During organic matter decomposition its molecular composition is modified and its first-order degradation rate constant decreases. Although, consistent compositional changes due to selective preservation of more refractory components have been observed, they have not yet been quantified, nor have they been linked to degradation kinetics. In this presentation we extend the protein amino acids based degradation index of Dauwe & Middelburg (1998) over a broader range of environmental conditions (including more deep sea deposits) and relate it to independent measures of degradation state such as enzymatic degradability (Dauwe et al., 1999) and first order rate constants. The results demonstrate that this degradation index not only correlates well with established compositional degradation state indicators, such as relative concentration of non-protein amino acids and hexosamines, but also with kinetically defined parameters such as the fraction of the total protein pool that is available for enzymatic decay and the first-order degradation rate constant of POM mineralization. The mole% concentration of non-protein amino acids (ß-alanine and <gamma>-aminobutyric acid) correlates negatively with the degradation index in the lower quality range, indicated by indexes between about -2 to 0. The mole% concentration hexosamines (galactosamine and glucosamine) correlates negatively over a broader range of indexes from about -2 to 1. The fraction of bio-available amino acids was positively correlated with the degradation index in the more labile material with indexes ranging between about 0 to 1.5. The logarithm of the first order degradation rate constant, giving the most direct information of the organic matter degradation state, was positively correlated over the whole range of investigated materials (total range of indexes between about -2 and 1.5). This novel direct link between organic matter degradation state, its molecular composition and its reactivity, provides a solid basis to predict the fate of particular organic matter in the ocean and its sediments and allows us to estimate the quality of organic matter based solely on its chemical composition.
Dauwe B & Middelburg JJ, Limnol. & Oceanography, 43, 782-798, (1998).
Dauwe, B, Middelburg JJ, Van Rijswijk, P, Sinke, J, Herman PMJ & Heip CHR, Jour. Marine Research, 57, 109-134, (1999).
The interdependencies between the bioturbated sediment and the watercolumn are investigated with a global geochemical circulation model (HAMOCC, Hamburg Ocean Carbon Cycle Model) on the basis of a velocity field provided by a dynamical Ocean General Circulation Model (LSG-OGCM). The model has a horizontal resolution of 3.5x3.5°, a 1-layer atmosphere reservoir, 11 or 22 layers in the water column, and 10 layers within the topsediment. The marine carbon and silicon cycles are simulated by use ofbasic parameterisations for the gas exchange between ocean and atmosphere, the particle export production (of particulate organic carbon, calcium carbonate, and opal), the vertical particle flux, remineralisation and redissolution in the water column and the sedimentpore waters, deposition and accumulation of sediment, pore water diffusion, and bioturbation. The model predicts the tracer distributions of DIC, TAlk, pkosphate, oxygen, and silic acid in the water column as well as the sediment porewaters, plus the weight percentages of the solid sediment tracers calcium carbonate, organic carbon, opal, and an inert component (clay). The model can be integrated over twenty thousand years within 10 CPU hours on a cray C90 computer. It is thus possible to use this model in multiple paleoclimate sensitivity experiments.
Sensitivity experiments were carried out with respect to thethe velocity field, the nutrient uptake kinetics, the particle flux profiles, the remineralisation and redissolution kinetics, the terrigenous input of matter (chemical reactive, inert), the bioturbation rate, and the rain ratios of organic carbon, calcium carbonate, and biogenic opal. These experiments indicate that resulting variations of the sediment water column interaction canpotentially induce considerable changes in the atmospheric carbon dioxide partial pressure.
The model results are compared with observations for the modern ocean. The agreement between simulated and observed tracer distributions in the water column as well as the bioturbated sedimentis encouraging. An outline of a method for a further systematic quantitative reconciliation of observed and simulated data is discussed. (This work was in part funded through EU MAST III project "SINOPS''MAS3-CT97-0141.)
We present simulation results of redox processes in marine surface sediments from different marine environments in the South Atlantic performed with the computer model CoTReM (Hamer & Sieger, 1994; Landenberger, 1998; Adler et al., sub.). CoTReM was developed for the simulation of one-dimensional transport of dissolved and solid phases and their interactions driven by bio-geochemical reactions and thermodynamic equilibria occurring in natural environments.
Our results comprise simulations of measured pore water and solid phase profiles from continental slope sediments in the South Atlantic. One major goal was the investigation of redox processes which are related to deep sulphate reduction occurring within a narrow sulphate-methane transition zone. This process is well described (cf. Niewöhner et al., 1998) and known as a typical phenomenon within the upper tens of meters of sediments in highly productive areas. In two case studies from cores taken off Angola and from the Amazon Fan we investigated the effects of this process on the diagenesis of iron minerals.
Stations on the Amazon Fan are characterized by high availability of reactive iron hydroxides in the solid phase. The sulfide produced by sulphate reduction is almost completely precipitated in form of iron-sulfides. Within the transition zone, however, small peaks of pore water hydrogen sulfide are measurable. To simulate the dynamics of the pore water system in sense of maintaining thermodynamic equilibria the reactivity of iron hydroxides was varied over depth by choosing different saturation indices for a standard Fe(OH)3 phase. Furthermore, iron sulfide phases were set to be in equilibrium resulting in a precipitation of mackinawite within the transition zone and different iron sulfides below.
In contrast, at a station from the upwelling region off southwest Africa high amounts of hydrogen sulfide are present in pore water below a depth of two meters implying that the precipitation of iron sulfides is limited by the availability of reactive iron hydroxides. A depletion of hydrogen sulfide at approximately two meters below sediment surface is preliminarily handled as a reoxidation by iron hydroxides. During simulation the equilibrium with calcite was maintained within the system revealing good approximations to measured Ca2+ and alkalinity profiles.
Apart from processes related to deep sulphate reduction, we plan to show simulation results of organic matter degradation and coupled calcite dissolution at the sediment-water interface. The simulations are carried out on the basis of in situ microelectrode data of O2, pCO2, pH, and Ca2+ from the west African continental margin and, therefore, suggest interesting new information for the understanding of preservation and dissolution of CaCO3 at the sea floor.
Adler M, Hensen C, Kasten S & Schulz HD, Geol Rdsch, (submitted).
Hamer K & Sieger R, Ernst & Sohn Verlag Berlin, (1994).
Landenberger H, Berichte, Fachbereich Geowissenschaften, Universitaet Bremen, (1998).
Niewöhner C, Hensen C, Kasten S, Zabel M & Schulz HD, Geochim Cosmochim Acta, 62(3), 455-464, (1998).
Sedimentary burial of organic matter in marginal areas of the ocean constitutes a major mechanism for removing nutrient elements from the oceans. Organics fluxes and their sedimentary records in the surface deposits (0-1 cm) have been studied on the NW Mediterranean margin between 600 and 1200 m depth. From east to west, investigations were focused in the Planier, Grand Rhône, Sète, and Lacaze-Duthiers canyon areas. The continental slope is described as a mosaic of different physiographic sites: head and axial channel (axis and sides) of the canyons, open slopes between two canyons (Buscail and Germain, 1997). At these sites, surface sediments were sampled seasonally and fluxes were measured by sequential sediment traps deployed during one year (Monaco et al., 1990, 1998; Heussner et al, 1996). We took organic carbon and biomarkers such as amino acids and sugars into account in determining the spatial and temporal variability of organic carbon burial and mineralization. A direct relationship was detected between near-bottom organic fluxes and organic contents in surface sediments. Qualitative seasonal differences were also evident in the biochemical characteristics of settling particles and surface deposits.The accumulation and mineralization rates of org C were estimated for the upper 1 cm. The sedimentation rates were evaluated from 210Pb dating of the deposits (Buscail et al, 1997; Radakovitch, 1995; Abassi 1998). Total mass flux is compared to the accumulation rate: only a proportion of the total mass flux measured near the bottom can be considered to contribute to the deposit. The complementary fraction is exported downward to the slope (advective transfer). These data are used for establishing the budget of org C. The budgets show that mineralization processes are dominant in the eastern part of the margin (Planier canyon axis and open slope and Grand Rhône canyon axis (73% of the org C input). In the western part, this proportion decrease to 50%. A direct relationship is detected between the increase of org C contents in sediment trap particles and mineralization rates. Accumulation rates of org C are higher in the western part. The general circulation is oriented from east to west in the Gulf of Lions (Liguro-Provençal current) and consists of a current which transfers and acts as a trap for particulate material along the continental slope (Durrieu de Madron et al. 1990). Budgets of organic carbon differ with respect to the dynamics of organic matter inputs. The NW Mediterranean margin appears as a mosaic of physiographic sites inducing a relative difference between the canyon axis and adjacent open slope. To explain these differences the quantity of carbon are not enough, but it is then necessary to consider the biochemical characteristics of the settling organic matter.
