A sulfur budget for the Black Sea anoxic zone

A sulfur budget for the Black Sea anoxic zone

Deep-Sea Research I 48 (2001) 2569–2593 A sulfur budget for the Black Sea anoxic zone Lev N. Neretina,*, Igor I. Volkovb, Michael E. Bo. ttchera, Vla...

339KB Sizes 1 Downloads 42 Views

Deep-Sea Research I 48 (2001) 2569–2593

A sulfur budget for the Black Sea anoxic zone Lev N. Neretina,*, Igor I. Volkovb, Michael E. Bo. ttchera, Vladimir A. Grinenkoc a

Department of Biogeochemistry, Max-Planck Institute for Marine Microbiology, Celsiusstrasse 1, D-28359 Bremen, Germany b P.P. Shirshov Institute of Oceanology RAS, Nakhimovskiy prosp. 37, 117851 Moscow, Russia c V.I. Vernadskiy Institute of Geochemistry and Analytical Chemistry, Kosygina st. 19, 117975 Moscow, Russia Received 11 December 2000; accepted 6 April 2001

Abstract A budget for the sulfur cycle in the Black Sea is proposed which incorporates specific biogeochemical process rates. The average sulfide production in the water column is estimated to be 30–50 Tg yr1, occurring essentially in the layer between 500 and 2000 m. About 3.2–5.2 Tg sulfide yr1 form during sulfate reduction in surface sediments of the anoxic zone. Total sulfur burial in anoxic sediments of 1 Tg yr1 consists of 10–70% (ca. 40–50% is the average) water column formed (syngenetic) component, the rest being diagenetic pyrite. As a maximum, between 3 and 5 Tg yr1 contribute sulfide to the bottom water or diffuse downward in the sediment. About 20–50 Tg yr1 sulfide is oxidized mostly at the chemocline and about 10–20% of this amount (4.4–9.2 Tg yr1) below the chemocline by the oxygen of the Lower Bosphorus Current. A model simulating the vertical distribution of sulfide in the Black Sea water column shows net consumption in the upper layers down to ca. 500 m, essentially due to oxidation at the chemocline, and net production down to the bottom. On the basis of the calculated budget anoxic conditions in the Black Sea are sustained by the balance between sulfide production in the anoxic water column and oxidation at the chemocline. On average the residence time of sulfide in the anoxic zone is about 90–150 yr, comparable to the water exchange time between oxic and anoxic zones. Hydrophysical control on the sulfur cycle appears to be the main factor regulating the extent of anoxic conditions in the Black Sea water column, rather than rates of biogeochemical processes. r 2001 Elsevier Science Ltd. All rights reserved. Keywords: Black sea; Water column; Sulfur budget; Anoxic conditions

*Corresponding author. E-mail address: [email protected] (L.N. Neretin). 0967-0637/01/$ - see front matter r 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 7 - 0 6 3 7 ( 0 1 ) 0 0 0 3 0 - 9

2570

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

1. Introduction The Black Sea is the largest anoxic basin in the world. The total sulfide pool of the contemporary Black Sea is about 4.6  103 Tg, the main part being located between 500 and 2000 m (Table 1). The spatial distribution of the oxic–anoxic transition in the sea is characterized by the presence of two dome-shaped structures located in the eastern and western central parts of the basin between 90 and 100 m. Along the periphery, the upper anoxic boundary deepens to 150– 180 m. (e.g., Skopintsev, 1975; Codispoti et al., 1991). Its location in the basin is strongly dependent on the hydrophysical structure of the pycnocline, and on average corresponds to a density surface of about 16.2 (e.g., Vinogradov and Nalbandov, 1990; Konovalov et al., 1999) independent of geographic position. The spatial distribution of hydrogen sulfide in the upper 500– 600 m of the anoxic column follows hydro-dynamical structures, both primary (main cyclonic gyres, Crimean and south-eastern anticyclones) and secondary (e.g., near-shore anticyclonic gyres . zsoy and U . nlu. ata, 1997), temporary gyres) (Neretin et al., 1997). (O Several hypotheses have been suggested to explain the temporal history of the Black Sea anoxia (Deuser, 1974; Glenn and Arthur, 1985; Calvert et al., 1987; Jones and Gagnon, 1994; Lane-Serff et al., 1997; Arthur and Dean, 1998). A comprehensive set of radiocarbon data by Jones and Gagnon (1994) reviewed an earlier hypothesis of a time-dependent evolution of the anoxic zone in the Black Sea. The observations by these authors, later supported by Arthur and Dean (1998), showed that water column anoxia developed simultaneously between 200 and 2220 m about 7.5– 7.8 ka BP. From this perspective, the Black Sea can serve as a case study how anoxic environments persist through geological time. In this paper we review data collected for several decades and estimate the fluxes for a complete sulfur budget of the Black Sea anoxic zone. Previous work considered only a part of the budget, or presented an inconsistent picture (Skopintsev, 1975; Lein et al., 1990; Lein and Ivanov, 1991; Bezborodov and Eremeev, 1993; Albert et al., 1995). In addition to compiling preexisting information, we used our own data to calculate hydrogen sulfide production and oxidation at the chemocline. The study also includes new sulfur isotope data and presents a model of the Black Sea anoxic zone, which improves our knowledge about this ecosystem. Table 1 Hydrogen sulfide inventory in the Black Sea (compiled from our data) Layer (m)

Weighted average SH2S (mmol l1)

SH2S  109 (t)

SH2S (mol m–2)

100–200 200–300 300–500 500–1000 1000–1500 1500–2000 2000–2200

21a 66 140 274 339 368 368b

0.022 0.068 0.283 1.315 1.469 1.220 0.210

2.0 6.5 28.1 132.1 159.1 150.7 40.7

4.587

519.2

Total a b

For the layer 150–200 m. For the layer 1500–2000 m.

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2571

The paper is divided into several parts according to the process considered. Hydrogen sulfide production in water column and sediments is presented first. In the discussion of sulfide production in the water column, gross production measured by radio-isotope label and net production derived by a modeling based on vertical distribution of total inorganic carbon are considered separately. Sulfide production in bottom sediments is calculated from literature data on sulfate reduction rates, and the average flux of sulfide at the sediment/water interface is estimated. We summarize all available information on sulfide oxidation rates at the chemocline, and present hypotheses about mechanisms controlling sulfide oxidation. Net production and consumption rates of sulfide in the Black Sea water column are calculated by the one-dimensional model. The role of syngenetic pyritization vs. diagenetic pyritization for the sulfide budget is estimated. The maximum estimate of sulfide removal rate by the Lower Bosphorus Current is given in the last part the paper. At the end we present the overall sulfur budget for the Black Sea anoxic zone. We show that hydrogen sulfide production in the Black Sea water column is the main source of sulfide, and anoxic sediments contribute to only a minor extent to the water column H2S pool. The dissolved oxygen carried by the Lower Bosphorus Current cannot explain more than 10–20% of sulfide consumption rate at the chemocline, and the latter is comparable with sulfide production rate in the water column. The rate of pyritization in water and sediments together does not exceed 3% of the sulfide production rate. Where appropriate, we try to point the way for future studies of the Black Sea anoxic zone.

2. Methods Data used to calculate average concentrations of sulfide in the Black Sea were obtained during a 6-year (1992–1997) field campaign in the northeastern part of the basin in different seasons (November 1993 RV Vityaz, September 1994, October 1994, March 1995 RV Akvanavt, June– July 1996 RV Yantar, September 1997 RV Petr Kotsov). The location of the stations and the itinerary of the cruises are presented elsewhere (Neretin, 1996). Briefly, water samples were obtained with a Sea-Bird CTD rosette system from 5 l bottles. All hydrogen sulfide determinations and sample fixation for isotope analyses were performed onboard the ship immediately after sample retrieval. New compositional data were collected for dissolved sulfide in sediments from the western part of the Black Sea. The locations of sampling stations for the isotope measurements of sulfur species in this study and previously published data are given in Fig. 1. Sulfide concentrations in the upper part of the anoxic zone (below 30 mmol l1) were determined by the methylene blue method (Cline, 1969) and higher concentrations by iodometric titration (Standard methods, 1992). 2.1. Sulfur isotopic analyses 34

S/32S values of pore water hydrogen sulfide from Stations 6–8 were measured by combustionisotope-ratio-monitoring mass spectrometry (C-irmMS) using a Carlo Erba EA 1108 elemental analyzer connected to a Finnigan MAT 252 mass spectrometer via Finnigan MAT Conflo II split interface, as described by Bo. ttcher et al. (1998). Reproducibility was 0.2%.

2572

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

Fig. 1. Stations sampled for pyrite and S species isotopic composition. The boxed-in rectangle in the NE part of the Black Sea is the area where most data for hydrogen sulfide concentrations were obtained. The shaded area within the boxed-in rectangle represents the polygon of five stations sampled by I.I. Volkov (unpublished data, 1993). Station numbers are given as in the original papers: Stt. 4740, 4745, 4747, 4750, 4752, 4753, 4754 (Vinogradov et al., 1962); 1990, 1991 (Migdisov et al., 1974); 1135, 1136 (Sweeney and Kaplan, 1980); 545, 546, 571, 580, 582, 584, 601, 620 (Vainshteyn et al., 1986); 55, 840 (Strizhov et al., 1989); BS4-14GC (Calvert et al., 1996) the same as St.14 (Lyons, 1997); 9, 15 (Lyons, 1997); 6–8 (our data).

