SEDIMENTARY ORGANIC MATTER PRESERVATION: A TEST FOR SELECTIVE DEGRADATION UNDER OXIC CONDITIONS

8/14/2019 SEDIMENTARY ORGANIC MATTER PRESERVATION: A TEST FOR SELECTIVE DEGRADATION UNDER OXIC CONDITIONS

1/27

SEDIMENTARY ORGANIC MATTER PRESERVATION: A TEST FORSELECTIVE DEGRADATION UNDER OXIC CONDITIONS

JOHN I. HEDGES*, FENG SHENG HU**, ALLAN H. DEVOL*,HILAIRY E. HARTNETT*, ELIZABETH TSAMAKIS*, and RICHARD G. KEIL*

ABSTRACT. We report here a test of the hypothesis that the extent of organicmatter preservation in continental margin sediments is controlled by the averageperiod accumulating particles reside in oxic porewater immediately beneath thewater/sediment interface. Oxygen penetration depths, organic element composi-tions, and mineral surface areas were determined for 16 sediment cores collectedalong an offshore transect across the Washington continental shelf, slope, andadjacent Cascadia Basin. Individual amino acid, sugar, and pollen distributionswere analyzed for a 11 to 12 cm horizon from each core, and 14C-based sedimentaccumulation rates and stable carbon isotope compositions were determinedfrom depth profiles within a subset of six cores from representative sites. Sedi-ment accumulation rates decreased, and dissolved O2 penetration depths in-creased offshore along the sampling transect. As a result, oxygen exposure times(OET) increased seaward from decades (mid-shelf and upper slope) to more thana thousand years (outer Cascadia Basin). Organic contents and compositions wereessentially constant within individual sediment cores but varied consistently withlocation. In particular, organic carbon/surface area ratios decreased progres-sively offshore and with increasing OET. Three independent compositionalparameters demonstrated that the remnant organic matter in farther offshoresediments is more degraded. Both concentration and compositional patternsindicated that sedimentary organic matter exhibits a distinct and reproducibleoxic effect. OET helps integrate and explain organic matter preservation inaccumulating continental margin sediments and hence provides a useful tool forassessing transfer of organic matter from the biosphere to the geosphere.

INTRODUCTION

Burial of organic matter in marine sediments directly links the global cycles ofcarbon, oxygen, and sulfur over geologic time (Berner, 1982, 1989). The organic matterpreserved in sedimentary deposits is important as a progenitor of fossil fuels, a recorderof Earth history, and the ultimate source of essentially all atmospheric O2 (Berner and

Canfield, 1989; Engel and Macko, 1993; Hunt, 1996). Although continental marginsaccount for only 10 percent of total ocean area and 20 percent of total ocean primaryproduction (Killops and Killops, 1993), greater than 90 percent of all organic carbonburial occurs in sediments depositing on deltas, continental shelves, and upper continen-tal slopes (Berner, 1989). Thus, to understand the processes and environmental condi-tions that control organic matter burial on a global basis, it is fundamentally important tostudy preservation along continental margins.

The question of what variables control organic matter preservation in marinesediments is one of the most complex and controversial issues in contemporarybiogeochemistry (Berner, 1980; Calvert and Pedersen, 1992; Canfield, 1994; Emersonand Hedges, 1988; Hedges and Keil, 1995; Henrichs, 1992; Henrichs and Reeburgh,1987). Suggested important factors include primary production rate (Calvert and Peder-sen, 1992), water column depth (Smith, 1978; Suess, 1980), organic matter sources(Hedges, Clark, and Cowie, 1988; Schubert and Stein, 1996) and reaction histories

(Berner and Westrich, 1985), sediment transport processes (Luckge and others, 1996;Pedersen, Shimmield, and Price, 1992), sediment accumulation rate (Henrichs, 1992;Muller and Suess, 1979), bottom water oxygen concentration (Betts and Holland, 1991;Demaison and Moore, 1980; Richards and Redfield, 1954), availability of O2 (Cai and

* School of Oceanography, Box 357940, University of Washington, Seattle, Washington 98195-7940** Department of Plant Biology, 265 Morrill Hall, University of Illinois, 505 South Goodwin Avenue,

Urbanna, Illinois 61801

[AMERICAN JOURNAL OF SCIENCE, VOL. 299, SEPT., OCT., NOV., 1999, P. 529555]

529


2/27

Reimers, 1995; Hulthe, Hulth, and Hall, 1998; Reimers and others, 1992) and otherdissolved electron acceptors (Brandes and Devol, 1995; Soetaert, Herman, and Middel-burg, 1996), cometabolization (Canfield, 1994), microbial dynamics (Lee, 1992; Smith,Walsh, and Jahnke, 1992), mixing and irrigation by macrobenthos (Aller, 1982), redoxfluctuations (Aller, 1994; Aller and others, 1996), and sorption (Keil and others, 1994a;Mayer, 1994a,b; Ransom and others, 1998). Identifying direct causative effects onsedimentary organic carbon preservation that can be assigned to a particular process orenvironmental condition is particularly challenging because many of the above factorsare interdependent and covary complexly in patterns that shift nonlinearly over time(Middelburg, Klug, and van der Nat, 1993) and space (Jahnke, 1996; Reimers and others,1992).

Here we report a specific test of the hypothesis that degradation under oxicconditions controls long-term preservation of organic matter in sediments accumulatingalong continental margins (Hedges and Keil, 1995). It has long been recognized thatopen ocean sediments depositing beneath deep water columns are poor in organicmatter, with typical mass-normalized percentages of organic carbon (percent OC) in therange of 0.1 to 0.3 (Premuzic and others, 1982). These organic concentrations are

roughly 10 percent of those observed in sediments of similar grain size depositing alongcontinental margins, a pattern that also holds when OC contents of sediments arenormalized to specific surface-area, as opposed to mass (Hedges and Keil, 1995; Weilerand Mills, 1965). As a result of this extreme depletion in OC, coupled with slow overallaccumulation rates, pelagic sediments contribute less than 5 percent of modern-dayglobal organic carbon preservation (Berner, 1989).

While there are many explanations for the low OC contents of offshore sediments,surface-oxidized marine turbidites demonstrate that long-term exposure to molecularoxygen is sufficient alone to cause such depletions. Numerous deep-sea turbidites havenow been described (Thomson and others, 1993). The relict f-turbidite from the MadeiraAbyssal Plain (MAP) region off Northwest Africa (de Lange, Jarvis, and Kuijpers, 1987;Weaver and Rothwell, 1987), however, has been best characterized organically (Prahland others, 1989, 1997). This initially homogeneous turbidite was emplaced about140,000 yrs before present and then subjected to in situ oxidation for the first10,000

yrs after deposition (Buckley and Cranston, 1988). O2 eventually diffused 0.5 m deepinto the 4-m thick deposit. Comparative analyses of sedimentary horizons above(oxidized, 0.1-0.2 percent OC) and below (unoxidized, 0.9-1.0 percent OC) the redoxboundary in two separate MAP cores indicated that80 percent of the original OC(Cowie and others, 1995) and all optically recognizable pollen (Keil and others, 1994b)were destroyed during long-term oxic exposure. Organic matter below the redoxboundary was moderately degraded, as is typical of upper continental margin deposits(Cowie and others, 1995). In contrast, organic materials in the oxidized interval wereheavily degraded and compositionally resembled mixtures from deep-sea sediments(Cowie and Hedges, 1994; Hedges and Keil, 1995).

Based on parallel offshore trends in pollen and organic contents along the continen-tal margin of Washington State (Keil and others, 1994c, 1998) and other regions(Premuzic and others, 1982; Reimers and others, 1992), it has been hypothesized thatoxic degradation also limits organic matter preservation in incrementally-depositing

surface marine sediments (Hedges and Keil, 1995; Reimers, 1989). The logic behind thispremise is that the average time of exposure of sedimentary organic matter to oxicconditions in the surface of depositing sediments corresponds roughly to the depth ofpenetration of O2 into porewaters divided by the average sediment accumulation rate atthe site (Hedges and Keil, 1995). Because O2 penetration depths generally increaseoffshore, as sediment accumulation rates become slower, it follows that in situ O2exposure times should increase farther offshore. In analogy to radiochemical decay, a

John I. Hedges and othersSedimentary organic matter preservation530


3/27

linear increase in time of exposure to porewater O2 should cause an exponentialdecrease in the OC content of the sediment accumulating below the oxic surface horizon(Hedges and Keil, 1995). Thus, offshore decreases in sedimentary organic content alongcontinental margins should be pronounced. This hypothesis was recently supported bythe observation that the burial efficiencies of OC in surface marine sediments accumulat-ing off the coasts of Washington, California, and Mexico decrease sharply as O2exposure times increase from days to millennia (Hartnett and others, 1998). This firstpublication, however, did not include information on the depositional settings, sources,vertical profiles, and surface area loadings of the sedimentary organic matter. Morecritically, molecular and pollen data were not presented to test whether organic remainsin more slowly depositing offshore sediments are in fact more degraded. We present herea comprehensive study of organic matter degradation in sediments depositing along theWashington State margin and discuss the implications of strong evidence for moreextensive degradation in the presence of O2.