Abassi A., Thèse de Doctorat de l'Université de Perpignan, 150 p, (1998).
Buscail R, Ambatsian P, Monaco A, Bernat M, Marine Geology, 137, 271-286, (1997).
Buscail R and Germain C, Limnology & Oceanography, 42 (2), 217-229, (1997).
Durrieu de Madron X, Nyffeler F, Godet CH, Cont Shelf Res, 10, 915-929, (1990).
Heussner S, Calafat A, Palanques A, EUROMARGE-NB Final Report, II, 68-95, (1996).
Monaco A, Courp T, Heussner S, Carbonne J, Fowler SW, Deniaux B, Cont Shelf Res, 10, 959-987, (1990).
In undisturbed marine sediments, the bacterially mediated oxidation of organic matter leads to a well-established depth sequence of redox reactions. In sediments where the relative positions of reducing and oxidizing components is shifted because of bioturbation, the concept of a vertical chemical zonation does not necessarily apply. This may temporary favour new reactions and change the depth sequence of redox reactions. Here we report new measurements of the vertical distributions of dissolved phosphate, manganese, nitrate, and ammonia, using pore water samples obtained by centrifugation, and reactive solid P, Fe and Mn obtained from chemical extraction with 1N HCl and ascorbate solution. We also measured profiles of total C and S, organic C, and radionuclides. Fine grained muddy sediments were collected using a multi-corer at 150, 250, 550, and 1000 m depth on a section across the sloping bottom of the Bay of Biscay in Oct. 1997, Jan., June, Jul., and Oct. 1998.
Excess 210-Pb indicated that the 150 m and 250 m stations were highly bioturbated within the top 20 cm, whereas the deepest stations showed an exponential decrease in 210-Pb below 1 cm. Organic carbon content of the top sediment decreased from 2.5 to 1.3% from the shallowest to the deepest station. The C-org content decreased with depth in the deepest station cores, but it remained high and constant in the shallowest one. These shallowest cores also contained black sediments below 2 cm depth, highly enriched in FeS (up to 0.120 mmol S/g dry sediment). Each core was enriched in HCl and ascorbate soluble manganese in the surface layer. The enrichment was due to the presence of manganese (III) and (IV). Ascorbate-Mn dropped close to 0 at depth, but acid-extractable Mn (Mn-carbonates) was trapped in the anoxic sediments of the shelf stations due to alkalinity increase caused by sulfato-reduction. Each of the cores contained appreciable amounts of reactive iron(III) compounds extractable with ascorbate. At depth, reactive Fe(III) was reduced. Fe was trapped as sulphide only in the shelf stations. Phosphate associated with ascorbate-Fe at the top was partly trapped as authigenic phases at depth in all the stations. Pore water profiles indicated that the sediment was a source of nitrate for the overlying waters of the slope stations (high nitrate production in the sediment oxic layer), but a sink in the shelf stations (high denitrification). The shape of the ammonia profiles suggests that ammonia was almost entirely oxidized by Mn-oxides to N2 rather than by oxygen to nitrate. This was the rule for the four stations. We also observed subsurface anaerobic nitrate production, which could be related to NH4+ oxidation with bioturbated Mn-oxide particles.
The biogeochemical cycling of silica in nearshore environments is typically characterized by the uptake of dissolved Si by diatoms near the mouths of rivers and coastal surface waters, partial dissolution in the water column and surface sediment, and eventual burial of undissolved debris. In the case of the Amazon river, burial of biogenic silica is considered minimal, accounting for only 4% of the annual riverine input of dissolved silica (DeMaster et al, 1983). It has been hypothesized (Wollast and de Broeu 1971, Michalopoulos and Aller 1995) that portions of sedimentary biogenic silica must be converted to authigenic aluminosilicate minerals in major deltaic depocenters and thus may represent an additional pool of stored biogenic silica.
Converted diatom frustules that retain the original discoid morphology of unaltered diatom cells have been isolated from the Amazon delta and the coastal zone of Amapá, N. Brazil. SEM elemental mapping and TEM studies (electron diffraction, analytical electron microscopy and morphological characteristics) demonstrate that diatom frustules convert to poorly crystalline aluminosilicate phases and microcrystalline clay minerals that primarily contain K, Fe and lesser amounts of Mg. Euhedral pseudohexagonal clay mineral crystallites have formed within the converted siliceous frustules. Two types of euhedral authigenic clays were found, a dominant K-rich phase and a less-ubiquitous K-free type. Under the TEM, relics of diatom frustule microarchitecture can be discerned in otherwise converted frustules.
The rate of diatom conversion is very rapid and can be complete in months to 2 years. This is directly demonstrated with laboratory incubation experiments with cultured diatoms inserted into Amazon anoxic muds. After 20 months of incubation, cultured diatom cells of the genus Coscinodiscous sp. convert to authigenic phases, and on occasion to euhedral K-Fe-rich clay minerals replacing the original siliceous frustule. In addition, converted diatom particles are found in recently deposited (~ months) sediments near the sediment water interface, corroborating the rapid nature of the conversion process.
Sediment profiles of diatom counts, operational measurements of biogenic silica and estimates of the degree of frustule alteration with an Alteration Index (A.I. = number of altered diatoms/ total number of diatoms) demonstrate that biogenic silica undergoes significant changes during early diagenesis in modern Amazon deltaic sediments, one of the Earth's largest deltas. The conversion process results in a 5-fold increase (minimum estimate) of the amount of biogenic silica stored in Amazon shelf sediments as compared to previous estimates. Similar conversion processes may be encountered in other tropical estuaries resulting in comparable increases in biogenic silica storage in these depositional environments.
Demaster DJ, Knapp GB & Nittrouer, CA, Geochim. Cosmochim. Acta, 47, 1713-1723, (1983).
Michalopoulos, P & Aller, RC, Science, 270, 614-617, (1995).
Wollast, R & de Broeu, F, Geochim. Cosmochim. Acta, 35, 613-620, (1971).
The Ba repartition was investigated in the Atlantic and Indian sectors of the Southern Ocean to assess the reliability of Ba-barite signal in surface Antarctic sediments as a proxy for (paleo)-productivity of the upper water column (Dymond et al., 1992; Paytan et al., 1996; Nürnberg et al., 1997). The barite fraction was assimilated to "excess Ba fraction". This biogenic Ba fraction is from bulk geochemistry (ICP-MS, ICP-AES measurements) using Al as detrital reference.
In the Atlantic sector along a 6°W transect, the highest excess Ba content (~1600 ppm) occurs in the open Antarctic area (50° to 60° of latitude S), south of the Polar Front. It contrasts with the highest diatom productivity and maximum particulate Ba in the water column which are both recorded North of the Polar Front (Smetacek et al., 1997). The excess Ba content depicts a decrease through water depth from 2500 m, in agreement with lower barite saturation index (Monnin et al., submitted). This suggests a strong control of the barite stability by hydrostatic pressure: the highest Ba content being recorded in the shallowest samples occurring in the Antarctic area.
In the Indian sector, the highest excess Ba content (~1800 ppm) occurs South of 60°S in the sea-ice influenced Antarctic area (Bareille et al., 1991). Most of the biogenic silica and organic carbon deposits occur in the open Antarctic Ocean area (50° to 60°S, Bareille et al., 1991). There is no general trend of the Ba content through depth, although a slight increase of the excess Ba content seems to characterise the deepest sites in the zone covered by spring ice of the Antarctic area. As biogenic dissolution rate reaches 95% in this area (Bareille et al., 1991), we suggest the high sedimentary excess Ba content is under control of the decaying diatom tests allowing favourable barite saturated micro-environments.
Our study emphasises that the maximum sedimentary excess Ba content does not coincide with the highest surface water productivity in modern Southern Ocean. This demonstrates that other processes such as dissolution of the barite through water column settling and in the sediment or different modes of formation of barite crystals within microenvironments have to be taken into account before use of sedimentary Ba content as a productivity tracer.
Bareille G, Labrachie M, Labeyrie L, Pichon J-J & Turon J-L, Mar. Chem., 35, 537-551, (1991).
Dymond J, Suess E & Lyle M, Paleoceanogr., 7, 163-181, (1992).
Nürnberg C, Bohrmann G, Schlüter M & Frank M, Paleoceanogr., 12, 594-603, (1997).
Monnin C, Jeandel C, Cattaldo T & Dehairs F, Mar. Chem, submitted
Paytan A, Kastner M & Chavez FP, Science, 274, 1355-1357, (1996).
