phosphorus cycling in marine sediments from the continental margin off namibia

Marine Geology 274 (2010) 95–106

Contents lists available at ScienceDirect

Marine Geology

j ourna l homepage: www.e lsev ie r.com/ locate /margeo

Phosphorus cycling in marine sediments from the continental margin off Namibia

Kathrin Küster-Heins a,⁎, Ekkehard Steinmetz a, Gert J. De Lange c, Matthias Zabel b

a Department of Geosciences, University of Bremen, Klagenfurter Strasse, 28334 Bremen, Germanyb Marum—Center for Marine Environmental Sciences, University of Bremen, Leobener Strasse, 28359 Bremen, Germanyc Institute of Earth Sciences—Geochemistry, University of Utrecht, Budapestlaan 4, 3584 Utrecht, The Netherlands

⁎ Corresponding author. Tel.: +49 421 218 65104.E-mail address: [email protected] (K. Küster-

0025-3227/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.margeo.2010.03.008

a b s t r a c t
a r t i c l e i n f o
Article history:Received 25 June 2009Received in revised form 25 February 2010Accepted 13 March 2010Available online 20 March 2010

Communicated by D.J.W. Piper

Keywords:benthic phosphorus cyclebenthic iron cycleparticle mixingBenguela Upwelling System

In this study we investigate benthic phosphorus cycling in recent continental margin sediments at three sitesoff the Namibian coastal upwelling area. Examination of the sediments reveals that organic and biogenicphosphorus are the major P-containing phases preserved. High Corg/Porg ratios just at the sediment surfacesuggest that the preferential regeneration of phosphorus relative to that of organic carbon has either alreadyoccurred on the suspension load or that the organic matter deposited at these sites is already ratherrefractory. Release of phosphate in the course of benthic microbial organic matter degradation cannot beidentified as the dominating process within the observed internal benthic phosphorus cycle. Dissolvedphosphate and iron in the pore water are closely coupled, showing high concentrations below theoxygenated surface layer of the sediments and low concentrations at the sediment–water interface. Theabundant presence of Fe(III)-bound phosphorus in the sediments document the co-precipitation of bothconstituents as P-containing iron (oxyhydr)oxides. However, highly dissolved phosphate concentrations inpore waters cannot be explained, neither by simple mass balance calculations nor by the application of anestablished computer model. Under the assumption of steady state conditions, phosphate release rates aretoo high as to be balanced with a solid phase reservoir. This discrepancy points to an apparent lack of solidphase phosphorus at sediment depth were suboxic conditions prevail. We assume that the known, active,fast and episodic particle mixing by burrowing macrobenthic organisms could repeatedly provide themicrobially catalyzed processes of iron reduction with authigenic iron (oxyhydro)oxides from the oxicsurface sediments. Accordingly, a multiple internal cycling of phosphate and iron would result before bothelements are buried below the iron reduction zone.

Heins).

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Phosphorus (P) is one of the essential elements for life on Earthand plays an important role in the biological productivity in oceansand on continents. Because of the close relation between oceanicproductivity and atmospheric CO2, changes in the oceanic P-cyclemayconsiderably affect the chemistry of the oceans and atmospherethroughout geological time (e.g. Van Cappellen and Ingall, 1994,1996). The major initial input into the oceans is by riverine transportof organic and inorganic P compounds (Benitez-Nelson, 2000). A greatportion of this P is trapped in estuarine and coastal shelf areas anddoes not reach the open ocean (Ruttenberg, 1993). Paytan andMcLaughlin (2007) estimated that almost 99% of particulate and 25%of dissolved P are buried in coastal shelf zones. The remainingparticulate and dissolved P is exported to the deep ocean. During thistransport the predominant portion of P, which is associated withorganic substances is decomposed microbially within the water

column (Heggie et al., 1990; Baturin, 2003; Paytan and McLaughlin,2007). As a result, P is released and subject to transformationprocesses (biological recycling, ad-, desorption) between dissolvedand particulate phases (Faul et al., 2005). However, a small proportionof the P becomes buried into the sediment as altered organiccompounds and as detrital and amorphous P mineral phases,adsorbed to particle surfaces (e.g. clay minerals) or iron (oxyhydr)oxides, or is transformed in situ into a mineralized form (variousauthigenic apatites)(Froelich, 1988; Compton et al., 2000; Paytan andMcLaughlin, 2007). After burial, the behaviour of P is controlled by anumber of biogeochemical processes. Release of phosphate to porewaters can result from: (1) microbial degradation of organic matter(Ingall et al., 1993; Jensen et al., 1995; Canfield et al., 2005); (2) bypolyphosphate accumulating bacterial species (Schulz and Schulz,2005; Diaz et al., 2008); (3) desorption from iron (oxyhydr)oxideswhen iron reduction takes place (e.g. Froelich et al., 1982; Baturin, 2003;Paytan and McLaughlin, 2007) and (4) dissolution of P-containingmineral forms during subsequent alteration (Suess, 1981; Froelich et al.,1988; Schenau and De Lange, 2001; Baturin and Dubinchuk, 2003).Under oxic conditions, phosphate concentrations in interstitial watersare predominantly controlled by a complex interplay of the release from

mailto:[email protected]

http://dx.doi.org/10.1016/j.margeo.2010.03.008

http://www.sciencedirect.com/science/journal/00253227

96 K. Küster-Heins et al. / Marine Geology 274 (2010) 95–106

organic debris, metabolic uptake, adsorption onto particle surfaces anddiffusive transport. Furthermore, bioturbative transport processes caninduce a permanent internal cycling (dissolution–reprecipitation) ofparticulate iron-phosphate phases within the oxidized and reducedsediment layer (Slomp and Van Raaphorst, 1993). Under sub- to anoxicconditions dissolved phosphate may continued to originate from thedecomposition of organic substances, butmay alsobe released from iron(oxyhydr)oxides when iron reduction takes place. However, assumingsteady state, the accumulation of particulate P and release of phosphateinto pore andbottomwaters shouldbebalancedby theburial rate of P inthe sediment.

The shelf area off Namibia, as well as off Chile/Peru, belongs to themost productive systems in the modern ocean, where strongupwelling of nutrient-rich waters causes high primary production insurfacewaters,which sustain large zooplankton andpelagicfish stockson the continentalmargin. Additionally, continentalmargin sedimentsare themain repository for remineralizedmarine organic carbon and aprimary sink for phosphate (Ruttenberg, 1993; Filippelli, 1997). So, thehigh availability of organic material and the resulting very intensebenthic activity make coastal high productive areas very suitable forinvestigations on the benthic P-cycle. Surface sediments of these shelfareas have previously been described as having favourable conditionsfor recent authigenic apatite formation (Burnett, 1977; Calvert andPrice, 1983; Froelich et al., 1988; Glenn andArthur, 1988; Bremner andRogers, 1990; Borchers et al., 2005; Schulz and Schulz, 2005; Arninget al., 2008). Apart from impermeable hardgrounds where phospho-genesis can also occur, an increase of pore water phosphate is dis-cussed as a prerequisite for the precipitation of authigenic apatite (e.g.Ingall and Jahnke, 1997; Canfield et al., 2005; Paytan and McLaughlin,2007; Diaz et al., 2008). Moreover, Schulz and Schulz (2005) havefound indications that the alteration between oxic and anoxicconditions in bottom waters and sediments in particular off Namibiais responsible for the redox-sensitive microbial formation and re-mineralization of polyphosphates (i.e. by Thiomargarita namibiensis).

