[A J S ., 1999, P. 556–588] ANOMALOUS ENRICHMENTS OF IRON...

ANOMALOUS ENRICHMENTS OF IRON MONOSULFIDE INEUXINIC MARINE SEDIMENTS AND THE ROLE OF H2S IN IRONSULFIDE TRANSFORMATIONS: EXAMPLES FROM EFFINGHAM

INLET, ORCA BASIN, AND THE BLACK SEA

MATTHEW T. HURTGEN*, TIMOTHY W. LYONS**, ELLERY D. INGALL***,and ANNA M. CRUSE****

ABSTRACT. It is well documented that sedimentary pyrite formation proceedsthrough an iron monosulfide (FeS) precursor. Laboratory studies have tradition-ally indicated that intermediate S species such as elemental sulfur (S0) or polysul-fides (Sx

2�) are responsible for the transformation of FeS to pyrite (FeS2). Recentexperimental work, however, has suggested that H2S may also be responsible forthe transformation. The present study extrapolates reaction pathways respon-sible for pyrite formation in the laboratory to two fundamentally differentmodern anoxic marine systems. We hypothesize that on decadal timescales, H2S(or HS�) is responsible for transformations of FeS into FeS2 in natural systemswhere intermediate S species are isolated from FeS production. The possibility ofprolonged coexistence of high levels of H2S and FeS, however, challenges recentexperimental predictions of extremely rapid transformations (that is, timescalesof hours or less) of FeS to FeS2 via the H2S pathway.

Sediments of Effingham Inlet (a fjord on Vancouver Island) and the OrcaBasin (an intraslope brine pool in the northern Gulf of Mexico) are both charac-terized by atypically high concentrations of FeS but contrasting levels of H2S andFeS2 production. In both systems, FeS formation is spatially decoupled fromintermediate S species as a consequence of either rapid deposition/burial orextreme water-column stratification. Within settings that promote this separa-tion, H2S may be the principal species responsible for pyrite formation. FeS toFeS2 transformations are favored by the high concentrations of H2S in sedimentsof Effingham Inlet. Additional results from the margin of the anoxic Black Seacorroborate the Effingham model for iron sulfide transformation.

H2S concentrations are controlled by the amount of bacterial sulfate reduc-tion and the availability of reactive Fe. H2S concentrations will be buffered to lowlevels via the production of FeS in systems with appreciable amounts of reactiveFe and therefore be unavailable to transform FeS to FeS2. Under conditionswhere H2S is the only species available for FeS to FeS2 transformations, somedegree of Fe limitation may promote pyrite formation by allowing H2S toaccumulate in the pore waters. Ultimately, this balance between H2S productionand reactive Fe availability may strongly influence the amount of pyrite formedin anoxic systems.

INTRODUCTION

Pyrite burial is the major sink of reduced sulfur and iron in the marine environmentand is linked to the biogeochemical cycling of O, C, S, and Fe via bacterial sulfatereduction. In particular, balanced changes in the reservoirs of pyrite and organic carbon(OC) have played a major role in regulating atmospheric O2 through geologic time(Holland, 1973, 1984; Berner and Raiswell, 1983; Kump and Garrels, 1986; Berner,1987). These factors and the widespread occurrence of pyrite in both modern andancient sedimentary environments have stimulated extensive research in the laboratory(Berner, 1964; Rickard, 1969, 1975, 1997; Pyzik and Sommer, 1981; Luther, 1991;Schoonen and Barnes, 1991a, b; Wilkin and Barnes, 1996; Canfield, Thamdrup, and

* Department of Geological Sciences, University of Missouri, Columbia, Missouri 65211. PresentAddress: Pennsylvania State Astrobiology Research Center, University Park, Pennsylvania 16802

** Department of Geological Sciences, University of Missouri, Columbia, Missouri 65211*** Marine Science Institute, University of Texas, Port Aransas, Texas 78373

**** Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institute, WoodsHole, Massachusetts 02543

[AMERICAN JOURNAL OF SCIENCE, VOL. 299, SEPT., OCT., NOV., 1999, P. 556–588]

556

Fleischer, 1998) and in the field (Berner, 1970, 1984; Goldhaber and Kaplan, 1974;Jørgensen, 1977; Howarth and Jørgensen, 1984; Raiswell and Berner, 1985; Morse andCornwell, 1987; Boesen and Postma, 1988; Middelburg, 1991; Canfield, Raiswell, andBottrell, 1992; Gagnon, Mucci, and Pelletier, 1995; Lyons, 1997; Wilkin and Barnes,1997). However, despite the abundance of previous work, reaction pathways responsiblefor low temperature pyrite formation in natural systems and, specifically, the transforma-tion of an iron monosulfide (‘‘FeS’’) precursor phase to pyrite (FeS2) remain unresolved.

It is well documented that pyrite formation is strongly related to: (1) rates ofsedimentation and bacterial sulfate reduction; (2) the amount and reactivity of Fe andOC; (3) water-column redox conditions, and (4) the presence and spatial distribution ofSO4

2�, H2S, and intermediate S species and the requisite oxidants (Berner, 1970, 1984).The present study examines how these factors control the reaction pathways of ironsulfide formation within Effingham Inlet and the Orca Basin, with additional reference toblack muds from the Black Sea margin.

Traditionally, laboratory and field studies have indicated that authigenic pyriteformation proceeds via an FeS precursor which forms on rapid timescales (com-monly9 1 yr) through the reaction of dissolved sulfide (H2S) or bisulfide (HS�) withreactive phases of Fe (fig. 1) (Berner, 1964; Rickard, 1969; Schoonen and Barnes,1991a, b; Canfield, Raiswell, and Bottrell, 1992). At typical marine pH values, mostdissolved sulfide is present as HS� which reacts with Fe2� to form FeS. Reactive Feavailable for FeS production is supplied via detrital iron minerals dominantly in the formof Fe oxides (Berner, 1984; Canfield, 1989). H2S is produced through bacterial sulfatereduction in the absence of oxygen and is promoted by the availability of reactiveorganic matter (Berner, 1980, 1984; Westrich and Berner, 1984; Middelburg, 1989).Previous workers have suggested that precipitation of pyrite can occur directly throughthe reaction of H2S and Fe oxides (Howarth, 1979) without an FeS precursor. However,Schoonen and Barnes (1991a, b) argued that this reaction is not significant at lowtemperatures unless preexisting pyrite nuclei are present.

Rates and pathways responsible for the transformation of FeS to FeS2, including therequisite chemical species, have been studied extensively. Despite this work, thesereactions remain more uncertain than those involved in the formation of FeS (fig. 1). FeSis an approximate stoichiometry for a class of minerals known collectively as ‘‘ironmonosulfide’’ which include disordered FeS, mackinawite (FeS1�X), and greigite (Fe3S4)(Berner, 1970; Morse and others, 1987). FeS, also termed acid-volatile sulfide (AVS), hasbeen operationally defined in the present study as those phases solubilized during a cold6 N HCl extraction, which may underestimate the quantity of recalcitrant ‘‘FeS’’ phasessuch as greigite (Chanton and Martens, 1985). AVS is generally considered to bemetastable under marine sedimentary conditions and is normally transformed to pyritewithin the uppermost centimeters proximal to the redox interface of a ‘‘typical’’ marinesediment profile (Berner, 1970). Traditionally, researchers have attributed FeS preserva-tion in modern sediments to a lack of the intermediate S species thought to be necessaryfor the transformation of FeS to FeS2 (Boesen and Postma, 1988; Middleburg, 1991;Gagnon, Mucci, and Pelletier, 1995; Lyons, 1997).

The transformation of FeS to FeS2 through the reaction of FeS with intermediatesulfur species such as S0 and Sx

2� (Berner, 1970; Rickard, 1975; Luther, 1991; Schoonenand Barnes, 1991a, b; Wilkin and Barnes, 1996) has been well documented in thelaboratory, for example:

FeS � S0= FeS2, (1)

FeS � Sx2�= FeS2 � S (x�1)

2� . (2)

Matthew T. Hurtgen and others 557

Production of intermediate S species, whether biotic or abiotic, is initiated through H2Soxidation near the redox boundary by oxidants such as O2, NO3

�, and Fe and Mn oxides(Brock and Gustafson, 1976; Howarth and Jørgensen, 1984; Aller and Rude, 1988). Alink to oxidation is consistent with the observation that for many environments themajority of pyrite forms early, near the redox boundary, either in the water column (ineuxinic settings which are characterized by anoxic-sulfidic bottom waters) or within thesediments (Goldhaber and Kaplan, 1974; Canfield, 1989; Middelburg, 1991; Canfield,Raiswell and Bottrell, 1992; Henneke and others, 1997; Lyons, 1997). By contrast, somemodern environments display pyrite formation or, more specifically monosulfide todisulfide transformation, that is either arrested or delayed and stratigraphically removed(tens of centimeters) from the redox interface within the sediments (Boesen and Postma,1988; Middelburg, 1991; Gagnon, Mucci, and Pelletier, 1995; Lyons, 1997). Generally,intermediate S species do not exist in appreciable amounts away from the oxic-anoxicinterface and the electron acceptors associated with their formation and therefore maylimit pyrite formation (Luther and others, 1986; Thode-Andersen and Jorgenesen, 1989;

Fig. 1. General model for sedimentary pyrite formation via an FeS precursor (after Berner, 1984). H2S isproduced anaerobically through bacterial sulfate reduction. The H2S reacts rapidly with Fe to form FeS.Laboratory studies have traditionally indicated that intermediate S species (for example, S0 and Sx

2�) areresponsible for the transformation of FeS to FeS2. Recent experimental work, however, suggests that H2S canalso be responsible for the direct transformation.

Matthew T. Hurtgen and others—Anomalous enrichments of558

Henneke, Luther, and De Lange, 1991). In such cases, an additional mechanism isrequired to explain observed down-core transformations of FeS to FeS2.

Recent experiments by Rickard (1997) and Rickard and Luther (1997) suggest that apathway involving reaction between FeS(s) with H2S(aq) may also be responsible for thetransformation as represented by the overall reaction:

FeS � H2S= FeS2 � H2. (3)

This ‘‘H2S pathway’’ reflects, mechanistically, a series of elementary reactions and wasfound to proceed more rapidly than the ‘‘polysulfide pathway’’ (reaction 2) and all otherpathways involving intermediate S species (Rickard, 1997; Rickard and Luther, 1997).Taylor, Rummery, and Owen (1979) and Drobner and others (1990) observed reaction(3) at 100°C; however, additional efforts to duplicate these results inorganically atambient marine temperatures have proven unsuccessful (Wilkin and Barnes, 1996;Benning and Barnes, 1998). Canfield, Thamdrup, and Fleischer (1998) determined thatin the presence of bacteria, pyrite formation occurred through reactions (2) and (3) atrates of up to 104 to 105 times faster than those reported by Rickard (1997) in the absence ofbacteral mediation, although the exact role bacteria play in these reactions was unspecified.The kinetic models of both Rickard (1997) and Canfield, Thamdrup, and Fleischer (1998)predict extremely rapid transformation of FeS to FeS2 via the H2S pathway (that is, timescalesof hours or less), which is in direct contrast to the results of the present study.