Smetacek V, de Baar HJ W, Bathmann UV, Lochte K & Rutgers van der Loeff M, Deep-Sea Res. II, 44, (1997).
The biogenic barite flux to the sediments of the Chilean continental slope generally correlates very well with the flux of organic carbon. However, this relationship breaks down during the Last Glacial Maximum (LGM) and most of the Holocene. The biogenic barite flux during the LGM is abnormally low and then increases sharply across Termination I, reaching a maximum during the early and middle Holocene and then declining to modern values. This pattern can neither be explained by productivity changes nor by changes in sediment provenance. Good correlation with other redox-sensitive elements (e.g. Fe, Mn) and the low organic carbon content of the sediment make it highly unlikely that the early Holocene barium peak is a result of a redox front. These peaks are present in three gravity cores from the Chilean continental slope, where they are identical on a common time scale but show different geometries when plotted against depth below sediment surface.
Assuming that organic carbon is a reliable proxy of paleoproductivity and knowing the flux of biogenic barite to the sediment, the algorithm proposed by (Dymond et al., 1992) can be solved for the concentration of barium dissolved in sea-water. The Ba concentration reached a minimum (70 nM/kg) during the LGM and then increased through most of the Holocene to values up to 170 nM/kg. During the late Holocene Ba concentrations in sea-water declined to the modern value of 110 nM/kg. This is matched by Ba/Ca ratios in benthic forams in the equatorial Pacific (Lea and Boyle, 1990). The shut-down of NADW production led to the development of water with a high ventilation age, enriched in barium in the North Atlantic. The resumption of NADW export from the North Atlantic led to an influx of water with a high ventilation age and elevated Ba concentrations through the Southern Ocean into the Pacific. This is supported by geochemical modelling (Lea and Boyle, 1990). The changes in the Ba concentration of sea-water in the Pacific are part of a global change in sea-water chemistry following the last glaciation. Biogenic barite can only be used as a proxy of paleoproductivity if the paleo-concentration of Ba in sea-water is known.
Dymond J, Suess E & Lyle M, Paleoceanography, 7 (2), 163-181, (1992).
Lea DW & Boyle EA, Paleoceanography, 5 (5), 719-742, (1990).
Examination of the solid phase of two sediment cores from the western and eastern equatorial Atlantic Ocean, Ceará and Sierra Leone Rises, respectively, revealed distinct barium peaks at glacial/interglacial boundaries (Kasten et al., subm.). These Ba spikes are unrelated to any other potential productivity proxy - as e.g. organic carbon, calcium carbonate or opal. Moreover they coincide with minima in organic carbon contents. Barium peaks of similar magnitude have also been reported to occur at glacial terminations in deposits of the northwest African margin (Matthewson et al., 1995), the Portuguese margin (Thomson et al., subm.) and the Ontong Java Plateau in the Pacific Ocean (Schwarz et al., 1996). Only the core from the Portuguese margin - characterized by comparatively high accumulation rates of > 10 cm/kyr - displays simultaneous maxima in the biogenic barium, organic carbon and diatom records. All other sampling sites mentioned above which have significantly lower sedimentation rates of < 5 cm/kyr show no positive correlation between barium and organic carbon contents.
We suggest that the prominent Ba spikes found at glacial terminations in sediments of the equatorial Atlantic Ocean formed in association with productivity pulses which were initiated by elevated nutrient contents in subsurface waters during deglacial periods. Evidence for short-duration high productivity episodes around glacial/interglacial transitions comes from (1) coincident peaks of biogenic barium and 13C values which have been proposed to be an indicator of nutrient contents in subsurface waters and (2) the coincidence in Ba and organic carbon maxima in the high accumulation sediments of the Portuguese margin (Thomson et al., subm.). The lack of correlation between Ba and organic carbon in the low sedimentation areas is attributed to an enhanced degradation efficiency of organic carbon along glacial/interglacial boundaries as a result of downward-moving oxidation fronts. These burn-down phenomena are likely to have been initiated by a decrease in bulk sedimentation rate and an increase in bottom water oxygen content during the transition from glacial to interglacial conditions. Moving downward into the previously suboxic sediments, the oxidation fronts led to an oxidation of the degradable organic carbon - leaving the solid phase distribution of Ba unaffected. This assumption is supported by the arrangement of the solid phase peaks of Ba, Fe and Mn along the glacial/interglacial boundaries which show striking similarities to the distribution of these elements in oxidized sapropel intervals in eastern Mediterranean sediments (van Santvoort et al., 1996).
Kasten S, Haese RR, Zabel M, Rühlemann C & Schulz HD, Chem Geol, (submitted).
Matthewson AP, Shimmield GB & Kroon D, Paleoceanography, 10, 677-692, (1995).
Schwarz B, Mangini A & Segl M, Geol Rdsch, 85, 536-545, (1996).
Thomson J, Nixon S, Summerhayes CP, Rohling EJ, Schönfeld J, Zahn R, Grootes P, Abrantes F, Gaspar L & Vaqueiro S, (submitted).
Van Santvoort PJM, De Lange GJ, Thomson J, Cussen H, Wilson TRS, Krom MD & Ströhle K, Geochim Cosmochim Acta, 60, 4007-4017, (1996).
Uranium and Manganese are two redox-sensitive trace elements with complementary behaviour. While U is immobile, Mn becomes mobile under anoxic conditions. Uranium dissolved in sea water thus diffuses into reducing sediments and is enriched there. The depth of the redoxcline and the sediment accumulation rate determine the enrichment of authigenic uranium. On the contrary, Mn diffuses from the reduced sediment sections and forms peaks of Mn-oxide just above the redoxcline. The depth of the subsurface peak of Mn, corresponds to the last event of shoaling of the redoxcline. This could be either caused by a higher flux of organic matter to the sediments or by lower oxygen concentration in the water column, or both. Modelling of the redoxcline as a function of the oxygen content of the water column and of the flux of organic matter to the sediments puts some constrains to these values during periods of extreme shallow redoxcline depths. Reconstruction of past ventilation from the sedimentary record is however difficult because enhanced oxidation of organic matter and deepening of the redoxcline during periods of better ventilation of deep water may alter significantly the organic matter and the uranium signals in the sediments. At present-day water conditions sediments accumulating at rates smaller than 2 cm/ka loose their redox signals due to burnd-down of organic matter. Focussing of sediments is an additional complication. The available data in the Atlantic Ocean rather suggests of periodic events of oxygen content in the water column as low as 100 µmol/l, significantly lower than today. They might be attributed to periodic flushing of the deep Atlantic with "older" southern component water during the last glacial. Peaks of U and Mn are also observed in glacial Pacific Ocean sediments.
Cu and Zn are trace elements in seawater. Dissolved Cu and Zn concentrations increase with depth in the water column, just like nutrients: the biological activity depletes them in surface while they are remineralized in the deeper waters through oxidation of organic matter by bacterial activity (Boyle et al., 1977; Bruland and Franks, 1983). Concentrations of dissolved Cu and Zn are substantially higher in old Pacific waters than in younger Atlantic waters as a result of the continuous remineralization of these elements. The stable geochemistry of these elements is poorly known because of the lack of an adequate analytical technique. Their isotope abundance can be now precisely measured by plasma source mass spectrometry on the Plasma 54 of Lyon in the multicollection mode. The 65Cu/63Cu and 66Zn/64Zn ratios are measured alternately on the samples and on a standard solution of the same element (Cu NIST 976 and a shelf JMC Zn solution). Instrumental mass fractionation is corrected with respect to the isotopic composition of a standard of a different element added to the sample (Zn JMC for a Cu sample, Cu NIST 976 for a Zn sample) previously purified by an ion-exchange chemistry. The external reproducibility of the isotopic measurements is 0.04 permil (Maréchal et al., in press).
Sediment trap material was collected in the Central Atlantic Ocean, west of the Mauritanian coast, under mesotrophic conditions (EUMELI program, French JGOFS), at interval of 10 days on a total period of about 2 years (from 1991 to 1992). Traps were deployed at 250, 1000 and 2500 m. The isotopic signals show regular variations, up to 0.12 permil for 66Zn and 0.25 permil for 65Cu. The variations of the two elements are somewhat consistent, but the modulation by the seasonal signal is ambiguous. During the bloom of March 1991, the particulate Cu concentration increases with depth whereas the Zn concentration decreases. Such a behavior documents both Cu scavenging and Zn remineralization during particles transit in the water column. A breakdown of concentrations and isotopic compositions of Cu and Zn between the different particulate components at the three depths shows that the residual organic material is depleted in the light isotope at 2500 m. Bacterial oxidation of the organic matter may be responsible of this depletion. We infer that variations of particulate 66Zn and 65Cu with time are due to water column processes rather than source effects. This assumption is corroborated by preliminary results showing that surface sediments have a higher isotopic value in the Pacific than in the Atlantic.
