Assessing Bioresources and Standing Stock of Zoobenthos (Key Species, High Taxa, Trophic Groups) in the Chukchi Sea

USAGE Permission is granted to copy this article for use in teaching and research. Republication, systematic reproduction, or collective redistribution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: info@tos.org or The Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA. Oceanography THE OFFICIAL MAGAZINE OF THE OCEANOGRAPHY SOCIETY

and distribution of ecological communi ties, including energy flows (Piraino et al., 2002). Biological metrics include faunal biomass, abundance, productivity, and function in the ecosystem.
The Chukchi Sea's marine resources ( Figure 1a) are not harvested commer cially, although local people fish, hunt marine mammals, and harvest benthic invertebrates for their own use. Locally potential harvested benthic species in the Chukchi Sea include the bivalves Mya, Serripes, Mytilus, Chlamys, and Macoma; several carnivorous gastro pods from the genera Buccinum and Neptunea; the snow crab Chionoecetes opilio; and sea urchins from the genus Strongylocentrotus. However, the exact quantity of this bio resource available for harvesting is not known.
Other bioresources include plants and animals consumed by marine mammals, fishes, and birds. The Chukchi Sea attracts large numbers of walruses, seals, bow head and gray whales, and eiders that are either resident to the area or migrate to specific regions every year to feed, some on benthic prey. Some of these benthic species are listed in the Red Data Book of the Russian Federation (Illiashenko and Illiashenko, 2000) and/or are pro tected under the Endangered Species Act of the United States.
Among the marine mammals, the main consumers of benthos in the Pacific Arctic region are Pacific walruses and gray whales (Moore et al., 2014). Walruses feed on bivalves and gastropods, crustaceans, worms, and ascidians (Krylov, 1971;Sheffield and Grebmeier, 2009), while gray whales prefer crustaceans, especially amphipods, and other high biomass ben thic species (Blokhin and Pavluchkov, 1996;Highsmith et al., 2006).
Benthic biomass in the north ern Pacific Ocean increases northward from the Sea of Japan to the Bering Sea (Zenkevich, 1951;Shuntov, 2001), increasing to high values in the south ern Chukchi Sea (Makarov, 1937;Sirenko and Koltun, 1992;Grebmeier et al. 2006). A considerable proportion of this benthic fauna is represented by mollusks, crus taceans, polychaetes, sipunculid worms, echinoderms, and ascidians, and known regions of high benthic prey abundance attracts walruses, seals, whales, and birds to these areas to feed.
In order to identify and quantify the bioresources of the Chukchi Sea, we pres ent a general assessment of the key species of benthic invertebrates, based on quan titative grab samples of macrobenthic fauna. This study also evaluates the biore source potential of the major higher taxa components (at the level of class or phy lum) and associated trophic groups. We present (1) key species identified by their contributions to the total zoobenthos biomass, and (2) spatial distribution of

INTRODUCTION
With increasing depletion of commercial fisheries, the number of newly harvested marine species is on the rise. Thus, there is an immediate need to estimate all avail able biological resources (termed bio resources in this paper) used by humans, including food for consumption, forage feed for domestic animals, and indus trial raw materials. These new resources and their potential economic importance require a quantitative assessment of the total biomass (abundance) and the spatial distribution of the resource base.
In order to monitor and study the longterm dynamics of zoobenthic pop ulations, it is necessary to analyze the spatial distributions of those species or higherlevel taxa that contribute most to the total biomass in an area of inter est. As a rule, the number of such spe cies in each Arctic sea does not exceed 20 (Denisenko, 2004), with 50% of the zoo benthos biomass being formed by fewer than 10 species. These few can be con sidered the "key species" influencing the structure and function of the associated ecosystems and they have key roles in controlling the structure, development, ABSTRACT. We describe total standing stock and spatial biomass distribution of the key macrofaunal species, as well as the main taxonomic and trophic groups of zoobenthos, in the Chukchi Sea based on data collected over the period 1986-2012. The dominant species, ranked by biomass, are the bivalves Macoma calcarea, Ennucula tenuis, Astarte borealis, Nuculana radiata, and Yoldia hyperborea; the sipunculid, Golfingia margaritacea; the polychaete Maldane sarsi; and the sea cucumber Psolus peroni. We discuss the influence of bottom sediments and water masses on zoobenthos bioresources and standing stock and present a hypothesis for the mechanism that facilitates the phenomenon of very high benthic biomass in the southern Chukchi Sea. The relationship between biomass of filter feeders and deposit feeders indicates that the Chukchi Sea should be classified as a eutrophic marine system.  Kosheleva and Yashin (1999); Grebmeier (2012) a b was necessary in order to compare differ ent biomass values in the same local max ima or minima. A significant difference between these values can considerably decrease Pearson correlation even when the spatial distributions under compari son are visually very similar. Organism feeding types were also used in the analysis and are based on previously published categorizations (Kuznetsov, 1980(Kuznetsov, , 1986.