In natural systems, progress of reactions (1) and (2) may be limited by spatialdecoupling between the production of intermediate S species and FeS. Within thiscontext, the H2S pathway (reaction 3) provides a mechanism that more plausiblyexplains pyrite formation via an FeS precursor under strictly reducing conditionsremoved from the redox interface in an oxidant-limited setting. The present effort is acomparative study of the transformation of FeS to FeS2 in two fundamentally differentanoxic marine environments, Effingham Inlet (a fjord on Vancouver Island) and theOrca Basin (an intraslope brine pool in the northern Gulf of Mexico). Additional datafrom the Black Sea that are analogous in many respects to those of the anoxic portion ofEffingham Inlet are briefly summarized. These natural laboratories are marked bydown-core persistence of FeS at wt percent levels for tens of centimeters. The sedimentsfrom the Orca Basin are characterized by persisting high levels of FeS in the absence ofH2S and FeS2 production, while sediments from Effingham Inlet (present study) andthose of the Black Sea margin (Lyons, 1991, 1992, 1997) show substantial subsurfacepyrite formation with systematic transformation of FeS to FeS2 on decadal timescalesunder conditions of comparatively high dissolved H2S (100–103 µM).

The rapid laboratory-based reaction kinetics (Canfield, Thamdrup, and Fleischer,1998; Rickard, 1997) argue against the coexistence of appreciable concentrations of FeSand H2S. The absence of both H2S and S intermediates preclude both the H2S pathwayof pyrite formation and those involving sulfur species of intermediate oxidation states(for example, the polysulfide pathway) in the Orca Basin. We propose that underconditions promoting the separation of H2S and intermediate S production, as inEffingham Inlet and the Black Sea margin, H2S may play a direct role in the transforma-tion of FeS to FeS2 over extended diagenetic timescales (101 to �102 yrs). However, therates and mechanisms of reaction appear to be slower and more complex than thosepredicted by the experimental conditions of Canfield, Thamdrup, and Fleischer (1998)and Rickard (1997).

SITE AND SAMPLE DESCRIPTIONS

Effingham InletEffingham Inlet is an anoxic fjord located on the west coast of Vancouver Island (fig.

2) which, unlike nearby Saanich Inlet, has only recently received the attention of

iron monosulfide in euxinic marine sediments 559

Fig. 2. Map of Effingham Inlet displaying sills and the inner (Site EF1) and outer (Site EF2) basins andcoring locations. The fjord is located on the west coast of Vancouver Island and opens into the Imperial EagleChannel which connects to the eastern Pacific Ocean.


researchers (Baumgartner and others, 1999). The fjord is characterized by steep valleywalls, a wet climate and a high input of sediment and terrestrial organic matter. The fjordis subdivided by two sills which inhibit lateral circulation and restrict ventilation of deepwater in the two major sub-basins. A well-developed pycnocline restricts vertical mixingbetween surface and deep waters. A stratified water column has developed, with thedeep portions of the inner and outer basins filled with cool, dense, saline water(Baumgartner and others, 1999). Sediments for this portion of the study were collectedby gravity coring at two sites (EF1 and EF2, fig. 2) in late May 1997 aboard the R/VBarnes.

The maximum water depths of the inner and outer basins are 120 and 210 m,respectively. At the time of sampling, O2 concentrations in the inner basin reached zeroat a water depth of approx 70 m, which is just below the depth of the inner sill.Baumgartner and others (1999) reported a similar O2 profile for the inner basin inDecember 1995. This O2 depletion is a consequence of consumption through organicdecomposition and restricted ventilation due to the stratified water column. Baumgart-ner and others (1999) found low but nonzero O2 concentrations (0.5 mL/L) in the deepportion (195 m) of the outer basin in December 1995. In May 1997, O2 concentrations ata depth of 130 m on the margin of the outer basin (near site EF2; fig. 2) were 0.2 mL/L.This subtle but meaningful disparity suggests that some seasonal flushing may occur,most likely between August and December when upwelling onto the Pacific shelfnormally occurs (Baumgartner and others, 1999). Baumgartner and others (1999) arguedthat seasonal upwelling occurs only sparingly in the inner basin because the water at theinner sill depth does not reach a density capable of displacing the deep basin water. Thisis also supported by the observation that O2 concentrations were zero in the inner basinboth in December and May.

One gravity core was collected from the anoxic deep portion of the inner basin(EF1-GC3; 120 m water depth). The organic-rich sediments at EF1-GC3 are overlain byan anoxic-sulfidic (euxinic) water column which precludes bioturbation. Consequently,the upper �28 to 30 cm consist of black microlaminated (mm-scale) mud. Below thissurface interval is a massive layer which extends to the base of the core (�87 cm) and isinterpreted to be a gravity flow originating from the steep walls of the basin margin(Baumgartner and others, 1999). Sediments above the gravity flow were emphasized inthe present study. Sedimentation rates were determined by 210Pb dating in a complemen-tary study (Ingall) to be approx 0.5 cm/yr in the anoxic inner basin.

An additional gravity core was collected along the oxic margin of the outer basin(EF2-GC7; 121 m water depth). It consists of homogenized, bioturbated, gray-brownmuds extending over the entire length of the core (28 cm). Small (mm-scale) openburrows were present at the sediment-water interface, and shell fragments were foundthroughout the core with increasing abundance in the basal intervals. Black mottles,likely iron monosulfide, occur in the upper few centimeters. Pine needles and small twigswere found throughout the core, illustrating the high degree of terrestrial organic inputand preservation. Preliminary sedimentation rates via the 210Pb method are less than afactor of two lower than those for the anoxic portion of the inner basin.

Orca BasinThe Orca Basin is an intraslope depression in the northern Gulf of Mexico (fig. 3)

characterized by extreme water-column stratification (Sheu, 1990). The Orca Basin is ahighly stable, anoxic brine pool associated with salt diapirism along the Louisianacontinental slope. Morphologically, the basin is a 400 km2 elbow-shaped depression withwater depths ranging from an upper limit of about 1800 to 2400 m in the deepestportions (Shokes and others, 1977). There are two sub-basins separated by a medialsaddle (fig. 3). The bottom 200 m of the Orca Basin, including the saddle, arepermanently anoxic and maintain a salinity of �260 per mill due to the dissolution of


near-seafloor salt diapirs (Shokes and others, 1977; Sheu and others, 1987). The seawateroverlying the basinal brine has a typical marine salinity of 35 per mill. The resultingdensity contrast at the seawater-brine interface between depths of 2200 and 2250 mmarks extreme stratification within the water column, with a density interface sufficientlydramatic to appear as a mid-water seismic reflector (Shokes and others, 1977; Trabantand Presley, 1978).

Water-column samples and six cores collected in September and November 1996aboard the R/V Longhorn were used in the present study. Water samples were retrievedusing 12 L Niskin bottles mounted on a CTD rosette. Four sediment cores (fig. 3) werecollected using a box corer (LH996-BC6, LH996-BC7, LH1196-BC7, and LH1196-BC5)and two using a Knight-type piston corer (LH996-KC8 and LH1196-KC2). Sedimenta-

Fig. 3. Geographic location and detailed map of Orca Basin showing coring locations (after Shokes andothers, 1977). The dashed contour delineates the anoxic brine interface.


tion rates as high as 61.3 cm/ka have been reported within the Orca Basin (Addy andBehrens, 1980). However, Behrens (1980) suggested that these high rates are misleadingdue to a very high water content of the sediments. They further suggested thatintrabasinal and extrabasinal rates would be similar if calculated using a constantporosity (�10 cm/ka at 50 percent porosity). The steep walls of the basin (3°–14°) suggestthat periodic slumping likely occurs (Trabant and Presley, 1978).

Core LH996-BC6 (32 cm total length; 2336 m water depth) was collected at ananoxic site on the medial saddle (fig. 3) and consists entirely of black microlaminated(mm- to cm-scale) mud. This core was processed at 2 cm intervals for geochemicalanalysis. Sediments from subcores LH1196-BC7 (38 cm; 2400 m water depth) andLH1196-BC5 (26 cm; 2395 m water depth) were collected from the deep portion of thesouthern sub-basin and also consist of black microlaminated mud. Despite the soupytexture of the black basinal sediments, X-radiographs reveal that the box cores were littledisturbed during collection. Fronds of unaltered Sargassum are ubiquitous throughoutthe sampled sediments. Carbonate and siliceous microfossils are well preserved in theseanoxic muds (Kennett and Penrose, 1978; present study).

Core LH996-BC7 (52 cm; 2240 m water depth) was collected where the water-column redox transition (chemocline) impinges with the basin margin. The upper 43.5cm of these sediments exhibit a dramatic brick red color mottled with light-gray mud;what appears as mottling may actually be accentuated by smearing of interlayeredlight-gray mud along the sides of the core barrel. X-radiography of a subcore reveals asubtle cm-scale banding within this red upper interval. A unit of light-gray mud extendsbelow this surface interval to the base of the subcore (52 cm).

The sediments from LH996-KC8 (45 cm; 2039 m water depth) were collected alongthe basin margin under an oxic water column and consist of a water-rich, homogenized,bioturbated, olive-brown mud spanning the upper 6 cm. Below this surface interval is amore cohesive (less porous) homogenized, bioturbated unit which is light gray in colorand extends to the base of the core. Sediments from core LH1196-KC2 (62 cm; 1994 mwater depth) were collected well above the redox transition zone and consist of a reddishbrown, homogenized, bioturbated unit spanning the upper 28 cm. Two homogenized,bioturbated units exist below this surface interval: a mottled brown and gray horizon(28–44 cm) and an olive-gray (44–62 cm) interval. Sargassum fronds were not found inLH996-KC8 nor in LH1196-KC2.

Black SeaHighlights from an earlier study of the Black Sea are summarized in the context of

the present investigation. Details, including sediment and site descriptions, methodolo-gies, and additional geochemical data, are available in Lyons (1991, 1992, 1997) andAnderson, Lyons, and Cowie (1994). Briefly, the Station 15 box core herein highlightedwas collected at an anoxic-sulfidic site on the basin margin at a water depth of 198 m. Thecore is approx 50 cm of very water-rich mud displaying a millimeter- to centimeter-scalecolor banding of alternating dark gray to jet black sediment. This banding is superim-posed on a finer-scale lamination, visible in X-radiograph. The X-radiograph reveals acomplete lack of bioturbation and shelly benthic fauna. A rapid sediment accumulationrate of 0.77 cm/yr is indicated by 210Pb analysis and is corroborated by a downcorepersistence of 137Cs, an anthropogenic radionuclide, to depths in excess of 30 cm (Mooreand O’Neill, 1991; Anderson, Lyons, and Cowie, 1994).

ANALYTICAL METHODS

Cores and subcores retrieved from both Effingham Inlet and the Orca Basin wereimmediately processed on board (extruded, 1 or 2 cm intervals) under an N2 atmosphere


at �8°C, the ambient bottom water temperature. Samples were taken from the center ofthe core to avoid portions smeared during coring and processing. Subsamples werestored in 50 mL centrifuge tubes and frozen immediately for later solid-phase chemicalanalyses.

