 2015) because of Ce3+ oxidation and Ce4+ fixation at sinking particles in oxic seawater.

Once Ce3+ is oxidized to Ce4+, it is usually less mobile and will participate less in exchange

reactions (Bau and Dulski, 1996a; Bau and Koschinsky, 2009; de Baar et al., 1985; Goldberg

et al., 1963; Piper, 1974b; Sholkovitz et al., 1993) unless the solution is rich in complexing

agents such as CO32– or siderophores (e.g., Bau et al., 2013; Kraemer et al., 2015; Möller and

Bau, 1993). This may lead to the development of a negative CeSN anomaly in the REYSN pattern

of the solution (e.g., seawater) and a positive CeSN anomaly in the solid material (e.g., marine

hydrogenetic Fe-Mn crusts and nodules; Fig. 1) (Elderfield and Greaves, 1982; Kasten et al.,

1998; Piper, 1974b). If, however, sinking Mn- and/or Fe-(oxyhydr)oxide particles with positive

CeSN anomaly cross a redox-cline and enter anoxic seawater, these particles re-dissolved and

produce anoxic seawater with a positive CeSN anomaly (Bau et al., 1997; de Baar et al., 1988).

Note however, that formation of the negative CeSN anomaly of oxic seawater is not confined to

the marine environment, but already develops during terrestrial weathering and characterizes

the truly dissolved REYSN patterns of river waters (Merschel et al., 2017; Pourret and Tuduri,

2017). Besides Ce, Eu can also show pronounced positive or negative anomalies in REYSN

patterns, as observed, for example, in high-temperature hydrothermal environments (Michard,

1989). In low-temperature systems such as pelagic sediments, however, Eu is present in the

trivalent oxidation state (Bau, 1991). After particle deposition, diagenesis can impact the

distribution of the REY in the sediment and pore water leading to deviations from the seawater

REYSN pattern (Abbott et al., 2016; Bright et al., 2009; Haley et al., 2004; Trotter et al., 2016).

The alteration of the pore-water REY pool during early diagenesis is also suggested by the

distinct Nd isotopic composition of pore waters, which differs from that of bottom water (Abbott

et al., 2016; Du et al., 2016). The dominant biogeochemical process during early diagenesis

in the upper 10 m of Clarion Clipperton Zone (CCZ) sediment was shown to be aerobic

respiration followed by denitrification and Mn oxide reduction (Mogollón et al., 2016; Volz et


100

al., 2018). For the sediments of the CCZ typical oxygen penetration depths of 1–4.5 m have

been reported (Mewes et al., 2014; Mogollón et al., 2016; Rühlemann et al., 2011; Volz et al.,

2018). Recent studies have also shown that in some areas within the CCZ the sediments

overlying the oceanic crust are fully oxic. This phenomenon was explained by sediment

alteration by upward diffusing oxygen from the basaltic crust in which low-temperature

circulation of seawater occurs, especially in areas close to seamounts and faults (Kuhn et al.,

2017; Mewes et al., 2016). This shows that deep oxygen penetration depths on meter scales

prevail in the study area. Previous studies of REY in sediments from the central Pacific

concluded that biogenic Ca phosphates such as apatite, are the main hosts of REY in Pacific

deep-sea sediments and that the biogenic phosphate is largely composed of fish debris, i.e.

bones and teeth (e.g.,Elderfield et al., 1981; Kon et al., 2014; Toyoda et al., 1990; Toyoda and

Masuda, 1991; Toyoda and Tokonami, 1990). Unaltered conodonts, defined by a low color

index (Bertram et al., 1992; Epstein et al., 1977), fish debris and marine phosphorites show

REYSN patterns similar to those of seawater, i.e. LaSN/YbSN <<1, a negative CeSN anomaly and

positive anomalies of LaSN, GdSN, and YSN, i.e. superchondritic Y/Ho ratios (marine phosphorite,

Fig. 1). Altered conodonts, defined by a high color index (Bertram et al., 1992; Epstein et al.,

1977), and fish debris show higher REY concentrations and MREY enrichment, deviating from

the seawater REYSN pattern (Fig. 1). Several authors suggested coupled substitution as the

process for REY uptake into apatite as Ca2+ in Ca phosphates is easily replaced through

coupled substitution by a REE3+ together with a monovalent element of similar size (e.g., Na+)

(Elderfield et al., 1981; Jarvis et al., 1994; Rønsbo, 1989).


101

Fig. 1. REYSN patterns of fish debris, fossil fish teeth, marine phosphorite, hydrogenetic Fe-

Mn crust, and seawater (PAAS from Taylor and McLennan, 1985, except Dy from McLennan,

1989). (See above-mentioned references for further information.)

In this contribution we aimed at elucidating the controls on the REY distribution in CCZ

sediments as well as the processes that might have caused the alteration of primary REYSN

patterns during oxic and suboxic early diagenesis.


102

2. Samples and methods

2.1. Geological setting of the study site

The CCZ is located in the central equatorial Pacific and bracketed by the Clarion Fracture Zone

to the North and the Clipperton Fracture Zone to the South. Both faults extend west-southwest

from the East Pacific Rise (EPR), where the Pacific crust forms and moves west from the

spreading center. The oceanic crust in the CCZ dates from the Late Cretaceous (~100 Ma) to

the Miocene (~5 Ma) (Barckhausen et al., 2013; Eittreim et al., 1992). Surface sediments are

dominated by siliceous ooze (mostly radiolarians and diatoms) and silty mud (Berger, 1974;

International Seabed Authority, 2010; Kuhn, 2015; Mewes et al., 2014; Müller et al., 1988; Volz

et al., 2018). The preservation of calcareous material in marine sediments depends on the

carbonate compensation depth (CCD) below which calcite dissolves. Today, the seafloor is

mostly below the CCD located at approx. 4500 m water depth (e.g., Bramlette, 1961 in Berger,

1970; Lyle, 2003) and the surface sediments show carbonate concentrations <1 wt.%

(Sharma, 2017). The particulate organic carbon (POC) flux is low and varies from

1 mg Corgm-2 d-1 in the northern part of the study area and 1.5 mg Corg m-2 d-1 in the central to

western part to 2 mg Corg m-2 d-1 in the south-eastern part (Lutz et al., 2007; Vanreusel et al.,

2016; Volz et al., 2018). Total organic carbon (TOC) contents in the surface sediments range

from 0.2-0.45 wt.% and under approx. 1 m depth, the TOC contents are below 0.2 wt.% (Volz

et al., 2018). Sedimentation rates are between 0.20 and 1.15 cm kyr-1 for the uppermost 50

cm of the sediment (Mewes et al., 2014; Mogollón et al., 2016; Volz et al., 2018).

Based on extrapolation of the sedimentation rate in the surface sediment (Volz et al., 2018),

the cores represent approx. 2.5–3 million years of depositional history. For core 194GC, we

assume relatively constant sedimentation rates of 0.2 cm kyr-1 (Volz et al., 2018) and no hiatus

is visible. Therefore, the age at the core bottom at ca. 580 cm could be around 2.9 Ma. The

high sedimentation rate of 1.15 cm kyr-1 for core 87GC might be due to sediment focusing or

the delivery of material from nearby seamounts (Volz et al., 2018). An extrapolation of the

sedimentation rate for the entire core length is therefore not appropriate. Beryllium isotope

dating for cores from the nearby BGR contractor area show that sedimentation rates were

about 0.3 cm kyr-1 until 2 Ma and significantly lower before (0.04 cm kyr-1) (Mewes et al., 2014).

The core bottom age could be between 2.6 and 3.6 Ma. Since core 165GC shows a hiatus at

790 cm, dating is even more complicated. Based on the sedimentation rate determined for

surface sediments (0.64 cm kyr-1, Volz et al., 2018) the sediment at 790 cm depth could be 1.2

Ma. Below the hiatus, lower TOC and higher Mn contents suggest the deposition of different

material, that was associated with the closure of the Panama Isthmus 2.5 Ma for sediment with


103

a similar composition below a hiatus at 420 cm in the BGR contractor area (Mewes et al.,

2014).

2.2. Sampling

The sediment samples were taken during RV SONNE cruise SO239 in 2015 in the CCZ as

part of the BMBF-EU Joint Programming Initiative Healthy and Productive Seas and Oceans

(JPI Oceans) pilot action ‘‘Ecological Aspects of Deep-Sea Mining (EcoMining/MiningImpact)”

(Fig. 2; Martínez Arbizu and Haeckel, 2015). A gravity corer (GC) was used for the retrieval of

sediment cores of up to 10 m length. A total of six GCs were selected for REY analyses of

which three cores with extensive carbonate or metalliferous (Fe, Mn) layers were excluded for

this study. The sediment cores 87GC and 165GC were retrieved in two European contract

areas for the exploration of manganese nodules, namely the area of the eastern European

consortium IOM (InterOceanMetal) and the French IFREMER area (Institut Français de

Recherche pour l’Exploitation de la Mer), respectively (Table 1). Furthermore, core 194GC

originates from the third of the nine Areas of Particular Environmental Interest (APEIs) located

around the contractor areas, which is excluded from any mining activities (Table 1). The solid

phase was sampled in 20–100 cm intervals for REY analyses, taking approx. 2 cm layers for

one subsample. An upward diffusion of oxygen from the basaltic crust as explained in the

introduction is expected to occur at sites 87GC and 165GC, indicated by decreasing dissolved

Mn2+ concentrations in the pore water in both cores at depth, suggesting oxidation of Mn2+ at

~20 m and 7 m, respectively (Volz et al., 2018).

Pore-water samples were obtained by filling sediment into acid pre-cleaned 50 mL centrifuge

vials and centrifuging at 2800 rpm for 40 min. Afterwards, the supernatant was filtered through

0.1 M suprapure hydrochloric acid (HCl) and deionized water pre-cleaned 0.2 µm cellulose

acetate syringe filters. The filter size was chosen because it is a common physical size limit for

the dissolved fraction. Samples were immediately acidified with 1 mL suprapure HCl (30%) per

1 mL sample.

In addition to REY analyses of bulk acid digestions described in Section 2.3, REY were

analyzed in six sequential extraction solutions corresponding to an upper and lower part in

each core and scanning electron microscope (SEM) analyses are presented for two samples

of core 165GC. Pore-water REY data could only reliably be reported for three samples of the

lower part of 194GC.


104

Fig. 2: Core sampling locations of 87GC and 165GC in the CCZ and of 194GC north of the

Clarion Fracture Zone. The map was created using GeoMapApp.

Table 1: Overview of GC sampling sites.

Sample ID Location Core length (cm) Contractor area Water depth (m)

87GC 11° 04.54’ N 119° 39.83’ W 942 IOM 4436

165GC 14° 02.63’ N 130° 08.39’ W 927 IFREMER 4922.7

194GC 18° 47.54’ N 128° 22.33’ W 576 APEI 3 4815.5

2.3. Analytical methods

Total acid digestions were performed at the Alfred Wegener Institute Helmholtz Centre for

Polar and Marine Research (AWI), Marine Geochemistry laboratory in Bremerhaven,

Germany. 50 mg of freeze-dried and homogenized sediments were digested in the MARS

Xpress (CEM) microwave system at 230°C after the procedure by Nöthen and Kasten (2011)

and Volz et al. (2018). A mixture of 3 mL sub-boiling distilled 65% nitric acid (HNO3), 2 mL sub-

boiling distilled 30% HCl, and 0.5 mL suprapur®_ 40% hydrofluoric acid (HF) was used for the

treatment. The digested solutions were evaporated and the solid residue subsequently re-

dissolved in 5 mL 1M HNO3 at 200°C. After full digestion, the residue was filled up to 50 mL

with 1 M HNO3.

Major element concentrations (Al, Ca, Fe, K, Mn, Na, P, S, Ti) were determined at AWI,

Bremerhaven, using an IRIS Intrepid ICP-OES, Thermo Elemental. The NIST-2702 was used

as certified reference material (CRM). Analytical accuracy was within 7% for all elements

(n=12) and the relative standard deviation (RSD) within 3% except for Na (9%) (for details see

Supplementary data 2, Table EA1). For the REY and Ba, ICP-MS measurements were carried


105

out at Jacobs University Bremen (JUB), using a Perkin Elmer Nexion 350x, for pore-water

analyses coupled to an Apex Q (ESI) for improved background and sensitivity. A multi-element

standard containing 2 ppb of all measured elements was used for external calibration and the

linearity was checked with a 4 ppb multi-element standard. A mixture of the elements Ru, Re,

and Bi was used as an internal standard to correct for sample matrix effects and machine drift.

Blanks were four to five orders of magnitude below REY sediment concentrations for LREY

and two to three orders of magnitude lower for HREY. Due to low REY concentrations in some

pore waters relative to blank concentrations (e.g., in several samples from shallow depths), it

was not possible to determine REY data for all porewater samples. Interferences from Ba and

some LREE oxides and hydroxides were corrected for by experimentally determined oxide and

hydroxide yields of molecular ions. NIST-2702 from the AWI digestions (n=10) and BHVO-2

from a different acid digestion performed at JUB (n=1) were used to check machine

performance. Accuracy was within 4% for both CRMs except for Y (6%) in BHVO-2 and Sm

(6%) and Nd (9%) in NIST-2702. The RSD was within 7% for NIST-2707 and 5% for BHVO-2

(see Supplementary data 2, Tables EA2 and EA3 for details; for more information see, for

example, Bau et al. (2018)). To the best of our knowledge no CRM REY data exist for pore

waters. Results for bulk sediments were corrected for the porewater salt matrix according to

Kuhn (2013) and Volz et al. (2018) to adjust for impairment of pore-water constituents on the

sediment composition. These correction factors range between 3 and 13% but the correction

does not affect the REYSN patterns.

2.4. Reporting

The CeSN anomaly was calculated after Bau and Dulski (1996a):

1. 𝐶𝑒

𝐶𝑒∗=

𝐶𝑒(𝑆𝑁)

(0.5∗𝐿𝑎(𝑆𝑁)+0.5∗Pr(𝑆𝑁))

REY concentrations in mg/kg were used for anomaly calculations of dry bulk sediment,

leachates, and pore water. For convenience, we refer to a negative CeSN anomaly if CeSN/CeSN*

is <0.9, to no anomaly if it is between 0.9 and 1.1, and to a positive anomaly if it is >1.1.

2.5. Sequential extraction

In order to differentiate between sedimentary Mn- and Fe-(oxyhydr)oxides as well as different

solid-phase Fe binding forms, an extraction protocol was developed combining the methods

presented by Koschinsky and Halbach (1995), Koschinsky et al. (2001a), and Poulton and

Canfield (2005) (Table 2; Köster, 2017). The sequential extractions were performed at room

temperature (ca. 22°C) at AWI, Bremerhaven. Fresh sediment portions equivalent to 100 mg

dry sediment were treated with 10 mL of the respective extraction reagent for specific reaction

times (Table 2). After centrifugation (4000 rpm, 5 min), the supernatant leaching solution was


106

filtered through a 0.2 µm polyethersulfone membrane for analysis while the residue was

subsequently treated with the next extraction reagent (Table 2). Solution carry-over was

minimized as much as possible during filtration but to avoid loss of sediment it could not be

prevented in all samples. The sediment was not washed with deionized water between

leaching steps. Comparison of bulk sediment data with the results of the combined leaching

steps resulted in up to 25% higher concentrations of major elements and REY for the added

sequential extraction steps, which might be due to error propagation of sequential extraction

measurements or due to contamination. Results above 100%, however, only occured for one

sample (87GC-827 cm). Iron, Mn, and P were measured at AWI, Bremerhaven and REY at

JUB as described in the analytical methods section above.

Table 2: Leaching scheme for the sequential extraction of Mn- and Fe-(oxyhydr)oxides

(adapted from Köster, 2017).

Step Associated mineral

phase Extraction reagent

Extraction

time [h] pH

Method

after

I

Pore-water and sorbed Fe,

Mn

Carbonate-associated Fe,

Mn

1 M Na-acetate 24 4.5 1

II Easily reducible Mn-oxides 0.1 M hydroxylamine-HCl 2 2 2,3

III Easily reducible Fe-oxides 1 M hydroxylamine-HCl 48 1

IV Reducible Fe.oxides Na-dithionite (50 g/L)/

0.2 M Na-citrate solution 2 4.8 1

V Magnetite 0.2 M NH4-oxalate/0.17 M

oxalic acid 6 1

1: Poulton and Canfield, 2005

2: Koschinsky and Halbach, 1995

3: Koschinsky et al., 2001a

2.6. Scanning electron microscopy

Untreated, fresh, carbon-coated sediment samples were analyzed by scanning electron

microscopy (SEM) combined with a QUANTAX energy-dispersive X-ray spectrometer (EDS,

Bruker) at AWI, Bremerhaven. The spectral imaging tool ESPRIT HyperMap was used to

create quantitative element maps (scan time: 25 min) including Ba, Ca, Fe, Mn, P, S, Sr, As,

Cr, Ni, Zn, Mo.


107

3. Results

3.1. Bulk solid-phase major elements

The major elements show a mostly conservative profile in the upper meters of all sediment

cores (Fig. 3). The detrital elements such as Al, K, and Ti decrease at depth, while Fe, Mn, Ba,

S, Ca, and P, concentrations increase: in 87GC below ca. 730 cm depth, in 165GC below ca.

790 cm depth and in 194GC below ca. 500 cm depth (Fig. 3, Supplementary data 3). At these

depths, the sediments also display a color change from tan to dark brown (Fig. 3). The

northernmost core, 194GC, shows overall higher concentrations of Al, K, Fe, and Ti but lower

Ba and S concentrations than the cores 87GC and 165GC. Except for Ce, REY concentrations

also increase with depth (Fig. 3).


108

Fig. 3: Top: Depth profiles of selected major elements and three representative REY (Ce, Nd,

Yb). Yb concentrations were multiplied by 10 to fit the scale of the figure. Core pictures depict

that the sediment gets darker with depth in all cores and has thin dark layers throughout.

Oxygen data from Volz et al. (2018). Bottom: Depth profiles of REY parameters HREE/LREE,

MREE/MREE*, Ce/Ce*, and Y/Ho for bulk sediment, the sequential extraction solutions (Na-

dithionite only for HREE/LREE and MREE/MREE* for 194GC 561 cm), and pore water (Y/Ho

only for 194GC-511 cm). HREE/LREE = (Ho + Er + Tm + Yb + Lu)/(La + Ce + Pr + Nd).

MREE/MREE* = (Sm + Eu + Gd + Tb + Dy)/((La + Ce + Pr + Nd + Ho + Er + Tm + Yb + Lu)*2).


109

3.2. REY concentrations and shale-normalized patterns of bulk solid

phase

Total REY concentrations (∑REY) in the CCZ vary between ca. 270 ppm and 850 ppm. REY

concentrations increase two- to threefold at depth in all cores: in 87GC at 700 cm, in 165GC

at 790 cm and in 194GC at 460 cm. The depth of this REY concentration increase appears to

coincide with the sediment color change to dark brown at the lower core end.

The REYSN patterns (Fig. 4) show an enrichment of MREY and HREY over the LREY and a

slight MREY enrichment: LaSN/GdSN ratios range between 0.35 and 0.60 and GdSN/YbSN ratios

between 1.19 and 1.47. There is a trend of increasing Gd/Yb ratio and decreasing La/Gd ratio

with depth, i.e. the enrichment of MREY over LREY and HREY increases. Most layers show a

negative CeSN anomaly with CeSN/CeSN* ratio from 0.27 to 0.89 and significant variation

between cores and with depth. The negative CeSN anomaly develops (165GC and 194GC) or

becomes increasingly negative with depth (87GC) (Fig. 5). It also increases in size with

increasing P concentration (Fig. 5). Pore-water REYSN patterns show slight MREY enrichment

and a negative CeSN anomaly, similar to the sediments (Fig. 6).


110

Fig. 4: REYSN patterns of selected sediment layers of the three cores investigated in this study

(PAAS from Taylor and McLennan, 1985, except for Dy from McLennan, 1989).All cores and

layers show a slight enrichment of MREY and HREY and most layers display a negative CeSN

anomaly.


111

Fig. 5: Increase of negative CeSN anomaly with depth and with increasing P concentration. See

Eq. (1) in chapter 2.4 for the calculation of the CeSN anomaly.

Fig. 6: Left: REYSN patterns of pore waters from 194GC (PAAS from Taylor and McLennan,

1985, except for Dy from McLennan, 1989). All patterns show an enrichment of the MREY and

a pronounced negative CeSN anomaly. Right: Bulk sediment and Ca phosphate phase

normalized to pore water.


112

3.3. Sequential extraction

Selected samples were leached according to the extraction protocol by Köster (2017; Table

2). All REY except Ce are predominantly leached in the first step with Na-acetate (Fig. 7,

Supplementary data 3). Cerium is partially leached in the Na-acetate (1-31% compared to bulk

sediment), and 0.1 M hydroxylamine-HCl (4-22% compared to bulk sediment) steps but mostly

in the 1 M hydroxylamine step (35-59% compared to bulk sediment). Small portions of REY

are leached in the Na-dithionite and NH4-oxalate steps. Similar to the REY except Ce, P is

predominantly dissolved in the Na-acetate step, while Mn is mostly leached in the 0.1 M

hydroxylamine-HCl step and Fe in the Na-dithionite step (Fig. 7; for details see Supplementary

data 3).

Fig. 7: REYSN patterns of sequential leaching solutions of selected sediment layers from the

three cores (PAAS from Taylor and McLennan, 1985, except Dy from McLennan, 1989). From

each core one sample from an upper and lower part of the core was selected. Some data

points are missing for the Na-dithionite and NH4-oxalate patterns due to concentrations below

the LOQ.

The Na-acetate REYSN patterns show a marked negative CeSN anomaly (0.01-0.13) in all layers

and a slight MREY enrichment (Fig. 7). The 0.1 M hydroxylamine-HCl REYSN patterns display

positive CeSN anomalies (1.1-2.8) in the upper parts of the cores and slightly negative CeSN

anomalies in the lower parts of cores 87GC and 194GC (0.56 and 0.86) or a slightly positive


113

CeSN anomaly (1.13) in core 165GC. For the 1 M hydroxylamine-HCl REYSN pattern, we can

discern pronounced positive CeSN anomalies (2.33-5.83) – except for 194GC-561 cm (1.03) –

and sub-chondritic Y/Ho ratios, i.e. negative YSN anomalies, in all layers (19.3-27.0; Fig. 7).

Positive CeSN anomalies are also found in the Na-dithionite extracts (Fig. 7). In the REYSN

patterns of the NH4-oxalate leach, only 165GC-812 cm has a positive CeSN anomaly (1.38),

the other layers show no anomaly.

3.4. Scanning electron microscopy

Scanning electron microscopy (SEM) images show that P and Ca are associated and occur in

the same particles (Fig. 8). Particles are of ca. 20-100 µm size.

Fig. 8: Scanning electron microscopy (SEM) images of particles rich in phosphorus and

calcium; examples from layers 165GC-792 cm (left and middle) and 165GC-812 cm (right).

4. Discussion

4.1. Controls on REY composition of bulk sediment

All cores show a positive correlation of P and Ca (Fig. 9), suggesting the presence of calcium

phosphates (“apatite”: Ca5[(F, Cl, OH)|(PO4)3]) in the sediment. Overall concentrations of

“apatite”, however, are low. Phosphorous and Nd, the latter representing the REY, correlate

positively in all cores (Fig. 9); correlations of REY with P are better than with Ca, because Ca

is also associated with other minerals, e.g., Ca carbonates. Exceptions can be seen for cores

165GC (792-912 cm) and 194GC (521-561 cm) where P concentrations are higher while REY

concentrations remain in the same range as in the other samples (Fig. 9). Even if the outliers


114

are included in the linear regression, the correlation of REY with P within each core is still

positive and better than for REY with other elements (Table 3), suggesting that Ca phosphates

may also be the dominating REY-controlling phase at depth in these cores.

Fig. 9: Left: P vs. Ca plot. Right: P vs. Nd plot. Nd represents the REY. Linear regression lines

for the cores in both graphs and Pearson R correlation coefficients in the legend. All cores

show positive correlations of P and Ca and P and Nd. The deepest layers in 165GC (792-

912 cm) and 194GC (521-561 cm) deviate from the linear regression due to a lower Nd/P ratio

(for further discussion see text).

As all cores show either no or even negative correlations between the REY and the detritus

indicators Al, K, and Ti (Table 3), detrital minerals are unlikely to host a significant fraction of

the total REY budget of the sediments. There are positive correlations of the REY with Fe and

Mn (Table 3) but if the REY were controlled by hydrogenous Fe- or Mn-(oxyhydr)oxides, they

should show positive CeSN (Toyoda et al., 1990) and negative YSN anomalies (Bau and

Koschinsky, 2009), which our REYSN patterns do not (compare Figs. 1 and 4). Additionally, an

impact of redox-zonation on the REYSN patterns could be expected if the REY were bound to

Mn-oxides, because Mn-oxides dissolve in the suboxic layer. Yet, there is no change in REYSN

patterns between oxic and suboxic sections of the sediments (see Figs. 3 and 4) and the

REYSN patterns are similar in the three cores regardless of redox-zonation. For core 165GC,

there is a strong positive correlation of Nd with Ba and S (Table 3), which could be an indicator

that barite (BaSO4) is a REY-controlling phase. Due to size constraints, however, the REY fit

better into the crystal structure of Ca phosphate than Ba sulfate (Guichard et al., 1979). The

same holds true for other biogenic minerals, such as calcite and opal (Piper, 1974a; Elderfield

et al., 1981; Grandjean and Albarède, 1989, and references therein).


115


elements of bulk sediment digestions. Data correlated for the completely analyzed core

sections.

Core Ca P Al Ti K Fe Mn Ba S

87GC 0.97 0.98 -0.46 -0.35 -0.71 0.16 0.73 0.18 -0.18

165GC 0.99 0.98 -0.91 -0.97 -0.85 0.92 0.07 0.97 0.97

194GC 0.91 0.97 -0.87 -0.93 -0.88 0.78 0.89 -0.34 0.86

Besides the positive correlation of REY and P, the Nd/P ratio (Fig. 10) remains similar with

depth, further supporting the association of REY with phosphates. Exceptions are the deep

layers in cores 165GC and 194GC (Fig. 10), which also deviate from the linear regression line

(Fig. 9). Phosphorous and Ca concentrations are extremely high at the bottom of those two

cores, potentially due to increased deposition of apatite material because of increased surface

water productivity (Toyoda et al., 1990). The REY, however, did not become enriched to the

same degree as P and Ca and are present in similar concentrations as in core 87GC at the

bottom. This suggests that REY uptake into the apatite was limited by REY availability in the

pore water.


116

Fig. 10: Nd/P ratio of bulk sediment at different depths for cores 87GC, 165GC and 194GC.