Boyle EA, Sclater FR, Edmond JM, Earth Plan Sc Let, 37, 38-54, (1977).
Bruland KW and Franks RP, Trace Metals in Seawater, Plenum, 395-414, (1983).
Maréchal CN, Télouk P, Albarède F, Chem Geol, (in press).
Mn and -D values of the interstitial water were determined at the lower plate reference (Site 1039) and the lowermost slope (Sites 1043, 1040 and 1041) of the Costa Rica margin. -D values range from -10.75 to +0.2 per mil at Site 1041 and from -2.14 to +2.4 per mil at Site 1039. -D and Cl values decrease from the oceanic domain to the apron. At Site 1040, the lower -D values and the negative Cl gradient in the upper 200 mbsf suggest a diffuse circulation of meteoric water through the pore media of the apron and the wedge. This diffusive circulation of fresh water could be less important at Site 1043, where channeled fluid flow occurs along the fluid conduit at 80 mbsf and also probably at 115 mbsf. The meteoric fluid is however, mixed with a deep fluid enriched in Ca and thermogenic methane. This deep fluid has an isotopic signature much heavier than that of the meteoric fluid, as shown by the D value of +2 per mil determined at Site 1043 wihin the décollement.
At Site 1039, dissolved Mn profile shows positive anomalies. The upper Mn peak could be related to the early diagenesis of the O.M.. At 100 mbsf, the positive Mn anomaly is associated with a methane-rich fluid. The residual Mn oxides are reduced by the fluid flow and then are released into the IW. In the deformed serie of Sites 1043 and 1040, two surprising features are observed:
(i)- No Mn peak is associated with Cl and C1,C2,C3 positive anomalies at depths of channeled fluid flow. The low solid phase Mn concentrations (<0.05%) are not controlled by Mn oxides, but more probably by carbonates. Also sustained by the high Mg/Ca ratios, and thermodynamic tests, this suggests Mg-Ca carbonate cristallizations with plausible Mn uptakes. The absence of Mn oxides may be partially explained by diagenetic reactions where dissolved Mn2+ is incorprated in carbonates, immediately it is released.
(ii)-The occurrence of a steep negative gradient of dissolved Mn, whereas alcalinity, PO4, NH4 increase and SO4 drops to zero. In comparison with Site 1039, a lower amount of Mn oxides might be integrated into the bulk sediment. Two conceptual models may be proposed. First, numerous fluid events and mud volcanoes were found on the lowermost slope. The venting of reducing fluid could provide a partial O2 consumption in the bottom SW. Second, the deformed wedge represents material of the sedimentary apron that has flowed downslope, and reducing IW could be expelled through the formation, and also contibuted to the partial O2 consumption of the bottom SW. It is of course possible that both processes occurred.
Distinct organic-rich intervals (Sapropels) are known to occur in a cyclic pattern in sediments of the eastern Mediterranean. Their occurrence has been reported to be associated with Monsoonal / Humidity climatic cycles. Stagnation / anoxic deep-water conditions and reversed circulation / enhanced productivity are amongst the most extreme mechanisms that have been hypothesized as a possible explanation for sapropel formation. In most cases, however, the one cannot be decoupled from the other; i.e. enhanced productivity almost automatically leads to reduced oxygen, or even anoxic, bottom water conditions. Our findings of high-resolution studies across sub-recent to Pliocene sapropels indicate that during all sapropel periods enhanced productivity did occur, whereas at least for some sapropel periods water column euxinic conditions occurred even up into the photic zone. Continued studies will be done to test the generality of the latter observations. Despite some mobilization that may occur due to extreme anoxic conditions, Barium appears to be a comparatively reliable indicator of relative organic carbon fluxes/paleo-productivities. The distribution of Ba and some trace elements on the one hand, and of the paleomagnetic parameters NRM/ARM on the other hand, appear to be accurate recorders of the initial presence of sapropels. In contrast, the presence of isorenieratene, the pyrite 34S isotope ratio, and the trace metal content of some sapropels points to a water column that is for the greater part anoxic. Clearly, sapropel formation takes place only during wet climatic conditions, whereas during non-sapropel conditions dry weather conditions appear to prevail.
This research was in part supported by the MAST contract #MAS2-CT93-0051, MAS3-CT97-00127, and GdL by GOA/ NWO grants 750.00.620-7290,
Slowly-accumulated (3-4 cm/ky) E. Mediterranean open basin cores from >2200 m water depth are compared with rapidly-accumulated (up to 20 cm/ky) cores to define the initiation and termination times of the most recent sapropel (S1). An underlying premise is that high deposition rates should first provide an improved time resolution, and second should minimise post-depositional effects which have thinned slowly-accumulated S1 units by oxidation of sapropel organic C. The validity of the Ba/Al ratio as a more persistent productivity index than organic C is proved in two S1 units accumulated at >15 cm/ky, where Ba/Al is directly related to the organic C content over the entire visual S1 units. AMS radiocarbon dating indicates a maximum duration for S1 from ~9,500-6,000 (uncorrected radiocarbon convention years b.p.) in the rapidly-accumulated cores and ~10,000-5,300 y b.p. in the slowly-accumulated cores. This difference between slowly- and rapidly-accumulated records is ascribed to a bioturbation mixing artefact affecting the slower-accumulated cores. In the two most rapidly-accumulated units, from the Adriatic and NE of Cyprus at water depths <1500 m, there is a "saddle" of lower values in both the organic C and Ba/Al profiles which makes the visual S1 unit appear as a doublet, approximately in the middle of the S1 productivity pulse and centred on 7500 y b.p. Geochemical and micropaleontological evidence indicates that this intervening period is best interpreted as an episode of increased water column oxygenation at intermediate depth rather than an interruption to sapropel formation, as productivity does not fall to the oligotrophic levels which preceded and post-dated S1. Like other Adriatic cores reported, our Adriatic S1 example has a distinctly shorter duration (8300-6300 y) than in any other core, further indicating the importance of new deep water on sapropel formation and preservation.
Pliocene to Holocene sediments in the eastern Mediterranean are characterised by the cyclic occurrence of organic-rich layers: sapropels. These layers are a few centimetres to decimetres thick. Apart from being enriched in organic carbon, sapropels contain pyrite (FeS2) as a result of bacterial sulphate reduction that prevailed during deposition of the sapropels. The most recent sapropel (Si2) is situated at a few decimetres below the sediment-water interface, and it was deposited between 5 and 9 kyr ago. After its deposition oxic sediments have buried it. At present, an oxidation front is situated at the top of Si2. At this front, organic carbon and pyrite are oxidised by oxidants that diffuse downward from the bottom waters into the sediments. As a result of this oxidation, the organic-rich layer becomes progressively thinner and iron oxides precipitate directly above the sapropel. These iron oxides record the geomagnetic field. We hope to resolve timing and mechanisms of formation of sedimentary iron compounds by studying magnetic properties in relation to geochemical data.
Three boxcores from the eastern Mediterranean were sampled at subcentimetre resolution (ABC26 from the Mediterranean Ridge, UM42 from Medina Rise, and BC19 from the Herodotus Abyssal Plain). A number of magnetic measurements have been performed (e.g. initial susceptibility, anhysteretic remanent magnetisation (ARM), isothermal remanent magnetisation (IRM), thermomagnetic analysis). Our first results show that different diagenetic zones are reflected by different magnetic properties. In ABC26 the initial susceptibility is low, dominated by paramagnetic silicates. ARM, however, reflects the behaviour of ferromagnetic minerals (predominantly magnetite). The increase of ARM with depth in the uppermost decimetre of ABC26 indicates the steady formation of magnetic iron oxides in the sediments, while a constant MDFARM (Median Destructive Field) indicates that the newly formed iron oxides are identical to the ones originally present. In between 10 and 20 cm depth, MDFARM is high (45 mT compared to 20 mT in the oxic top and 30 mT in the pyritized sediments below). This zone with high MDF coincides with the zone where iron oxides have precipitated above Si2. Within and below Si2 pyrite can be identified in thermomagnetic analysis in air through its oxidation via magnetite to hematite. We can identify two types of pyrite; one oxidises below 450oC and the other above 450oC. The higher pyrite oxidation temperature is predominantly found below the sapropel. This may be related to the microtexture of pyrite, which is euhedral below sapropels and mainly framboidal within sapropels.
