Washington State Department of Fish and Wildlife Fish Program, Science Division Genetics Lab

Washington State Department of Fish and Wildlife Fish Program, Science Division Genetics Lab 19 June 2003 To: Curt Leigh, WDFW Frank C. Shrier, PacifiCorp Diana Gritten-MacDonald, Cowlitz PUD From: Janet Loxterman Subject: Lewis River Oncorhynchus mykiss I evaluated the genetic structure of rainbow trout and steelhead stocks in the Lewis River watershed to determine 1) the genetic composition of stocks in the Lewis River watershed, 2) how the different stocks are related to one another, and 3) the most suitable steelhead stock for reintroduction into the upper Lewis River. To address these questions, native rainbow trout and steelhead were sampled from several localities in the Lewis River watershed (Table 1). For comparative purposes, these native samples were analyzed in conjunction with samples from various hatchery strains. The additional stocks included: Goldendale Hatchery rainbow trout, South Tacoma Hatchery rainbow trout, Spokane Hatchery rainbow trout, and Merwin Hatchery winter and summer run steelhead (Table 1). Microsatellite Data Collection and Analysis DNA was extracted from 837 tissue samples representing five rainbow trout subpopulations, six steelhead subpopulations, and three rainbow/steelhead subpopulations (Table 1). Nine microsatellite loci were assayed using fluorescently labeled primers following multiplex protocols developed in the WDFW Genetics Lab (Table 2). Genotypes were generated from the resulting PCR products using an Applied Biosystems (ABI) 3730 automated sequencer. Microsatellite alleles were sized using an internal GS 500 LIZ (ABI) size standard. I used GENEMAPPER (Version 3.0) software to collect and analyze the microsatellite data. General measures of within-population genetic diversity including average heterozygosity, allelic richness, and estimates of inbreeding were computed for each subpopulation using FSTAT (Version 2.9.3.2, Goudet 1995). Tests for Hardy-Weinberg proportions for each locus and genotypic linkage disequilibrium between all pairs of loci within each subpopulation were performed using GENEPOP (Version 3.3, Raymond and Rousset 1995) and statistical significance was evaluated using a Bonferroni correction of P-values (Rice 1989). To assess population structure, I computed several pairwise estimates of genetic differentiation between subpopulations. I generated estimates of genotypic population Page AQU 15-1

differentiation using GENEPOP 3.3. In addition, I used ARLEQUIN (Version 2.000, Schneider et al. 2000) to compute measures of population subdivision between all pairs of subpopulations. These estimates use allelic and genotypic frequency data to assess differences between subpopulation pairs. Statistical significance of F ST estimates was tested using 10,000 permutations and was evaluated using a Bonferroni correction of P- values (Rice 1989). I performed various jackknife assignments to evaluate the cohesiveness of the subpopulations. I used WHICHRUN (Version 4.1, Banks and Eichert 2000) to implement a jackknife procedure that removes each fish from the dataset, recalculates allele frequencies of the baseline subpopulations, and assigns the fish to the most likely group. In addition to jackknifing the baseline subpopulations, I used WHICHRUN to assign individuals from the three rainbow/steelhead subpopulations to the most likely baseline subpopulation to identify possible relationships between the mixed samples and the baseline subpopulations. Finally, genetic distance between subpopulation pairs was estimated using the Cavalli- Sforza and Edwards chord distance (Cavalli-Sforza and Edwards 1967) as performed in MICROSAT (Version 1.5, Minch et al. 1997). Distance was estimated first between all pairs of subpopulations and then for only the known pure rainbow trout and steelhead subpopulations (excluding the rainbow/steelhead subpopulations). The distance matrices (1000 bootstrap matrices) were used to construct neighbor-joining trees using the NEIGHBOR function as implemented in PHYLIP (Version 3.572, Felsenstein 1993). Bootstrap consensus trees were constructed using the CONSENSE option in PHYLIP. Trees were imported and drawn with their associated bootstrap values using the TREEVIEW program (Version 1.6.5, Page 1996). Results and Discussion All nine microsatellite loci were polymorphic with the number of alleles ranging from 8 (Ots-103) to 36 (One-108). Average heterozygosity ranged from 0.5962 in South Tacoma Hatchery rainbow trout (STcHat) to 0.8210 in Muddy River rainbow/steelhead (MRrbst). In general, the lowest heterozygosity values were found in the hatchery collections and the highest in two rainbow/steelhead subpopulations (Table 3). Lower allelic richness and heterozygosity in the hatchery subpopulations was not unexpected since hatchery populations tend to exhibit reduced genetic diversity. Higher values for the Lewis River rainbow/steelhead (LRrbst) and MRrbst subpopulations are likely related to these groups being a combination of different subpopulations and potentially different forms of O. mykiss. In addition, the increased estimates of heterozygosity and allelic richness in the MRrbst group may reflect the relatively small sample size (N = 33). Another within subpopulation estimate is F IS, which provides an indication of the degree of inbreeding in each subpopulation. Estimates of F IS for these subpopulations indicated significant levels of inbreeding in five subpopulations, one rainbow trout subpopulation, one steelhead subpopulation, and the three rainbow/steelhead subpopulations (Table 3). Like the estimates of allelic richness and heterozygosity, the significant F IS in the LRrbst Page AQU 15-2