STUDY SITE

Patterns of sediment distribution along the convergent plate margin of WashingtonState generally parallel the coastline in response to local input and transport processes(Gross and others, 1972). The southern continental shelf (0-200 m water depth) is 25 to60 km wide and incised by marine canyons (fig. 1). The mid-shelf (100 m) is covered bya silt deposit that separates OC-poor modern sands inshore and relict sands offshore(Nittrouer and Sternberg, 1981). Sediments in this region derive primarily from theColumbia River (White, 1970) and are carried northward parallel to the shore byresuspension during winter storms (Kachel and Smith, 1989). Finer silt- and clay-sizeparticles are winnowed out and transported offshore to the continental slope andCascadia Basin (fig. 1). The local continental slope (200-2500 m) is contacted by a

Fig. 1. Washington margin study region. Circled numbers indicate that14C-based sediment accumulationrates were determined for that site.

a test for selective degradation under oxic conditions 531


4/27

pronounced O2 minimum (15 percent saturation) between water depths of500 to2000 m (Carpenter, 1987). Slope sediments in this region are relatively rich in organiccarbon (2 percent OC) and accumulate more slowly than on the adjacent shelf(Carpenter and Peterson, 1989). Surface sediments of Cascadia Basin (2500-3000 m)are gradually depositing hemipelagic muds that contain less pollen and bulk organiccarbon than texturally similar deposits on the upper slope (Hedges and Keil, 1995).Cascadia Basin is bound about 500 km to the west by the Juan de Fuca Ridge, beyondwhich surface sediments deposit slowly and contain low organic carbon contents (0.1-0.3percent OC) typical of open ocean deposits (Gross and others, 1972).

SAMPLE COLLECTION AND ANALYSIS

Sediments (fig. 1) were collected July 20 to August 17, 1994 during Cruise 94-07B ofthe R/V Wecoma (Lambourn, Hartnett, and Devol, 1996). A 20 30 cm Soutar boxcorer was used to recover sediment blocks up to 0.7 m deep that were then subsampledon deck with 10-cm diam core tubes. Sediments were extruded from core tubes in aN2-filled glove box and subsampled at 0.5 cm intervals from 0 to 2 cm, 1.0 cm intervalsfrom 2 to 10 cm, 2.0 cm intervals to 25 cm, and at 3.0 cm for deeper intervals. The bulksubsamples, along with small portions for 14C analysis, were immediately frozen.Dissolved oxygen profiles were determined by the whole-core squeezing method ofLambourn, Devol, and Murray (1991), using an in-line polarographic oxygen microelec-trode. Oxygen penetration depths into sediment porewaters were calculated (relativeprecision 15 percent) as described by Brandes and Devol (1995). Sediment porositieswere determined by water loss upon drying at 60C, assuming a sediment density of2.4 g/cm3 and a porewater salinity of 35 percent.

Nitrogen-specific mineral surface areas were determined using the one-point BETmethod on a Quantachrome Monosorb surface area analyzer. Prior to determination,sediments were cleaned of organic matter via hydrogen peroxide treatment (Mayer,1994a; Keil and others, 1994c). The average precision for the samples reported here was3 permil.

Elemental compositions were measured with a Carlo Erba model 1106 CHN

analyzer (Hedges and Stern, 1984). Weight percentages of organic carbon (percent OC)were determined (relative precision 2 percent) following vapor phase acidification.Weight percentages of total nitrogen (percent TN) and total carbon were determinedsimilarly, but without acidification. Weight percentages of inorganic carbon (percent IC)were calculated as the difference between total carbon and organic carbon. The 14Ccontents of all bulk sedimentary organic samples were determined at the National OceanSciences AMS Facility (Woods Hole Oceanographic Institution). Samples were com-busted (after acidification) to CO2, which was converted to graphite and analyzed alongwith NBS Oxalic Acid I and II standards (plus process blanks). Counting errors in thereported 14C contents (app. 1) were estimated by NOSAMS and averaged2 percent ofthe measured age. The stable carbon isotopic composition in a fraction of the sameCO2 was determined by NOSAMS and is reported here as the permil deviation (13C)from the PDB standard material (Stuiver and Polach, 1977). This measurement has aprecision of0.10.2 permil.

Pollen analyses were made on 1 cm3 of sediment taken from the 11 to 12 cm horizonof selected cores. All sediments were sieved through a 7-m mesh screen to remove fineparticles, after which pollen was isolated followed standard techniques (Faegri andothers, 1989). A known amount of Eucalyptus pollen was added as an internal standardto each sample to determine natural pollen concentrations. Pollen grains in up to fiveslides of each sample were counted in oil immersion under 400x and 1000x magnifica-tion. Pollen sums ranged from 15 to 250 grains. Individual grains were classified as



5/27

degraded if they exhibited any of the types of physical deterioration described byCushing (1967).

Neutral sugars (aldoses) were analyzed by the method of Cowie and Hedges (1984).Dry samples were pretreated with 12 M H2SO4 at room temperature for 2 hrs to facilitateglucose yield from cellulose. The acid was then diluted to 1.2 M and used to hydrolyzeindividual samples for 3 hrs at 100C. Hydrolysate solutions were neutralized to a pH of6.5 with Ba(OH)2 and centrifuged to remove BaSO4. The supernatant was deionized bypassage through a column of mixed anion-cation exchange resins and rotoevaporatednearly to dryness. The residue was dissolved in pyridine and anomerically equilibratedwith LiClO4 catalyst at 60C for 48 hrs. The equilibrated aldoses were converted totrimethylsilyl derivatives with BSTFA (Regis Chemical Company) and quantified with aprecision of5 to 10 percent by gas chromatography on a 30 m by 0.25 mm I.D. quartzcapillary column coated with DB-1 (J&W Scientific) methylsilicone liquid phase.

Amino acids were measured by high-pressure liquid chromatography using charged-matched recovery standards and the general procedure described by Cowie and Hedges(1992a). Aqueous hydrolyses were done in 6 N HCl under N 2 for 70 min at 150C. Thehydrolysis mixture was dried, dissolved in water, and the component amino acids wereconverted to fluorescent o-phthaldialdehyde (OPA) derivatives with a Gilson model 231automated injector. The OPA derivatives were injected after 1 min of reaction onto a15 cm 4.6 mm-I.D. column operated in reverse-phase mode with 5-m C18 packing.Using this method, individual amino acids can be measured with a precision of5 to 10percent.

RESULTS AND DISCUSSION

To test definitively the hypothesis that degradation under oxic conditions exerts amajor control on organic matter preservation in surface sediments off the WashingtonState coast, it is necessary to: (1) directly measure varying oxygen exposure times (OET)at representative sites across the local continental margin, (2) determine (and properlyexpress) the corresponding sedimentary organic matter concentrations, (3) demonstratethat the sedimentary deposits being compared contain similar types of preserved organicmatter and are vertically uniform, and (4) test whether any decreases in organic

concentrations that might attend increases in OET are supported by independentmeasurements of advanced degradation. These issues will be sequentially addressed inthe following sections.

Offshore Trends in Physical Characteristics of SedimentsThe southern Washington State margin (fig. 1) provides a uniformly varying coastal

environment well suited to the goal of relating organic preservation to physical setting.Nine sediment cores were collected from a region of the Washington continental shelfand slope that lies between 46.4 and 48.8N latitude and extends20 to 50 km offshore.Seven additional cores were taken along a transect that extended another 100 kmoffshore through Cascadia Basin along approximately 46.7N latitude. Water depths atthe sites where the shelf/slope cores were collected decreased almost linearly offshorefrom 100 to 2000 m, whereas depths along the basin sequence deepened graduallyfrom2500 to 2750 m (fig. 2A). This shelf/slope and basin terminology will be used

hereafter, with the latter set of deepwater samples being indicated in figures by bold fonts(note that core numbers are not sequential with depth).

In the following discussion and illustrations most geographic trends are presented interms of distance offshore, as opposed to water depth. Distance was chosen simplybecause it corresponds more straightforwardly to a chart. The same results could havebeen presented as a function of water depth. Given local bathymetry ( fig. 2A), the effectof a depth-based format would be to stretch trends among nearshore stations and



6/27

Fig. 2. Physical properties of sediment core samples: (A) Water depth versus distance offshore, (B) O2penetration depth versus distance offshore.



7/27

Fig. 2(C) Average sediment accumulation rate versus distance offshore, and (D) Calculated oxygenexposure time (OET) versus distance offshore.



8/27

compress those for offshore samples. None of the following major trends, or theirinterpretation, would be appreciably altered by normalization to depth versus distanceoffshore.

Oxygen penetration depths generally increase offshore (fig. 2B), from values of0.5to 1.0 cm on the mid-shelf (Cores 4 and 8) and continental slope (Cores 11, 13, 2, 15, 6,20, and 3), to a range of1.5 to 3.5 cm in Cascadia basin (Cores 5, 23, 19, 9, 17, 1 and16). Core 16 has an unusually low O2 penetration depth for its extreme offshore location,possibly because of recent local deposition of labile organic matter. The overall offshoreincrease in O2 penetration below the water/sediment interface likely results from thecombined effects of lower primary production farther off the Washington coast (Perry,Bolger, and English, 1989) and greater attenuation of organic particles raining throughlonger water columns (Martin and others, 1987). Impingement of the oxygen minimumzone between500 to 2000 m water depth (Carpenter, 1987) may explain the unusuallylow O2 penetration depths measured in Cores 13, 6, 15, 20, and 3 from the upper slope.

