Summer phytoplankton production and transport along the shelf break in the Bering Sea

-A general model is presented for the production and fate of phytoplankton during summer in two regions over the continental shelf of the Bering Sea. We propose that both regions of productivity are supported by nutrients transported into the area with the Bering Slope Current and that the fate of the phytodetritus produced is significantly affected by advection. We hypothesize that one system of primary productivity is initiated at the Bering Sea shelf-break front and continues into the northern Bering Sea as part of the modified Bering Shelf water mass. Phytodetritus produced in this system is transported north through Anadyr and Shpanberg Straits and we estimate that in 1987 it supplied 26% of the daily carbon demand of the benthos in the Chirikov Basin. The second region of primary productivity is located in the northern Bering Sea. Nutrients from the Anadyr Current, the northern branch of the bifurcated Bering Slope Current, support a highly productive phytoplankton bloom throughout the summer. Phytodetritus produced in this surface bloom is probably advected into the southern Chukchi Sea and deposited in the sediments. I N T R O D U C T I O N OUR understanding of the magnitude, location and fate of primary production in the vicinity of the Bering Sea continental shelf is increasing rapidly. Results from recent studies in the region suggest that processes contributing to annual production on the shelf include: production due to epontic ice algae (McRov and GOERING, 1974); spring iceedge blooms (NIEBAUER and ALEXANDER, 1985), which follow the receding ice edge from deep water onto the shelf; spring blooms over the shelf, which are maintained into the summer by occasional wind mixing events (SAMBROTFO et al., 1986); spring blooms at the shelf-break front (IvERSON et al., 1979a); and summer production on the northeastern shelf (SPRINGER, 1988). The northern Bering and southern Chukchi Seas have been the subject of extensive oceanographic surveys since 1985 as part of the Inner Shelf Transfer and Recycling (ISHTAR) program. The program was designed to assess the fate of the dissolved and particulate phases of carbon, nitrogen, phosphorus and silicon in the northern Bering and southern Chukchi Seas (WALSH et al., 1989). A primary motivation for this effort * Institute of Marine Science, School of Fisheries and Ocean Sciences, University of Alaska Fairbanks, Fairbanks, AK 99775-1080, U.S.A. % Correspondence and present address: Institute of Marine Sciences, University of California, Santa Cruz, CA 95064, U.S.A. :~ Department of Marine Science, University of South Florida, St. Petersburg, FL 33701, U.S.A. § School of Oceanography, WB-10, University of Washington, Seattle, WA 98195, U~S.A. II University of Texas, Marine Science Institute, Port Aransas, TX 78373, U.S.A. 1085 1086 D.A. HANSELL et al. was the desire to understand the processes sustaining the rich ecosystem known to exist in this subpolar region. Field studies conducted over the past few decades have resulted in a nearly comprehensive understanding of the regional physical oceanography (COACHMAN et al . , 1975). A survey of primary productivity in the area (SAMBROTrO et al . , 1984) provided an initial estimate of the potential magnitude of primary productivity but an incomplete explanation for the processes controlling it. Early in the ISHTAR program, regions exhibiting major phytoplankton blooms were mapped, and rates of primary production quantified, in the northern Bering and southern Chukchi Seas (SPRINGER, 1988). These highly productive areas were assumed to be the major sources of organic carbon and nitrogen necessary to support the rich highertrophic populations known to thrive in the Chirikov Basin, of the northern Bering Sea, and in the southern Chukchi Sea (GREBMEIER, 1987; SPRINGER, 1988). However, the accepted paradigm for the production and fate of organic matter did not, in our view, completely survive close scrutiny. For instance, a spatial discrepancy existed between the location of the major bloom in the northern Bering Sea (SPRINGER, 1988) and the location of maximum benthic biomass in the same region (GREBMEIER et al . , 1988). The highest concentrations of benthic biomass were located 20-30 km to the east of the major phytoplankton bloom. It did not appear physically possible, in this strong northerly advective system, to deposit organic carbon and nitrogen such a distance to the east of the bloom. It was clear that we had not fully elucidated the sources and fate of particulate organic nitrogen and carbon over the northern Bering Sea shelf. Earlier studies of the southeastern Bering Sea shelf (PROBES) resulted in models of the general circulation (COACHMAN, 1986) and primary productivity patterns (SAMBROTTO et al . , 1986) upstream of the ISHTAR study area. IVERSON et al. (1979b) noted that a spring chlorophyll maximum found over the shelf-break front in the southeastern Bering Sea persisted for about 1 month, slowly sinking at a rate of about 1 m d -1. We propose that this shelf-break system of productivity continues along the front to the northwest toward Cape Navarin, and with the Bering Slope Current, extends across the Gulf of Anadyr and through Anadyr and Shpanberg Straits. This production and transport process is analogous to a continuous culture system where, as the plants are transported and grazed, a ready supply of nutrients maintain the phytoplankton at high levels. MALONE et al. (1983) described a similar production and transport process at the shelf break of the New York Bight. They found that development of stratification in nutrientrich offshore water between storm events results in high growth rates and biomass near the surface on the shelf side of the shelf-break front. During the summer, and as we propose to be the case at the Bering Sea shelf break, plant growth occurred at the pycnocline. Unlike the Bering Sea, however, where transport is from the shelf-break front to the inner shelf, up to 35% of annual production over the continental shelf of the New York Bight is exported from shelf to slope water. A second important region of summer production is associated with the western boundary current (the northern extension of the Bering Slope Current known as the Anadyr Current) south of Bering Strait. Nutrients from the Anadyr Current support intense surface production following passage through Anadyr Strait (SPRINGER, 1988). We propose that this second region of productivity results from continuation of the system responsible for production at the shelf break and, further, that the phytodetritus production in this region is transported north through Bering Strait for deposit in the southern Chukchi Sea. Summer phytoplankton production 1087 We support the hypotheses presented here with data collected over the last decade from spatially disparate regions of the Bering and Chukchi Seas. This large area is by no means static, being subject to both seasonal and interannual variability. As a result, our interpretation of the data and the model presented are subject to the uncertainty imposed by this variability. M E T H O D S Data was collected during cruises on the R.V. Alpha Helix (July 1985 and October 1987) and the R . V . T . G . Thompson (July-August 1987) as part of the Inner Shelf Transfer and Recycling (ISHTAR) project. The study area included the regions surrounding St. Lawrence Island in the south to approximately 69°N latitude and from the Alaskan coast in the east to the United States-Soviet Union 1867 Convention Line in the west (Fig. 1). Water column conductivity and temperature were determined with a Neil Brown CTD. A Turner-Designs fluorometer and an acetone extraction procedure