group and MRrbst groups likely reflects the mixed nature of these subpopulations. While significant F IS in MRrbst and LRrbst may not indicate a significant level of inbreeding in these subpopulations, the significant F IS in Pine Creek rainbow trout (PinCrb), EF Lewis River winter-run steelhead (EFLRWR), and EF Lewis River rainbow/steelhead (96EFLR) suggests some degree of inbreeding is occurring in these subpopulations. Tests for Hardy-Weinberg equilibrium revealed some deviations from expected proportions (Table 4). Significant deviations in the PinCrb, LRrbst, MRrbst, and 96EFLR are not surprising given the significant F IS estimates in these subpopulations. Similarly, tests for linkage disequilibrium between pairs of loci within each subpopulation indicated some significant disequilibrium (Table 5). However, except for the Merwin Hatchery winter-run steelhead (MHatWR) subpopulation, most subpopulations exhibiting significant disequilibrium had only one or two pairs of loci deviating from equilibrium. This pattern does not likely reflect physical linkage between loci since linkage disequilibrium was not present in all subpopulations. Rather, this pattern may result from small effective population sizes, admixture among populations, population bottlenecks, or merely by chance due to the large number of multiple comparisons (504). Since the deviations in Hardy-Weinberg proportions and linkage equilibrium were distributed among subpopulations and loci, all loci and subpopulations were retained for analysis. Assignment tests for the known subpopulations (excluding the rainbow/steelhead groups) provided information regarding the relationships among the different subpopulations (Table 6). All five hatchery subpopulations had greater than 75% of the fish assign back to the hatchery of origin. The PinCrb had a large proportion assign back to PinCrb (50%) with the remaining proportion assigning primarily to the GdnHat and Spokane Hatchery (SpoHat) subpopulations. This subpopulation, located above Swift Reservoir, has been supplemented with hatchery stocks, which is clearly reflected in the assignment test. Conversely, the NF Lewis River rainbow trout (NFLRrb) subpopulation does not reflect a large hatchery influence with 85% of the fish assigning back to the NFLRrb subpopulation and only one fish assigning to a hatchery subpopulation. In general, the rainbow trout subpopulations had a large proportion of individuals assigning to the original subpopulation with some individuals assigning to other rainbow trout subpopulations. The pattern among the steelhead subpopulations was less consistent. Most of the steelhead subpopulations cross-assigned with several other subpopulations (Table 6). The EF Lewis River summer-run steelhead (EFLRSR) had several individuals assigned to the Merwin Hatchery summer-run steelhead (MHatSR) subpopulation, suggesting a hatchery influence in this subpopulation. Similarly, the two Cedar Creek subpopulations (96CeWR and 03CeWR) had several fish cross-assign to the MHatWR subpopulation, as well as to one another. The remaining steelhead subpopulation (EFLRWR) was largely assigned back to EFLRWR with a large component assigned to the EFLRSR subpopulation. Page AQU 15-3