Partly sponsored by the EU-MAST contracts Marflux (MAS1-CT90-0022), Palaeoflux (MAS2-CT93-0051) and SAP (MAS3-CT97-0137).
Organic biogeochemical aspects relevant to the link between water column processes and sedimentary record were investigated in a high frequence series of sinking particles and surficial sediments on the continental margin of the Gulf of Lions (NW Mediterranean Sea) as part of the European project MATER.
Bulk organic geochemical descriptors (organic carbon, nitrogen, main biochemicals) along with molecular lipid and amino acid biomarkers and associated fluxes have been considered in order to discern 1) the main allochthonous and autochthonous organic matter supplies 2) their evolution and fate during settling and on the sediment surface 3) the quantitative and qualitative links between downward fluxes and accumulation in the sediment.
Higher accumulation of sedimentary organic carbon is observed on the continental slope, and follows the accumulation of fine materials, a common trend in marginal areas (Buscail et al., 1997). A large part of the organic matter derives from the adjacent continent, as revealed by C/N ratios and specific terrestrial lipid biomarkers (n-alkanes, n-alcohols) and it appears to accumulate on the slope. Its composition reveals significant reworking which has mainly occurred on the shelf and during a dominant advectif transport (Buscail et al., 1990; Buscail and Germain, 1997). The extent of alteration is correlated to the width of the adjacent shelf. In contrast, components deriving from marine algal sources appear to accumulate near the shelf break. The relative importance of marine versus terrestrial material decreases seawards reflecting both the persistant nature of the terrestrial organic matter and the spatial patterns of the biological productivity in the area. Among the bulk marine organic supply several specific contributors from primary producers have been recognized by means of source-specific lipid biomarkers. Zooplankton grazing and bacterial reworking have altered the original lipid signature at various degrees and with different spatial patterns. Such alterations have been also identified in the sinking material along with additional signatures deriving by secondary processes.
In spring, one station was sampled at high frequency (ca. 7 days) for both sinking material and surficial sediment. A noticeable short term variability was observed for both fluxes and biogeochemical compositions of the sinking material. Similarly, significant variations were observed in the sedimentary organic matter contents. Individual rganic constituents such as main biochemicals (lipids, amino acids and sugars) as well as detailed biomarker abundance and composition also varied reflecting a) differences in the relative abundance of terrestrial and marine inputs b) variations of the various algal sources of the marine organic matter, c) variations of degradation processes at the water/sediment interface.
Qualitative and quantitative discrepancies in the biogeochemical characteristics of sinking particles and surficial sediments reflect complex aspects of the biogeochemical cycling (degradation, primary/secondary production, early diagenesis) and particle dynamics.
Buscail R, Pocklington R, Daumas R & Guidi L, Continental Shelf Res, 10, 1089-1122, (1990).
Buscail R & Germain C, Limnol. Oceanogr, 42, 217-229, (1997).
Buscail R, Ambatsian P, Monaco A & Bernat M, Mar. Geol, 137, 271-286, (1997).
For the past 400 000 years, at least 12 organic rich layers (Corg>2%) have been deposited in the eastern Mediterranean. These sapropel layers are preserved under anoxic conditions that may be due to a high gradient salinity between surface and bottom waters.
For the most recent sapropel S1 (9000-5000 years BP), several hypothesis have been proposed for the decrease of surface salinity, of which the most reasonable is presented by Rossignol-Strick et al. (1982) who incriminates abundant rainfall over Africa leading to a high Nile discharge.
In order to obtain information about the origin of the sedimented material and therefore on the climatic changes that triggered the deposition of sapropels, 143Nd/144 Nd and Rare Earth Element concentrations were measured in sediments from three different cores (18°, 25° and 29° long E) both for the total and leachable phases (HCl 1 M).
(i) The Nd isotopes results on the total phase confirm the East-West trend observed by Michard et al. (1995) with higher <epsilon>Nd0 in the East to lower <epsilon>Nd0 in the West.
(ii) <epsilon>Nd0 in the present sapropel and its oxidized fraction (as identified by double Mn peak (Higgs et al., 1994) or Ba Gaussian shape (Thomson et al., 1995)) have a similar signature and are higher than in the adjacent sediments. At one site (35° N, 18°E), however, we do not observe such variations.
(iii) The HCl leachings show more radiogenic <epsilon>Nd0 values (around 5<epsilon>Nd units higher) than the residues for all sampling sites.
Sediment samples were submitted to sequential leachings in an attempt to:
(1) understand the differences in <epsilon>Nd0 between the HCl leaching phase and the residue
(2) how Rare Earth Elements are distributed between the different phases (carbonates, iron and manganese hydroxides, organic matter and silicate residue).
Higgs N.C., Thomson J., Wilson T.R.S. and Croudace I.W., Geology, 22, 423-426, (1994).
Michard A., Thomson J. and De Lange G., EUG Conference, (1995).
Rossignol-Strick M., Nesteroff W., Olive P and Vergnaud-Grazzini C, Nature, 295, 105-110, (1982).
Thomson J, Higgs NC, Wilson TRS, Croudace IW, De Lange GJ and Van Santvoort PJM, Geochim., Cosmochim. Acta, 59, 3487-3501, (1995).
Nitrogen isotope distributions in the water column and sediments can be used to track present and past changes in the marine nitrogen cycle, including changes in paleoproductivity and paleodenitrification. The use of nitrogen isotopes rests on two constrains; an expected relationship between isotopic fractionation and nutrient uptake, where extensive fractionation results when nutrients are abundant, and secondly that, despite diagenesis, bulk organic matter retains the nitrogen isotopic signal imprinted on PON at the time of synthesis. However, the expected relationship between nutrient uptake and isotope fractionation has been difficult to replicate in laboratory cultures, and field data are often complicated by the presence of detrital material and by the alteration of 15N-PON by diagenesis. In order to circumvent those problems we have been developing analytical methodologies for making compound specific stable N-isotope measurements on chlorophyll. Chlorophyll is a light harvesting pigment present in all marine photoautotrophs, and as such is an excellent tracer for algal biomass. We are attempting to determine the relationship between 15N-PON and 15N-Chl, and thus exploit the biomarker potential of chlorophyll as a benchmark for algal nitrogen isotope values unaffected by detritus and early diagenetic alteration.To determine a relationship between nutrient uptake and isotope fractionation under different conditions, we are combining 15N measurements in DON, PON, DIN and chlorophyll-a with determinations of nutrient concentration in both controlled laboratory cultures and in the field. The field work being carried out in the Mediterranean Sea will be used to establish a budget of N-derived productivity in the basin.Preliminary results from the Mediterranean Sea are as follow: 1) 15N of nitrate at 1000 m depth was depleted (-0.05‰) with respect to the average value for the world ocean (5.7‰). This suggests that either NO3- is produced from oxidation of organic matter depleted in 15N, or that nitrogen fixation is an important primary production pathway (15N of N2 = 0‰). 2) 15N of PON was more depleted in the Eastern Basin (-0.6 ± 0.1‰) than in the Western Basin (+1.2 ± 1.4‰) and the Alboran Sea (+3.0 ± 0.9‰). This pattern is consistent with a scenario in which N2 fixation is more intense in the eastern Basin, where the supply of NO3- carried with Atlantic waters has been depleted. 3) 15N[PON-chlorophyll] was +5.8 ± 1.8‰. This offset is similar to 15N [phytoplankton-chlorophyll] observed in laboratory cultures (+5.2 ± 2.4‰). These preliminary results suggest that chlorophyll can be used to infer 15N of phytoplankton biomass. Current work is being done to constrain the effect of varying growth rates of phytoplankton under conditions of variable nutrient concentration and source on the isotopic signal.
During the last decade the application of microsensors has changed our view of biogeochemical dynamics in sediments as steep concentration gradients driven by microbial activity were discovered. Based on numerous techniques different electrodes were constructed to measure single constituents. To study the interaction of different redox-sensitive species (O2, HS-, Mn2+, Fe2+, I-) at the same time and depth a new microelectrode was developed (Brendal and Luther 1995). For the SMILE project (Sedimentary Manganese and Iron cycLEes) we apply this voltammetric microelectrode to coastal, deep sea and cold vent influenced sediments.