34

S/32S ratios in the paper are given in the d-notation with respect to the Vienna CDT (VCDT) standard: " 34

d S¼

ð34 S=32 SÞsample ð34 S=32 SÞVCDT

#  1 1000%:

Silver sulfide (IAEA-S-1) (d34S=0.3%) was used for the calibration of the mass spectrometer, and a d34S value of +20.5970.08% (n ¼ 10) was obtained for the IAEA barium sulfate NBS-127. It should be noted that a comparison of some isotope measurements from this study with data from the literature include some uncertainties, because the recommended standardization of stable sulfur isotope ratios changed from CDT to VCDT (Coplen and Krouse, 1998). Previous studies often did not report isotope results for international standards to make their data comparable to other laboratories. For instance, Sweeney and Kaplan (1980) reported their results for California coastal seawater sulfate to be +19.8% vs. CDT compared to +20.5% vs. VCDT found for coastal seawater of the North Sea (Bo. ttcher et al., 1997). Additionally, the isotopic inhomogeneity of the CDT material itself is at least 0.4% (Coplen and Krouse, 1998). In order to avoid additional uncertainties in the present study, we report the former results on the CDT scale, and compare data reported for the CDT with the data of the same isotopic scale. Our recent data reported for the VCDT standard are consistent with each other, but cannot be precisely compared

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2573

with the data reported for CDT scale. We estimate that the uncertainty is less than 1%, significantly smaller than the range of the reported isotope values for dissolved sulfide and pyrite in Black Sea sediments. 2.2. Simulation of hydrogen sulfide vertical profile by one-dimensional numerical model We use the modeling software PROFILE of Berg et al. (1998) to calculate the depth distribution of sulfate reduction and hydrogen sulfide consumption rates in the Black Sea water column. The model is a robust numerical procedure for the analysis of vertical profiles in sediment pore waters. As a first approximation, without considering an advection term for the one-dimensional mass conservation equation, this model can be used also for the numerical simulation of vertical profiles in the water column (bioturbation and bioirrigation terms are also neglected). The numerical procedure involves finding of least square fits to the measured concentration profile, and comparison of these fits through F-tests. The objective of the model is to select the simplest net production–consumption profile that resembles the measured concentrations of the modeled solutes. An average hydrogen sulfide vertical distribution of the Black Sea water column (Fig. 2A) was chosen for the simulation. A profile of the vertical diffusion coefficient Kz used in the model is presented in Fig. 2B. There are several numerical approaches to calculate this coefficient. We used the model developed by Boguslavsky and Kotovschikov (1984) and later refined by Boguslavsky and Ivaschenko (1990). The model takes into account the vertical gradient of salinity. The vertical profile of Kz in the anoxic zone has a pronounced minimum at the oxic/anoxic interface of about

Fig. 2. The average depth distribution of hydrogen sulfide (our data) (A) and vertical diffusion coefficient Kz (from Boguslavsky and Ivaschenko, 1990) (B) in the Black Sea water column.

2574

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

0.2 cm2 s1 corresponding to the maximum salinity gradient; at the intermediate depths Kz values increase up to 3.2 cm2 s1, and then decrease again to about 1.3 cm2 s1. The model divides the whole water column into several layers with similar rates of net production/consumption, and during subsequent numerical approximations chooses the best fit to the observed profile based on F-testing. We started all numerical simulations from eight zones. Four different boundary conditions were tested: I. constant H2S concentrations of 0 mmol l1 at the top (100 m) and 369 mmol l1 at the bottom (2200 m) of the model; II. constant H2S concentration (369 mmol l1) and zero sulfide flux at the bottom; III. constant H2S concentration at the top (0 mmol l1) and zero sulfide flux at the bottom; IV. constant H2S concentration at the top (0 mmol l1) and sulfide flux from bottom sediments into the water column of 0.31 mol m2 yr1. The different boundary conditions all gave similar results, and only the results for the boundary conditions I are discussed in the paper. The modeled net consumption (negative values) and production (positive values) rates are expressed in nmol cm3 day1, or depth-integrated rates in nmol cm2 s1. 2.3. Sulfide production in deep-sea sediments First measurements of sulfate reduction rates (SRR) in sediments of the Black Sea by Sorokin (1962) showed significant spatial variability with maximum rates at the periphery of the basin between 475 and 1241 mmol m2 yr1 and rates between 11 and 44 mmol m2 yr1 in the central parts. The intermediate zone between the two main Black Sea gyres was characterized by values between 146 and 256 mmol m2 yr1. Based on these measurements, Deuser (1971) calculated an average annual sulfide production in Black Sea sediments of 3.6 Tg. No sulfate reduction was measured below the uppermost 5 cm of sediment (Sorokin, 1962). Unlike these first data, Vainshteyn and co-authors (1986) showed for the western Black Sea that sulfate reduction takes place throughout the whole Holocene and upper Pleistocene sediment sequence. Measured SRR varied between 52 and 85 mmol m2 yr1 depending on the sediment and water depth. Lein et al. (1990) used these and new data to calculate the average hydrogen sulfide production in the anoxic sediments of the Black Sea of about 560 mmol m2 yr1, or 5.9 Tg yr1. This estimate is higher than Deuser’s because the whole Holocene sequence was considered. Recent measurements by Albert and co-authors (1995) gave an average sulfide production in the upper 20 cm (including a 2 cm fluffy layer) of about 5.2 Tg yr1. Critical for the hydrogen sulfide budget of the water column is the amount of sulfide diffusing from the pore waters into the bottom waters, which depends on the concentration gradient at the sediment/water interface. The balance between sulfate reduction rates, input of reactive iron, rate of pyritization, and physical factors, such as sediment slides and turbidites at the continental slope, and advective pore water exchange in situ (cf. Hu. ttel et al., 1998), will influence sulfide concentration in pore water and the concentration gradient. Significant spatial variability of all parameters, for example reactive iron content (Rozanov et al., 1974), yields different flux directions at different localities. The estimate for this flux is presented below.

2575

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2.4. Hydrogen sulfide production in the Black Sea water column All measurements of SRRs in the Black Sea water column are summarized in Table 2. Generally, a maximum in sulfate reduction rates was observed in the upper (200–300 m down to 600–700 m) part of the anoxic column, and in the layers adjacent to the bottom. The highest rate measured for the upper anoxic zone so far is 1569 nmol l1 day1 (Il’chenko and Sorokin, 1991). The lowest SRRs in the water column were reported by Albert et al. (1995) and did not exceed 3.5 nmol l1 day1. With a sensitivity of the method of about 0.2–0.6 nmol l1 day1 (Lein et al., 1990; Albert et al., 1995), reduction of sulfate in the intermediate zone (600–700 down to 2000 m) comprising the main part of the Black Sea hydrogen sulfide pool (Table 1) is not revealed at all (Sorokin, 1962; Lein et al., 1990), or SRRs in these layers are 1–2 orders of magnitude lower than they are near the upper anoxic boundary (Gulin, 1991; Albert et al., 1995). In general, vertical profiles of SRRs show significant variability, possibly due to activity changes associated with seasonally and spatially variable organic detritus input. Mean sulfide production values were calculated over the total volume of the anaerobic zone (Table 2). Sulfate reduction rates given in the literature on a volumetric basis were first integrated over entire water column on a square meter basis, and then calculated over the total surface area of the Black Sea anoxic zone (306.3  109 m2). Mean sulfide production rates obtained by this approach are biased by water column depth where measurements have been obtained (Table 2, Remarks). Nevertheless, for the calculation of the sulfide production in the water column, we used all these data sets in spite of the fact that they did exhibit significant seasonal variation and different depth resolution to show the range of sulfide production rates obtained by experimental measurements. They yield the average sulfide production of 48738 (95%CL) Tg yr1. Excluding the data reported by Albert and co-authors (1995) 0.9–4.6 Tg H2S yr1, most other data fall in the range 8–200 Tg yr1. The question about seasonal variability of sulfate reduction in the water column is still unresolved. The only data acquired in wintertime, by Sorokin et al. (1992), showed

Table 2 Hydrogen sulfide production in the Black Sea water column from experimental measurements with

35

SO2 4

Season

Number of Maximum SRR Mean (range) Reference stations (nmol day1) sulfide production (Tg yr1)

Remarks

Winter Spring Spring

9 7 2

Data above 300–400 m Data below 300 m Entire water column

Summer 9

1569

61 (8–190)

Summer 3

86

34 (25–44)

Sorokin et al. (1992) Gulin (1991) Albert et al. (1995) with Jrgensen et al. (1991) Il’chenko and Sorokin (1991) Sorokin (1962)

Summer 4 Autumn 2

623 129

94 (54–135) 20 (15–24)

Pimenov et al. (2000) Lein et al. (1990)

Mean

309 117 3.5

17 (0–105) 109 (11–207) 2.7 (0.9–4.6)

48738 (95%CL)

Data above 1000 m With temperature correction factor 2.5; entire water column Data above 600 m Entire water column