RESULTS
Comparison of benthic faunal biomass, statistically weighted by area at each sta tion, revealed eight macrofaunal spe cies contributing up to 50% of the total zoobenthos bioresources of the Chukchi Sea (Table 1). Five of these species were bivalve mollusks: Macoma calcarea, Ennucula tenuis, Astarte borealis, Nuculana radiata, and Yoldia hyperborea. Of these, M. calcarea was the most important and abundant, with a total bio mass (i.e., bioresources) several times greater than those of any other bivalve species evaluated separately.
The zoobenthos biomass and the five key bivalve species were unevenly dis tributed. The highest values of total macrofaunal biomass were observed in the southern and northwestern areas of the sea (Figure 3a), and М. calcarea had the highest biomass of all species in the southern and the southwestern areas ( Figure 3b). The highest biomass val ues of a second bivalve, E. tenuis, also occurred in the southwestern portion of the Chukchi Sea ( Figure 3c). The high est biomass of A. borealis was recorded in the northern part of the sea as well as in the south, in Koluchin Bay and Kotzebue Sound (Figure 3d). The major settlements of N. radiata extended toward the west ern part of the Chukchi Sea (Figure 3e). High biomass values of the sipunculid Golfingia margaritaceum were mostly recorded in northern and the northeast ern areas (Figure 3f).
Significant Spearman correlations indi cated some degree of spatial distribution similarity, by biomass, of М. calcarea and months in formalin.
Stationwise species lists of abun dance and biomass for all data were pooled into a common data set. Based on these files, maps of the spatial distribu tions of species, taxonomic groups, and trophic groups were constructed using Ocean Data View 4.7.3 (Schlitzer, 2015). The total benthic biomasses of compo nents were displayed spatially using geo statistical interpolations performed with Surfer 8 and MapViewer 7.2 (http:// www.goldensoftware.com). Kriging, with a spherical variogram and search ellipse with a radius of -5° longitude and -1.8° latitude (Software Surfer, v8), was used as a universal interpolator. These parameters were experimentally found to be optimal with respect to the devia tion of calculated values from the actual data. On average, the calculated deviation made up not more than 10% of the actual benthic faunal biomass found at the study sites. The map surface areas were based on the northern border of the Chukchi Sea being located along 72°30'N°, with a total area of ca. 335,000 km 2 . The west ern border was located to the north of Wrangel Island along 180°W and further to the south from Cape Blossom to Cape Yakan. The eastern border was located along 158°W ( Figure 1).
In order to assess the similarity of spa tial distributions of individual species, taxonomic groups, and trophic groups, the Spearman rank coefficient rather than the Pearson correlation was used. This these species using geographical imagery, together with an evaluation of the quanti tative distribution of the dominant benthic taxa and their trophic levels in the food web. A companion paper by Grebmeier et al. (2015b) in this special issue dis cusses total benthic community composi tion of benthic macro and megafauna in the context of environmental factors.

