752 



Abstract— We used allozyme. mic- 

 rosatellite, and mitochondrial DNA 

 (nitDNA) data to test for spatial and 

 interannual genetic diversity in wall- 

 eye pollock [Theragra chalcogramma) 

 from six spawning aggregations rep- 

 resenting three geographic regions: 

 Gulf of Alaska, eastern Bering Sea, and 

 eastern Kamchatka. Interpopulation 

 genetic diversity was evident primar- 

 ily from the mtDNA and two allozyme 

 loci (SOD'2*, MPI*). Permutation tests 

 indicated that F^j. values for most allo- 

 zyme and microsatellite loci were not 

 significantly greater than zero. The 

 microsatellite results suggested that 

 high locus polymorphism may not be a 

 reliable indicator of power for detecting 

 population differentiation m walleye 

 pollock. The fact that mtDNA revealed 

 population structure and most nuclear 

 loci did not suggests that the effective 

 size of most walleye pollock popula- 

 tions is large (genetic drift is weak I 

 and migration is a relatively strong 

 homogenizing force. The allozymes and 

 mtDNA provided mostly concordant 

 estimates of patterns of spatial genetic 

 variation. These data showed signifi- 

 cant genetic variation between North 

 American and Asian populations. In 

 addition, two spawning aggregations 

 in the Gulf of Alaska, in Prince Wil- 

 liam Sound, and off Middleton Island, 

 appeared genetically distinct from 

 walleye pollock spawning in the She- 

 likof Strait and may merit manage- 

 ment as a distinct stock. Finally, we 

 found evidence of interannual genetic 

 variation in two of three North Ameri- 

 can spawning aggregations, similar 

 in magnitude to the spatial variation 

 among North American walleye pol- 

 lock. We suggest that interannual 

 genetic variation in walleye pollock 

 may be indicative of one or more of 

 the following factors: highly variable 

 reproductive success, adult philopatry, 

 source-sink metapopulation structure, 

 and intraannual variation (days! in 

 spawning timing among genetically 

 distinct but spatially identical spawn- 

 ing aggregates. 



An examination of spatial and temporal 

 genetic variation in walleye pollock 

 (Theragra chalcogrammd) 

 using allozyme, mitochondrial DNA, 

 and microsatellite data* 



Jeffrey B. Olsen 

 Susan E. Merkouris 

 James E. Seeb 



Gene Conservation Laboratory 



Alaska Department of Fish and Game 



333 Raspberry Road 



Anchorage Alaska 99518-1599 



E-mail address (lor James E Seeb, contact author! |im_seebmifishgame stale ak us 



Manuscript accepted 2.5 March 2002. 

 Fish. Bull. 100:7,52-764 (20021. 



Detecting spatial structure in the ge- 

 netic variation of some marine fishes 

 is challenging because populations 

 are often closely related due to high 

 gene flow, and the relationships be- 

 tween populations may change over 

 years (Hedgecock, 1994; Shaklee and 

 Bentzen, 1998; Waples, 1998). For 

 these species, independent population 

 studies of genetic variation may result 

 in conflicting evidence on the extent 

 and complexity of population structure 

 (McQuinn, 1997; Shaklee and Bentzen, 

 1998). Interpreting these results may 

 be further complicated if the loci and 

 classes of genetic markers differ by 

 study and year. In addition, the discor- 

 dant patterns of population structure 

 may reflect complex population biology 

 and ecology or merely differences in 

 the resolving power of loci and marker 

 classes to detect spatial structure. 

 The walleye pollock iTheragra chalco- 

 gramma) is a marine fish that typifies 

 this situation. 



Walleye pollock inhabit basin, slope, 

 and shelf waters of four major seas in 

 the North Pacific Ocean: Sea of Japan, 

 Sea of Okhotsk, Bering Sea, and Gulf 

 of Alaska (Bailey et al., 1997). Their 

 abundance is greatest in the eastern 

 Bering Sea and the Sea of Okhotsk, but 

 large aggregations are also found in 

 the Sea of Japan, western Bering Sea, 

 and the Gulf of Alaska. Smaller ag- 

 gregations occur along coastal waters 

 in bays and fjords such as in Prince 

 William Sound, Alaska, and Puget 



Sound in Washington State. Popula- 

 tion boundaries for walleye pollock 

 may be correlated with the margins 

 of spatially distinct spawning aggre- 

 gates that occur throughout the species 

 range in predictable locations during 

 late winter and early spring (Bailey 

 et al., 1999). Many of these spawning 

 aggregates can be distinguished by 

 spawning time and habitat, and by 

 meristic and morphometric characters 

 (e.g. Iwata, 1975b; Hinckley, 1987; Mul- 

 ligan et a!., 1989). Nevertheless, genet- 

 ic support for discrete populations is 

 equivocal, and the extent of population 

 structuring in walleye pollock remains 

 a controversial and unresolved issue 

 for management and consei-vation in 

 U.S. and international waters (Bailey 

 etal., 1999). 



Independent attempts to define ge- 

 netic population structure in walleye 

 pollock on a regional scale have shown 

 mixed results. Some allozyme loci show 

 genetic differentiation between walleye 

 pollock from contiguous sea basins (e.g, 

 eastern Bering Sea and Gulf of Alaska; 

 Grant and Utter, 1980), and a single 

 locus (SOD) appears to reveal a major 

 east-west division between Asian and 

 North American populations (Iwata, 

 1975a, 1975b; Grant and Utter, 1980). 

 This major division is supported by a 



' Contribution PP-217 of the Alaska Depart- 

 ment of Fish and Game, Division of Com- 

 mercial Fisheries, Juneau, AK 99802-5526. 