In addition to examining the relationships among the baseline subpopulations, assignment tests were performed using the three rainbow/steelhead subpopulations as unknown. The results of this analysis are quite revealing (Table 6). All three subpopulations had individuals assigned to several different subpopulations. Not surprisingly, fish from two of the three mixed subpopulations were assigned both to steelhead subpopulations and to rainbow trout subpopulations. Based on the assignment test, the LRrbst subpopulation has both rainbow trout and steelhead components with a large proportion of the LRrbst assigned to the GdnHat subpopulation, suggesting a hatchery influence. Conversely, the MRrbst and 96EFLR groups appear to have little hatchery influence. However, while the MRrbst was divided among rainbow trout and steelhead subpopulations, the 96EFLR subpopulation was largely assigned to the EFLRSR subpopulation with the remaining individuals assigned to other steelhead subpopulations. While the results from the assignment test provide some information about the relationships among the different subpopulations, examining the degree of population differentiation using allele frequency and genotype differences among the different subpopulations can further assess the strength of these interrelationships among subpopulations. Pairwise tests of population differentiation indicated strong population genetic structuring among these rainbow trout and steelhead subpopulations (Tables 7 and 8). Except for the two Cedar Creek steelhead subpopulations, all subpopulation pairs exhibit significant differences in allele frequencies (Table 7). Furthermore, 77 of 91 pairwise comparisons of genetic differentiation were statistically significant (Table 8). Twelve of the 14 comparisons that were not significant involved one of the rainbow/steelhead subpopulations. Like the allele frequency differences, the remaining non-significant comparisons of F ST involved the Cedar Creek subpopulations and the winter and summer run EF Lewis River subpopulations (EFLRSR and EFLRWR). Based on these results, there is little or no genetic exchange occurring among these subpopulations of rainbow trout and steelhead resulting in significant population subdivision among subpopulations. The neighbor-joining tree is another means to illustrate interrelationships among subpopulations (Fig. 1). Like the assignment test, both trees group PinCrb, GdnHat, and SpoHat together with strong bootstrap support (99%), reflecting the hatchery influence in this subpopulation. In addition, the EFLRSR subpopulation is strongly associated with the MHatSR subpopulation, and these subpopulations grouped with EFLRWR with 99% bootstrap support, indicating a hatchery influence in these subpopulations as well. There is also strong support for grouping the two Cedar Creek subpopulations (92% bootstrap support). The STcHat subpopulation is not grouped with any of the subpopulations suggesting this hatchery subpopulation is not largely used to supplement populations in the Lewis River watershed. A more interesting result is the NFLRrb subpopulation, which does not group with any of the hatchery subpopulations or the rainbow trout subpopulations (100% bootstrap support). In conjunction with the other analyses, this result strongly suggests that this rainbow trout subpopulation is genetically distinct and possibly represents a native stock. When the rainbow/steelhead subpopulations were incorporated into the neighbor-joining tree (Fig. 1b), the LRrbst and MRrbst group Page AQU 15-4

together with 78% bootstrap support, while the 96EFLR subpopulation has little bootstrap support. Conclusions In general, these subpopulations of rainbow trout and steelhead are genetically distinct with minimal gene flow occurring among different stocks. The results do reflect some hatchery influence in the Pine Creek rainbow trout subpopulation, as well as in the Cedar Creek and EF Lewis River steelhead subpopulations. Further, except for the hatchery subpopulations and the Pine Creek rainbow trout subpopulation, these subpopulations of rainbow trout and steelhead do not appear to be genetically depauperate indicating that they are currently not being strongly influenced by genetic drift or small effective population size. An important result, particularly relevant in management and possible supplementation efforts in the Lewis River watershed, is the apparent genetic distinction in the NF Lewis River rainbow trout. While this subpopulation represents a genetically distinct stock, it is unclear as to whether this stock represents a native rainbow trout population or a residualized steelhead stock that was landlocked when the dams were constructed. In either case, since resident rainbow trout and steelhead do interbreed (Currens et al. 1990, Leider et al. 1995, Busby et al. 1996), the introduction of non-native steelhead (i.e. Merwin Hatchery) into the NF Lewis River could compromise the genetic integrity of the NF Lewis River O. mykiss subpopulation. Although, using NF Lewis River O. mykiss as broodstock for any supplementation effort could mitigate this effect. Acknowledgments I would like to thank Jennifer Von Bargen, Alice Pichahchy, Cherril Bowman, and Norm Switzler for their technical contributions in the laboratory. I gratefully acknowledge WDFW staff and other Lewis River Relicensing Participants for collecting the tissue samples. The DNA data collection and analyses described herein were conducted with funds obtained from PacifiCorp. Page AQU 15-5