Despite the aforementioned suitable conditions for phosphogen-esis in shelf sediments, we investigate sampling sites without anyevidence for active and ongoing precipitation of authigenic apatite. Inthis study we therefore focus on the benthic phosphorus cycle incontinental margin sediments at three locations off the coast ofNamibia under consideration of (1) the speciation of sedimentary P incontinental margin surface sediments in a high productivity regionwith oxygenated bottom waters, (2) the comparison of distributionsof P species with concentration profiles of interstitial waters, (3)obtaining a further insight into the sources and sinks of particulate Pand (4) the interpretation and quantification of processes determin-ing the distribution and behaviour of P. For the last, a model was usedwhich has been developed by Slomp et al. (1996).

2. Location and oceanic environment

Sediment sampling has been carried out during the RV Meteorcruise M34/2 (Schulz et al., 1996) along the Namibian coastal region.The upwelling area off Namibia is regarded as the most productive ofthe four major eastern boundary systems on Earth (BenguelaUpwelling System; Bianchi et al., 1999). Intense marine productivityis induced by the persistent SE trade wind field, which drives surfacewaters offshore and causes the upwelling of the nutrient-rich SouthAtlantic CentralWater in at least fourmajor cells. Due to seasonality inwind force, these cells show seasonal patterns with a correspondingchange of the geochemical conditions in bottomwaters (Shannon andNelson, 1996). As a result, coastal sediments are enriched in organicsubstances. A number of studies has described sediments in this areaas mainly consisting of organic-rich, marine, calcareous facies (i.e.foraminifers and coccoliths), barely diluted by terrigenous material(Calvert and Price, 1983; Bremner and Willis, 1993; Wefer et al.,1998). But, the offshore extension of phytoplankton filaments, lateral

particle advection in the water column, and redistribution of re-suspended particulate organic matter, lead to high total organiccarbon contents in upper margin sediments as well. A reflection ofthese processes is a distinct depocenter of organic-rich sedimentsbetween 600 and 1500 m water depth, centered at about 25.5 °S (cf.Fig. 1; Inthorn et al., 2006; Mollenhauer et al., 2007).

3. Materials and methods

3.1. Sampling

Analytical work was done on material that has been recoveredfrom about 1300 m water depth at three locations along thecontinental margin off Namibia (Fig. 1, Table 1). Sediment wassampled using a multi-corer. Sediments retrieved in this area are ofHolocene age (Mollenhauer et al., 2002). Cores were transferred to therefrigerated on-board laboratory immediately after recovery andwere processed at a temperature of about 4 °C. Overlying bottomwaters were sampled, filtered and stored for analysis. For pressurefiltration under argon atmosphere (5 bar), Teflon- and PE-squeezersand 0.2 μm cellulose acetate filters were used to collect the porewater. Sub-sampling for further solid phase investigations was doneat every 0.5 cm (top 0–3 cm of the core), 1 cm (between 3 and 5 cm)and 5 cm (below 5 cm).

3.2. Pore water analyses

Onboard RVMeteor, pore water phosphate, ferrous iron and nitratewere measured photometrically immediately after pore watersampling, with an autoanalyzer using standard methods (Stricklandand Parsons, 1968; Grasshoff et al., 1983; Schulz, 2006). In situdetermination of dissolved oxygen in pore water was carried outusing microelectrodes (Wefer et al., 1997).

3.3. Solid phase analyses

Bulk concentrations of trace elements (P and iron (Fe)) weredetermined by total digestion of 50 mg dried and ground sample in aHNO3 (65%), HCl (30%) and HF (47–51%) mixture using a Microwavesystem (Zabel et al., 1999; Schulz, 2006). Dissolved elementconcentrations were measured using an inductively coupled plasmaatomic emission spectrometer ICP-OES (Perkin Elmer Optima 3000).Application of standard reference material assured the accuracy ofmeasurements; the precision was better than 5%.

The speciation of sedimentary phosphorus in the sediments wasdetermined using a five-step sequential extraction scheme, where∼125 mg of dried and ground sediment sample was successivelywashed with 25 ml (1) 2 M NH4Cl (pH 7; 10 times at 4 h), (2) citratedithionite buffer (pH 7.5; 8 h), (3) 1 M sodium acetate buffer (pH 4;6 h), (4) 1 M HCl (24 h), and (5) 1 M HCl (24 h) after ignition at550 °C (2 h)(Table 2). After steps (2) and (3) the samples were rinsedrepeatedly with 25 ml 2 M NH4Cl and 25 ml demineralized water(2 h) to prevent the readsorption of phosphate. In addition, to test if asignificant portion of total extractable P is exchangeable or looselysorbed, we extracted ∼125 mg of a sub-sample with 25 ml 0.35 MNaCl ((1b); 2 h). The extraction analysis is in accordance with thescheme of Ruttenberg (1992), and the modification of step (1) bySchenau and De Lange (2000) (Table 2). P concentrations inextraction solutions were measured photometrically (Perkin Elmer550SE Spectrophotometer). Solutes from step (1) were measured forP and calcium and solutes from step (2) for P and Fe using aninductively coupled plasma atomic emission spectrometer ICP-OES(Perkin Elmer Optima 3000). Precision was generally better than 5%,except for step (2) (3–15%).

Biogenic apatite (i.e. hydroxyapatite in bones, scales, shells:biogenic debris) is soluble with NH4Cl extraction once all soluble

Fig. 1. A map of the working area showing the approximate position of investigated GeoB-sites (dotes). The grey-lined areas indicate two distinct areas off Namibia: the mudbelt(Schulz and Schulz, 2005) and the depocenter (Inthorn et al., 2006; Mollenhauer et al., 2007; Aspetsberger et al., 2007).

97K. Küster-Heins et al. / Marine Geology 274 (2010) 95–106

calcium carbonate was extracted. Therefore, ten successive extrac-tions with NH4Cl are required to determine in which NH4Cl step thebiogenic apatite fraction is extracted (Fig. 2; Schenau and De Lange,2000). High concentrations of P and Ca were measured after the firstextraction with NH4Cl, indicating the presence of exchangeable P and/or Ca-associated P (Table 2). A second maximum was measured afterthe sixth extraction, probably extracting P associated with biogenic

Table 1Location and water depth (m), surface Corg and CaCO3 content (wt.%).

GeoB site Latitude Longitude Water depth(m)

Corg(wt.%)

CaCO3

(wt.%)

3707 21°37.4′S 12°11.6′E 1357 4.0 56.33718 24°53.7′S 13°09.7′E 1323 6.6 63.53702 26°47.6′S 13°27.2′E 1306 3.7 58.5

Table 2Sequential extraction scheme (Ruttenberg, 1992; Schenau and De Lange, 2000).

Step Reagents

(1) Biogenic 25 ml 2 M NH4Cl (pH 7)

(1b) Exchangeable 25 ml 0.35 M NaCl(2) Fe(III)-bound 25 ml citrate dithionite buffer (pH 7.5), 25 ml 2 M NH4Cl, 25 m(3) Authigenic 25 ml 1 M Na-acetate (pH 4), 25 ml 2 M NH4Cl, 25 ml dem. wa(4) Detrital 25 ml 1 M HCl, 25 ml dem. water(5) Organic-bound After ignition at 550 °C 25 ml 1 M HCl

apatite. The remaining P after the tenth extraction with NH4Cl mayresult from incomplete dissolution caused by saturation of theextraction solution.

Total organic carbon content was determined via the difference oftotal carbon and inorganic carbon by a Coulomat 702-GA/LI(Ströhnlein Instruments).