Nd represents the REY. Similar values with depth suggest that the ratio of REY to P stays the

same except in the deep layers (165GC 792–912 cm and 194GC 521–561 cm) where lower

Nd/P values suggest that P is more enriched than the REY. The relative uncertainty of Nd/P

based on NIST-2702 digestions (n = 12 for P and n = 10 for Nd) and measurements is 6.27%.

The SEM results validated the association of Ca and P (Fig. 8), and together with the REY-P

correlations (Fig. 9) as well as the constant Nd/P ratio with depth (Fig. 10), suggest that Ca

phosphates are the overall dominating REY host phase in these sediments over ca. 1000 km

and irrespective of redox-conditions, corroborating the conclusions of other studies (e.g.,

Elderfield et al., 1981; Toyoda et al., 1990; Toyoda and Tokonami, 1990; Toyoda and Masuda,

1991; Ziegler and Murray, 2007; Kon et al., 2014).

4.2. Sequential extraction: REY in phosphate phases

Sequential extraction protocols specifically designed for phosphorous leaching (Filippelli and

Delaney, 1996; Ruttenberg, 1992; Schenau and De Lange, 2000) show comparable steps to

the sequential extraction used in this study. The 1 M Na-acetate solution (step 1 in Table 2)

has been frequently used to extract biogenic and authigenic phosphate (step 3 in Ruttenberg,


117

1992; Filippelli and Delaney, 1996; Schenau and De Lange, 2000). Fish debris (apatite present

as Ca5OH(PO4)3) can be converted to authigenic carbonate fluorapatite (CFA; Ca5F(PO4)3)

during diagenesis through substitution of fluoride for hydroxyl ions, which makes the apatite

more stable and difficult to dissolve (Schenau and De Lange, 2000, and references therein).

Reducible Fe-oxides such as goethite and hematite were leached in the Na-dithionite

extraction (Poulton and Canfield, 2005), step 4 in this study. Sodium-dithionite was also used

to extract Fe-bound P by Ruttenberg (1992), Filippelli and Delaney (1996), and Schenau and

De Lange (2000) as step 2 of their sequential extraction protocols. Hydrogenous Mn- and Fe-

oxides except for goethite were leached using 0.1 M hydroxylamine-HCl in acetic acid by

Toyoda and Masuda (1991) for REE leaching, which is comparable to steps 2 and 3 of the

leaching protocol by Köster (2017; Table 2).

Approximately 40-100% of REY except Ce and ca. 80-100% of P occur in the Na-acetate

extraction fraction compared to the bulk sediment. Exceptions are 194GC-157 cm and -561 cm

with 59% and 54% P, respectively, and 87GC-827 cm, with basically all P and REY in the Na-

acetate fraction. Differences in Na-acetate extraction can be seen for the different REY groups:

LREY ca. 40-100%, MREY ca. 70-100%, and HREY ca. 60-100%. In the remaining steps, ca.

2-20% are leached in the 0.1 M and 1 M hydroxylamine-HCl steps each, and ca. 0-1% in each

the Na-dithionite and NH4-oxalate steps. For further details see Supplementary data 3. This

suggests that P is predominantly present in the phosphate phase and that the REY are

associated with this P phase. Only Ce is associated with oxidic Mn and Fe phases (0.1 and 1

M hydroxylamine-HCl extract, 4-22% and 35-59% Ce, respectively). The REY in the phosphate

fraction (Fig. 7 Na-acetate) control the overall REYSN patterns in our samples (Fig. 4), revealing

that both the pronounced negative CeSN anomaly and the MREY enrichment can be attributed

to this phase.

4.3. Alteration of biogenic Ca phosphates during early diagenesis

The impact of diagenetic processes on REY patterns in apatite has been extensively discussed

for fossil apatite and the applicability of its REY patterns for paleoceanographic reconstructions

(Bright et al., 2009; Chen et al., 2015; Grandjean and Albarède, 1989; Reynard et al., 1999;

Trotter et al., 2016). Various researchers (Grandjean and Albarède, 1989; Reynard et al., 1999;

Wright et al., 1984) proposed that the seawater REY pattern could be recorded in biogenic

apatite without fractionation, which would lead to a preservation of the seawater pattern (in

modern seawater: negative CeSN anomaly, HREY enrichment). This is, however, not what we

see in the samples in this study. The REYSN patterns presented here are rather similar to

altered fish debris (compare Figs. 1 and 4), a main component of which is biogenic Ca

phosphate (Elderfield et al., 1981; Kon et al., 2014; Toyoda et al., 1990). Unaltered Ca


118

phosphates, i.e. fish teeth and marine phosphorites (Fig. 1), show a seawater-derived pattern.

Once they get altered after deposition, (i) the positive YSN anomaly disappears, (ii) the REY

concentrations increase, and (iii) the MREY enrichment develops.

The missing positive YSN anomaly clearly shows that the Ca phosphates did not get their REYSN

pattern from seawater, which typically has super-chondritic Y/Ho weight-ratios (e.g., Y/Ho ca.

54-56, Alibo and Nozaki, 1999; 47-77, Bau and Dulski, 1994) in contrast to slightly sub-

chondritic to chondritic Y/Ho ratios of 24.7-28.7 from the Ca phosphates in this study.

Apatite in living organisms has REY concentrations that are too low to explain the enrichment

in sedimentary Ca phosphate. Moreover, the deviation from the seawater REYSN pattern points

to a post-mortem modification during diagenesis (Auer et al., 2017; Bright et al., 2009;

Elderfield and Pagett, 1986; Reynard et al., 1999; Wright et al., 1984). In addition, Toyoda and

Tokonami (1990) showed that REE accumulation in apatite is too slow to reach the high

concentrations found in sedimentary apatite before burial and that the REY must have been

scavenged from ambient pore waters. Pore-water REY concentrations are one to two orders

of magnitude higher than seawater concentrations (Alibo and Nozaki, 1999; Soyol-Erdene and

Huh, 2013), contributing to the observed enrichment of REY in diagenetic Ca phosphates.

Therefore, transport to the seafloor by settling particles carrying REY, such as organic particles

or Mn- and Fe-(oxyhydr)oxides, and release to the pore water during organic matter

degradation or desorption from Mn- and Fe-(oxyhydr)oxides under oxic conditions is more

likely (Elderfield et al., 1981; Kashiwabara et al., 2018). At the seafloor, REY get incorporated

into Ca phosphates (Chen et al., 2015; Elderfield et al., 1981; Grandjean and Albarède, 1989;

Haley et al., 2004) through Ca2+ ↔ REY3+ ion exchange (Auer et al., 2017; Elderfield et al.,

1981). The fact that Na concentrations in the bulk sediment increase at the same depth as

REY concentrations (Fig. 3) corroborates the suggestion that coupled substitution of Ca2+ by

REY3+ and Na+ is the uptake mechanism of REY into apatite. As there is only minor

fractionation of the REY during their incorporation into Ca minerals, such as Ca carbonates

and phosphates by coupled substitution (Nothdurft et al., 2004; Webb and Kamber, 2000), this

suggests that ambient pore water is the source of REY in the Ca phosphates. The matching

pore-water REYSN patterns for core 194GC and the bulk sediments (compare Figs. 4 and 6)

as well as the smooth line when normalizing the Ca phosphate phase from the Na-acetate

leaching step to pore water (Fig. 6), confirms that the trivalent REY are incorporated without

major fractionation. The MREY enrichment of the Ca phosphates is therefore also taken over

from the pore water. The Ca phosphate signature, hence, develops at the sediment-water

interface during aerobic early diagenesis. Because of the low sedimentation rate in pelagic

sediments, the settled particles stay at the seafloor surface for centuries to millennia and


119

undergo degradation and dissolution during early diagenesis (Elderfield et al., 1981; Reynard

et al., 1999). The MREY enriched pattern is already developed at shallow depth (194GC-5 cm),

which shows that the shift away from the seawater pattern happens at the sediment-water

interface and the pattern intensifies with depth because of slightly better fitting ionic radii of the

MREY than the LREY and HREY into the Ca phosphate crystal structure when substituting

REY for Ca (Wright, 1990). The increase of the MREY enrichment with depth is a sign of further

diagenetic overprint. The MREY enrichment in recent sediments is much less pronounced than

in fossil apatites (e.g., Grandjean and Albarède, 1989; compare Figs. 1 and 4). All three

observed shifts away from the seawater REYSN pattern, i.e. sub-chondritic Y/Ho, high ∑REY

concentrations, and MREY enrichment, prove that Ca phosphates are not a suitable archive

for paleoceanographic reconstructions for e.g., seawater chemistry. Our results corroborate

previous findings that fossil apatite REY patterns resemble modern pore-water patterns

(Trotter et al., 2016).

4.4. Pore-water REY pool

As we described in Section 4.3 that the MREY enrichment in the bulk sediment and Ca

phosphate phase is pore water derived, it remains to discern how the MREY enrichment in the

pore water develops as well as what the REY source to the pore water is. A variety of solid-

phase fractions could be potential sources of REY to the pore water: Ca carbonates and

phosphates, POC, Mn- and Fe-(oxyhydr)oxides. REY in the pore water cannot originate from

Ca carbonate or Ca phosphate phases because those show a primary seawater REYSN pattern

(HREY enrichment) (Bau and Alexander, 2006; Nothdurft et al., 2004; Webb and Kamber,

2000) when they get deposited and would release that seawater pattern to the pore water,

which we do not see for the 194GC pore-water REYSN patterns (Fig. 6). The MREY enriched

pore-water pattern could result from the release of REY from decaying POC at the sediment-

water interface that are subsequently stabilized by dissolved organic carbon (DOC) and lead

to a MREY enriched pattern in the dissolved pore-water pool (Soyol-Erdene and Huh, 2013;

Tang and Johannesson, 2010). There is, however, too little TOC (<0.2–0.45 wt.% Volz et al.,

2018) to explain the extent of REY increase. In addition, the highest REY concentrations in the

pore water and the bulk solid phase should then be visible in the upper 1 m of the cores where

highest TOC concentrations prevail. The aging of Mn- and Fe-(oxyhydr)oxides may release

REY to the pore water, also under oxic conditions (Kashiwabara et al., 2018). Fe-Mn crusts,

for example, show a decrease in REY concentrations from the outside to the inside (Bau et al.,

1996), i.e. with increasing age the concentration decreases, pointing to a release of REY from

the solid phase. This might be due to progressive crystallization of the amorphous precipitates

and exclusion of REY from the crystal lattices, similar to dehydration (e.g., Koschinsky and

Halbach, 1995). These REY are then redistributed to phosphate phases via pore water. The


120

REY source also has to have insoluble Ce(IV) compounds, because Ce is not released to the

Ce-depleted pore water (Fig. 6). This is typical for Mn- and Fe-(oxyhydr)oxides, at the surface

of which Ce3+ can be oxidized to Ce4+ (Bau and Koschinsky, 2009; Kashiwabara et al., 2018;

Takahashi et al., 2007). The Mn and Fe phases of the sequential extraction (0.1 and 1 M

hydroxylamine-HCl) also show slight MREY enrichment (Fig. 7), suggesting Mn and Fe phases

as the REY source to the pore water. The released REY can then reach other layers by

diffusion (e.g., Elderfield and Sholkovitz, 1987) so that the REY redistribution should not only

be restricted to the layer where the REY originate from. A REY source might be located at

even greater depth from which the REY diffuse upwards.

4.5. Increase of negative CeSN anomaly with depth

Our CeSN anomaly values fit well with data from other studies conducted in the Pacific (0.48-

0.79: Toyoda and Masuda, 1991). Besides the development of the MREY enrichment and the

concentration increase at depth, a decrease of CeSN/CeSN* values with depth is observed

(Fig. 5).

One aspect that impacts the distribution of CeSN anomalies is bioproductivity: Negative CeSN

anomalies are found in sediments with higher surface water productivity but not in sediments

underlying less productive surface waters (Toyoda et al., 1990). A fish debris sample from a

low bioproductivity area even showed a positive CeSN anomaly (Toyoda and Masuda, 1991).

The primary productivity in the equatorial Pacific Ocean decreases from east to west and

coincides with the occurrence of negative CeSN anomalies in the upper 100 cm of sediment

(Toyoda et al., 1990). Following this pattern, low latitudes in the Pacific Ocean display negative

CeSN anomalies and high latitudes show positive CeSN anomalies, and a shift of negative CeSN

anomalies to no CeSN anomalies from east to west in surface sediments can be observed

(Toyoda et al., 1990). In the eastern core 87GC from this study, CeSN anomalies in the surface

sediments are around 0.8. Farther west and north, away from the equator, cores 165GC and

194GC show no CeSN anomalies in surface sediments (Fig. 11). At depth, however, both cores

(165GC and 194GC) have negative CeSN anomalies (0.27 and 0.43) similar to core 87GC from

the eastern CCZ.

Decreasing Ce/Ce* values with depth have been found in a 70 m long ODP core from the

Central Pacific (26°N, 147.5°W), where they were largely explained by different mineralogy

(red clay vs nannofossil ooze) (Ziegler and Murray, 2007). In this study, however, we see a

smooth decrease of CeSN/CeSN* values in the upper meters of the cores, but this decrease

intensifies with depth (Fig. 11). Layers with similar CeSN/CeSN* values can be seen in the three

cores, but they occur at different depths. In core 165GC, the layer with CeSN/CeSN* values of


121

0.7-0-36 (Fig. 11) is missing which might be due to a hiatus located at 790 cm, which is

common in the Late Miocene and Paleo-Pleistocene sections in the CCZ (International Seabed

Authority, 2010; Johnson, 1972; von Stackelberg and Beiersdorf, 1991). The transition from

lighter to darker sediment at this depth is distinct. There must have been a drastic change in

sedimentation rate, non-deposition, or erosion of previously deposited material that lead to the

abrupt change visible in the sediment record (Craig, 1979; Mewes et al., 2014; Volz et al.,

2018; von Stackelberg and Beiersdorf, 1991). Deposition similar to the other areas studied

here must have commenced again afterwards as the major element and REY profiles are in

the same range as in cores 87GC and 194GC. The change is visible for other parameters as

well: change in sediment color, jump in Ca vs P concentration as well as concentration

increases for the other major elements, as well as a change in the MREY pattern (small GdSN

anomaly appears) while it stays the same for all layers in the upper part of core 165GC. The

transition at the bottom of core 194GC is much smoother and there is no hiatus in this core.

The age at the bottom of the core might still be similar to that at the bottom of core 165GC

below the hiatus, but in core 194GC it is due to the low sedimentation rate (0.2 cm kyr-1

compared to 0.64 cm kyr-1 in core 165GC, Volz et al., 2018).


122

87GC 165GC 194GC Color legend

(cm) CeSN/CeSN* (cm) CeSN/CeSN* (cm) CeSN/CeSN*

5 *1.07 1.07

85 *1.07 1.03

157 *1.03 1.02

201 *1.02 1

261 *1.02 0.97

47 *0.95 321 *0.96 0.96

127 *0.97 381 *0.9 0.95

187 *0.95 0.9

267 0.83 0.89

367 0.88 0.88

407 0.83 0.87

467 *1 0.85

35 0.8 517 0.87 461 0.85 0.83

123 0.69 612 0.89 0.8

163 0.67 672 0.75 0.75

203 0.75 732 0.72 0.72

243 0.7 0.7

283 0.69 0.69

343 0.64 0.67

383 0.6 0.64

443 0.56 0.6

483 0.57 0.57

607 0.57 0.56

667 0.57 0.54

521 0.54 0.43

727 0.42 561 0.43 0.42

767 0.4 0.4

827 0.39 0.39

887 0.36 0.36

792 0.3 0.31

832 0.31 0.3

912 0.27 0.27

Fig. 11: Ce/Ce* values for each layer. Ce/Ce* was calculated according to equation (1) in the

text. Yellow star symbols denote no CeSN anomaly. Values decrease with depth in all three

cores, starting with different Ce/Ce* values at the top of the sediment cores.

In contrast to all other REY which are sequentially extracted with phosphate in the Na-acetate

leaching step, Ce is mostly leached with Mn and Fe phases. Other studies also found REY

(except for Ce) in the easily leachable fraction (0.5mol/L HCl or 0.2mol/L H2SO4) of Pacific

sediments (e.g., Kato et al., 2011) and studies of Fe-Mn crusts proved that the trivalent REY

are associated with phosphate and/or Fe phases while Ce is linked to Mn phases (De Carlo,

1991) or as found here, with both, the Mn- and Fe-(oxyhydr)oxide phase (Bau and Koschinsky,

2009). This association of Ce with a different phase than the other REY also indicates that the


123

CeSN anomaly is also sensitive to concentration variations of the trivalent REY and not only to

changes in Ce concentration (De Carlo, 1991). In our samples, the Ce concentration shows

only little change, but the trivalent REY concentrations increase significantly and systematically

with depth. The CeSN anomaly is, therefore, mostly a result of increasing concentrations of

trivalent REY, corroborating results of Elderfield et al. (1981) who also found that there is no

significant diagenetic remobilization of Ce compared to the other REY.

An explanation for the increase of the CeSN anomaly with depth is that Ce3+ adsorbs to Mn-

and Fe-(oxyhydr)oxides (as we see from the sequential extractions; see also Bau and

Koschinsky, 2009) and only then is oxidized at the surface of the metal (oxyhydr)oxides. These

Ce4+ compounds, however, cannot be released to pore water in the oxic part of the sediment

column close to the surface, in contrast to the REY3+. The trivalent REY in the Ce-depleted

pore water are later incorporated into the Ca phosphate component, producing negative CeSN

anomalies. Chen et al. (2015) expected the CeSN anomaly to become smaller with depth as

Mn-oxides are reduced in the suboxic zone and release Ce. This change, however, cannot be

observed in our sample set in which the solid-phase Ce concentration does not increase when

conditions become suboxic, i.e. where hydrogen sulfide and oxygen are absent and Mn (IV)

and nitrate reduction take place (450 cm 165GC (O2 < 10 µmol/L) and 300 cm 87GC (O2 < 20

µmol/L); and 300 cm 87GC; Volz et al., 2018). This could be explained by the fact that Ce may

mostly be bound to clusters of discrete Ce (IV) oxide phases of low solubility or Fe-

oxyhydroxides (as indicated by sequential extraction), which were shown to be preserved and

not subject to reductive dissolution in the suboxic zone of the CCZ sediments at depths down

to ca. 10 m (Heller et al., 2018; Volz et al., 2018).

5. Conclusions

We suggest Ca phosphates as the host phase for REY in siliceous-ooze-rich silty clay

sediments in the abyssal Pacific Ocean. The phosphates become overprinted during early

diagenesis: a MREY enrichment over the LREY and HREY develops at the sediment–water

interface by uptake of REY from pore water, the positive YSN anomaly disappears, and a

negative CeSN anomaly becomes increasingly negative with depth or develops where no

negative CeSN anomaly is present at the sediment surface. The bulk sediment solid-phase Ce

concentrations remain relatively constant, while trivalent REY concentrations increase with

depth due to continuous incorporation of REY from the pore water by coupled substitution,

creating a larger negative CeSN anomaly. This is because Ce is mostly bound to Mn- and Fe-

(oxyhydr)oxides as Ce(IV) while the other trivalent REY are bound to Ca phosphates.


124

The matching pore water, Na-acetate sequential extraction, and bulk solid-phase REYSN

patterns show that the apatite takes over the pore-water REYSN pattern with only minor

fractionation. Our findings are hence also relevant for paleoceanographic work because

biogenic apatites in fossil records are frequently used to reconstruct ancient seawater

chemistry. Our results, however – like those of some other studies (Bright et al., 2009; Chen

et al., 2015; Elderfield and Pagett, 1986; Reynard et al., 1999; Sholkovitz et al., 1989; Toyoda

and Tokonami, 1990) – advise caution when using REY patterns for paleoceanographic

reconstructions since the pattern seems to be strongly diagenetically influenced and

determined by the pore-water REY pool. Additionally, the negative CeSN anomaly often

interpreted as a sign of oxic seawater in paleoenvironmental reconstructions was produced

during early diagenesis in our samples and does not reflect the negative CeSN anomaly of

modern seawater.

Even though we found similar REY patterns and apatite as the REY host phase over large

geographic distances of ca. 1000 km, it has to be considered that sediment composition in the

study area can be highly variable on different spatial scales. More research is necessary to

describe REY behavior during early diagenesis in CCZ sediments with different mineralogy,

e.g., extensive carbonate or metalliferous layers.

There is still little knowledge about REY cycling in pore waters and the exchange between pore

waters and different sediment fractions. Analyses of the Nd isotopic composition might hence

be beneficial to constrain REY sources to deep-sea pore waters.

Acknowledgements

This work was made possible by the captain and crew of RV SONNE and the chief scientist

Pedro Martínez Arbizu on cruise SO239 and their help during the sampling campaign. Our

appreciation goes to Annika Moje and Inken Preuss from Jacobs University Bremen (the latter

now at GEOMAR Kiel) and Ingrid Dohrmann and Ingrid Stimac from the Alfred Wegener

Institute Helmholtz Centre for Polar and Marine Research for help during sampling onboard

RV SONNE and in the home laboratories. We thank Susann Henkel for help with developing

the sequential extraction protocol and Gerhard Kuhn and Ruth Cordelair for providing and

helping with the SEM analyses. This work was funded by the German Federal Ministry of

Education and Research in the framework of the JPI Oceans project MiningImpact (grant

numbers 03F0707F+G). We thank the three reviewers and associate editor Tina van de Flierdt

for their helpful comments and suggestions to improve this manuscript.


125

Appendix A. Supplementary Material Supplementary data to this article can be found online at

https://doi.org/10.1016/j.gca.2019.02.019.


126

Chapter 5 – REY in metalliferous and carbonate-rich sediments of the CCZ

127

Chapter 5 - Rare earth elements and yttrium in

metalliferous and calcium-carbonate-rich

sediments from the central equatorial Pacific

Title

Rare earth elements and yttrium in metalliferous and calcium-carbonate-rich sediments from


Authors

Sophie A. L. Paul, Michael Bau, Thomas Kuhn, Jessica B. Volz, Inken Preuss, Sabine Kasten,

Andrea Koschinsky

First draft


128


129

Rare earth elements and yttrium in metalliferous

and calcium-carbonate-rich sediments from the

central equatorial Pacific

Sophie A. L. Paul1, Michael Bau1, , Thomas Kuhn2, Jessica B. Volz3, Inken Preuss1,4, Sabine

Kasten3,5, Andrea Koschinsky1


2German Federal Institute for Geosciences and Natural Resources, Hannover, Germany

3Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research, Bremerhaven,

Germany


5Faculty of Geosciences, University of Bremen, Bremen, Germany

Abstract

Rare earth elements and yttrium (REY) are frequently used as indicators for environmental

conditions and to determine origins of and changes undergone by sediments, rocks, and water.

REY in sediments have often been studied in the past. Nevertheless, the phase associations

of REY in deep-sea sediments dominated by different material, such as siliceous ooze,

calcareous ooze, clay minerals, and alterations taking place during early diagenesis, are not

yet well understood. Here we present solid phase REY data for sediments from the Clarion

Clipperton Zone (CCZ) in the central equatorial Pacific, that in parts consist of Ca-carbonate-

or metalliferous-rich layers. REY concentrations are lowest in Ca-carbonate-rich layers and

enriched in Mn-rich layers. Correlations of REY with major elements, however, point to a Ca-

phosphate control of REY in these sediments, which is similar to cores in the central Pacific

without carbonate or metalliferous layers. Shale-normalized (SN) REY patterns show middle

REY enrichment over the light and heavy REY (LaSN/GdSN=0.34-0.54 and GdSN/YbSN=1.05-

1.60), as well as pronounced negative CeSN anomalies (CeSN/CeSN*=0.19-0.84). These

patterns suggest diagenetic alteration of the Ca phosphate REYSN pattern during early

diagenesis due to uptake of REY from ambient pore waters.

Keywords: rare earth elements, calcium phosphates, metalliferous sediments, calcium

carbonate layers, CCZ


130

1. Introduction

Rare earth elements (REE) can be used to determine sediment provenance as well as past

and present environmental conditions (e.g., Bau and Dulski, 1996a; Bright et al., 2009). Due

to their chemical properties, i.e. similar ionic radii and trivalent charge, they behave coherently

in the environment. Yttrium, which is also trivalent and of similar size as Ho, is often included

with the REE, forming the REY. The small differences in ionic radii are, however, significant

enough to lead to fractionation during chemical complexation or during incorporation into

crystal lattices. The resulting fractionation is visible in the shale-normalized (SN) patterns as

enrichments and anomalies. Cerium and Eu in particular can form pronounced anomalies

because of their redox sensitivity, meaning they can occur in the tetravalent and divalent state,

respectively.

Many studies have investigated REY in pelagic sediments (Elderfield et al., 1981; Kato et al.,

2011; Kon et al., 2014; Paul et al., 2019; Piper, 1974b, 1974a; Piper et al., 1988; Toyoda et

al., 1990; Toyoda and Masuda, 1991; Toyoda and Tokonami, 1990;). Over large areas of the

Pacific, REY are associated with Ca phosphates, except in hydrothermally influenced

metalliferous sediments e.g., close to the East Pacific Rise, where they are associated with Fe

(oxyhydr)oxides (Kato et al., 2011; Kon et al., 2014; Ruhlin and Owen, 1986; Toyoda et al.,

1990). Sediments in close proximity to hydrothermally active sites as well as particles from

hydrothermal plumes display positive EuSN anomalies (German et al., 1990; Michard, 1989;

Ruhlin and Owen, 1986). The Fe (oxyhydr)oxides scavenge REY from seawater and develop

a seawater REYSN pattern with increasing distance from the hydrothermally active site

(German et al., 1990), i.e. heavy REY enrichment over the light REY (HREY and LREY,

respectively), negative CeSN, as well as positive LaSN, GdSN, and YSN anomalies. The Ca

phosphate REYSN patterns show middle REY (MREY) enrichment over the LREY and HREY,

and no or negative CeSN anomalies (Paul et al., 2019; Toyoda et al., 1990; Toyoda and

Masuda, 1991). These patterns develop during early diagenesis, where the Ca phosphates

accumulate REY from ambient pore waters and lose their seawater REYSN pattern (Paul et al.,

2019). Consequently, diagenetically altered phosphates have high REY concentrations

compared to unaltered phosphates.