Literature Cited Banks, M. A., M. S. Blouin, B. A. Baldwin, V. K. Rashbrook, H. A. Fitzgerald, S. M. Blankenship, and D. Hedgecock. 1999. Isolation and inheritance of novel microsatellites in chinook salmon (Oncorhynchus tschawytscha). Journal of Heredity 90:281-288. Banks, MA and Eichert W (2000) WHICHRUN (Version 3.2) a computer program for population assignment of individuals based on multilocus genotype data. Journal of Heredity. 91:87-89. Busby P. J., T. C. Wainwright, G. J. Bryant et al. 1996. Status review of west coast steelhead from Washington, Oregon, and California. U.S. Department of Commerce, NOAA-NWFSC Technical Memorandum 27. National Marine Fisheries Service, Seattle. Cavalli-Sforza, L. L., and A. W. F. Edwards. 1967. Phylogenetic analysis: models and estimation procedures, American Journal of Human Genetics 19:233-257. Currens, K. P., C. B. Schreck, and H. W. Li. 1990. Allozyme and morphological divergence of rainbow trout (Oncorhynchus mykiss) above and below waterfalls in the Deschutes River, Oregon. Copeia 1990:730-746. Felsenstein, J. 1993. PHYLIP - phylogeny inference package (Version 3.5). Univ. of Washington, Seattle. Goudet, J. 1995. Fstat version 1.2: a computer program to calculate F-statistics. Journal of Heredity 86:485.486. Leider, S. A., S. R. Phelps, and P. L. Hulett. 1995. Genetic analysis of Washington steelhead: Implications for revision of genetics conservation management units. WDFW Program Report. Minch, E. 1997. MICROSAT 1.5d. Stanford Univ., http://lotka.standford.edu /microsat.html. Morris D. B., K. R. Richard, J. M Wright. 1996. Microsatellites from rainbow trout (Oncorhynchus mykiss) and their use for genetic study of salmonids. Canadian Journal of Fisheries Aquatic Sciences 53:120 126. Olsen, J.B., S.L. Wilson, E.J. Kretschmer, K.C. Jones, and J.E. Seeb. 2000. Characterization of 14 tetranucleotide microsatellite loci derived from sockeye salmon. Molecular Ecology 9:2185-2187. Page, R. D. M. 1996. TREEVIEW: an application to display phylogenetic trees on personal computers. Computer Application Biosciences 12:351-358. Page AQU 15-6

Raymond, M., and F. Rousset. 1995. GENEPOP (ver. 1.2): A population genetics software for exact test and ecumenicism. Journal of Heredity 86:248-249. Rexroad III, C. E., R. L. Coleman, A. M. Martin, W. K. Hershberger, and J. Killefer. 2001. Thirty-five polymorphic microsatellite markers for rainbow trout (Oncorhynchus mykiss). Animal Genetics 32:316.331. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. Schneider, S., J. M. Kueffer, D. Roessli, and L. Excoffier. 2000. Arlequin version 2.000: a software for population genetic data analysis. Genetics and Biometry Laboratory, Univ. of Geneva, Geneva, Switzerland. Small, M.P., T.D. Beacham, R.E. Withler, and R.J. Nelson. 1998. Discriminating coho salmon (Oncorhynchus kisutch) populations within the Fraser River, British Columbia. Molecular Ecology. 7:141-155. Page AQU 15-7

Table 1. Subpopulation information including location, abbreviation, sample size (N), life form, and collection code for 14 subpopulations of O. mykiss. Subpopulation Abbreviation N Form Collection Code Spokane Hatchery SpoHat 96 rainbow trout 00DF Pine Creek PinCrb 50 rainbow trout 01FT Goldendale Hatchery GdnHat 50 rainbow trout 01JB South Tacoma Hatchery STcHat 50 rainbow trout 02BK Lewis River @Eagle cliff trap LRrbst 48 rainbow/steelhead 02CB EF Lewis River (2002) EFLRSR 49 summer-run steelhead 02FN Muddy River MRrbst 33 rainbow/steelhead 02FX Cedar Creek (2003) 03CeWR 60 winter-run steelhead 03AS Merwin Hatchery MHatWR 80 winter-run steelhead 03CD Merwin Hatchery MHatSR 80 summer-run steelhead 03CE EF Lewis River 96EFLR 60 rainbow/steelhead 96DK EF Lewis River (1996) EFLRWR 60 winter-run steelhead 96DL Cedar Creek (1996) 96CeWR 60 winter-run steelhead 96DS NF Lewis River NFLRrb 61 rainbow trout 99ZG Table 2.. PCR amplification conditions and primer references for 10 microsatellite loci used to genotype O. mykiss. Dye Annealing Primer Multiplex Locus Reference Label T ( C) Conc (um) Omy B One-102 Olsen et al. 2000 6fam 55 0.08 One-114 Olsen et al. 2000 hex 55 0.20 Ots-100 Olsen et al. 2000 ned 55 0.07 Omy C One-108 Olsen et al. 2000 6fam 55 0.03 Ots-103 Small et al. 1998 hex 55 0.03 One-101 Olsen et al. 2000 ned 55 0.04 Omy D Ots-1 Banks et al. 1999 6fam 49 0.07 Omy-77 Morris et al. 1996 hex 49 0.08 Ots-3M Banks et al. 1999 ned 49 0.05 Page AQU 15-8