3.4. Diagenetic model for P cycling in surface sediments

To identify and quantify the interrelating processes in thesesediment pore water systems, we applied a steady state diageneticmodel for P cycling developed by Slomp et al. (1996). The steady statemodel describes the one-dimensional concentration change withdepth of dissolved pore-water phosphate, sedimentary oxide-associ-ated P, authigenic P and organic P. Assuming that extraction with

P component extracted

Remaining pore water P, exchangeable P, biogenic apatite, apatiteprecursor mineral, carbonate-associated PExchangeable, loosely sorbed P

l dem. water Adsorbed and reducible/reactive Fe(III)-bound Pter Authigenic P

Detrital POrganic P

Fig. 2. Sequential extraction of phosphorus (P; dots) and calcium (Ca; crosses) with NH4Cl (μmol/g dry sediment), multiple analysis of each sample. Solid lines represent the averagevalue of each single extraction step. Note the different scale for P concentration for GeoB 3718.


NH4Cl and NaCl (step 1 and step 1b, respectively, Table 2) determinesimilar exchangeable P contents, the difference between step 1 and 1bgives a better estimation of P solely bound to biogenic apatite and Ca-associated P.We used biogenic P (step 1minus 1b) plus the authigenicP fraction (extracted during step 3) for the model calculation asauthigenic P concentration. A constant background organic Pconcentration is used for a better fit for extracted organic P. Thisorganic, residual P is solely transported through the sediment column,is not subject to degradation processes and does not contribute topore water phosphate. The sediment column is divided into threezones: an oxidized surface zone (zone I: 0≤x≥L1), a reducedsediment zone with bioturbation (zone II: L1≤x≥L2) and a reducedsediment zone without bioturbation (zone III: xNL2). Transport ofsedimentary P fractions occurs through bioturbation or physicalmixing (biodiffusion) and sediment accumulation. The first-orderreactions, with reaction rate constants ks, kg, km and ka, respectively,are: (1) phosphate release from organic matter due to organic matterdegradation in all three zones; (2) reversible sorption of phosphate toiron (oxyhydr)oxides in the first zone; (3) phosphate release due toreductive dissolution of iron (oxyhydr)oxides in the bioturbatedzones; and (4) authigenic P precipitation in the reduced sedimentzones. The differential equations used for each P reservoir arepresented in Slomp et al. (1996). Detailed descriptions of the valuesand their sources for constant and variable parameters used for themodel calculation are displayed in Tables 3 and 4. Values oftemperature, the depths of layers 1–3, Φ, Dsed, Db, ω, Co, CS, ps werefixed on the basis of existing data. The other parameters (JMX, JGX, JAX,ks, kg, km and ka) were varied to fit the model to measured profiles.

Table 3Values of constant parameter.

Constant values

Parameter Description

Temp. Top sediment sampleLayer 1 Boundary oxidized/reduced zone with bioturbationLayer 2 Boundary bioturbation/ reduced zoneLayer 3 Reduced zone, no bioturbationϕ Sediment porosityDsed Sediment diffusion coefficientDb Sediment mixing or biodiffusion coeff.ω Sedimentation rateC0 Bottom water HPO4

2−conc.Cs Equilibrium conc. for P sorptionρs Density for conversion factor theta (g cm−3 dry sed. per cm−3 pore wat

*Source: (1) onboard measurement; (2) depth where NO3−starts to decrease (Fig. 3); (3) ass

and Gregory 1974; (7)Middelburg et al. 1997; (8)Wefer et al. 1998; (9) Froelich 1988, Slompto be constant with depth in each relevant zone.

4. Results

4.1. Distribution of major diagenetic pore water compounds in surfacesediments

The pore-water profiles obtained at sites from the continentalmargin off Namibia reflect the pore-water changes during oxic andsuboxic diagenesis of organic matter (Fig. 3). The measured bottomwater oxygen concentration was around 200 μmol/l at all stations;oxygen penetration into the sediment varies between 8 mm (GeoB3718) and 17 mm (GeoB 3702). At all stations, nitrification leads to apeak in nitrate within the uppermost sediment layer. At depth nitrateis completely removed by biogeochemical processes (e.g. denitrifica-tion and/or dissimilatory nitrate reduction). Phosphate and ferrousiron concentrations reached a maximum around 2.5–4 cm at allstations (19–26 μmol/l and 2.9–7.2 μmol/l, respectively), whereconditions become more reducing. Below this depth phosphate andiron are removed from pore water as deduced from decreasing porewater concentrations.

4.2. Distribution of sedimentary P and Fe in surface sediments

In all three cores total sedimentary P (Ptotal) concentrations arehighest close to the sediment surface (27.6–33.8 μmol/g) anddecrease very slightly with sediment depth (Fig. 4). Compared toother continental margin sediments, Ptotal concentrations are higherthan previously measured contents (Filippelli and Delaney, 1996; Vander Zee et al., 2002; Cha et al., 2005). The maximum in total

Unit GeoB 3707 GeoB 3718 GeoB 3702 Source*

°C 6 6 5 1cm 1.8 1.4 1 2cm 5.5 6.5 5 3cm 20 20 20 4cm3 cm−3 0.88 0.91 0.92 5cm2 d−1 4.3⁎10−5 2.8⁎10−5 3⁎10−5 6cm2 yr−1 5.36 5.36 5.36 7cm yr−1 0.01 0.02 0.01 8mol m−3 1.96⁎10−3 2⁎10−3 2.3⁎10−3 1mol m−3 9⁎10−4 1⁎10−3 1⁎10−6 9

er) g cm−3 0.095 0.11 0.11 10

uming a uniformly mixed surface zone; (4) xNL2; (5) ∅ of top 5 cm; (6) Dsed HPO42−, Li

and Van Raaphorst 1993; (10) average, **theta=ρs[(1−ϕ)/ϕ]; (5)–(7) were assumed

Table 4Values for variable parameter.

Variable values

Parameter* Description Unit GeoB 3707 GeoB 3718 GeoB 3702

JMX=0 Flux of Fe-bound P to sediment μmol cm2 d−1 7.5⁎10−3 1⁎10−3 1.7⁎10−2

JGX=0 Flux of organic P to sediment μmol cm2 d−1 0.75 2.49 0.26JAX=0 Flux of authigenic P to sed. μmol cm2 d−1 0.15 0.68 0.22ks Rate constant for P sorption d−1 0.6 0.49 0.48kg Rate constant for organic P decomposition d−1 9.1⁎10−5 7.4⁎10−4 3.8⁎10−5

km Rate constant for Fe-bound P release d−1 5⁎10−3 2.6⁎10−3 2.2⁎10−3

ka Rate constant for apatite precipitation d−1 1.1⁎10−5 1.9⁎10−4 1⁎10−4

*Constant fluxes from the overlying water to the sediment were assumed at x=0; rate constants were assumed to be constant with depth in each relevant zone.


particulate Fe (Fetotal) occurs at the sediment surface (180.4–295.1 μmol/g). in 2–4 cm sediment depth Fetotal concentrationdecreases, followed by a slight increase with depth.

Mean concentrations and percentages of sedimentary phosphorusspecies are given in Table 5. In the core from the northern site GeoB

Fig. 3. Depth profiles of dissolved phosphate, iron, nitrate and oxygen in pore water (μmo(oxyhydr)oxides. In the right column: solid lines mark the boundary between sediment an

3707 concentrations range from 0.81 μmol/g for detrital P (Pdet) to9.51 μmol/g for organic P (Porg). In GeoB 3718 P bearing componentsrange from 0.76 μmol/g for authigenic P (Paut) to 14.29 μmol/g forbiogenic P (Pbio), and at the southern site GeoB 3702 mean con-centrations range from 0.91 μmol/g for Pdet to 8.72 μmol/g for Pbio.

l/l). In the left column: the dashed bars mark the area of reductive dissolution of irond water. Note the different depth scale for oxygen.