Sediment accumulation rates were determined for the six cores listed in table 2 andcircled in figure 1. Because 4 to 7 measurements of14C were required per core, additionaldating was prohibitively expensive for this study. The six pro filed cores were selected onthe basis of their uniformity and greater length, with an eye toward representing a rangeof water depths and distances offshore. Although 210Pb profiles were also measured foralmost all cores, the calculated accumulation rates were invariably greater than 14C-based rates (when available) for the same deposits. Due to mixing and the relatively shorthalf-life of 210Pb, this difference became greater with increasing distance offshore. Inslowly accumulation sediments from Cascadia Basin, 210Pb-based values often exceeded14C-counterparts by more than two orders of magnitude.

14C-based sediment accumulation rates decrease progressively offshore (fig. 2C)from average values near 15 cm/ky on the upper slope (Cores 13 and 15) to 3 cm/ky inthe outer basin (Core 1). This seaward decrease is expected from previous studies in thisregion (Carpenter, Peterson, and Bennett, 1982; Carson, 1971). All three of the deepestcores (17, 1, and 16) from the outer basin penetrated a gray clay horizon at a depth of 15to 20 cm (W94-B07 cruise log) whose age (at the surface) is on the order of 9000 to 13,000yrs (Carson, 1973). The deepest dated sample (29-30 cm) from Core 1 was from withinthe gray clay horizon (about 10 cm below the interface) and had a14C age of 11,500 years(app. 1). A 15 to 20 cm glacial/post-glacial boundary with an age of 10,000 yrscorresponds to an average sediment accumulation rate of roughly 2 cm/ky. This value isclose to the mean rate of 3 cm/ky determined from the 14CprofileforCore 1 (table2) andto deposition rates (2 cm/ky) reported for this offshore region by Carson (1973). Thus,14C appears to be a reliable radionuclide for determining comparative average sedimentaccumulation rates within this study region (fig. 1).

Oxygen exposure times, calculated by dividing the previously discussed O2 penetra-tion depths by the corresponding 14C-based sediment accumulation rates, increaseexponentially away from the coast (fig. 2D). Calculated OET values range from decadeson the slope, to centuries at the inner basin, up to a millenium (Station 1) in outerCascadia Basin. The overall increase by a factor of about 30 results from a 6 offshoreincrease in O2 penetration and a corresponding 5 decrease in accumulation rate.Because these two contributing trends are nearly linear (fig. 2B,C), OET data for the

other 10 (undated) cores should follow a general trend similar to that in figure 2D.Neither O2 penetration depth nor accumulation rate alone explains the observedincrease in OET, a parameter that integrates both in situ particle dynamics as well as O2supply and demand (Hartnett and others, 1998).

Offshore Trends in Organic ContentWeight percentage of organic carbon is the traditional measurement of organic

matter content in both sediments and soils (Hedges and Oades, 1997). Full profiles



9/27

(12-19 samples per core) of percent OC, percent TN, and percent IC were determinedfor each of the 16 Washington Margin sediment sequences, amounting to almost 250analyses of each element overall. Inorganic carbon was a trace component (overallaverage of 0.2 wt percent) of these sediments and will not be discussed further. As istypical for much of the Washington margin (Carpenter, 1987; Hedges and Mann, 1979),depth profiles of percent OC and percent TN were relatively featureless, except for thefew cores that penetrated OC-poor glacial gray clay at depth. The percent standarddeviations in percent OC and percent TN among sediments (excluding gray clays) fromindividual cores averaged 11 and 22 percent, respectively, with the greater variabil-ity in percent TN resulting largely from the challenge of measuring its lower concentra-tion. Due to this overall vertical uniformity, only mean percent OC, percent TN and(C/N)a values are reported (and plotted) for each core (table 1).

Average mass-normalized organic contents of the Washington margin sedimentcores ranged from about 1 to 3 percent OC and 0.1 to 0.3 percent TN (table 1) and fall inthe range of earlier analyses (Gross and others, 1972; Hedges and Mann, 1979; Prahl andothers, 1994). Because percent TN closely parallels percent OC (later discussion), we willfocus here on the more abundant element. An offshore decrease in percent OC isdiscernable in figure 3A. No sediment farther than 100 km from shore has values1.5 wtpercent. Cores 4, 8, and 2 exhibit lower organic contents than the other near-shoredeposits (fig. 3A). Sediments from these same three cores (table 1) also exhibit unusuallylow (0.7) porosites and surface areas (20 m2 g), indicating that their component grainsare relatively coarse. Because (A) most organic matter in sediments from the Washingtonmargin is associated with mineral surfaces (Hartnett and others, 1998; Keil and others,1994a,c), and (B) coarse minerals have small surface/mass ratios (Mayer, 1994a,b), thelow percent OC values of these samples probably result from their coarser textures.

Surface area-normalized organic loadings (mgOC/m2) are less susceptible totextural influences than mass-normalized counterparts and in this case yield a smoothoffshore decrease in organic content (fig. 3B). In particular, Cores 8 and 4 are shown tobe the most organic-rich on a surface area basis. A plot of OC/SA versus distanceoffshore reveals a break near 100 km (2500 m water depth) that is consistent with otherstudies in this region (Carpenter and Peterson, 1989; Prahl, 1985). Organic loadingsdecrease rapidly offshore from the mid-shelf and across the slope but then diminishgradually along the floor of Cascadia Basin (fig. 3B). Most OC/SA ratios of the slopesamples span the range of 0.5 to 1.0 mgOC/m2, as is typical of non-deltaic uppercontinental margin sediments (Keil and others, 1994c; Mayer 1994a,b). In contrast, allfive cores from outer Cascadia Basin have average OC/SA ratios near 0.3, whichindicate substantially reduced preservation efficiencies. Representative profiles of OC/SAdown the six 14C-dated cores (fig. 4) illustrate the uniformity of organic carbon loadingbelow the upper few centimeters of sediment in these cores, although the deepestanalyzed horizons represent ages since deposition of roughly 2000 to 8000 yrs. Thisuniformity suggests minimal organic matter degradation below the upper few centime-ters of sediment (see later discussion). Such results can be considered robust, however,only if the sources of organic matter preserved in these deposits is demonstrably similarand independent evidence can be presented for an offshore increase in diageneticalteration.

Sedimentary Organic Matter SourcesStable carbon isotope compositions of the organic matter in the six dated Washing-

ton margin cores provide information on the both the origin and spatial uniformity of theassociated organic matter. The variability of13C values for organic matter in differenthorizons of each dated sediment core averaged 0.2 permil, which is close to analyticalreproducibility. Over the six separate sites (fig. 1), 13C averages of individual coresranged only from22.4 to23.0 permil, with no evidence of an offshore trend (table 2).



10/27

Fig. 3. Average organic carbon contents of individual sediment cores given as (A) Weight percentages,%OC, versus distance offshore, and (B) Organic surface area loadings, OC/SA, versus distance offshore.



11/27

Fig.4.

VerticalprofilesofOC/SAdownthesixcore

sthatwere

14C-dated(fig.

1,app1).Age11.5

representsthecalculatedage(yrssincedepositio

n)

ofthe11to

12cm

sedimenthorizonroutinelyusedform

olecular-levelbiochemicalanalyses.



12/27

Thus, the organic mixtures in these sediments are remarkably uniform in their stablecarbon isotope composition, and hence likely in their percentages of marine andterrestrial organic matter. While assignments of endmember 13C values to marinesedimentary mixtures are dangerous (Hedges and Prahl, 1993), respective averages of21.5 and 25.5 permil for marine- and terrestrially-derived organic matter arereasonably well established for the Washington margin (Hedges and Mann, 1979; Prahland others, 1994). Given these boundary conditions, sediments from the dated coresappear to comprise roughly two-thirds marine and one-third terrestrial organic matter(see also Prahl and others, 1994).

Atomic C/N ratios of the 16 sediment cores fall into two groups (table 1). Cores 4, 8,and 15 from the mid-shelf and upper slope have (C/N)a ratios between 17 and 18(mean 17.3 0.3). The (C/N)a ratios of the other 13 cores fall without pattern in therange of 10 to 13 (mean 11.8 1.0) across the entire slope-basin span. Prahl andothers (1994) reported similar (C/N)a ratios for surface sediments from this region. Aplot of core-average percent OC versus percent TN illustrates the close relationship ofthese two elements. Best-fit regression lines to the averaged elemental data are illustratedin figure 5 for the three shallow (percent OC 14.7*percent TN 0.03, r2 0.998) and13 other cores (percent OC 9.28*percent TN 0.13, r2 0.926). Both regressionlines have intercepts near the origin, indicating that inorganic nitrogen is a minorcomponent of TN, such that the numeric mean and slope-derived (C/N)a values arerepresentative of the bulk of the organic matter in these two size classes (Bergamaschiand others, 1997).

The unusually high (C/N)a values of the three shallow-water sediments may beassociated with their depositional settings. Cores 4 and 8 are from the southernWashington mid-shelf silt deposit, whose sediments are characterized by relatively high

Fig. 5. Average weight percentage organic carbon (%OC) versus weight percentage total nitrogen (%TN)for all Washington margin sediment cores.