2576

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

measurable SRR in the upper 200–300 m only at three of nine sampling stations. The authors attributed the absence of sulfate reduction at the other stations during this season to the disappearance of the anaerobic bacterial community in the central parts of the basin due to intense mixing at the chemocline during winter convection. A comparison of the published data and methods suggests that the variability in the SRR is not an analytical artifact. The only data that deviate significantly from all other published data are the results by Albert et al. (1995). In the absence of detailed seasonal studies of SRRs in the Black Sea water column, we can only speculate that the range of the rates observed by these authors in May 1988 represents conditions after winter mixing and diminished supply of particle-associated organic carbon to anoxic depths. These data seem to reflect the absence of a microbial biomass maximum at the interface between oxygenated and sulfide-rich waters (Bird and Karl, 1991) seen in previous years in the Black Sea (e.g., Sorokin, 1983; Zubkov et al., 1992; Rozanov et al., 1998), a characteristic feature at or near the redoxcline of other basins with anoxic water columns. Sulfate reduction in the deeper parts of the anoxic column may be particle-associated. The vertical distribution of this process is related to sinking rate and size of the particles, and can have a pronounced seasonal pattern and spatial patchiness. Seasonal and spatial variations in the particulate organic carbon flux over one order of magnitude in the Black Sea are reported (Hay et al., 1990). Concentrations and turnover times of potentially important substrates for sulfatereducers (acetate, lactate, formate) in the anaerobic layers are well above the threshold levels and additionally support particle-associated processes (Albert et al., 1995). Competition with acetoclastic methanogens or iron limitation were considered as limiting factors for the active development of sulfate reducers in anoxic layers (Albert et al., 1995). Detailed studies made by Albert and co-authors with high resolution sampling in the chemocline clearly showed that the upper 10–15 m band of high SRRs just below the interface was characterized by an order of magnitude greater activity compared with deep values (36 nmol l1 day1 and 3.5 nmol l1 day1, respectively), but when expressed on a depth-integrated basis, deep SR accounted for the 85% of total water column activity. The same conclusions can be derived from our calculations (Table 3, last column). Because of the shortcomings of SRR determinations for gross sulfide production estimates, we decided to use an alternative approach based on concentration profiles and residence time of sulfide in the anaerobic zone. This method gives an estimate of sulfide accumulation in the water column for a complete cycle of the anoxic zone, which lasts between 100 and 200 yr (Skopintsev, 1975; Ovchinnikov et al., 1993), the time of water exchange between oxic and anoxic parts. The production rate derived by this method will give net sulfide production. Bezborodov and Eremeev  (1993) applied this approach using the decrease in the SO2 4 =Cl ratios with depth and calculated sulfide production in different layers of the anoxic zone. They demonstrated that most sulfide production takes place in a layer between 500 and 2200 m depth, with an average annual production of 20–40 Tg, compared to 2–4 Tg in the upper 200–500 m. In this study sulfide production was calculated from the total inorganic carbon accumulation in the Black Sea (Neretin and Volkov, 1999), assuming near-Redfield stoichiometry (C/S=2.1, R2 ¼ 0:99) for organic carbon decomposition by sulfate reduction observed in the anoxic zone (Volkov et al., 1998). The insignificant deviation from the Redfield ratio seems to be related to the additional sulfate reduction TIC production during fermentation. The anoxic zone was divided into several layers. Accumulation of TIC during sulfate reduction in each layer (DTIC) was

Depth (m) SH2S TIC (mmol kg1) (mmol kg1)

100 150 200 250 300 500 750 1000 1250 1500 1750 2000

0 12 36 61 86 174 263 307 333 345 369 368

TIC increase DTIC (mmol kg1)

1

2

1

2

3201 3238 3339 3365 3399 3607 3781 3870 3896 3876 4004 3958

3221 3222 3232 3286 3336 3523 3736 3842 3933 3934 3968 4001

0 37 138 164 198 406 580 669 695 675 803 757

0 1 11 65 115 302 515 621 712 713 747 780

1

2

Sulfide production SH2S increase related HSPa to DTIC (taken as av. 1 bt. (1) and (2) (mmol l ) Tg yr1 %

74 171 328 581 680 749 757

5 71 232 515 688 750 780

20 60 137 267 332 364 373

Layer (m) Weight-averaged DTIC (mmol kg1)

100–200 200–300 300–500 500–1000 1000–1500 1500–2000 2000–2200 Total

0.14 0.41 1.85 8.54 9.59 8.04 1.42 30710

0.5 1.4 6.2 28 32 27 4.7 100

P HSPðTg yr1 Þ ¼ H2 S increaseðmmol l1 ÞVlayer ðkm3 Þ34 gmol1 106 =150 yr; where Vlayer (km3)Fvolumes of the layers in the anoxic zone were taken from Skopintsev (1975), and 150 yr is the average of 100–200 yr, given by Skopintsev (1975), Ovchinnikov et al. (1993). a

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

Table 3 The calculation of hydrogen sulfide production (HSP) based on total inorganic carbon (TIC) accumulation in the Black Sea water column. TIC data from Skopintsev (1975)F(1) and Dyrssen (1985)F(2) and SH2S data from Table 1 are used. The calculated SH2S increase is based on TIC/ SH2S=2.1 (Volkov et al., 1998) and 150 years as the time required to form the observed sulfide pool in each layer, the time for exchange between oxic and anoxic parts of the basin

2577

2578

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

calculated as the difference between average TIC content at 100 m (3201 and 3221 mmol kg1) and average TIC of a particular layer (Table 3). It was assumed that the observed profile of TIC and TIC pools are formed during the average time of water exchange between oxic and anoxic layers (150 yr). The weight factor for weight-averaged DTIC values was the layer width. The latter were used to calculate SH2 S increase for the TIC=SH2 S ¼ 2:1: The annual hydrogen sulfide production (HSP) was estimated to be 30710 Tg (or 3.471.1 mol m2 yr1), 8% being accumulated within the upper 500 m (Table 3). These results are similar to the data of Bezborodov and Eremeev (1993), but derived by an independent method. Summarizing data on SRRs and results of the calculation based on TIC=SH2 S ratio, the total sulfide production in the Black Sea water column could be estimated to be within the range of 30 to 50 Tg yr1. This is one order of magnitude higher than the potential flux of hydrogen sulfide from bottom sediments. Qualitatively, the predominant role of sulfide production in the water column compared to the flux from the sediments in the Black Sea was also derived by Vinogradov et al. (1962) based on the difference in the isotopic composition of sulfide in sediment pore waters and water column. The important conclusion drawn from the above discussion of SRRs and budget calculations is that most of the hydrogen sulfide is produced in the entire water column (below 500 m). Despite the fact that the highest activity is observed in the vicinity of the chemocline, on a depthintegrated basis most of the water column sulfide production (about 92%, Table 3) occurs in the middle and lower parts of the sulfidic zone.

2.5. Sulfide oxidation at the oxic/anoxic interface Measurements of sulfide oxidation rates in the Black Sea chemocline yielded integrated values between 53 and 125 Tg yr1 (Sorokin, 1972, 1983; Jorgensen et al., 1991). The range agrees with data by Sorokin et al. (1992, 1995) and Bezborodov and Eremeev (1993). These authors showed that the zone of active oxidation does not exceed 10–20 m and its lower boundary is located as a maximum at 10–15 m below the upper anoxic boundary. Most of the sulfide is oxidized at or above the boundary (Bezborodov and Eremeev, 1993; Volkov et al., 1997; Rozanov et al., 1998). However, experimental data with radio-labeled sulfide demonstrate a maximum in sulfide oxidation below the boundary, and are thus in disagreement with the vertical sulfide distribution. These seem to represent artifacts of the radiotracer method as demonstrated by JØrgensen et al.

(1991). Ayzatullin and Skopintsev (1974) used the rate constants of chemical sulfide oxidation measured in the laboratory by Cline and Richards (1969) to estimate sulfide oxidation rates in the Black Sea chemocline of 4.1 mol m2 yr1 corresponding to 42 Tg yr1. An approximation of the sulfide flux at the boundary is influenced by the turbulence and advection coefficients used and varies between 0.8 and 200 Tg yr1 (Spencer and Brewer, 1971; Bezborodov and Eremeev, 1993; Neretin, 1996). Mechanisms controlling sulfide oxidation in the Black Sea chemocline are still unresolved. Sorokin (1972, 1983) suggested that sulfide oxidation occurs in two stages: in the first step sulfide is chemically oxidized to elemental sulfur and thiosulfate, and in the second step thiosulfate oxidation to sulfate is mediated by bacteria.

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2579

However, sulfur-oxidizing bacteria isolated from the Black Sea chemocline oxidize sulfide and other reduced sulfur species, and are not limited to thiosulfate oxidation. Obligatory chemolithoautotrophic bacteria were isolated from the Black Sea chemocline that are related to the genus Thiomicrospira sp. When present in sufficiently high cell density these organisms may be able to compete with abiological sulfide oxidation (Jannasch et al., 1991). Thiomicrospira species are able to oxidize reduced sulfur species (hydrogen sulfide, thiosulfate, elemental sulfur and tetrathionate) to sulfate (e.g., Brinkhoff et al., 1999). Oxidation of reduced sulfur species to sulfate was also demonstrated with heterotrophic strains isolated from the Black Sea interface zone (Sorokin and Lysenko, 1993). Morphologically conspicuous, pelagic, large prokaryotic cells have been observed in the Black Sea chemocline, and were tentatively identified as members of the usually benthic, sulfur-oxidizing genera Thiovulum (Zubkov et al., 1992) and Achromatium (Bird and Karl, 1991); since no microphotographs have been published, definite identifications of these organisms are still missing. There is also strong evidence for anaerobic photosynthesis in the Black Sea chemocline. Increased concentrations of bacteriochlorophyll-e occur at the oxic/anoxic interface (Repeta et al., 1989), and diurnal monitoring of chlorophyll fluorescence (Karabashev, 1995) pointed to the potential contribution of this process to sulfide oxidation. Overmann et al. (1992) isolated five strains of the phototroph Chlorobium phaeobacteroides from the chemocline. All together, these data suggest that chemolithoautotrophic or heterotrophic oxidation in the Black Sea chemocline may successfully compete with abiological oxidation. However, chemical oxidation of sulfide by manganese (iron) oxides can also not be neglected (Luther III et al., 1991; Yao and Millero, 1995). In a simple vertical steady-state model of electron gradients at the oxic/anoxic interface, Murray et al. (1995) showed that electron flux of Mn (II) and Fe(II) would not produce sufficient particulate metal oxides to oxidize sulfide. The oxide deficit could be compensated only when particulate metal (Fe, Mn) oxides are produced at the boundaries of the basin and transported into the interior. Studies of the redox nepheloid layer at the oxic/anoxic interface in the coastal zone of the Black Sea (summarized by Rozanov et al., 1998) showed that most (70–95%) of the suspended matter was particulate organic matter, manganese oxyhydroxides contributing about 3%. The turbidity maximum at the interface was strongly dependent on the distance from the coast: the closer to the coast, the deeper and the larger the maximums were. It is reasonable to suspect that the average contribution of reactive manganese and iron species varies strongly with the distance from the shore. Thus, in the overall budget of sulfide the contribution of metals in its oxidation can have a significant spatial aspect. Future research on the horizontal dynamics between the coastal zone and the central parts of the basin coupled with chemical measurements could give new insights to explain the potential contribution of manganese and iron hydroxides to hydrogen sulfide oxidation at the Black Sea chemocline. 2.6. Simulation of hydrogen sulfide vertical distribution with a one-dimensional model Starting from eight equally spaced zones, the final model calculation was three zones (Fig. 3). The modeled sulfide distribution fit the data very well (R2 ¼ 0:9986). The depth-integrated value of sulfide production in zone II (512–934 m) and III (934–2200 m) is 5.28 mol m2 yr1, which is about 54.9 Tg yr1. From the upper model boundary (100 m) to the depth of 512 m, the model