Fig. 4. Depth profiles of total particulate P and Fe (μmol/g dry sediment); dashed line bars mark the area of reductive dissolution of iron (oxyhydr)oxides.


Mean total extracted sedimentary P concentrations of the cores are inthe range of 22.9–28.5 μmol/g (∼0.23–0.29 wt.%). Comparison of Ptotalcontents obtained by total digestion to those in the sequentialextraction solutions (bulk P, Fig. 5) shows that from all samples fromGeoB 3707 and 3718 nearly all sedimentary P (on average 96%) wasrecovered. Extraction of sedimentary P phases from GeoB 3702 doesnot result in complete dissolution (Fig. 5). Possibly, this discrepancyresults from the presence of a refractory phase which does notdissolve after ashing at 550 °C and treatment with strong acid(Ruttenberg, 1992; Lopez, 2004). Nevertheless, the relations betweenthe individual P fractions are very similar to both other cores andtherefore are likely reliable. Despite the fact that the composition ofsedimentary P varies from site to site, most of the extractable P inthese sediments is present in Pbio (on average 32–50%; Table 5). Thisfraction is relatively constant with depth, except for site GeoB 3718.Another quantitative important carrier of P is Porg (24–41%).Concentrations are high in the uppermost sediment layer of thecores and decrease slightly with depth. Phosphate bound to iron(oxyhydr)oxides (PFe) makes up on average 16–22% of sedimentary P.The resulting depth profiles show clear differences between the upperand the deeper sediment column supporting that geochemicaltransformation processes proceed (Fig. 5). But overall, mean PFeconcentration is quite similar at all sites, compared to the largevariation in total Fe content (Fig. 4). Exchangeable P (Pex, not shownhere) and nonreactive detrital P (Pdet) are very small and ratherconstant P fractions. The amount of P associatedwith authigenic apatiteis also a less relevant burial form in these sediments, accounting atmost6% of total extractable P.

4.3. Carbon to phosphorus ratios in surface sediments

A quantitatively important proportion of P is delivered to thesediment–water interface as Porg in organic carbon particles (Corg).Therefore, the geochemical behaviour of P in sediments is closelylinked to that of Corg (e.g. Ingall and Van Cappellen, 1990; Ingall andJahnke, 1997; Anderson et al., 2001; Filippelli, 2001). In the studiedCorg-rich sediments (3.6–6.6 wt.%, Table 1), sedimentary molarorganic C/P ratios are much higher (312–1115) than known from

Table 5Mean P concentration (μmolP g−1) and percentage (%).

GeoB site # samples Biogenic P % Oxide P %

3707 12 7.56 32 5.30 223718 11 14.29 50 4.53 163702 12 8.72 38 5.08 22

fresh planktonic organic matter (106–140; Redfield et al., 1963;Takahashi et al., 1985). These date imply that Porg has preferentiallybeen removed from the organic matter. Without relevant data from thewater column,we cannot distinguishwhere themain depletion processtakes place, alreadyduring particle transport or after their accumulationon the sea floor. C/N ratios between 14 and 18 for the suspension loadjust above the sea floor (M. Inthorn, unpubl. data) may indicate thatorganic matter is significantly depleted in organic P, before it is finallydeposited. However, calculated molar ratios are in the range of thosereported elsewhere for this area. For instance, Hartmann et al. (1976)calculated organic C/P ratios which are N480 in rapidly depositingsediments (with a sedimentation rate of 0.01 cm yr−1 in sedimentsfrom600 to 3700 mwater depth) at theNWAfrican continentalmargin.Wefer et al. (1998) observed organic C/P ratios ∼1000 in sediments offSW Africa. However, in contrast to sites GeoB 3707 and GeoB 3702,molar organic C/P ratios are much higher at site GeoB 3718, with amaximum of 1115 in 12 cm sediment depth (Fig. 6). Bioavailablereactive forms of sedimentary P (Preactive) are potentially regeneratedfrom the sediment to fuel primary production (Ruttenberg, 1992; Faulet al., 2005). These forms have been defined as the sum of Pbio, Piron, Pautand Porg. Expressed as Corg/Preactive ratios in sediments this ratio mayreflect the diagenetic transformation of reactive P during burial. MolarCorg/Preactive ratios range from 107 to 253 and increase slightly withincreasing sediment depth (Fig. 6).

5. Discussion

5.1. Controls on sedimentary P cycling

Benthic P cycling is controlled by the input of P into the sediment.Our results show that most sedimentary P deposited at the sediment–water interface along the Namibianmargin is associatedwith biogenicP. In the close vicinity to one of the highest productive areas in theglobal ocean, this is not very surprising. So, a substantial proportion oftotal biogenic P is associated with biogenic debris (i.e. 48–63% of theextracted biogenic P fraction that is accumulated on the sedimentsurface is associated with biogenic debris, Fig. 2). Since—for instance—hard parts of fish debris mainly consist of hydroxyapatite crystals (e.g.

Authigenic P % Detrital P % Organic P %

0.52 2 0.81 3 9.51 410.76 3 2.10 7 6.79 241.29 6 0.91 4 6.77 30

Fig. 5. Sedimentary P distribution (μmol/g dry sediment), as deduced from sequential extraction of biogenic, Fe(III)-bound, authigenic, detrital and organic-bound P compared tobulk P content.


Suess, 1981; Nriagu, 1983; Baturin and Dubinchuk, 2003; Nemliheret al., 2004), in principle their dissolution could be an importantsource for pore water phosphate (Suess, 1981; Froelich et al., 1988;Schenau et al., 2000). But our data reveal only a slight decrease of thebiogenic P fraction with sediment depth (Fig. 5), which indicatesrather good preservation of this component. It seems obvious that thehigh burial efficiency of biogenic P is favoured also by the relativelyhigh sedimentation rates in the study area (∼0.01 cm yr–1; Weferet al., 1998; Inthorn et al., 2006). However, we ascertain that biogenicP plays only a minor role within the benthic P-cycle at all threesampling sites at least on short terms. Nevertheless, as depicted inFig. 2, an even larger proportion of biogenic P may consist of Ca-rich Pphases (i.e. 37–52% of biogenic P), which is preserved afteraccumulation. These phases could act as a precursor for subsequentprecipitation of Ca–P minerals (Föllmi, 1996; Monbet et al., 2007).

The next, second most important P fraction is associated withorganicmatter (Porg). Most recently by using different biogeochemicalmethods like radiocarbon dating or determination of degradationindices, Inthorn et al. (2006), Mollenhauer et al. (2007) andAspetsberger et al. (2007) could characterized the organic matter insurface sediments along a transect across the depocenter, where siteGeoB 3718 is located. All three studies conclude that the bulk of

Fig. 6. Depth profiles of molar organic C/Porg (note the differe

organic debris within the depocenter has frequently been re-suspended and, as a result of bacterial leaching, is relatively exhaustedin components, which are available for microbes. Therefore, theassociated preferential loss of P may explain why the molar organic C/P ratio is much higher within the depocenter (GeoB 3718) than at thetwo other sites outside (Fig. 6). However, expressed by the nearlyconstant values for Porg, the organic fraction may not represent amajor source for the observed amount of dissolved phosphate.Additionally, the much lower Corg/Preactive than Corg/Porg ratiosindicate that at least some P is subsequently transformed.