13/27

concentrations of vascular plant debris (Hedges and Mann, 1979). These deposits areknown to exhibit higher average (C/N)a values (15) than more offshore slope and basin(means 9-11) sediments (Prahl and others, 1994). Core 15 is from a submarine channelassociated with the Columbia River (figs. 1, 2A) and hence also may preferentiallyreceive vascular plant debris (Gross and others, 1972; Harmon, 1972). Size fractionationof surface sediments from the Washington shelf and slope (Keil and others, 1994c; 1998)demonstrate that the terrigenous organic components of sands and silts are enriched invascular plant remains [(C/N)a 30], whereas more nitrogen-rich soil organic matter[(C/N)a 15] is the major terrigenous component of clays. Local vascular plant debrisand soil organic matter have similarly light (13C 25 to 26 permil) stable carbonisotope compositions (Keil and others, 1994c). Thus, by slightly increasing the ratio ofwoody debris to soil organic matter within a constant total amount of terrigenous organicmaterial, it is possible to elevate the (C/N)a of the nearshore sediments withoutappreciably changing their 13C values (for example, Core 15). Overall, the predomi-nantly marine-derived organic mixtures from the Washington margin sediments (tables1 and 2) appear sufficiently similar in organic composition to allow a reasonablecomparison of potential diagenetic influences throughout this region of contrasting OET.

Offshore Trends in Organic Matter CompositionIf in fact the offshore drop in OC/SA (fig. 3B) is due to cumulative in situ

degradation near the sediment surface, then the biochemical compositions of these samesediments should provide independent evidence of advanced organic alteration at moreoffshore sites. To investigate this issue the amino acid and neutral sugar (aldose)compositions of sediments collected from the 11 to 12 cm horizon of each core weredetermined by hydrolysis and molecular level analyses. This intermediate sedimentinterval was chosen to be below the depths of O2 penetration and rapid particle mixing,

TABLE 1

Locations and physical characteristics of sediment samples from the Washington Statecontinental margin

Originally designated core numbers (in bold) are 200 units higher (for example, 201, 202 . . .), Spl# the number of samples used to calculate whole core averages, * Indicates values averaged for whole core,shadedrows correspond to 14C dated cores, porosity, IC inorganic carbon, OC organic carbon, TNtotal nitrogen, C/N(a) atomic organic carbon/total nitrogen (calculated directly from tabulated meanpercentOC and percent TN), Spl. sample, SA surface area, Penet. penetration.



14/27

without being so far removed from the surface that comparisons to current O2 penetra-tion depths and sedimentation rates would be unduly compromised. Correspondingdepth profiles of sugars and amino acids were also determined within selected cores toexamine the effects of prolonged in situ degradation that could bias comparisons amongcores of different ages. Amino acids and aldoses were chosen because both biochemicaltypes derive primarily from marine sources and can provide useful diagenetic informa-tion (Cowie and Hedges, 1994; Cowie and others, 1995).

Non-protein amino acids provide one of the most reliable biochemically-basedindicators of organic matter degradation. In particular, Cowie and others (1995) demon-strated that pronounced elevations in the mole percentages of-alanine plus -aminobu-tyric acid, percent (BALA GABA), occur when sedimentary mixtures are subjected toslow oxic degradation in deep-sea turbidites. In addition, percent (BALA GABA) has

proven useful for comparing degradation stages in rivers (Hedges and others, 1994,submitted) and coastal marine environments (Cowie, Hedges, and Calvert, 1992b; Keiland others, 1998). A plot of percent (BALA GABA) versus distance offshore (fig. 6A)demonstrates a highly significant linear increase, [percent (BALA GABA) km*0.0291 1.64, r2 0.90]. In comparison to the intercept of 1.6 percent at zerodistance offshore, fresh organic matter has a percent (BALA GABA) near 1 (Cowieand Hedges, 1994). To test for progressive in situ degradation, depth profiles of percent(BALA GABA) were measured at six horizons each in Cores 9 and 19. The total rangeof variability within both these Cascadia Basin cores was less than 10 percent of thecorresponding mean (fig. 6A) and hence comparable in magnitude to analytical preci-sion. Minimal vertical gradients in bioactive materials are typical of sediment cores fromthe Washington margin (Carpenter, 1987; Hedges and Mann, 1979), partially as a resultof physical mixing of biologically active surficial sediments (Carpenter, Peterson, andBennett, 1982). Because 210Pb-based mixing depths average about 5 cm in these cores

(unpublished data), it is difficult to determine where in the upper 5 cm or so of thesedeposits their diagenetic signatures are imprinted.

Glucose weight percentages (percent GLC) among total aldoses generally decreasewith organic matter degradation, whereas percentages of the 2 six-carbon deoxy sugars,rhamnose (RHA) and fucose (FUC), increase (Cowie, Hedges, and Calvert, 1992;Hamilton and Hedges, 1988; Hernes and others, 1996). Both these trends are observedwith increasing distance from the Washington coast (fig. 6B). All three derivative

TABLE 2

Average13C values, sediment accumulation rates and oxygen exposure times for six 14C-datedsediment cores

Uncertainty intervals are standard deviations for 13C (defined in text), standard errors in line fitting foraccumulation rates, and calculated standard deviations for O2 exposure time (using the listed errors in theaccumulation rates and an estimated 15% uncertainty in oxygen penetration depth measurements).



15/27

parameters, percent (FUC RHA), 100/wt percent GLC, and (FUC RHA)/GLC,increase toward deeper water. The simplest of these parameters, 100/percent GLC, has avalue near 1.5 to 2.5 for fresh plankton and vascular plant tissues (Cowie and Hedges,1984; Hernes and others, 1996). Values of 100/percent GLC increase sharply offshorefrom near 3 (30 wt percent glucose) for sediments on the shelf and inner slope to 4.5 to5.5 (20 wt percent glucose) for those from outer Cascadia Basin (fig. 6B). Theobservation that 100/wt percent GLC increases more sharply on the shelf and innerslope than does percent (BALA GABA) demonstrates that different indicator com-pounds can provide contrasting sensitivities to progressive diagenetic alteration (Cowieand Hedges, 1994; Wakeham and others, 1997).

The variability of 100/wt percent GLC within a 6-sample sequence from Core 19(fig. 6B) is comparable to analytical precision. Thus, in situ carbohydrate alteration also isnot apparent, even within the biologically active upper portion of this sediment core.The overall lack of biochemical evidence for progressive degradation (fig. 6), or OC loss(for example, fig. 4), throughout the varying lengths of the profiled cores indicates that itis valid in this case to compare whole-core averages among sediment sequences ofdifferent average age (table 2 and fig. 4). As would be expected if exposure to oxic

conditions is critical, the diagenetic differences evident in figures 3 and 6 appear to beimprinted in the well-mixed upper few centimeters of the sediments. Although fermenta-tion and sulfate reduction are possible throughout the top meter of Washington margindeposits, these processes do not occur sufficiently fast to leave discernable depletions (fig.4) or chemical imprints over the time intervals represented by the lengths of the14C-dated cores (2000-8000 yrs). This evidence for minimal deep degradation is consis-tent as well with the general observation that ancient shales exhibit organic carboncontents similar to those of modern fine-grained marine sediments that have accumu-lated below the diagenetically active upper 5 to 10 cm (Hunt, 1996).

Pollen affords an independent (nonchemical) method for assessing the relativeextents to which organic materials in these sediments have been exposed to diageneticalteration. Previous research has shown that pollen is sensitive to long-term oxicdegradation and completely disappears under extreme conditions of O2 exposure inoxidized deep-sea turbidites (Keil and others, 1994b). This observation is striking

because pollen walls, sporopollenins, are among the most resistant biological remainsknown and persist for millions of years in anoxic sediment and peat deposits (Brooks andShaw, 1978). Pollen grains, however, are sorted differentially during sedimentarytransport and tend to concentrate with silt-size mineral grains (Keil and others, 1998). Forthis reason, and because of variable inputs over time, it is difficult in degradation studiesto assess changes in pollen abundances per unit volume of bulk sediment. It is feasible,however, to apply the percentage of physically degraded pollen grains in a sample(Cushing, 1967) as a qualitative degradation indicator.

Pollen was counted in the 11 to 12 cm intervals of nine of the Washington marginsediment cores (fig. 6C). Pollen grains were observed in all the analyzed sedimentsamples. While a detailed description of pollen compositions is beyond the scope of thispaper, the assemblages were primarily of mixed conifer types (Pinaceae, Pinus, Picea,Tsuga) along with abundant alder (Alnus), as is typical of the Paci fic Northwest regionover the last 6000 yrs (Heuser, 1985). The four continental shelf and slope samples

contained 100,000 to 250,000 pollen grains/cm3 of sediment, whereas all five sedimentsfrom Cascadia Basin contained 20,000 grains/cm3. The most prevalent forms ofphysical alteration exhibited by the sedimentary pollen grains were diffuse surfaceerosion, pitting, and perforation. The observed percentages of physically altered grainsincreased from values near 20 percent on the continental slope to 40 to 60 percent inCascadia Basin. The observation that the basin sediments contained lower concentra-tions of more corroded pollen suggests that these recalcitrant terrestrially-derived



16/27

Fig. 6. Distance from shore of individual sediment cores versus average degradation indicators,including: (A) Mole percent of-alanine plus -aminobutyric acid, %(BALA GABA); (B) One hundredtimes the inverse of weight percent glucose, 100/%GLC; and



17/27

particles are also subject to severe in situ sedimentary degradation. It seems unlikely thatmore degraded pollen would be subject to greater offshore transport or that suchextensive degradation would occur during relatively brief periods of marine transport.Long term degradation on land, however, could account for the lower threshold of 20

percent degradation observed in all cores. Nevertheless, pollen adds a third, unique lineof compositional evidence in support of an in situ oxic effect in Washington coastsediments.

