Spatial and Temporal Variability in Size at Maturity of Walleye Pollock in the Eastern Bering Sea

—Size at maturity is incorporated into stock assessments of walleye pollock Theragra chalcogramma. For eastern Bering Sea walleye pollock, however, this important biological variable has not been examined since 1976 and possible interannual and geographic variations have never been evaluated. Maturity condition and fish length data were collected from 10,197 walleye pollock aboard factory trawlers during winter 2002 and 2003. We also analyzed a smaller, previously unanalyzed data set on maturity collected annually by National Marine Fisheries Service scientists during eight echo integration–trawl surveys over 1989–2002. Length at 50% maturity (L 50 ) was estimated by year and area by logistic regression using maximum likelihood methods. Spatial (within subareas) and temporal (interannual) variability were found. For instance, fish matured at the smallest lengths north of the Pribilof Islands and in the years 1989, 1991, and 1995 and at the largest lengths in the southeastern Bering Sea and in the years 2001 and 2002. We found evidence that the variability in size at maturity was related to biological mechanisms. Length at 50% maturity appeared to be directly related to growth as measured by length at age. Moreover, an inverse relationship between walleye pollock biomass and L 50 suggests that growth is density dependent. This spatiotemporal variability in size at maturity has implications for estimates of spawning stock biomass and should be incorporated into the annual stock assessments for setting annual catch specifications. The goal of this study was to estimate the spatial and temporal variability in the size of maturity of walleye pollock Theragra chalcogramma in the eastern Bering Sea. Until now, published maturity schedules for this important commercial species have come from a single year (1976) of observations pooled over a large area of the eastern Bering Sea shelf (Bakkala and Smith 1978). Although some data were collected over a limited portion of the eastern Bering Sea during echo integration–trawl surveys during 1989–2002, these data have not yet been analyzed for potential geographic or interannual shifts in size at maturity. Walleye pollock are a subarctic species ranging from the northwest Pacific Ocean, including the Sea of Japan, the Okhotsk Sea, and the Commander Islands, north through the Bering and Chukchi seas and south into the Gulf of Alaska to Puget Sound (and rarely as far south as central California) (Mecklenburg et al. 2002). Population structure remains somewhat unresolved within the eastern Bering Sea. Recent studies of allozymes, microsatellites, and mitochondrial DNA and analyses of DNA sequence data from the pantophysin locus indicate weak structuring over large geographical distances, conforming to an isolation-by-distance pattern of neutral divergence. This suggests either that gene flow is high or that the genetic drift since separation has been insufficient for significant levels of differentiation to accumulate (Olsen et al. 2002; O’Reilly et al. 2004; Canino et al. 2005). Morphometrics and life history characteristics imply that there are three to five stocks in the eastern Bering Sea (Hinckley 1987; Dawson 1994), and length at age (Hinckley 1987), juvenile otolith composition (Mulligan et al. 1992), and size composition and growth (Lynde et al. 1986; Shuck 2000) differ among one or more subareas. For fishery management purposes, walleye pollock from the U.S. portion of the Bering Sea are assessed and managed as three separate stocks: the eastern Bering Sea shelf from Unimak Pass to the U.S.–Russia Convention line, the Aleutian Islands shelf region (1708W to the U.S.–Russia Convention line), and the central Bering Sea–Bogoslof Island region (includes the Aleutian Basin) (Ianelli et al. 2005). Our study focuses on walleye pollock collected during National Marine Fisheries Service (NMFS) fishery-independent surveys and from commercial fisheries targeting the eastern Bering Sea shelf stock. Because our spatial analyses subsequently revealed that some samples were actually collected off the continental shelf in basin waters, we also examined our results with respect to stock delineation, as these results could have broader implications for fishery stock assessment and management. * Corresponding author: [email protected] 1 Present address: Alaska Department of Fish and Game, Division of Commercial Fisheries, Post Office Box 110024, Juneau, Alaska 99811, USA. Received April 27, 2007; accepted March 24, 2008 Published online September 29, 2008 1543 Transactions of the American Fisheries Society 137:1543–1557, 2008 Copyright by the American Fisheries Society 2008 DOI: 10.1577/T07-099.1 [Article]