The third P fraction which has to be considered to unravel the localbenthic P-cycle is Fe(III)-bound. In contrast to both fractionsdiscussed before, the total amount of iron (oxyhydr)oxides withinthe oxic zone cannot be used to estimate the preliminary rain rate ofthis mineral fraction. However, pore water profiles and the Ptotal andFetotal concentration profiles give clear evidence of an additionaldiagenetic source. Due to the well known affinity of phosphate toadsorb on iron (oxyhydr)oxides, pore water phosphate concentra-tions have been described as closely coupled with the redox-sensitivebehaviour of iron, at least under oxic to suboxic conditions (e.g. Slompet al., 1996; Anschutz et al., 1998; Poulton and Canfield, 2006). Ourresults clearly demonstrate that this is also the case in margin

nt scale for GeoB 3718) and molar Corganic/Preactive ratios.


sediments off Namibia (Fig. 7). The general shapes of phosphate andferrous iron concentration profiles are very similar at all three sitesindicating a simultaneous release of both constituents at the samedepth intervals around 4 cmbsf. From here phosphate and ferrous irondiffuse in both directions, upward to the surface layer and downwardinto deeper sediments. Within the oxic zone ferrous iron is reoxidizedand form iron (oxyhydr)oxides which offer ideal new surfaces for the(re-)adsorbtion of pore water phosphate. This co-precipitation isexpressed by the concentrations of particulate P and Fe extractedwithCDB (PFe, FeCDB), which increase congruently (Fig. 7).

Previous pore water results in this region (e.g. Zabel et al., 1998)suggest a sink for phosphorus just below the zone of iron reduction(Fig. 3). Base on the solid phase extractions in our study, aquantitatively important formation of authigenic P minerals can beprecluded (Fig. 5). Our model results document that there is a clearexcess of phosphate release against the rate of authigenic P formation(Fig. 8, Table 6; discussed below). With the exception of siteGeoB 3718, also residual iron oxides seem not to be involved. How-ever, the nearly constant Fetotal contents obtained by total digestion(Fig. 4) give evidence that Fe is converted into amore stable phase likepyrite that is not dissolvable with the extraction procedure used(Borchers et al., 2005; Monbet et al., 2007). But, this gives noexplanation for the PFe concentrations when FeCDB is nearly notdetectable at sites GeoB 3707 and 3702. Presumably, this findingmight be partly attributed to the dissolution of residual Pbio duringCDB application (Jensen et al., 1998; Monbet et al., 2007). After tensuccessive extractions with NH4Cl the remaining content of P is onaverage 0.1–0.52 μmol/g (Fig. 2). Otherwise, part of the PFe fractioncould also be a post-sedimentarily precipitated, not well-definedphase which was additionally dissolved during CDB application (e.g.vivianite (Fe3(PO4)28H2O); Rosenqvist, 1970; März et al., 2008).

Despite the general confirmation of the close linkage between thebenthic cycles of P and Fe, uncertainties mentioned above alreadyshow that this connection is much more complex and cannot beexplained by simple dissolution and adsorption processes alone.Especially balances between transfer rates, as indicated by diffusivefluxes, and the available amount of solid phases, as indicated byresults from the extraction scheme, reveal great discrepancies. So,assuming steady state conditions in a classical approach (1—notemporal change and 2—transport occurs only by accumulation,conservative mixing, and molecular diffusion), the release ofphosphate during iron reduction exceeds by far the concentration

Fig. 7. Depth profiles of CDB extractable P and Fe content (Step (2); μmol/g dry s

which would correlate with the available amount of phosphateadsorbed on iron (oxyhydr)oxides. So, based on the diffusive fluxes ascalculated by applying Fick's First Law, the release of phosphateduring Fe reduction is about 8, 13 and 12 mmol P m−2 yr−1 at thethree sites GeoB 3702, 3707 and 3718. Assuming a realistic wet bulkdensity of 0.5 g cm−3 and a mean sedimentation rate of 0.01 cm yr−1,the value of 8 mmol P m−2 yr−1 for the example corresponds to atotal amount of about 0.26 g P yr−1 which releases from a sedimentslice of 1 cm thickness and 1 m² expansion. But examining our data,the total PFe reservoir of this sediment volume for 100 yrs is only0.75 g P. Therefore, in this case the pore water profile indicates arelease of P which is 34-fold higher than the existing content of Passociatedwith (oxyhydr)oxides and even 5 timesmore than the totalP content bound onto mineral phases. For the sediments at both othersites this apparent deficit is even larger. In the same way it can beshown that there is also no counterpart in the solid phase, whichwould be equivalent to the P sink in the oxic surface layers.

Although our ferrous iron data have to be handled cautiouslybecause of a potential underestimation of concentrations on account ofre-oxidation during the sampling procedure (De Lange et al., 1992),they give additional support to this imbalance. The sediment datareveal that the molar FeCDB/PFe ratios within the iron reduction zoneare ∼4 (GeoB 3702), ∼2 (GeoB 3707), and ∼8 (GeoB 3718). Althoughthese values seem to be low in comparison to theoretical calculationswhich are based on determinations of the specific surface area of Fe(oxyhydr)oxides (∼60; Crosby et al., 1983; Haese, 2006), they are in agood correspondence with ratios calculated for real sediments (2–10;Slomp et al., 1996; Matthiesen et al., 2001; Gunnars et al., 2002;Virtasalo et al., 2005). However, there is a clear tendency that FeCDB/PFeratios decrease with increasing sediment depth. Assuming the quiterealistic preferential dissolution of fine grained, newly formed, lessconsolidated iron (oxyhydr)oxides during burial, just the opposite is tobe expected. A reduction of the specific surface area for P adsorptionaccompanied to such a coarsening should result in higher FeCDB/PFeratios. From the background of the high phosphate release associatedwith iron reduction, the low gradients in phosphate concentrationsacross the benthic sediment water interface also indicate the highbuffer capacity for phosphate to iron (oxyhydr)oxides (Anschutz et al.,1998).

However, the discrepancies stated for the balances between thesediment composition and specific microbial and/or geochemicaltransfer rates seem to reflect non-steady state conditions. In contrast,

ediment) and pore water phosphate and ferrous iron concentration (μmol/l).

Fig. 8. Calculated depth profiles (solid lines) to the depth profiles of pore water phosphate (μmol/l), solid phase Fe(III)-bound P, organic-bound P and authigenic P (i.e. sum ofbiogenic P and authigenic P; cf. Section 3.4)(μmol/g dry sediment). The dashed horizontal lines in each graph mark the boundaries of each sediment zone used for the modelcalculation (cf. Section 3.4). The vertical dashed line in the authigenic P depth profile represents authigenic P that has not been formed in-situ (background P; note the different scalefor GeoB 3718).


the local, nearly constant environmental conditions and the similar-ities between the main features at the three locations rather support asteady state system. Assuming that the much faster reacting porewater system represents a steady state situation, this phenomenon ofan apparent deficit of iron (oxyhydr)oxides within several centi-metres sediment depth could easily be explained by an age-dependent, non-steady state sediment mixing. This process could bea macrobenthic sediment transport as a frequently disregarded formof bioturbation than the conventionally assumed homogenous

sediment mixing. This, at least occasionally occurring additionaldownward transport of iron (oxyhydr)oxide particles will refill thesedimentary reservoir with material from the oxic zone over and overagain. There is no doubt that such transport processes can have astrong effect on the sediment composition (e.g. Pope et al., 1996;Fernandes et al., 2006). It could clearly be shown that the combinationof rapid biodiffusive mixing and downward transport allows a muchmore rapid mixing of surface material into the seabed than estimatedby a conventional combination of sedimentation rate and biodiffusion.