2580

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

Fig. 3. Depth distribution of the hydrogen sulfide net production and consumption rates calculated by PROFILE modeling of the average H2S distribution (open circles) in the Black Sea water column (from Fig. 2A).

gives a total consumption rate of 4.97 mol m2 yr1 (51.8 Tg yr1). The net consumption rate in the upper 500 m of the anoxic water column is thus almost balanced by sulfide production below 500 m, 78% of which is produced at intermediate depths between 500 and 950 m. Quantitatively, these results are in fairly good agreement with model calculations based on TIC accumulation rates (ca. 55 Tg yr1 vs. 30710 Tg yr1). The difference is that in the TIC model most of the sulfide is produced in the layer 500–2000 m. However, both model approaches show the same tendencyFsulfide production in the Black Sea water column is not restricted to the upper part of the anoxic zone as supposed earlier (e.g., Sorokin, 1983), but proceeds continuously to the bottom. Sulfide production in the lower part of the Black Sea water column (500–2200 m) is balanced by net consumption in the upper 500 m (Fig. 3). This result principally contradicts the measurements of SRRs which are highest in the same depth range (e.g., Albert et al., 1995). The profile model and the TIC model both give net hydrogen sulfide production and consumption rates, as opposed to gross sulfate reduction rates measured with isotopically labeled sulfate, or sulfide. From a geochemical point of view, the observed distribution of solutes in the natural environment results from a tight balance between H2S net production and consumptionFa cumulative result of biogeochemical and physical processes. In our opinion, the approach based on TIC/H2S

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2581

stoichiometry in combination with measured SRRs gives a more realistic picture of hydrogen sulfide transformation than the SRRs alone, because the latter is biased by significant temporal and spatial variability. 2.7. Fe sulfide formation in the Black Sea water column Pyrite formation in the water column of the Black Sea was the subject of several studies (Tambiev and Zhabina, 1988; Muramoto et al., 1991; Volkov, 1995; Cutter and Kluckhohn, 1999). Muramoto et al. (1991) suggested that most syngenetic pyrite forms in the upper part of the anoxic zone (below 200 m), where the authors observed non-conservative (relative to salinity) sulfide distribution. This argument is supported by the similarity in the isotopic composition of pyrite sulfur in surface sediments of the central Black Sea and of hydrogen sulfide in the upper anoxic column (Calvert et al., 1996; Lyons, 1997). However, there is some evidence that the zone of pyrite formation in the water column is wider, 1000–1500 m depth (Tambiev and Zhabina, 1988; Neretin and Volkov, 1995). Relative enrichment of 32S in total sulfur in sediment traps decreasing from 32.7% at 477 m to 39.4% at 1065 m (Muramoto et al., 1991) may also suggest continuous sulfidization in the anoxic layer. These values may have been produced by continuous pyrite growth during settling through the water column, because there is a general observed tendency in hydrogen sulfide to become enriched in the lighter isotope at intermediate depths in the anoxic zone between 300(400) and 2000 m (Fry et al., 1991; Neretin et al., 1996). However, the sulfur isotope data in sediment traps are sparse, and the existing evidence needs to be confirmed by additional studies. Recent results of Cutter and Kluckhohn (1999) strongly support earlier conclusions based on moored sediment trap collections that iron sulfides are formed in the Black Sea water column. The higher concentrations of three measured iron sulfide fractions (FeS, Fe3S4, FeS2) were above 1000 m with the lower values in bottom layers. The highest concentrations of pyrite, and acid volatile sulfide fraction and greigite were observed at or below the oxic/anoxic interface, respectively. However, because no size fractionation studies (only the fraction above 0.4 mm) were performed, it is difficult to assess whether measured iron sulfide pools represent in situ inventories. In contrast to ‘‘normal’’ marine sediments deposited under oxic benthic conditions, where pyrite is formed exclusively within the sediment, in euxinic environments pyrite can also form at or below the sediment–water interface. For the following discussion about relative contribution of a syngenetic component to the total pyrite pool and sulfur budget calculations including Black Sea sediments, we need to consider the quantitative roles played by these two pyrite components in the anoxic sediments. Several authors provided evidence for the preferential contribution of the syngenetic component to the pyrite pool of laminated sediments in the central basin (Calvert et al., 1996; Canfield et al., 1996; Lyons, 1997; Wilkin et al., 1997). However, the d34S values for the pyrite of the marginal sediments suggested that iron sulfidization is occurring not only within the water column but also within the sediments at some locations. Several studies (Canfield et al., 1996; Lyons, 1997; Raiswell and Canfield, 1998) showed that the origin of sedimentary pyrite formed in anoxic basins will be strongly influenced by the relative contributions of detrital Fe phases containing reactive Fe that is scavenged in the water column. These ratios will be strongly influenced by the lateral gradient of the iron detrital flux. Particulate iron fluxes measured by

2582

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

Muramoto et al. (1991) at two stations in the southwestern Black Sea showed significant seasonal and geographical variations, and non-sulfidic iron apparently was a significant proportion of the total iron. A detailed comparison of C–S–Fe systematics of the Black Sea sediments with sediments from oxygenated basins (Rozanov et al., 1974; Volkov, 1984, 1995) and all available data on pyrite isotopic composition (Ohmoto et al., 1990; Volkov, 1995; Ohmoto and Goldhaber, 1997) (summarized in Table 4) give strong evidence for an important contribution of a diagenetic pyrite component even in the central part of the basin. The average total reduced sulfur (TRS) concentration in the Black Sea surface sediments below 100 m is 1.26% by dry weight (without Sorg : 1.15 dw%). Reported mean TRS concentrations in trap material are 0.38% (Tambiev and Zhabina, 1988), and 0.95% by dw (Muramoto et al., 1991). Data of Cutter and Kluckhohn (1999) are given on a volumetric basis for the size fraction above 0.4 mm, with a maximum FeS2 concentration of 88 nmol S l1 at the anoxic interface. The sum of iron sulfides in the Black Sea water column averaged only about 11% of the total particulate sulfur, and organic sulfur compounds were the predominant sulfur forms. These results contradict the data of Tambiev and Zhabina (1988), who observed only pyrite sulfur. The average of seasonal data obtained by Muramoto et al. (1991) gave the following composition in the traps: AVS: 3%, S0: 49%, FeS2: 48%. All this information provides strong evidence for the significant compositional variability of particulate sulfur fluxes in the water column, and consequently contribution of a syngenetic pyrite to the anoxic sediments. However, there is some information on the absolute values for the particulate sulfur flux in the Black Sea water column. Tambiev and Zhabina (1988) reported value of 39 to 50 mmol m2 yr1, corresponding to 0.4–0.5 Tg yr1. The data by Muramoto et al. (1991) are slightly lower 0.1– 0.4 Tg yr1. Since the total sulfur burial in the anoxic Black Sea sediments is about 1 Tg yr1 (Lein and Ivanov, 1983), about 10–50% of the sulfur buried may come from the syngenetic component. The isotope composition of reduced sulfur species in the anoxic sediments can also be used to calculate the average potential contribution of a syngenetic pyrite. Pyrite sulfur concentration formed during in situ diagenesis (Cdiag ) can be estimated from the mass-isotope balance equation: Cdiag ¼

Csed d34 Ssed  Ctrap d34 Strap ; d34 Sdiag

where C is concentration. The average TRS (mostly pyrite) content in anoxic sediments of the Black Sea is 1.15 dw% (Volkov, 1995). Averaging of all available literature data on pyrite isotope composition (mostly in the central part of the basin) gives a value of 30.7% (Table 4). We assume that the isotopic composition of diagenetic pyrite sulfur is close to the isotopic composition of sulfide in pore waters, 28.4% (Table 5). The average content of total reduced sulfur in trap material TRStrap is 0.67 dw% (the average between Tambiev and Zhabina (1988) and Muramoto et al. (1991) data), and d34S(TRStrap)Ed34S(FeS2syng)=37.6% (Muramoto et al., 1991). Cdiag ¼ 0:36 dw% for the average values of the input parameters in the above equation. At the same time, concentration of diagenetic pyrite can vary between 0.23 and 0.83 dw%, if the highest and the lowest values for the sulfur isotope composition of sulfide in pore water and sulfur in sediment traps are used. Thus, the mass-balance equation shows that syngenetic pyrite constitutes between 20% and 72% of total sedimentary pyrite. Additional data for S contents and isotopic composition of sediment trap material, and complementary data on the isotopic

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2583

Table 4 Pyrite sulfur isotopic composition vs. CDT in the Black Sea surface sediments (after Volkov (1995) with supplements) d34S–FeS2 (%) 30.8 29.3 27.2 32.7 26.9 24.9 34.0 33.3

(33.7 (29.4 (27.4 (40.0 (36.3 (29.7

to to to to to to

Number of stations 26.3) 29.2) 27.0) 30.6) 17.5) 19.8)

6 2 2 8 2 5 1 3

(37.2 to 27.8)