OVERVIEW AND IMPLICATIONS

While the evidence presented is consistent with the hypothesis that the lowerorganic contents of more offshore Washington margin sediments (fig. 3A, B) result fromlonger in situ degradation under oxic conditions (fig. 2D), a number of overriding pointsand interpretive constraints merit further discussion.

Relationships between organic parameters and with environmental variables otherthan distance offshore have not been stressed to this point. It is mechanisticallyinformative, however, to test directly how well the amino acid- and carbohydrate-baseddiagenetic parameters correlate with surface-normalized organic content. Plots of OC/SAversus percent (BALAGABA) and 100/percent GLC (fig. 7A and 7B, respectively)

illustrate that the average organic contents of the Washington margin cores decrease withincreasing degradation. Most of the discernableorganic matter alteration and loss (figs. 3and 6), however, occurs over a relatively short distance on the shelf/slope (20-50 kmoffshore, 100-2500 m water depth), as opposed to along the 100-km long sequence ofcores on the floor of Cacadia Basin. In contrast, OET increases more among surfacesediments from Cascadia Basin (fig. 2D). The apparent insensitivity of degradationextent to OET values greater than 200 yrs is particularly evident in a plot of OC/SA

Fig. 6(C) Percent of degraded pollen.



18/27

Fig. 7. Organic carbon/surface area ratios (OC/SA) plotted versus: (A) %(BALA GABA), (B) 100/%GLC, and



19/27

versus log of OET (fig. 7C), where higher carbon loadings persist in the most offshoresediments (Core 1) than would be expected for simple first-order decay.

These long-tailed decreases in organic matter versus time are the rule, rather than

the exception, for sediments of many types and ages (Boudreau, 1997; Westrich andBerner, 1984). Such ever-slowing kinetics can be variously described with multi-Gmodels for sums of different organic components degrading with contrastingfirst-orderrates (Berner, 1980, 1981; Jrgensen, 1978) or as power (Middelburg, 1989) orreaction continuum models (Boudreau and Ruddick, 1991), in which remineralizationrate constants decrease as a function of time (Tarutis, 1993). The present data set is toosmall and geographically diverse to backward project multiple organic componentswith characteristically different reaction rates (Van Liew, 1962). Even these six data,however, indicate that the OC/SA trend is, as theoretically expected (Hedges and Keil,1995), exponential in form. The best least-squares fit of the data (excluding Core 1)indicates a half-life for oxic organic matter remineralization on the order of 100 yrs (fig.7C, assumingfirst-order degradation). This dynamic, however, is largely constrained bythe time scale represented by the shelf/slope cores (Middelburg, Vlug and van der Nat,1993) and does not rule out faster (or slower) kinetics (Boudreau and Ruddick, 1991). In

fact, Hartnett and others (1998) recently reported a negative exponential relationshipbetween sedimentary organic carbon burial efficiency and OET that extends fromcenturies to days. In contrast to previous speculation that OETs beyond a threshold ofcenturies to millennia might be needed to affect organic matter preservation along theWashington margin (Hedges and Keil, 1995), oxic effects are now apparent atexposure times of decades or less (Hartnett and others, 1998). Thus, selective degrada-tion under oxic conditions need not be slow.

Fig. 7(C) log of OET. See figure 6 for definitions of the two organic parameters.



20/27

Uncertainties about the physical history of particulate materials comprising thesesediments necessarily cascade into parallel ambiguities as to the sites and mechanisms oftheir alteration. For example, it is not clear when and how degradation characteristic ofmore offshore sedimentary organic matter (figs. 3, 6) occurred. Since the componentmineral particles are predominantly clastic, they must (like pollen) be derived eventuallyfrom land. After being carried to the coastal ocean by rivers or wind, these particles mayhave been subject to multiple deposition/resuspension cycles in the course of offshoretransport to their present sites (Nittrouer and Sternberg, 1981). Thus, their total history ofexposure to O2 may be substantially greater than would be estimated from their final(sampled) depositional site alone. Such multi-step transport should lead to more ad-vanced degradation at more offshore sites, and thus produce degradation trends thatwould be difficult to discriminate from simple aoxic effect at one point of deposition.

Several characteristics of sedimentary organic matter transport and degradationwithin this (and other) depositional region, however, should mitigate large cumulativeeffects of sequential transport. First, the sediments that comprise most of our data setwere sampled from water depths 500 m (fig. 2A) where periods of sufficient energy forsediment resuspension should be relatively rare. This notion of relatively quiescentdeposition is supported by uniformly high total surface areas (fig. 7) and porosities (table1), indicative of a paucity of sand in all sediments from depths greater than 500 m.Second, extensive resuspension and offshore relocation of sedimentary particles along abenthic transport pathway should result in relatively constant deposition rates, or in theextreme, more extensive offshore accumulation. In contrast, sediments of our studyregion accumulate at rates that decrease progressively offshore (fig. 2C) and to depthsthat are consistent with gradual accumulation at relatively constant rates (see 14C results).Since Cascadia Basin is bound seaward by a continuous ocean ridge, sedimentaryparticles are not likely to abridge this depositional record by escaping farther offshore.Moreover, the OET values calculated for a particular site will only be directly affected ifresuspension erodes particles deeper than the oxic sediment horizon and if the resus-pended particles are then transported to a deposition site with an appreciably differentO2 penetration depth or average accumulation rate. The relatively uniform compositions

down the sediment cores in this study (figs. 4 and 6), and the smooth geographic trendsamong individual cores (figs. 2, 3, 6, and 7), all point toward relatively uniformaccumulation histories.

Bioturbation and irrigation can complicate interpretations of oxygen exposurehistories as well. Calculated oxygen exposure times should not be greatly affected bybioturbation, which (if nonselective) should move particles into and out of the oxygen-ated surface horizon of marine sediments with comparable probabilities. Under condi-tions such as off the Washington coast, however, where bioturbation intervals can besubstantially deeper than O2 penetration depths, the number of times an individualparticle passes between oxic and anoxic zones will be much greater with bioturbationthan without (Aller, 1994). Periodic irrigation of anoxic porewaters with oxygenatedbottom water should have the same local effect. Given evidence that re-exposure oforganic matter from anoxic sediments to oxygenated conditions facilitates biodegrada-

tion (Hulthe, Hulth, and Hall, 1998), OET calculations based on steady state conditionsmay substantially underestimate cumulative degradation under oscillating redox condi-tions (Aller, 1994). Thus periodic resuspension/redeposition and bioturbation/irrigationevents will both cause OET to conservatively estimate total biodegradation history. Theobservation that OET provides a useful guideline for the extent of cumulative degrada-tion in Washington margin sediments (fig 7C) suggests, but does not prove, that average



21/27

depositional conditions predominate over scattered periodic events in determining theamounts and types of organic materials that are preserved. Significantly, all thesescenarios focus primarily on conditions and processes near the sediment/water interface,as opposed to those occurring near the surface of the overlying water column.

The issue of where oxic degradation stops on the Washington margin is presentlyunresolved. Although the organic matter accumulating on the floor of Cascadia Basin ismeasurably altered (figs. 3, 6, and 7), it may not yet have reached full degradationmaturity. One observation supporting this notion is that the average percent OC (1.1)and OC/SA (0.3 mgOC/m2) of Cascadia Basin sediments are distinctly higher than thecorresponding contents (percent OC 0.1-0.2; OC/SA 0.01-0.10 mgOC/m2) typicalof open ocean deposits (Hedges and Keil, 1995). In fact, Gross and others (1972) reportpercent OC values less than 0.5 percent on the Tufts Plain to the west of the Juan de FucaRidge. In addition, all the Cascadia Basin sediments contain ample pollen in contrast tolittle or no sporopollenin in sediments from the oxidized MAP turbidite horizon andopen ocean (Keil and others, 1994a). Finally, the few deep-water sediments that havebeen analyzed yield much higher abundances of non-protein amino acids (Lee, Wake-ham, and Hedges, submitted; Whelan, 1977), a sensitive compositional indicator ofadvanced oxic degradation (Cowie and others, 1995). The limited seaward extent ofCascadia Basin, which is bound west of the sampling transect by the Juan de Fuca ridgecrest system (fig. 1), interrupts the typical offshore decrease in productivity, sedimentaccumulation rate, and hence OET, that would characterize most continental margins.Analyses of sediments from a transect extending farther offshore from an unbound landmass will be necessary to test for more severe sedimentary degradation that might attendporewater O2 exposure on time scales of thousands to hundreds of thousands of years(Middelburg, Vlug and van der Nat, 1993).