Calcareous-ooze-rich sediments have low REY concentrations, approx. ≤200 ppm (Kato et al.,

2011; Pattan and Higgs, 1995). This is either due to the fast deposition of calcareous material

wherefore the sediments do not have time to accumulate high REY concentrations (Kato et

al., 2011) or because the REY are primarily bound in other phases, e.g., clay or Mn and Fe

(oxyhydr)oxides (Pattan and Higgs, 1995), which are diluted by the carbonate. Calcareous

sediment, as well as foraminifera specifically, show seawater-like REYSN patterns, suggesting


131

that the REY are taken up without fractionation by the skeletal material (Elderfield et al., 1981;

Pattan and Higgs, 1995). But REY in CaCO3 skeletal material are often impacted by diagenesis

after deposition, during which higher concentrations of REY are accumulated and REY

fractionate, obscuring the seawater REYSN pattern (Haley et al., 2005; Johannesson et al.,

2006; Webb and Kamber, 2000), via a similar process to that undergone by biogenic Ca

phosphates, as introduced above. Haley et al. (2005) proposed that foraminiferal tests

preserve the seawater REYSN pattern in oxic environments but get overprinted during anoxic

diagenesis, leading to higher REY concentrations and an MREY enrichment over the LREY

and HREY. If the MREY enriched pattern is found, they propose that it is a sign of anoxic

diagenesis in the past (Haley et al., 2005).

Most studies in the past focused on REY in surface sediments and excluded layers that

deviated from the general pattern (Elderfield et al., 1981; Toyoda et al., 1990). Our recent work

detected a wider variety of REY distribution in deep-sea sediments and that cores with

extensive metalliferous and carbonate layers do not follow the trend of enrichment and

increasing negative CeSN anomaly with depth. These exceptions will be the focus of this study.

2. Methods

2.1. Sampling

Cores were taken during RV SONNE cruises SO239 and SO240 in 2015 (Kuhn, 2015;

Martínez Arbizu and Haeckel, 2015). A gravity corer (GC or SL) was used for sampling

sediments down to 13 m depth. To capture the surface layer of 117SL, of which ca. 1.5 m were

lost during GC sampling, a multicorer (MUC) was taken at the same site. Cores 69SL, 117SL,

and 116MUC were taken in the German (BGR: Bundesanstalt für Geowissenschaften und

Rohstoffe) contractor area for manganese nodule exploration and 122GC in the Belgian (GSR:

Global Sea Mineral Resources NV) area (Table 1 and Figure 1).


132

Table 1: Overview of core samples. SL=Schwerelot (German: gravity core), GC=gravity core,

MUC=multicore.

Sample ID Location Area (deep-sea mining contractors)

Core length (cm)

Water depth (m)

SO240_69SL 12° 39.855’ N 119° 13.374’ W

BGR 1265 4275

SO240_116MUC 13° 11.098’ N 118° 06.003’ W

BGR 31 4270

SO240_117SL 13° 11.103’ N 118° 05.992’ W

BGR Ca. 600 4271

SO239_122GC 13° 51.22’ N 123°15.29’ W

GSR 761 4517

Figure 1: Map of sampling locations. The map was created using GeoMapApp.


Bulk sediment acid pressure digestions were conducted at the Alfred Wegener Institute

Helmholtz Centre for Polar and Marine Research (AWI), Bremerhaven, Germany. 50 mg

sediment, previously freeze-dried and homogenized, were digested using a MARS Xpress

(CEM) microwave system at 230°C, following procedures described by (Nöthen and Kasten,

2011). A mixture of 3 mL nitric acid (HNO3, 65%, sub-boiling distilled), 2 mL hydrochloric acid

(HCl, 30%, sub-boiling distilled), and 0.5 mL hydrofluoric acid (HF, 40%, suprapure) was used.

After the digestion, solutions were evaporated at 200°C and re-dissolved in 5 mL 1 M HNO3.

The fully digested sample was filled up to 50 mL with 1 M HNO3.


133

Major element concentrations (Al, Ba, Ca, Fe, K, Mn, P, S, Ti) were measured at AWI with

ICP-OES (IRIS Intrepid, Thermo Elemental). Certified reference material (CRM) NIST-2702

was used to check method and machine performance. Analytical accuracy was within 7% for

all elements except S (13%) (n=12) and method precision, given as relative standard deviation

(RSD) was within 2% for all elements except K (3%) and Fe (4%) (for details see

Supplementary Material 1). The REY were measured at Jacobs University Bremen (JUB) with

an ICP-MS (Perkin Elmer Nexion 350x), checking accuracy and precision with CRMs NIST-

2702 and a BHVO-2 from a different digestion completed at JUB. Analytical accuracy was

within 6% except Nd (9%) (n=6) and method precision was within 2% RSD for all REY (for

details see Supplementary Material 1).

2.3. Reporting

Results for bulk solid phase data were corrected for the pore-water salt matrix according to

Kuhn (2013) to adjust for impacts of pore-water constituents on the solid-phase concentration.

All REY patterns are normalized to PAAS, using REY data from Taylor and McLennan (1985),

except for Dy from McLennan (1989).

The CeSN anomaly was calculated after Bau and Dulski (1996a):

(1) 𝐶𝑒

𝐶𝑒∗=

𝐶𝑒(𝑆𝑁)

(0.5∗𝐿𝑎(𝑆𝑁)+0.5∗Pr(𝑆𝑁))

We define a negative CeSN anomaly for CeSN/CeSN* values < 0.9, no CeSN anomaly for

CeSN/CeSN* values between 0.9 and 1.1, and a positive CeSN anomaly for CeSN/CeSN* values

> 1.1.


134

3. Results

3.1. Ca-, Mn-, and Fe-rich layers

Figure 2: Depth profiles of selected major elements and three representative REY (Ce, Nd,

Yb). Yb concentrations were multiplied by 10 to fit the scale of the figure. Note that ca. 1.5 m

of 117SL were lost during sampling. Therefore, two layers (7.5 cm and 23 cm) of the

corresponding MUC (116MUC) were included to represent surface sediment. The sediments

are oxic throughout (Kuhn, 2015; Volz et al., 2018).

The two cores from the eastern CCZ, 117SL and 69SL were found to have Ca-rich layers at

about 500-800 cm depth (Figure 2). Approximately 1.5 m of the top of core 117SL was lost

during retrieval, which is not corrected for in Figure 2. Therefore, the Ca-rich layer with

concentrations up to 30 wt.%, visible at approx. 350 cm in Figure 2, actually starts at ca.

500 cm seafloor depth. We also included two layers of the corresponding MUC (116MUC)

which was taken at the same location to have indicative data for the surface sediments – those

are the two uppermost values in the 117SL profiles in Figure 2. Core 122GC from a more

westerly location does not show this Ca-rich layer. Manganese, Fe, Ba, and Al concentrations

are lower in the Ca-rich core areas. Barium concentrations are highest in the middle of the

cores between approx. 200 to 600 cm. Phosphorus concentrations are relatively constant

throughout, with small peaks in both cores.

Two metalliferous areas were identified within 122GC: Mn concentrations as well as the Mn/Al

ratio increased at 220 cm and remained high throughout the lower part of the core; a Fe-rich

phase started at 546 cm and continued to the bottom of the core. In 69SL, Mn concentrations

peak at 205 cm and then show an increasing trend with depth, while they are constant


135

throughout in 117SL (< 1 wt.%) except for a peak between 105 to 165 cm (Figure 2). The Mn-

and Fe-rich layers do not coincide with oxygen penetration depths as all cores were oxic

throughout (Kuhn, 2015; Volz et al., 2018).

3.2. Rare earth elements and yttrium (REY)

REY concentrations were lowest in the Ca-rich parts of cores 69SL and 117SL (Figure 2). For

117SL, REY concentrations were highest (~400 ppm) at the top (MUC) and in the middle of

the GC (around 205 cm), with an exceptionally high concentration (829 ppm) measured at

105 cm. At 145 cm, Mn and Fe concentrations showed a peak and REY concentrations are

comparatively low. There is a Fe-rich phase at 305 cm, where REY concentrations were again

slightly higher than in the surrounding sediment. The REY concentration peaks at 105, 205,

and 305 cm corresponded with peaks in the P profile.

In 69SL, REY concentrations increased until 265 cm, decreased with depth down to 625 cm

and then increased again to remain relatively stable around 260 ppm. Highest concentrations

are found at 265 cm where there was also a concentration peak in Mn concentrations (Figure

2).

In core 122GC, highest concentrations were found in the middle of the core (200-506 cm), in

the Mn-rich part. The concentration jump from 180 cm to 200 cm could also be seen for Ca

and P.

REYSN patterns show MREY enrichment over LREY and HREY in all cores and all layers

(Figure 3) with LaSN/GdSN ratios of 0.34 to 0.54 and GdSN/YbSN ratios of between 1.05 and 1.60.

LaSN/GdSN ratios had the tendency to increase with depth for 117SL and 69SL, while GdSN/YbSN

ratios decreased with depth. For 122GC LaSN/GdSN ratios were constant at approx. 0.4 until

546 cm and elevated below (> 0.4). In contrast, GdSN/YbSN ratios were highest in the Mn-rich

part between 200 cm and 546 cm with values between 1.42 and 1.60. Additionally, all layers

displayed pronounced negative CeSN anomalies (CeSN/CeSN*=0.19-0.84). At 116MUC/117SL

and 69SL, CeSN/CeSN* ratios are highest in the surface sediments (CeSN/CeSN*=0.61-0.76) and

then consistent at 0.3 between 105 and 500 cm depth in 117SL, with a decrease from approx.

0.4 to 0.2 in 69SL. Y/Ho ratios are slightly sub-chondritic throughout with values between 25.0

and 27.5.


136

Figure 3: REYSN patterns of the three cores from this study. All cores show MREY enrichment

and negative CeSN anomalies.


137

4. Discussion

4.1. REY control phase

In the Ca-rich layers of 69SL and 117SL, where Ca reaches concentrations of up to 30 wt.%,

Ca was present as carbonate. The carbonate-rich sediment layers in 69SL and 117SL were

dated to approximately 15 Ma using Be isotopes, and the area must have moved away from

the equatorial high productivity area afterwards (Kuhn, 2015). As core 122GC does not contain

carbonate layers, it may have never been in the equatorial high productivity zone or the

carbonate dissolved or eroded. Today, the calcite compensation depth in the CCZ is at approx.

4500 m (e.g., Bramlette, 1961 in Berger, 1970; Lyle, 2003) and consequently 122GC is located

below this depth (Table 1). These high Ca concentrations stand in marked contrast to Ca

concentrations of only 0.5 – 2 wt.% in the remaining layers of the Ca-rich cores and 122GC.

Where Ca was present in small quantities (Ca < 3 wt.%), the cores show positive correlations

of Ca and P, suggesting the presence of a Ca phosphate phase (Figure 4). Since most Ca was

bound in CaCO3 in the Ca-rich regions, the Ca-P correlation cannot be held up in these parts

because the Ca-phosphate content is significantly lower than the CaCO3 content.

Figure 4: P-Ca plot of the three cores from this study. Pearson R coefficients for linear

regressions in the Ca-poor parts of 69SL and 117SL, as well as for the entire 122GC core,

show a positive correlation of P and Ca.

REY (representatively using Nd) and P showed positive correlations in 117SL (Figure 5) and

the correlation would have been even stronger if the data point with exceptionally high P and

Nd concentrations was included (Pearson R=0.92). In contrast, Nd and P do not correlate in


138

69SL, not even if the outlier was included (Figure 5). Core 122GC has no positive Nd-P

correlation throughout the entire core, but if the core was split into two parts at 546 cm, where

the Fe-rich layer starts, two distinct parts can be discerned with positive correlations of Nd and

P in each (Figure 5). The Nd-P correlation suggests the control of REY by Ca phosphates,

which is a commonly reported phenomenon in the Pacific (Elderfield et al., 1981; Kon et al.,

2014; Paul et al., 2019; Toyoda et al., 1990; Toyoda and Masuda, 1991; Toyoda and

Tokonami, 1990). Additionally, REYSN patterns of diagenetically altered Ca phosphates that

continuously incorporate trivalent REY from the pore water show MREY enrichment and

negative CeSN anomalies also found in the cores from this study (Figure 3). The MREY

enrichment and marked negative CeSN anomaly are typical for diagenetically altered Ca

phosphates in pelagic sediments (Toyoda et al., 1990; Toyoda and Masuda, 1991; Toyoda

and Tokonami, 1990).

Figure 5: Nd (representing the REY) vs. P plots for the three cores from this study. An outlier

was excluded from the correlation analyses in cores 117SL and 69SL each. Correlations for

122GC were conducted in two parts and the core split at 546 cm, where the Fe-rich layer starts.

Similar to the Ca-P correlations, in the Ca-rich layers of 69SL (Ca-Nd: Pearson R=-0.85) and

117SL (Ca-Nd: Pearson R=-0.94) Ca correlate linearly negatively with the REY. A negative

linear correlation between REY and CaCO3 was also found by Wang et al. (1986) in deep-sea

sediments from the South Atlantic and by Pattan and Higgs (1995) in the Indian Ocean,

whereas a negative non-linear relationship was discovered by Toyoda et al. (1990) in Pacific

deep-sea sediments.

Since REY concentrations are lowest in the carbonate layers, the carbonate must have a

dilution effect on the REY carrying phase (e.g., Glasby et al., 1987; Kato et al. 2011). Kato et

al. (2011) also found low REY concentrations in carbonate layers in the Pacific (approx.

200 ppm), concluding that the rapid deposition of carbonate leads to the low REY


139

concentrations. This is supported by studies from the Indian Ocean, where REE concentrations

in calcareous sediments were also <<200 ppm, but the investigators concluded that the REY

are bound in clay as well as Mn and Fe oxide phases (Pattan and Higgs, 1995). A dilution

effect of carbonate on REY concentration as well as a negative linear correlation of Ca and

REY has also been observed in deep-sea sediments of the Peru Basin (Paul et al., in prep).

There are no correlations of Nd with Fe, Mn, Al, Ti, and K in any core, excluding the possibility

of Fe and Mn oxide, as well as detrital material acting as REY controlling phases (Table 2).

There is, however, a correlation of Nd with Ba and S in the eastern cores (117SL and 69SL;

Table 2), potentially indicating barite as a REY controlling phase. Due to the size of the REY’s

ionic radii, they fit better into the Ca phosphate than into the barite crystal lattice. If the Ca

phosphate and the barite are both biogenic (Dymond et al., 1992; Toyoda et al., 1990; Toyoda

and Tokonami, 1990), similar deposition corresponding to surface water productivity and

therefore also similar correlations with REY are plausible explanations for these positive

correlations, even though the REY are not incorporated into the barite.


elements.

Fe Mn Ba S Al Ti K

117SL 0.15 0.29 0.98 0.87 0.75 0.78 0.71

69SL -0.04 0.23 0.89 0.67 0.64 0.67 0.62

122GC -0.24 0.37 0.80 0.10 0.35 (below 506 0.79)

0.34 (below 506: 0.91)

0.23 (below 506 0.49)

Since no positive correlation of Nd with Mn and Fe in the metalliferous 122GC core was

observed, these metal oxide phases do not seem to play are role in controlling the REYSN

pattern. Usually metalliferous layers are associated with hydrothermal inputs (Bischoff and

Rosenbauer, 1977; Ruhlin and Owen, 1986), but the sequence of Mn-rich layer and Fe-rich

layer may indicate remnants of oxic/suboxic fronts. Hydrothermal influence can be excluded

for 122GC with no EuSN anomalies, indicative of hydrothermally influenced REY (Michard,

1989), visible. The upward diffusion of dissolved Mn and Fe from suboxic pore waters is

stopped at the oxic/suboxic boundary, where oxides precipitate and form Fe (oxyhydr)oxide

and Mn oxide layers (Thomson et al., 1993). These layers are stabilized in oxic sediments, so

that preservation of oxic/suboxic fronts is most likely the process responsible for the visible

layers in 122GC.

From the correlation analysis, we conclude that the REY are associated with a Ca phosphate

phase in 117SL and 122GC, but we cannot identify a carrier phase for 69SL.


140

4.2. REY concentrations with depth

We previously established that REY concentrations generally increase with depth in siliceous-

ooze-rich central equatorial Pacific deep-sea sediments due to continuous incorporation of

REY from ambient pore waters (Paul et al., 2019). In 117SL, the REY concentrations decrease

with depth, but this is only due to the increase of CaCO3 which dilutes the REY-carrying phase.

This is also visible in 69 SL, where the CaCO3-rich layers have lowest REY concentrations.

REY concentrations were highest in the middle part of 122GC, in the Mn-rich layer, as well as

in the Mn enriched layer at 265 cm in 69SL. This shows that Mn oxide layers accumulate REY,

but Fe (oxyhydr)oxide layers do not: the Fe-rich layer in 122GC does not display substantial

REY enrichment. The relationship of Fe, Mn and REY concentrations in 117SL is more

obscured, but neither phase seems to control the high REY concentrations, e.g., at 105 cm.

There, the REY concentration peaks correlate with P (Figure 2).

4.3. MREY enrichment

The slight MREY enrichment observed in the samples (Figure 3) is typical for diagenetically

overprinted Ca phosphates (e.g., Bright et al., 2009; Paul et al., 2019). The increase of

LaSN/GdSN ratios and decrease of GdSN/YbSN ratios in cores 117SL and 69SL indicate an

intensification of the MREY enrichment with depth. This is a sign of further incorporation of

REY from ambient pore waters and continuing diagenetic overprinting as the REYSN pattern

shifts away from a HREY enriched seawater pattern and to a diagenetic MREY enriched

pattern. The MREY enrichment has, however, also been found in diagenetically altered

foraminiferal tests (Haley et al., 2005), suggesting that calcareous material shows the same

diagenetically altered patterns as Ca phosphates. However, based on the correlations in

section 4.1. we refute the association of REY with carbonates in these cores.

The variations in LaSN/GdSN ratios and GdSN/YbSN ratios in core 122GC indicate a decoupling

between the LREY and the HREY because the ratio changes appear within different parts of

the core. LaSN/GdSN ratios were elevated (> 0.4) in the lower, Fe-rich part of the core, while

GdSN/YbSN ratios were highest (> 1.42) in the middle, Mn-rich part. The changes are primarily

associated with changes in Gd, i.e. MREY enrichment, but we nevertheless propose further

downcore fractionation between LREY and HREY, possibly due to different scavenging by Mn

oxides compared to Fe (oxyhydr)oxides. LREY are preferentially incorporated into Mn and Fe

phases (Koeppenkastrop and De Carlo, 1992), which is also seen by increasing LaSN/YbSN

ratios with depth.


141

4.4. CeSN anomaly

Another difference between the REYSN patterns found in cores from this study and REYSN

patterns in is constant with depth (approx. 0.3 in 117SL and approx. 0.43 in 122GC below

180 cm). The Ce concentration changes with depth together with the concentrations of the

trivalent REY and is not consistent as it was the case in cores from our previous study (Paul

et al., 2019). The constant CeSN anomaly suggests that Ce is bound to the same carrier phase

as its trivalent neighbors. The steady CeSN/CeSN* ratio in 117SL could, however, also be a

result of the core being much shorter than 69SL and that if it were longer, the CeSN/CeSN*

values would also decrease at the core base. Core 69SL displays stable CeSN/CeSN* ratios

between 2 and 9 m but lower values below 10 m (CeSN/CeSN* ~ 0.2). For 122GC, where

CeSN/CeSN* ratios are consistent in the Mn- and Fe-rich parts, sufficient Ce may be bound in

these phases, known to predominantly host Ce as a consequence of oxidative scavenging

(Bau and Koschinsky, 2009; De Carlo, 1991; Koeppenkastrop and De Carlo, 1992), so that the

concentration difference to the trivalent REY in the Ca phosphate phase is masked.

In all cores, the CeSN anomaly is smallest in the surface layers with CeSN/CeSN* ratios of 0.7 to

0.8. These values are similar to other cores from the eastern CCZ, where surface sediment

CeSN/CeSN* values were around 0.7, and values then decrease towards the west, with

decreasing particulate organic carbon flux (Toyoda et al., 1990; Paul et al., 2019; Volz et al.,

2018).

5. Conclusions

REYSN patterns with large negative CeSN anomalies and MREY enrichment prevail throughout

the three cores from the eastern CCZ, irrespective of sediment composition. Calcium-P and

Nd-P correlations confirm for two cores, that Ca phosphates are the likely REY carrier phase.

We therefore conclude that the REY are also bound to Ca phosphates that are diagenetically

altered by uptake of REY from ambient pore water, as has been similarly observed in other

regions of the central equatorial Pacific. Impacts of carbonate and metalliferous layers can,

however, clearly be seen in REY concentrations measured within the respective layers: Mn-

rich layers accumulate high REY concentrations while CaCO3 layers dilute REY

concentrations.

The comparison of these three cores showed the high spatial variability of deep-sea sediments

from the CCZ with respect to geochemical composition. This heterogeneity proves that a single

core is not representative of an area and can often not be fully explained on its own, but only


142

as part of a larger sample set. Then, patterns that are widely applicable, such as the Ca

phosphate control of REY, can be determined and verified.

Chapter 6 – Conclusions and Outlook

143


This cumulative PhD thesis focused on improving our levels of understanding of natural

geochemical variability in deep-sea sediments of the Pacific – solid phase and pore water –

and potential impacts of deep-sea mining on these sediments. Intraregional variability and

possible impacts on the seafloor were analyzed in the Peru Basin and compared with the CCZ

for interregional variability assessment. Clear heterogeneity can be seen in oxygen penetration

depths, POC contents, pore water metal concentrations, solid phase metal concentrations, and

REYSN patterns.

Small changes in bathymetry are sufficient to influence deposition and subsequent early

diagenesis, leading to small-scale variability in POC content, nitrate penetration depth, and

Fe(II) oxidation in the clay minerals in the Peru Basin (Chapter 2). Two of the seven GCs that

were analyzed from the Peru Basin showed layers rich in U, Mo, V, As (and Cu) in the solid

phase and pore water. These two cores were located in lower lying areas, suggesting that

bathymetry may have led to the deposition of different materials than at more elevated sites.

Such small-scale variability needs to be taken into account when sampling sediment cores and

using them to extrapolate results across larger areas, since the sampled core might not be as

representative as previously thought. Buried nodules that were found in the suboxic part of the

sediment cores are expected to impact the geochemical cycling as well, releasing metals such

as Mn and Co to the pore water during dissolution (Chapter 2). The buried nodules also locally

impact their surrounding sediment as they provide oxides in the suboxic environment and the

sediment in the vicinity of the nodule gets oxidized. The insights that could be gained from

dissolving nodules were based on few samples and should be further investigated in the future.

The quantification of metals released from buried nodules and the resulting impact on upward

metal cycling towards the oxic-suboxic boundary or even the sediment-water interface may

lead to a better understanding of the metal fluxes and contributions of pore water metals to the

oceanic metal budget.

The results from the heterogeneity analysis showed that a thorough analysis of prospective

mining areas and associated reference sites is necessary to (1) understand the regional

biogeochemical system and (2) to subsequently choose representative reference sites to

determine impacts of and to monitor regeneration after scientific disturbance experiments or

industrial deep-sea mining in the future (Chapters 2 and 3). The comparison of 26-year and

five-week old disturbance tracks with undisturbed sites in the Peru Basin showed that pore

water metals are still out of equilibrium after five weeks but have reached a new equilibrium

after 26 years, while solid phase metal distribution can still be impacted 26 years after physical


144

disturbance and different impacts are visible depending on the gear that was used to create

the disturbance (i.e. plow harrow or EBS). Amino acids and DOC, two indicators of organic

matter degradation, showed no significant differences between undisturbed and disturbed sites

after 26 years (Chapter 3).

With respect to the development of mining regulations, we propose Mn and O2 as key

parameters for monitoring of biogeochemical environmental impacts of deep-sea mining. As

could be demonstrated for the DISCOL area, Peru Basin, Mn and O2 are most useful in

determining changes to the redox zonation in the upper 20-30 cm, where mining impacts are

expected to occur (Chapter 3). Changes in O2 penetration depth are indicative of the behavior

of metals and the O2 penetration depth determines whether the metals are in the solid or

dissolved phase. Solid phase Mn concentrations clearly show how the upper Mn-oxide-rich

sediment layer has been impacted: sediment removal, sediment mixing, and sediment

redeposition could be distinguished (Chapter 3). Additionally, other cations such as Co, Mo,

Ni, and to a lesser extent Cu are bound to the Mn oxides in this surface layer, so that the

distribution of Mn allows us to draw conclusions about the distribution patterns of the

associated trace metals as well. This also shows the importance of properly conducted studies

of a natural system and how this helps to define key monitoring parameters. Dissolved Mn in

the pore water is a less suitable key parameter because only short-term (weeks to months)

impacts can be seen. Due to vertical and lateral diffusion of oxygen, the metals are removed

from the pore water by precipitation of oxides and sorption onto oxide surfaces (Chapter 3).

The “quick” re-equilibration of the pore water therefore shows a misleading picture of the entire

system, where the solid phase is still impacted for longer-time scales (decades to centuries).

A schematic drawing of these main parameters and their regeneration after a mining-like

disturbance is shown in Figure 1.


145

Figure 1: Schematic drawing of the behavior of key parameters (dissolved oxygen, solid phase

and pore water Mn) in surface sediments of the Peru Basin before and after a mining-like

disturbance.

Various metals, e.g., Cu and V, are enriched in the pore waters in the surface layer (approx.

2 cm) of the sediments due to the degradation of organic matter (Chapter 3). A disturbance of

the surface layer leads to a release of these elevated dissolved metal concentrations to the

bottom seawater and the re-establishment of these enrichments takes more than a few weeks.

We can hence conclude from these metal profiles how well the microbial processes that

degrade organic matter are recovering. Any release of dissolved metals from the pore water

due to a disturbance will be short-lived and the metals will be removed within the oxic seawater

or by diffusing oxygen in the pore water. It has been previously shown that metals released to

the oxic water column are quickly scavenged by particles (Koschinsky et al., 2003).

It was not possible, however, to determine exact values for the key parameters in the

framework of this PhD thesis project. More samples need to be analyzed to statistically define

robust values. We could show, however, that the system is heterogeneous on small- and large-

scales, in surface sediments and down to 10 m, so that extensive baseline research at each

site where mining is planned must be conducted (Paul et al., 2018; Paul et al., in prep.;

Chapters 2 and 3). From a policy perspective, it would be best to define key parameters that

are suitable for a wide range of geochemical conditions.

The impact studies carried out so far focused on relatively small areas (e.g., one track such as

an EBS track) or areas where multiple tracks were plowed but not in a coherent area (DISCOL

area, 1989 experiment). Therefore, lateral effects may still influence geochemical cycling since

e.g., oxygen also diffuses from the adjacent sediment into the disturbed site and not only

vertically from the overlying bottom water (personal communication, M. Haeckel). A larger

scale mining test such as the collector test planned in the CCZ for 2019, where an area of


146

approx. 300x300 m will be harvested, will help to reduce such lateral effects of dissolving

oxygen and other constituents into the experimental test area. As a next step in the

environmental assessment of impacts of deep-sea mining on the sediment, such a larger-scale

project including the collector test itself and environmental monitoring over several years will

help develop our understanding of potential industrial-scale impacts.