We will present an overview of results achieved for the SMILE project. Pore water data from the oligotrophic eastern Mediterranean Sea with O2 penetration depths > 20 cm as well as from a coastal area of the southern Netherlands with an O2 penetration depth of less than 3 mm will be shown. At the latter site we found local (over 2 mm) Mn2+ release in the sulfidic zone implying that local mineralization may significantly contribute to total mineralization. The heterogeneity of organic matter mineralization shall be further investigated by multi-profiling within one core. A third aspect of our multi-component microelectrode application in the field will be carried out during a submersible expedition to cold vents in the eastern Mediterranean Sea in December 1998. In addition, preliminary experimental results demonstrate the potential to study complex and nucleation processes involving Fe and HS- under controlled conditions.
Brendal PJ & Luther III GW, Environ. Sci. Technol, 29, 751-761, (1995).
The role of authigenic (Ca,Mn)-carbonate solid-solutions as a proxy for salt water ingressions from the North Sea into the anoxic environment of the central Baltic has been discussed by several authors. A general model of their formation includes oxidation of Mn(II) to Mn(IV) and precipitation of MnO2 which accumulates in Mn-rich layers at the sediment surface. When the bottom water becomes anoxic again MnO2 degradation releases Mn2+ into the pore water, and alkalinity increases as well during organic matter mineralization. Thus, (Mn,Ca)-carbonates are precipitated close to the sediment-water-interface where pore water are supersaturated with respect to such carbonate phases well discernible by light laminae in the dark sediment. In July 1997 surface sediment cores from the Gotland basin were sampled in order to study the formation of such mixed carbonates and their potential to be a quantitative proxy to reconstruct the paleohydrographical conditions in the deep basins of the Baltic Sea. The Mn/Ca ratio was 2.8 for five distinct laminae at 7 cm and at 53 cm sediment depth corresponding well with the thermodynamic energy minimum for a rhodochrosite/calcite solid-solution with a Mn/Ca ratio of approximately 3 (Prof. Kersten, Univ. Mainz, personal communication). As a product of the latest salt water inflows in 1993 and 1994 light laminae should occur at about 1 cm sediment depth. The former event in 1975/76 was expected to be documented at a depth of 6 - 7 cm. Surprisingly, the Mn-content of the uppermost 6.5 cm was below 0.02 M and light laminae could not be found. The 1975/76 event was reflected by a significant Mn-enrichment of 1.8 M at 7 cm, although light laminae were difficult to distinguish from the surrounding sediment. Furthermore Mn-contents in surficial sediments taken in August 1994 reach contents of about 0.5 M at the same location. The decrease of the Mn-content at the sediment surface from 1994 to 1997 and the absence of distinct light laminae in the uppermost layers may be explained by a re-dissolution of the precipitated (Mn,Ca)-carbonates. Due to increasing eutrophication in the last few decades the TOC content of the uppermost sediment has increased up to 12% leading to an extreme unstable, fluffy layer with a thickness of 5 - 6 cm at the top of the sediment column. Considering a possible re-suspension of this material an effective mixing with bottom water which is significantly undersaturated with respect to Mn- and Ca-carbonate phases may result in the re-dissolution of the (Mn,Ca)-carbonate.
Biogenic Ba contents in sediments has been widely used as a proxy of oceanic paleoproductivity (e.g. Dymond et al., 1992). Moreover, as barite is authigenic and well preserved in sediments, its geochemical signature could be a powerful tool to reconstruct paleoceanographic environments (e.g., seawater strontium isotope composition, Paytan et al., 1993). In this study, we suggest to isolate the barite-rich phases by using sequential chemical leaching on trap material and surface Antarctic sediments in order to identify their geochemical signatures and to have a better knowledge of Ba transfer from seawater to sediment. Several sequences of acid leaching (including HCl, HNO3, HF, acetic acid) with variable concentrations were tested. Elemental analyses of leachates (including Sr, Ca, REE, Al, Th, Rb) were determined by ICP-MS or ICP-AES. Presence of barite and its state of preservation was controlled by SEM investigations on residues.
In sediments, the use of excess acid during removal of the carbonate phase turns out to dissolve, even at low concentration (e.g. HCl 0.05N), a large fraction of barite. This is evidenced by high Ba content of leachates and by a lower abundance of, partly dissolved, barite crystals. Our results show that barites (1-5 µm) are easily dissolved in dilute acid (i.e., HNO3 0.5N, HCl 0.4N or HF 0.3N). After this step, Ba content of the solid phase leached reaches ca. 6%, which is much lower than an expected pure barite phase (59%). This emphasises that such separation technique does not completely separate barite from other diluting phases, especially from detrital particles, as already suggested by Martin et al. (1995). However, barite geochemical signatures can be identified by using elemental ratios such as Ba/Sr, which are different from one sediment to another. Contrary to sediments, in sub-surface trap material, Ba is associated with organic phases, as shown by the important part of Ba leached with H2O2.
Our approach strongly precludes the use of concentrated acid in chemical separation of barite from recent sedimentary material. It sheds light on the repartition and transfer of Ba from particles to sediment. In particular, our approach demonstrates that the biogenic Ba fraction is not only present as barite but is present also in organic material and carbonates with variable contents. This Ba repartition has to be taken into account when using excess Ba in sediment as a paleoproductivity proxy.
Dymond J, Suess E & Lyle M, Paleoceanography, 7, 163-181, (1992).
Paytan A, Kastner M, Martin EE, Macdougall JD & Herbert T, Nature, 366, 445-449, (1993).
Martin EE, Macdougall JD, Herbert TD, Paytan A & Kastner M, Geochim. Cosmochim. Acta, 59, 1353-1361, (1995).
The aim of the present study was the evaluation of the applicability of biogenic barium as a proxy for productivity. For this purpose 190 surface sediment samples from the South Atlantic Ocean which were collected in the frame of the Special Research Project SFB 261 at the University of Bremen were analysed for their barium and aluminium concentrations. The biogenic barium excess was calculated by subtracting the calculated terrigenous barium - obtained from the terrigenous Ba/Al ratio and the amount of Al in the sample - from the total Ba content in the sample. Based on the accumulation rates of biogenic barium export production was estimated using the three different algorithms proposed by Dymond et al. (1992), Francois et al. (1995) and Nürnberg et al. (1997). Primary productivity was calculated from these different export productions and compared with the actual primary productivity in the overlying surface waters.
Only the primary productions calculated on the basis of the algorithm of Dymond et al. (1992) yielded productivity values comparable to those existing in surface waters. This study further revealed that it is not valid to use a constant and generally applicable organic carbon/biogenic barium ratio as postulated by Francois et al. (1995). This ratio has to be regionally assessed - a result which has also been confirmed by examinations of sediment trap material. For the sediments of the Cape Basin in the eastern South Atlantic Ocean a new algorithm was developed which gives plausible primary productivities for the overlying surface waters.
Dymond J, Suess E & Lyle M, Paleoceanography, 7(2), 163-181, (1992).
Francois R, Honjo S, Manganini SJ & Ravizza GE, Global Biogeochem Cycles, 9, 289-303, (1995).
Nürnberg CC, Bohrmann G & Schlüter M, Paleoceanography, 12(4), 594-603, (1997).
Eastern Mediterranean Sea sediments consist of an alternation of organic-poor (<0.5 wt%), hemi-pelagic layers and organic-rich (up to 30 wt%), sapropel units. The formation of these sapropels is thought to be the result of increased productivity in the surface waters that led to higher settling rates of organic matter, and enhanced preservation of organic matter due to bottom water anoxia. The episodes of high productivity were presumably initiated by an increase in the continental runoff of freshwater with nutrients. It has also been hypothesized that enhanced benthic regeneration of phosphorus in the oxygen depleted bottom waters has helped to sustain the high productivity. This should be reflected in a reduced burial effiency (the ratio of the burial and depositional flux) of phosphorus in sapropel-times. Assuming that most of the phosphorus reaches the sediment in the form of organic matter, it is possible to estimate the depositional flux from paleo proxies. The burial flux of phosphorus is difficult to quantify, however, since sediment phosphorus undergoes strong diagenetic alterations. This particularly holds for sapropels that have been partly oxidized. To obtain more insight in the diagenesis of phosphorus during and after sapropel formation, high resolution pore water and solid phase data for two sites in the eastern Mediterranean Sea were collected. These data are used to determine (1) the exact position of the present and original sapropel unit S1, which is the most recent sapropel in the eastern Mediterranean Sea, (2) the position of the active oxidation front, and (3) the sequence(s) of events that could have lead to the depth distributions of organic, Fe-bound, detrital Ca-bound and authigenic Ca-bound phosphorus in these sediments. Finally, the potential importance of sediment phosphorus release under anoxic conditions for sapropel formation is discussed.
This work was funded by the EU-MAST-III programme MAS3-CT97-0137 (SAP).