MATERIALS AND METHODS
This study was based on quantitative zoo benthos samples collected at 640 stations in the Chukchi Sea by specialists from the Zoological Institute of the Russian Academy of Sciences and scientists from the University of Alaska Fairbanks and the University of Maryland Center for Environmental Science over the period 1976-2012 ( Figure 2). Macrofauna were collected during multiple cruises in both Russian and US waters through programs that included the RussianAmerican Longterm Census of the Arctic (RUSALCA). Three to five replicates (with a few tworeplicate stations) were col lected using either a 0.25 m 2 Okean grab or a 0.1 m 2 van Veen grab, with sedi ment sieved over primarily 1 mm screens (a few over 0.8 mm screens). The remain ing macrofauna were preserved in 4-10% seawater formalin buffered with borax or hexamethylenetetramine. Macrofauna were later identified to genus or spe cies levels (or to the lowest taxon pos sible), with the taxa subsequently counted and weighed after three to four E. tenuis (S = 0.51), and of A. borealis and G. margaritacea (S = 0.52). Bivalves clearly dominate in biomass compared with the other large macrofaunal taxa in the Chukchi Sea (Table 2). They made up more than 50% of the zoobenthos bioresources by both wet and dry weight biomass. Even considering that the cal cium carbonate shell accounts for about 30-50% of the molluskan wet bio mass (Alexey V. Golikov, Kazan Federal University, pers. comm., 2015;Stoker, 1978;Brey, 2001;Denisenko, 2013), bivalves are still the dominant taxa by biomass. This finding was also supported by recalculat ing the wet biomass on an organic carbon basis (Stoker, 1978;Grebmeier, 2012). The contribution of other taxonomic groups was an order of magnitude lower than that of bivalves (see Table 2). Polychaetes accounted for ~11% of total zoobenthos bioresources, as did echinoderms (holo thurians and sea urchins). The distribution of bivalve biomass ( Figure 4a) resembled that of the total zoobenthos biomass (see Figure 3a). However, the biomass values for Bivalvia were much lower in the northwestern and, especially, in the northeastern part of the sea relative to the high southern Chukchi Sea biomass values, although they can be regionally important. with sea urchins, mostly formed biomass aggregations in the southern and north ern parts of the Chukchi Sea, where water dynamics were the greatest. Judging from the location of the local biomass max ima of these groups, these varying taxon TABLE 1. Contribution of different species into total bioresources (by mass, tons) and total specific biomass (g wet weight m -2 ) of zoobenthos over the whole Chukchi Sea. R% = Proportion of the total bioresources. R,t = Bioresources (tons). B = Average specific total biomass (g wet weight m -2 ) within regions. B Err = Statistical error of the average specific total biomass.

Species
Feeding mode R% R,t  (Kuznetsov, 1986).  settlements did not overlap each other. Due to this separation of groups, the sig nificant Spearman correlation coeffi cients, which indicate similarity of the spatial distribution of fauna by biomass, were only obtained for Ascidiacea and Echinoidea (S = 0.77). Filter feeders (suspension feeders) pre dominate overall for much of the spatial extent of the Chukchi Sea. Their propor tion exceeded 44% of the total biomass, making up approximately 40 million tons, with an average wet biomass of 141 g m -2 ( Table 3). The proportion of deposit feed ers (including both surface and subsur face deposit feeders) is about equal to the filter feeders, making up 45.2% of the overall biomass or approximately 40 mil lion tons. The bioresources proportion of surface deposit feeders to subsurface deposit feeders was 3:2. Omnivorous and predatory species contributed less than 10%, and made up 7.5 million tons. On the whole, the average biomass value of each trophic group was better correlated with its individual bioresource value than biomass of the large taxa or the key spe cies groups with their individual biore source values (see Tables 1, 2, and 3).
The spatial distribution of trophic groups shows a tendency toward replace ment of communities where filter feeders dominate (mostly in the southernmost and northwesternmost parts; Figure 6a) with communities in which surface deposit feeders dominate (in the northeastern and southwestern parts; Figure 6b). Slow moving subsurface deposit feeders domi nate in the northeastern Alaska Chukchi coastal waters, around Hanna Shoal, and to the east and south of Wrangel Island (Figure 6c) within siltyclay sedi ments (Figure 1b). The highest biomass of predators coincides partly with the total zoobenthos biomass.

Benthic Communities Defined by Biomass Spatial Patterns
Visual or computerized demarcation of benthic communities based on the results of clustering or ordination is the most TABLE 2. Contribution of higher taxa into total bioresources (by mass, tons) and total specific weight (g wet weight m -2 ) of zoobenthos in the Chukchi Sea. R% = Proportion of the total bioresources. R,t = Bioresources (tons). B = Average specific total biomass (g wet weight m -2 ) within regions. B Err = Statistical error of the average specific total biomass. Note: All mollusks from the family Tellinidae were considered to be an intermediate group between filter feeders and surface deposit feeders (Kuznetsov, 1986).  Note: All mollusks from the family Tellinidae were considered to be an intermediate group between filter feeders and surface deposit feeders (Kuznetsov, 1986).