Besides further environmental impact studies of potential adverse impacts of deep-sea mining,

it is as pertinent to investigate other options of gaining metals or using less metals to reduce

the need for these new, deep mining sites. The deep-sea would only add a source for metals

to land-based reserves (Hein et al., 2013), to e.g., take off pressure from sensitive terrestrial

ecosystems such as the rainforest and Arctic regions. Before venturing into either of these

vulnerable ecosystems, the potential for recycling should be exhausted and the consumer

behavior of today’s society rethought, where people frequently buy new laptops, smartphones,

and cars that are constructed partially with non-renewable high-technology metals.

The REY were also used in this thesis project to compare small- and large-scale variability

within each of the study areas as well as between the CCZ and the Peru Basin due to their

suitability to reveal sediment origin as well as impacts of transport and alteration processes.

Differences could be seen within the CCZ, where the REY are predominantly bound to Ca

phosphates (Chapter 4). The REY show similar SN patterns across the central equatorial

Pacific (CCZ), with MREY enrichment and no or negative CeSN anomalies that develop or

increase with depth, but cores containing carbonate (Ca up to 30 wt.%) or metalliferous layers

deviate from the typical enrichment with depth. Carbonate layers dilute REY concentrations,

whereas layers rich in Mn and Fe partly accumulate REY or also dilute REY concentrations.

The Ca phosphate control typical for deep-sea sediments in this area is also not clear, shown

by lower Pearson R values for P-Nd correlations (Chapter 5). These findings substantiate the

idea that the deep-sea floor is heterogeneous on spatially large and small scales.

Sequential extractions proved that the trivalent REY are primarily associated with the Ca

phosphate phase, while Ce is associated with Mn and Fe phases (Chapter 4). Therefore, the

Ce concentration remains relatively constant with depth because Ce is oxidatively scavenged

by Mn and Fe (oxyhydr)oxides in the surface layer. As a result, the negative CeSN anomaly

develops or increases with depth due to increasing concentrations of the Ce neighboring

trivalent REY. The CeSN anomaly only does not change with depth in core 117SL, indicating

that Ce might not be decoupled from its trivalent neighbors and potentially sits in the same

phase as the other REY.


147

The MREY development at the sediment-water interface, the concentration increase with

depth, and the increasing CeSN anomaly with depth are signs that the seawater pattern is not

recorded in Ca phosphates in the sediments and that the REYSN patterns are clearly

diagenetically overprinted. Therefore, caution needs to be exercised when using Ca

phosphates, such as biogenic apatite and conodonts, as paleoceanographic proxies.

The solid phase and pore water REYSN patterns obtained from the Peru Basin display HREY

enrichment, concentration increase with depth, and negative CeSN as well as positive LaSN,

GdSN, and YSN anomalies (Chapter 2), thereby resembling the seawater REYSN pattern but not

the solid phase and pore water REYSN pattern from the CCZ. Nevertheless, phosphates may

play a role as a REY carrier phase, but nontronites should not be fully excluded either.

Even though we were only able to provide few reliable deep-sea pore water REYSN patterns

(Chapters 2 and 4), the data show that models drafted based on near-shore data (e.g., Haley

et al., 2004) might not be suitable for (oxic) deep-sea pore waters. Especially in the CCZ, oxic

pore water REYSN patterns show a MREY enrichment which corresponds with the sediment

REYSN patterns. This is, however, contrary to the interpretation of MREY enrichment on

continental margins, where this pattern was interpreted to be indicative of anoxic environments

(Haley et al., 2004). Still, the MREY bulge in the oxic sediment could be a sign of anoxic

diagenesis in the past (Haley et al., 2005).

For further work on pore water REY, which is crucial as it remains an understudied topic and

pore water is a potentially large source of REY to the seawater (Abbott et al., 2015), method

development is needed. Due to small sample volumes and the impeding salt matrix of pore

water samples, analyses are challenging. SeaFAST analysis similar to seawater REY analysis

(Hathorne et al., 2012) or pooling of samples and preconcentration using a column pre-

concentration method (Bau and Dulski, 1996b; Shabani et al., 1992) to separate the REY from

the salt matrix prior to ICP-MS measurements are possible approaches. This would be

important to get vertical concentration profiles of REY in deep-sea sediments and to determine

REYSN patterns in a higher depth resolution.

The deep-sea remains an understudied ecosystem where many processes and potential

anthropogenic impacts still need to be understood. This thesis helped, however, to fill some

gaps and provides thorough baseline data for (trace) elements in deep-sea sediments (Al, Ca,

Fe, Mn, Mo, Ni, Co, Cu, U, V, Cd, Pb, Zn, REY) and pore waters (Mn, Co, Cu, Mo, U, V, Cd,

As, REY). These are crucial to better understand the natural processes occurring in these

nodule provinces, to support informed environmental management decisions, and to be able

to identify, predict, and quantify anthropogenic impacts in the future.


148

Chapter 7 – Related scientific work

149


1. Research cruises and sampling campaigns

1.1. Research cruises

April-May 2018, M147

Five-week research cruise with RV METEOR (M147, Amazon GEOTRACES) in the Amazon

estuary, equatorial western Atlantic. Trace metal clean (TM) sampling of seawater with a TM-

CTD as well as sediment and pore water sampling with a multi-corer. Pore water was extracted

using a centrifuge and syringe filters. Selected seawater samples were ultrafiltered. Samples

were frozen or acidified for the transport and subsequent analyses in the home laboratories in

Germany.

July-October 2015, SO242 legs 1 and 2

Ten-week research cruise with RV SONNE (SO242/1 and 2; DISCOL revisited) to the DISCOL

experimental area in the Peru Basin, south east equatorial Pacific. Sampling of sediment, pore

water and seawater using gravity corers, multi-corers, ROV push cores, and CTD rosette. Pore

water was extracted using a centrifuge and syringe filter as well as rhizons. Samples were

frozen or acidified for the transport and subsequent analyses in the home laboratories in

Germany.

1.2. Land-based sampling campaigns

July 2017, Kangerlussuaq, Greenland

Two-week sampling campaign with Dr. Katja Schmidt for the project: Riverine fluxes of high

field strength elements Zr, Hf, Nb, Ta into the oceans and the impact on Hf isotopic composition

of seawater. Sampling of large volumes of river water from the Watson river and its tributaries

from the Russell and Leverett glaciers, the Kangerlussuaq fjord and a melt water lake on the

ice sheet in western Greenland. Temperature, conductivity, and pH were measured on site.

Samples were ultrafiltered and acidified for metal, isotope and DOC analyses.

February 2017, Luleå, Sweden

One-week sampling campaign with Dr. Katja Schmidt for the project: Riverine fluxes of high

field strength elements Zr, Hf, Nb, Ta into the oceans and the impact on Hf isotopic composition

of seawater. Sampling of large volumes of river water from the Råne and Kalix rivers in

northern Sweden. Temperature, conductivity, and pH were measured on site. Samples were

ultrafiltered and acidified for metal, isotope and DOC analyses.


150

2. Conferences

Paul, S. and Koschinsky, A. Sediment and pore water geochemistry in the deep-sea and

potential impacts of manganese nodule mining. Gordon Research Conference 2018, Hong

Kong, China, 08.-13.07.2018. Poster.

Paul, S. and Koschinsky, A. Sediment and pore water geochemistry in the deep-sea and

potential impacts of manganese nodule mining. Gordon Research Seminar 2018, Hong Kong,

China, 07.-08.07.2018. Talk and Poster.

Paul, S. and Koschinsky, A. Geochemical assessment of heavy metals in nodule mining

disturbance sites: implications for standard development and monitoring. 46th Underwater

Mining Conference (UMC), Berlin, Germany, 24.-29.09.2017. Talk.

Paul, S., Koschinsky, A., Gaye, B., and Daehnke, K. Long-term impacts of a manganese

nodule mining experiment on sediment and pore water geochemistry. Goldschmidt 2017,

Paris, France, 13.-18.08.2017. Talk.

Paul, S. and Koschinsky, A. 26-year Impact Assessment of Heavy Metal Distributions in

Sediment and Pore Water at a Deep-Sea Nodule Mining Disturbance Site. 14th International

Symposium on the Interaction between Sediments and Water (IASWS), Taormina, Italy, 17.-

22.06.2017. Talk.

Paul, S., Bau, M., Kuhn, T., Volz, J., Kasten, S., and Koschinsky, A. Controls on the

Distribution of Rare Earth Elements and Yttrium in Siliceous Ooze Rich Sediments from the

Pacific Ocean. ASLO 2017 Aquatic Sciences Meeting, Honolulu, Hawai’i, USA, 26.02.-

03.03.2017. Talk.

Post, J., Koschinsky, A. and Paul, S. Benthic Impact Experiment DISCOL in the manganese

nodule area of the Peru Basin – First results of a revisit after 26 years. Underwater Mining

Conference (UMC), St. Petersburg, Florida, USA, 01.-06.11.2015. Talk.


151

3. Co-supervised Bachelor thesis and guided research projects

Jessica Münch (2018) Working title: Elementverteilung in der Festphase im Amazonasdelta

(Major element distribution in the solid phase of the Amazon estuary). Bachelor Thesis.

University of Bremen.

Charlotte Opatz (2017) Sustainable Tourism and its development potential for Small Island

States – A case study of the Seychelles Islands. Bachelor Thesis. Jacobs University Bremen.

Marissa Menzel (2016) Geochemical Analysis of Deep-Sea Pore Water from the DISCOL

Experimental Area in the Peru Basin. Guided Research Project. Jacobs University Bremen.

Rajina Bajracharya (2016) A Geochemical Heterogeneity Analysis of Deep-Sea Sediments in

the DISCOL Experimental Area of the Peru Basin. Bachelor Thesis. Jacobs University Bremen.


152

Acknowledgements

153

Acknowledgements

First, thank you to my thesis committee members Andrea Koschinsky, Michael Bau, Sabine

Kasten, and Thomas Kuhn.

Especially, thank you to my PhD supervisor Andrea Koschinsky, who supported me every step

of the way, let me go on cruises, sampling trips, and conferences, and guided me through this

research project while always leaving space for my own ideas and interests. Thank you,

Andrea, for that!

Michael Bau, thanks for the interesting discussions about my REY data. Many concepts and

ideas were new to me but after a scientific conversation with you the world makes more sense.

I am looking forward to more discussions with you.

Thanks to my committee members Sabine Kasten and Thomas Kuhn, who were always there

for me with scientific advice, paper writing tips, and a lot of knowledge about the CCZ and

nodules.

Thanks also to my other co-authors: Jessica, Matthias, Birgit, Male, Inken, and Rajina for

providing fruitful discussions and input during paper writing.

I would also like to thank Autun Purser for proofreading parts of this thesis.

Thanks go to my work group at Jacobs: Luise, Kathy, Annika, Erika, Sandra, Ann, Franzi,

Charly, David, Dennis, Nadine, Ben, Dam, Song and our former members Inken, Adilah,

Daniela, Nathalie, Gila, Fumi, and Katja. Because of you, I liked coming to work and we had a

lot of fun during coffee and lunch breaks, christmas parties and summer BBQs. Special thanks

to Dennis and Kathy, for providing a lot of answers to OES and REE questions; to Annika, who

answered all my paranoid lab questions and still takes me to Werder Bremen soccer games;

and to Katja, who introduced me to the world of the geochemistry lab, all the techniques

including handling our MS, and took me on two extraordinary sampling trips to Sweden and

Greenland that took my mind off the deep-sea and sparked my interest in rocks and rivers.

I would also like to thank everyone who made the sampling cruises during my PhD time

unforgettable experiences: Inken, Tim, Seinab, Kristin, Manu, Jan, Anne, Autun, Yann, Lidia,

Lisa, Luise, Jessi, Juli, Rebecca, Patrick and many more.

Acknowledgements

154

During my time as a PhD student, I received a lot of support from GLOMAR – thanks for the

helpful courses, seminars, and conference funding.

Thanks to my friends in Bremen and all over the place as well as my family, who supported

me during this time and offered distraction when necessary. Thanks especially to my parents,

who always believe in me and gave me so much support during this thesis time and my whole

life before. Special thanks to my mom, for still picking up the phone when she sees my number

after all the calls I made during this thesis when I needed an open ear!

References

155

References

Abbott, A.N., Haley, B.A., McManus, J., 2016. The impact of sedimentary coatings on the diagenetic Nd flux. Earth Planet. Sci. Lett. 449, 217–227. https://doi.org/10.1016/j.epsl.2016.06.001

Abbott, A.N., Haley, B.A., McManus, J., Reimers, C.E., 2015. The sedimentary flux of dissolved rare earth elements to the ocean. Geochim. Cosmochim. Acta 154, 186–200. https://doi.org/10.1016/j.gca.2015.01.010

Alibo, D.S., Nozaki, Y., 1999. Rare earth elements in seawater: Particle association, shale-normalization, and Ce oxidation. Geochim. Cosmochim. Acta 63, 363–372. https://doi.org/10.1016/S0016-7037(98)00279-8

Alt, J.C., 1988. Hydrothermal oxide and nontronite deposits on seamounts in the eastern Pacific. Mar. Geol. 81, 227–239. https://doi.org/10.1016/0025-3227(88)90029-1

Anders, E., Grevesse, N., 1989. Abundances of the elements: Meteoritic and solar. Geochim. Cosmochim. Acta 53, 197–214. https://doi.org/10.1016/0016-7037(89)90286-X

Andreae, M.O., 1979. Arsenic speciation in seawater and interstitial waters: The influence of biological-chemical interactions on the chemistry of a trace element. Limnol. Oceanogr. 24, 440–452.

Andreae, M.O., 1977. Determination of Arsenic Species in Natural Waters. Anal. Chem. 49, 820–823. https://doi.org/10.1021/ac50014a037

Arndt, S., Jørgensen, B.B., LaRowe, D.E., Middelburg, J.J., Pancost, R.D., Regnier, P., 2013. Quantifying the degradation of organic matter in marine sediments: A review and synthesis. Earth-Science Rev. 123, 53–86. https://doi.org/10.1016/j.earscirev.2013.02.008

Auer, G., Reuter, M., Hauzenberger, C.A., Piller, W.E., 2017. The impact of transport processes on rare earth element patterns in marine authigenic and biogenic phosphates. Geochim. Cosmochim. Acta 203, 140–156. https://doi.org/10.1016/j.gca.2017.01.001

Auguste, M., Mestre, N.C., Rocha, T.L., Cardoso, C., Cueff-Gauchard, V., Le Bloa, S., Cambon-Bonavita, M.A., Shillito, B., Zbinden, M., Ravaux, J., Bebianno, M.J., 2016. Development of an ecotoxicological protocol for the deep-sea fauna using the hydrothermal vent shrimp Rimicaris exoculata. Aquat. Toxicol. 175, 277–285. https://doi.org/10.1016/j.aquatox.2016.03.024

Barbier, E.B., Moreno-Mateos, D., Rogers, A.D., Aronson, J., Pendleton, L., Danovaro, R., Henry, L.-A., Morato, T., Ardron, J., Van Dover, C.L., 2014. Protect the deep sea. Nature 505, 475–477.

Barckhausen, U., Bagge, M., Wilson, D.S., 2013. Seafloor spreading anomalies and crustal ages of the Clarion-Clipperton Zone. Mar. Geophys. Res. 34, 79–88. https://doi.org/10.1007/s11001-013-9184-6

Bau, M., 1991. Rare-earth element mobility during hydrothermal and metamorphic fluid-rock interaction and the significance of the oxidation state of europium. Chem. Geol. 93, 219–230. https://doi.org/10.1016/0009-2541(91)90115-8

Bau, M., Alexander, B., 2006. Preservation of primary REE patterns without Ce anomaly during dolomitization of Mid-Paleoproterozoic limestone and the potential re-establishment of marine anoxia immediately after the “Great Oxidation Event.” South African J. Geol. 109, 81–86. https://doi.org/10.2113/gssajg.109.1-2.81

References

156

Bau, M., Dulski, P., 1999. Comparing yttrium and rare earths in hydrothermal fluids from the Mid-Atlantic Ridge: implications for Y and REE behaviour during near-vent mixing and for the Y/Ho ratio of Proterozoic seawater. Chem. Geol. 155, 70–90.

Bau, M., Dulski, P., 1996a. Distribution of yttrium and rare-earth elements in the Penge and Kuruman iron-formations, Transvaal Supergroup, South Africa. Precambrian Res. 79, 37–55. https://doi.org/10.1016/0301-9268(95)00087-9

Bau, M., Dulski, P., 1996b. Anthropogenic origin of positive gadolinium anomalies in river waters. Earth Planet. Sci. Lett. 143, 245–255. https://doi.org/10.1016/0012-821X(96)00127-6

Bau, M., Dulski, P., 1994. Evolution of the yttrium-holmium systematics of seawater through time. Mineral. Mag. 58A, 61–62. https://doi.org/10.1180/minmag.1994.58A.1.35

Bau, M., Koschinsky, A., 2009. Oxidative scavenging of cerium on hydrous Fe oxide: Evidence from the distribution of rare earth elements and yttrium between Fe oxides and Mn oxides in hydrogenetic ferromanganese crusts. Geochem. J. 43, 37–47. https://doi.org/10.2343/geochemj.1.0005

Bau, M., Koschinsky, A., Dulski, P., Hein, J.R., 1996. Comparison of the partitioning behaviours of yttrium, rare earth elements, and titanium between hydrogenic marine ferromanganese crusts and seawater. Geochim. Cosmochim. Acta 60, 1709–1725.

Bau, M., Möller, P., Dulski, P., 1997. Yttrium and lanthanides in eastern Mediterranean seawater and their fractionation during redox-cycling. Mar. Chem. 56, 123–131. https://doi.org/10.1016/S0304-4203(96)00091-6

Bau, M., Schmidt, K., Pack, A., Bendel, V., Kraemer, D., 2018. The European Shale: An improved data set for normalisation of rare earth element and yttrium concentrations in environmental and biological samples from Europe. Appl. Geochemistry 90, 142–149. https://doi.org/10.1016/j.apgeochem.2018.01.008

Bau, M., Tepe, N., Mohwinkel, D., 2013. Siderophore-promoted transfer of rare earth elements and iron from volcanic ash into glacial meltwater, river and ocean water. Earth Planet. Sci. Lett. 364, 30–36. https://doi.org/10.1016/j.epsl.2013.01.002

Beck, M., Dellwig, O., Schnetger, B., Brumsack, H.J., 2008. Cycling of trace metals (Mn, Fe, Mo, U, V, Cr) in deep pore waters of intertidal flat sediments. Geochim. Cosmochim. Acta 72, 2822–2840. https://doi.org/10.1016/j.gca.2008.04.013

Berger, W.H., 1974. Deep-sea sedimentation, in: Burk, C.A., Drake, C.L. (Eds.), The Geology of Continental Margins. Springer, New York, pp. 213–241.

Berger, W.H., 1970. Planktonic Foraminifera: Selective solution and the lysocline. Mar. Geol. 8, 111–138. https://doi.org/10.1016/0025-3227(70)90001-0

Berger, W.H., Adelseck, C.G., Mayer, L.A., 1976. Distribution of carbonate in surface sediments of the Pacific Ocean. J. Geophys. Res. 81, 2617–2627. https://doi.org/10.1029/JC081i015p02617

Bernat, M., 1975. Les isotopes de l’uranium et du thorium et les terres rares dans l’environnement marin. Cah. ORSTOM Ser. Geol 7, 65–83.

Bertram, C.J., Elderfield, H., Aldridge, R.J., Conway Morris, S., 1992. 87Sr/86Sr, 143Nd/144Nd and REEs in Silurian phosphatic fossils. Earth Planet. Sci. Lett. 113, 239–249. https://doi.org/10.1016/0012-821X(92)90222-H

Bischoff, J.L., Heath, G.R., Leinen, M., 1979. Geochemistry of Deep-Sea Sediments from the Pacific Manganese Nodule Province: DOMES Sites A, B, and C, in: Bischoff, J.L., Piper, D.Z. (Eds.), Marine Geology and Oceanography of the Pacific Manganese Nodule

References

157

Province. Springer US, Boston, MA, pp. 397–436. https://doi.org/10.1007/978-1-4684-3518-4_12

Bischoff, J.L., Rosenbauer, R.J., 1977. Recent metalliferous sediment in the North Pacific manganese nodule area. Earth Planet. Sci. Lett. 33, 379–388. https://doi.org/10.1016/0012-821X(77)90089-9

Bluhm, H., 2001. Re-establishment of an abyssal megabenthic community after experimental physical disturbance of the seafloor. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3841–3868. https://doi.org/10.1016/S0967-0645(01)00070-4

Boetius, A., 2015. RV SONNE Fahrtbericht/Cruise Report SO242-2 JPI OCEANS Ecological Aspects of Deep-Sea Mining: DISCOL Revisited. https://doi.org/10.3289/GEOMAR_REP_NS_27_2015

Boudreau, B.P., 1997. Diagenetic models and their implementation: Modelling transport and reactions in aquatic sediments, Springer, New York. https://doi.org/0.I007/97S-3-642-60421-S

Bright, C.A., Cruse, A.M., Lyons, T.W., MacLeod, K.G., Glascock, M.D., Ethington, R.L., 2009. Seawater rare-earth element patterns preserved in apatite of Pennsylvanian conodonts? Geochim. Cosmochim. Acta 73, 1609–1624. https://doi.org/10.1016/j.gca.2008.12.014

Brockmeyer, B., Spitzy, A., 2013. Evaluation of a Disc Tube Methodology for Nano- and Ultrafiltration of Natural Dissolved Organic Matter (DOM). Int. J. Org. Chem. 3, 17–25. https://doi.org/10.4236/ijoc.2013.31A002

Brown, A., Thatje, S., Hauton, C., 2017a. The Effects of Temperature and Hydrostatic Pressure on Metal Toxicity: Insights into Toxicity in the Deep Sea. Environ. Sci. Technol. 51, 10222–10231. https://doi.org/10.1021/acs.est.7b02988

Brown, A., Wright, R., Mevenkamp, L., Hauton, C., 2017b. A comparative experimental approach to ecotoxicology in shallow-water and deep-sea holothurians suggests similar behavioural responses. Aquat. Toxicol. 191, 10–16. https://doi.org/10.1016/j.aquatox.2017.06.028

Burdige, D.J., 1993. The biogeochemistry of manganese and iron reduction in marine sediments. Earth-Science Rev. 35, 249–284. https://doi.org/10.1016/0012-8252(93)90040-E

Burdige, D.J., Komada, T., 2015. Sediment Pore Waters, in: Hansell, D.A., Carlson, C.A. (Eds.), Biogeochemistry of Marine Dissolved Organic Matter. Academic Press, Burlington, pp. 535–577. https://doi.org/10.1016/B978-012323841-2/50015-4

Burdige, D.J., Martens, C.S., 1990. Biogeochemical cycling in an organic-rich coastal marine basin: 11. The sedimentary cycling of dissolved, free amino acids. Geochim. Cosmochim. Acta 54, 3033–3052. https://doi.org/10.1016/0016-7037(90)90120-A

Cabon, J.Y., Cabon, N., 2000. Determination of arsenic species in seawater by flow injection hydride generation in situ collection followed by graphite furnace atomic absorption spectrometry Stability of As(III). Anal. Chim. Acta 418, 19–31. https://doi.org/10.1016/S0003-2670(00)00948-X

Cai, W.-J., Sayles, F.L., 1996. Oxygen penetration depths and fluxes in marine sediments. Mar. Chem. 52, 123–131. https://doi.org/10.1016/0304-4203(95)00081-X

Cantrell, K.J., Byrne, R.H., 1987. Rare earth element complexation by carbonate and oxalate ions. Geochim. Cosmochim. Acta 51, 597–605. https://doi.org/10.1016/0016-7037(87)90072-X

Chen, J., Algeo, T.J., Zhao, L., Chen, Z.-Q., Cao, L., Zhang, L., Li, Y., 2015. Diagenetic uptake

References

158

of rare earth elements by bioapatite, with an example from Lower Triassic conodonts of South China. Earth-Science Rev. 149, 181–202. https://doi.org/10.1016/j.earscirev.2015.01.013

Cole, T.G., Shaw, H.F., 1983. The nature and origin of authigenic smectites in some recent marine sediments. Clay Miner. 18, 239–252. https://doi.org/10.1180/claymin.1983.018.3.02

Colley, S., Thomson, J., 1985. Recurrent uranium relocations in distal turbidites emplaced in pelagic conditions. Geochim. Cosmochim. Acta 49, 2339–2348. https://doi.org/10.1016/0016-7037(85)90234-0

Colley, S., Thomson, J., Wilson, T.R.S., Higgs, N.C., 1984. Post-depositional migration of elements during diagenesis in brown clay and turbidite sequences in the North East Atlantic. Geochim. Cosmochim. Acta 48, 1223–1235. https://doi.org/10.1016/0016-7037(84)90057-7

Craig, J.D., 1979. The relationship between bathymetry and ferromanganese deposits in the north equatorial Pacific. Mar. Geol. 29, 165–186. https://doi.org/10.1016/0025-3227(79)90107-5

Cronan, D.S., Rothwell, G., Croudace, I., 2010. An ITRAX geochemical study of ferromanganiferous sediments from the Penrhyn Basin, South Pacific Ocean. Mar. Georesources Geotechnol. 28, 207–221. https://doi.org/10.1080/1064119X.2010.483001

Cullers, R.L., Chaudhuri, S., Arnold, B., Lee, M., Wolf, C.W., 1975. Rare earth distributions in clay minerals and in the clay-sized fraction of the Lower Permian Havensville and Eskridge shales of Kansas and Oklahoma. Geochim. Cosmochim. Acta 39, 1691–1703. https://doi.org/10.1016/0016-7037(75)90090-3

Davies, A.J., Roberts, J.M., Hall-Spencer, J., 2007. Preserving deep-sea natural heritage: Emerging issues in offshore conservation and management. Biol. Conserv. 138, 299–312. https://doi.org/10.1016/j.biocon.2007.05.011

Davis, J., Kaiser, K., Benner, R., 2009. Amino acid and amino sugar yields and compositions as indicators of dissolved organic matter diagenesis. Org. Geochem. 40, 343–352. https://doi.org/10.1016/j.orggeochem.2008.12.003

de Baar, H.J.W., Bacon, M.P., Brewer, P.G., Bruland, K.W., 1985. Rare earth elements in the Pacific and Atlantic Oceans. Geochim. Cosmochim. Acta 49, 1943–1959. https://doi.org/10.1016/0016-7037(85)90089-4

de Baar, H.J.W., German, C.R., Elderfield, H., van Gaans, P., 1988. Rare earth element distributions in anoxic waters of the Cariaco Trench. Geochim. Cosmochim. Acta 52, 1203–1219. https://doi.org/10.1016/0016-7037(88)90275-X

De Carlo, E.H., 1991. Paleoceanographic implications of rare earth element variability within a Fe-Mn crust from the central Pacific Ocean. Mar. Geol. 98, 449–467. https://doi.org/10.1016/0025-3227(91)90116-L

de Stigter, H., 2015. Gravity core descriptions, in: Greinert, J. (Ed.), RV SONNE Fahrtbericht/Cruise Report SO242-1 JPI OCEANS Ecological Aspects of Deep-Sea Mining: DISCOL Revisited. GEOMAR Helmholtz Centre for Ocean Research Kiel. https://doi.org/10.3289/GEOMAR_REP_NS_26_2015

Deng, Y., Ren, J., Guo, Q., Cao, J., Wang, H., Liu, C., 2017. Rare earth element geochemistry characteristics of seawater and porewater from deep sea in western Pacific. Sci. Rep. 7, 1–13. https://doi.org/10.1038/s41598-017-16379-1

Ding, X., Henrichs, S.M., 2002. Adsorption and desorption of proteins and polyamino acids by clay minerals and marine sediments. Mar. Chem. 77, 225–237.