To compare absolute extents of degradation among different sediments and deposi-tional sites, consistent relationships between organic content and alteration are needed.Given the varying sensitivities of biochemical-based degradation indicators over differ-ent stages (and types?) of alteration (fig. 7), multiple indicators of diagenetic stage may benecessary. A plot of percent OC versus SA (fig. 8), however, does illustrate a potentially

useful pattern of organic matter loss and alteration at higher values of percent(BALA GABA) corresponding to greater OETs. In this format, OC/SA values ofWashington margin sediments sweep down clockwise with increasing degradationalong rays whose lengths depend on specific surface areas. While it is premature toattempt a rigorous fit, percent OC values for a given surface area do decrease as percent(BALA GABA) values increase. Thus surface area and diagenetic state can becombined to constrain the organic content of these sediments. This relationship may wellextrapolate downward to the low OC/SA and high percentages of non-protein aminoacids typical of extremely organic-poor open ocean sediments. Because the percent(BALA GABA) values of the upper shelf/slope sediments are already near 1, as istypical of fresh organic matter, another parameter will be necessary to extend thisrelationship to sediments with OC/SA values greater than one (Hartnett and others,1998; Hedges and Keil, 1995).

Related to the issue of where organic degradation stops is the question of howvalid it is to compare degradation state among sediment cores of different average ages,which result from analyzing comparable depth intervals among deposits accumulating atcontrasting average rates (fig. 2C). If appreciable degradation continues below thebioturbated surface zone of different deposits, then organic materials in more slowlyaccumulating sediments will be more degraded simply because they are on average



22/27

older, as opposed to having initially passed through the surface oxic zone more slowly.Inspection of OC/SA ratios down the six cores chosen for 14C dating (fig. 4), however,

indicates little evidence for consistent in situ degradation below the upper few centime-ters of sediment. This evidence for minimal in situ degradation below the surfacebioturbated layer is supported by the narrow range of variability and lack of adiscernable downward increase, in the diagenetic parameters measured over the lengthsof cores 9 and 19 (fig. 6A,B). Such featureless profiles are typical of the other cores in thisstudy (table 1) and most sediments from the Washington margin (Hedges and Mann,1979; Prahl and others, 1994). The general observation that shales and modern marinesediments of similar texture both contain an average of about 1 weight percent of organiccarbon (Hunt, 1996) argues against extensive organic matter losses from sedimentarydeposits on geologic time scales (Hedges and Keil, 1995). It appears, therefore, that theprocesses that imprint the amount and types of organic preserved in these deposits occurpredominantly within the upper few centimeters of these sediment and thus are largelyindependent of the overall length of the studied cores.

The mechanisms ofoxic degradation indicated by the previous relationships are

not yet clear. While oxic conditions appear to be necessary in sedimentary porewatersfor extensive degradation to occur, this relationship does not necessarily mean that O 2 isthe causative electron acceptor or enzyme intermediate. In fact, the vertical separationbetween oxic, suboxic, and anoxic conditions in these continental margin sediments is sonarrow (and fluctuating) that any number of compounds could act as redox agents andcatalysts. It is also likely that specific microorganisms enzymatically facilitate key

Fig. 8. A degradation fan plot of %OC versus specific surface area, SA, for average data from all 16Washington margin cores and other reference samples. The corresponding trend in %(BALAGABA) is thebest-fit line to the given diagenetic arc.



23/27

electron exchanges, and that some macrobenthic organisms are especially effective increating physical conditions conducive to these reactions (Aller, 1994). Thus oxicdegradation is more appropriately viewed as a set of as yet unknown processes thatcharacteristically prevail under sedimentary conditions where O2 is present, as opposedto a specific mechanism that directly involves molecular oxygen. Oxygen exposure timesimply incorporates benthic conditions and dynamics into a parameter that predicts theextent of organic matter preservation near the sediment/water interface where the leakof OC from the biosphere to geosphere is directly controlled. Future mechanistic insightsinto preservation processes will no doubt lead to new and better predictors.

ACKNOWLEDGMENTS

We thank the Captain and crew of the R/V Wecoma for their assistance andpatience during Cruise 94-07B. Dean Lambourn, Crystal Thimsen, Jay Brandes, and ElliTsamakis helped with many aspects of sample collection and analysis. Constructivecomments by Drs. Robert Aller and Marc Alperin substantially improved this manu-script. This manuscript was completed while JH was a Fellow of the Hanse Wissen-schaftskolleg, Delmenhorst, Germany. Our research was supported by NSF grant OCE

9401903 to JH and OCE 9116275 to AD and OCE-9402081 to RK.

APPENDIX 1

Isotope and age characteristics of individual sediment horizons from the six14C-datedsediment cores

Core#

SedimentDepth (cm)

13C()PDB

Age(yr BP)

Age Error(yr BP)

1 2.5 22.7 4470 551 5.5 22.9 4590 351 7.5 22.9 4680 401 9.5 23.0 5220 401 11.5 22.9 5510 301 14.0 22.9 6100 651 29.5 23.4 11500 603 1.8 22.4 1600 353 7.5 22.2 2150 253 11.5 22.4 2370 303 17.0 22.8 2400 553 29.0 22.3 2660 253 43.0 22.1 2590 359 2.5 23.3 4280 659 9.5 22.5 5110 309 18.0 22.8 6750 409 29.5 22.9 9270 459 39.5 23.1 10550 40

13 4.5 24.2 1490 5513 11.5 22.5 1390 4513 22.0 22.1 2210 5013 32.5 22.2 2710 5515 1.8 22.9 785 4515 7.5 23.0 805 2515 14.0 22.9 1370 2015 18.0 22.9 1740 2515 22.0 n.d. 1640 3015 32.5 22.3 1610 2519 1.8 22.5 3090 9519 5.5 22.3 3280 8019 18.0 22.4 4570 6019 29.0 22.7 6330 9519 43.0 24.8 8150 45

Bold data indicated core horizons used for sediment accumulate rate calculations.



24/27

REFERENCESAller, R. C., 1982, The effects of macrobenthos on chemical properties of marine sediment and overlying

water, in McCall, P. L., and Tevesz, M. J. S., editors, Animal-Sediment Relations: New York, Plenum,p. 53102.

1994, Bioturbation and remineralization of sedimentary organic matter: effects of redox oscillation:

Chemical Geology, v. 114, p. 331345.Aller, R. C., Blair, N. E., Xia, Q., and Rude, P. D., 1996, Remineralization rates, recycling, and storage of

carbon in Amazon shelf sediments: Continental Shelf Research, v. 16, p. 753786.Bergamaschi, B. A., Tsamakis, E., Keil, R. G., Eglinton, T. I., Montlucon, D. B., and Hedges, J. I., 1997, The

effect of grain size and surface area on organic matter, lignin and carbohydrate concentrations andmolecular compositions in Peru Margin sediments:Geochimicaet Cosmochimica Acta, v. 61,p. 12471260.

Berner, R. A., 1980, Early Diagenesis: A Theoretical Approach: Princeton, New Jersey, Princeton UniversityPress, 231 p.

1981, A rate model for organic matter decomposition during bacterial sulfate reduction in marinesediments, inDumas,R., editor, Biogeochimie de la Matiere Organique a lInterface Eau-Sediment Marin:Colloques Internationaux du C. N. R. S. no. 293, p. 3544.

1982, Burial of organic carbon and pyrite sulfur in the modern ocean: Its geochemical and environmen-tal significance: American Journal of Science, v. 282, p. 451473.

1989, Biogeochemical cycles of carbon and sulfur and their effect on atmospheric oxygen overPhanerozoic time: Palaeogeography, Palaeoclimatology, and Palaeoecology, v. 73, p. 97122.

Berner, R. A., and Canfield, D. E., 1989, A new model for atmospheric oxygen over Phanerozoic time:American Journal of Science, v. 289, p. 333361.

Berner, R. A., and Westrich, J. T., 1985, The role of sedimentary organic matter in bacterial sulfate reduction:the G model tested: Limnology and Oceanography, v. 29, p. 236249.

Betts, J. N., and Holland, H. D., 1991, The oxygen content of ocean bottom waters, the burial efficiency oforganic carbon, and the regulation of atmospheric oxygen: Palaeogeography, Palaeoclimatology, andPalaeoecology, v. 97, p. 518.

Boudreau, B. P., 1997, Diagenetic Models and Their Implementation: Modelling Transport and Reactions inAquatic Sediments: Berlin, Springer-Verlag, 414 p.

Boudreau, B. P., and Ruddick, B. R., 1991, On a reactive continuum representation of organic matterdiagenesis: American Journal of Science, v. 291, p. 507538.

Brandes, J. A., and Devol, A. H., 1995, Simultaneous nitrate and oxygen respiration in coastal sediments:Evidence for discrete diagenesis: Journal of Marine Research, v. 53, p. 771797.

Brooks, J., and Shaw, G., 1978, Sporopollenin: a review of its chemistry, palaeochemistry and geochemistry:Grana, v. 17, p. 9197.

Buckley, D. E., and Cranston, R. E., 1988, Early diagenesis in deep-sea turbidites: The imprint of paleo-oxidation zones: Geochimica et Cosmochimica Acta, v. 52, p. 29252939.

Cai, W-J., and Reimers, C. E., 1995, Benthic oxygen flux, bottom water oxygen concentration and core toporganic carbon content in the deep northeast Pacific Ocean: Deep-Sea Research, v. 42, p. 16811995.