The numerical model CoTReM (Column Transport and Reaction Model) is developed for the simulation of one-dimensional transport of solute and mineral phases and their interactions driven by bio-geochemical reactions and thermodynamic equilibria occurring in natural environments. CoTReM is the further developed and updated version of the model CoTAM (Hamer & Sieger, 1994). It is based on the calculation concept of operator-splitting (or mixing-cell approach; Schulz & Reardon, 1983) solving independently the different terms for transport and reaction. The program enables us to perform simulations with a chooseable set of dissolved and solid species. At present, a number of 30 species (minerals and solutes) are considered in CoTReM. Hydrodynamic transport simulation is covered by using the Crank-Nicolson (finite) differences scheme considering the advective, dispersive and diffusive mechanisms as well as ad- and desorption processes. Minerals are similar handled by sediment advection and bio-diffusion (bioturbation). Non-local mixing of solutes (bioirrigation) is solved separately. The important redox reactions are incorporated by a rate driven subroutine for presently 27 pre-defined full reaction equations. For the calculation of mineral equilibria the geochemical standard software PHREEQC (Parkhurst, 1995) is implemented. One of the most exceptional advantages is the possibility of modeling non-steady state processes within an inhomogeneous model area. The existing model-setting allows the simulation of various applications concerning processes in marine sediments and groundwater environments with complex conditions which have to be investigated via numerical simulations rather than analytical methods.
Pore water concentration profiles of marine sediments with a distinct sulfate-methane transition zone are presented. Deep sulfate reduction by methane occurs in these marine sediments (Niewöhner et al., 1998). This redox reaction produces sulfide, bicarbonate and water. The sulfide may be instantly precipitated due to high amounts of available iron minerals in the transition zone, e.g. the gravity core GeoB 4417 on the Amazon Fan, or sulfide may be dissolved in pore water (no sufficient amounts of reactive iron minerals) and distribute along the core controlled by diffusion. The upward limitation of the sulfide distribution in the latter case is given by the bottom water sediment boundary or a sulfate reoxidation zone, e. g. the gravity core GeoB 3714 in the upwelling area off Namibia.
Hamer K & Sieger R, Ernst & Sohn Verlag, Berlin, (1994).
Niewöhner C, Hensen C, Kasten S, Zabel M & Schulz HD, Geochim Cosmochim Acta, 62(3), 455-464, (1998).
Parkhurst DL, US Geol Survey, Water Resource Invest, 95-4227, (1995).
Schulz HD & Reardon EJ, Water Resources Res, 19, 493-502, (1983).
High resolution analysis of a deep sea sediment core from the Biotrans station in the NE Atlantic (Porcupine Basin, 47° 11´ N, 19° 34´ W, 4564 m water depth) allows to estimate the sedimentation rates over the last 12,000 years. Assuming a constant supply of 230Th from the water column to the sediment at the location of the core, a change in total sedimentation rate from 2.2 cm/ka for recent conditions to 5.8 cm/ka for the oldest sample (37.5 cm deep) could be demonstrated. The model reveals an almost constant flux of biogenic carbonate for the whole core, while the flux of the terrigenous fraction changed drastically. The time scale derived from the model shows a peak sedimentation of terrigenous matter 11.8 ka ago. This increase probably reflects the influence of the Younger Dryas on the sedimentation in the NE Atlantic. The fact that Heinrich layers were not encountered confirms the time scale derived from the model.
In the terrigenous fraction of the lower section of the core, fluxes of most major and minor elements were 5-10 times higher compared to recent flux rates. In particular the elements Ti, Zr and Hf had 15 fold higher fluxes during that time period. Basalt as a Ti-rich (and generally Mg-rich) source seems improbable because the change of the Mg/Al ratio with time does not reflect the change of the Ti/Al ratio. The enhanced deposition of Ti, Zr, Hf may be attributed to the increased input of ice-rafted detritus and/or a stronger eolian influence.
Multicorer cores from the western Arabian Sea were investigated with respect to the elemental distributions of Fe, Mn, Ca, Al, Ti, Ba, Cd, Cu and Pb in the solid phase and Fe, Mn, Si, Ca, Mg, Cd, Cu and Pb in the pore water. The cores were taken on two adjacent stations: WAST-Top (16° 10,5' N, 59° 46' E; 1915 m waterdepth) on the Owen Ridge and WAST-Plain (16° 13' N, 60° 16'E; 4045 m waterdepth) located about 30 nm east of WAST-Top. The sediments on both stations differ significiantly: WAST-Top consists of light olive grey coloured, carbonate rich (70-80 wt% CaCO3) foraminiferal sand. Due to the sediment structure the length of the core is less than 20 cm. The accumulation seems to be continuous through time. WAST-Plain is characterized by a turbidity sequence. Its upper boundary is at about 8 to 10 cm (or less?) below the sediment surface. The carbonate content is within a range of 40-55 wt%. The turbidity sequence has high water contents (80 Vol% at a depth of 30 cm), a noticeable dark olive grey color and a very fine grain size. The lower boundary of the turbidite was not reached within the multicorer, although some cores reached depths of more than 50 cm below sediment surface. At WAST-Plain the content of organic carbon increases from the benthic interface to a core depth of about 5 cm, where a level of about 2,3 wt% is reached. This level continues to the end of the cores.
Two different methods for pore water sampling were used: i. pressure filtration in teflon sqeezers under argon atmosphere in a glove box (the remaining sediment was used for solid phase investigations). ii. diffusive gradient in thin films technique (DGT) with a time of application of about 24 hours in a multicorer core, which was stored in a cooling lab during sampling.
The element concentrations in the spolid phase are mainly influenced by the carbonate content. When Ti (and also Fe in the youngest part of the cores) is normalized to the Al content, the concentrations show nearly the same ranges in the non-marine fraction on both stations. This indicates the same source for terrigenous matter (fluviatil input and dust). Sedimented Fe-and Mn-minerals undergo alteration (solution and precipitation) during imbedding by diagenetic processes, which are closely connected to the degradation of organic matter. Trace metals, which are bound to these minerals and to the organic substance are released or fixed according to these processes. Investigations of the pore water concentration profiles show characteristics of the geochemical behaviour of Cd, Cu and Pb in dependence on the respective environmental condition. The high contents of organic matter in the turbidity sequence at WAST-Plain give evidence to the presumption that the sediment/pore water system (in the upper parts of the cores) is not in a "steady state" due to the singularity of the tubidite event.
Methane seeps in the northwestern Black Sea were found to be sites of intense carbonate precipitation (Luth et al., 1998). Samples were collected at locations on the Romanian (120 m depth) and Ukrainian (180 - 200 m depth) shelf and slope, respectively. Methane gas plumes were detected by echosounding and seep deposits were obtained with a 2.5 m beam trawl. The layered carbonate crusts are formed of (i) carbonate-cemented siliciclastic sediments containing the bivalve Dreissena caspia crassa ANDRUSOV, (ii) microcrystalline Mg-calcite, and (iii) aragonitic cements. The Dreissena sediments are subrecent and date from the early Holocene freshwater-phase of the Black Sea. These sediments were affected by the seepage which induced intergranular precipitation of methane-derived carbonates. Micrites contain detritical quartz, framboidal pyrite, and glauconite. They exhibit an intense autofluorescence which is probably related to a high content in organic residues. Aragonitic cements form either isopachous layers or botryoids. Precipitation of micrites and aragonites preserved microbial filaments about 10 to 20 µm in diameter and up to 900 µm in length. These filaments are significantly larger than mat-forming methanogens which have been described from the same location (Pimenov et al., 1998). The carbonates are extremely depleted in 13C. Micrites range from -27 to -41‰ PDB, and botryoidal aragonites range from -26 to -38‰ PDB. The extreme 13C depletion indicates that the carbonates were derived from the microbial oxidation of methane. The 18O values of carbonates exhibit an extraordinarily narrow range from +0,1 to 0,0‰ PDB. Carbonates are accompanied by blocks and crusts composed of pyrite. The framboidal sulfides exhibit a palisade-like texture with framboids being arranged to parallel pillars. Sulfur isotopic ratios (34S) averaging +18‰ CDT indicate that the sulfur is not derived from the 34S-depleted H2S of the water column or the uppermost sediment layers. Most likely, H2S seeping from deeper horizons is the source of sulfur in the iron sulfides. Bacterial reduction of sulfate in a closed system causes the preferential loss of 32S from the reservoir. Obviously, the sulfur of the precipitates is derived from a residual pool in which the preceding sulfate consumption led to an enrichment of 34S.