Weighted average: 30.7%

Reference Vinogradov et al. (1962) Migdisov et al. (1974) Sweeney and Kaplan (1980) Vainshteyn et al. (1986) Strizhov et al. (1989) Volkov, unpubl. data (1993) Calvert et al. (1996) Lyons (1997)

29

Table 5 The isotopic composition of sulfide in sediment pore water of anoxic Black Sea. All data vs. VCDT except for Vinogradov et al. (1962) data, which are vs. CDT Station

Station depth (m)

Sediment depth (cm)

d34S–H2S (%)

4740 4751 6

2008 2216 394

7

1204

0–10 0–10 2–4 35–52 2–10 10–18 10–18

22.5 Vinogradov et al. (1962) 16.2 Vinogradov et al. (1962) 38.7 This paper 34.4 36.1 This paper 35.5 40.4 This paper 28.4710.6 (95%CL)% (38.7 to 16.2) n ¼ 4

8 2045 Mean for the uppermost 10 cm

Reference

compositions of hydrogen sulfide in pore waters, are necessary to refine our estimates of the average contribution of water column formed pyrite to total sedimentary pyrite. However, the preliminary data on the sulfur fluxes and our mass-balance calculation indicate that water column pyritization does not necessarily account for most of the sedimentary pyrite in the Black Sea anoxic sediments, as postulated earlier (e.g., Lyons, 1997). We speculate that the contribution of syngenetic pyrite depends strongly on seasonal sedimentation patterns (e.g., Muramoto et al., 1991), which vary from site to site. Lein and Ivanov (1983) estimated total sulfide burial in the Black Sea to be 2.4 Tg yr1, including about 1 Tg yr1 that is buried in the anoxic zone. Reduced sulfur accumulation in the Black Sea sediments comprises two components. The maximum estimate for the hydrogen sulfide flux from the sediments into the water column can be made assuming that most of the buried sulfur (80% from the mass-isotope balance equation 0.8 Tg) comes from the water column. To calculate the averaged flux of sulfide at the sediment/water interface, the diffusion path length of sulfide for one year is needed. The diffusion path length for hydrogen sulfide L can be calculated from the equation L ¼ ð2Ds tÞ1=2 ; where Ds is the sediment diffusion coefficient for hydrogen sulfide, and t is time. Ds (HS) for the upper Black Sea sediments can be calculated from porosity

2584

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

j (0.8–0.9) (our data), temperature and salinity data (81C, 22%), Ds (01C, 0%)=9.75  106 cm2 s1 (Li and Gregory, 1974), and Ds ¼ Ds (81C, 22%)/(1ln(j)2) derived by Boudreau (1997). For the time of one year, the diffusion path length will not exceed 23 cm. For the upper 20 cm an average annual SRR in Black Sea sediments falls in the range 3.2–5.2 Tg (Lein et al., 1990; Albert et al., 1995). The total amount of sulfide produced in the sediment would constitute an annual sulfide flux of 3–5 Tg into the water column. However, it is likely to be less, because the contribution of diagenetic pyrite to the total pyrite pool of Black Sea anoxic sediments can vary from site to site (see discussion above), and sulfide diffusion downward in the deeper sediment layers was neglected in the calculation. In the preceding discussion we did not consider the specific situations that are observed on the steep continental slopes of the Black Sea bordering the Caucasus and Anatolia. At some sites in these areas limnic Late Pleistocene clays are characterized by high concentrations of reactive iron and low concentrations of dissolved sulfide in pore waters below 30 mM (Volkov et al., 1971). These sediments may serve as an additional sink for hydrogen sulfide formed in the deep sea. The spatial distribution of these areas and their influence on the hydrogen sulfide budget need to be addressed in future research.

2.8. Sulfide removal by the Lower Bosphorus Current The evolution of hydrogen sulfide in the Black Sea water column is strongly regulated by physical processes on seasonal (e.g., Neretin et al., 1997) and also on geological (e.g., Deuser, 1974) time scales. Changes in the Mediterranean inflow and outflow through the Bosphorus Strait . zsoy and U . nlu. ata, play a crucial role in the salinity of the Black Sea and the water budget (O 1997), but may also significantly influence the chemical composition of the water column. We attempt to estimate the potential contribution of sulfide oxidation by Bosphorus water for the overall budget of hydrogen sulfide, and the relative importance of this process in comparison with H2S removal rates in the chemocline. The annual Bosphorus flux into the Black Sea is . nlu. ata et al. (1990). estimated to be 120–312 km3, the highest value being the latest estimate by U The Bosphorus Current operates in two-directional transient fluxes, occasionally changing direction even within one day (Latif et al., 1991). Temporary blockage of the Mediterranean flow into the Black Sea can occur in spring and summer during periods of high freshwater discharge. Otherwise, the blockage of the Black Sea outflow can be observed during southwesterly winds in . zsoy et al., 1993). winter (O The Mediterranean effluent follows along the canyon and the shelf, producing a delta-like structure in the shelf region. The vertical spread in the interface region between shelf and continental slope is the result of mixing of Bosphorus waters with CIL (e.g., Di Iorio and Yu. ce, 1999). Upon passage over the shelf Mediterranean Water (MW) is altered to ‘‘shelf-modified . zsoy et al., 1993) obtaining a signature distinctly MW’’ (SMMW) (terminology suggested by O different from the initially warm, saline characteristics of MW. Cascades of dense SMMW sink down the continental slope to 500–600 m, where they can still be distinguished from the ambient . zsoy et al., 1993). To our knowledge, there is no quantitative estimate waters by a cold anomaly (O of SMMW intrusions to the anoxic part of the basin in the literature.

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2585

Considering the significant variability in the magnitude and direction of water exchange through the Bosphorus, and the absence of quantitative information on the volumes of SMMW entrainment during the cascade-like intrusions to the anoxic zone, estimates of sulfide removal rates by the Lower Bosphorus Current are only approximate at the moment. At the same time, these values give the approximate magnitude, and in our opinion can be used for the overall sulfide budget. We assume that SMMW intrusion into the anoxic zone is more likely when the ratio between CIL and MW is low, which will mean higher density of the resulting flux and deeper sinking. Reported ratios for SMMW : MW in the literature are between 3 : 1 and 6 : 1 (Murray et al., . zsoy et al., 1993), which give total fluxes of 936 km3 yr1 and 1872 km3 yr1, respectively. 1991; O The oxygen content in the core of the CIL is about 300750 mmol l1 (Neretin, unpublished data). Skopintsev (1975) reports an average concentration of 200 mmol l1 for the Mediterranean water. Thus, the resulting oxygen concentrations of modified Bosphorus waters (SMMW) are 267 mmol l1 and 283 mmol l1 for the low and high entrainment ratios, respectively. The corresponding oxygen fluxes are 2.5  1011 mol O2 yr1 (SMMW : MW=3 : 1) and 5.3  1011 mol O2 yr1 (SMMW : MW=6 : 1). Our budget calculations are influenced by two possibilities: (1) SMMW intrudes into the anoxic zone, which will be the case for high entrainment ratio, or (2) the SMMW spreads above the oxic/ anoxic boundary in the case of low entrainment ratios. The CIL of the Black Sea on the . zsoy et al., 1993), which is far above the upper continental shelf occurs between 50 and 75 m (O anoxic boundary at 120–160 m at the periphery of the basin (Neretin et al., 1997). Temporal and spatial variability in the SMMW flow and its relation to atmospheric and sea level changes in the Bosphorus are well documented (e.g., Di Iorio and Yu. ce, 1999). These data provide evidence for significant annual changes in the transformed Bosphorus waters as well as sinking depths even at a given SMMW : MW ratio. If sulfide is oxidized according to the following equation: þ H2 S þ 2O2 ¼ SO2 4 þ2H ;

then the sulfide consumption for the entrainment ratios of 3 : 1 and 6 : 1 should be about 1.3 and 2.7  1011 mol yr1, corresponding to 4.4 and 9.2 Tg yr1, respectively. This estimate is roughly 10–20% of the average sulfide oxidation at the oxic/anoxic interface. As discussed earlier, the lower estimate is a more realistic value as it is based on the lower entrainment ratio. These results have important implications for assessing the role of the Lower Bosphorus Current in ventilating the Black Sea deep layers. As shown earlier in the paper in the model simulating vertical distribution of sulfide in the Black Sea, net consumption rate of sulfide in the upper anoxic layers (ca. 100–500 m) of about 52 Tg yr1 cannot be explained by SMMW intrusions alone, and additional mechanisms need to be considered. For a long time the Mediterranean influx from the Bosphorus was considered the only source of ventilation and renewal of anoxic waters in the deep Black Sea (e.g., Skopintsev, 1975), but investigations by Russian oceanographers in the Black Sea during the last decade (Ovchinnikov et al., 1993) put forth new hypotheses that are based on the predominant role of the lateral exchange between coastal and interior waters. According to the data presented by Ovchinnikov and his co-authors, ventilation of the anoxic waters was primarily associated with the existence of a Nearshore Convergence Zone (NCZ) along the whole periphery of the basin. In some deep

2586

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

anticyclonic gyres associated with the NCZ observed in the western Black Sea (Andrianova and Ovchinnikov, 1991), cold intermediate layer waters were observed down to 1000–1200 m. Downwelling of surface waters (mostly oxygen enriched Cold Intermediate Layer (CIL) waters) can provide an additional mechanism for the ventilation of the anoxic zone. Winter mixing processes in the central parts of the main Black Sea gyres can also play an important role in the ventilation of the anoxic water column. This mechanism has been suggested to induce the formation of CIL in the Black Sea (Ovchinnikov and Popov, 1987). During severe winter conditions main pycnocline waters were observed close to the surface at 20–35 m depth in the photic zone. If the upper boundary of hydrogen sulfide is also at the density surface of 16.2 in winter (Neretin, 1996), that would mean that during strong winters hydrogen sulfide-enriched waters will be in direct contact with well oxygenated waters of CIL. These conditions have an inter-annual cycle of 15–20 years (Ovchinnikov et al., 1993). On a longer-time scale, however, winter mixing processes may be responsible for most of the ventilation of anoxic waters and oxidation of hydrogen sulfide mediated chemically.