Table 6Phosphate production and removal rates as calculated with the diagenetic P model(rates are in μmolm2 d−1).

GeoB 9510 GeoB 9519 GeoB 9518

Phosphate productionOrganic P release 0.71 2.39 0.21Fe-bound P desorption 20.16 12.52 12.46Diffusive P flux at SWIa – – 0.23Total 20.87 14.91 12.90

Phosphate removalDiffusive P flux at SWI 0.40 0.44 –

Diffusive P flux at bottom 0.22 1.52 0.25Authigenic P formation 0.02 0.25 0.10P adsorption to iron phases 20.23 12.70 12.55Total 20.87 14.91 12.90

a SWI: sediment water interface.


In addition, a recent study by Küster-Heins et al. (2010) give evidencethat sedimentary systems off Senegal are balanced by an additionaltransport only.

5.2. Quantification of P transformation

To test our previous interpretations of the local benthic P-cycle, adiagenetic transport and reaction model was used (cf. Section 3.4).Despite some simplifications, all important net reactions which mayaffect the distribution of porewater and solid phase P inmarine surfacesediments are considered. This allows for the application of the modelto various data sets by changing the boundary conditions, input andtransfer rates (Table 3 and 4). Given that the model assumes steadystate, the burial rate of P is always equal to the total accumulation rateof P on the sediment surface and total phosphate release is balancedwith the sum of corresponding processes of its removal.

Best fit calculations of sedimentary P for the GeoB sites agreereasonably well with the extracted mineralogical fractions (Fig. 8),especially for sites GeoB 3707 and GeoB 3702. Table 6 shows thecorresponding calculated contributions of the various processes andreservoirs that play a role in P cycling at the three different sites. Thecalculations give strong support to our assumption that the cycling of Pin these sediments is dominated by release from and binding to iron(oxyhydr)oxides. They contribute 85–97% to the internal total Ptransfers. Only 0.03–3.3% of the primary P input is associatedwith ironminerals, however. Accordingly, model results confirm the assump-tion that themajor flux of reactive P to the sediment occurs in the formof organic-bound and ‘authigenic’ P that has been formed elsewhere,here assumed to be biogenic based on our extraction results.

However, a prominent discrepancy occurs between the calculatedand the measured pore water phosphate concentrations below theoxic zone (Fig. 8: L1). In correspondence with the interpretation in theprevious chapter, this approves that we have ignored an additionalprocess that is responsible for the increased release of phosphateduring microbial iron reduction and removal at depth. The processwould not only explain the apparent discrepancy between the releaseand transfer rates of (ferrous) iron with the available amount of iron(oxyhydr)oxides. Since these iron phases act as carrier for thetransport of phosphate as well, excess phosphate concentrationswould be the consequence. Furthermore, depending on the number ofsuch internal loops the FeCDB/PFe ratio would decrease due to thesurplus of phosphate which continuously enters the cycle by themicrobial degradation of organic matter via oxygen respiration, deni-trification and iron reduction.

6. Conclusions

A combination of sequential extractions and pore water analyseswere used to identify specific P-containing compounds responsible for

the distribution of phosphorus in sedimentary settings off the highproductive Namibian area. Results show that Ptotal is relatively homo-geneously distributed in these continental margin sediments, andsolid phase P speciation varies only slightly from site to site.Nevertheless, Pbio and Porg pools were the dominant reservoirs for Pburied in sediments and they do not act as important sources for porewater phosphate during early diagenesis. Depth profiles of dissolvedpore water phosphate and iron, and reactive P species suggest a rapidreorganization of P, where P released by mineralization is incorpo-rated into a PFe phase rather than into a Paut phase. But, a simpleinterplay of dissolution and adsorption processes alone cannotexplain the observed high transformation rates and pore waterconcentrations of P in these sediments. The release of phosphate inthe course of iron reduction exceeds by far the identified content ofiron (oxyhydr)oxides. We assume that this apparent imbalance couldbe caused by the disregard of macrobenthic particle mixing. Un-fortunately, such directed, episodic particle transport processes cannotbe detected by the methods which were available. The much fasterreplacement of iron (oxyhydr)oxides in the iron reduction zone thandue to sediment accumulation and conservativemixing (bioturbation)alone may be a key prerequisite for the observed elevated dissolvedphosphate concentrations.

Acknowledgements

The DFG International Graduate College Proxies in Earth History(EUROPROX) funded this study. The authors would like to thankSilvana Pape, Karsten Enneking, Susanne Siemers, Niklas Allroggen,Lars Hoffmann, Rinke Knoop and Javier Díaz-Ochoa for laboratoryassistance and analytical support. Special thanks to Christian Märzand Peter Kraal for helpful discussions.We thank Caroline P. Slomp forimproving an earlier version of this manuscript.

References

Anderson, L.D., Delaney, M.L., Faul, K.L., 2001. Carbon to phosphorus ratios insediments: implications for nutrient cycling. Global Biogeochemical Cycles 15,65–79.

Anschutz, P., Zhong, S., Sundby, B., Mucci, A., Gobeil, C., 1998. Burial efficiency ofphosphorus and the geochemistry of iron in continental margin sediments.Limnology and Oceanography 43, 53–64.

Arning, E.T., Birgel, D., Schulz-Vogt, H.N., Holmkvist, L., Jørgensen, B.B., Larson, A.,Peckmann, J., 2008. Lipid biomarker patterns and phosphogenic sediments fromupwelling regions. Geomicrobiology Journal 25, 69–82.

Aspetsberger, F., Zabel, M., Ferdelman, T., Struck, U., Mackensen, A., Ahke, A., Witte, U.,2007. Instantaneous benthic response to different organic matter quality: in situexperiments in the Benguela Upwelling System. Marine Biology Research 3,342–356.

Baturin, G.N., 2003. Phosphorus cycle in the ocean. Lithology and Mineral Resources 38,101–119.

Baturin, G.N., Dubinchuk, V.G., 2003. The composition of phosphatized bones in recentsediments. Lithology and Mineral Resources 38, 313–323.

Benitez-Nelson, C.R., 2000. The biogeochemical cycling of phosphorus in marinesystems. Earth-Sciences Reviews 51, 109–135.

Bianchi, G., Carpenter, K.E., Roux, J.-P., Molloy, F.J., Boyer, D., Boyer, H.J., 1999. FAOSpecies Indentification Field Guide For Fishery Purposes. The Living MarineResources of Namibia. FAO Food and Agricultural Organization of the UnitedNations, Rome. 5 pp.

Borchers, S.L., Schnetger, B., Böning, P., Brumsack, H.-J., 2005. Geochemical signatures ofthe Namibian diatom belt: perennial upwelling and intermittent anoxia. Geo-chemistry, Geophysics, Geosystems 6, Q06006. doi:10.1029/2004GC000886.

Bremner, J.M., Rogers, J., 1990. Phosphorite Deposits on the Namibian Continental Shelf.In: Burnett, W.C., Riggs, S.R. (Eds.), Phosphate deposits of the world. University ofCambridge, Cambridge, pp. 143–152.

Bremner, J.M., Willis, J.P., 1993. Mineralogy and geochemistry of the clay fraction ofsediments from the Namibian continental margin and the adjancent hinterland.Marine Geology 115, 85–116.

Burnett, W.C., 1977. Geochemistry and origin of phosphorite deposits from off Peru andChile. GSA Bulletin 88, 813–823.