Higher taxa R% R,t (tons) B (gww m -2 ) B Err (gww m -2 )
common method used for local zonation of marine benthic biota and for compar ing qualitative and quantitative character istics of zoobenthos in various regions and over temporal periods (Denisenko, 2007). Most researchers define a "biological community" as the combination of popu lations of organisms interacting with each other and with the environment in space and time (Möbius, 1877;Whittaker, 1975;Bigon et al., 2006). However, this defini tion is somewhat vague because the com munities usually have no distinct bor ders but rather transition into each other imperceptibly (Pianka, 1994). Besides, the use of different approaches and software yields different, and often not compara ble, results describing the dominance of a given group in a given area. Certain methods can be limiting in their demarcation of benthic communi ties because of assumptions that species always have the same quantitative rep resentation in communities (Vorobyev, 1949;Clarke and Warwick, 2001). The possibility of detecting changes in the abundance and biomass of populations can be obscured, although longterm fluctuations and quantitative indices of many benthic invertebrates (Pearson and Rosenberg, 1978;Gerasimova and Maksimovich, 2000;Kortsch et al. 2012;Denisenko, 2013), plankton (Krause Jensen et al., 2012), and fish (Klyashtorin and Lubushin, 2005) are well known. As a result, comparing the distributions of the socalled communities in different marine areas and for different research periods might be reduced to a compari son of the degree of dominance of a few key species, such as indicated in our pres ent study. A cyclic decrease or increase in the abundance or biomass of these species, caused by any environmental changes, may result in a misclassification of communities on the basis of species dominance (Denisenko, 2013).
Our research indicates that the most abundant species can also dominate by biomass in benthic communities (Table 1; Denisenko, 2004Denisenko, , 2007Sirenko et al., 2009;Grebmeier et al., 2015a). These species often serve as the major food source for numerous benthicfeeding mammals and birds. Therefore, the assessment of macrobenthos biomass should be based on direct calculations of the quantitative spatial distributions of those key species, higher taxa, and trophic groups that dominate the bio mass (i.e., bioresources) in a study area. The importance of these species as future human resources may better be assessed by a simple interpolation of biomass on a map. In contrast to various methods of "community" classification, includ ing visual recognition of dominant spe cies, our simplified approach will more directly facilitate comparative analysis in the future. In addition, using interpola tion approaches, the outliers on the maps are to some degree smoothed, which compensates for possible sampling errors and the mosaic nature of organismal dis tributions on the seafloor.
Using this approach of statistically  Sirenko and Gagaev, 2007) coincide with the locations of known particle deposi tions sites to the north of Bering Strait (Sirenko and Koltun, 1992;Cooper et al., 2005;Grebmeier et al. 2006;Pisareva et al., 2015, in this issue). This high ben thic biomass zone may result from slow ing current speeds (Ratmanov, 1937) and possible cyclonic features that are pres ent in the autumn (Pickart et al., 2013; Figure 7). A nearly continuous input of nutrients through Bering Strait allows the phytoplankton community to func tion actively throughout the period of sufficient light and to provide the under lying sediments with a continuous supply of fresh phytodetritus, the major source of food for benthic invertebrates. In con trast, in other Arctic regions, spring development of phytoplankton blooms is subsequently inhibited by depletion of nutrients in the surface layers of stratified water columns (Sakshaug 2004). Thus, during most of the season when light is available, the sinking of phytodetritus to the bottom in these other regions is weak, thus limiting the production and biomass of underlying zoobenthic communities. The exact mechanism for this strong pelagicbenthic coupling is uncertain, but its role in the annual formation of a zone of high benthic biomass in the southern Chukchi Sea is critical. The existence of periodic cyclonic type cir culation is supported by hydrographic section data. However, we disagree with Ratmanov's (1937) interpretation that the cyclonic gyre arises from the inter action of northwardflowing Pacific water and a southwestflowing East Siberian Current. More recent data indicate a vari able, more ephemeral flow from the East Siberian Sea along the Chukchi Sea coast to the southern Chukchi Sea (Münchow et al., 1999;Weingartner et al., 2005). It is more likely that the greater depth to the northwest helps facilitate a cyclonic type circulation in the upper portion of Central Valley. This alternative mech anism would result in enrichment of the photosynthetic water layer by nutri ents and enhanced carbon export to the weighted spatial benthic data analysis, it was determined that 75-80% of the total macrofaunal biomass of the Barents Sea during a defined study period was made up of 15-20 taxa, with 7-10 species mak ing up 40-50% of the total benthic bio mass (Denisenko, 2004;Wassmann et al., 2006). By comparison, using the same statistical methods with a large suite of data in the Chukchi Sea, we found fewer taxa were dominant, with eight species making up nearly 50% of the total ben thic biomass (or bioresources).