References

159

https://doi.org/10.1016/S0304-4203(01)00085-8

Drodt, M., Trautwein, A.X., König, I., Suess, E., Bender Koch, C., 1997. Mössbauer spectroscopic studies on the iron forms of deep-sea sediments. Phys. Chem. Miner. 24, 281–293. https://doi.org/10.1007/s002690050040

Du, J., Haley, B.A., Mix, A.C., 2016. Neodymium isotopes in authigenic phases, bottom waters and detrital sediments in the Gulf of Alaska and their implications for paleo-circulation reconstruction. Geochim. Cosmochim. Acta 193, 14–35. https://doi.org/10.1016/j.gca.2016.08.005

Dymond, J., Suess, E., Lyle, M., 1992. Barium in Deep-Sea Sediment: A Geochemical Proxy for Paleoproductivity. Paleoceanography 7, 163–181. https://doi.org/10.1029/92PA00181

Eittreim, S.L., Ragozin, N., Gnibidenko, H.S., Helsley, C.E., 1992. Crustal age between the Clipperton and Clarion Fracture Zones. Geophys. Res. Lett. 12, 2365–2368.

Elbaz-Poulichet, F., Nagy, A., Cserny, T., 1997. The distribution of redox sensitive elements (U, As, Sb, V and Mo) along a river-wetland-lake system (Balaton Region, Hungary). Aquat. Geochemistry 3, 267–282. https://doi.org/10.1023/A:1009616214030

Elderfield, H., Greaves, M.J., 1982. The rare earth elements in seawater. Nature 296, 214–219. https://doi.org/10.1038/296214a0

Elderfield, H., Hawkesworth, C.J., Greaves, M.J., Calvert, S.E., 1981. Rare earth element geochemistry of oceanic ferromanganese nodules and associated sediments. Geochim. Cosmochim. Acta 45, 513–528. https://doi.org/10.1016/0016-7037(81)90184-8

Elderfield, H., Pagett, R., 1986. Rare earth elements in ichthyoliths: variations with redox conditions and depositional environment. Sci. Total Environ. 49, 175–197. https://doi.org/10.1016/0048-9697(86)90239-1

Elderfield, H., Sholkovitz, E.R., 1987. Rare earth elements in the pore waters of reducing nearshore sediments. Earth Planet. Sci. Lett. 82, 280–288. https://doi.org/10.1016/0012-821X(87)90202-0

Elderfield, H., Whitfield, M., Burton, J.D., Bacon, M.P., Liss, P.S., 1988. The Oceanic Chemistry of the Rare-Earth Elements. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 325, 105–126. https://doi.org/10.1098/rsta.1988.0046

Emerson, S.R., Huested, S.S., 1991. Ocean anoxia and the concentrations of molybdenum and vanadium in seawater. Mar. Chem. 34, 177–196. https://doi.org/10.1016/0304-4203(91)90002-E

Epstein, A.G., Epstein, J.B., Harris, L.D., 1977. Conodont color alteration: An index to organic metamorphism, Geological Survey Professional Paper. https://doi.org/10.3133/pp995

Filippelli, G.M., Delaney, M.L., 1996. Phosphorus geochemistry of equatorial Pacific sediments. Geochim. Cosmochim. Acta 60, 1479–1495. https://doi.org/10.1016/0016-7037(96)00042-7

Fritsche, U., Koschinsky, A., Winkler, A., 2001. The different diffusive transport behaviours of some metals in layers of Peru Basin surface sediment. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3653–3681. https://doi.org/10.1016/S0967-0645(01)00061-3

Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N. a., Heath, G.R., Cullen, D., Dauphin, P., Hammond, D., Hartman, B., Maynard, V., 1979. Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis. Geochim. Cosmochim. Acta 43, 1075–1090. https://doi.org/10.1016/0016-7037(79)90095-4

Fütterer, D.K., 2006. The Solid Phase of Marine Sediments, in: Schulz, H.D., Zabel, M. (Eds.),

References

160

Marine Geochemistry. Springer, Berlin, Heidelberg, pp. 1–25.

Gaye, B., Nagel, B., Dähnke, K., Rixen, T., Lahajnar, N., Emeis, K.C., 2013. Amino acid composition and δ15N of suspended matter in the Arabian Sea: Implications for organic matter sources and degradation. Biogeosciences 10, 7689–7702. https://doi.org/10.5194/bg-10-7689-2013

GEOMAR, 2017. JPI Oceans - Ecological Aspects of Deep-Sea Mining [WWW Document]. URL https://jpio-miningimpact.geomar.de/ (accessed 9.15.17).

German, C.R., Klinkhammer, G.P., Edmond, J.M., Mitra, A., Elderfield, H., 1990. Hydrothermal scavenging of rare-earth elements in the ocean. Nature 345, 516–518. https://doi.org/10.1038/345516a0

Gingele, F.X., Kasten, S., 1994. Solid-phase manganese in Southeast Atlantic sediments: Implications for the paleoenvironment. Mar. Geol. 121, 317–332. https://doi.org/10.1016/0025-3227(94)90037-X

Gingele, F.X., Zabel, M., Kasten, S., Bonn, W.J., Nürnberg, C.C., 1999. Biogenic Barium as a Proxy for Paleoproductivity: Methods and Limitations of Application, in: Fischer, G., Wefer, G. (Eds.), Use of Proxies in Palaeoceanography: Examples from the South Atlantic. Springer-Verlag, Berlin, Heidelberg, pp. 345–364.

Glasby, G.P., Gwozdz, R., Kunzendorf, H., Friedrich, G., Thijssen, T., 1987. The distribution of rare earth and minor elements in manganese nodules and sediments from the equatorial and S.W. Pacific. Lithos 20, 97–113. https://doi.org/10.1016/0024-4937(87)90001-6

Glasby, G.P., Stoffers, P., Sioulas, A., Thijssen, T., Friedrich, G., 1982. Manganese nodule formation in the Pacific Ocean: a general theory. Geo-Marine Lett. 2, 47–53. https://doi.org/10.1007/BF02462799

Glover, A.G., Smith, C.R., 2003. The deep-sea floor ecosystem: current status and prospects of anthropogenic change by the year 2025. Environ. Conserv. 30, 219–241. https://doi.org/10.1017/S0376892903000225

Goldberg, E.D., Koide, M., Schmitt, R.A., Smith, R.H., 1963. Rare-Earth distributions in the marine environment. J. Geophys. Res. 68, 4209–4217.

Gollner, S., Kaiser, S., Menzel, L., Jones, D.O.B., Brown, A., Mestre, N.C., van Oevelen, D., Menot, L., Colaço, A., Canals, M., Cuvelier, D., Durden, J.M., Gebruk, A., Egho, G.A., Haeckel, M., Marcon, Y., Mevenkamp, L., Morato, T., Pham, C.K., Purser, A., Sanchez-Vidal, A., Vanreusel, A., Vink, A., Martinez Arbizu, P., 2017. Resilience of benthic deep-sea fauna to mining activities. Mar. Environ. Res. 129, 76–101. https://doi.org/10.1016/j.marenvres.2017.04.010

Grandjean, P., Albarède, F., 1989. Ion probe measurement of rare earth elements in biogenic phosphates. Geochim. Cosmochim. Acta 53, 3179–3183. https://doi.org/10.1016/0016-7037(89)90097-5

Grasshoff, K., Kremling, K., Ehrhardt, M., 1999. Methods of Seawater Analysis. Wiley-VCH, Weinheim.

Greinert, J., 2015. RV SONNE Fahrtbericht/Cruise Report SO242-1 JPI OCEANS Ecological Aspects of Deep-Sea Mining: DISCOL Revisited. https://doi.org/10.3289/GEOMAR_REP_NS_26_2015

Guichard, F., Church, T.M., Treuil, M., Jaffrezic, H., 1979. Rare earths in barites: distribution and effects on aqueous partitioning. Geochim. Cosmochim. Acta 43, 983–997. https://doi.org/10.1016/0016-7037(79)90088-7

References

161

Haeckel, M., König, I., Riech, V., Weber, M.E., Suess, E., 2001. Pore water profiles and numerical modelling of biogeochemical processes in Peru Basin deep-sea sediments . Deep. Res. Part II Top. Stud. Oceanogr. 48, 3713–3736. https://doi.org/10.1016/S0967-0645(01)00064-9

Halbach, P., Scherhag, C., Hebisch, U., Marchig, V., 1981. Geochemical and mineralogical control of different genetic types of deep sea nodules from the Pacific Ocean. Miner. Depos. 16, 59–84.

Haley, B.A., Klinkhammer, G.P., McManus, J., 2004. Rare earth elements in pore waters of marine sediments. Geochim. Cosmochim. Acta 68, 1265–1279. https://doi.org/10.1016/j.gca.2003.09.012

Haley, B.A., Klinkhammer, G.P., Mix, A.C., 2005. Revisiting the rare earth elements in foraminiferal tests. Earth Planet. Sci. Lett. 239, 79–97. https://doi.org/10.1016/j.epsl.2005.08.014

Halfar, J., Fujita, R.M., 2002. Precautionary management of deep-sea mining. Mar. Policy 26, 103–106. https://doi.org/10.1016/S0308-597X(01)00041-0

Hathorne, E.C., Haley, B., Stichel, T., Grasse, P., Zieringer, M., Frank, M., 2012. Online preconcentration ICP-MS analysis of rare earth elements in seawater. Geochemistry, Geophys. Geosystems 13. https://doi.org/10.1029/2011GC003907

Heggie, D., Kahn, D., Fischer, K., 1986. Trace metals in metalliferous sediments, MANOP Site M: interfacial pore water profiles. Earth Planet. Sci. Lett. 80, 106–116. https://doi.org/10.1016/0012-821X(86)90023-3

Heggie, D., Lewis, T., 1984. Cobalt in pore waters of marine sediments. Nature 311, 453–455. https://doi.org/10.1038/311453a0

Hein, J.R., Koschinsky, A., 2014. Deep-Ocean Ferromanganese Crusts and Nodules, in: Treatise on Geochemistry. Elsevier, pp. 273–291. https://doi.org/10.1016/B978-0-08-095975-7.01111-6

Hein, J.R., Mizell, K., Koschinsky, A., Conrad, T.A., 2013. Deep-ocean mineral deposits as a source of critical metals for high- and green-technology applications: Comparison with land-based resources. Ore Geol. Rev. 51, 1–14. https://doi.org/10.1016/j.oregeorev.2012.12.001

Heller, C., Kuhn, T., Versteegh, G.J.M., Wegorzewski, A. V., Kasten, S., 2018. The geochemical behavior of metals during early diagenetic alteration of buried manganese nodules. Deep. Res. Part I Oceanogr. Res. Pap. 142, 16–33. https://doi.org/10.1016/j.dsr.2018.09.008

https://jpio-miningimpact.geomar.de, n.d. JPI Oceans - Ecological Aspects of Deep-Sea Mining [WWW Document]. URL https://jpio-miningimpact.geomar.de/ (accessed 9.15.17).

International Seabed Authority, 2018. Deep Seabed Minerals Contractors [WWW Document]. URL https://www.isa.org.jm/deep-seabed-minerals-contractors (accessed 7.17.18).

International Seabed Authority, 2014. Deep Seabed Minerals Contractors [WWW Document]. URL https://www.isa.org.jm/deep-seabed-minerals-contractors (accessed 4.22.15).

International Seabed Authority, 2010. A Geological Model of Polymetallic Nodule Deposits in the Clarion Clipperton Fracture Zone. Kingston, Jamaica.

Ittekkot, V., Degens, E., 1984. The role of organic matter in the Wadden Sea - Tracking of marine dissolved and particulate organic carbon - environmental implications, in: Netherlands Institute for Sea Research Publication Series, 10. pp. 179–194.

Jahnke, R.A., Jackson, G.A., 1992. The Spatial Distribution of Sea Floor Oxygen Consumption

References

162

in The Atlantic and Pacific Oceans, in: Rowe, G.T., Pariente, V. (Eds.), Deep-Sea Food Chains and the Global Carbon Cycle. Springer Netherlands, Dordrecht, pp. 295–307. https://doi.org/10.1007/978-94-011-2452-2_18

Jamieson, A., 2015. The Hadal Zone: Life in the Deepest Oceans. Cambridge University Press, Cambridge. https://doi.org/10.1017/CBO9781139061384

Jarvis, I., Burnett, W.C., Nathan, Y., Almbaydin, F.S.M., Attia, A.K.M., Castro, L.N., Flicoteaux, R., Hilmy, M.E., Husain, V., Qutawnah, A.A., Serjani, A., Zanin, Y.N., 1994. Phosphorite geochemistry: State-of-the-art and environmental concerns. Eclogae Geol. Helv. 87, 643–700.

Johannesson, K.H., Hawkins, D.L., Cortés, A., 2006. Do Archean chemical sediments record ancient seawater rare earth element patterns? Geochim. Cosmochim. Acta 70, 871–890. https://doi.org/10.1016/j.gca.2005.10.013

Johnson, D.A., 1972. Ocean-floor erosion in the equatorial Pacific. Bull. Geol. Soc. Am. 83, 3121–3144. https://doi.org/10.1130/0016-7606(1972)83[3121:OEITEP]2.0.CO;2

Jones, D.O.B., Kaiser, S., Sweetman, A.K., Smith, C.R., Menot, L., Vink, A., Trueblood, D., Greinert, J., Billett, D.S.M., Martinez Arbizu, P., Radziejewska, T., Singh, R., Ingole, B., Stratmann, T., Simon-Lledó, E., Durden, J.M., Clark, M.R., 2017. Biological responses to disturbance from simulated deep-sea polymetallic nodule mining. PLoS One 12, e0171750. https://doi.org/10.1371/journal.pone.0171750

Kashiwabara, T., Toda, R., Nakamura, K., Yasukawa, K., Fujinaga, K., Kubo, S., Nozaki, T., Takahashi, Y., Suzuki, K., Kato, Y., 2018. Synchrotron X-ray spectroscopic perspective on the formation mechanism of REY-rich muds in the Pacific Ocean. Geochim. Cosmochim. Acta 240, 274–292. https://doi.org/10.1016/j.gca.2018.08.013

Kasten, S., Glasby, G.P., Schulz, H.D., Friedrich, G., Andreev, S.I., 1998. Rare earth elements in manganese nodules from the South Atlantic Ocean as indicators of oceanic bottom water flow. Mar. Geol. 146, 33–52. https://doi.org/10.1016/S0025-3227(97)00128-X

Kato, Y., Fujinaga, K., Nakamura, K., Takaya, Y., Kitamura, K., Ohta, J., Toda, R., Nakashima, T., Iwamori, H., 2011. Deep-sea mud in the Pacific Ocean as a potential resource for rare-earth elements. Nat. Geosci. 4, 535–539. https://doi.org/10.1038/ngeo1185

Klewicki, J.K., Morgan, J.J., 1998. Kinetic behavior of Mn(III) complexes of pyrophosphate, EDTA, and citrate. Environ. Sci. Technol. 32, 2916–2922. https://doi.org/10.1021/es980308e

Klinkhammer, G., Heggie, D.T., Graham, D.W., 1982. Metal diagenesis in oxic marine sediments. Earth Planet. Sci. Lett. 61, 211–219. https://doi.org/10.1016/0012-821X(82)90054-1

Klinkhammer, G.P., 1980. Early diagenesis in sediments from the eastern equatorial Pacific, II. Pore water metal results. Earth Planet. Sci. Lett. 49, 81–101. https://doi.org/10.1016/0012-821X(80)90151-X

Koeppenkastrop, D., De Carlo, E.H., 1992. Sorption of rare-earth elements from seawater onto synthetic mineral particles: An experimental approach. Chem. Geol. 95, 251–263. https://doi.org/10.1016/0009-2541(92)90015-W

Kon, Y., Hoshino, M., Sanematsu, K., Morita, S., Tsunematsu, M., Okamoto, N., Yano, N., Tanaka, M., Takagi, T., 2014. Geochemical characteristics of apatite in heavy REE-rich Deep-Sea Mud from Minami-Torishima Area, Southeastern Japan. Resour. Geol. 64, 47–57. https://doi.org/10.1111/rge.12026

König, I., Drodt, M., Suess, E., Trautwein, A.X., 1997. Iron reduction through the tan-green color transition in deep-sea sediments. Geochim. Cosmochim. Acta 61, 1679–1683.

References

163

https://doi.org/10.1016/S0016-7037(97)00007-0

König, I., Haeckel, M., Drodt, M., Suess, E., Trautwein, A.X., 1999. Reactive Fe(II) layers in deep-sea sediments. Geochim. Cosmochim. Acta 63, 1517–1526. https://doi.org/10.1016/S0016-7037(99)00104-0

König, I., Haeckel, M., Lougear, A., Suess, E., Trautwein, A.X., 2001. A geochemical model of the Peru Basin deep-sea floor - and the response of the system to technical impacts. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3737–3756. https://doi.org/10.1016/S0967-0645(01)00065-0

Koschinsky, A., 2001. Heavy metal distributions in Peru Basin surface sediments in relation to historic, present and disturbed redox environments. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3757–3777. https://doi.org/10.1016/S0967-0645(01)00066-2

Koschinsky, A., Fritsche, U., Winkler, A., 2001a. Sequential leaching of Peru Basin surface sediment for the assessment of aged and fresh heavy metal associations and mobility. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3683–3699. https://doi.org/10.1016/S0967-0645(01)00062-5

Koschinsky, A., Gaye-Haake, B., Arndt, C., Maue, G., Spitzy, A., Winkler, A., Halbach, P., 2001b. Experiments on the influence of sediment disturbances on the biogeochemistry of the deep-sea environment. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3629–3651. https://doi.org/10.1016/S0967-0645(01)00060-1

Koschinsky, A., Halbach, P., 1995. Sequential leaching of marine ferromanganese precipitates: Genetic implications. Geochim. Cosmochim. Acta 59, 5113–5132. https://doi.org/10.1016/0016-7037(95)00358-4

Koschinsky, A., Heinrich, L., Boehnke, K., Cohrs, J.C., Markus, T., Shani, M., Singh, P., Smith Stegen, K., Werner, W., 2018. Deep-sea mining: Interdisciplinary research on potential environmental, legal, economic, and societal implications. Integr. Environ. Assess. Manag. 14, 672–691. https://doi.org/10.1002/ieam.4071

Koschinsky, A., Winkler, A., Fritsche, U., 2003. Importance of different types of marine particles for the scavenging of heavy metals in the deep-sea bottom water. Appl. Geochemistry 18, 693–710. https://doi.org/10.1016/S0883-2927(02)00161-0

Köster, M., 2017. Reactivity and spatial variation of Mn and Fe mineral phases in sediments of the eastern Clarion- Clipperton Zone, Pacific Ocean. Universität Bremen.

Kowalski, N., Dellwig, O., Beck, M., Grunwald, M., Fischer, S., Piepho, M., Riedel, T., Freund, H., Brumsack, H.J., Böttcher, M.E., 2009. Trace metal dynamics in the water column and pore waters in a temperate tidal system: response to the fate of algae-derived organic matter. Ocean Dyn. 59, 333–350. https://doi.org/10.1007/s10236-009-0192-7

Kraemer, D., Kopf, S., Bau, M., 2015. Oxidative mobilization of cerium and uranium and enhanced release of “immobile” high field strength elements from igneous rocks in the presence of the biogenic siderophore desferrioxamine B. Geochim. Cosmochim. Acta 165, 263–279. https://doi.org/10.1016/j.gca.2015.05.046

Ku, T.-L., Knauss, K.G., Mathieu, G.G., 1977. Uranium in open ocean: concentration and isotopic composition. Deep Sea Res. 24, 1005–1017. https://doi.org/10.1016/0146-6291(77)90571-9

Kuhn, G., 2013. Don’t forget the salty soup : Calculations for bulk marine geochemistry and radionuclide geochronology, in: Goldschmidt 2013 Conference Abstracts. Florence, Italy, p. 1519. https://doi.org/10.1180/minmag.2013.077.5.11

Kuhn, T., 2015. RV SONNE SO240 Cruise Report/Fahrtbericht. https://doi.org/10.2312/cr_so240

References

164

Kuhn, T., Versteegh, G.J.M., Villinger, H., Dohrmann, I., Heller, C., Koschinsky, A., Kaul, N., Ritter, S., Wegorzewski, A. V., Kasten, S., 2017. Widespread seawater circulation in 18-22 Ma oceanic crust: Impact on heat flow and sediment geochemistry. Geology 45, 799–802. https://doi.org/10.1130/G39091.1

Kulaksiz, S., Bau, M., 2013. Anthropogenic dissolved and colloid/nanoparticle-bound samarium, lanthanum and gadolinium in the Rhine River and the impending destruction of the natural rare earth element distribution in rivers. Earth Planet. Sci. Lett. 362, 43–50. https://doi.org/10.1016/j.epsl.2012.11.033

Lahajnar, N., Rixen, T., Gaye-Haake, B., Schäfer, P., Ittekkot, V., 2005. Dissolved organic carbon (DOC) fluxes of deep-sea sediments from the Arabian Sea and NE Atlantic. Deep. Res. Part II Top. Stud. Oceanogr. 52, 1947–1964. https://doi.org/10.1016/j.dsr2.2005.05.006

Lallier, L.E., Maes, F., 2016. Environmental impact assessment procedure for deep seabed mining in the area: Independent expert review and public participation. Mar. Policy 70, 212–219. https://doi.org/10.1016/j.marpol.2016.03.007

Lee, C., 1992. Controls on organic carbon preservation: The use of stratified water bodies to compare intrinsic rates of decomposition in oxic and anoxic systems. Geochim. Cosmochim. Acta 56, 3323–3335. https://doi.org/10.1016/0016-7037(92)90308-6

Luther, G.W., 2005. Manganese(II) Oxidation and Mn(IV) Reduction in the Environment—Two One-Electron Transfer Steps Versus a Single Two-Electron Step. Geomicrobiol. J. 22, 195–203. https://doi.org/10.1080/01490450590946022

Luther, G.W., Sundby, B., Lewis, B.L., Brendel, P.J., Silverberg, N., 1997. Interactions of manganese with the nitrogen cycle: Alternative pathways to dinitrogen. Geochim. Cosmochim. Acta 61, 4043–4052. https://doi.org/10.1016/S0016-7037(97)00239-1

Lutz, M.J., Caldeira, K., Dunbar, R.B., Behrenfeld, M.J., 2007. Seasonal rhythms of net primary production and particulate organic carbon flux to depth describe the efficiency of biological pump in the global ocean. J. Geophys. Res. Ocean. 112. https://doi.org/10.1029/2006JC003706

Lyle, M., 2003. Neogene carbonate burial in the Pacific Ocean. Paleoceanography 18. https://doi.org/10.1029/2002PA000777

Lyle, M., 1983. The brown-green color transition in marine sediments: A marker of the Fe(III)-Fe(II) redox boundary. Limnol. Oceanogr. 28, 1026–1033. https://doi.org/10.4319/lo.1983.28.5.1026

Lyons, T.W., Werne, J.P., Hollander, D.J., Murray, R.., 2003. Contrasting sulfur geochemistry and Fe/Al and Mo/Al ratios across the last oxic-to-anoxic transition in the Cariaco Basin, Venezuela. Chem. Geol. 195, 131–157. https://doi.org/10.1016/S0009-2541(02)00392-3

Madison, A.S., Tebo, B.M., Luther, G.W., 2011. Simultaneous determination of soluble manganese(III), manganese(II) and total manganese in natural (pore)waters. Talanta 84, 374–381. https://doi.org/10.1016/j.talanta.2011.01.025

Madison, A.S., Tebo, B.M., Mucci, A., Sundby, B., Luther, G.W., 2013. Abundant porewater Mn(III) is a major component of the sedimentary redox system. Science. 341, 875–878. https://doi.org/10.1126/science.1241396

Marchig, V., Von Stackelberg, U., Hufnagel, H., Durn, G., 2001. Compositional changes of surface sediments and variability of manganese nodules in the Peru Basin. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3523–3547. https://doi.org/10.1016/S0967-0645(01)00055-8

Marchig, V., von Stackelberg, U., Wiedicke, M., Durn, G., Milovanovic, D., 1999. Hydrothermal

References

165

activity associated with off-axis volcanism in the Peru Basin. Mar. Geol. 159, 179–203.

Martínez Arbizu, P., Haeckel, M., 2015. RV SONNE Fahrtbericht/Cruise Report SO239 EcoResponse Assessing the Ecology, Connectivity and Resilience of Polymetallic Nodule Field Systems. https://doi.org/10.3289/GEOMAR_REP_NS_25_2015

Martins, I., Goulart, J., Martins, E., Morales-Román, R., Marín, S., Riou, V., Colaço, A., Bettencourt, R., 2017. Physiological impacts of acute Cu exposure on deep-sea vent mussel Bathymodiolus azoricus under a deep-sea mining activity scenario. Aquat. Toxicol. 193, 40–49. https://doi.org/10.1016/j.aquatox.2017.10.004

Mascarenhas-Pereira, M.B.L., Nath, B.N., 2010. Selective leaching studies of sediments from a seamount flank in the Central Indian Basin: Resolving hydrothermal, volcanogenic and terrigenous sources using major, trace and rare-earth elements. Mar. Chem. 121, 49–66. https://doi.org/10.1016/j.marchem.2010.03.004

McLennan, S.M., 1989. Rare Earth Elements in Sedimentary Rocks: Influence of Provenance and Sedimentary Processes, in: Lipin, B.R., McKay, G.A. (Eds.), Geochemistry and Mineralogy of Rare Earth Elements, MSA Reviews in Mineralogy. pp. 169–200.