Table 3. Genetic diversity estimates including average heterozygosity (Avg Het), allelic richness (A O ), and inbreeding (F IS ) for 14 subpopulations of O. mykiss. Asterisk indicates statistical significance. Subpopulation Avg Het Ao F IS SpoHat 0.6876 4.7003 0.0160 PinCrb 0.7500 6.3950 0.1070* GdnHat 0.7127 4.7520 0.0250 STcHat 0.5962 3.9210-0.0090 LRrbst 0.8044 7.7074 0.1490* EFLRSR 0.7282 6.1506 0.0680 MRrbst 0.8210 7.1514 0.2520* 03CeWR 0.7570 6.5457 0.0370 MHatWR 0.7127 5.9873 0.0210 MHatSR 0.7100 5.6422 0.0230 96EFLR 0.6971 5.6886 0.1070* EFLRWR 0.7538 6.4448 0.0780* 96CeWR 0.7710 6.8653 0.0560 NFLRrb 0.7754 6.7259 0.0160 Table 4. Significant probability values for Hardy-Weinberg tests (heterozygote deficiencies) tested for nine microsatellite loci in 14 subpopulations of O. mykiss. Population Locus P-value PinCrb Ots-100 0.0000 LRrbst One-114 0.0003 LRrbst One-108 0.0000 MRrbst One-114 0.0000 MRrbst One-108 0.0000 MRrbst Ots-1 0.0000 MHatWR One-114 0.0000 MHatWR Omy-77 0.0000 96EFLR Ots-1 0.0000 96CeWR Ots-1 0.0000 96CeWR Omy-77 0.0000 Page AQU 15-9

Table 5. Pairs of microsatellite loci exhibiting significant linkage disequilibrium in subpopulations of O. mykiss. Population Locus 1 Locus 2 P-value SpokHat Ots-100 One-101 0.0000 03CeWR One-102 One-108 0.0000 03CeWR One-108 Ots-1 0.0000 MHatWR Ots-100 One-114 0.0000 MHatWR One-102 Ots-1 0.0000 MHatWR One-108 Ots-1 0.0000 MHatWR One-102 Omy-77 0.0000 MHatWR One-108 Omy-77 0.0000 MHatWR Ots-3M Omy-77 0.0000 MHatSR Ots-103 Ots-3M 0.0000 96EFLR Ots-100 One-108 0.0000 96EFLR Ots-100 Omy-77 0.0000 96CeWR One-102 One-108 0.0000 Page AQU 15-10

Table 6. Results of the jackknife assignments for 14 subpopulations of O. mykiss. The subpopulations located in the lower portion (below bold line) are the rainbow/steelhead subpopulations that were analyzed as unknown subpopulations. Source Population to which individuals were assigned Pop SpoHat PinCrb GdnHat STcHat EFLRSR 03CeWR MHatWR MHatSR EFLRWR 96CeWR NFLRrb SpoHat 98.96 1.04 PinCrb 12.00 50.00 32.00 4.00 2.00 GdnHat 100.00 STcHat 100.00 EFLRSR 57.14 6.12 2.04 16.33 16.33 2.04 03CeWR 1.67 16.67 41.67 16.67 16.67 6.67 MHatWR 2.50 7.50 82.50 1.25 1.25 5.00 MHatSR 15.00 83.75 1.25 EFLRWR 31.67 8.33 8.33 6.67 45.00 96CeWR 3.33 10.00 25.00 18.33 3.33 6.67 30.00 3.33 NFLRrb 1.64 8.20 1.64 3.28 85.25 LRrbst 4.65 16.28 30.23 4.65 9.30 4.65 6.98 20.93 2.33 MRrbst 6.25 6.25 18.75 3.13 12.50 6.25 15.63 3.13 28.13 96EFLR 1.75 1.75 1.75 56.14 5.26 3.51 10.53 17.54 1.75 Page AQU 15-11