Environmental Factors Influencing Benthic Community Structure
The composition and biomass of macro benthos in the Chukchi Sea are deter mined by the locations of water masses and associated frontal zones, cur rents, seawater temperature and salin ity, and the content of organic carbon in the sediments and in the water col umn (Grebmeier et al., 1989;Feder et al., 2005Feder et al., , 2007. Seafloor topography, com bined with the current regime, is the crit ical factor influencing the characteristics of bottom sediments and ultimately the associated concentration of organic car bon in them (Gorshkova, 1975; also see Pisareva et al., 2015, in this issue).
Comparison of the distributions of bottom sediment grain size, total benthic biomass, and the biomass of the domi nant taxa in the Chukchi Sea shows that sediments with similar contents of sand, silt, and clay fractions are the most favor able for the development of zoobenthos settlements, and they are also character ized by a large amount of bioresources (see Figures 1b and 3a). Faunal abun dance and availability of food have major influences on the quantitative develop ment of the zoobenthos. Average annual primary production in the Chukchi Sea varies from 20 to 200 gC m -2 (Sakshaug, 2004;Codispoti et al., 2013). Regions with increased concentrations of sus pended and freshly deposited organic matter depend on the intensity of the nutrients supplied via currents from the Bering Sea as well as locations of frontal water masses (Grebmeier, 1993(Grebmeier, , 2012 and of the ice edge (Denisenko, 2002).
The Chukchi Sea is three to four times shallower (average depth of 50 m) than the western seas of the Russian Arctic, and the amount of detritus deposited on the bot tom from the surface layers of the water is thus higher. The shallow depths of the Chukchi Sea ensure abundant food is sup plied to benthic invertebrates (Grebmeier et al., 1989;Grebmeier, 1993;Iken et al., 2010). As a result, very high zoobenthos biomass, exceeding 1 kg m -2 , is formed even in the open sea areas (see Figure 3a). The bottom of the Chukchi Sea is a flat tened plain that dips slightly toward the center and has an average depth of less than 80 m. Contrary to other Arctic seas, there is a distinct dependence of zoobenthos distribution on phytoplankton biomass (Grebmeier, 1993(Grebmeier, , 2012. The bottoms of other Arctic seas (e.g., the Barents Sea) are highly uneven and characterized by a considerable (hundreds of meters) depth range between shallow banks and deep channels (average depth of 250 m). Thus, different areas of the seafloor receive dif ferent amounts of biologically degraded organic matter produced by phytoplank ton. In the Chukchi Sea, where depth vari ations are much less pronounced, there is tighter pelagicbenthic coupling of upper water column production to the under lying benthos. However, a recent synthesis by Grebmeier et al. (2015a) found a defi nite latitudinal decline in export produc tion and benthic biomass overall from the southern to the northern Chukchi Sea due to reduced levels of primary production and water mass variability. The complex water mass structure and current flow in the northern Chukchi Sea, with its shoals and canyons, adds to the heterogeneity of ecosystem dynamics and underlying ben thic populations in that region.

Potential Mechanisms for Maintenance of High Benthic Biomass Zones
The biomass maxima of benthic inverte brates recorded in the southern part of the Chukchi Sea (Grebmeier et al., 2006; underlying sediments that support the high bivalve biomass in this region. This southern Chukchi Sea gyre may exist as a persistent feature, regardless of the pres ence or absence of an opposed current, with only the size of the cyclonic gyre varying with variable current flows of the different water masses in the region.

