bottom with older detrital material. The work of Coyle and 

 Cooney (1988) indicates that a large portion of the eastern 

 Bering Sea ice-edge bloom is ungrazed by zooplankton and 

 sinks directly to the benthos. These workers also found that a 

 separate zooplankton community on the middle shelf of the 

 southeastern Bering Sea did not graze a significant proportion 

 ofthe spring carbon production (Cooney & Coyle, 1982). Here 

 again, a large percentage of the organic carbon settles to the 

 benthos to support a rich benthic community (Feder et al.. 

 1980;Feder&Jewett, 1981;Walsh& McRoy, 1986). Likewise, 

 In the northern Bering and Chukchi Seas, a significant portion 

 of the high carbon production in the Bering Shelf and Anadyr 

 waters reaches the benthos, supporting a high benthic biomass 

 of infaunal invertebrates and high sediment carbon 

 mineralization (Grebmeier et al., 1988, 1989; Grebmeier & 

 McRoy, 1989; Walsh et al., 1989). Although sediment grain 

 size composition is the dominant factor in determining benthic 

 community composition on the continental shelf of the Bering 

 and Chukchi Seas (Stoker, 1978; Grebmeier f/fl/., 1989), food 

 supply is the major factor influencing benthic biomass 

 (Grebmeier f/fl/., 1988: Grebmeier & McRoy, 1989). 



Carbon/nitrogen ratios in surface sediments can provide 

 an indication of the quality of organic material arriving at the 

 sea bottom, although water column nutrient concentration, 

 zooplankton grazing, and bacterial degradation can influence 

 these values (Parsons et al., 1977; Valiela, 1984). Past work 

 has used surface sediment C/N values to separate areas of high 

 and low particulate organic carbon loss to the benthos (Walsh 

 et al.. 1981). Grebmeier el al. (1988) used C/N ratios, in 

 combination with sediment respiration rates, to investigate the 

 quality and quantity of organic material available to benthic 

 populations in the northern Bering and Chukchi Seas. Low 

 surface sediment C/N ratios (5-7 wt./wt.) suggest a higher 

 quality, nitrogen-rich organic material deposition to the benthos 

 under the highly productive Bering Shelf-Anadyr water 

 .(280 g C m - yr '; Walsh et al., 1989), compared with lower 

 quality, higher C/N ratios (8-14 wt./wt.), indicative of less 

 labile, more refractory marine and terrestrial organic matter in 

 the sediment under the less productive Alaska Coastal water 

 (60gCm-yr ';Walshcfa/., 1989). In the southeastern Bering 

 Sea, variations in the cross-shelf distributions of C/N ratios in 

 surface sediments have been attributed to different proportions 

 ofdetritus (Walsh ('/«/.. 1981; Walsh & McRoy, 1986). 



Lead-210(-"'Pb; half-life = 22.3 yr) is a particle-reactive, 

 naturally-occurring radionuclide produced during the 

 -'"Uranium decay series. Since -"Rn is a nonreactive, noble 

 gas, some escapes to the atmosphere before it decays to -'"Pb, 

 which is then washed out of the atmosphere via precipitation, 

 forming a measurable flux of -'"Pb in excess of that produced 

 solely from ---Rn decay within the water column and sediments. 

 Lead-2 10 rapidly adsorbs onto particulate matter and descends 

 to the bottom sediments. It has been used successfully in both 

 freshwater and marine systems to quantify physical and 

 biogeochemical processes on time scales of months to decades 

 (Krishnaswami et a!., 1980; Walsh, 1988). In sedimentary 

 studies, measures of excess -'"Pb ( - '"Pb-ex) in sediments provide 

 an indicator of sediment accumulation, since -'"Pb 

 concentrations in excess of that supported by decay in the 



sediments alone indicate accumulation of allocthonous 

 materials. Used coincidently with organic carbon measurements 

 in the sediment, -'"Pb-ex can provide information on 

 sedimentation and accumulation rates of organic carbon in the 

 environment. 



Stable oxygen isotope composition is normally expressed 

 as '^O/'^O ratios in the standard 5 notation: 



5 "<0 = (''standard/'*sample')x 10' 7„,, where R = '"0/"'0 and 

 standard is standard mean ocean water (SMOW). The stable 

 oxygen isotope composition of seawater (5 '"O) varies 

 temporally and spatially in regions of the ocean, such as on 

 shallow continental shelves influenced by freshwater input, 

 which is depleted in the heavier oxygen isotope ('*0), relative 

 to the lighter oxygen isotope ( '"O), particularly at high latitudes 

 where fractionation is intensified (Ferronsky & Polyakov, 

 1982). Stable oxygen isotope variability in surface marine 

 waters has been used to study oceanic circulation, and when 

 combined with salinity and temperature data — water 

 contributions from rivers, evaporated surface ocean waters, 

 melting glaciers, and melting sea ice, can be separated and 

 water types characterized (Epstein & Mayeda, 1953: Redfield 



6 Friedman, 1969; Tan & Strain, 1980; Bedard et al., 1981: 

 Fertonsky & Polyakov, 1982). Salinity is the predominant 

 factor determining seawater density and water mass 

 characteristics in the northern Bering and Chukchi Seas 

 (Coachman et al., 1975). There is a known relationship 

 between '"O content and salinity in ocean waters, with similar 

 processes influencing both salinity and "'O content in tandem 

 (Epstein & Mayeda, 1953: Fen-onsky & Polyakov, 1982). 

 Thus, the major water masses in our study should be 

 distinguishable by both 5 "*0 values and salinity concentrations. 

 However, the salinity-5 ""O relationship can become decoupled 

 when multiple freshwater sources of differing 5 "*0 values mix 

 with saline water, leading to different 5 "O values but similar 

 salinities for the mixtures. Another deviation from the salinity- 



5 '"O relationship can occur when sea ice forms and the 

 resultant brine injection increases the underlying waters' 

 salinities but does not significantly change 6 ' '*0 values over the 

 whole water column, although sea ice itself is affected ( Redfield 



6 Friedman, 1969; Vetshteyn et al., 1974; Fertonsky & 

 Polyakov, 1982). Thus '"O data can allow tracing of known 

 water mass distributions in polar seas despite changes in 

 salinity over the winter period. 



Oxygen removed from seawater by organisms reflects 

 oceanic circulation processes in many circumstances. The 

 oxygen isotope composition of cellulose is directly related to 

 the oxygen isotope composition of water available to submerged 

 aquatic plants and to members ofthe marine urochordate class 

 Ascidiacea (tunicates), which synthesize cellulo.se (Epstein 

 et al., 1977: DeNiro & Epstein, 1979, 1981). DeNiro and 

 Epstein (1981) observed that aquatic plant and tunicate cellulose 

 5 "*0 values were 27 ± 3 "/„o more positive than the 5 "*0 values 

 of the growth media. No significant temperature effects on 

 isotopic fractionation were observed during cellulose synthesis 

 in freshwater plants (DeNiro & Epstein, 1981) or during the 

 carbonyl exchange reactions prior to cellulose synthesis that 

 may govern the fractionation observed (Sternberg & DeNiro, 

 1983). Although tunicates have not been cultured under 



244 