Calvert, S. E., and Pedersen, T. F., 1992, Organic carbon accumulation and preservation in marine sediments:How important is anoxia? in Whelan, J., and Farrington, J. W., editors, Organic Matter: New York,University Press, p. 231263.

Canfield, D. E., 1989, Sulfate reduction and oxic respiration in marine sediments: implications for organic

carbon preservation in euxinic environments: Deep-Sea Research, v. 6: p. 121138. 1994, Factors influencing organic carbon preservation in marine sediments: Chemical Geology, v. 114,p. 315329.

Carpenter, R., 1987, Has man altered the cycling of nutrients and organic C on the Washington continentalshelf and slope? Deep-Sea Research, v. 34, 881896.

Carpenter, R., and Peterson, M. L., 1989, Chemical cycling in Washingtons coastal zone, inLaundry, M. R.,and Hickey, B. M., editors, Coastal Oceanography of Washington and Oregon: New York, Elsevier,p. 367509.

Carpenter, R. C., Peterson, M. L., and Bennett, J. T., 1982, 210Pb derived sediment accumulation and mixingrates for the Washington continental slope: Marine Geology, v. 48, p. 135164.

Carson, B., ms, 1971, Stratigraphy and depositional history of Quaternary sediments in northern CascadiaBasin and Juan de Fuca abyssal plain, North Pacific Ocean: Ph.D. thesis: Department of Oceanography,University of Washington, Seattle.

Carson, B., 1973, Acoustic stratigraphy, structure, and history of Quaternary deposition in Cascadia Basin:Deep-Sea Research, v. 20, p. 387396.

Cowie, G. L., and Hedges, J. I., 1984, Carbohydrate sources in a coastal marine environment: Geochimica etCosmochimica Acta, v. 48, p. 20752087.

1992a, Improved quantification of amino acids in natural samples: charge-matched recovery standardsand reduced analysis time: Marine Chemistry, v. 37, p. 223238.

1992b, Sources and reactivities of amino acids in a coastal marine environment: Limnology andOceanography, v. 37, p. 703724. 1994, Biochemical indicators of diagenetic alteration in natural organic matter mixtures: Nature, v. 369,

p. 304307.Cowie, G. L., Hedges, J. I., and Calvert, S. E., 1992, Sources and relative reactivities of amino acids, neutral

sugars, and lignin in an intermittently anoxic sediment: Geochimica et Cosmochimica Acta, v. 56,p. 19631978.

Cowie, G. L., Hedges, J. I., Prahl, F. G., and de Lange, G. J., 1995, Elemental and major biochemical changesacross an oxidation front in a relict turbidite: An oxygen effect: Geochimica et Cosmochimica Acta, v. 59,p. 3346.



25/27

Cushing, E. J., 1967, Evidence for differential pollen preservation in Late Quaternary sediments in Minnesota:Review of Palaeobotany and Palynology, v. 4, p. 87101.

De Lange, G. J., Jarvis, I-., and Kjuipers, A., 1987, Geochemical characteristics and provenance of lateQuaternary sediments from the Madeira Abyssal Plains, N. Atlantic. inWeaver, P. P. E. and Thomson, J.editors, Geology and Geochemistry of Abyssal Plains: Blackwood, p. 147165.

Demaison, G. J., and Moore, G. T., 1980, Anoxic environments and oil source bed genesis: AmericanAssociation of Petroleum Geologists Bulletin, v. 64, p. 11791209.Emerson, S., 1985, Organic carbon preservation in marine sediments, inSundquist, E. T., and Broecker, W. S.,

editors, The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present: AmericanGeophysical Union, Geophysical Monograph Series, v. 32, p. 7887.

Emerson S.,and Hedges, J. I., 1988, Processescontrolling the organic carbon content of open ocean sediments:Paleoceanography, v. 3, p. 621634.

Engel,M. H.,and Macko,S. A., 1993, Organic Geochemistry: Principles andApplications: NewYork, PlenumPress, 861 p.

Faegri, K., Iversen, J., Kaland, P. E., and Krzywinski, K., 1989, Textbook of Pollen Analysis: New York,Hafner, 328 p.

Gross, M. G., Carey, A. G., Fowler, G. A., and Kulm, L. D., 1972, Distribution of organic carbon in surficialsediment, northeast Pacific Ocean, in Pruter, A. T., and Alverson, D. L., editors, The Columbia RiverEstuary and Adjacent Ocean Waters: Bioenvironmental Studies: Seattle, University of Washington Press,p. 254264.

Hamilton, S. E., and Hedges, J. I., 1988, The comparative geochemistries of lignins and carbohydrates in ananoxic fjord: Geochimica et Cosmochimica Acta, v. 52, 129142.

Harmon, R. A., 1972, Distribution of microbiogenic sediment near themouth of the Columbia River, inPruter,A. T., and Alverson, D. L., editors, The Columbia River Estuary and Adjacent Ocean Waters:Bioenvironmental Studies: Seattle University of Washington Press, p. 265278.

Hartnett,H. E., Keil, R.G., Hedges, J.I., and Devol,A. H., 1998, Influence of oxygenexposuretime on organiccarbon preservation in continental margin sediments: Nature, v. 391, p. 572574.

Hedges, J. I., Clark, W. A., and Cowie, G. L., 1988, Fluxes and reactivities of organic matter in a coastal marinebay: Limnology and Oceanography, v. 33, p. 11161136.

Hedges, J. I., Cowie, G. L., Richey, J. E., Quay, P. D., Benner, R., Strom, M., and Forsberg, B., 1994, Originand processing of organic matter in the Amazon River as indicated by carbohydrates and amino acids:Limnology and Oceanography, v. 39, p. 743761.

Hedges, J. I., and Keil, R. G., 1995, Sedimentary organic matter preservation: an assessment and speculativesynthesis: Marine Chemistry, v. 49, p. 81115.

Hedges, J. I., and Mann, D. C., 1979, The lignin geochemistry of marine sediments from the southernWashington coast: Geochimica et Cosmochimica Acta, v. 43, p. 18091818.

Hedges, J. I., Mayorga, E., Tsamakis, E., McClain, M. E., Quay, P. D., Richey, J. E., Benner, R., Opsahl, S.,Black, B., Pimentel, T., Aquirre, J. Q., and Maurice, L., submitted. Organic matter in Bolivian tributariesof the Amazon River: A comparison to the lower mainstream: Limnology and Oceanography.

Hedges, J. I., and Oades, J. M., 1997, Comparative organic geochemistries of soils and marine sediments:Organic Geochemistry, v. 27, p. 319361.

Hedges, J. I., and Prahl, F. G., 1993, Early diagenesis, consequences for applications of molecular biomarkers.inEngel, M. H., and Macko, S. A., editors, Organic Geochemistry: New York, Plenum Press, p. 237253.

Hedges, J. I., and Stern, J. H., 1984, Carbon and nitrogen determinations of carbonate containing solids:Limnology and Oceanography, v. 29, p. 657663.

Henrichs, S. M., 1992, The early diagenesis of organic matter in marine sediments: Progress and perplexity:Marine Chemistry, v. 39, p. 119149.

1993, Early diagenesis of organic matter: The dynamics (rates) of cycling of organic compounds, inEngel, M. H., and Macko, S. A., editors, Organic Geochemistry: New York, Plenum Press, p. 101117.

Henrichs, S. M., and Reeburgh, W. S., 1987, Anaerobic mineralization of marine sediment organic matter:Rates and the role of anaerobic processes in the organic carbon economy: Geomicrobiology, v. 5,p. 191237.

Hernes, P. J., Hedges, J. I., Peterson, M. L., Wakeham, S. G., and Lee, C., 1996, Neutral carbohydrategeochemistry of particulate organic material in the central equatorial Pacific: Deep-Sea Research, v. 43,p. 11811204.

Heusser, L. E., 1985, Quaternary palynology of marine sediments in the northeast Pacific, northwest Atlantic,andGulf of Mexico, inBryant, Jr., V. M.,and Holloway,R. G.,editors, PollenRecords of Late-QuaternaryNorth American Sediments: Dallas, Texas, American Association of Straitigraphic Palynologists,p. 386403.

Hulthe, G., Hulth, S., and Hall, P. O., 1998, Effect of oxygen on degradation rate of refractory and labileorganic matter in continental margin sediments: Geochimica et Cosmochimica Acta, v. 62, p. 13191328.

Hunt, J. M., 1996, Petroleum Geochemistry and Geology: New York, Freeman Press, p.Jahnke, R. A., 1996, The global flux of particulate organic carbon: Areal distribution and magnitude: GlobalBiogeochemical Cycles, v. 10, p. 7188.

Jorgensen, B. B., 1978, A comparison of methods for the quantification of bacterial sulfate reduction in coastalmarine sediments. II. Calculations from mathematical models: Gemicrobiology, v. 1, p. 2947.

Kachel, N. B., and Smith, J. D., 1989, Sediment transport and deposition on the Washington shelf, inLandry,M. R., and Hickey, B. M., Coastal Oceanography of Washington and Oregon: New York, Elsevier,p. 287342.

Keil, R. G., Hu, F. S., Tsamakis, E., and Hedges, J. I., 1994b, Pollen grains deposited in marine sediments aredegraded only under oxic conditions: Nature, v. 369, p. 639641.



26/27

Keil, R. G., Mayer, L. M., Quay, P. D., Richey, J. E., and Hedges, J. I., 1997, Loss of organic matter fromriverine particles in deltas: Geochimica et Cosmochimica Acta, v. 61, p. 15071511.

Keil, R. G., Montlucon, D. B., Prahl, F. G., and Hedges, J. I., 1994a, Sorptive preservation of labile organicmatter in marine sediments: Nature, v. 370, p. 549552.

Keil, R. G., Tsamakis, E., Fuh, C. B., Giddings, J. C., and Hedges, J. I., 1994c, Mineralogical and textural

controls on organic composition of coastal marine sediments: Hydrodynamic separation using SPLITTfractionation: Geochimica et Cosmochimica Acta, v. 57, 879893.Keil, R. G., Tsamakis E., Giddings, J. C., and Hedges, J. I., 1998, Biochemical distributions (amino acids,

neutral sugars and lignin phenols) among size-classes of modern marine sediments from the WashingtonCoast: Geochimica et Cosmochimica Acta, 62, p. 13471364.

Killops, S. D., and Killops, V. J., 1993, An Introduction to Organic Geochemistry: Essex, United Kingdom,Longman, 707 p.

Lambourn, L. D., Devol, A. H., and Murray, J. W., 1991, R/V New Horizon 90-5 cruise report: Water columnand porewater data: Seattle, Washington, University of Washington, Special Report No. 110, ReferenceA91-1, p.

Lambourn,L. D.,Hartnett, H.,and Devol,A. H.,1996,R/V Wecoma WE94-07Bcruise report: Porewaterdatafrom the Washington shelf and slope, Seattle, Washington, University of Washington, Special Report No.113, Reference A96-1, p.

Lee, C., 1992, Controls on organic carbon preservation: The use of stratified water bodies to compare intrinsicrates of decompositionin oxicand anoxic systems: Geochimica et CosmochimicaActa, v. 56, p. 33233335.

Lee, C., Wakeham, S. G., and Hedges, J. I., submitted, Composition and flux of particulate amino acids andpigments in Equatorial Pacific seawater and sediments: Deep-Sea Research.

Luckge, A., Boussafir, M., Lallier-Verges, E., and Littke, R., 1996, Comparative study of organic matterpreservation in immature sediments along the continental margin of Peru and Oman. Part I: Results ofpetrographical and bulk geochemical data: Organic Geochemistry, v. 24, p. 437451.

Martin, J. H., Knauer, G. A., Karl, D. M., and Broenkow, W. W., 1987, VERTEX: carbon cycling in thenortheast Pacific: Deep-Sea Research, v. 34, p. 267285.

Mayer, L. M., 1994a, Surface area control of organic carbon accumulation in continental shelf sediments:Geochimica et Cosmochimica Acta, v. 58, p. 12711284.

1994b, Relationships between mineral surfaces and organic carbon concentrations in soils andsediments: Chemical Geology, v. 114, p. 347363.

Middelburg, J. J., 1989, A simple rate model for organic matter decomposition in marine sediments:Geochimica et Cosmochimica Acta, v. 53, p. 15771581.

Middelburg, J. J., Vlug, T., and Van der Nat, F. J. W. A., 1993, Organic matter mineralization in marinesystems: Global Planetary Change, v. 8, p. 4758.

Muller, P. J., and Suess, E., 1979, Productivity, sedimentation rate, and sedimentary organic matter in theoceans-I. Organic carbon preservation: Deep-Sea Research, v. 26, p. 13471362.

Nittrouer, C. A., and Sternberg, R. W., 1981, The formation of sedimentary strata in an allochthonous shelfenvironment: the Washington continental shelf: Marine Geology, v. 42, p. 201232.

Pedersen, T. F., Shimmield, G. B., and Price, N. B., 1992, Lack of enhanced preservation of organic matter insediments under the oxygen minimum on the Oman Margin: Geochimica et Cosmochimica Acta, v. 56,p. 545551.

Perry, M. J., Bolger, J. P., and English, D. C., 1989, Primary production in Washington coastal waters, inLandry, M. R., and Hickey B., editors, Coastal Oceanography of Washington and Oregon: New York,Elsevier, p. 117138.

Prahl, F. G., 1985, Chemical evidence of differential particle dispersal in the southern Washington coastalenvironment: Geochimica et Cosmochimica Acta, v. 49, p. 25332539.

Prahl, F. G., de Lange, G. J., Lyle, M., and Sparrow, M. A., 1989, Post-depositional stability of long-chainalkenones under contrasting redox conditions: Nature, v. 341, p. 434437.

Prahl, F. G., de Lange, G. J., Scholten, S., and Cowie, G. L., 1997, A case of post-depositional aerobicdegradation of terrestrial organic matter in turbidite deposits from the Madeira Abyssal Plain: OrganicGeochemistry, v. 27, p. 141152.

Prahl, F. G., Ertel, J. R., Goni, M. A., Sparrow, M. A., and Eversmeyer, B., 1994, Terrestrial organic carboncontributions to sediments on the Washington margin: Geochimica et Cosmochimica Acta, v. 58,p. 30353048.

Premuzic, E. T., Benkovitz, C. M., Gaffney, J. S., and Walsh, J. J., 1982, The nature and distribution of organicmatter in surface sediments of world oceans and seas: Organic Geochemistry, v. 4, p. 6377.

Ransom, B., Kim, D., Kastner, M., and Wainwright, S., 1998, Organic matter preservation on continentalslopes: Importanceof mineralogy and surface area: Geochimica Cosmochimica Acta, v. 62, p. 13291345.

Reimers, C. E., 1989, Control of benthic fluxes by particulate supply, inBerger, W. H., Smetacek, V. S., andWefer, G., editors, Productivity of the Ocean: Present and Past: New York, John Wiley Sons, p. 217233.

Reimers, C. E., Jahnke, R. A., and McCorkle, D. C., 1992, Carbon fluxes and burial rates over the continentalslope and rise off central California with implications for the global carbon cycle: Global BiogeochemicalCycles, v. 6, p. 199224.

Richards, F. A., and Redfield, A. C., 1954, A correlation between the oxygen content of seawater and theorganic content of marine sediments: Deep-Sea Research, v. 1, p. 279281.

Schubert, C. J., and Stein, R., 1996, Deposition of organic carbon in Arctic Ocean sediments: terrigenoussupply vs. marine productivity: Organic Geochemistry, v. 24, p. 421436.

Smith, C. R., Walsh, I. D., and Jahnke, R. A., 1992, Adding biology to one-dimensional models ofsediment-carbon degradation: The multi-B approach, inRowe, G. T., and Pariente, V., editors, Deep-SeaFood Chains and the Global Carbon Cycle: The Netherlands, Kluwer Academic Publishers, p. 395 400.



27/27

Smith, K. L., Jr., 1978, Benthic community respiration in the N. W. Atlantic Ocean: in situ measurements from40 to 5299 m: Marine Biology, v. 47, p. 337347.

Soetaert, K., Herman, P. M. J., and Middelburg, J. J., 1996, A model of early diagenetic process from the shelfto abyssal depths: Geochimica et Cosmochimica Acta, v. 60, p. 10191040.

Stuiver, M., and Polach, H. A., 1977, Discussion: Reporting of14C data: Radiocarbon, v. 19, p. 355363.

Suess, E., 1980, Particulate organic carbon flux in the oceanssurface productivity and oxygen utilization:Nature, v. 288, p. 260263.Tarutis, Jr., W. J., 1993, On the equivalence of the power and reactive continuum models of organic matter

diagenesis: Geochimica et Cosmochimica Acta, v. 57, p. 13491350.Thomson, J., Higgs, N. C., Croudace, I. W., Colly, S., and Hydes, D. J., 1993, Redox zonation of elements at an

oxic/post-oxic boundary in deep-sea sediments: Geochimica et Cosmochimica Acta, v. 57, p. 579595.Van Liew, H. D., 1962, Semilograthmic plots of data which reflect a continuum of exponential processes:

Science, v. 138, p. 682693.Wakeham, S. G., Lee, C., Hedges, J. I., Hernes, P. J., and Peterson, M. L., 1997, Molecular indicators of

diagenetic status in marine organic matter: Geochimica et Cosmochimica Acta, v. 61, 53635369.Weaver, P. P. E., and Rothwell, R. G., 1987, Sedimentation on the Madeira Abyssal Plain over the last 300,000

years. in Weaver, P. P. E., and Thomson, J., editors, Geology and Geochemistry of Abyssal Plains:London, Blackwell Scientific, p. 7186.

Weiler, R. R., and Mills, A. A., 1965, Surface properties and pore water structure of marine sediments:Deep-Sea Research, v. 12, p. 511529.

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, 236249.

Whelan, J. K., 1977, Amino acids in a surface sediment core of the Atlantic abyssal plain: Geochimica etCosmochimica Acta, v. 41, 803810.

White, S., 1970, Mineralogy and geochemistry of continental shelf sediments off the Washington-Oregoncoast: Journal of Sedimentary Petrology, v. 40, p. 3854.


Documents

SEDIMENTARY ORGANIC MATTER PRESERVATION: A TEST FOR SELECTIVE DEGRADATION UNDER OXIC CONDITIONS