3. Conclusions The most important reactions within the sulfur cycle of the Black Sea are sulfide formation from bacterial reduction in the water column and hydrogen sulfide oxidation at the oxic/anoxic interface. The annual fluxes are estimated to be in the range 20–50 Tg. The maximum contribution of modified Bosphorus waters carrying dissolved oxygen into sulfide zone does not exceed 10– 20% of the total sulfide production (4.4–9.2 Tg yr1); significantly less is consumed by iron sulfidization in the water column (up to 0.7 Tg yr1). The most reasonable average estimate for sulfide flux at the sediment/water interface is about 3–5 Tg yr1. A general scheme of the processes within the sulfur cycle with their assigned annual fluxes is presented in Fig. 4. The estimated annual accumulation of total reduced sulfur, derived mostly from dissimilatory sulfate reduction, in ocean sediments is in the range 97–118 Tg (Volkov and Rozanov, 1983; Volkov, 1984), and additionally less than 18% of this amount (ca. 17–21 Tg yr1) participates in the internal sulfate sulfur turnover. The latter estimate is rather approximate, because it does not take into account re-oxidation processes at the sediment/water interface. We can assume that the total sulfur budget for the world ocean sediments is about 120–140 Tg yr1. The Black Sea sulfide production itself contributes about one third of the global sedimentary sulfur budget. These figures are rather flexible, but they clearly show the significance of the Black Sea for the global sulfur cycle. The balance between sulfide production in the water column and oxidation at the chemocline has an important feedback for the whole Black Sea ecosystem. The ratio between these two processes can vary seasonally. Increased sulfide oxidation at the chemocline during winter may be balanced by accelerated sulfate reduction during summer. These processes are independent of the Bosphorus influence. The model presented in this paper approximates multi-annual conditions. Taking into account the seasonality of different biogeochemical (e.g., variations in primary productivity) and physical (e.g., pycnocline erosion during winter mixing leading to the dispersion of the bacterial community at the interface) processes in the Black Sea, deviations from the presented figures for the fluxes and pathways of the sulfur cycle are to be expected. The main

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2587

Fig. 4. Sulfur budget for the Black Sea anoxic zone. The width of arrows and dimension of the boxes represent the relative magnitudes of respective processes. SMMW stands for ‘‘Shelf Modified Mediterranean Water’’. Process rates are in Tg yr1.

purpose of this study was to show the importance and relative magnitudes of different processes in the anoxic Black Sea. Further investigations need to address seasonal effects to provide a better understanding of the whole ecosystem. Our budget calculations give some clues for the interpretation of the present state of the Black Sea ecosystem. Temporal variations in the average depth of the chemocline in the Black Sea are mainly the result of climatic changes in the density structure of the water column. The upper anoxic boundary location vs. density for this basin did not change over the period from 1910 to 1995 (Bezborodov and Eremeev, 1993; Konovalov et al., 1999). But at the same time, the latter report showed a prominent increase in sulfide concentrations, as well as nutrient levels, within the anoxic zone (at 1000–2000 m) due to anthropogenic impact. Does the trend of increasing sulfide concentrations for the lower anoxic layers over the last 30 years observed by Konovalov et al. (1999) represent the general situation observed in the Black Sea? Are there definite unidirectional changes in the ecosystem? If the average sulfide production in the Black Sea is 30–50 Tg yr1 and represents an average annual figure and if the total sulfide inventory is about 4.6  103 Tg, the residence time of hydrogen sulfide in the water column is

2588

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

about 90–150 yr. This value is comparable to the time for water exchange between oxic and anoxic layers (e.g., Skopintsev, 1975). Unidirectional changes in the Black Sea anoxic zone suggested from the past 30 years by comprehensive Black Sea surveys may only reflect interannual climatic variability or sample heterogeneity. This work puts forth the importance of ventilation processes in the Black Sea anoxic zone. The Bosphorus flux cannot be considered a main factor for deep basin ventilation as suggested by the sulfide budget. More attention in future studies should be paid to near-shore meso-scale dynamics and their influence on chemocline processes and lateral exchange between near-shore and open waters (e.g., Kempe et al., 1991; Rozanov et al., 1998). We would like to stress the importance of hydrodynamics for the specific biogeochemical pathways within the sulfur cycle in the Black Sea. Severe winter conditions accompanied by strong horizontal and vertical mixing can initiate pronounced erosion of the pycnocline, which with the coastal processes occurring in wintertime, may determine the dynamics of the hydrogen sulfide zone (e.g., Ovchinnikov et al., 1993). Field data for the winter season are critical for progress in our understanding of the Black Sea ecosystem. Not considered in this paper in detail, but obviously among the most important questions for the understanding of the Black Sea sulfur cycle, are the processes at the oxic/anoxic interface (redox nepheloid layer). Further studies need to be done to investigate the role of oxygen and manganese for hydrogen sulfide oxidation, as well as the structure and diversity of the bacterial community responsible for the oxidation-reduction reactions. Direct SRR measurements (e.g., Albert et al., 1995), as well as the sulfur isotope values for sulfate and sulfide in the vicinity of chemocline (e.g, Fry et al., 1991; Neretin, 1996), suggest active sulfate reducing bacteria in the layers underneath the upper anoxic boundary. At the same time, these bacteria are also spatially located close to an abundant chemosynthetic community responsible for sulfide oxidation (e.g., Teske et al., 1996; Zopfi et al., 2001). Research on the Black Sea chemocline may thus provide information about functional relationships between sulfate-reducing bacteria and chemolithotrophic bacteria and possible symbiotic relationships.

Acknowledgements Special thanks to T.P. Demidova and N.N. Zhabina (P.P. Shirshov Institute of Oceanology, Moscow) for invaluable analytical help during cruises. Collaboration with the captains and the crews of RV Akvanavt, Yantar and Petr Kotzov is gratefully acknowledged. Volker Bru. chert (MPI for Marine Microbiology, Bremen) is specially acknowledged for critical reading and many efforts to improve the manuscript. We are grateful to Michael P. Bacon for editorial assistance. The authors wish to thank Prof. Bo Barker Jorgensen (MPI for Marine Microbiology, Bremen) for stimulating discussions. L.N.N. thanks Dr. Ivan M. Ovchinnikov (Southern Branch of P.P. Shirshov Institute of Oceanology, Gelendzhik, Russia) for fruitful discussions of the hydrophysical background for this work. M.E.B. wishes to thank Prof. Ju. rgen Rullko. tter from the Institute of Chemistry and Biology of the Marine Environment (University of Oldenburg, Germany) for permission to use the mass spectrometer. The authors gratefully acknowledge three anonymous reviewers for their very constructive comments and suggestions. This work was financially supported by the Russian Fund for Basic Research (grants 96-05-66169 and 96-15-

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2589

98440), EU INCO-Copernicus Program (project ERBIC15CT960113) and by the Max-PlanckSociety.

References Albert, D.B., Taylor, G., Martens, C., 1995. Sulfate reduction rates and low molecular weight fatty acid concentrations in the water column and surficial sediments of the Black Sea. Deep-Sea Research 42, 1239–1260. Andrianova, O.P., Ovchinnikov, I.M., 1991. Transformation of fresh waters in the western Black Sea. Meteorologiya i Gidrologiya 7, 74–79 (in Russian). Arthur, M.A., Dean, W.E., 1998. Organic matter production and preservation and evolution of anoxia in the Holocene Black Sea. Paleoceanography 13, 395–411. Ayzatullin, T.A., Skopintsev, B.A., 1974. Investigation of hydrogen sulfide oxidation in the Black Sea water. Okeanologiya 14, 403–420. Berg, P., Risgaard-Petersen, N., Rysgaard, S., 1998. Interpretation of measured concentration profiles in sediment pore water. Limnology & Oceanography 43, 1500–1510. Bezborodov, A.A., Eremeev, V.N., 1993. Chernoe more. Zona vzaimodeystviya aerobnykh i anaerobnykh vod (The Black Sea: oxic/anoxic interface zone). Marine Hydrophysical Institute AS of the Ukraine, Sevastopol, Ukraine, 299pp (in Russian). Bird, D.F., Karl, D.M., 1991. Microbial biomass and population diversity in the upper water column of the Black Sea. Deep-Sea Research I 38 (2A), S1069–S1082. Boguslavsky, S.G., Ivaschenko, I.K., 1990. Vertical mixing in the Black Sea. Morskoy Gidrophizicheskiy Zhurnal 6, 34–41. Boguslavsky, S.G., Kotovschikov, B.B., 1984. Formation of the contemporary salinity field in the Black Sea. Okeanologiya 24, 410–416. Bo. ttcher, M.E., Rusch, A., Ho. pner, T., Brumsack, H.-J., 1997. Stable sulfur isotope effects related to local intense sulfate reduction in a tidal sandflat (southern North Sea): results from a loading experiment. Isotopes Environmental Health Studies 33, 109–129. Bo. ttcher, M.E., Brumsack, H.-J., de Lange, G.J., 1998. Sulfate reduction and related stable isotope (34S, 18O) variations in interstitial waters of the Eastern Mediterranean. Proceedings ODP Scientific Results 160, 365–373. Boudreau, B.P., 1997. Diagenetic Models and their Implementation: Modeling Transport and Reactions in Aquatic Sediments, Springer, Berlin, 414pp. Brinkhoff, Th., Muyzer, G., Wirsen, C.O., Kuever, J., 1999. Thiomicrospira kuenenii sp. nov. and Thiomicrospira frisia sp. nov., two mesophilic obligately chemolithoautotrophic sulfur-oxidizing bacteria isolated from an intertidal mud flat. International Journal of Systematic Bacteriology 49, 385–392. Calvert, S.E., Vogel, J.S., Southon, J.R., 1987. Carbon accumulation rates and the origin of the Holocene sapropel in the Black Sea. Geology 151, 918–921. Calvert, S.E., Thode, H.G., Yeung, D., Karlin, R.E., 1996. A stable isotope study of pyrite formation in the Late Pleistocene and Holocene sediments of the Black Sea. Geochemica et Cosmochimica Acta 60, 1261–1270. Canfield, D.E., Lyons, T., Raiswell, R., 1996. A model for iron deposition to euxinic Black Sea sediments. American Journal of Science 296, 818–834. Cline, J.D., 1969. Spectrophotometric determination of hydrogen sulfide in natural waters. Limnology & Oceanography 14, 454–458. Cline, J.D., Richards, F.A., 1969. Oxygenation of hydrogen sulphide in sea water at constant salinity, temperature and pH. Environmental Science and Technology 3, 838–843. Codispoti, L.A., Friederich, G.E., Murray, J.W., Sakamoto, C.M., 1991. Chemical variability in the Black Sea: implications of continuous vertical profiles that penetrated the oxic–anoxic interface. Deep-Sea Research II 38 (2A), S691–S710. Coplen, T.B., Krouse, H.R., 1998. Sulphur isotope data consistency improved. Nature 392, 32.