Calvert, S.E., Price, N.B., 1983. Geochemistry of Namibian shelf sediments. In: Suess, E.,Thiede, J. (Eds.), Coastal Upwelling—Its Sediment Record, Part A. NATO conferenceseries. IV. Marine sciences, New York, pp. 337–375.

Canfield, D.E., Kristensen, E., Thamdrup, B., 2005. The phosphorus cycle. In: Southward,A.J., Tyler, P.A., Young, C.M., Fuiman, L.A. (Eds.), Advances in Marine Biology.Elsevier Academic Press, London, pp. 419–440.


Cha, H.J., Lee, C.B., Kim, B.S., Choi, M.S., Ruttenberg, K.C., 2005. Early diageneticredistribution and burial of phosphorus in the sediments of the southwestern EastSea (Japan Sea). Marine Geology 216, 127–143.

Compton, J., Mallinson, D., Glenn, C.R., Filippelli, G., Foellmi, K., Shields, G., Zanin, Y.,2000. Variations in the global phosphorus cycle. In: Glenn, C.R., Prevot-Lucas, L.,Lucas, J. (Eds.), Marine Authigenesis: From Global to Microbial. SEPM (Society forSedimentary Geology), Tulsa, Oklahoma, pp. 21–33.

Crosby, S.A., Glasson, D.R., Cuttler, A.H., Butler, I., Turner, D.R., Whitfield, M., Millward,G.E., 1983. Surface areas and porosities of Fe(III)–Fe(II)-derived oxyhydroxides.Environmental Science Technology 17, 709–713.

De Lange, G.J., Cranston, R.E., Hydes, D.H., Boust, D., 1992. Extraction of pore water frommarine sediments: a review of possible artifacts with pertinent examples from theNorth Atlantic. Marine Geology 109, 53–76.

Diaz, J., Ingall, E., Benitez-Nelson, C., Paterson, D., De Jonge, M.D., McNulty, I., Brandes, J.A.,2008. Marine polyphosphate: a key player in geological phosphorus sequestration.Science 320, 652–655.

Faul, K.L., Paytan, A., Delaney, M.L., 2005. Phosphorus distribution in sinking oceanicparticulate matter. Marine Chemistry 97, 307–333.

Fernandes, S., Meysman, F.J.R., Sobral, P., 2006. The influence of Cu contamination onNeris diviersicolor bioturbation. Marine Chemistry 102, 148–158.

Filippelli, G.M., 1997. Controls on phosphorus concentration and accumulation inoceanic sediments. Marine Geology 139, 231–240.

Filippelli, G.M., 2001. Carbon and phosphorus cycling in anoxic sediments of theSaanich Inlet, British Columbia. Marine Geology 174, 307–321.

Filippelli, G.M., Delaney, M.L., 1996. Phosphorus geochemistry of equatorial Pacificsediments. Geochimica et Cosmochimica Acta 60, 1479–1495.

Föllmi, K.B., 1996. The phosphorus cycle, phosphogenesis and marine phosphate-richdeposits. Earth-Science Reviews 40, 55–124.

Froelich, P.N., 1988. Kinetic control of dissolved phosphate in natural rivers andestuaries: a primer on the phosphate buffer mechanism. Limnology and Oceanogra-phy 33, 649–668.

Froelich, P.N., Bender, M.L., Luedtke, N.A., 1982. Themarine phosphorus cycle. AmericanJournal of Science 282, 474–511.

Froelich, P.N., Arthur, M.A., Burnett, W.C., Deakin, M., Hensley, V., Jahnke, R., Kaul, L.,Kim, K.-H., Roe, K., Soutar, A., Vathakanon, C., 1988. Early diagenesis of organicmatter in Peru continental margin sediments: phosphorite precipitation. MarineGeology 80, 309–343.

Glenn, C.R., Arthur, M.A., 1988. Petrology and major element geochemistry of Perumargin phosphorites and associated diagenetic minerals: authigenesis in modernorganic-rich sediments. Marine Geology 80, 231–267.

Grasshoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis. VerlagChemie, Weinheim, Germany. 317 pp.

Gunnars, A., Blomqvist, S., Johansson, P., Andersson, C., 2002. Formation of Fe(III)oxyhydroxides colloids in freshwater and brackish seawater, with incorpo-ration of phosphate and calcium. Geochimica et Cosmochimica Acta 66,745–758.

Haese, R.R., 2006. The biogeochemistry of iron, In: Schulz, H.D., Zabel, M.(Eds.), Marine Geochemistry, 2nd edition. Springer-Verlag, Berlin, Heidelberg,pp. 241–270.

Hartmann, M.P., Müller, P.J., Suess, E., Van der Weijden, C.H., 1976. Chemistry of LateQuaternary sediments and their interstitial waters from the N.W. Africancontinental margin. ‘Meteor’ Forschungsergebnisse Reihe C 24. UniversitätHamburg, Hamburg. 67 pp.

Heggie, D.T., Skyring, G.W., O´Brien, G.W., Reimers, C., Herczeg, A., Moriarty, D.J.W.,Burnett, W.C., Milnes, A.R., 1990. Organic carbon cycling and modern phosphoriteformation on the East Australian continental margin. In: Notholt, A.J.G., Jarvis, I.(Eds.), Phosphorite Research and Development. Geological Society SpecialPublication, London, pp. 87–117.

Ingall, E.D., Jahnke, R., 1997. Influence of water-column anoxia on the elementalfractionation of carbon and phosphorus during sediment diagenesis. MarineGeology 139, 219–229.

Ingall, E.D., Van Cappellen, P., 1990. Relation between sedimentation rate and burial oforganic phosphorus and organic carbon in marine sediments. Geochimica etCosmochimica Acta 54, 373–386.

Ingall, E.D., Bustin, R.M., Van Cappellen, P., 1993. Influence of water column anoxia onthe burial and preservation of carbon and phosphorus inmarine shales. Geochimicaet Cosmochimica Acta 57, 303–316.

Inthorn, M., Wagner, T., Scheeder, G., Zabel, M., 2006. Lateral transport controlsdistribution, quality, and burial of organic matter along continental slopes in high-productivity areas. Geology 34, 205–208.

Jensen, H.S., Mortensen, P.B., Andersen, F.O., Rasmussen, E., Jensen, A., 1995.Phosphorus cycling in a coastal marine sediment, Aarhus Bay, Denmark. Limnologyand Oceanography 40, 908–917.

Jensen, H.S., McGlathery, K.J., Marino, R., Howarth, R.W., 1998. Forms and availability ofsediment phosphorus in carbonate and sand of Bermuda seagrass beds. Limnologyand Oceanography 43, 799–810.

Küster-Heins, K., De Lange, G.J., Zabel, M., 2010. Benthic phosphorus and ironbudgets for three NWAfrican slope sediments: a balance approach. Biogeosciences 7,1–12.

Li, Y.-H., Gregory, S., 1974. Diffusion of ions in sea water and in deep-sea sediments.Geochimica et Cosmochimica Acta 38, 703–714.

Lopez, P., 2004. Spatial distribution of sedimentary P pools in a Mediterraneancoastal lagoon ‘Albufera d'es Grau’ (Minorca Island, Spain). Marine Geology 203,161–176.

März, C., Hoffmann, J., Bleil, U., De Lange, G.J., Kasten, S., 2008. Diagenetic changesof magnetic and geochemical signals by anaerobic methane oxidation in sedi-

ments of the Zambesi deep-sea fan (SW Indian Ocean). Marine Geology 255,118–130.