McManus, J., Berelson, W.M., Klinkhammer, G.P., Johnson, K.S., Coale, K.H., Anderson, R.F., Kumar, N., Burdige, D.J., Hammond, D.E., Brumsack, H.J., McCorkle, D.C., Rushdi, A., 1998. Geochemistry of barium in marine sediments: implications for its use as a paleoproxy. Geochim. Cosmochim. Acta 62, 3453–3473. https://doi.org/10.1016/S0016-7037(98)00248-8

Mengerink, K.J., Van Dover, C.L., Ardron, J., Baker, M., Escobar-Briones, E., Gjerde, K., Koslow, J.A., Ramirez-Llodra, E., Lara-Lopez, A., Squires, D., Sutton, T., Sweetman, A.K., Levin, L.A., 2014. A Call for Deep-Ocean Stewardship. Science. 344, 696–698. https://doi.org/10.1126/science.1251458

Merschel, G., 2017. Trace Element and Isotope Geochemistry of Particle- Reactive Elements in River Waters of the Amazon River Basin. Jacobs University Bremen.

Merschel, G., Bau, M., Schmidt, K., Münker, C., Dantas, E.L., 2017. Hafnium and neodymium isotopes and REY distribution in the truly dissolved, nanoparticulate/colloidal and suspended loads of rivers in the Amazon Basin, Brazil. Geochim. Cosmochim. Acta 213, 383–399. https://doi.org/10.1016/j.gca.2017.07.006

Mevenkamp, L., Brown, A., Hauton, C., Kordas, A., Thatje, S., Vanreusel, A., 2017. Hydrostatic pressure and temperature affect the tolerance of the free-living marine nematode Halomonhystera disjuncta to acute copper exposure. Aquat. Toxicol. 192, 178–183. https://doi.org/10.1016/j.aquatox.2017.09.016

Mewes, K., Mogollón, J.M., Picard, a., Rühlemann, C., Kuhn, T., Nöthen, K., Kasten, S., 2014. Impact of depositional and biogeochemical processes on small scale variations in nodule abundance in the Clarion-Clipperton Fracture Zone. Deep. Res. Part I Oceanogr. Res. Pap. 91, 125–141. https://doi.org/10.1016/j.dsr.2014.06.001

Mewes, K., Mogollón, J.M., Picard, A., Rühlemann, C., Eisenhauer, A., Kuhn, T., Ziebis, W., Kasten, S., 2016. Diffusive transfer of oxygen from seamount basaltic crust into overlying sediments : An example from the Clarion – Clipperton Fracture Zone. Earth Planet. Sci. Lett. 433, 215–225. https://doi.org/10.1016/j.epsl.2015.10.028

Michard, A., 1989. Rare earth element systematics in hydrothermal fluids. Geochim. Cosmochim. Acta 53, 745–750. https://doi.org/10.1016/0016-7037(89)90017-3

Mine and monitor impacts, 2015. . Nat. Geosci. 8, 161. https://doi.org/10.1038/ngeo2390

Mogollón, J.M., Mewes, K., Kasten, S., 2016. Quantifying manganese and nitrogen cycle coupling in manganese-rich, organic carbon-starved marine sediments: Examples from

References

166

the Clarion-Clipperton fracture zone. Geophys. Res. Lett. 43. https://doi.org/10.1002/2016GL069117

Möller, P., Bau, M., 1993. Rare-earth patterns with positive cerium anomaly in alkaline waters from Lake Van, Turkey. Earth Planet. Sci. Lett. 117, 671–676. https://doi.org/10.1016/0012-821X(93)90110-U

Morford, J.L., Emerson, S., 1999. The geochemistry of redox sensitive trace metals in sediments. Geochim. Cosmochim. Acta 63, 1735–1750. https://doi.org/10.1016/S0016-7037(99)00126-X

Morford, J.L., Emerson, S.R., Breckel, E.J., Kim, S.H., 2005. Diagenesis of oxyanions (V, U, Re, and Mo) in pore waters and sediments from a continental margin. Geochim. Cosmochim. Acta 69, 5021–5032. https://doi.org/10.1016/j.gca.2005.05.015

Morris, A.W., 1975. Dissolved molybdenum and vanadium in the northeast Atlantic Ocean. Deep Sea Res. Oceanogr. Abstr. 22, 49–54. https://doi.org/10.1016/0011-7471(75)90018-2

Müller, P.J., Hartmann, M., Suess, E., 1988. The chemical environment of pelagic sediments, in: Halbach, P., Friedrich, G., von Stackelberg, U. (Eds.), The Manganese Nodule Belt of the Pacific Ocean : Geological Environment, Nodule Formation, and Mining Aspects. Ferdinand Enke Verlag, Stuttgart, pp. 70–90.

Murnane, R., Clague, D.A., 1983. Nontronite from a low-temperature hydrothermal system on the Juan de Fuca Ridge. Earth Planet. Sci. Lett. 65, 343–352. https://doi.org/10.1016/0012-821X(83)90172-3

Nautilus Minerals, 2018. About Nautilus - Overview [WWW Document]. URL www.nautilusminerals.com/irm/content/overview.aspx?RID=252&RedirectCount=1 (accessed 8.10.18).

Nicomel, N., Leus, K., Folens, K., Van Der Voort, P., Du Laing, G., 2015. Technologies for Arsenic Removal from Water: Current Status and Future Perspectives. Int. J. Environ. Res. Public Health 13, 1–24. https://doi.org/10.3390/ijerph13010062

Nothdurft, L.D., Webb, G.E., Kamber, B.S., 2004. Rare earth element geochemistry of Late Devonian reefal carbonates, Canning Basin, Western Australia: Confirmation of a seawater REE proxy in ancient limestones. Geochim. Cosmochim. Acta 68, 263–283. https://doi.org/10.1016/S0016-7037(03)00422-8

Nöthen, K., Kasten, S., 2011. Reconstructing changes in seep activity by means of pore water and solid phase Sr/Ca and Mg/Ca ratios in pockmark sediments of the Northern Congo Fan. Mar. Geol. 287, 1–13. https://doi.org/10.1016/j.margeo.2011.06.008

O’Neill, H.S.C., 2016. The smoothness and shapes of chondrite-normalized rare earth element patterns in basalts. J. Petrol. 57, 1463–1508. https://doi.org/10.1093/petrology/egw047

Oebius, H.U., Becker, H.J., Rolinski, S., Jankowski, J.A., 2001. Parametrization and evaluation of marine environmental impacts produced by deep-sea manganese nodule mining. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3453–3467. https://doi.org/10.1016/S0967-0645(01)00052-2

Oldham, V.E., Owings, S.M., Jones, M.R., Tebo, B.M., Luther, G.W., 2015. Evidence for the presence of strong Mn(III)-binding ligands in the water column of the Chesapeake Bay. Mar. Chem. 171, 58–66. https://doi.org/10.1016/j.marchem.2015.02.008

Pälike, H., Lyle, M.W., Nishi, H., Raffi, I., Ridgwell, A., Gamage, K., Klaus, A., Acton, G., Anderson, L., Backman, J., Baldauf, J., Beltran, C., Bohaty, S.M., Bown, P., Busch, W., Channell, J.E.T., Chun, C.O.J., Delaney, M., Dewangan, P., Dunkley Jones, T., Edgar, K.M., Evans, H., Fitch, P., Foster, G.L., Gussone, N., Hasegawa, H., Hathorne, E.C.,

References

167

Hayashi, H., Herrle, J.O., Holbourn, A., Hovan, S., Hyeong, K., Iijima, K., Ito, T., Kamikuri, S., Kimoto, K., Kuroda, J., Leon-Rodriguez, L., Malinverno, A., Moore Jr, T.C., Murphy, B.H., Murphy, D.P., Nakamura, H., Ogane, K., Ohneiser, C., Richter, C., Robinson, R., Rohling, E.J., Romero, O., Sawada, K., Scher, H., Schneider, L., Sluijs, A., Takata, H., Tian, J., Tsujimoto, A., Wade, B.S., Westerhold, T., Wilkens, R., Williams, T., Wilson, P.A., Yamamoto, Y., Yamamoto, S., Yamazaki, T., Zeebe, R.E., 2012. A Cenozoic record of the equatorial Pacific carbonate compensation depth. Nature 488, 609–614. https://doi.org/10.1038/nature11360

Pattan, J.N., Higgs, N.C., 1995. Rare earth element studies of surficial sediments from the southwestern Carlsberg Ridge, Indian Ocean. Proc. Indian Acad. Sci. - Earth Planet. Sci. 104, 569–578. https://doi.org/10.1007/BF02839297

Pattan, J.N., Parthiban, G., 2007. Do manganese nodules grow or dissolve after burial? Results from the Central Indian Ocean Basin. J. Asian Earth Sceinces 30, 696–705. https://doi.org/10.1016/j.jseaes.2007.03.003

Paul, S.A.L., Gaye, B., Haeckel, M., Kasten, S., Koschinsky, A., 2018. Biogeochemical Regeneration of a Nodule Mining Disturbance Site: Trace Metals, DOC and Amino Acids in Deep-Sea Sediments and Pore Waters. Front. Mar. Sci. 5, 1–17. https://doi.org/10.3389/fmars.2018.00117

Paul, S.A.L., Volz, J.B., Bau, M., Köster, M., Kasten, S., Koschinsky, A., 2019. Calcium phosphate control of REY patterns of siliceous-ooze-rich deep-sea sediments from the central equatorial Pacific. Geochim. Cosmochim. Acta 251, 56–72. https://doi.org/10.1016/j.gca.2019.02.019

Pedro, G., Carmouze, J.P., Velde, B., 1978. Peloidal nontronite formation in recent sediments of Lake Chad. Chem. Geol. 23, 139–149. https://doi.org/10.1016/0009-2541(78)90071-2

Perkin Elmer, n.d. The 30-Minute Guide to ICP-MS.

Petersen, S., Krätschell, A., Augustin, N., Jamieson, J., Hein, J.R., Hannington, M.D., 2016. News from the seabed - Geological characteristics and resource potential of deep-sea mineral resources. Mar. Policy 70, 175–187. https://doi.org/10.1016/j.marpol.2016.03.012

Piper, D.Z., 1974a. Rare earth elements in the sedimentary cycle: A summary. Chem. Geol. 14, 285–304. https://doi.org/10.1016/0009-2541(74)90066-7

Piper, D.Z., 1974b. Rare earth elements in ferromanganese nodules and other marine phases. Geochim. Cosmochim. Acta 38, 1007–1022. https://doi.org/10.1016/0016-7037(74)90002-7

Piper, D.Z., 1973. Origin of metalliferous sediments from the East Pacific Rise. Earth Planet. Sci. Lett. 19, 75–82. https://doi.org/10.1038/251465a0

Piper, D.Z., Baedecker, P.A., Crock, J.G., Burnett, W.C., Loebner, B.J., 1988. Rare earth elements in the phosphatic-enriched sediment of the Peru Shelf. Mar. Geol. 80, 269–285. https://doi.org/10.1016/0025-3227(88)90093-X

Poulton, S.W., Canfield, D.E., 2005. Development of a sequential extraction procedure for iron: implications for iron partitioning in continentally derived particulates. Chem. Geol. 214, 209–221. https://doi.org/10.1016/j.chemgeo.2004.09.003

Pourret, O., Tuduri, J., 2017. Continental shelves as potential resource of rare earth elements. Sci. Rep. 7, 1–6. https://doi.org/10.1038/s41598-017-06380-z

Prudêncio, M.I., Figueiredo, M.O., Cabral, J.M.P., 1989. Rare earth distribution and its correlation with clay mineralogy in the clay-sized fraction of Cretaceous and Pliocene sediments (central Portugal). Clay Miner. 24, 67–74. https://doi.org/10.1180/claymin.1989.024.1.06

References

168

Ramirez-Llodra, E., Tyler, P. a., Baker, M.C., Bergstad, O.A., Clark, M.R., Escobar, E., Levin, L.A., Menot, L., Rowden, A.A., Smith, C.R., Van Dover, C.L., 2011. Man and the last great wilderness: Human impact on the deep sea. PLoS One 6. https://doi.org/10.1371/journal.pone.0022588

Reitz, A., Pfeifer, K., De Lange, G.J., Klump, J., 2004. Biogenic barium and the detrital Ba/Al ratio: A comparison of their direct and indirect determination. Mar. Geol. 204, 289–300. https://doi.org/10.1016/S0025-3227(04)00004-0

Reynard, B., Lécuyer, C., Grandjean, P., 1999. Crystal-chemical controls on rare-earth element concentrations in fossil biogenic apatites and implications for paleoenvironmental reconstructions. Chem. Geol. 155, 233–241. https://doi.org/10.1016/S0009-2541(98)00169-7

Roje, V., 2010. Multi-elemental analysis of marine sediment reference material MESS-3: one-step microwave digestion and determination by high resolution inductively coupled plasma-mass spectrometry (HR-ICP-MS). Chem. Pap. 64, 409–414. https://doi.org/10.2478/s11696-010-0022-x

Rønsbo, J.G., 1989. Coupled substitutions involving REEs and Na and Si in apatites in alkaline rocks from Ilímaussaq intrusion, South Greenland, and the petrological implications. Am. Mineral. 74, 896–901. https://doi.org/10.4103/0019-5545.140618

Rühlemann, C., Kuhn, T., Wiedicke, M., Kasten, S., Mewes, K., Picard, A., 2011. Current status of manganese nodule exploration in the German license area, in: Proceedings of the Ninth (2011) ISOPE Ocean Mining Symposium. Maui, Hawaii, pp. 168–173.

Ruhlin, D.E., Owen, R.M., 1986. The rare earth element geochemistry of hydrothermal sediments from the East Pacific Rise: Examination of a seawater scavenging mechanism. Geochim. Cosmochim. Acta 50, 393–400. https://doi.org/10.1016/0016-7037(86)90192-4

Russell, J., Goodman, B., Fraser, A., 1979. Infrared and Mossbauer Studies of Reduced Nontronites. Clays Clay Miner. 27, 63–71. https://doi.org/10.1346/CCMN.1979.0270108

Ruttenberg, K.C., 1992. Development of a sequential extraction method for different forms of phosphorus in marine sediments. Limnol. Oceanogr. 37, 1460–1482. https://doi.org/10.4319/lo.1992.37.7.1460

Sawlan, J.J., Murray, J.W., 1983. Trace metal remobilization in the interstitial waters of red clay and hemipelagic marine sediments 64, 213–230. https://doi.org/10.1016/0012-821X(83)90205-4

Schenau, S.J., De Lange, G.J., 2000. A novel chemical method to quantify fish debris in marine sediments. Limnol. Ocean. 45, 963–971. https://doi.org/10.4319/lo.2000.45.4.0963

Schindler, D.E., Hilborn, R., 2015. Prediction, precaution, and policy under global change. Science. 347, 953–954. https://doi.org/10.1126/science.1261824

Scholz, F., Hensen, C., Noffke, A., Rohde, A., Liebetrau, V., Wallmann, K., 2011. Early diagenesis of redox-sensitive trace metals in the Peru upwelling area - response to ENSO-related oxygen fluctuations in the water column. Geochim. Cosmochim. Acta 75, 7257–7276. https://doi.org/10.1016/j.gca.2011.08.007

Schriever, G., Koschinsky, A., Bluhm, H., 1996. Cruise Report ATESEPP Impacts of potential technical interventions on the deep-sea ecosystem of the southeast Pacific off Peru (SONNE Cruise 106), Berichte aus dem Zentrum für Meeres- und Klimaforschung der Universität Hamburg.

Schulz, H.D., 2006. Quantification of Early Diagenesis: Dissolved Consitutents in Pore Water and Signals in the Solid Phase, in: Schulz, H.D., Zabel, M. (Eds.), Marine Geochemistry.

References

169

Springer, Berlin, Heidelberg, pp. 73–124. https://doi.org/10.1017/CBO9781107415324.004

Seeberg-Elverfeldt, J., Schlüter, M., Feseker, T., Kölling, M., 2005. Rhizon sampling of porewaters near the sediment-water interface of aquatic systems. Limnol. Oceanogr. Methods 3, 361–371. https://doi.org/10.4319/lom.2005.3.361

Seifert, R., Emeis, K.C., Michaelis, W., Degens, E.T., 1990. Amino acids and carbohydrates in sediments and interstitial waters from Site 681, Leg 112, Peru Continental Margin, in: Proceedings of the Ocean Drilling Program, Scientific Results. pp. 555–566.

Seitz, M., Oliver, A.G., Raymond, K.N., 2007. The lanthanide contraction revisited. J. Am. Chem. Soc. 129, 11153–11160. https://doi.org/10.1021/ja072750f

Seth, M., Dolg, M., Fulde, P., Schwerdtfeger, P., 1995. Lanthanide and Actinide Contractions: Relativistic and Shell Structure Effects. J. Am. Chem. Soc. 117, 6597–6598. https://doi.org/10.1021/ja00129a026

Shabani, M.B., Akagi, T., Masuda, A., 1992. Preconcentration of Trace Rare-Earth Elements in Seawater by Complexation with Bis(2-ethylhexyl) Hydrogen Phosphate and 2-Ethylhexyl Dihydrogen Phosphate Adsorbed on a C18Cartridge and Determination by Inductively Coupled Plasma Mass Spectrometry. Anal. Chem. 64, 737–743. https://doi.org/10.1021/ac00031a008

Shannon, R.D., 1976. Revised Effective Ionic Radii and Systematic Study of Interatomic Distances in Halides and Chalcogenides. Acta Crystallogr. A32, 751–767. https://doi.org/10.1107/s0567739476001551

Sharma, R., 2017. Deep-sea mining: Resource potential, technical and environmental considerations. https://doi.org/10.1007/978-3-319-52557-0

Shaw, T.J., Gieskes, J.M., Jahnke, R.A., 1990. Early diagenesis in differing depositional environments: The response of transition metals in pore water. Geochim. Cosmochim. Acta 54, 1233–1246. https://doi.org/10.1016/0016-7037(90)90149-F

Sholkovitz, E.R., Church, T.M., Arimoto, R., 1993. Rare earth element composition of precipitation, precipitation particles, and aerosols. J. Geophys. Res. 98, 20587–20599. https://doi.org/10.1029/93JD01926

Sholkovitz, E.R., Landing, W.M., Lewis, B.L., 1994. Ocean particle chemistry: The fractionation of rare earth elements between suspended particles and seawater. Geochim. Cosmochim. Acta 58, 1567–1579. https://doi.org/10.1016/0016-7037(94)90559-2

Sholkovitz, E.R., Piepgras, D.J., Jacobsen, S.B., 1989. The pore water chemistry of rare earth elements in Buzzards Bay sediments. Geochim. Cosmochim. Acta 53, 2847–2856. https://doi.org/10.1016/0016-7037(89)90162-2

Soyol-Erdene, T.O., Huh, Y., 2013. Rare earth element cycling in the pore waters of the Bering Sea Slope (IODP Exp. 323). Chem. Geol. 358, 75–89. https://doi.org/10.1016/j.chemgeo.2013.08.047

Spectro, 2004. Spectro Ciros Vision - Optical emission spectrometer with inductively-coupled plasma excitation.

Stummeyer, J., Marchig, V., 2001. Mobility of metals over the redox boundary in Peru Basin sediments. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3549–3567. https://doi.org/10.1016/S0967-0645(01)00056-X

Takahashi, Y., Manceau, A., Geoffroy, N., Marcus, M.A., Usui, A., 2007. Chemical and structural control of the partitioning of Co, Ce, and Pb in marine ferromanganese oxides. Geochim. Cosmochim. Acta 71, 984–1008. https://doi.org/10.1016/j.gca.2006.11.016

References

170

Tang, J., Johannesson, K.H., 2010. Ligand extraction of rare earth elements from aquifer sediments: Implications for rare earth element complexation with organic matter in natural waters. Geochim. Cosmochim. Acta 74, 6690–6705. https://doi.org/10.1016/j.gca.2010.08.028

Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its Composition and Evolution. An Examination of the Geochemical Record Preserved in Sedimentary Rocks. Blackwell Science, Oxford. https://doi.org/10.1017/S0016756800032167

Telfeyan, K., Breaux, A., Kim, J., Cable, J.E., Kolker, A.S., Grimm, D.A., Johannesson, K.H., 2017. Arsenic, vanadium, iron, and manganese biogeochemistry in a deltaic wetland, southern Louisiana, USA. Mar. Chem. 192, 32–48. https://doi.org/10.1016/j.marchem.2017.03.010

Thiel, H., 2001. Use and protection of the deep sea - An introduction. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3427–3431. https://doi.org/10.1016/S0967-0645(01)00050-9

Thiel, H., Forschungsverbund Tiefsee-Umweltschutz, 2001. Evaluation of the environmental consequences of polymetallic nodule mining based on the results of the TUSCH Research Association. Deep. Res. Part II Top. Stud. Oceanogr. 48, 3433–3452. https://doi.org/10.1016/S0967-0645(01)00051-0

Thiel, H., Schriever, G., 1990. Deep-sea mining, environmental impact and the DISCOL project. Ambio 19, 245–250.

Thomson, J., Higgs, N.C., Croudace, I.W., Colley, S., Hydes, D.J., 1993. Redox zonation of elements at an oxic/post-oxic boundary in deep-sea sediments. Geochim. Cosmochim. Acta 57, 579–595. https://doi.org/10.1016/0016-7037(93)90369-8

Toole, J., Thomson, J., Wilson, T.R.S., Baxter, M.S., 1984. A sampling artefact affecting the uranium content of deep-sea porewaters obtained from cores. Nature 308, 263–266. https://doi.org/10.1038/311525a0

Toyoda, K., Masuda, A., 1991. Chemical leaching of pelagic sediments: Identification of the carrier of Ce anomaly. Geochem. J. 25, 95–119. https://doi.org/10.2343/geochemj.25.95

Toyoda, K., Nakamura, Y., Masuda, A., 1990. Rare earth elements of Pacific pelagic sediments. Geochim. Cosmochim. Acta 54, 1093–1103. https://doi.org/10.1016/0016-7037(90)90441-M

Toyoda, K., Tokonami, M., 1990. Diffusion of rare-earth elements in fish teeth from deep-sea sediments. Nature 345, 607–609. https://doi.org/10.1038/346183a0

Trotter, J.A., Barnes, C.R., McCracken, A.D., 2016. Rare earth elements in conodont apatite: Seawater or pore-water signatures? Palaeogeogr. Palaeoclimatol. Palaeoecol. 462, 92–100. https://doi.org/10.1016/j.palaeo.2016.09.007

Van Dover, C.L., 2011. Tighten regulations on deep-sea mining. Nature 470, 31–33.

Van Dover, C.L., Aronson, J., Pendleton, L., Smith, S., Arnaud-Haond, S., Moreno-Mateos, D., Barbier, E., Billett, D., Bowers, K., Danovaro, R., Edwards, A., Kellert, S., Morato, T., Pollard, E., Rogers, A., Warner, R., 2014. Ecological restoration in the deep sea: Desiderata. Mar. Policy 44, 98–106. https://doi.org/10.1016/j.marpol.2013.07.006

Vanreusel, A., Hilario, A., Ribeiro, P.A., Menot, L., Martínez Arbizu, P., 2016. Threatened by mining, polymetallic nodules are required to preserve abyssal epifauna. Sci. Rep. 6, 1–6. https://doi.org/10.1038/srep26808

Viehmann, S., Bau, M., Hoffmann, J.E., Münker, C., 2015. Geochemistry of the Krivoy Rog Banded Iron Formation, Ukraine, and the impact of peak episodes of increased global magmatic activity on the trace element composition of Precambrian seawater.

References

171

Precambrian Res. 270, 165–180. https://doi.org/10.1016/j.precamres.2015.09.015

Volz, J.B., Mogollón, J.M., Geibert, W., Martínez Arbizu, P., Koschinsky, A., Kasten, S., 2018. Natural spatial variability of depositional conditions, biogeochemical processes and element fluxes in sediments of the eastern Clarion-Clipperton Zone, Pacific Ocean. Deep. Res. Part I Oceanogr. Res. Pap. 140, 159–172. https://doi.org/10.1016/j.dsr.2018.08.006

von Stackelberg, U., Beiersdorf, H., 1991. The formation of manganese nodules between the Clarion and Clipperton fracture zones southeast of Hawaii. Mar. Geol. 98, 411–423. https://doi.org/10.1016/0025-3227(91)90113-I

Wang, D., 2012. Redox chemistry of molybdenum in natural waters and its involvement in biological evolution. Front. Microbiol. 3, 1–7. https://doi.org/10.3389/fmicb.2012.00427

Wang, Y.L., Liu, Y.G., Schmitt, R.A., 1986. Rare earth element geochemistry of South Atlantic deep sea sediments: Ce anomaly change at ~54 My. Geochim. Cosmochim. Acta 50, 1337–1355. https://doi.org/10.1016/0016-7037(86)90310-8

Webb, G.E., Kamber, B.S., 2000. Rare earth elements in Holocene reefal microbialites: A new shallow seawater proxy. Geochim. Cosmochim. Acta 64, 1557–1565. https://doi.org/10.1016/S0016-7037(99)00400-7

Weber, M.E., Pisias, N.G., 1999. Spatial and temporal distribution of biogenic carbonate and opal in deep-sea sediments from the eastern equatorial Pacific: implications for ocean history since 1.3 Ma. Earth Planet. Sci. Lett. 174, 59–73. https://doi.org/10.1016/S0012-821X(99)00248-4

Weber, M.E., Von Stackelberg, U., Marchig, V., Wiedicke, M., Grupe, B., 2000. Variability of surface sediments in the Peru basin: Dependence on water depth, productivity, bottom water flow, and seafloor topography. Mar. Geol. 163, 169–184. https://doi.org/10.1016/S0025-3227(99)00103-6

Weber, M.E., Wiedicke, M., Riech, V., Erlenkeuser, H., 1995. Carbonate preservation history in the Peru Basin: Paleoceanographic implications. Paleoceanography 10, 775–800. https://doi.org/10.1029/95PA01566

Wegorzewski, A., Kuhn, T., 2014. The influence of suboxic diagenesis on the formation of manganese nodules in the Clarion Clipperton nodule belt of the Pacific Ocean. Mar. Geol. 357, 123–138. https://doi.org/10.1016/j.margeo.2014.07.004

Whittaker, J.M., Goncharov, A., Williams, S.E., Müller, R.D., Leitchenkov, G., 2013. Global sediment thickness data set updated for the Australian-Antarctic Southern Ocean. Geochemistry, Geophys. Geosystems 14, 3297–3305. https://doi.org/10.1002/ggge.20181

Wright, J., 1990. Conodont Apatite: Structure and Geochemistry, in: Carter, J.G. (Ed.), Skeletal Biomineralization: Patterns, Processes and Evolutionary Trends (Volume 1). Van Nostrand Reinhold, New York, pp. 445–459. https://doi.org/10.1029/SC005p0149

Wright, J., Seymour, R.S., Shaw, H.F., 1984. REE and Nd isotopes in conodont apatite: Variations with geological age and depositional environment. Geol. Soc. Am. Special Pa, 325–340.

Wyrtki, K., 1981. An Estimate of Equatorial Upwelling in the Pacific. J. Phys. Oceanogr. 11, 1205–1214. https://doi.org/10.1175/1520-0485(1981)011<1205:AEOEUI>2.0.CO;2

Zhang, L., Algeo, T.J., Cao, L., Zhao, L., Chen, Z.Q., Li, Z., 2016. Diagenetic uptake of rare earth elements by conodont apatite. Palaeogeogr. Palaeoclimatol. Palaeoecol. 458, 176–197. https://doi.org/10.1016/j.palaeo.2015.10.049

Ziegler, C.L., Murray, R.W., 2007. Geochemical evolution of the central Pacific Ocean over the

References

172

past 56 Myr. Paleoceanography 22. https://doi.org/10.1029/2006PA001321

Appendix I

173

Appendix I

Supplementary Material to Small-scale heterogeneity of trace metals including REY in

deep-sea sediments and pore waters of the Peru Basin, SE equatorial Pacific (Chapter 2).

Appendix I

174

Supplementary Material 1

Averages, standard deviation, accuracy, and method precision calculated from averages of

each digested sample.

Table S1: Replicate analyses of MESS-3 reference material during ICP-OES measurements (n=13 digested samples, 10 ICP-OES runs).

LOQ# [mg/kg]

Element MESS-3 reference [mg/kg]

MESS-3 measured [mg/kg]

Accuracy (%)

Method precision (%RSD)

19-231 Al 85900±2300 68879±13622 -20 20

13-158 Ca 14700±600 13291±1094 -10 8

5-24 Cu 33.9±1.6 33±1 -3 4

98-379 Fe 43400±1100 39219±2264 -10 6

9-101 Mn 324±12 302±22 -7 7

86-741 P 1200a 1257±128 +5 10

2-16 Sr 129±11 114±17 -11 15

8-28 V 243±10 233±17 -4 7

4-15 Zn 159±8 147±6 -7 4 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a information value

Table S2: Replicate analyses of BHVO-2 reference material during ICP-OES measurements (n=13 digested samples, 10 ICP-OES runs).

LOQ# [mg/kg]

Element BHVO-2 reference [mg/kg]

BHVO-2 measured [mg/kg]

Accuracy (%) Method precision (%RSD)

19-231 Al 71600±800 74535±1889 +4 3

13-158 Ca 81700±1200 83017±2188 +2 3

5-24 Cu 127±7 135±5 +7 3

98-379 Fe 86300±1400 87880±2323 +2 3

9-101 Mn 1290±40 1347±35 +4 3

86-741 P 1200±100 1307±143 +9 11

2-16 Sr 389±23 392±10 +1 3

8-28 V 317±11 315±16 +22 19

4-15 Zn 103±6 117±5 +14 5 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Appendix I

175

Table S3: Replicate analyses of MESS-3 reference material during ICP-MS measurements (n=13 digested samples, 8 ICP-MS runs; one less for Mo).

Average LOQ# [mg/kg]




0.25 Co 14.4±2 14.5±0.5 +1 4

0.25 Ni 46.9±2.2 49.1±2 +5 4

0.67 Mo 2.78±0.07 2.78±0.16 0 6

0.07 U 4b 3.3±0.7 -17 21 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Table S4: Replicate analyses of BHVO-2 CRM during ICP-MS measurements (n=13 digested

samples; 9 ICP-MS runs; one less for Co, Ni, Ba, Eu, and Tb).


Element BHVO-2 recommended [mg/kg]


Accuracy (%)

Precision (%)

0.25 Co 45±3 46±1 +2 3 0.25 Ni 119±7 125±4 +5 3 0.58 Ba 130±13 129±6 -1 5 0.10 Y 26±2 25±1 -5 3 0.05 La 15±1 16±1 +4 5 0.05 Ce 38±2 39±2 +3 5 0.04 Pr --- 5.5±0.3 --- 5 0.28 Nd 25.0±1.8 25±1 0 5 0.24 Sm 6.2±0.4b 6.3±0.3 +1 5 0.09 Eu --- 2.1±0.1 --- 5 0.15 Gd 6.3±0.2b 6.7±0.4 +7 6 0.05 Tb 0.9b 0.96±0.04 +7 5 0.15 Dy --- 5.5±0.3 --- 5 0.05 Ho 1.04±0.04b 1.00±0.04 -4 4 0.13 Er --- 2.6±0.1 --- 5 0.05 Tm --- 0.34±0.02 --- 5 0.15 Yb 2.0±0.2b 2.0±0.1 +2 4 0.04 Lu 0.28±0.01b 0.28±0.02 -2 6

# LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Appendix I

176

Table S5: Replicate analyses of NASS-6 seawater certified reference material during ICP-MS

measurements.

LOQ#

[µg/kg]

Element n

(number

of ICP-

MS runs)

NASS-6

reference

[µg/kg]

NASS-6

measured

[µg/kg]

Accuracy

(%)

Method

precision

(%RSD)

0.029-

0.80

Cd 3 0.0303±0.0019 0.037±0.0012 +22 3.1

0.02-

0.17

U 5 3a 2.73±0.1 -9 3.7

0.44-

2.90

V (KED) 4 1.42±0.16 1.45±0.08 +2 5.5

0.56-

2.20

Mn (KED) --- 0.516±0.047 ---

0.07-

0.26

Co (KED) --- 0.015a ---

0.12-3 Cu (KED) --- 0.242±0.025 ---

0.88-4.8 As (KED) 2 1.40±0.12 1.22±0.02 -13 1.5

0.3-5.7 Mo (KED) 5 9.66±0.70 9.87±0.27 +2 2.7

# LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a information value

Table S6: Replicate analyses of NASS-7 seawater certified reference material during ICP-MS

measurements.

LOQ#

[µg/kg]

Element n

(number

of ICP-

MS runs)

NASS-7

reference

[µg/kg]

NASS-7

measured

[µg/kg]

Accuracy

(%)

Method

precision

(%RSD)

0.029-

0.80

Cd 2 0.0157±0.0016 0.021±0.003 +34 14

0.02-

0.17

U 7 2.81±0.16 2.65±0.23 -6 8.6

0.44-

2.90

V (KED) 5 1.27±0.08a 1.38±0.15 +9 11

0.56-

2.20

Mn (KED) 1 0.74±0.06 0.73 -1 ---

0.07-

0.26

Co (KED) --- 0.0143±0.0014 ---

0.12-3 Cu (KED) --- 0.195±0.014 ---

0.88-4.8 As (KED) 4 1.23±0.06a 1.14±0.12 -7 11

0.3-5.7 Mo (KED) 7 9.10±0.40 8.99±0.43 -1 4.8

# LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a reference value

Appendix I

177

Table S7: Replicate analyses of SLEW-3 estuarine water reference material during ICP-MS

measurements (n= 7 ICP-MS runs). Runs for which the values were far below the LOQ were

excluded.

LOQ#

[µg/kg]

Element n

(number

of ICP-

MS runs)

SLEW-3

reference

[µg/kg]

SLEW-3

measured

[µg/kg]

Accuracy

(%)

Method

precision

(%RSD)

0.029-0.80 Cd 5 0.047±0.004 0.052±0.003 +10 6.4

0.02-0.17 U 7 1.8b 1.51±0.12 -16 8.2

0.44-2.90 V (KED) 7 2.54±0.31 2.75±0.29 +8 10.7

0.56-2.20 Mn (KED) 6 1.59±0.22 1.29±0.22 -19 17.4

0.07-0.26 Co (KED) --- 0.040±0.010 ---

0.12-3 Cu (KED) 6 1.53±0.12 1.51±0.45 -1 30

0.88-4.8 As (KED) 2 1.34±0.09 1.27±0.03 -5 2.2

0.3-5.7 Mo (KED) 7 5.1b 4.7±0.3 -8 6.7

# LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Appendix II

178

Appendix II

Supplementary Material to Biogeochemical regeneration of a nodule mining disturbance

site: trace metals, DOC and amino acids in deep-sea sediments and pore waters (Paul

et al., 2018; Chapter 3).

Originally published by Frontiers

Appendix II

179

Supplementary material 1

LOQ given as a range from individual LOQs from all runs. CRM means and standard deviations calculated from averages from each run. Run averages

only based on data above LOQ for that specific run.

Accuracy and method precision calculated for average CRM values from the runs.

Table 1: Replicate analyses of IAPSO seawater reference material during ICP-OES measurements (n=29 measurements, 8 ICP-OES runs)

LOQ# [mg/kg]

Element IAPSO (OSIL) reference [mg/kg]

IAPSO measured [mg/kg]


0.11-0.96 B 4.5 4.5±0.12 0 3

0.83-9.83 Br 67 65±2 -2 3

0.34-14.36 Ca 412 430±8 4 2

2.19-10.43 K 399 401±8 1 2

0.36-12.10 Mg 1290 1427±41 11 3

1.11-72.53 S 904 914±16 1 2

0.42-2.82 Si <0.02-5 1.4±0.2 --- 17

0.20-2.98 Sr 7.9 7.5±0.3 -5 3 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Table 2: Replicate analyses of MESS-3 reference material during ICP-OES measurements (n=38 measurements, 11 digested samples, 9 ICP-OES runs)

LOQ# [mg/kg]



Accuracy (%)


28-401 Al 85900±2300 74546±15405 -13 21

23-183 Ca 14700±600 13757±1095 -6 8

5-37 Cu 33.9±1.6 33.7±1.2 -1 4

98-808 Fe 43400±1100 40447±2277 -7 6

76-542 K 26000a 25425±1273 -2 5

1-62 Mg 16000a 15874±2059 -1 13

3-100 Mn 324±12 311±16 -4 5

86-741 P 1200a 1306±125 9 10

4-15 Zn 159±8 149±4 -7 3 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a information value

Appendix II

180

Table 3: Replicate analyses of BHVO-2 reference material during ICP-OES measurements (n=24 measurements, 11 digested samples, 9 ICP-OES runs)

LOQ# [mg/kg]




28-401 Al 71600±800 74874±1842 5 2

23-183 Ca 81700±1200 82833±1115 1 1

5-37 Cu 127±7 133±4 5 3

98-808 Fe 86300±1400 87784±1633 2 2

76-542 K 4300±100 4309±80 0 2

1-62 Mg 43600±700 44625±986 2 2

3-100 Mn 1290±40 1354±28 5 2

86-741 P 1200±100 1338±167 12 13

4-15 Zn 103±6 115±5 12 4 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Table 4: Replicate analyses of IAPSO seawater reference material during ICP-MS measurements (n=25, 8 ICP-MS runs)

LOQ# [µg/kg] Element IAPSO (OSIL) reference [µg/kg]

IAPSO measured [µg/kg]

Accuracy (%)


0.12-1.58 Mo 10.5 13.07±0.99 25 8

0.004-0.452 U 3.3 3.09±0.15 -6 5 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Table 5: Replicate analyses of NASS-6 seawater reference material during ICP-MS measurements (n=50, 8 ICP-MS runs)

LOQ# [µg/kg]

Element NASS-6 reference [µg/kg]

NASS-6 measured [µg/kg]

Accuracy (%)


0.12-1.58 Mo 9.66±0.70 11.17±0.89 16 8

0.001-0.040 Cd 0.0303±0.0019 0.0480±0.0054 58 11

0.004-0.452 U 3a 2.81±0.16 -6 6

# LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a information value

Appendix II

181

Table 6: Replicate analyses of NASS-6 seawater reference material during ICP-MS measurements (n=24, 7 ICP-MS runs)

LOQ# [µg/kg] Element NASS-6 reference [µg/kg]

NASS-6 measured [µg/kg]

Accuracy (%)


0.20-0.90 V (KED) 1.42±0.16 1.42±0.06 0 4 0.73-3.34 Mn (KED) 0.516±0.047 Below LOQ --- --- 0.08-0.22 Co (KED) 0.015a Below LOQ --- ---

0.21-1.35 Cu (KED) 0.242±0.025 Below LOQ --- ---

0.17-1.65 As (KED) 1.40±0.12 1.28±0.10 -9 8 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a information value

Table 7: Replicate analyses of SLEW-3 estuarine water reference material during ICP-MS measurements (n=15, 7 ICP-MS runs)

LOQ# [µg/kg]

Element n SLEW-3 reference [µg/kg]

SLEW-3 measured [µg/kg]

Accuracy (%)


0.38-3.55 Mn 8 (5 runs) 1.59±0.22 1.90±0.39 19 21

0.20-0.90 V (KED) 15 (7 runs) 2.54±0.31 2.94±0.12 16 4

0.73-3.34 Mn (KED) 8 (4 runs) 1.59±0.22 1.57±0.13 -2 9

0.08-0.22 Co (KED) 0.040±0.010 Below LOQ --- ---

0.21-1.35 Cu (KED) 15 (7 runs) 1.53±0.12 1.34±0.39 -13 29

0.17-1.65 As (KED) 12 (6 runs) 1.34±0.09 1.39±0.17 4 13

# LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Table 8: Replicate analyses of SLRS-6 river water reference material during ICP-MS measurements (n=12, 7 ICP-MS runs)

LOQ# [µg/kg]

Element n SLRS-6 reference [µg/kg]

SLRS-6 measured [µg/kg]

Accuracy (%)


0.20-0.90 V (KED) 0.352±0.006 Below LOQ --- ---

0.73-3.34 Mn (KED) 6 (4 runs) 2.12±0.10 2.26±0.18 7 8

0.21-1.35 Cu (KED) 12 (7 runs)

24.0±1.8 30.5±0.95 27 3

0.17-1.65 As (KED) 0.57±0.08 Below LOQ --- --- # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Offset could be explained by a matrix effect (no salt matrix added to SLRS-6).

Appendix II

182

Table 9: Replicate analyses of MESS-3 reference material during ICP-MS measurements (n=24 measurements, 10 digested samples, 7 ICP-MS runs)

LOQ# [mg/kg]




0.03-0.36 Co 14.4±2 14.1±0.1 -2 1

0.10-2.18 Ni 46.9±2.2 48.1±1.1 3 2

0.07-2.78 Mo 2.78±0.07 2.69±0.09 -3 3

0.05-4.93 Pb 21.1±0.7 19.6±5.2 -8 26 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Table 10: Replicate analyses of BHVO-2 reference material during ICP-MS measurements (n=71 measurements, 11 digested samples, 8 ICP-MS runs)

LOQ# [mg/kg]




0.03-0.36 Co 45±3 46±2 3 3

0.10-2.18 Ni 119±7 126±4 6 3 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run

Appendix II

183

Supplementary material 2: Sediment major element profiles of the four undisturbed sites and

six disturbed sites.

Appendix II

184

Supplementary material 3: Bottom water and pore water element profiles of undisturbed and

disturbed sites. The uppermost values refer to bottom water concentrations measured in the

supernatant retrieved above the sediment surface in the MUC liner.

Appendix II

185

Supplementary material 4: Dissolved amino acid (DAA) spectra of three undisturbed and two

disturbed sites.

Appendix II

186

Supplementary material 5:

F=-øDsed(δC/δx) Flux calculation

Dsed=Dsw/(ϴ^2) diffusion coefficient in sediment

ϴ^2=1-ln(ø^2) tortuosity

Diffusive flux across sediment-water interface

Location average porosity (ø) δx [m] Mn Cu V Mo Mn Cu V Mo

Reference South 0.93 0.01 37.5 73.7 215 146 92.3 14.6 74.0 121

Reference West 0.94 0.01 18.0 69.5 242 161 0 0 59.3 155

outside DEA East plow track 0.93 0.01 - 96.6 200 138 - 20.8 75.4 130

outside EBS track 0.93 0.01 22.2 177 353 154 31.0 37.0 85.3 126

DEA West plow track 0.93 0.01 - 53.1 136 156 - 17.1 63.8 127

DEA East plow track 0.91 0.01 - 100 214 142 - 0 69.1 117

DEA East ripple 0.91 0.01 29.6 113 249 162 48.5 53.2 96.5 169

DEA East valley 0.92 0.01 14.5 125 195 167 30.1 33.2 86.4 147

DEA East white patch 0.92 0.01 17.7 171 222 144 94.0 243 184 139

EBS track 0.89 0.01 87.8 53.3 95.3 116 56.6 27.4 54.6 149

Potential diffusive flux across sediment-water interface when oxic layer removed

Pore

water

Bottom

water

Pore

water

Bottom

water

Pore

water

Bottom

water

Location average porosity (ø) δx [m] Mn Mn average porosity (ø)δx [m] Cu Cu average porosity (ø)δx [m] Co Co

DEA East plow track 0.88 0.06 8632 0 0.89 0.08 73.3 0 0.86 0.06 7.61 0

EBS track 0.87 0.12 4846 56.6 0.87 0.08 101 27.4 0.87 0.12 7.23 0

For bottom water values below the detection limit, 0 was used for calculations. Comparison calculations with bottom water data from other cores or low pore water

concentrations before the profile increase showed no major differences in final flux results.

Pore water nmol/L Bottom water nmol/L

DSW Mn=3.02E-10m^2s^-1; DSW Cu=3.22E-10m^2s^-1; DSW Co=3.15E-10m^2s^-1;

DSW V=5E-10m^2s^-1; DSW Mo=5E-10m^2s^-1

Dsw at temperature 0°C from Schulz, 2006 after Boudreau, 1997, closest diffusion coefficient

values considering deep-sea temperatures of 1.85°C (Brown et al., 2017b), except for V and Mo,

where only general diffusion coefficients published in (Emerson and Huested, 1991) and (Scholz et

al., 2011) were used

Appendix III

187

Appendix III

Supplementary data to Calcium phosphate control of REY patterns of siliceous-ooze-rich

deep-sea sediments from the central equatorial Pacific (Chapter 4).

Originally published by Elsevier.

Appendix III

188

Table EA1: Replicate analyses of NIST-2702 certified reference material (CRM) during ICP-

OES measurements (n=12 digested samples; 14 measurements in 5 ICP-OES runs).

LOQ# [mg/L] Element NIST-2702 certified [mg/kg]

NIST-2702 measured [mg/kg]

Accuracy (%)

Precision (%)

30-120 Al 84100±2200 79735±1525 -5 2

8.5-118 Caa 3430±240 3486±69 2 2

6.5-103.5 Fe 79100±2400 76739±2430 -3 3

60.5-951 K 20540±720 19500±386 -5 2

0.5-17.1 Mn 1757±58 1728±41 -2 2

169-2983 Na 6810±200 6485±568 -5 9

7.5-45 P 1552±66 1537±39 -1 3

8.5-148 Sb 15000 16001±314 7 2

4.7-18.5 Ti 8840±820 8302±189 -6 2 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a reference value b information value

Table EA2: Replicate analyses of NIST-2702 CRM during ICP-MS measurements (n=10

digested samples; 26 measurements in 2 ICP-MS runs). Accuracy and precision calculated

from averages of each digested sample.


Element NIST-2702 certified [mg/kg]


Accuracy (%)

Precision (%)

0.122 Ba 397.4±3.2 393±27.4 -1 7

0.074 Y --- 36.8±2.1 6

0.030 La 73.5±4.2 71.6±4.4 -3 6

0.030 Ce 123.4±5.8 125.5±8.2 2 7

0.043 Pr --- 16.5±1.0 6

0.051 Ndb 56 61±3.5 9 6

0.050 Smb 10.8 11.4±0.7 6 6

0.050 Eu --- 2.0±0.1 5

0.041 Gd --- 9.0±0.5 5

0.042 Tb --- 1.3±0.1 5

0.036 Dy --- 7.4±0.4 6

0.033 Ho --- 1.4±0.1 5

0.029 Er --- 4.0±0.2 5

0.023 Tm --- 0.6±0.0 5

0.017 Yb --- 3.7±0.2 5

0.018 Lu --- 0.5±0.0 4 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Appendix III

189

Table EA3: Replicate analyses of BHVO-2 CRM during ICP-MS measurements (n=1 digested

sample; 2 measurements during 1 ICP-MS run). Accuracy and precision calculated from

multiple measurements of 1 digested sample.

Average

LOQ# [mg/kg] Element BHVO-2

recommended [mg/kg]


Accuracy (%)

Precision (%)

0.122 Ba 130±13 129±6.1 -1 5

0.074 Y 26±2 24.5±0.4 -6 2

0.030 La 15±1 15.5±0.8 3 5

0.030 Ce 38±2 39±2 3 5

0.043 Pr --- 5.4±0.28 5

0.051 Nd 25.0±1.8 24.8±1.1 -1 4

0.050 Smb 6.2±0.4 6.3±0.3 1 5

0.050 Eu --- 2.1±0.08 4

0.041 Gdb 6.3±0.2 6.3±0.2 1 4

0.042 Tbb 0.9 0.9±0.03 2 4

0.036 Dy --- 5.4±0.3 5

0.033 Hob 1.04±0.04 0.98±0.03 -4 3

0.029 Er --- 2.6±1.12 5

0.023 Tm --- 0.3±0.01 3

0.017 Ybb 2.0±0.2 2.0±0.07 1 4

0.018 Lub 0.28±0.01 0.28±0.01 -2 4 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Appendix III

190

Supplementary data 3: Major elements.

Major element concentrations and Fe/Al and Mn/Al ratios

Salt mass corrected

Sample ID Al [wt.%] Ba [ppm] Ca [wt.%] Fe [wt.%] K [wt.%] Mn [wt.%] Na [wt.%] P [ppm] S [ppm] Ti [ppm] Fe/Al Mn/Al

SO239-87-35cm 6.71 11747 0.84 4.54 2.25 0.33 4.87 1030 5386 3288 0.68 0.05

SO239-87-123cm 7.51 11070 0.91 5.05 2.4 0.14 3.83 1226 5172 3678 0.67 0.02

SO239-87-163cm 7.5 12549 0.95 4.84 2.42 0.58 4.05 1347 5660 3615 0.65 0.08

SO239-87-203cm 7.3 14579 0.85 4.82 2.37 0.09 3.82 1075 5868 3586 0.66 0.01

SO239-87-243cm 7.96 10556 0.97 5.28 2.58 0.06 4.02 1370 5001 3911 0.66 0.01

SO239-87-283cm 7.58 11338 0.95 4.99 2.4 0.08 3.87 1224 4999 3736 0.66 0.01

SO239-87-343cm 7.38 12973 0.98 4.84 2.33 0.13 3.92 1479 5242 3570 0.66 0.02

SO239-87-383cm 7.16 12332 0.96 4.65 2.28 0.24 3.87 1423 5378 3484 0.65 0.03

SO239-87-443cm 7.39 10457 1.03 5.08 2.42 0.07 3.89 1709 5010 3547 0.69 0.01

SO239-87-483cm 7.11 16073 1 4.91 2.36 0.17 3.74 1713 5945 3409 0.69 0.02

SO239-87-607cm 7.19 14798 1 5.24 2.33 0.06 3.59 1599 5871 3594 0.73 0.01

SO239-87-667cm 7.22 12948 0.96 4.89 2.3 0.06 3.41 1484 5294 3471 0.68 0.01

SO239-87-727cm 6.93 10602 1.15 4.94 2.21 0.24 3.50 2229 4872 3366 0.71 0.03

SO239-87-767cm 7.11 13043 1.3 5.18 2.25 0.89 3.59 2690 5426 3482 0.73 0.12

SO239-87-827cm 6.92 13614 1.3 4.81 2.1 0.9 3.59 2708 5423 3455 0.7 0.13

SO239-87-887cm 6.84 13207 1.37 4.87 2.13 0.92 3.39 3100 5108 3423 0.71 0.13

SO239-165-47cm 7.84 5139 0.75 4.94 2.59 0.26 3.31 944 3241 3911 0.63 0.03

SO239-165-127cm 8.24 4646 0.74 5.16 2.69 0.48 3.43 850 3248 4100 0.63 0.06

SO239-165-187cm 7.94 5084 0.74 5.03 2.6 0.58 3.64 833 3511 3921 0.63 0.07

SO239-165-267cm 7.9 5610 0.85 5.02 2.6 0.43 3.47 1282 3497 3949 0.64 0.05

SO239-165-367cm 7.95 5146 0.77 4.98 2.67 0.61 3.50 1019 3349 4009 0.63 0.08

SO239-165-407cm 6.61 5910 0.83 3.98 2.28 6.63 3.53 804 3125 3485 0.60 1.00

SO239-165-467cm 7.92 6522 0.72 5.08 2.61 0.11 3.49 810 3736 3960 0.64 0.01

SO239-165-517cm 8.02 5366 0.76 5.16 2.63 0.05 3.42 916 3323 3934 0.64 0.01

Appendix III

191

Supplementary data 3: Major Elements continued

Sample ID Al [wt.%] Ba [ppm] Ca [wt.%] Fe [wt.%] K [wt.%] Mn [wt.%] Na [wt.%] P [ppm] S [ppm] Ti [ppm] Fe/Al Mn/Al

SO239-165-612cm 8.08 4756 0.76 5.24 2.6 0.05 3.43 915 3217 3988 0.65 0.01

SO239-165-672cm 7.83 5856 0.84 5.17 2.52 0.27 3.46 1316 3631 3779 0.66 0.03

SO239-165-732cm 7.84 6342 0.88 5.17 2.52 0.16 3.59 1471 3803 3758 0.66 0.02

SO239-165-792cm 6.17 14757 1.8 6.86 2.21 0.68 4.79 4726 7035 2667 1.11 0.11

SO239-165-832cm 5.69 14560 1.89 7.22 2.33 1.56 5.10 5112 7237 2377 1.27 0.27

SO239-165-912cm 5.14 14305 2.04 7.46 2.14 1.08 5.15 6149 7300 2110 1.45 0.21

SO239-194-5cm 8.77 3275 0.78 5.41 3.03 0.54 2.33 961 1943 4762 0.62 0.06

SO239-194-85cm 8.74 3027 0.73 5.48 3.06 0.6 2.38 920 1887 4792 0.63 0.07

SO239-194-157cm 8.61 3626 0.72 5.27 3.03 0.53 2.38 972 2027 4713 0.61 0.06

SO239-194-201cm 8.48 3587 0.72 5.24 3.02 0.64 2.23 1001 1906 4579 0.62 0.08

SO239-194-261cm 8.45 3771 0.76 5.35 3 0.78 2.24 1074 1884 4695 0.63 0.09

SO239-194-321cm 8.45 3911 0.78 5.29 2.98 0.85 2.18 1181 2006 4550 0.63 0.1

SO239-194-381cm 8.64 3976 0.8 5.6 3.05 0.93 2.28 1244 2047 4750 0.65 0.11

SO239-194-461cm 8.24 3383 0.88 5.43 2.97 1.2 2.38 1497 1874 4604 0.66 0.15

SO239-194-521cm 8.49 3305 1.73 4.87 3 0.77 3.30 3004 2134 4088 0.57 0.09

SO239-194-561cm 6.8 3106 1.71 11.77 2.76 2.08 3.33 5110 2554 3024 1.73 0.31

Appendix III

192

Supplementary data 3: REY sediment

REY concentrations in CCZ sediment samples.

All concentrations are given in ppm.

Salt mass

corrected

Sample ID La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu ΣREY Ce/

Ce* Y/Ho

HREE/

LREE

MREE/

MREE*

SO239-

87GC-35cm 38.7 72.0 11.1 47.6 11.6 2.5 11.6 1.8 10.9 52 2.11 5.99 0.83 5.60 0.83 276 0.8 24.9 0.091 0.104

87GC-123cm 55.6 91.6 16.8 70.7 17.3 3.6 17.1 2.6 15.5 76 2.97 8.37 1.17 7.69 1.10 388 0.69 25.7 0.091 0.110

87 GC -163cm 58.6 93.7 17.6 73.1 17.9 3.7 17.7 2.7 16.2 81 3.19 8.94 1.22 8.01 1.18 405 0.67 25.4 0.093 0.109

87 GC -203cm 54.0 96.0 16.1 67.1 16.4 - 16.4 2.5 15.1 75 2.90 8.22 1.10 7.53 1.13 - 0.75 25.7 0.090 0.099

87 GC -243cm 63.3 106.2 19.2 79.5 19.4 4.3 19.5 2.9 17.9 87 3.46 9.81 1.36 8.95 1.34 444 0.7 25.1 0.093 0.109

87 GC -283cm 59.6 98.6 17.9 74.7 18.3 4.0 18.4 2.8 17.0 84 3.25 9.32 1.33 8.63 1.26 419 0.69 25.8 0.095 0.110

87 GC -343cm 70.9 110.1 21.6 90.6 22.6 4.8 22.5 3.4 20.8 102 4.06 11.56 1.58 10.76 1.59 498 0.64 25.1 0.101 0.115

87 GC -383cm 64.5 95.7 20.3 85.4 21.2 4.6 21.1 3.3 19.4 95 3.84 10.75 1.50 9.75 1.43 458 0.6 24.7 0.103 0.118

87 GC -443cm 67.9 90.5 20.5 85.3 21.1 4.7 22.0 3.4 20.2 100 3.97 11.43 1.58 10.42 1.52 465 0.56 25.3 0.109 0.122

87 GC -483cm 75.2 103.9 23.0 96.0 23.5 5.0 24.3 3.7 22.5 107 4.32 12.38 1.73 11.19 1.65 515 0.57 24.8 0.105 0.120

87 GC -607cm 71.2 99.8 22.3 92.2 23.1 4.9 23.7 3.6 21.4 107 4.08 11.67 1.62 10.28 1.47 498 0.57 26.1 0.102 0.122

87 GC -667cm 73.3 103.8 23.5 97.8 24.5 5.3 24.9 3.8 22.0 106 4.23 11.76 1.56 10.27 1.53 515 0.57 25.1 0.098 0.123

87 GC -727cm 99.2 102.7 31.5 133.9 33.6 7.8 34.4 5.1 30.8 148 5.91 16.41 2.19 14.14 2.08 668 0.42 25 0.111 0.137

87 GC -767cm 114.3 108.9 34.2 145.5 36.1 8.2 38.2 5.6 34.1 177 6.78 18.86 2.57 16.28 2.41 749 0.4 26.1 0.116 0.136

87 GC -827cm 116.7 110.2 36.0 153.4 36.8 8.7 40.0 5.9 36.4 191 7.26 19.88 2.76 17.72 2.51 785 0.39 26.3 0.120 0.137

87 GC -887cm 129.9 112.7 39.4 167.5 40.5 9.4 43.4 6.5 39.7 208 7.73 21.58 2.93 18.68 2.73 851 0.36 26.9 0.119 0.139

165 GC -47cm 45.1 98.1 12.5 49.8 11.7 2.5 10.9 1.8 10.3 54.9 2.04 5.79 0.83 5.54 0.81 312 0.95 26.9 0.073 0.084

165 GC -127cm 46.0 101.6 12.7 50.3 11.8 2.5 11.0 1.7 10.2 51.7 1.98 5.57 0.80 5.23 0.75 314 0.97 26.1 0.068 0.083

165 GC -187cm 46.9 103.2 13.3 52.3 12.1 2.6 11.4 1.7 10.6 53.4 2.03 5.80 0.80 5.32 0.79 322 0.95 26.3 0.068 0.083

165 GC -267cm 55.3 105.6 15.5 63.1 15.0 3.3 15.1 2.3 14.2 74.2 2.75 7.98 1.11 7.32 1.07 384 0.83 27.0 0.084 0.096

165 GC -367cm 53.5 108.1 15.1 60.4 14.5 3.1 13.8 2.2 13.0 64.9 2.49 7.08 0.99 6.53 0.98 367 0.87 26.1 0.076 0.091

Appendix III

193

Supplementary data 3: REY sediment continued

Sample ID La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu ΣREY Ce/

Ce* Y/Ho

HREE/

LREE

MREE/

MREE*

SO239-

165 GC -407cm 51.0 100.6 15.0 60.9 14.7 3.2 13.8 2.2 12.7 59.9 2.41 6.91 0.97 6.43 0.95 352 0.83 24.8 0.078 0.095

165 GC -467cm 44.3 103.0 12.6 50.2 12.0 2.5 11.2 1.7 10.5 52.8 2.03 5.97 0.80 5.27 0.78 316 1.00 26.0 0.071 0.084

165 GC -517cm 48.7 99.5 13.9 55.6 13.1 2.8 12.5 1.9 11.7 57.8 2.26 6.46 0.91 5.94 0.89 334 0.88 25.6 0.076 0.090

165 GC -612cm 48.3 100.6 14.1 55.9 13.5 3.0 13.0 2.0 12.1 60.4 2.33 6.75 0.94 6.23 0.91 340 0.89 25.9 0.078 0.092

165 GC -672cm 56.3 99.2 16.5 68.9 16.7 3.8 16.7 2.5 15.7 82.0 3.05 8.82 1.25 8.08 1.21 401 0.75 26.9 0.093 0.105

165 GC -732cm 63.7 110.8 19.3 79.3 19.4 4.4 19.4 3.0 18.1 90.2 3.45 10.02 1.39 9.07 1.37 453 0.72 26.1 0.093 0.108

165 GC -792cm 110 75.1 29.5 125 28.0 6.0 30.2 4.4 27.7 155 5.58 16.11 2.25 13.97 2.07 631 0.30 27.9 0.118 0.127

165 GC -832cm 107 77.6 30.1 127 28.9 6.3 30.5 4.4 27.8 149 5.54 15.93 2.17 13.76 2.01 628 0.31 26.9 0.115 0.128

165 GC -912cm 127 75.0 33.0 140 30.4 6.8 33.8 4.8 30.9 184 6.42 18.31 2.51 15.79 2.27 711 0.27 28.7 0.121 0.127

194 GC -5cm 52.0 123.7 13.6 53.1 11.9 2.5 10.6 1.7 10.0 54.7 2.0 5.7 0.8 5.2 0.8 348 1.07 27.4 0.060 0.071

194 GC -85cm 51.7 124.8 13.8 54.1 12.3 2.7 11.1 1.7 10.4 54.5 2.0 5.8 0.8 5.3 0.8 352 1.07 27.3 0.060 0.074

194 GC -157cm 54.6 127.1 14.7 57.1 12.8 2.8 12.0 1.8 11.1 56.0 2.1 6.1 0.9 5.7 0.8 366 1.03 26.1 0.062 0.075

194 GC -201cm 55.6 129.6 15.3 61.1 14.0 3.0 12.9 1.9 12.1 60.9 2.3 6.7 0.9 6.1 0.9 383 1.02 26.0 0.065 0.079

194 GC -261cm 62.8 146.1 17.4 69.2 16.0 3.5 14.9 2.2 13.7 68.0 2.6 7.5 1.0 6.8 1.0 433 1.02 25.8 0.065 0.080

194 GC -321cm 61.2 138.6 17.8 72.1 17.2 3.8 16.2 2.4 14.7 75.0 2.9 8.0 1.1 7.1 1.0 439 0.96 26.1 0.070 0.088

194 GC -381cm 66.9 141.9 19.5 78.6 18.6 4.2 17.8 2.7 16.0 81.3 3.1 8.8 1.2 7.8 1.1 470 0.90 26.0 0.072 0.090

194 GC -461cm 77.5 157.3 23.3 95.7 22.8 5.3 22.2 3.3 20.1 99.4 3.8 10.6 1.5 9.3 1.4 553 0.85 26.0 0.075 0.097

194 GC -521cm 94.7 122.9 28.5 118.8 28.1 6.6 28.6 4.2 25.5 136.3 5.0 14.3 2.0 12.6 1.8 630 0.54 27.2 0.098 0.116

194 GC -561cm 138.0 137.0 38.8 163.1 36.7 8.7 37.9 5.5 34.1 187.2 6.8 19.7 2.7 16.9 2.5 836 0.43 27.3 0.102 0.117

Appendix III

194

Supplementary data 3: Leaching

Major element and rare earth element data for leached samples [ppm]

nm: not measured

Sample ID Fe Mn P La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu Ce/

Ce*

Y/

Ho

HREE/

LREE

MREE/

MREE*

Na acetate

87_163 17.6 80.6 1122 37.2 13.1 13.2 57.8 15.3 3.85 16.8 2.52 15 79.9 2.87 8.02 1.08 7.1 1.06 0.13 27.8 0.166 0.189

Na acetate

87_827 3.2 570 3247 117 34.4 37.9 168 42.6 10.7 47.6 6.98 42.3 235 8.2 23.1 3.1 19.7 2.95 0.12 28.7 0.160 0.181

Na acetate

165_407 5.87 457 623 22 4.75 8.52 37.6 10.2 2.55 10.7 1.59 9.42 43.9 1.73 4.81 0.637 4.16 0.612 0.08 25.4 0.164 0.203

Na acetate

165_812 21.7 112 4662 96 9.52 27.9 122 28.3 6.96 31.4 4.58 28 153 5.55 15.9 2.12 13.3 1.97 0.04 27.6 0.152 0.169

Na acetate

194_157 9.08 8.28 576 20.5 4.3 7.42 32.9 8.67 2.17 9.25 1.39 8.01 42.5 1.54 4.24 0.567 3.59 0.535 0.08 27.6 0.161 0.195

Na acetate

194_561 3 6.37 2748 73.9 1.3 22.8 108 27.2 6.69 31 4.4 26.8 152 5.36 14.9 1.93 11.9 1.79 0.01 28.4 0.174 0.199

0.1M HA

87_163 355 4999 nm 1.25 3.55 0.422 1.81 0.461 0.123 0.468 0.0694 0.422 2.16 0.079 0.228 0.0281 0.208 0.0257 1.11 27.3 0.081 0.102

0.1M HA

87_827 290 8553 nm 4.22 5.96 1.38 5.89 1.47 0.37 1.56 0.226 1.4 7.15 0.256 0.736 0.1 0.683 0.0973 0.56 27.9 0.107 0.130

0.1M HA

165_407 568 55286 nm 1.49 6.69 0.476 1.99 0.476 0.123 0.469 0.074 0.451 1.67 0.0842 0.251 0.0343 0.259 0.0347 1.81 19.8 0.062 0.070

0.1M HA

165_812 1686 15842 nm 5.87 15.3 1.64 6.37 1.34 0.315 1.31 0.193 1.18 6.33 0.23 0.683 0.0942 0.63 0.0878 1.13 27.5 0.059 0.070

0.1M HA

194_157 1180 4530 nm 0.943 6.66 0.31 1.27 0.315 0.075 0.324 0.0498 0.293 1.43 0.0551 0.165

below

LOQ 0.146

below

LOQ 2.80 26.0 0.040 0.055

0.1M HA

194_561 4792 17706 nm 16 30.9 4.24 16.1 2.97 0.641 2.69 0.394 2.39 12.3 0.469 1.4 0.202 1.33 0.183 0.86 26.2 0.053 0.064

1M HA

87_163 2032 827 68 3.33 42.1 0.897 3.31 0.712 0.164 0.663 0.102 0.609 2.79 0.111 0.333 0.0456 0.358 0.0474 5.60 25.1 0.018 0.022

1M HA

87_827 1776 1936 130 9.97 62.5 2.7 9.63 2.05 0.488 2 0.326 1.97 9.77 0.361 1.05 0.155 1.08 0.146 2.77 27.1 0.033 0.039

1M HA

165_407 1977 4225 109 3.46 39.9 0.932 3.45 0.788 0.17 0.671 0.113 0.658 2.27 0.118 0.356 0.0537 0.387 0.0553 5.11 19.2 0.020 0.025

1M HA

165_812 6302 2287 144 8.17 40.9 2 6.67 1.21 0.247 1.03 0.166 1.04 4.99 0.2 0.596 0.0887 0.619 0.0763 2.33 25.0 0.027 0.031

1M HA

194_157 3658 666 87 4.09 53.1 1.07 3.9 0.845 0.17 0.747 0.118 0.693 2.84 0.127 0.383 0.0582 0.402 0.0538 5.85 22.4 0.016 0.020

1M HA

194_561 22188 2487 692 36 82.6 9.45 32.1 5.98 1.27 5.02 0.88 5.42 23.6 1.04 3.14 0.486 3.34 0.449 1.03 22.7 0.053 0.055

Appendix III

195

Supplementary data 3: Leaching continued

Sample ID Fe Mn P La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu Ce/

Ce*

Y/

Ho

HREE/

LREE

MREE/

MREE*

Na-dithionite

87_163 6371 81.3

Too

low 0.469 2.4

below

LOQ 0.303

below

LOQ

below

LOQ -

below

LOQ - 0.3

below

LOQ -

below

LOQ

below

LOQ

below

LOQ

Na-dithionite

87_827 6563 192

Too

low 0.889 4.19

below

LOQ 0.626

below

LOQ -

below

LOQ

below

LOQ

below

LOQ 0.598

below

LOQ

below

LOQ

below

LOQ

below

LOQ

below

LOQ

Na-dithionite

194_561 56683 209 1263 2.27 6.62 0.575 2.07 0.399

below

LOQ 0.369

below

LOQ 0.461 2.15

below

LOQ 0.294

below

LOQ 0.386

below

LOQ 0.059 0.050

NH4 oxalat

87_163 2664 34

Too

low 0.601 1.21 0.129 0.484 0.0859

below

LOQ 0.0679

below

LOQ 0.05 0.202

below

LOQ

below

LOQ

below

LOQ -

below

LOQ 1.00 0.042

NH4 oxalat

87_827 3213 114

Too

low 1.07 2.09 0.248 0.864 0.151 0.03 0.123

below

LOQ 0.104 0.377

below

LOQ 0.04

below

LOQ 0.04

below

LOQ 0.94 0.019 0.046

NH4 oxalat

165_407 2271 42.7

Too

low 0.614 1.26 0.134 0.49 0.0843

below

LOQ 0.0711

below

LOQ 0.0467 0.192

below

LOQ -

below

LOQ 0.0219

below

LOQ 1.01 0.009 0.040

NH4 oxalat

165_812 4054 18.9

Too

low 0.218 0.61

0.048

1 0.19 0.0312 - 0.0262

below

LOQ

below

LOQ

below

LOQ

below

LOQ -

below

LOQ

below

LOQ

below

LOQ 1.37 0.000 0.027

NH4 oxalat

194_157 2883 19.4

Too

low 0.882 1.83 0.206 0.719 0.129

below

LOQ 0.104

below

LOQ 0.0768 0.291

below

LOQ

0.034

5

below

LOQ 0.0313

below

LOQ 0.99 0.018 0.042

NH4 oxalat

194_561 3956 18.7

Too

low 0.399 0.876

0.092

9 0.322 0.0634

below

LOQ 0.0566

below

LOQ 0.0568 0.208

below

LOQ

0.026

4

below

LOQ 0.0269

below

LOQ 1.05 0.032 0.051

Na-dithionite

87_163 6371 81.3

Too

low 0.469 2.4

below

LOQ 0.303

below

LOQ

below

LOQ -

below

LOQ - 0.3

below

LOQ -

below

LOQ

below

LOQ

below

LOQ

Na-dithionite

87_827 6563 192

Too

low 0.889 4.19

below

LOQ 0.626

below

LOQ -

below

LOQ

below

LOQ

below

LOQ 0.598

below

LOQ

below

LOQ

below

LOQ

below

LOQ

below

LOQ

Na-dithionite

165_407 5813 237

Too

low 0.491 2.65

below

LOQ 0.349 - -

below

LOQ

below

LOQ - 0.308

below

LOQ -

below

LOQ -

below

LOQ

Sum leaching

87_163 1190 42.9 62.4 14.6 63.7 16.6 4.1 18.0 2.7 16.1 85.4 3.1 8.6 1.2 7.7 1.1

87_827 3377 133.1 109.1 42.2 185.0 46.3 11.6 51.3 7.5 45.8 252.9 8.8 24.9 3.4 21.5 3.2

165_407 732 28.1 55.3 10.1 43.9 11.5 2.8 11.9 1.8 10.6 48.3 1.9 5.4 0.7 4.8 0.7

165_812 4806 110.6 68.2 31.6 135.5 30.9 7.5 33.8 4.9 30.2 164.6 6.0 17.2 2.3 14.5 2.1

194_157 663 27.1 69.5 9.0 39.2 10.0 2.4 10.4 1.6 9.1 47.6 1.7 4.8 0.6 4.2 0.6

194_561 4703 128.6 122.3 37.2 158.6 36.6 8.6 39.1 5.7 35.1 190.3 6.9 19.8 2.6 17.0 2.4

Appendix III

196

% leached

Sample ID P La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu Na acetate 87_163 83 63 14 75 79 85 105 95 95 93 99 90 90 89 89 90 Na acetate 87_827 120 100 31 105 110 116 124 119 118 116 123 113 116 112 111 118 Na acetate 165_407 77 43 5 57 62 70 80 77 74 74 73 72 70 66 65 65 Na acetate 165_812 97 87 13 95 98 101 115 104 103 101 98 100 99 94 95 95 Na acetate 194_157 59 38 3 51 58 68 78 77 77 72 76 72 69 63 63 65 Na acetate 194_561 54 54 1 59 66 74 77 82 80 79 81 78 76 72 70 73 0.1M HA 87_163 2 4 2 2 3 3 3 3 3 3 2 3 2 3 2 0.1M HA 87_827 4 5 4 4 4 4 4 4 4 4 4 4 4 4 4 0.1M HA 165_407 3 7 3 3 3 4 3 3 4 3 3 4 4 4 4 0.1M HA 165_812 5 20 6 5 5 5 4 4 4 4 4 4 4 5 4 0.1M HA 194_157 2 5 2 2 2 3 3 3 3 3 3 3 3

0.1M HA 194_561 12 23 11 10 8 7 7 7 7 7 7 7 7 8 7 1M HA 87_163 5 6 45 5 5 4 4 4 4 4 3 3 4 4 4 4 1M HA 87_827 5 9 57 8 6 6 6 5 6 5 5 5 5 6 6 6 1M HA 165_407 14 7 40 6 6 5 5 5 5 5 4 5 5 6 6 6 1M HA 165_812 3 7 54 7 5 4 4 3 4 4 3 4 4 4 4 4 1M HA 194_157 9 7 42 7 7 7 6 6 7 6 5 6 6 7 7 6 1M HA 194_561 14 26 60 24 20 16 15 13 16 16 13 15 16 18 20 18 Na-dithionite 87_163 1 3 0 0

Na-dithionite 87_827 1 4 0 0




Na-dithionite 194_561 25 2 5 1 1 1 1 1 1 1 2

Appendix III

197

% leached continued

Sample ID P La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu

NH4 oxalat 87_163 1 1 1 1 0 0 0 0

NH4 oxalat 87_827 1 2 1 1 0 0 0 0 0 0 0

NH4 oxalat 165_407 1 1 1 1 1 1 0 0 0

NH4 oxalat 165_812 0 1 0 0 0 0

NH4 oxalat 194_157 2 1 1 1 1 1 1 1 1 1

NH4 oxalat 194_561 0 1 0 0 0 0 0 0 0 0

Bulk 87_163 1347 58.6 93.7 17.6 73.1 17.9 3.68 17.7 2.65 16.2 81.1 3.19 8.94 1.22 8.01 1.18 87_827 2708 117 110 36 153 36.8 8.65 40 5.92 36.4 191 7.26 19.9 2.76 17.7 2.51

165_407 804 51.0 100.6 15.0 60.9 14.7 3.2 13.8 2.2 12.7 59.9 2.4 6.9 1.0 6.4 0.9

165_812 (REY from 792cm) 4793 110 75.1 29.5 125 28.0 6.0 30.2 4.4 27.7 155 5.58 16.11 2.25 13.97 2.07

194_157 972 54.6 127.1 14.7 57.1 12.8 2.8 12.0 1.8 11.1 56.0 2.1 6.1 0.9 5.7 0.8 194_561 5110 138.0 137.0 38.8 163.1 36.7 8.7 37.9 5.5 34.1 187.2 6.8 19.7 2.7 16.9 2.5

No bulk REY data for 165GC 812 cm. To calculate the approximate % that was leached, REY data from 792 cm was used. REY concentrations in this

layer are similar to the next sampled layers below 812 cm and it can therefore be assumed that the REY content is representative.

Appendix III

198

Supplementary data 3: REY pore water

REY concentrations in pore water samples from 194GC. All

concentrations are given in mg/kg.

Sample ID

SO239 La Ce Pr Nd Sm Eu Gd Tb Dy Y Ho Er Tm Yb Lu

194GC 511cm 3.20E-04 4.85E-05 8.55E-05 3.96E-04 8.83E-05 1.04E-04 9.47E-05 6.35E-04 2.01E-05 5.44E-05 6.22E-06 4.68E-05 5.66E-06

194GC 551cm 5.52E-05 5.63E-05 1.23E-05 6.62E-05 1.50E-05 1.22E-05 1.21E-05 7.93E-05 4.22E-06

194GC 571cm 7.57E-05 3.86E-05 1.79E-05 7.81E-05 1.58E-05 1.58E-05 1.56E-05 1.17E-04 1.01E-05 8.14E-06

Average Blank 6.64E-06 3.06E-05 2.51E-06 9.83E-06 6.61E-06 3.21E-06 6.23E-06 5.53E-06 2.52E-06 8.04E-06 1.52E-06 2.49E-06 9.72E-07 2.52E-06 1.66E-06

Sample ID

SO239 Ce/Ce* Y/Ho

HREE/

LREE

MREE/

MREE*

194GC 511cm 0.07 3.16E+01 0.157 0.146

194GC 551cm 0.50 0.022 0.101

194GC 571cm 0.24 0.087 0.103

Appendix IV

199

Appendix IV

Supplementary material to Rare earth elements and yttrium in metalliferous and calcium-

carbonate-rich sediments from the central equatorial Pacific (Chapter 5).

Appendix IV

200

Table 1: Replicate analyses of NIST-2702 certified reference material (CRM) during ICP-OES

measurements (n=12 digested samples, each measured once, 2 ICP-OES runs).

LOQ# [mg/L] Element NIST-2702 certified [mg/kg]


Accuracy (%)

Precision (%)

99-324 Al 84100±2200 77954±1526 -7 2

2-11 Ba 397.4±3.2 376±8.2 -5 2

19-34 Caa 3430±240 3464±63 1 2

33-101 Fe 79100±2400 75811±3079 -4 4

369-892 K 20540±720 19312±525 -6 3

0.5-1 Mn 1757±58 1679±21 -4 1

21-30 P 1552±66 1527±28 -2 2

217-414 Sb 15000 13006±306 -13 2

9-41 Ti 8840±820 8368±164 -5 2 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run a reference value b information value

Table 2: Replicate analyses of NIST-2702 CRM during ICP-MS measurements (n=6 digested

samples; 24 measurements in 2 ICP-MS runs). Accuracy and precision calculated from

averages of each digested sample.

LOQ# [mg/kg] Element NIST-2702 certified [mg/kg]


Accuracy (%)

Precision (%)

0.029-0.047 Y --- 36.4±0.6 2

0.005-0.009 La 73.5±4.2 70.7±1.5 -4 2

0.019-0.032 Ce 123.4±5.8 124.4±3.0 1 2

0.004-0.007 Pr --- 16.3±0.4 2

0.021 Ndb 56 61±1.4 9 2

0.012-0.017 Smb 10.8 11.4±0.2 6 2

0.005-0.007 Eu --- 2.02±0.03 2

0.010-0.011 Gd --- 9.65±0.17 2

0.008 Tb --- 1.34±0.03 2

0.009-0.013 Dy --- 7.59±0.10 1

0.004-0.01 Ho --- 1.43±0.02 2

0.005-0.01 Er --- 4.04±0.05 1

0.002-0.009 Tm --- 0.57±0.01 2

0.008-0.009 Yb --- 3.77±0.05 1

0.004-0.009 Lu --- 0.55±0.01 1 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Appendix IV

201

Table 3: Replicate analyses of BHVO-2 CRM during ICP-MS measurements (n=1 digested

sample; 8 measurements during 1 ICP-MS run). Accuracy and precision calculated from

averages of multiple measurements of 1 digested sample.

LOQ# [mg/kg] Element BHVO-2 recommended [mg/kg]


Accuracy (%)

Precision (%)

0.029-0.047 Y 26±2 24.3±0.1 -7 0.4

0.005-0.009 La 15±1 15.3±0.1 2 0.8

0.019-0.032 Ce 38±2 39±0.2 2 0.5

0.004-0.007 Pr --- 5.4±0.03 0.5

0.021 Nd 25.0±1.8 25.0±0.07 0 0.3

0.012-0.017 Smb 6.2±0.4 6.28±0.03 1 0.5

0.005-0.007 Eu --- 2.14±0.01 0.3

0.010-0.011 Gdb 6.3±0.2 6.53±0.02 4 0.3

0.008 Tbb 0.9 0.96±0.004 6 0.4

0.009-0.013 Dy --- 5.5±0.04 0.8

0.004-0.01 Hob 1.04±0.04 1.00±0.01 -4 0.7

0.005-0.01 Er --- 2.61±0.01 0.4

0.002-0.009 Tm --- 0.33±0.004 1.3

0.008-0.009 Ybb 2.0±0.2 2.04±0.02 2 0.9

0.004-0.009 Lub 0.28±0.01 0.28±0.001 -1 0.3 # LOQ: limit of quantification; 10*standard deviation of acid blanks for each run b information value

Documents

Biogeochemistry of Pacific deep-sea sediments and