2590

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

Cutter, G.A., Kluckhohn, R.S., 1999. The cycling of particulate carbon, nitrogen, sulfur, and sulfur species (iron monosulfide, greigite, pyrite, and organic sulfur) in the water columns of Framvaren Fjord and the Black Sea. Marine Chemistry 67, 149–160. Deuser, W.G., 1971. Organic carbon budget of the Black Sea. Deep-Sea Research 18, 995–1004. Deuser, W.G., 1974. Evolution of anoxic conditions in Black Sea during Holocene. In: Degens, E.T., Ross, R.A. (Eds.), The Black SeaFGeology, Chemistry and Biology, Memoirs, American Association of Petroleum Geologists, 20, Tulsa, Oklahoma, USA, pp. 133–136. Di Iorio, D., Yu. ce, H., 1999. Observations of Mediterranean flow into the Black Sea. Journal of Geophysical Research 104 (C2), 3091–3108. Dyrssen, D., 1985. Some calculations on the Black Sea chemical data. Chemica Scripta 25, 199–205. Fry, B., Jannasch, H.W., Molyneaux, S.J., Wirsen, C., Muramoto, J., King, S., 1991. Stable isotope studies of the carbon, nitrogen and sulfur cycles in the Black Sea and the Cariaco Trench. Deep-Sea Research 38 (2A), S1003– S1019. Glenn, C.R., Arthur, M.A., 1985. Sedimentary and geochemical indicators of productivity and oxygen contents in modern and ancient basins: The Holocene Black Sea as the ‘‘type’’ anoxic basin. Chemical Geology 48, 325–354. Gulin, M.B., 1991. Izuchenie bakterialnikh processov sulfatredukcii i khemosinteza v vodnoy srede Chernogo morya (The investigation of bacteria mediated sulfate reduction and chemosynthesis in the Black Sea water column), Ph.D. Thesis abstract, Institute of biology of the southern seas, Sevastopol, Ukraine (in Russian). Hay, B.J., Honjo, S., Kempe, S., Ittekkot, V.A., Degens, E.T., Konuk, T., Izdar, E., 1990. Interannual variability in particle flux in the southwestern Black Sea. Deep-Sea Research I 37, 911–928. Hu. ttel, M., Ziebis, W., Forster, S., Luther III, G.W.I., 1998. Advective transport affecting metal and nutrient distributions and interfacial fluxes in permeable sediments. Geochimica et Cosmochimica Acta 62, 613–631. Il’chenko, S.V., Sorokin, Yu.I., 1991. K otsenke intensivnosti obrazovaniya serovodoroda v Chernom more (To the estimate of hydrogen sulfide production in the Black Sea). In: Vinogradov, M.E. (Ed.), Izmenchivost’ ekosistemy Chernogo morya: estestvenniye i anropogenniye faktory (The Black Sea ecosystem variability: natural and anthropogenic factors). Nauka, Moscow, pp. 73–77 (in Russian). Jannasch, H.W., Wirsen, C.O., Molyneaux, S.J., 1991. Chemoautotrophic sulfur-oxidizing bacteria from the Black Sea. Deep-Sea Research 38 (2A), S1105–S1120. Jones, G., Gagnon, A., 1994. AMS radiocarbon dating of sediments in the Black Sea. Deep-Sea Research I 41, 531–557. Jrgensen, B.B., Fossing, H., Wirsen, C.O., Jannasch, H.W., 1991. Sulfide oxidation in the anoxic Black Sea chemocline. Deep-Sea Research 38 (2A), S1083–S1103. Karabashev, G.S., 1995. Diurnal rhythm and vertical distribution of chlorophyll fluorescence as evidence of bacterial photosynthetic activity in the Black Sea. Ophelia 40, 229–238. Kempe, S., Diercks, A.R., Liebezeit, G., Prange, A., 1991. Geochemical and structural aspects of the pycnocline in the Black Sea (R/V Knorr 134-8, Leg 1, 1988). In: Izdar, E., Murray, J.M. (Eds.), Black Sea Oceanography, NATO ASI Series C, Vol. 351. Kluwer Academic Publishers, Dordrecht, pp. 89–110. Konovalov, S.K., Eremeev, V.N., Suvorov, A.M., Khaliulin, A.Kh., Godin, E.A., 1999. Climatic and anthropogenic variations in the sulfide distribution in the Black Sea. Aquatic Geochemistry 5, 13–27. Lane-Serff, G.F., Rohling, E.J., Bryden, H.L., Charnock, H., 1997. Postglacial connection of the Black Sea to the Mediterranean and its relation to the timing of sapropel formation. Paleoceanography 12, 169–174. . zsoy, E., Oguz, T., U . nlu. ata, U., 1991. Observations of the Mediterranean inflow into the Black Sea. Latif, M.A., O Deep-Sea Research II 38 (2A), S677–S723. Lein, A.Yu., Ivanov, M.V., 1983. Reduced sulphur accumulation in sediments of marine basins with high rates of sulphate reduction. In: The Global Biogeochemical Sulphur Cycle, SCOPE 19, Wiley & Sons, Chichester, UK, pp. 413–420. Lein, A.Yu., Ivanov, M.V., 1991. On the sulfur and carbon balances in the Black Sea. In: Izdar, E., Murray, J.W. (Eds.), The Black Sea Oceanography. Kluwer Academic Publishers, Amsterdam, pp. 307–318. Lein, A.Yu., Ivanov, M.V., Vaynshtein, M.B., 1990. Hydrogen sulfide balance in the deep-water zone of the Black Sea. Microbiologiya 59, 656–664. Li, Y.-H., Gregory, S., 1974. Diffusion of ions in seawater and in deep-sea sediments. Geochimica et Cosmochimica Acta 38, 703–714.