Functional Feeding Groups
Our schematic maps indicate that the Chukchi Sea can be separated into two regions of almost equal size: (1) southern and northwestern with an active current regime, and (2) northern with slower, but more variable, nearbottom hydro dynamics (Figure 1). This hydrodynamic separation is reflected in the distribution of suspension feeders vs. deposit feed ers, whose nearly equal bioresources are divided in distribution and follow the hydrodynamic regimes (also see Pisareva et al., 2015, in this issue).
Hydrodynamics are most intense in the southern area just north of Bering Strait, along the Alaska and Chukotka coastlines, and within Herald Canyon to the east of Wrangel Island. Coarse sandy deposits (Figure 1b) on the seafloor in this region are better suited for attached organisms such as ascidians, barnacles, and several filterfeeding amphipods (Figure 5a-c). These areas are characterized by moder ately high values of zoobenthos biomass (Figure 3a). The presence of suspension feeders indicates high organic matter con tent in the nearbottom water layer. The suspended particulate matter may origi nate both autochthonously, resulting from the development of local phytoplank ton, and allochthonously, as organic mat ter transported into the Chukchi Sea via the Bering Strait inflow. Surface deposit feeders generally dominate in the upper Herald Valley and in the south central Chukchi Sea, with decreasing biomass to the northwest (Figure 6b).
The prevalence of subsurface deposit feeders in the northeastern and northern Chukchi Sea areas indicates a horizontal input of fresh organic matter. This func tional group is especially characteristic of eastern areas (Feder et al., 2005;Blanchard et al., 2013a,b;Blanchard and Feder, 2014) and depends on the transfer of the fresh organic matter from southern areas (Grebmeier et al., 2015a). A second region of organic matter accumulation is observed to the northwest in Herald Canyon (Kosheleva and Yashin, 1999). Increased concentrations of organic mat ter in the near bottom water layer along the canyon slopes ensure ample develop ment of filterfeeding organisms, while the accumulation and burial of organic matter in the bottom sediments provide food for subsurface deposit feeders.
Finally, the highest biomass of pred ators ( Figure 6d) coincides with areas of highest zoobenthos biomass (i.e., south east Chukchi Sea and in Herald Canyon; Figure 3a). The distributions of Chukchi Sea trophic groups identified in this study, and the fact that they are determined by environmental factors, agree well with the results of earlier studies carried out in the eastern part of the sea (Feder et al., 2005(Feder et al., , 2007. In the southern Chukchi Sea, as in the Pechora Sea, the shallow part of the Barents Sea, the distribution of fil ter feeders was not just confined to the coastal areas, as occurs in the Barents Sea (Denisenko, 2007), but they were often most abundant in the offshore regions.
Observations of almost equal propor tions of suspension and deposit feed ers are consistent with the findings of Neiman (1961), who concluded that such a distribution of trophic zones is char acteristic of seas with a broad and gen tly sloping continental shelf. According to Neiman (1988), the trophic structure of zoobenthos in the Chukchi Sea, on the whole, may be considered eutrophic. Such a structure usually develops due to a weakly balanced trophic cycle in the pelagic community where zooplankton are insufficient to consume the available organic carbon (Grebmeier et al., 2006).
Indeed, all the key zoobenthos spe cies of the Chukchi Sea have a broad or a cosmopolitan borealArctic distribu tion. Astarte borealis may reach an age of 26 years (Denisenko, 1997), Macoma calcarea 15-17 years (Petersen, 1978;Denisenko, 1997), Yoldia hyperborea 13 years (Rusanova, 1963), Golfingia margaritacea at least six years (Denisenko, 2006), and Maldane sarsi at least four years (Denisenko, 2006). A limited, but detailed, study of the growth rate of Y. hyperborea (Box 1) indicates that bivalve growth rate is tied to climate forcing parame ters, such as the Arctic Oscillation; thus, the future stability of bivalve populations may change. Since bivalves are a major bioresource in the Chukchi Sea for upper

Box 1. Relationship of Growth Patterns in a Chukchi Sea Bivalve Population to the Arctic Oscillation By Stanislav G. Denisenko and Vladimir V. Skvortsov
Forty live specimens of Yoldia hyperborea ( Figure B1), one of the dominant bivalve taxa in the Chukchi Sea, were collected at one station (CS12) on the Chukchi South line in the southern Chukchi Sea during the RUSALCA September 2012 expedition. These specimens were used to estimate individual growth rates by measuring annual increments of carbonate added to their shells. To test the hypothesis that climatic factors have a significant impact on the growth of mollusks, the following climatic parameters were examined in relation to measured growth rates: a 10-year time series of bottom water temperatures observed from moorings in Bering Strait (Temp), the annual flux of incoming water flowing through Bering Strait (FW), the Arctic Oscillation (AO), and the Pacific/North American (PNA) pattern (Wallace and Gutzler, 1981;Woodgate et al., 2012).
We plotted individual mollusk growth increment curves ( Figure B2) and calculated a theoretical growth curve using the Gompertz formula (Bowers et al., 1997). The Gompertz equation was used to describe the growth rates of the 40 individuals collected. This equation is based on growth rates (L), the growth coefficient (k), and the inflection point when growth rates change (i) ( Table B1). We also calculated the standardized growth rates of these mollusks, that is, the ratio of difference and derivative (RDD) (Denisenko, 2013). We determined that these standardized rates of growth for the mollusks collected (RDD) show long-term oscillations ( Figure B3).