Matthiesen, H., Leipe, T., Laima, M.J.C., 2001. A new experimental setup for studying theformation of phosphate binding iron oxides in marine sediments, preliminaryresults. Biogeochemistry 52, 79–92.

Middelburg, J.J., Soetaert, K., Herman, P.M.J., 1997. Empirical relationships for use inglobal diagenetic models. Deep-Sea Research I 44, 327–344.

Mollenhauer, G., Schneider, R.R., Müller, P.J., Spieß, V., Wefer, G., 2002. Glacial/Interglacial variability in the Benguela upwelling system: spatial distribution andbudgets of organic carbon accumulation. Global and Biogeochemical Cycles 16,1134. doi:10.1029/2001GB001488.

Mollenhauer, G., Inthorn, M., Vogt, T., Zabel, M., Sinninghe Damste, J.S., Eglinton, T.I.,2007. Aging of marine organic matter during cross-shelf lateral transport in theBenguela upwelling system revealed by compound-specific radiocarbon dating.Geochemistry, Geophysics, Geosystems 8, Q09004. doi:10.1029/2007GC001603.

Monbet, P., Brunskill, G.J., Zagorskis, I., Pfitzner, J., 2007. Phosphorus speciationin the sediment and mass balance for the central region of the Great BarrierReef continental shelf (Australia). Geochimica et Cosmochimica Acta 71,2762–2779.

Nemliher, J.G., Baturin, G.N., Kallaste, T.E., Murdmaa, I.O., 2004. Transformation ofhydroxyapatite of bone phosphate from the ocean bottom during fossilization.Lithology and Mineral Resources 39, 468–479.

Nriagu, J.O., 1983. Rapid decomposition of fish bones in Lake Erie sediments.Hydrobiology 106, 217–222.

Paytan, A., McLaughlin, K., 2007. The oceanic phosphorus cycle. Chemical Review 107,563–576.

Pope, R.H., DeMaster, D.J., Smith, C.R., Seltmann, Jr., 1996. Rapid bioturbation inequatorial Pacific sediments evidence from excess 234Th measurements. Deep-SeaResearch II 43, 1339–1364.

Poulton, S.W., Canfield, D.E., 2006. Co-diagenesis of iron and phosphorus inhydrothermal sediments from the southern East Pacific Rise: implications for theevaluation of paleoseawater phosphate concentrations. Geochimica et Cosmochi-mica Acta 70, 5883–5898.

Redfield, A.C., Ketchum, B.H., Richards, F.A., 1963. The influence of organisms in thecomposition of seawater. In: Hill, M.N. (Ed.), The Sea. Interscience, New York,pp. 26–77.

Rosenqvist, I.T., 1970. Formation of vivianite in Holocene clay sediments. Lithos 3,327–334.

Ruttenberg, K.C., 1992. Development of a sequential extraction method for differentforms of phosphorus in marine sediments. Limnology and Oceanography 37,1460–1482.

Ruttenberg, K.C., 1993. Reassessment of the oceanic residence time for phosphorus.Chemical Geology 107, 405–409.

Schenau, S.J., De Lange, G.J., 2000. A novel chemical method to quantify fish debris inmarine sediments. Limnology and Oceanography 45, 963–971.

Schenau, S.J., De Lange, G.J., 2001. Phosphorus regeneration vs. burial in sediments ofthe Arabian Sea. Marine Chemistry 75, 201–217.

Schenau, S.J., Slomp, C.P., De Lange, G.J., 2000. Phosphogenesis and active phosphoriteformation in sediments from the Arabian Sea oxygen minimum zone. MarineGeology 169, 1–20.

Schulz, H.D., 2006. Quantification of early diagenesis: dissolved constituents inpore water and signals in the solid phase, In: Schulz, H.D., Zabel, M.(Eds.), Marine Geochemistry, 2nd edition. Springer-Verlag, Berlin, Heidelberg,pp. 75–124.

Schulz, H.D., cruise participants, 1996. Report and preliminary results of METEOR-Cruise M 34/2, Walvis Bay—Walvis Bay, 29.01.–18.02.1996. Berichte aus demFachbreich Geowissenschaften der Universität Bremen 78. Universität Bremen,Bremen. 133 pp.

Schulz, H.D., Schulz, H.N., 2005. Large sulfur bacteria and the formation of phosphorite.Science 307, 416–418.

Shannon, L.V., Nelson, G., 1996. The Benguela: large scale features and processes andsystem variability. In: Wefer, G., Berger, W.H., Siedler, G., Webb, G.J. (Eds.), TheSouth Atlantic: present and past circulation, pp. 163–210.

Slomp, C.P., Van Raaphorst, W., 1993. Phosphate adsorption in oxidized marinesediments. Chemical Geology 107, 477–480.

Slomp, C.P., Epping, E.H.G., Helder, W., Van Raaphorst, W., 1996. A key role for iron-bound phosphorus in authigenc apatite formation in North Atlantic continentalplatform sediments. Journal of Marine Research 54, 1179–1205.

Strickland, J.D.H., Parsons, T.R., 1968. A Practical Handbook of Seawater Analysis.Fisheries Research Board of Cananda, Ottawa. 293 pp.

Suess, E., 1981. Phosphate regeneration from sediments of the Peru continentalmargin by dissolution of fish debris. Geochimica et Cosmochimica Acta 45,577–588.

Takahashi, T., Broecker, W.S., Langer, S., 1985. Redfield ratio based on chemical datafrom isopycnal surfaces. Journal of Geophysical Research 90, 6907–6924.

Van Cappellen, P., Ingall, E.D., 1994. Benthic phophorus regeneration, net primaryproduction, and ocean anoxia: a model of the coupled marine biogeochemicalcycles of carbon and phosphorus. Paleoceanography 9, 677–692.

Van Cappellen, P., Ingall, E.D., 1996. Redox stabilization of the atmosphere and oceansby phosphorus-limited marine productivity. Science 271, 493–496.

Van der Zee, C., Slomp, C.P., Van Raaphorst, W., 2002. Authigenic P formation andreactive P burial in sediments of the Nazare canyon on the Iberian margin (NEAtlantic). Marine Geology 185, 379–392.

Virtasalo, J.J., Kohonen, T., Vuorinen, I., Huttula, T., 2005. Sea bottom anoxia in theArchipelago Sea, northern Baltic Sea—implications for phosphorus remineraliza-tion at the sediment surface. Marine Geology 224, 103–122.


Wefer, G., Bleil, U., Schulz, H.D., Fischer, G., 1997. Geo Bremen South Atlantic 1996,Cruise No. 34, 21. Februar—15. April 1996, Meteor-Berichte 97-2. University ofHamburg, Hamburg. 268 pp.

Wefer, G., Berger, W.H., Richter, C., Shipboard Scientific Party, 1998. Facies patterns andauthigenic minerals of upwelling deposits off southwest Africa. Proceedings ofthe Ocean Drilling Program, Initial Reports, Vol. 175, A&M University Texas,pp. 487–504.

Zabel, M., Dahmke, A., Schulz, H.D., 1998. Regional distribution of diffusive phosphateand silicate fluxes through the sediment–water interface: the eastern SouthAtlantic. Deep-Sea Research I 45, 277–300.

Zabel, M., Bickert, T., Dittert, L., Haese, R.R., 1999. The significance of the sedimentary Al/Ti ratios as indicator for reconstructions of terrestrial input to the equatorialAtlantic. Paleoceanography 14, 789–799.

phosphorus cycling in marine sediments from the continental margin off namibia

Documents