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2591

Luther III, G.W., Church, T.M., Powell, D., 1991. Sulfur speciation and sulfide oxidation in the water column of the Black Sea. Deep-Sea Research 38 (2A), S1121–S1138. Lyons, T., 1997. Sulfur isotopic trends and pathways of iron sulfide formation in upper Holocene sediments of the anoxic Black Sea. Geochemica et Cosmochimica Acta 61, 3367–3382. Migdisov, A.A., Cherkovskiy, S.L., Grinenko, V.A., 1974. The correlation between sulfur isotopic composition and formation conditions in humic sediments. Geokhimiya 10, 1482–1502. Muramoto, J., Honjo, S., Fry, B., Hay, B.J., Howarth, R.W., Cisne, J.L., 1991. Sulfur, iron and organic carbon fluxes in the Black Sea: sulfur isotopic evidence for origin of sulfur fluxes. Deep-Sea Research 38 (2A), S1151–S1189. . zsoy, E., 1991. Hydrographic properties and ventilation of the Black Sea. Deep-Sea Research Murray, J.W., Top, Z., O II 38 (2A), S663–S689. Murray, J.W., Codispoti, L.A., Friederich, G.E., 1995. Oxidation-reduction environments: the suboxic zone in the Black Sea. In: Huang, C., et al. (Ed.), Aquatic Chemistry. Kluwer Academic Publishers, Amsterdam, pp. 157–176. Neretin, L.N., 1996. Sovremennoe sostoyanie serovodorodnoy zony Chernogo morya (Contemporary state of the hydrogen sulfide zone in the Black Sea), Ph.D. Thesis abstract, P.P. Shirshov Institute of Oceanology RAS, Moscow, 26pp (in Russian). Neretin, L.N., Volkov, I.I., 1995. On the vertical distribution of hydrogen sulfide in deep waters of the Black Sea. Okeanologiya 35, 60–65. Neretin, L.N., Volkov, I.I., 1999. The calculation of hydrogen sulfide production in the Black Sea based on total inorganic carbon content. Doklady Earth Sciences 365A (3), 398–401. Neretin, L.N., Grinenko, V.A., Volkov, I.I., 1996. Isotopic composition of hydrogen sulfide in deep water of the Black Sea. Transactions (Doklady) of the Russian Academy of Sciences 349A, 1015–1019. Neretin, L.N., Demidova, T.P., Volkov, I.I., 1997. Some aspects of hydrogen sulfide spatial variability near the NorthCaucasus coast of the Black Sea. Okeanologiya 37, 365–372. Ohmoto, H., Goldhaber, M.B., 1997. Sulfur and carbon isotopes. In: Barnes, H.L. (Ed.), Geochemistry of Hydrothermal Ore Deposits. Wiley, New York, pp. 517–611 (Chapter 11). Ohmoto, H., Kaiser, C.J., Geer, K.A., 1990. Systematics of sulphur isotopes in recent marine sediments and ancient sediment-hosted basemetal deposits. In: Herbert, H.K., Ho, S.E. (Eds.), Stable Isotopes and Fluid Processes in Mineralization. Perth: University of Western Australia Publication, pp. 70–120 (Vol. 23). Ovchinnikov, I.M., Popov, Yu.I., 1987. Formation of the Black Sea cold intermediate layer. Okeanologiya 27, 739–747. Ovchinnikov, I.M., Popov, Yu.I., Titov, V.B., 1993. Gidrofizicheskaya koncepciya na razvitie ekologicheskoy situatsii v Chernom more (Hydrophysical conception on ecological situation development in the Black Sea). Deposited in VINITI, No.278-B93, Moscow, 35pp (in Russian). Overmann, J., Cypionka, H., Pfennig, N., 1992. An extremely low-light-adapted phototrophic sulfur bacterium from the Black Sea. Limnology & Oceanography 37, 150–155. . zsoy, E., U . nlu. ata, U., Top, Z., 1993. The Mediterranean water evolution, material transport by double diffusion O intrusions, and interior mixing in the Black Sea. Progress in Oceanography 31, 275–320. . zsoy, E., U . nlu. ata, U., 1997. Oceanography of the Black Sea: a review of some recent results. Earth Science Reviews O 42, 231–272. Pimenov, N.V., Rusanov, I.I., Yusupov, S.K., Friedrich, J., Lein, A.Yu., Wehrli, B., Ivanov, M.V., 2000. Microbial processes at the oxic-anoxic interface in the deep-water zone of the Black Sea. Mikrobiologiya 69, 527–540. Raiswell, R., Canfield, D.E., 1998. Sources of iron for pyrite formation in marine sediments. American Journal of Science 298, 219–245. Repeta, D.J., Simpson, D.J., Jrgensen, B.B., Jannasch, H.W., 1989. Evidence for anoxygenic photosynthesis from the distribution of bacterio-chlorophylls in the Black Sea. Nature 342, 69–72. Rozanov, A.G., Volkov, I.I., Yagodinskaya, T.A., 1974. Forms of iron in surface layer of Black Sea sediments. In: Degens, E.T., Ross, D.A. (Eds.), The Black SeaFGeology, Chemistry, and Biology. American Association of Petroleum Geologists, Tulsa, Oklahoma, pp. 532–541. Rozanov, A.G., Neretin, L.N., Volkov, I.I., 1998. Redox nepheloid layer (RNL) of the Black Sea: its location, composition and origin. In: Ivanov, L.I., Oguz, T. (Eds.), Ecosystem Modeling as a Management Tool for the Black Sea, Vol. 1. Kluwer Academic Publishers, Amsterdam, pp. 77–91.

2592

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

Skopintsev, B.A., 1975. Formirovanie sovremennogo khimicheskogo sostava vad Chernogo morya (Formation of contemporary chemical composition of the Black Sea waters), Hidrometeoizdat, Leningrad, 336pp (in Russian). Sorokin, Yu.I., 1962. Experimental investigation of bacterial sulfate reduction in the Black Sea using 35S. Microbiologiya 31, 329–335. Sorokin, Yu.I., 1972. The bacterial population and the processes of hydrogen sulfide oxidation in the Black Sea. Journal du Conseil International pour l’Exploration de la Mer 34, 423–454. Sorokin, Yu.I., 1983. The Black Sea. In: Ketchum, B.H. (Ed.), Ecosystems of the World. Estuaries and Enclosed Seas, Vol. 26. Elsevier, Amsterdam, pp. 253–292. Sorokin, Yu.I., Sorokin, P.Yu., Sorokina, O.V., Sorokin, D.Yu., Sukhomlin, A.V., 1992. Raspredelenie i funktsionalnaya aktivnost’ mikroflory v tolsche vody Chernogo morya zimoy i v nachale vesny 1991g. (The distribution and functional activity of bacteria in the Black Sea water column during winter-beginning spring 1991). In: Vinogradov, M.E. (Ed.), Zimnee sostoyanie ekosistemy otkrytoy chasti Chernogo morya (The Open Black Sea Ecosystem During Wintertime). P.P. Shirshov Institute of Oceanology, Moscow, pp. 89–102 (in Russian). Sorokin, D.Yu., Lysenko, A.M., 1993. Heterotrophic bacteria from the Black Sea oxidizing reduced sulfur compounds to sulfate. Microbiologiya 62, 594–602. Sorokin, Yu.I., Sorokin, P.Yu., Avdeev, V.A., Sorokin, D.Yu., Ilchenko, S.V., 1995. Biomass, production and activity of bacteria in the Black Sea, with special reference to chemosynthesis and the sulfur cycle. Hydrobiologia 308, 61–76. Spencer, D.W., Brewer, P.G., 1971. Vertical advection diffusion and redox potentials as controls on the distribution of manganese and other trace metals dissolved in waters of the Black Sea. Journal of Geophysical Research 76, 5877– 5892. Standard methods for the examination of water and wastewater, 1992. American Public Health Organization. 18th edition, Washington, DC. Strizhov, V.P., Nikolaev, S.D., Gurina, N.V., 1989. The isotopic composition of sulfur, oxygen and carbon in the Black Sea sediments and possible reasons of its alteration in the Late Quaternary time. Okeanologiya 29, 979–986. Sweeney, R.E., Kaplan, I.R., 1980. Stable isotope composition of dissolved sulfate and hydrogen sulfide in the Black Sea. Marine Chemistry 9, 145–152. Tambiev, S.B., Zhabina, N.N., 1988. Pyritization in the Black Sea anoxic water: its scale and influence on recent sediments. Doklady Akademii Nauk SSSR 299, 1216–1221. Teske, A., Wawer, C., Muyzer, G., Ramsing, N.B., 1996. Distribution of sulfate-reducing bacteria in a stratified fjord (Mariager Fjord, Denmark) as evaluated by most-probable-number counts and denaturing gradient gel electrophoresis of PCR-amplified ribosomal DNA fragments. Applied and Environmental Microbiology 62, 1405–1415. . zsoy, E., 1990. On the physical oceanography of the Turkish straits. In: Pratt, . nlu. ata, U., Oguz, T., Latif, M.A., O U L.G. (Ed.), The Physical Oceanography of Sea Straits. Kluwer Academic Publishers, Amsterdam, pp. 25–60. Vainshteyn, M.B., Tokarev, V.G., Shakola, V.A., Lein, A.Yu., Ivanov, M.V., 1986. The geochemical activity of sulfatereducing bacteria in sediments in the western part of the Black Sea. Geochemistry International 1, 110–122. Vinogradov, A.P., Grinenko, V.A., Ustinov, V.I., 1962. Isotopic compositions of sulfur compounds in the Black Sea. Geochemistry International 10, 973–997. Vinogradov, M.E., Nalbandov, Yu.R., 1990. Effect of changes in water density on the profiles of phisiochemical and biological characteristics in the pelagic ecosystem of the Black Sea. Okeanologiya 30, 567–573. Volkov, I.I., 1984. Sulfur Geochemistry in Ocean Sediments, Nauka, Moscow, 272pp (in Russian). Volkov, I.I., 1993. Unpublished data. Volkov, I.I., 1995. Serovodorod i vosstanovlennye soedineniya sery v Chernom more: sravnitelniy analiz (Hydrogen sulfide and reduced sulfur species in the Black Sea: comparative analysis). In: Bordovskiy, O.K. (Ed.), Khimiya morey i okeanov (The Chemistry of Seas and Oceans). Nauka, Moscow, pp. 266–287 (in Russian). Volkov, I.I., Rozanov, A.G., Yagodinskaya, T.A., 1971. Pyrite micro-nodules in the Black Sea sediments. Doklady Akademii Nauk SSSR 197, 195–198. Volkov, I.I., Rozanov, A.G., 1983. The sulphur cycle in oceans. In: Ivanov, M.V., Freney, J.R. (Eds.), The Global Biochemical Sulphur Cycle. Wiley, New York, pp. 357–448.

L.N. Neretin et al. / Deep-Sea Research I 48 (2001) 2569–2593

2593

Volkov, I.I., Kontar, E.A., Lukashev, Yu.F., Neretin, L.N., Nyffeler, F., Rozanov, A.G., 1997. The upper hydrogen sulfide boundary and the origin of the redox nepheloid layer (RNL) in waters of the Caucasus continental slope of the Black Sea. Geokhimiya 6, 618–629. Volkov, I.I., Dyrssen, D., Rozanov, A.G., 1998. Water alkalinity and anaerobic mineralization of organic matter in the Black Sea. Geochemistry International 36, 540–550. Wilkin, R.T., Arthur, M.A., Dean, W.E., 1997. History of water column anoxia in the Black Sea indicated by pyrite framboid size distributions. Earth and Planetary Science Letters 148, 517–525. Yao, W., Millero, F.H., 1995. Oxidation of hydrogen sulfide by Mn(IV) and Fe(III) (hydr)oxides in seawater. In: Vairavamurthy, M.A., Schoonen, M.A.A. (Eds.), Geochemical Transformation of Sedimentary Sulfur. ACS Symposium Series 612, 260–279. Zopfi, J., Ferdelman, T.G., Jrgensen, B.B., Teske, A., Thamdrup, B., 2001. Influence of water column dynamics on sulfide oxidation and other major biogeochemcal processes in the chemocline of Mariager Fjord (Denmark). Marine Chemistry 74, 29–51. Zubkov, M.V., Sazhin, A.F., Flint, M.V., 1992. The microplankton organisms at the oxic-anoxic interface in the pelagial of the Black Sea. FEMS Microbiology Ecology 101, 245–250.