Rates of bacterial sulfate reduction in the sediments were determined using the 35Sradiotracer technique described by Jørgensen (1978). During core processing under anN2 atm, 10 mL syringes with their ends removed were packed with sediment, sealed withrubber septa, and injected with 80 µL (0.8 µCi) of a 35SO4

2� solution. The sediments wereincubated in the dark at �8°C to approximate ambient conditions for roughly 12 hr, afterwhich they were frozen to terminate reaction. Additionally, �50 percent of the sedi-ment-filled syringes were injected with Zn-acetate prior to freezing to fix sulfide andpoison bacterial reduction of sulfate. Our results indicate no meaningful difference insulfate reduction rates among samples from the same interval with and withoutthe Zn-acetate poison. In the laboratory (Columbia, MO), H2

35S was liberated via anAVS distillation of the freshly-thawed sediment using 6 N HCl containing 15 per-cent SnCl2 (room temperature, stirred, 1.5-hr extraction; Berner, Baldwin, andHoldren, 1979; Chanton and Martens, 1985; Morse and Cornwell, 1987). The H2S gaswas driven via an N2 carrier into �30 mL of a 2 percent Zn-acetate with 6 percentNH4OH where it precipitated as ZnS. Separate aliquots of the dispersed ZnS precipitate(H2

35S) and the remaining 35SO42�-containing solution in the reaction vessels (centri-

fuged to isolate solid and liquid phases) were withdrawn for liquid scintillation counting.Rates of sulfate reduction were calculated based on the H2

35S/35SO42� activity ratio, the

SO42� concentrations as determined independently (see below), and the duration of

incubation. A fractionation factor (�) of 1.06 was used in the rate calculations (Jørgensen,1978).

Pore waters were isolated via centrifugation. Two subsamples of each filteredpore-water interval were fixed on board using a 3 percent Zn solution inside an N2-filledglove bag immediately following centrifugation. These fixed samples were stored at�8°C until analysis in Columbia, Missouri. Sulfate was quantified in one of the twosubsamples using ion chromatography. Analytical precision of 5 percent was obtainedon replicate SO4

2� analyses of an International Association for the Physical Sciences ofthe Oceans (IAPSO) standard and duplicate unknowns. In the other fixed subsample,total dissolved sulfide was determined spectrophotometrically using the methylene bluetechnique of Cline (1969) with a detection limit of 3 µM.

A sequential sulfur extraction scheme was employed to determine sediment concen-trations of AVS-S (S bound as FeS) and pyrite-S. AVS was extracted from freshly-thawedsediment using 6 N HCl containing 15 percent SnCl2 (room temperature, stirred, 1.5 hrextraction; Berner, Baldwin, and Holdren, 1979; Chanton and Martens, 1985; Morseand Cornwell, 1987; Lyons, 1997). The H2S was driven via an N2 carrier gas into �30mL of 3 percent Zn acetate solution with 10 percent NH4OH and trapped as ZnS forquantification by iodimetric titration. This HCl-SnCl2 extraction yielded sulfur recover-ies averaging 98.6 percent for CdS standards. The remaining sediment was then filteredand analyzed for pyrite-S using the chromium reduction method described by Canfieldand others (1986). The chromium reduction method is specific to reduced inorganicforms of S (pyrite S � AVS-S � S0) [Canfield and others, 1986]. No separate determina-tion of S0 was performed (see discussion below). Sulfur yields on the order of 96 percentand higher have been achieved for pyrite standards using this technique.

Degree-of-sulfidation (DOS) values were determined to assess the extent of reactionbetween H2S and ‘‘reactive’’ Fe. DOS is directly analogous to the more widely useddegree-of-pyritization (DOP) (see Berner, 1970; Raiswell and others, 1988) but isformulated to account for appreciable concentrations of AVS (Boesen and Postma, 1988)


and is calculated as follows:

DOS �(AVS-Fe) � (pyrite Fe)

(pyrite Fe) � (reactive Fe).

‘‘HCl-Fe’’ is traditionally treated as the remaining unsulfidized portion of the Fe poolwith the potential to react with H2S. By standard convention, ‘‘HCl-Fe’’ is defined as thefraction of total Fe extractable during a boiling 12 N HCl distillation for 1 min usingdried, homogenized sediments (Berner, 1970; Raiswell and others, 1988). It is assumedthat during drying, all AVS oxidizes to phases soluble in boiling HCl. Recent studieshave shown that the HCl-soluble fraction, often includes Fe-bearing mineral phases (forexample, silicate Fe) that require prolonged exposure to dissolved sulfide (�102 yr) and,consequently, may be an overestimation of the most readily reactive Fe (Canfield,Raiswell, and Bottrell, 1992; Raiswell, Canfield, and Berner, 1994; Raiswell and Can-field, 1996). The quantities of AVS-Fe and pyrite-Fe are calculated from the sulfur dataassuming stoichiometries of FeS and FeS2, respectively. Concentrations of extracted Fewere quantified spectrophotometrically using the ferrozine method of Stookey (1970).Duplicate HCl-Fe extractions using unknowns yielded standard deviations of less than10 percent of the mean.

For a selected group of samples from Effingham Inlet, extractions using buffereddithionite were also performed (sodium dithionite in a pH 4.8 buffer comprising 0.35 Macetic acid and 0.2 M sodium citrate; room temperature) (Canfield, 1989; Raiswell,Canfield and Berner, 1994). The phases extracted via this method are dominated by Feoxides and iron oxyhydroxides and, consequently, are a more accurate representation ofthe Fe component that is reactive on very short timescales (9102 yr) than the boilingHCl approach (Canfield, 1989; Canfield, Raiswell, and Bottrell, 1992; Raiswell, Can-field, and Berner, 1994). As with the boiling HCl, it assumed that Fe originally present asAVS oxidizes quantitatively to phases soluble in the buffered dithionite (Raiswell,Canfield, and Berner, 1994). Dithionite-extracted Fe was quantified using an inductivelycoupled plasma-optical emission spectrometer.

Stable sulfur isotope compositions of AVS and FeS2 for a representative suite ofsediments from the inner basin of Effingham Inlet were measured on Ag2S precipitates ofthe sulfide liberated during a sequential extraction involving the AVS distillation andchromium reduction method as described above (also Lyons, 1997). Aliquots of Ag2Swere combusted in the presence of cupric oxide under vacuum for a quantitativeconversion to sulfur dioxide and analyzed via mass spectrometry. Sulfur isotope resultsare expressed as per mill deviations from the S isotope composition of the troilite phaseof the Canon Diablo meteorite (CDT) using the conventional delta (�34S) notation. Sulfurisotopic results were generally reproducible within �0.1 to 0.2 per mill.

Concentrations of total and inorganic C were determined via combustion (950°C)and acidification (2 N HCl), respectively. The evolved CO2 was quantified by coulomet-ric titration (Huffman, 1977); organic carbon (OC) was calculated by difference. Analyti-cal error of less than 1 percent was obtained on replicate C analyses of a pure CaCO3standard and duplicate unknowns. Finally, all solid-phase data collected from within thebrine-filled Orca Basin have been salt corrected to account for mass contributions fromoriginally dissolved pore-water salts.

RESULTS

Effingham InletSite EF1: anoxic-sulfidic inner basin.At the time of sampling (5/97), the water column

of the fjord’s inner basin was anoxic-sulfidic (euxinic) with O2 values reaching zero at 70m and H2S values reaching �160 µM at the sediment-water interface (120 m water


depth). The first observed H2S occurred at 65 m water depth (2.83 µM). Pore-water andsolid-phase data are summarized in table 1.

Figure 4A illustrates that rates of sulfate reduction are as high as 800 nmol/cm3/dayin the microlaminated sediments of the inner basin. These rates drop off dramaticallywithin the upper few centimeters despite high concentrations of OC which actuallyincrease down core (fig. 4A; table 1). In particular, there is a significant increase in OCfrom 6.2 to 9.1 wt percent across the boundary at �28 to 30 cm from the blackmicrolaminated mud to the underlying massive unit interpreted to be a gravity flow.Nevertheless, sulfate reduction rates remain fairly uniform and low across this sameinterface. The pore-water profile displays SO4

2� depletion concurrent with H2S enrich-ment with increasing depth (fig. 5A; table 1). H2S values are high over the full length ofthe core (up to 6.08 mM).

Anomalously high concentrations of FeS persist to depth within the inner basin butshow a gradual down-core drop starting at a maximum value of 0.37 wt percent for the 2to 4 cm interval (fig. 6A; table 1). At Site EF1, increases in pyrite concentration withdepth are coincident with FeS depletion. More specifically, the ratio of AVS-S to pyrite-Sdrops dramatically over the first �14 cm suggesting a diagenetic transformation of FeS toFeS2 (fig. 6B; table 1). This is not, however, a simple one-to-one down-core transforma-tion of monosulfide to disulfide. AVS-S shows a systematic, approximately threefolddecrease in concentration over the microlaminated interval (table 1), but the correspond-ing increase in pyrite-S far exceeds that predicted by a simple transformation (that is,2 AVS-S loss). Calculated concentrations of total reactive Fe (pyrite-Fe plus HCl-Fe,which includes AVS-Fe) start at 1.50 wt percent in the uppermost sample and increasedown core to 2.99 wt percent by the approximate base of the microlaminated zone(�28–30 cm). Intermediate DOS values were measured in the microlaminated mud atEF1 with values ranging from 0.52 to 0.66, and despite a general down-core increase intotal iron sulfide (0.91–1.54 wt percent, there is a corresponding trend of decreasingDOS in the lower portion of the core starting with 12 to 14 cm interval (table 1). Theisotope data of FeS and FeS2 (fig. 6C; table 2) display nonsystematic trends with theexception of a progressive 34S enrichment for AVS below the uppermost layers.

Site EF2: Oxic margin—outer basin.Rates of sediment accumulation and the waterdepths for the margin of the outer basin are similar to those in the anoxic site (fig. 2; table1). The main difference between EF1 and EF2 is that the margin of the outer basin isoverlain by a weakly oxic water column despite the fact that it sits below the outer silldepth. The down-core persistence of bioturbation suggests a long-term dominance byoxic conditions. Sediment and pore-water geochemical properties of Site EF2 aresummarized in table 1.

Rates of sulfate reduction in the upper few centimeters are lower than those fromSite EF1 by almost a factor of four (fig. 4B). A maximum rate of 216.1 nmol/cm3/dayoccurs in the upper 2 cm. Below this depth, rates decrease significantly but are relativelyconstant and somewhat higher than those from Site EF1 over the same range of depths.Also, OC concentrations are in the same range as those in the microlaminated upper�28 cm from the euxinic inner basin, with a mean value lower by about 0.5 wt percentrelative to EF1 (compare fig. 4A and B). OC values remain fairly constant down core atEF2 with only a subtle decrease at the base. H2S values are substantially lower than thoseat Site EF1, and SO4

2� levels do not drop significantly over the interval of interest (table1), a relationship facilitated by bioirrigation (Aller, 1980). DOS values are low tointermediate and range from 0.11 to 0.30 (table 1). FeS-S concentrations are low in theupper few centimeters (0.12 wt percent) and drop to zero by �14 cm (table 1).

Orca BasinAnoxic deep basin.Sediment and pore-water geochemistry for the Orca Basin are

outlined in table 3. Rates of sulfate reduction and concentrations of OC are plotted


versus depth in figure 7. Uniformly low rates of bacterial sulfate reduction were found inthe sediments, with a maximum value of 38.4 nmol SO4

2�/cm3/day found in the upperfew centimeters (LH1196-BC7, table 4). Our water-column results suggest reductionrates in the redox transition zone are higher than the sediment maximum whencompared on a simple total-volume to total-volume basis. Sediment OC values rangefrom 1.1 to 4.1 wt percent, with scattered values generally increasing to a depth of 20 cmand then decreasing to the base of the core. Pore-water sulfate concentrations areconsiderably higher (up to 42.8 mM) than those found in typical deep waters of the Gulfof Mexico (compare LH1196-BC7 and LH1196-BC5 with LH1196-KC2, table 4) andnormal seawater (�29 mM). This enrichment is attributed to dissolution from thesurrounding near-surface, gypsum-containing salt diapirs. The concentrations remainfairly constant with increasing burial depth, exhibiting only a subtle decrease. Thesedata, combined with undetectable concentrations of H2S (3 µM), are consistent withthe low rates of sulfate reduction reported above.

Concentrations of AVS-S and pyrite-S are plotted versus depth in figure 8. Salt-corrected AVS concentrations reach high values in the surface layers which persist (up to1.9 wt percent) over the full 32 cm length of the core. However, very low pyrite-Sconcentrations were found within the sediments except in the 8 to 10 cm interval. DOSvalues range from 0.42 to 1.07 (table 3). The latter value represents procedural errorlinked to our assumption that all AVS-Fe oxidizes to a phase(s) extractable with theboiling 12 N HCl but nevertheless records an extreme extent of Fe sulfidation. A DOSvalue of 1.00 (the theoretical maximum) indicates that 100 percent of the originalreactive Fe (as defined above) has been sulfidized.

Transition Zone and Oxic MarginSediment and pore-water geochemical characteristics for the transition zone and

oxic margin are summarized in table 3. Sediments collected where the transition zoneimpinges along the basin margin contain elevated levels of reactive Fe (sulfide-Fe plusresidual unreacted HCl-Fe ranging up to 5.3 wt percent compared to an average value of2.2 wt percent for sediments below the interface). Pyrite-S and AVS-S concentrations arebelow detection (that is, 10�2 wt percent). OC values range from 0.38 to 0.63 wtpercent.

The gray-colored sediments from the oxic outer margin of the basin are character-ized by low rates of sulfate reduction (see LH1196-KC2, table 4) and undetectable levelsof pyrite-S and AVS-S. Sulfate concentrations range from 25.6 to 27.6 mM and decreaseonly slightly with depth. OC values are low compared to those of the anoxic sedimentsand range from 0.53 to 0.95 wt percent (table 3).

DISCUSSION

Comparison with Other AreasSediments at sites within Effingham Inlet, the Orca Basin, and the Black Sea contain

unusually high concentrations of sulfur as FeS which persist to depth. Additional coastalmarine settings with atypical FeS enrichment are discussed in Boesen and Postma (1988)and Gagnon, Mucci, and Pelletier (1995). Pyrite formation is coincident with FeSdepletion in sediments at Site EF1 in Effingham Inlet, which manifests as dramaticdown-core decreases in the ratio of AVS-S to pyrite-S over upper �14 cm (fig. 6B) andsuggests diagenetic transformation of FeS to FeS2 under complex non-steady-stateconditions. Additionally, down-core increases in �34S values for FeS and FeS2 (fig. 6C)suggest that some AVS and pyrite formation is occurring below the sediment-waterinterface, removed from the redox transition within the water column. A similarrelationship has been recognized along the Black Sea margin (Lyons, 1997, and fig. 9;table 5).


TA

BL

E1

Geo

chem

ical

char

acte

risti

csof

thes

edim

ents

and

pore

wat

ersw

ithin

thei

nner

and

oute

rbas

inso

fEffi

ngha

mIn

let.

AV

S-Fe

and

FeS 2

-Fec

once

ntra

tions

arec

alcu

late

dfr

omth

eSda

taas

sum

ing

stoic

hiom

etri

esof

FeS

and

FeS 2

,res

pect

ivel

y.H

Cl-

Fean

dD

-Fe

(dith

ioni

teex

trac

tabl

eFe)

valu

esin

clude

AV

S-Fe

(see

text

ford

etai

ls)


Using sulfur isotopic trends as recorded in the microlaminated (Unit 1) sediments ofthe deep, central basin of the Black Sea, Lyons (1997) determined that appreciable levelsof pyrite were formed rapidly in the water column in association with the chemoclinewhere the production of intermediate S species (that is, S0 and Sx

2�) is maximized.However, the �34S values for bulk pyrite from the more rapidly accumulating anoxicBlack Sea margin show 34S enrichments relative to the Unit 1 sediments of the deep basinsuggesting that iron sulfidation is occurring not only within the water column and at thesediment-water interface (syngenetic component) but also early within the sediments(diagenetic component). Lyons (1997) determined that the persistence of AVS withdepth, rather than the efficient transformation to FeS2 within the surface intervals, is afunction of rapid sediment accumulation at the basin margin site with a rate of 0.77cm/yr (Anderson, Lyons, and Cowie, 1994). In simple terms, along the basin margin ofthe Black Sea, rapid deposition and burial favor AVS preservation through rapidremoval of AVS from the vicinity of the redox boundary and the associated formation ofintermediate S species that form via H2S oxidation. In contrast to Effingham Inlet and theBlack Sea margin, the Orca Basin displays almost no pyrite formation with depth (that is,conversion of FeS to FeS2) despite atypically high concentrations of FeS.

Sulfate Reduction Rates and Organic Carbon in Effingham InletFigure 4A illustrates that rates of sulfate reduction are as high as 800 nmol/cm3/day

in the euxinic inner basin of Effingham Inlet (compare Lin and Morse, 1991; Canfield,

Fig. 4(A) Down-core distributions of rates of bacterial sulfate reduction and organic carbon (OC) for theanoxic Site EF1-GC3 in Effingham Inlet. OC is expressed as weight percent. (B) Down-core distributions ofrates of bacterial sulfate reduction and organic carbon (OC) for the oxic Site EF2-GC7 in Effingham Inlet.


1991; for typical coastal marine values). Sulfate reduction rates are controlled by thequantity and quality of organic matter (Berner, 1984) which are strongly influenced bysedimentation rate (Toth and Lerman, 1977; Berner, 1978; Canfield, 1989). Underconditions of rapid deposition/burial, sediment is quickly removed from the oxic (andanoxic) portion of the water column and efficiently delivered to the dissimilatory obligateanaerobes below the sediment-water interface. Therefore, the very high rates of sulfatereduction in the upper few centimeters at Effingham Inlet most likely represent reminer-alization of labile, marine-dominated OC phases effectively protected from water-column degradation through rapid deposition and burial (approx 0.5 cm/yr) in acomparatively shallow basin. However, rates of sulfate reduction drop off dramaticallybelow the top few centimeters despite significant concentrations of OC that increase tovalues as high as 9.1 wt percent with increasing depth below the sediment-waterinterface. This relationship suggests that the OC reservoir is dominated by a refractoryterrigenous component whose source is the adjacent highly vegetated steep valley walls(see fig. 2). The down-core increase in OC, as opposed to a more typical decrease,represents a non-steady-state system with differing levels of terrestrial input throughtime. Temporal variations in the character of the terrigenous flux are also manifested inthe trend for total reactive Fe defined as (py � AVS � HCl) which shows an appreciabledown-core increase (table 1) rather than a more typical steady-state profile with increas-ing sulfidation of a roughly constant flux of reactive Fe.

Fig. 5(A) Sulfide and sulfate concentrations versus depth for the anoxic Site EF1-GC3 in Effingham Inlet.(B) Sulfide and sulfate concentrations versus depth for the oxic Site EF2-GC7 in Effingham Inlet.


Fig.

6(A

)D

epth

dist

ribu

tions

ofA

VS-

San

dpy

rite

-Sfo

rgra

vity

core

EF1

-GC

3,an

oxic

site

,Effi

ngha

mIn

let.

Sulfu

rext

ract

eddu

ring

chro

miu

mre

duct

ion

ofth

eH

Cli

nsol

uble

resi

due

isas

sum

edto

repr

esen

tpyr

ite-S

.S0

valu

esar

ein

ferr

edto

bene

glig

ible

(see

text

for

disc

ussi

on).

(B)R

atio

ofA

VS-

Sto

pyri

te-S

vers

usde

pth

for

grav

ityco

reE

F1-G

C3.

(C)�

34S

valu

esfo

rA

VS

and

pyri

tefo

rgr

avity

core

EF1

-GC

3.


The geochemical character of the oxic margin of the outer basin supports the notionthat the OC reservoir is swamped by a comparatively non-reactive terrestrial fraction.The low rates of sulfate reduction in the upper few centimeters of the sediment relative tothose of the inner basin indicate that the more labile marine-dominated OC phases weredegraded aerobically within the water column or at the sediment-water interface (fig.4B). Below the upper few centimeters, rates are relatively constant and somewhat higherthan those from the euxinic inner basin at the same depths below the sediment-waterinterface. OC concentrations are only slightly lower than those at anoxic Site EF1. Thus,the labile marine phases are being consumed rapidly by sulfate reducers in the euxinicinner basin and aerobic organisms in the outer oxic system. The fact that rates of sulfatereduction fall off so quickly in the euxinic system and never reach high levels in the oxicsystem supports the premise that the OC pool is dominated by refractory land-derivedcomponents.

Degree of Sulfidation in Effingham InletAs a result of bacterial sulfate reduction, pore-water profiles for sediments of the

euxinic inner basin display SO42� depletion concurrent with H2S enrichment with depth

(fig. 5A). �H2S values are anomalously high (up to 6.08 mM) when compared to the oxicmargin of Effingham Inlet (fig. 5B) and other marginal marine systems (compareGoldhaber and Kaplan, 1974; Lyons and Berner, 1992), particularly given the abun-dance of terrestrial organic matter, suggesting at least short-term reactive Fe limitation.This is not atypical for euxinic systems where reactive Fe is often the limiting factor iniron sulfide formation (Berner, 1984; Raiswell and Berner, 1985; Raiswell and others,1988; Calvert and Karlin, 1991; Middelburg, 1991; Lyons and Berner, 1992). The Feavailable to react with H2S (or HS�) in marine environments on comparatively shorttimescales mainly occurs as iron oxides and oxyhydroxides (Canfield, 1989) but mayalso include ‘‘reactive’’ iron-bearing silicates (Canfield, Raiswell, and Bottrell, 1992).Canfield, Raiswell, and Bottrell (1992) found that this H2S-reactive Fe component iscomposed of Fe phases that may react on time scales of hours (ferrihydrite) to 102 yr (Fein magnetite and reactive sheet silicates).

The DOS values of the inner basin, ranging from only 0.42 to 0.55 (table 1) despiteexposure on decadal timescales to high concentrations of dissolved sulfide, also indicatethat the residual Fe reservoir is dominated by a comparatively nonreactive componentthat reacts only on timescales of �102 yr. Once corrected for contributions from oxidizedAVS-Fe, buffered sodium dithionite extractions (as outlined in the Methods section)yielded negligible Fe concentrations, corroborating short-term Fe limitation. Euxinicsettings are commonly associated with very high DOS values that approach 1.0 (Raiswelland others, 1988). Traditionally, these high DOP values have been attributed to theubiquity of sulfide and prolonged exposure under conditions of slow accumulation

TABLE 2

Sulfur isotopic compositions for AVS and pyrite for sediments collected from the inner basinof Effingham Inlet (EF1-GC3)

Depth (cm)

�34S permil* � EF1-GC3

(AVS) (FeS2)

2–4 �22.6 �22.36–8 �23.4 �22.8

10–12 �23.2 �23.916–18 �21 �2122–24 �17.2 �20.730–32 �13.8 �24.9


TA

BL

E3

Geo

chem

ical

char

acte

risti

csof

thes

edim

ents

and

pore

wat

ersw

ithin

and

surr

ound

ing

theO

rca

Bas

in.A

VS-

Fean

dFe

S 2-F

eco

ncen

trat

ions

arec

alcu

late

dfr

omth

eSda

taas

sum

ing

stoic

hiom

etri

esof

FeS

and

FeS 2

,res

pect

ivel

y.H

Cl-

Feva

lues

inclu

deA

VS-

Fe


(Raiswell and Berner, 1985; Raiswell and others, 1988). Recent work, however, haslinked spatial variation in DOP within the anoxic portions of basins to fundamentallateral differences in the Fe reservoir (Canfield, Lyons, and Raiswell, 1996; Lyons, 1997;Raiswell and Canfield, 1998). This recent work has shown that sedimentary pyrite formsin anoxic basins through reactions involving both detritally delivered Fe phases and

Fig. 7. Down-core distributions of rates of bacterial sulfate reduction (LH1196-BC7) and OC (LH996-BC6) for anoxic sites within the Orca Basin.


highly reactive Fe that is scavenged in the water column (that is, dissolved Fe and theoxidized products) and is decoupled from the local detrital flux. The relative contribu-tions of these two reservoirs ultimately influence the styles and extents of Fe sulfidationsuch that DOP (DOS) values can vary by as much as a factor of two under conditions ofconstant water-column anoxia across strong lateral gradients in siliciclastic flux (Lyons,1997).

Fig. 8(A) Depth distributions of AVS-S and pyrite-S for LH996-BC6, an anoxic site in the Orca Basin.Sulfur extracted during chromium reduction of the HCl-insoluble residue is assumed to represent pyrite-S. S0

values are inferred to be negligible (see text for discussion). (B) Ratio of AVS-S to pyrite-S versus depth forLH996-BC6.

TABLE 4

Sulfate concentrations and rates of sulfate reduction for the sediments and water column of theOrca Basin


DOS values for Effingham Inlet, despite the euxinic conditions, are intermediatebecause of their close proximity to the basin margin and the relatively short period oftime over which they have accumulated. Accordingly, the readily sulfidized water-column Fe signal is flooded by a less reactive detrital Fe component that is reactive onlyon longer timescales (�102 yrs) but is still extracted via the HCl method. DOS valuesequaling 1.0 are attained substituting sodium dithionite extractable Fe in the DOScalculation.

Pyrite Formation in Effingham InletEffingham Inlet displays pyrite formation or, more specifically, monosulfide to

disulfide transformation in the absence of appreciable concentrations of S intermediates.The stratified water column controls the spatial distribution of the chemical speciesresponsible for iron sulfidation. In the inner basin of Effingham Inlet, the O2-H2Sinterface occurs at a water depth of �70 m which is 50 m above the sediment-waterinterface. It is at this redox boundary that intermediate S species are formed through theoxidation of H2S and FeS by dissolved oxidants such as O2, NO3

�, and Fe and Mn oxides(Brock and Gustafson, 1976; Howarth and Jørgensen, 1984). Additionally, Sx

2� can formthrough the reaction of H2S and S0 (Cloke, 1963; Teder, 1971). Aller and Rude (1988)determined that reduced forms of sulfur (H2S and FeS) could be oxidized under anoxicconditions when Mn-oxides are present. However, only trace amounts of Mn oxides

Fig. 9(A) Depth distributions of AVS-S and pyrite-S for Station 15 on the anoxic basin margin of theBlack Sea (after Lyons, 1997). Sulfur extracted during chromium reduction of the HCl-insoluble residue isassumed to represent pyrite-S. S0 values are inferred to be negligible (see text for discussion). (B) Ratio ofAVS-S to pyrite-S versus depth for Station 15.


were detected in Effingham Inlet—based on buffered sodium-dithionite extractions whichshould provide an accurate estimate of Mn-oxide availability—and therefore do not playa significant role in oxidizing reduced sulfur in the subsurface.

In laboratory experiments, Pyzik and Sommer (1981) found that sulfur intermedi-ates (S0 and S2O3

2�) can form during reactions of H2S with Fe particles. The implicationis that reactive Fe oxides persisting below the redox boundary can provide a mechanismfor the production of S intermediates. However, buffered sodium-dithionite extractions(D-Fe in table 1), once corrected for contributions from oxidized AVS-Fe, suggested thatthese reactive Fe oxides are negligible in the anoxically deposited sediments of Effing-ham Inlet. Therefore, intermediate S species likely exist in appreciable amounts onlynear the redox boundary in the water column not within the sediments. Studies ofadditional anoxic environments have shown that S0 and Sx

2� exist only at very lowconcentrations away from the redox interface and therefore may limit pyrite formation(Luther and others, 1986; Thode-Andersen and Jørgensen, 1989; Henneke, Luther, andDe Lange, 1991).

Bacterial production of H2S in Effingham Inlet initiates within the water column justbelow the chemocline (fig. 10). Reaction between H2S (or HS�) and reactive Fe,including dissolved Fe, results in FeS formation below the redox boundary in the watercolumn, at the sediment-water interface, and within the sediments. In an averagesteady-state system, the observed down-core increase in total sulfide Fe would beparalleled by sympathetic increases in DOS as the reactive Fe reservoir becomesprogressively sulfidized; however, the DOS relationships and down-core trends for totalreactive Fe at Site EF1 highlight the complexity of the system. Despite this complexity,AVS-S concentrations decrease down core with corresponding increases in pyrite-S.This relationship is manifested in a systematic drop in the ratio of AVS-S to pyrite-S,particularly over the top 14 cm, in the presence of high concentrations of H2S but withlevels of intermediate S that are inferred to be negligible. While a portion of the pyriteincrease likely reflects direct precipitation on preexisting pyrite nuclei (Schoonen andBarnes, 1991a, b), and the Fe reservoir is complex in its temporal pattern, the data areconsistent with sulfide transformation with depth.

TABLE 5

Geochemical characteristics of the sediments and pore waters on the basin margin of the Black Sea(Station 15). AVS-Fe and FeS2-Fe concentrations are calculated from the S data assuming

stoichiometries of FeS and FeS2, respectively. HCl-Fe values include AVS-Fe


With the exception of the deepest measured interval for pyrite at EF1, down coreincreases in �34S values for FeS and FeS2 are in agreement with the ‘‘restricted-system’’reservoir effects expected for pore waters within rapidly accumulating, nonbioturbatedsediments (fig. 6C). This relationship suggests that pyrite and some AVS, despite thecorresponding decrease in the concentration of AVS-S, is forming below the sediment-water interface, which further corroborates the model for subsurface monosulfide todisulfide transformation. However, the comparatively narrow range of DOS valuesargues against substantial Fe sulfidation (AVS formation) at depth.

A simple mass balance calculation helps quantify the subsurface FeS formation.Bacterial sulfate reduction produces H2S at the rate of 40 nmol/cm3/day for the 18 to 20cm interval of EF1-GC3 (table 1), which is �0.13 mmoles/cm3 since initial deposition (18cm � 0.5 cm/yr � 9 yrs) or 0.7 percent S. This calculation assumes there are no additionsor losses via diffusion. Additionally, this number represents a minimum since thecalculation is based on a deep rate of sulfate reduction. It is possible that new FeS isforming concurrently with FeS transformation to FeS2. For example, 0.26 percent S(FeS-S for the 18–20 cm interval) of the total 0.7 percent S produced may be used to formnew FeS while 0.44 percent S may be utilized to convert FeS to pyrite.

Regardless of where the FeS is formed, the very high rates of deposition rapidlyremove it from the intermediate S species produced at the chemocline in the watercolumn. Additionally, high burial rates further isolate FeS from the intermediate Sspecies and deliver more-reactive Fe and organic carbon to the subsurface, encouragingsulfate reduction and some FeS formation at depth within the sediments. As a result ofthe collective physical and chemical conditions, the intermediate S species are largelydecoupled from FeS production, and pyrite formation cannot proceed effectivelythrough the reaction of FeS with these products of partial sulfide oxidation.

Despite this spatial decoupling between the availability of intermediate S speciesand FeS production and accumulation, FeS to FeS2 transformation appears to beoccurring at depth. This conclusion suggests that H2S may be the principal speciesresponsible for FeS transformation to pyrite in the euxinic sediments of Effingham Inlet.The amount of H2S within the pore waters is controlled by the balance between theavailability of reactive Fe and the production of H2S. Ultimately, this balance betweenreactive Fe and the production of H2S may have significant implications for the amountof pyrite formed. Therefore, under conditions where H2S is the critical species respon-

Fig. 10. Schematic model of pyrite formation in Effingham Inlet, anoxic Site EF1.


sible for FeS transformations into FeS2, some degree of Fe limitation may promote pyriteformation by allowing H2S to accumulate in the pore waters. Furthermore, contrary tothe rapid reaction kinetics predicted by Rickard (1997) and Canfield, Thamdrup, andFleischer (1998) for the H2S pathway, significant quantities of AVS can persist in thepresence of high concentrations of H2S for timescales that may extend to decades.

These results are consistent with those generated for the basin margin of the BlackSea (Lyons, 1991, 1992, 1997—see Station 15; table 5 of the present study) where the H2Spathway appears to be even more strongly suggested. The Black Sea site is characterizedby a number of properties similar to those of EF1: (1) rapid sediment accumulation rates(about 0.77 cm/yr as determined by 210Pb), (2) significant levels of H2S (39–144 µM), (3)an Fe reservoir that is dominated by less reactive detrital phases, (4) decreasing AVS-Sand increasing pyrite-S with increasing depth as reflected in a systematic drop in the ratioof AVS-S to pyrite-S (compare figs. 6B and 9B), and (5) �34S values showing a subtleincrease down core—albeit only by a few per mill (Lyons, 1997). Consequently, like EF1,the sediments exhibit low to intermediate DOS values (Station 15, 0.33–0.47) andcontain high FeS concentrations but also show significant pyrite formation (FeS to FeS2transformation; fig. 9A) (Lyons, 1991, 1992, 1997). Unlike EF1, however, both totalsulfide Fe and total reactive Fe remain comparatively constant down core which isreflected in a systematic down-core crossover between AVS-S and pyrite-S (fig. 9A)where the pyrite increase is actually a few tenths of a wt percent less than that predictedfrom AVS-S loss (that is, 2 AVS-S loss), and DOS values scatter but show a generaldecrease down core across a fairly narrow range.

As for Effingham Inlet, pyrite formation at the Black Sea site continues with depthdespite sedimentation rates that facilitate rapid removal of FeS production from theintermediate S species that form near the redox interface in the water column. Clearly,this is not a simple, progressive, one-to-one down-core transformation of early-formedAVS to pyrite. Furthermore, there are system complexities analogous to those of EF1,such as temporal variability in the quantity and quality of the reactive Fe and OC fluxes,varying extents of direct pyrite precipitation on pyrite seeds and variations in the relativefluxes of detrital Fe and Fe scavenged in the water column through syngenetic ironsulfide formation that is independent of the local siliciclastic flux.

Previous extractions (Lyons, 1992) using 1 M hydroxylamine hydrochloride in 25percent acetic acid as a measure of Mn oxides and highly labile Fe oxyhydroxides(Canfield, 1988, 1989) suggest minimal readily available Fe(III) and Mn(IV) in the BlackSea sediments. (Recall that the HCl extraction of Fe used in the calculation of DOS is anoverestimation of the labile Fe(III) availability.) This relationship, in combination withrapid deposition/burial and corresponding removal from the chemocline and othersources of oxidants necessary for the formation of intermediate S species, suggests thatH2S may be the principal species responsible for the FeS transformation to FeS2 alongthe basin margin of the Black Sea.

Lyons (1997), in comparing the black basin-margin muds and turbidites in the BlackSea, developed an isotopic mass balance model that further supports H2S as a mecha-nism for the sulfidation of FeS. A broad suite of geochemical properties for muddyturbidites of the abyssal plain in the Black Sea indicates that the sediments derive fromthe anoxic basin margin (for example, from sediments like those at Station 15 describedabove) (Lyons, 1991, 1992). The reworked sediments of the abyssal plain show DOP(DOS) values similar to those of the basin margin suggesting that additional ironsulfidation did not occur. However, the AVS originating from the marginal muds wascompletely transformed to pyrite during transport and redeposition in the likely absenceof sufficient intermediate S species (Lyons, 1997). Lyons (1997) determined that the Sresponsible for this conversion to pyrite had a �34S value equal to the ambient H2S of thedeep anoxic water column. In a very general sense, the timing of turbidite emplacement


as addressed by Crusius and Anderson (1991) using 210Pb is consistent with theprotracted rates of H2S-driven FeS to FeS2 transformation proposed in the present study.

Sulfate Reduction Rates and Organic Carbon in the Orca BasinRates of sulfate reduction (table 4; fig. 7) in the anoxic Orca Basin are low relative to

the anoxic sediments in Effingham Inlet, despite OC values ranging from 1.13 to 4.06 wtpercent. Past researchers have proposed that these low rates result from the high salinitywithin the Orca Basin and/or a more refractory OC reservoir (Wiesenburg, Brooks, andBernard, 1985; Sheu, 1987, 1990). The strong density contrast defined by the transitionfrom brine to the overlying normal seawater effectively traps organic material at theredox boundary and promotes localized sulfate reduction. Trefry and others (1984)found that 60 percent of the particulates suspended at the brine-seawater interface wereorganic matter. The organic matter suspended at the density interface will decrease inreactivity over time because microbial organisms will preferentially degrade the morelabile compounds first, leaving more refractory phases behind. The organic matter thateventually reaches the sediment is enriched in the refractory component.

Klinkhammer and Lambert (1989) argued that microbial activity in hypersalinewaters is severely inhibited. By contrast, Henneke and others (1997) determined thatrates of bacterial sulfate reduction within the hypersaline basins in the MediterraneanSea were similar to those found along continental margins and, therefore, determinedthat the brines do not appear to inhibit sulfate reduction rates. Additionally, LaRock andothers (1979) determined that in spite of hypersaline conditions within the Orca Basin,appreciable levels of microbial activity are present at the oxic-anoxic boundary, followedby a sharp decrease below the interface and then increase again in activity within thebottom waters. They further noted that this pattern is similar to other anoxic basinslacking hypersaline waters such as the Cariaco Basin and the Black Sea (Jannasch,Truper, and Tuttle, 1974; Karl, LaRock, and Shultz, 1977; Albert, Taylor, and Martens,1995).

Our results are consistent with the findings of LaRock and others (1979) andillustrate that sulfate reduction rates are highest near the brine-seawater interface and arelower within the sediments (table 4). Sediments deposited under the oxic water columncollected along the margin have even lower rates of sulfate reduction (table 4) and OCvalues that range from 0.53 to 0.95 wt percent despite the likelihood of shorterwater-column residence times in the absence of the strong density interface. Addition-ally, the fronds of Sargassum which were so prevalent in the sediments deposited withinthe anoxic basin are absent along the basin margin. The relatively low rates of sulfatereduction indicate that most of the labile OC phases, including the Sargassum fronds,were consumed at the sediment-water interface via aerobic respiration in combinationwith the slow sediment accumulation rates. Therefore, assuming that intrabasinal andextrabasinal sedimentation rates and OC fluxes are similar, this relationship suggests thatoxic degradation is more efficient than hypersaline anoxic degradation and that OCpreservation is enhanced under these anoxic conditions.

Degree of Sulfidation in the Orca BasinH2S in the Orca water column is largely restricted to low concentrations in the

region just below the brine-seawater interface. Its near absence deeper in the watercolumn has been attributed to the low rates of sulfate reduction as well as its reaction withreactive Fe to form FeS (McKee and others, 1978; Trabant and Presley, 1978). Evidencefor particulate FeS enrichment has been documented in the lower portion of the redoxtransition zone (Trefry and others, 1984; Van Cappellen and others, 1999). For example,Van Cappellen and others (1999) detected the highest concentrations of solid phase Fe atwater depths of 2240 to 2250 m. Additionally, we retrieved filtered black particulatesfrom this depth range that likely represent FeS. The reactive Fe necessary for FeS


formation is scavenged by dissolved H2S (or HS�) in the lower portion of the redoxtransition zone. The sediment collected where the chemocline impinges with the basinmargin is a brick red color from iron oxide enrichment, with reactive Fe (HCl-soluble)values as high as 5.3 wt percent, documenting the extent of redox cycling of Fe in thechemocline region. Furthermore, dissolved Fe concentrations reach �35 µM within theupper portions of the Orca brine, which is 1000 times higher than the overlying seawater(Trefry and others, 1984).

H2S was not detected (3 µM) within the sediments, suggesting effective consump-tion by Fe in combination with low production. As noted above, this balance betweenthe availability of reactive Fe and the production of H2S may have significant implica-tions for the amount of pyrite formed via mechanisms involving both H2S and intermedi-ate S species.

DOS values within the sediments of the Orca anoxic sites are high, ranging from0.42 to 1.07 (1.00) with an average value of 0.76, indicating that much of the availablereactive Fe has been sulfidized. This high extent of sulfidation is likely a function of theFe reservoir itself, which reflects mainly very reactive phases that are redox cycled andsequestered at the redox boundary, as opposed to Effingham which experiences a rapidinflux of refractory detritally delivered Fe phases.

Pyrite Formation in the Orca BasinConcentration relationships (saturation states) and the extreme stratification within

the water column of the Orca Basin result in an appreciable water-column gap betweenthe location of sulfide oxidation and concomitant production of intermediate S speciesand the production of FeS deeper in the transition zone (that is, separate isopycnalsurfaces for the production of the two phases). In particular, this phenomenon mayreflect the long residence time of particulate S0 along a density interface which minimizesthe potential for polysulfide production via the reaction of S0 with H2S. This effectiveseparation of the production of intermediate S species and FeS is controlled by processesgoverning the spatial distribution of O2, H2S, and Fe2� as reported by Van Cappellenand others (1999) (fig. 11). The redox transition zone extends from 2200 to 2250 m, withcorresponding salinities ranging from 50 to 260 per mill, respectively. The production ofintermediate S species may initiate in the upper portion of the transition zone via H2Soxidation and the most likely electron acceptor responsible for appreciable sulfideoxidation is O2 which is completely consumed by 2225 m (salinities 70 per mill). Inaddition, Mn oxides are present in the upper portion of the transition zone (2180–2210 mwater depth; Van Cappellen and others, 1999; Sheu, 1990). These are also capable ofoxidizing H2S (Aller and Rude, 1988). Therefore, H2S oxidation and the production ofintermediate S species, whether oxidized by O2, Mn oxides or other electron acceptors(Jørgensen and others, 1991), occurs dominantly in the upper portion of the transitionzone.

Moderate amounts of H2S are produced in the middle and lower portion of thetransition zone (�2235 m, 130 per mill) through bacterial sulfate reduction. A portion ofthe dissolved sulfide most likely diffuses upward and is oxidized in the upper portion ofthe transition zone. Sheu and others (1987) using sulfur and oxygen isotope compositionsof dissolved sulfate in the water column determined that about 23 percent of the sulfideproduced near the transition zone must be oxidized to sulfate, likely involving theproduction of intermediate S species. Dissolved Fe is first measured in the lower portionof the transition zone at a salinity of 200 per mill (2240 m). This Fe2� reacts with H2S (orHS�) yielding FeS. Therefore, FeS appears to occur largely in the lower part of thetransition zone and is vertically decoupled from the intermediate S species whichmust exist dominantly in the upper portion due to their comparative instability understrongly reducing conditions and the paucity of available oxidants deeper in the watercolumn.


In an analogous water-column study in the Black Sea, Lewis and Landing (1991)observed a dissolved Fe maximum tens of meters below the first appearance of sulfide.Furthermore, they documented a suspended particulate Fe maximum also below thesulfide interface and coincident with the down-profile decrease for the dissolved Femaximum. The coincidence between increasing particulate Fe and decreasing dissolvedFe was thought to reflect solubility controls on FeS precipitation. It is important to note,however, that the extreme salinities and subsequent stratification observed in the OrcaBasin are not a factor in the transformation of FeS to FeS2 in the Black Sea. While theBlack Sea is stratified, salinities are 18 per mill in the surface waters and 22 per mill in thedeep waters (compared to 50 and 260 per mill for the Orca Basin). As a result, the densitycontrast is not sufficient to trap particulate S0 along the oxic-anoxic interface, and asignificant gap never develops between the production of intermediate S species andFeS. Consequently, FeS to FeS2 transformation is markedly more effective.

The black microlaminated sediments collected from the saddle region in the OrcaBasin contain an anomalously high concentration of sulfur as FeS with only minor pyritevalues (fig. 8A). This exceptionally long preservation of FeS on timescales of 103 yr is dueto the extreme stratification within the water column and the effective separation ofintermediate reduced S species from FeS production. Additionally, the absence ofpore-water H2S inhibits the transformation of FeS to FeS2 is observed in the sulfidicsediments of Effingham Inlet and the Black Sea margin. H2S levels are undetectablewithin the sediments partly due to the low rates of sulfate reduction. Of course, lowsulfide concentrations in the pore waters also hinder the production of appreciableintermediate S regardless of the availability of oxidants.

Fig. 11. Schematic model of pyrite formation in the Orca Basin.


Sheu and Presley (1986) examined the Orca Basin on a lamina-by-lamina basis.They argued that the black and gray layers exhibit differing levels of FeS transformationand organic carbon concentration. The gray layers display lower OC values and a higherextent of FeS to FeS2 transformation, which they attributed to periods of enhancedmixing at the brine-seawater interface brought about by winter storms. The enhancedmixing might encourage the oxidation of Fe2� and H2S and the production of intermedi-ate S species. By contrast, they attributed the high concentrations of organic carbon andthe negligible transformation of FeS to FeS2 to spring productivity events which supplylarger amounts of organic matter to the brine-seawater interface. The present study didnot examine the sediments at equivalent resolution. Using a 2 cm sampling scheme wefound only one interval (8–10 cm) with significant transformation of FeS to FeS2. Thisinterval may represent a higher concentration of the gray muds due to enhanced mixingwithin the transition zone as reported by Sheu and Presley (1986). However, it is unclearto us why winter storms would affect such deep waters.

CONCLUSIONS

Effingham Inlet and the analogous Black Sea margin and the Orca Basin define twofundamentally different environments of anoxic marine deposition which, despite thedifferences, are marked by similar persistence of appreciable FeS to depths of tens ofcentimeters but contrasting levels of FeS2 and H2S production. Effingham Inlet and theBlack Sea margin are characterized by rapid accumulation rates, high rates of sulfatereduction, and, as a consequence, moderate to very high H2S values which persist todepth. The resulting sediments contain concentrations of sulfur as FeS atypically high formarine sediments but show significant downcore transformation of FeS to FeS2. TheOrca Basin is characterized by an extreme water-column stratification, low accumulationrates, moderate to low rates of sulfate reduction, and levels of H2S within the sedimentsthat are below detection. Orca Basin sediments contain a high concentration of FeS withonly minor pyrite values.

FeS persists to depth in these environments for different but related reasons. Inanoxic marine systems, polysulfides and other intermediate reduced sulfur speciescannot exist in appreciable quantities below the redox boundary and therefore decreasein importance for iron sulfide transformation. In Effingham Inlet and the Black Sea, rapiddeposition favors FeS preservation through rapid removal of FeS from the redoxboundary in the water column and, most importantly, from the intermediate S specieswhose formation initiates with H2S oxidation. Additionally, high burial rates furtherisolate FeS from the intermediate S species and enhance the delivery of reactive Fe andorganic carbon to the subsurface, which encourages sulfate reduction and perhaps someFeS formation at depth within the sediments. In the Orca Basin, FeS is also decoupledfrom intermediate reduced sulfur species but as a result of an extreme stratificationwithin the water column which prevents communication between the two species.Under conditions promoting the spatial decoupling (vertical separation) of FeS andappreciable levels of intermediate S, H2S may play a direct role in the transformation ofFeS to FeS2 over extended diagenetic timescales (101 to 102� yr).

ACKNOWLEDGMENTS

Financial support for this project was provided by NSF (E. Ingall). We thank thecrews of the R/V Barnes and R/V Longhorn. L. Clark, M. Clementz, and P. VanCappellen shared in the sample collection and data generation and were sources ofvaluable discussions. R. Wilkin, L. Benning, C. Kelley, and R. Glaser provided insightfulcomments on a preliminary draft of this manuscript. We are grateful to R. Raiswell andH. Passier for insightful reviews.


REFERENCES

Addy, S. K., and Behrens, E. W., 1980, Time of accumulation of hypersaline anoxic brine in Orca Basin (Gulfof Mexico): Marine Geology, v. 37, p. 241–252.

Albert, D. B., Taylor, C. D., and Martens, C. S., 1995, Sulfate reduction rates and low molecular weight fattyacid concentrations in the water column and surficial sediments of the Black Sea: Deep-Sea Research,v. 42, p. 1239–1260.

Aller, R. C., 1980, Quantifying solute distributions in the bioturbated zone of marine sediments by defining anaverage microenvironment: Geochimica et Cosmochimica Acta, v. 44, p. 1955–1965.

Aller, R. C., and Rude, P. D., 1988, Complete oxidation of solid phase sulfides by manganese and bacteria inanoxic marine sediments: Geochimica et Cosmochimica Acta, v. 52, p. 751–765.

Anderson, R. F., Lyons, T. W., and Cowie, G. L., 1994, Sedimentary record of a shoaling of the oxic/anoxicinterface in the Black Sea: Marine Geology, v. 116, p. 373–384.

Baumgartner, T., Crusius, J., Thomson, R., Holmgren, D., Francis, R., Soutar, A., Ferreira, V., and Field, J.,1999, A sedimentary record of small pelagic fish abundances from the west coast of Vancouver Island: inpress.

Behrens, E. W., 1980, On sedimentation rates and porosity: Marine Geology, v. 35, p. M11–M16.Benning, L. G., and Barnes, H. L., 1998, In-situ determination of the stability of iron monosulphides and

kinetics of pyrite formation: Mineralogical Magazine, v. 62A, p. 151–152.Berner, R. A., 1964, Iron sulfides formed from aqueous solution at low temperatures and atmospheric pressure:

Journal of Geology, v. 72, p. 293–306.————— 1970, Sedimentary pyrite formation: American Journal of Science, v. 268, p. 1–23.————— 1978, Sulfate reduction and the rate of deposition of marine sediments: Earth and Planetary Science

Letters, v. 37, p. 492–498.————— 1980, Early Diagenesis: A Theoretical Approach: Princeton, New Jersey, Princeton University Press,

241 p.————— 1984, Sedimentary pyrite formation: an update: Geochimica et Cosmochimica Acta, v. 48, p. 605–615.————— 1987, Models for carbon and sulfur cycles and atmospheric oxygen: application to Paleozoic geologic

history: American Journal of Science, v. 287, p. 177–196.Berner, R. A., Baldwin, T., and Holdren, G. R., Jr., 1979, Authigenic iron sulfides as paleosalinity indicators:

Journal of Sedimentary Petrology, v. 49, p. 1345–1350.Berner, R. A., and Raiswell, R., 1983, Burial of organic carbon and pyrite sulfur in sediments over Phanerozoic

time: a new theory: Geochimica et Cosmochimica Acta, v. 47, p. 855–862.Boesen, C., and Postma, D., 1988, Pyrite formation in anoxic environments of the Baltic: American Journal of

Science, v. 285, p. 193–206.Brock, T. D., and Gustafson, J., 1976, Ferric iron reduction by sulfur- and iron-oxidizing bacteria: Applied and

Environmental Microbiology, v. 32, p. 567–571.Calvert, S. E., and Karlin, R. E., 1991, Relationships between sulphur, organic carbon, and iron in the modern

sediments of the Black Sea: Geochimica et Cosmochimica Acta, v. 55, p. 2483–2490.Canfield, D. E., ms 1988, Sulfate reduction and the diagenesis of iron in anoxic marine sediments: Ph.D.

dissertation, Yale University, New Haven, Connecticut.————— 1989, Reactive iron in marine sediments: Geochimica et Cosmochimica Acta, v. 53, p. 619–632.————— 1991, Sulfate reduction in deep-sea sediments: American Journal of Science, v. 291, p. 177–188.Canfield, D. E., Lyons, T. W., and Raiswell, R., 1996, A model for iron deposition to euxinic Black Sea

sediments: American Journal of Science, v. 296, p. 818–834.Canfield, D. E., Raiswell, R., and Bottrell, S., 1992, The reactivity of sedimentary iron minerals toward sulfide:

American Journal of Science, v. 292, p. 659–683.Canfield, D. E., Raiswell, R., Westrich, J. T., Reaves, C. M., and Berner, R. A., 1986, The use of chromium

reduction in the analysis of reduced inorganic sulfur in sediments and shales: Chemical Geology, v. 54,p. 149–155.

Canfield, D. E., Thamdrup, B., and Fleischer, S., 1998, Isotope fractionation and sulfur metabolism by pureand enrichment cultures of elemental sulfur-disproportionating bacteria: Limnology and Oceanography,v. 43, p. 253–264.

Chanton, J. P., and Martens, C. S., 1985, The effects of heat and stannous chloride addition on the activedistillation of acid volatile sulfide from pyrite-rich marine sediment samples: Biogeochemistry, v. 1,p. 375–383.

Cline, J. D., 1969, Spectrophotometric determination of hydrogen sulfide in natural water: Limnology andOceanography, v. 14, p. 454–458.

Cloke, P. L., 1963, The geologic role of polysulfides-Part I: Geochimica et Cosmochimica Acta, v. 27,p. 1265–1298.

Crusius, J., and Anderson, R. F., 1991, Immobility of 210Pb in Black Sea sediments: Geochimica etCosmochimica Acta, v. 55, p. 327–333.

Drobner, E., Huber, H. J., Wachterhauser, G., Rose, D., and Stetter, K. O., 1990, Pyrite formation linked withhydrogen evolution under anaerobic conditions: Nature, v. 246, p. 742–744.

Gagnon, C., Mucci, A., and Pelletier, E., 1995, Anomalous accumulation of acid-volatile sulphides (AVS) in acoastal marine sediment, Saguenay Fjord, Canada: Geochimica et Cosmochimica Acta, v. 59, p. 2663–2675.

Goldhaber, M. B., and Kaplan, I. R., 1974, The sulfur cycle, in Goldberg, E. D., editor, The Sea, Volume 5:New York, Wiley-Interscience, p. 569–655.

Henneke, E., Luther, G. W., III, and De Lange, G. J., 1991, Determination of inorganic sulphur speciation withpolarographic techniques: some preliminary results for recent hypersaline anoxic sediments: MarineGeology, v. 100, p. 115–123.

Henneke, E., Luther, G. W., III, De Lange, G. J., and Hoefs, J., 1997, Sulphur speciation in anoxic hypersalinesediments from the eastern Mediterranean Sea: Geochimica et Cosmochimica Acta, v. 61, p. 307–321.


Holland, H. D., 1973, Systematics of the isotopic composition of sulfur in the oceans during the Phanerozoicand its implications for atmospheric oxygen: Geochimica et Cosmochimica Acta, v. 37, p. 2605–2616.

————— 1984, The Chemical Evolution of the Atmosphere and Oceans: Princeton, New Jersey, PrincetonUniversity Press, 582 p.

Howarth, R. W., 1979, Pyrite: Its rapid formation in a salt marsh and its importance in ecosystem metabolism:Science, v. 203, p. 49–51.

Howarth, R. W., and Jorgensen, B. B., 1984, Formation of 35S-labelled elemental sulfur and pyrite in coastalmarine sediments (Limfjorden and Kysing Fjord, Denmark) during short-term 35SO4

2� reduction measure-ments: Geochimica et Cosmochimica Acta, v. 48, p. 1807–1818.

Huffman, E. W. D., Jr., 1977, Performance of a new carbon dioxide coulometer: Microchemical Journal, v. 22,p. 567–573.

Jannasch, H. W., Truper, H. G., and Tuttle, J. H., 1974, The microbial sulfur cycle of the Black Sea, in Degens,E. T. and Ross, D. A., editors, The Black Sea-Geology, Chemistry and Biology: Tulsa, Oklahoma,American Association of Petroleum Geologists, p. 419–425.

Jørgensen, B. B., 1977, The sulfur cycle of a coastal marine sediment (Limfjorden, Denmark): Limnology andOceanography, v. 22, p. 814–832.

————— 1978, A comparison of methods for the quantification of bacterial sulfate reduction in coastal marinesediments. I. Measurement with radiotracer techniques: Geomicrobiology Journal, v. 1, p. 11–28.

Jørgensen, B. B., Fossing, H., Wirsen, C. O., and Jannasch, H. W., 1991, Sulfide oxidation in the anoxic BlackSea chemocline: Deep-Sea Research, v. 41, p. 531–557.

Karl, D. M., LaRock, P. A., and Shultz, D. J., 1977, Adenosine triphosphate and organic carbon in the CariacoTrench: Deep-Sea Research, v. 24, p. 105–113.

Kennett, J. P., and Penrose, N. L., 1978, The occurrence of fossil Holocene seaweed and attached calcareouspolychaetes in an anoxic basin, Gulf of Mexico: Nature, v. 276, p. 172.

Klinkhammer, G. P., and Lambert, C. E., 1989, Preservation of organic matter during salinity excursions:Nature, v. 339, p. 271–274.

Kump, L. R., and Garrels, R. M., 1986, Modeling atmospheric O2 in the global sedimentary redox cycle:American Journal of Science, v. 286, p. 337–360.

LaRock, P. A., Lauer, R. D., Schwartz, J. R., Watanabe, K. K., and Wiesenburg, D. A., 1979, Microbial biomassand activity distribution in an anoxic, hypersaline basin: Applied and Environmental Microbiology, v. 37,p. 466–470.

Lewis, B. L., and Landing, W. M., 1991, The biogeochemistry of manganese and iron in the Black Sea:Deep-Sea Research, v. 38, p. S773–S803.

Lin, S., and Morse, J. W., 1991, Sulfate reduction and iron sulfide mineral formation in Gulf of Mexico anoxicsediments: American Journal of Science, v. 291, p. 55–89.

Luther, G. W., III, Church, T. M., Giblin, A. E., and Howarth, R. W., 1986, Speciation of dissolved sulfur in saltmarshes by polarographic methods, in Sohn, editor, Organic Marine Geochemistry; American ChemicalSociety Symposium Series 305, p. 339–355.

Luther, G. W., III., 1991, Pyrite synthesis via polysulfide compounds: Geochimica et Cosmochimica Acta,v. 55, p. 2839–2849.

Lyons, T. W., 1991, Upper Holocene sediments of the Black Sea: Summary of Leg 4 box cores (1988 Black SeaOceanographic Expedition), in Idzar, E., and Murray, J. W., editors, Black Sea Oceanography: Boston,Massachusetts, Kluwer Academic Publishers, p. 401–441.

————— 1992, Contrasting sediment types from oxic and anoxic sites of the modern Black Sea: Geochemical andsedimentological criteria: Ph.D. dissertation, Yale University, New Haven, Connecticut.

————— 1997, Sulfur isotopic trends and pathways of iron sulfide formation in upper Holocene sediments of theanoxic Black Sea: Geochimica et Cosmochimica Acta, v. 61, p. 3367–3382.

Lyons, T. W., and Berner, R. A., 1992, Carbon-sulfur-iron systematics of the uppermost deep-water sedimentsof the Black Sea: Chemical Geology, v. 99, p. 1–27.

McKee, T. R., Jeffrey, L. M., Presley, B. J., and Whitehouse, U. G., 1978, Holocene sediment geochemistry ofcontinental slope and intraslope basin areas, northwest Gulf of Mexico, in Bouma, A. H., Moore, G. T.,and Coleman, J. M., editors, Framework, Facies, and Oil-Trapping Characteristics of the Upper Continen-tal Margin, American Association Petroleum Geologist Studies in Geology, no. 7, p. 313–326.

Middelburg, J. J., 1989, A simple rate model for organic matter decomposition: Geochimica et CosmochimicaActa, v. 53, p. 1577–1581.

————— 1991, Organic carbon, sulphur, and iron in recent semi-euxinic sediments of Kau Bay, Indonesia:Geochimica et Cosmochimica Acta, v. 55, p. 815–828.

Moore, W. S., and O’Neill, D. J., 1991, Radionuclide distributions in recent Black Sea sediments, in Idzar, E.,and Murray, J. W., editors, Black Sea Oceanography: Boston, Massachusetts, Kluwer Academic Publish-ers, p. 343–359.

Morse, J. W., and Cornwell, J. C., 1987, Analysis and distribution of iron sulfide minerals in recent anoxicmarine sediments: Marine Chemistry, v. 22, p. 55–69.

Morse, J. W., Millero, F. J., Cornwell, J. C., and Rickard, D., 1987, The chemistry of the hydrogen sulfide andiron sulfide systems in natural waters: Earth Science Reviews, v. 24, p. 1–42.

Pyzik, A. J., and Sommer, S. E., 1981, Sedimentary iron monosulfides: kinetics and mechanism of formation:Geochimica et Cosmochimica Acta, v. 45, p. 687–698.

Raiswell, R., and Berner, R. A., 1985, Pyrite formation in euxinic and semi-euxinic sediments: AmericanJournal of Science, v. 285, p. 710–724.

Raiswell, R., Buckley, F., Berner, R. A., and Anderson, T. F., 1988, Degree of pyritization of iron as apaleoenvironmental indicator of bottom-water oxygenation: Journal of Sedimentary Petrology, v. 58,p. 812–819.


Raiswell, R., and Canfield, D. E., 1996, Rates of reaction between silicate iron and dissolved sulfide in PeruMargin sediments: Geochimica et Cosmochimica Acta, v. 60, p. 2777–2787.

————— 1998, Sources of iron for pyrite formation in marine sediments: American Journal of Science, v. 298,p. 219–245.

Raiswell, R., Canfield, D. E., and Berner, R. A., 1994, A comparison of iron extraction methods for thedetermination of degree of pyritisation and the recognition of iron-limited pyrite formation: ChemicalGeology, v. 111, p. 101–110.

Rickard, D. T., 1969, The chemistry of iron sulphide formation at low temperatures: Stockholm, Contributionsto Geology, v. 20, p. 67–95.

————— 1975, Kinetics and mechanism of pyrite formation at low temperatures: American Journal of Science,v. 275, p. 636–652.

————— 1997, Kinetics of pyrite formation by the H2S oxidation of iron (II) monosulfide in aqueous solutionsbetween 25 and 125°C: The rate equation: Geochimica et Cosmochimica Acta, v. 61, p. 115–134.

Rickard, D., and Luther, G. W., III., 1997, Kinetics of pyrite formation by the H2S oxidation of iron(II)monosulfide in aqueous solutions between 25 and 125°C: The mechanism: Geochimica et CosmochimicaActa, v. 61, p. 135–147.

Schoonen, M. A. A., and Barnes, H. L., 1991a, Reactions forming pyrite and marcasite from Solution: I.Nucleation of FeS2 below 100°C: Geochimica et Cosmochimica Acta, v. 55, p. 1495–1504.

————— 1991b, Reactions forming pyrite and marcasite from solution: II. Via FeS precursors below 100°C:Geochimica et Cosmochimica Acta, v. 55, p. 1505–1514.

Sheu, D.-D., 1987, Sulfur and organic carbon contents in sediment cores from the Tyro and Orca Basins:Marine Geology, v. 75, p. 157–164.

————— 1990, The anoxic Orca Basin (Gulf of Mexico): Geochemistry of brines and sediments: Reviews inAquatic Science, v. 2, p. 491–507.

Sheu, D.-D., and Presley, B. J., 1986, Variations of calcium carbonate, organic carbon and iron sulfides inanoxic sediment from the Orca Basin, northern Gulf of Mexico: Marine Geology, v. 70, p. 103–118.

Sheu, D.-D., Shakur, A., Pigott, J. D., Wiesenburg, D. A., Brooks, J. M., and Krouse, H. R., 1987, Sulfur andoxygen isotopic compositions of dissolved sulfate in the Orca Basin: implications for origin of thehigh-salinity brine and oxidation of sulfides at the brine-seawater interface: Marine Geology, v. 78,p. 303–310.

Shokes, R. F., Trabant, P. K., Presley, B. J., and Reid, D. R., 1977, Anoxic, hypersaline basin in the northernGulf of Mexico: Science, v. 196, p. 1443–1446.

Stookey, L. L., 1970, Ferrozine—a new spectrophotometric reagent for iron: Analytical Chemistry, v. 42,p. 779–781.

Taylor, P., Rummery, T. E., and Owen, D. G., 1979, Reactions of iron monosulfide solids with aqueoushydrogen sulfide up to 160°C: Journal of Inorganic Nuclear Chemistry, v. 41, p. 1683–1687.

Teder, A., 1971, The equilibrium between elementary sulfur and aqueous polysulfide solutions: Acta Chem.Scand., v. 25, p. 1711–1728.

Thode-Andersen, S., and Jørgensen, B. B., 1989, Sulfate reduction and the formation of 35S-labeled FeS, FeS2and S0 in coastal marine sediment: Limnology and Oceanography, v. 34, p. 793–806.

Toth, D. J., and Lerman, A., 1977, Organic matter reactivity and sedimentation rates in the ocean: AmericanJournal of Science, v. 277, p. 265–285.

Trabant, P. K., and Presley, B. J., 1978, Orca Basin, anoxic depression on the continental slope, northwest Gulfof Mexico, in Bouma, A. H., Moore, G. T., and Coleman, J. M., editors, Framework, Facies, andOil-Trapping Characteristics of the Upper Continental Margin: American Association Petroleum Geolo-gists Studies in Geology, no. 7, p. 303–316.

Trefry, J. H., Presley, B. J., Keeney-Kennicutt, W. L., and Trocine, R. P., 1984, Distribution and chemistry ofmanganese, iron, and suspended particulates in Orca Basin: Geo-Marine Letters, v. 4, p. 125–130.

Van Cappellen, P., Viollier, E., Roychoudhury, A., Clark, L., Ingall, E., Lowe, K., and DiChristina, T., 1998,Biogeochemical cycles of manganese and iron at the oxic-anoxic transition of a stratified marine basin(Orca Basin, Gulf of Mexico): Environmental Science and Technology, 1998, v. 32, p. 2931–2939.

Westrich, J. T., and Berner, R. A., 1984, The role of sedimentary organic matter in bacterial sulfate reduction:the G model tested: Limnology and Oceanography, v. 29, p. 236–249.

Wiesenburg, D. A., Brooks, J. M., and Bernard, B. B., 1985, Biogenic hydrocarbon gases and sulfate reductionin the Orca Basin brine: Geochimica et Cosmochimica Acta, v. 49, p. 2069–2080.

Wilkin, R. T., and Barnes, H. L., 1996, Pyrite formation by reactions of iron monosulfides with dissolvedinorganic and organic sulfur species: Geochimica et Cosmochimica Acta, v. 60, p. 4167–4179.

Wilkin, R. T., and Barnes, H. L., 1997, Pyrite formation in an anoxic estuarine basin: American Journal ofScience, v. 297, p. 620–650.

Matthew T. Hurtgen and others588

[A J S ., 1999, P. 556–588] ANOMALOUS ENRICHMENTS OF IRON...

Documents

Transcript of [A J S ., 1999, P. 556–588] ANOMALOUS ENRICHMENTS OF IRON...