Donald M. Van Doornik

Genetic Stock Identification of Steelhead in the Columbia River Basin: An Evaluation of Different Molecular Markers

Abstract Protein genetic markers (allozymes) have been used during the last decade in a genetic stock identification (GSI) program by state and federal management agencies to monitor stocks of steelhead Oncorhynchus mykiss in the Columbia River basin. In this paper we report new data for five microsatellite and three intron loci from 32 steelhead populations in the three upriver evolutionarily significant units (ESUs) and compare the performance of allozyme, microsatellite, and intron markers for use in GSI mixture analyses. As expected, microsatellites and introns had high total heterozygosity (H T) values; but there was little difference among marker classes in the magnitude of population differentiation as estimated by Wrights fixation index (F ST), which ranged from 0.041 (microsatellite loci) to 0.047 (allozyme loci) and 0.050 (intron loci). For allozyme and microsatellite loci, the relationships among populations followed the patterns of geographic proximity. In computer-simulated mixture analyses,...

A centralized model for creating shared, standardized, microsatellite data that simplifies inter-laboratory collaboration

We demonstrate an efficient model for standardizing microsatellite DNA data among laboratories studying Oncorhynchus mykiss. Eight laboratories standardized 13 microsatellite loci following allele nomenclature of a central laboratory (average inter-laboratory genotyping concordance >98%). Following this central model, we have currently standardized 298 alleles from throughout the species native range. Although we focus here on O. mykiss, our experiences and recommendation apply equally to other broadly distributed species that may benefit from multi-laboratory collaborative data collection.

Genetic population structure of central Oregon Coast coho salmon (Oncorhynchus kisutch)

We surveyed microsatellite variation from 22 spawning populations of coho salmon (Oncorhynchus kisutch) from the Oregon Coast to help identify populations for conservation planning. All of our samples were temporally replicated, with most samples obtained in 2000 and 2001. We had three goals: (1) to confirm the status of populations identified on the basis of spawning location and life history; (2) to estimate effective population sizes and migration rates in order to determine demographic independence at different spatial scales; and (3) to determine if releases of Washington hatchery coho salmon in the 1980s into Oregon Coast streams resulted in measurable introgression into nearby wild Oregon Coast coho populations. For the last question, our study included a hatchery broodstock sample from 1985, after the Puget Sound introduction, and a 1975 sample taken from the same area prior to the introduction. Our results generally supported previously hypothesized population structure. Most importantly, we found unique lake-rearing groups identified on the basis of a common life-history type were genetically related. Estimates of immigrant fraction using several different methods also generally supported previously identified populations. Estimates of effective population size were highly correlated with estimates of spawning abundance. The 1985 hatchery sample was genetically similar to contemporary Washington samples, and the contemporary Oregon Coast samples were similar to the 1975 Oregon Coast sample, suggesting that introductions of Washington coho salmon did not result in large scale introgression into Oregon populations.

An Update of Genetic Stock Identification of Chinook Salmon in the Pacific Northwest: Test Fisheries in California

Analyzing the stock composition of mixed-stock fisheries using genetic stock identification (GSI) procedures was developed for chinook salmon in the early 1980s when an incipient baseline was focused in the Columbia River basin. The current coastwide baseline of allozyme (protein) genetic loci includes approximately 75% of the major contributing populations from California to western Alaska for over 30 polymorphic loci. We review the 2-decade history of GSI studies of chinook salmon in fishery applications in the Pacific Northwest. By summarizing GSI results for four test fisheries in California, we demonstrate the use of a regional segment of the current baseline to monitor critically depleted stocks. Average FST is 0.099, with 12 loci FST > 0.05, in a 28 locus/53 population dataset from southern Oregon to southern California. Seven stock groups are recognized based on a multilocus pattern of differentiation that coincides with watersheds and coastal affiliations. Simulated mixture analyses indicate that stock groups are well resolved: percent correct assignment is 83% for Central Valley spring-run, but >92% for all other groups. Simulated mixture analyses also indicated that it is difficult to distinguish between mixtures with low levels (<1%) of Central Valley winter- or spring-run stocks given the present dataset. GSI estimates for four test fisheries indicate that Central Valley fall- and late fall-run chinook salmon comprised the majority of each mixture (89 to 95%). Critical or endangered stock groups were detected in test fisheries in the 1997 Point Conception fishery (3% Sacramento River winter-run) and in the 1999 Bodega Bay fishery (2.8% Upper Klamath-Trinity rivers). Preliminary regional baselines for intron and microsatellite loci show promise for added stock discrimination among chinook salmon populations. GSI projects are increasingly involving multiple agencies and using multicharacter procedures. Continued GSI

Genetic diversity in the Snake River sockeye salmon captive broodstock program as estimated from broodstock records

Snake River sockeye salmon spawning in Redfish Lake, Idaho are one of the most endangered taxa of Pacific salmon. The wild population nearly went extinct in the 1990s, and all surviving fish were incorporated into a captive broodstock program at that time. We used pedigree analysis to evaluate the effectiveness of the breeding program in retaining genetic variation from 1991 through 2008. Broodstock records document which males were crossed with which females, but fish from multiple crosses were frequently raised in the same tank so the exact pedigree of the population is unknown. Therefore, a simulation-based approach was used to estimate how much genetic diversity was retained by this breeding program. Results indicate that in 2008, after 5.5 generations of breeding, the average inbreeding coefficient was probably about 0.056. We estimated the inbreeding effective population size to be 41 over the entire program and 115 for the most recent generation. This amount of inbreeding is substantially less than has occurred in many high-profile captive breeding programs. Our results depend on several assumptions regarding the relatedness of fish in the breeding program, but simulations suggest our main results are relatively insensitive to these assumptions.