Correlation of Growth Rates with Climatic Parameters
The statistical analysis of the relationship between the growth rates of the 40 Y. hyperborea and climatic factors was carried out using correlation analysis, multiple regression analysis, and time series analysis. Multiple regression analysis was performed in two ways: stepwise multiple regressions and ridge multiple regressions (using Statistica 6). We used both raw data (untransformed) and transformed data (ln (x+2)) to determine the best approximation for the dependence of the growth rate on climatic variables. Significant correlations (p <0.05) were found between the transformed values of the RDD and water temperature and the AO (partial correlations are 0.680 and 0.674, respectively, for the transformed data). As a result of multiple regression analysis undertaken in the model analysis, only one predictor-water temperature (Table B2)provided satisfactory results with good agreement between predicted and observed values ( Figure B4). The initial data were treated as a time series and analyzed based on cross correlation between the RDD index and the climatic parameters. Figure B5 shows good coherence between the variability of the RDD and the AO indices. However, shifting the relationship of the standardized growth rate of Y. hyperborea (RDD) relative to the AO index provides a better match if a two-year delay (R 2 = 0.782) is assumed (Table B3, Figure B6). The relationship between observed RDD and modeled RDD also improves when a two-year lag is assumed between the AO and the RDD ( Figure B7). The AO index is a measure of climatic influence, including water temperature (lag = one year, R 2 = 0.903) and water flow (lag = two years, R 2 = 0.964), as evidenced by the high values of the coefficient of determination. Therefore, we conclude that the AO index is the most reliable predictor of standardized rate of growth for clams in this population ( Figures B6 and B7).    trophicfeeding marine mammals such as walrus, timeseries studies on spe cific benthic animals, or at select sites (Grebmeier et al., 2015b, in this issue), are needed. Finally, there are no data on the maximum life span or the average age of Nuculana radiata, Ennucula tenuis, and Psolus peroni, but, judging from their sizes in the Chukchi Sea, it is likely they also live for at least several years. The implication of these findings is that, on the large spatial scale, benthic popula tions are stable, with slowly changing gen erations (Neiman, 1988). Our studies, reported here, support previous classifica tions of trophic characteristics and taxo nomic structures of the benthic fauna of the Chukchi Sea. However, we see impacts of changing environmental conditions on bivalve growth rates (Box 1) and declin ing bivalve biomass in the rich southern Chukchi Sea region (Grebmeier et al., 2015b, in this issue). These two results suggest that although the overall ben thic population structure may be stable on the large scale, smallscale studies are needed at specific sites to track the status of and changes in bivalve populations that are a critical component of the food web in the Chukchi Sea.

CONCLUSIONS
Chukchi Sea benthic ecosystems are among the most productive in the seas of the Russian and US sectors of the Arctic. The average biomass of zoobenthos in the Chukchi Sea is two times greater than in the Barents Sea. Because the Chukchi Sea is relatively shallow, pelagic benthic coupling of upper water column production is stronger than in the Barents Sea. Maximum values of zoobenthos bio mass in the Chukchi Sea were found in the southern region due to a possible cyclonic gyre where seasonal changes in hydrodynamics can result in high pri mary production, with phyto detritus being subsequently and efficiently exported to the underlying benthic com munities. The persistence of this hydro graphic feature is likely influenced by the northward flow of cold, nutrientrich Pacific water, acrossshelf differences in current speed, and variable bathymetric steering of these waters. Due to the result ing deposition of large amounts of fresh phytodetritus to the benthos, the propor tion of surface deposit feeders in the zoo benthos is rather high, while that of fil ter feeders is smaller than in the Barents Sea. Analysis of the spatial distribution of the key species may be useful not only for assessment of bioresources but also for studies of longterm fluctuations of zoo benthos under the influence of various natural and anthropogenic factors.