Table 7. Pairwise test for population genetic differentiation for 17 subpopulations of O. mykiss. The test was performed for all subpopulation pairs at each locus and a global χ 2 value is reported. Subpopulations not significantly differentiated are indicated in bold type. Subpopulation 1 Subpopulation 2 Chi2 df P-value PinCrb SpoHat Infinity 18 Highly sign. GdnHat SpoHat Infinity 18 Highly sign. GdnHat PinCrb Infinity 18 Highly sign. STcHat SpoHat Infinity 18 Highly sign. STcHat PinCrb Infinity 18 Highly sign. STcHat GdnHat Infinity 18 Highly sign. LRrbst SpoHat Infinity 18 Highly sign. LRrbst PinCrb Infinity 18 Highly sign. LRrbst GdnHat Infinity 18 Highly sign. LRrbst STcHat Infinity 18 Highly sign. EFLRSR SpoHat Infinity 18 Highly sign. EFLRSR PinCrb Infinity 18 Highly sign. EFLRSR GdnHat Infinity 18 Highly sign. EFLRSR STcHat Infinity 18 Highly sign. EFLRSR LRrbst Infinity 18 Highly sign. MRrbst SpoHat Infinity 18 Highly sign. MRrbst PinCrb Infinity 18 Highly sign. MRrbst GdnHat Infinity 18 Highly sign. MRrbst STcHat Infinity 18 Highly sign. MRrbst LRrbst Infinity 18 Highly sign. MRrbst EFLRSR Infinity 18 Highly sign. 03CeWR SpoHat Infinity 18 Highly sign. 03CeWR PinCrb Infinity 18 Highly sign. 03CeWR GdnHat Infinity 18 Highly sign. 03CeWR STcHat Infinity 18 Highly sign. 03CeWR LRrbst Infinity 18 Highly sign. 03CeWR EFLRSR Infinity 18 Highly sign. 03CeWR MRrbst Infinity 18 Highly sign. MHatWR SpoHat Infinity 18 Highly sign. MHatWR PinCrb Infinity 18 Highly sign. MHatWR GdnHat Infinity 18 Highly sign. MHatWR STcHat Infinity 18 Highly sign. MHatWR LRrbst Infinity 18 Highly sign. MHatWR EFLRSR Infinity 18 Highly sign. MHatWR MRrbst Infinity 18 Highly sign. MHatWR 03CeWR Infinity 18 Highly sign. MHatSR SpoHat Infinity 18 Highly sign. MHatSR PinCrb Infinity 18 Highly sign. MHatSR GdnHat Infinity 18 Highly sign. MHatSR STcHat Infinity 18 Highly sign. MHatSR LRrbst Infinity 18 Highly sign. MHatSR EFLRSR Infinity 18 Highly sign. MHatSR MRrbst Infinity 18 Highly sign. MHatSR 03CeWR Infinity 18 Highly sign. MHatSR MHatWR Infinity 18 Highly sign. Page AQU 15-12

Table 7. Continued Subpopulation 1 Subpopulation 2 Chi2 df P-value 96EFLR SpoHat Infinity 18 Highly sign. 96EFLR PinCrb Infinity 18 Highly sign. 96EFLR GdnHat Infinity 18 Highly sign. 96EFLR STcHat Infinity 18 Highly sign. 96EFLR LRrbst Infinity 18 Highly sign. 96EFLR EFLRSR 46.618 18 0.00024 96EFLR MRrbst Infinity 18 Highly sign. 96EFLR 03CeWR Infinity 18 Highly sign. 96EFLR MHatWR Infinity 18 Highly sign. 96EFLR MHatSR Infinity 18 Highly sign. EFLRWR SpoHat Infinity 18 Highly sign. EFLRWR PinCrb Infinity 18 Highly sign. EFLRWR GdnHat Infinity 18 Highly sign. EFLRWR STcHat Infinity 18 Highly sign. EFLRWR LRrbst Infinity 18 Highly sign. EFLRWR EFLRSR 45.877 18 0.00031 EFLRWR MRrbst Infinity 18 Highly sign. EFLRWR 03CeWR 87.621 18 0.00000 EFLRWR MHatWR Infinity 18 Highly sign. EFLRWR MHatSR Infinity 18 Highly sign. EFLRWR 96EFLR 58.292 18 0.00000 96CeWR SpoHat Infinity 18 Highly sign. 96CeWR PinCrb Infinity 18 Highly sign. 96CeWR GdnHat Infinity 18 Highly sign. 96CeWR STcHat Infinity 18 Highly sign. 96CeWR LRrbst Infinity 18 Highly sign. 96CeWR EFLRSR Infinity 18 Highly sign. 96CeWR MRrbst Infinity 18 Highly sign. 96CeWR 03CeWR 21.696 18 0.24575 96CeWR MHatWR Infinity 18 Highly sign. 96CeWR MHatSR Infinity 18 Highly sign. 96CeWR 96EFLR Infinity 18 Highly sign. 96CeWR EFLRWR Infinity 18 Highly sign. NFLRrb SpoHat Infinity 18 Highly sign. NFLRrb PinCrb Infinity 18 Highly sign. NFLRrb GdnHat Infinity 18 Highly sign. NFLRrb STcHat Infinity 18 Highly sign. NFLRrb LRrbst Infinity 18 Highly sign. NFLRrb EFLRSR Infinity 18 Highly sign. NFLRrb MRrbst Infinity 18 Highly sign. NFLRrb 03CeWR Infinity 18 Highly sign. NFLRrb MHatWR Infinity 18 Highly sign. NFLRrb MHatSR Infinity 18 Highly sign. NFLRrb 96EFLR Infinity 18 Highly sign. NFLRrb EFLRWR Infinity 18 Highly sign. NFLRrb 96CeWR Infinity 18 Highly sign. Page AQU 15-13

Table 8. Matrix of pairwise F ST estimates testing for population subdivision among subpopulations of O. mykiss. Gray shading indicates pairwise F ST estimates that were not statistically significant. SpoHat PinCrb GdnHat STcHat LRrbst EFLRSR MRrbst 03CeWR MHatWR MHatSR 96EFLR EFLRWR 96CeWR NFLRrb SpoHat --- PinCrb 0.0652 --- GdnHat 0.1270 0.0572 --- STcHat 0.2237 0.1414 0.1692 --- LRrbst 0.0731 0.0070 0.0556 0.1302 --- EFLRSR 0.1595 0.0751 0.1299 0.1638-0.0019 --- MRrbst 0.1161 0.0448 0.0783 0.1727 0.0546 0.0363 --- 03CeWR 0.1550 0.0720 0.1334 0.1853 0.0126 0.0146 0.0419 --- MHatWR 0.1554 0.0772 0.1476 0.1957 0.0174 0.0328 0.0528 0.0132 --- MHatSR 0.1501 0.0822 0.1512 0.1822 0.0108 0.0210 0.0577 0.0379 0.0515 --- 96EFLR 0.1212 0.0247 0.0939 0.1494 0.0224-0.0800 0.1338-0.0295-0.0187-0.0254 --- EFLRWR 0.1421 0.0656 0.1201 0.1604 0.0054 0.0034 0.0390 0.0130 0.0314 0.0264-0.0379 --- 96CeWR 0.1510 0.0714 0.1343 0.1830 0.0098 0.0180 0.0251 0.0044 0.0184 0.0412-0.0284 0.0147 --- NFLRrb 0.1211 0.0977 0.1251 0.1967 0.0263 0.0735 0.0136 0.0549 0.0717 0.0877 0.0362 0.0480 0.0476 --- Page AQU 15-14

a) MHatSR 03CeWR 96CeWR EFLRSR EFLRWR 98 99 96 92 64 99 MHatWR NFL Rrb 100 GdnHat 95 100 STcHat PinCrb 0.1 SpoHat b) STcHat PinCrb GdnHat MRrbst 83 99 SpoHat 99 LRrbst 78 96EFLR 96CeWR 59 62 75 89 62 81 98 EFLRWR NFLRrb 03CeWR EFLRSR 0.1 MHatWR MHatSR Figure 1. Consensus tree from 1000 neighbor-joining trees based on Cavalli-Sforza and Edwards genetic distance for subpopulations of O. mykiss a) excluding the rainbow/steelhead subpopulations and b) including all 14 subpopulations. Page AQU 15-15

This page intentionally blank. Page AQU 15-16