Patterns of Temporal Genetic Variation in Coho Salmon: Estimates of the Effective Proportion of 2-Year-Olds in Natural and Hatchery Populations

Abstract The majority of coho salmon Oncorhynchus kisutch south of central British Columbia spawn when they are 3 years old, so gene flow among brood years occurs predominately when precocious 2-year-old males (jacks) spawn successfully. To determine how much gene flow among brood years is occurring, we evaluated temporal differences in allozyme allele frequencies in four coho salmon populations in Washington State. Forty-three loci were polymorphic in the 26 samples, totaling 2,328 fish, that were collected from three naturally spawning populations and one hatchery population over 7 years. Our results indicate a greater degree of geographic variation among populations than temporal variation within populations, suggesting that population genetic analyses that depend on spatial variation can be done on natural coho salmon populations even in the absence of temporal sampling. Using a modification of the temporal method for estimating effective population size, we estimated the effective proportion of 2-yea...

Multiple paternity is prevalent in Pacific ocean perch (Sebastes alutus) off the Oregon coast, and is correlated with female size and age

The need to rebuild Pacific ocean perch, Sebastes alutus, populations on the west coast of the United States has precipitated a need to better understand the life history characteristics of this rockfish species. One such characteristic is mating behavior, which has the potential to influence the amount of genetic diversity in a population. We documented and examined the frequency of multiple mating in Pacific ocean perch collected off the Oregon coast using five microsatellite loci. We found that 47 of 66 (71.2%) females examined had broods sired by multiple males. The mean number of sires per brood was 1.92 (SD = 0.76) and ranged from 1–4. Polyandrous females were significantly larger and had an older average age than monogamous females. Our results suggest that polyandrous behavior among female Pacific ocean perch off the coast of Oregon is prevalent, is related to female size and age, and should be preserved by maintaining a natural age structure in this population.

Spawning by Female Chinook Salmon Can Be Detected by Electromyogram Telemetry

Abstract New methods to detect spawning of anadromous salmonids in their natural environment are needed to improve understanding of breeding behavior patterns, natural selection on reproductive traits (e.g., spawn timing), and interactions between artificially propagated and wild fish. We implanted maturing female Chinook salmon Oncorhynchus tshawytscha with coded electromyogram (CEMG) transmitters and continuously recorded spawning activity to develop an algorithm capable of accurately detecting spawning events from CEMG data. Marked increases in female digging frequencies immediately after spawning (cover digging) strongly correlated with CEMG values. The algorithm detected averages of 65% and 86% of the actual spawning events in 2003 and 2004, respectively. The algorithm accurately detected zero spawning events for the two female salmon that did not spawn. The presence of CEMG transmitters did not affect the digging frequency, number of nests constructed, or the reproductive life span of implanted fish...

The Effects of Exercise on Behavior and Reproductive Success of Captively Reared Steelhead

Abstract Steelhead Oncorhynchus mykiss collected as eyed eggs from the Hamma Hamma River, Washington, and reared in high (∼1 body length/s) or low current velocity (∼0.25 body lengths/s) were tested for differences in reproductive behavior in both an experimental spawning channel and their natal river. We conducted continuous (24-h/d) behavioral observations in the spawning channel and applied DNA pedigree analyses to explain the variation in individual reproductive success. Female steelhead reared in high and low current velocities did not exhibit differences in reproductive behavior in the spawning channel, but females reared in low current velocity were more frequently observed constructing nests in the Hamma Hamma River. Males reared in low-velocity tanks were more frequently observed courting females in the spawning channel and in the river. A pedigree analysis revealed no significant differences in reproductive success between the rearing treatments. Male reproductive success in the spawning channel...

Allozyme Studies of Pacific Salmonids with Nonlethal Sampling of Fin Tissue

Abstract We evaluated the use of nonlethal sampling of bony fin tissue to obtain allozyme data for Pacific salmonids Oncorhynchus spp. Enzyme activity was detected for 53 loci, including many polymorphic loci, in caudal fin tissue sampled from chinook salmon O. tshawytscha, chum salmon O. keta, coho salmon O. kisutch, sockeye salmon O. nerka, and steelhead O. mykiss. This represents 61% of all the loci (N = 87) routinely analyzed using eye, heart, liver, and muscle tissue. Anal, dorsal, caudal, pelvic, and pectoral fin tissues all expressed the same loci with approximately the same strength of activity. Fin tissue sampled from adult salmonids yielded a slightly greater number of loci than did juvenile samples. Sufficient allozyme activity was observed for 9 mg of juvenile chinook salmon and steelhead fin tissue for some loci; however, the minimal amount of tissue needed to obtain all detectable loci was 54 mg. We performed computer simulations to test the feasibility of using allozyme data from fin tissue...