Thiel H, Luth C, Luth U, in: (Luth U, Luth C, Thiel H, eds), Ber. Zentr. Meeres. Klimaforsch., Reihe E, 14, 5-9, (1998).
Pimenov NV, Rusanov II, Poglazova MN, Mityushina LL, Sorokin DY, Khmelenina VN, Trotsenko YA, in: (Luth U, Luth C, Thiel H, eds), Ber. Zentr. Meeres. Klimaforsch., Reihe E, 14, 37-50, (1998).
The composition of organic matter (OM) in the marine environment is affected by endogenous and exogenous factors, such as primary production, heterotrophic activity, particle dynamics and circulation, and riverine and eolian inputs. During settling in the water column and deposition to the sediment, it is further modified by complex mineralization processes and biochemical transformations.
This study focuses on the assessment of the various sources of organic matter (OM), both marine and terrestrial, in an oligotrophic marine area of the Eastern Mediterranean (Cretan Sea). Alteration processes which affect the abundance and composition of OM during settling in the water column and accumulation in the sediment are also investigated. In order to address these questions, a detailed study of seven distinct lipid compound classes in sinking material and surface sediments collected from the nearshore towards the deep Cretan basin was carried out.
In both sinking particles and sediments, concentrations and fluxes of marine markers, such as sterols, long-chain alkenones and fatty acids were generally low. The relatively elevated contributions of these markers during spring, reflects a higher productivity occurring in this period. Considering the contributions from algal species, coccolithophorids were significantly more abundant in April, as evidenced by alkenone and brassicasterol higher fluxes. In contrast, the seasonal variation of dinosterol and long chain diols and keto-ols indicated a higher contribution of dinoflagellates and nanoplankton species in summer. The biogenic terrestrial organic markers (long chain n-alkanes, n-alkanols and n-alkanoic acids) were an important component of the lipid fraction, underlining the significance of the allochthonous inputs in this oligotrophic marine area. Fluxes of anthropogenic compounds (PAH) were low compared to those found in the Western Mediterranean Sea. They showed a pronounced pyrolytic character during winter and spring periods, while during summer, an admixture of both petrogenic and pyrolytic sources was recorded.
Planktonic markers decrease highly during settling and sedimentation, due to degradation and heterotrophic alteration processes. This was confirmed by the corresponding increase of alteration products and microbial marker fluxes (isoprenoid compounds, stanols, microbial alcohols and fatty acids) with depth. On the other hand, terrestrial markers were found to be more resistant to degradation processes in the water column and accumulated in the underlying sediment in proportions one to three orders of magnitude higher than planktonic markers. This can be attributed to: i) the oligotrophic character of the study area, which does not favour the rapid transport, accumulation and preservation of marine OM in the sediment and ii) the refractory character of the terrestrial OM. As a consequence, the coupling between the water column and the surface sediment is more pronounced for the terrestrially-derived OM than for the marine one.
The eastern Mediterranean Holocene sedimentary sequences are characterized by the occurrence of a dark colored organic-rich layer (sapropel S1) within normal marine sediment. The depositio of S1 appears to be related to significant changes in climate, water circulation and biogeochemical cycling. To obtain insights on the mechanism of S1 formation and to reconstruct its original thickness micropaleontological quantitative analyses was performed on calcareous nannofossils, planktonic foraminiferal and pteropods assemblages in two boxcores, BC19 and BC42, aligned in an E-W transect. A 1 cm sampling was used for the micropaleontological investigations. Distinctive changes in the abundance of some nannofossils taxa were identified. In BC19 Helicosphaera carteri increases in abundance below S1 reaching the maximum value (22%) in the lower part of S1 and remain low through the rest of the interval. In BC42 H. carteri shows a peak (23%) above the visual top of S1. In both boxcores the increase of H. carteri coincides with a decline of Syracosphaera sp.1, which is virtually absent in S1 up to few cm above it. This change doesn't coincide with lithological variation. Florisphaera profunda increases below S1 reaching its highest peak within S1 in BC19 and a lower peak above the S1 visual top in BC42. F. profunda proliferates in the lower photic zone within the DCM and its abundance is inversely correlated with the abundance of all other coccolitophorids. In both cores faunal assemblages start and continue to change before the lithological evidence of S1. There is a corrispondence with foraminifera and pteropods fluctuations. The planktonic foraminifera assemblage of S1 differs considerably from those in the sediments above and below. There is an increase in abundance of Truncorotalia truncatulinoides, Globorotalia inflata, Globigerina bulloides and dextral Neogloboquadrina pachyderma prior to S1; the middle part of S1 is characterized by a peak in frequency of Globigerinoides ruber (var. rosea and alba) and some species of SPRUDTS group (=warm peak) that continue some cm above S1 in both cores. G. bulloides occurs predominantly in cold subpolar water and it is prolific below the thermocline; in the modern assemblages peaks of N. pachyderma (dextral) are at the deep of DCM. So the presence of this association indicates periods of high fertility. Moreover, in BC42, about 5-6 cm above S1, planktonic foraminiferal assemblage is characterized by peaks of N. pachyderma, G. bulloides, Turborotalia quinqueloba and Globorotalia scitula, whereas warm species decrease, indicating the onset of cold condition as supported by heavier 18 O values measured on G. ruber. Pteropods among other mollusks have provided excellent climate indicators as Limacina inflata, L. trochiformis (warm indicators) and L. retroversa (cold indicator). Data from the fossil group suggest that S1 was originally thicker than now visible.
Three vertically moored sediment traps were deployed at 48oN21oW at 3 depths for 1 year from April 1989 to April 1990 (Joint Global Ocean Flux Program (JGOFS) North Atlantic Bloom Experiment (NABE)). We present here the coccolithophore results obtained from 3.7 km water depth (sea floor at 4.4 km) and from the top sediments of six box cores, recovered between 44-47oN and 20-24oW, at water depths ranging between 3.2 and 4.4 km. Calcium carbonate was the largest component of the settling particles (61%), particulate organic matter and opal followed (Honjo and Manganini, 1993). In the calcareous ooze underneath the trap location, the carbonate represented 77-90% of the total mass and with fine size fraction (<32µm) contributing about 70% to the total carbonate (Van Kreveld et al., 1995). The foraminiferal lysocline in this area is at about 4000 m water depth. As far the dominant coccolith species are concerned, the coccolith taxonomic composition of the trap samples roughly reflects the assemblages in the sediment samples. However, selective dissolution of coccoliths does occur, and species diversity is lower in the fossil record. The total and the fine fraction carbonate, the coccolith accumulation rates and different dissolution indexes will be discussed. We propose a dissolution index related to the modification of coccolith morphology (mainly E. huxleyi and gephyrocapsids) by dissolution and fragmentation (due to mechanical breakage) in the surface sediments. In the studied depth transect we estimate the maximum loss due to dissolution to beless or equal 50%. Under these circumstances coccolith accumulation rates and changes in nannofossil assemblage proved unsatisfactory as indicators of dissolution. As an alternative, the ultrastructural breakdown of E. huxleyi (the most abundant living coccolithophore species) provides a potential dissolution indicator that can be numerically expressed and calibrated with carbonate loss. So, we propose a dissolution index related to the modification of coccolith morphology (mainly E. huxleyi and gephyrocapsids) by dissolution and fragmentation (due to mechanical breakage) in the surface sediments.
Honjo S & Manganini SJ, Deep-Sea Research II, 40, 587-607, (1993).
Van Kreveld SA, Ganssen GM, van Hinte JE, Melkert MM, Troelstra SR, van der Borg K & de Jong A, Radiocarbon, 37, 585-592, (1995).
The shelf of the Great Australian Bight has been the site of cool-water carbonate sedimentation since Eocene time, resulting in an almost 1-km-thick succession, and it is now the largest area on the globe composed of such sediments. The high primary depositional permeability of winnowed grainstones of the Eucla Shelf and the lack of early cementation suggest that significant groundwater circulation may occur. Porewater profiles measured during Leg 182 of the Ocean Drilling Program, indicate that the margin of the ramp contains a complex system of different brines. The preliminary data allows us to distinguish between two brines, both derived from seawater, but with different salinities (80 and 105). We can additionally identify the influence of porefluids which are characterized by a distinctly different sodium/chlorinity ratio, possibly derived from a halite deposit.In contrast to the deep seated high salinity brines, the sodium/chlorinity anomaly is confined to shallow depths (< 500 mbsf) and a bathymetric interval between 250 and 600 m water depth. The brines, being a major sulfate source, allow for the development of extended sulfate reduction zones, producing alkalinity values as high as 140 and large amounts of H2S.
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications