Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Migration and survival of Okanagan River Sockeye Salmon Oncorhynchus nerka, 2012–2019

Migration and survival of Okanagan River Sockeye Salmon Oncorhynchus nerka, 2012–2019 Background: Okanagan River Sockeye Salmon Oncorhynchus nerka (Okanagan Sockeye) are one of two remain- ing self-sustaining Sockeye Salmon populations in the Columbia River Basin. We used detection histories of smolts implanted with passive integrated transponder (PIT ) tags between 2012 and 2019 to estimate survival and behavioral metrics during reintroduction efforts and changing environmental conditions over the monitoring period. Results: Smolts migrating to McNary Dam, whose route includes 130 km of the Okanagan River and 388 km of the Columbia River, generally had high survival (mean of 87.0% per 100 km) and fast migration speeds (up to 50 km/day) relative to other salmonids in the region. Smolt-to-adult returns (SARs) ranged from 0.4 to 6.1% and were greater for fish originating from Skaha Lake compared to cohorts tagged in Osoyoos Lake. Most adults returned after 2 years in the ocean (69%), followed by jacks (27%), and adults that spent 3 years at sea (4%), though Skaha Lake adults had a significantly younger age structure than cohorts from Osoyoos Lake. Survival of adults from Bonneville Dam (rkm 235) upstream to Wells Dam (rkm 830) was generally high (80–92%), and migration speed decreased in upstream reaches. Survival from Wells Dam to the Okanagan River was only estimable in 2018, where 64% of adults survived to the spawning grounds. The upstream migration of adult Okanagan Sockeye was significantly compromised during the drought of 2015 when less than 5% of Okanagan Sockeye that returned to the Columbia River reached spawning grounds. Conclusions: Our results indicate that Okanagan Sockeye have exceptional survival and migratory ability relative to other salmonids, though poor ocean conditions combined with warming water temperatures in freshwater habitats in recent years have the potential to devastate the population. The success of reintroduction efforts to increase spatial structure and diversity of Okanagan Sockeye is, therefore, critical to maintaining the population in years to come. Keywords: Sockeye Salmon, PIT tags, Columbia River Basin, Survival inhabited the Columbia River Basin. Sockeye Salmon Background once existed in eight Columbia River tributaries, includ- Okanagan River Sockeye Salmon Oncorhynchus nerka ing at least 12 nursery lakes above the Snake River con- (Okanagan Sockeye) originating from Osoyoos Lake fluence [21, 24, 47], and peak returns likely ranged from (Canada), and Lake Wenatchee Sockeye Salmon (United 2.5 to 3.2 million adults [12, 21]. Degradation of and States) comprise the last two self-sustaining populations blocked access to habitat, urban and industrial devel- of anadromous O. nerka among several that formerly opment, and overexploitation have greatly diminished returns [27, 36], and by the late 1900s, Columbia River *Correspondence: jmurauskas@fourpeaksenv.com Sockeye Salmon had been virtually extirpated except for Four Peaks Environmental Science & Data Solutions, Wenatchee, USA populations in the Okanagan (spelled Okanogan in the Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 2 of 16 U.S.) and Wenatchee River basins [21, 29, 47]. Despite ongoing threats, Okanagan Sockeye have demonstrated perhaps the greatest recovery of any salmonid population in the Pacific Northwest as returns have increased by an order of magnitude over the last 30  years. Mean counts at Wells Dam increased from 17,083 adults during the 1990s to 199,907 adults during the 2010s [13]. The recov - ery is largely a response to water resource management in the Okanagan Basin, increased smolt survival through hydroelectric projects, a period of favorable ocean con- ditions, and the Skaha Lake reintroduction program, including the kł cp̓əlk̓ stim̓ Hatchery in Penticton, British Columbia [33, 69]. While Okanagan Sockeye have demonstrated excep- tional resilience, anthropogenic actions continue to influence the population at all life stages. Urbanization, agricultural operations, and water quality and quantity in spawning and early rearing habitats still dominate management discussions; although, anticipating risks to the species has become increasingly complicated under increasing water temperatures [50]. For example, more than 90% of Okanagan Sockeye that entered the Colum- bia River did not survive to the spawning grounds dur- ing record high water temperatures observed in 2015 [22, 34, 38]. The effects of the 2015 drought persisted through Fig. 1 Overview of the Columbia River migratory corridor (top right); 2019 when the dominant year class of that brood year the Okanagan River Basin (bottom right); and juvenile release sites (open circles), PIT arrays (open triangles), and dams (closed circles) returned at roughly 25% of the recent 10-year average in the Okanagan River Basin (left). Dams passed during migration [11]. With temperatures expected to continue increasing include (1) Bonneville, (2) The Dalles, (3) John Day, (4) McNary, (5) in the Columbia River Basin [71], supporting the recov- Priest Rapids, (6) Wanapum, (7) Rock Island, (8) Rocky Reach, (9) Wells, ery of imperiled Pacific Salmon stocks is a critical goal (11) Zosel, (12) McIntyre, (13) Skaha, and (14) Penticton dams. Chief of water management agreements between the United Joseph Dam (10) is shown for reference but does not have upstream passage facilities States and Canada [15, 53] and litigation concerning hydroelectric development in the Columbia River [1]. Within the commonly referenced Viable Salmonid Population (VSP) framework, the persistence of salmo- and decreasing hypolimnetic oxygen concentrations nid populations is characterized in terms of their abun- resulting from anthropogenic eutrophication greatly dance, productivity, spatial structure, and diversity [48]. reduce suitable nursery habitat for juveniles during late Okanagan Sockeye have reached record levels of abun- summer each year [61]. This situation emphasizes the dance and productivity, but the lack of spatial structure importance of interventions to increase the diversity and and diversity presents risk to the population, especially spatial structure for Okanagan Sockeye. with increases in water temperatures observed in recent Beginning in the late 1990s, efforts to reintroduce years. Habitat in the upper Okanagan River Basin Okanagan Sockeye to historical spawning and rearing became inaccessible to anadromous fishes following habitat in the upper Okanagan River Basin was initiated construction of several dams in the 1900s that blocked by a diverse group of stakeholders led by the Okanagan access to Vaseux Lake (McIntyre Dam, 1921), Skaha Nation Alliance, a First Nations Council in Canada, Lake (Okanagan Falls Dam, 1958), and Okanagan Lake with critical support from Fisheries and Oceans Canada (Penticton Dam, 1915; Fig.  1). Most of the 20th-century (DFO), public utility districts in the United States, and spawning and rearing habitat of Okanagan Sockeye in the the Habitat Conservation Trust Foundation. Reintroduc- Okanagan River Basin [57] was, therefore, restricted to tion of Okanagan Sockeye into Skaha Lake commenced approximately 20 km of the Okanagan River near Oliver, in 2004 [9]. By the 2010s, up to 10% of adult returns to British Columbia, and Osoyoos Lake [33]. Most juvenile the Okanagan River Basin comprised hatchery-origin Okanagan Sockeye rear in portions of Osoyoos Lake, fish released as fry from the newly constructed the kł where increasing water temperatures of the epilimnion M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 3 of 16 cp̓ lk̓ stim̓ Hatchery and upstream passage was improved downstream and pass through the Osoyoos Lake Nar- or reestablished at McIntyre Dam below Vaseux Lake rows, which connects the central and north basins of (2009) and Okanagan Falls Dam below Skaha Lake (2011) the lake. From Zosel Dam at the outlet of Osoyoos Lake, [33]. To track the success of these efforts, a monitor - the Okanagan River flows south through the United ing and evaluation program was implemented in 2005. States, converging with the Columbia River near Brew- Okanagan Nation Alliance staff later worked to create ster, Washington, in the reservoir created by Wells Dam. an extensive passive integrated transponder (PIT) tag Smolts then migrate through nine hydroelectric projects monitoring network through the freshwater migration along the Columbia River to reach the Pacific Ocean corridor. The combined efforts to protect, enhance, and (Fig.  1). Okanagan Sockeye then spend anywhere from 1 monitor Okanagan Sockeye have proven critical to suc- to 3 years in the ocean before returning to migrate up the cessful management of one of the most culturally, eco- Columbia River to spawning grounds in the Okanagan logically, and economically important fisheries in the River. Interior Columbia River Basin. The purpose of this manuscript is to provide the first Tagging methods and release sites comprehensive analysis of the migration behaviors and Okanagan Sockeye smolts were captured and PIT-tagged population dynamics of Okanagan Sockeye based on data at five sites from 2012 to 2019 within the Okanagan from juveniles that were PIT-tagged and released in the River Basin to measure juvenile outmigration sur- Okanagan Basin. These data may be used to provide esti - vival and travel time from release site to McNary Dam, mates of smolt survival during the downstream migra- SAR from the release site back to Bonneville Dam, and tion, age-at-maturity, smolt-to-adult returns (SARs), adult survival and travel time upstream from Bonnev- conversion rates of adults during the upstream migra- ille Dam through the Columbia and Okanagan rivers. tion, and a variety of related behavioral metrics, such as Tagging and release sites were identified by codes des - travel time and date of arrival [4, 9, 52]. Results presented ignated by the Columbia Basin PIT Tag Information Sys- here will provide survival estimates and other data across tem (PTAGIS; www. ptagis. org), and included: Osoyoos critical life stages that can be used to inform recovery Lake (OSOYOL), Osoyoos Lake at Haynes Point Camp- strategies that optimize measures in the VSP framework. ground (OSOYHA), Osoyoos Lake Narrows Highway Specifically, smolt survival, SARs, and adult conversion 3 Bridge (OSOYBR), the tailrace approximately 0.5  km rates to spawning habitat exert profound influences on downstream of Skaha Dam (SKATAL), and Skaha Lake salmon abundance and productivity. In addition, behav- (SKAHAL) (Fig.  1). Fish labeled as released at SKA and ioral information and survival of Okanagan Sockeye orig- SKATAL sites were pooled in each year as this was the inating from re-established habitat are critical to plans same release site. Smolts were captured using either fyke to improve spatial structure and diversity of the popula- nets, rotary screw traps, purse seines, or a combination tion. Finally, recommendations for expanding the current at each site over the study years (Table  1). From 2013 to monitoring framework are provided so managers can 2017, all or a portion of smolts from Osoyoos Lake were understand how to best allocate limited resources as con- captured using a floating fyke net at the OSOYBR site servation efforts evolve in coming years. (see methods in 2). Starting in 2017 purse seine methods were primarily used in all Osoyoos sites [20]. An 8.5  m Methods purse seiner fishing with 183 m seine net (1.27 cm knot - Study area ted mesh) was used to capture smolts at all Osoyoos Lake The Okanagan River is a major tributary to the Columbia and SKAHAL sites from 2016 to 2019. Purse seines fished River and has an approximate length of 185  km (37  km to a depth of 12 m, concentrating in the central basin of Canadian section, 148  km United States section). Wild- Osoyoos Lake and in the southern region of Skaha Lake origin Okanagan Sockeye spawn primarily in the river where most Okanagan Sockeye smolts congregate. From between Osoyoos and Skaha lakes (Fig.  1) during Octo- 2012 to 2016, smolts emigrating from Skaha Lake were ber [62]. Fry emerge from late April to early May and captured using two rotary screw traps that operated generally spend 1  year rearing in Osoyoos Lake, and/ intermittently throughout the outmigration period [19] or the Okanagan River before emigrating as smolts [7, downstream of Okanagan Falls Dam. The success of the 33]. Hatchery- and, more recently, wild-origin smolts program resulted in the increase of target number of PIT- that rear in Skaha Lake pass through Okanagan Falls tagged smolts from a combined total of 5,000 for Osoy- Dam, continuing their journey down the Okanagan oos and Skaha lakes to 5,000 for each lake (Table 1) [70]. River through McIntyre Dam and Osoyoos Lake (Fig. 1). Smolts were anesthetized using a 40  mg/l solution of Okanagan Sockeye smolts from both lakes begin their tricaine methanesulfonate (MS-222) and tagged with outmigration at similar times (April–May). Smolts travel Biomark HPT 12 PIT tags (134.2  kHz; 12  mm length) Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 4 of 16 Table 1 Number of PIT-TAGGED Okanagan Sockeye by PTAGIS release site code, capture method, and release year (2012–2019) Year OSOYOL OSOYHA OSOYBR SKA + SKATAL SKAHAL TOTAL Purse seine Purse seine Fyke net Purse seine Screw trap Purse seine Combined 2012 534 534 2013 2783 1203 3986 2014 3706 1348 5054 2015 1741 5435 7176 2016 3044 1754 3101 2338 10,237 2017 8794 152 2,642 11,588 2018 1521 3562 5860 10,943 2019 4968 4114 9082 OSOYBR Osoyoos Lake Narrows Highway 3 Bridge, OSOYHA Osoyoos Lake at Haynes Point Campground, OSOYOL Osoyoos Lake, SKA + SKATAL Tailrace of Skaha Dam, SKAHAL Skaha Lake according to procedures outlined by PTAGIS and Bio- detections of PIT-tagged adults at various points. The mark [2]. Tags were implanted with an MK-25 Rapid first detection site through which all adult Okanagan Implant Gun along with APT12 pre-loaded needles (Bio- Sockeye must pass along their return migration are the mark, Boise, Idaho). In addition to receiving a PIT tag, fish ladders at Bonneville Dam. These detection points a proportion of smolts were measured for fork length (PTAGIS Site Code) include arrays in the Bradford Island (mm) each year. Following tagging, all mortalities were Ladder (BO1), Bonneville Cascades Island Ladder (BO2), removed from the tagged population. Visible determi- Bonneville WA Shore Ladder (BO3), and Bonneville WA nation of hatchery versus wild origin was not possible Ladder slots (BO4). The next upstream detection site because hatchery Okanagan Sockeye were released as used in this study is McNary Dam where PIT-tagged fry when adipose clips were not possible. Therefore, fish adults are detected at the McNary Oregon Shore Lad- origin was not considered in estimation of survival and der (MC1) or McNary Washington Shore Ladder (MC2). travel times. Continuing into the upper Columbia River from McNary Dam, the next upstream site used in this study is Wells Detection sites for juveniles and adults Dam where adults are detected at the Adult Ladders Juvenile survival, detection probabilities, and travel time (WEA). After passing Wells Dam, Okanagan Sockeye estimates were based on detections of Okanagan Sock- travel up the Okanagan River where the first adult PIT- eye from release sites to McNary Dam (Table 1). McNary tag detection site used in this evaluation is at Okanagan Dam was used as the first point of detection for travel Channel at Vertical Drop Structure-3 (OKC), installed time and survival estimates because it is most capable of in 2009 (Fig. 1). The last upstream detection point in the providing reasonably precise estimates of survival over Okanagan River Basin, the Penticton Channel PIT array the greatest distance of the juvenile freshwater migration (OKP), was installed and operational in 2018. corridor. During their downstream migration, smolts pass a maximum of four diversion dams in the Okanagan Juvenile survival, detection probabilities, and travel time River before reaching the Columbia River, where they analyses then pass five hydroelectric projects (Wells [river kilom - Survival, detection probabilities, and travel time of eter, rkm 830], Rocky Reach [rkm 762], Rock Island [rkm Okanagan Sockeye smolts were estimated for each PIT- 730], Wanapum [rkm 669], and Priest Rapids [rkm 639]) tagged release group to McNary Dam (2012–2019; before reaching McNary Dam (rkm 470). After McNary Table  1). Estimates of juvenile survival and travel time Dam, juveniles pass through John Day (rkm 347), The were generated from the Columbia River Data Acqui- Dalles (rkm 308), and Bonneville (rkm 235) dams before sition in Real Time (DART) query program [13]. The reaching the Pacific Ocean (rkm 0). The farthest down - DART query retrieves PIT-tag data files of individual- stream detection site was in the Columbia River estuary, based capture history from PTAGIS (for tag files used where the National Marine Fisheries Service operated a see Additional file  1: Appendix S1) and uses these files paired-trawl detection system [44]. to generate survival estimates, detection probabilities, Survival and travel time of adult Okanagan Sockeye and travel times for each release site in each year. Sur- migrating through the Columbia River and Okanagan vival estimates, detection probabilities, and associated River back to their spawning grounds was based on standard error for each release group in each year were M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 5 of 16 estimated in DART using the Cormack–Jolly–Seber Skaha Lake (SKAHAL) in 2019. Because these omissions (CJS) model [14, 40, 60]. Assumptions for the CJS model appeared random across years for each release site, we include (1) individuals marked for the study are a repre- interpolated average length for these releases groups by sentative sample from the population of interest; (2) sur- averaging values across years where length data were pre- vival and capture probabilities are not affected by tagging sent from the same site (SKA + SKATAL: n = 4, standard or detection; (3) all detections are instantaneous; (4) the deviation 3.21  mm; SKAHAL: n = 2, standard deviation fate of each tagged individual is independent of the fate of 5.21 mm.) Exploratory analyses were conducted prior to all others; (5) all tagged individuals alive at the beginning model fitting to ensure distributional assumptions were of a reach have the same probability of surviving until the met and to avoid issues in multi-collinearity [72]. All end of that reach; (6) all tagged individuals alive at the modeling and validation were conducted in R (v. 3.6.1, R beginning of a detection site have the same probability Core Team 2019). of being detected at that site; (7) each individual detected Travel-time estimates were based on the release date at a particular detection site has the same probability for each PIT-tagged fish and the first detection date at of being removed, and (8) the probability of removal is McNary Dam. Harmonic mean travel time was used to independent of the survival process. Data files contain - describe each group’s (by release site and year) rate of ing records from all Okanagan release sites from PTAGIS travel from release site to McNary Dam. This statistical were examined for erroneous records, inconsistencies, summary of central tendency is more robust to the pres- and anomalies. Data files of releases of juvenile Okanagan ence of outliers (i.e., a very fast or slow fish) than arith - Sockeye in Osoyoos Lake (OSOYOL, OSOYOHA, and metic mean and is computed following Eq. (2): OSOYBR) and Skaha Lake and Skaha Dam tailrace (SKA, t = SKATAL, and SKAHAL) since 2012 were analyzed. Sur- 1 (2) i=1 t vival estimates were not adjusted for juveniles that did i not emigrate, tag failure, tag loss, or other factors, which where t is the computed harmonic mean travel time; n could result in fish surviving but not being detected at a is the observed number of unique fish detected at the downstream location. Due to these factors, actual sur- detection sites; t is the observed reach travel time for vival may be higher than the reported estimates. each fish i through n. Survival estimates across all years and sites were used in a regression-model framework to better understand the factors associated with juvenile survival. Multiple Smolt to adult survival rates and proportion of ocean‑age weighted logistic regression was used to explore how adults returning variation in juvenile survival is explained by outmigration Smolt-to-adult survival rates (SARs) and associated year, release site, release date, and smolt size (i.e., average standard errors were calculated as the number of PIT- length). The modeling framework followed Eq. (1): tagged smolts released from each lake of origin (Skaha Lake or Osoyoos Lake) in each year (2012–2016) that logit(p ) ∼ β + β x + ε , ε ∼ N 0, σ survived to return as adults to Bonneville Dam as ocean i 0 k k i i (1) age-1, age-2, or age-3 adults in 2013–2019. This approach k=1 assumes that smolts emigrate to the ocean the year they where p is the probability of juvenile survival observed are tagged and is supported by the negligible proportion across all individuals from group i (i = 1, …, 18); β is the of smolts that hold over an additional year following tag- intercept; β is the weighted slope term for each covariate ging [23]. PIT-tagged fish released in years 2017–2019 x ; ε is the normally distributed random error associated were excluded from the analysis due to incomplete age- with group i. structure data sets (i.e., return data only includes through Each observation was weighted using the number of 2019; therefore, no age-2 and age-3 data are available individuals within the release group (i.e.,  w = n ) , and i i for 2018 and 2017, respectively) and small sample sizes these weights were used to adjust the covariance matrix in those years. Data used in the analysis were queried to give higher weights to those site and year combinations using the PTAGIS interrogation summary tool. A query where the number of smolts was highest. Forward selec- reporting all instances of Okanagan Sockeye detected at tion based on AICc [8] was used to find the most parsi - the Bonneville adult ladders (BO1, BO2, BO3, and BO4) monious model amongst a suite of candidate models, from each release site and each year was generated. Bon- including outmigration year, release site, release date, and neville Dam provides the first points of detection for all smolt length. Two out of 18 release groups were initially adults as they begin their upriver migration to spawn- lacking length data: individuals released from Skaha Dam ing grounds and has nearly 100% detection efficiency (SKA + SKATAL) in 2015 and individuals released from when pooling data from the four PIT-tag interrogation Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 6 of 16 sites [63]. Duplicate tag numbers were deleted from each certainly resident O. nerka (i.e., kokanee). Included in data set to eliminate duplicate detections within each these 33 samples were eight fish detected only at OKP detection site. All data from the PTAGIS interrogation in 2019; these fish were not included in calculations of summaries were examined for erroneous records, incon- survival estimates. Survival estimates of adults return- sistencies, and anomalies. The equation used to calculate ing to OKC were, therefore, only possible in 2018 due SAR for each basin release group for each year across all to lack of data or exclusion of individuals that were not ages of adults returning (SAR ) is provided in the follow- confirmed to emigrate to the ocean. An additional 180 ij ing equation: individuals were eliminated from the analysis because they were juveniles detected in adult fishways dur - ij ing the year of release as smolts. A total of 395 adult SAR = ij (3) ij Okanagan Sockeye were ultimately used in the mark– recapture analysis to estimate adult survival and detec- where A is the total number of adults originating from ij tion probabilities upstream through the Columbia either Skaha or Osoyoos lakes and detected at Bonneville River to the Okanagan River Basin. Dam (i) in each year  (j); R is the total number of juve- ij Our analysis relied on survival rates derived in a mark– niles released from each lake (i) in each year (j). recapture framework to understand important factors Data were further analyzed to determine length of driving juvenile survival. Parameter estimation proceeds time of ocean residence to calculate proportion of ocean by first estimating detection probabilities, and then using age-1, age-2, and age-3 Okanagan Sockeye that return to these estimates to provide an estimate of survival proba- Bonneville Dam. Ocean age was determined by subtract- bility based of apparent detection. Therefore, our survival ing year of release from year of return. A Chi-squared probabilities were sensitive to differences in detection test was conducted to determine whether proportions efficiency across years and sites and were thus likely differed by return year and tagging basin. affected by factors that can influence detection efficiency (e.g., flow). Although a more robust approach would include covariates within a mark–recapture framework Adult upstream survival estimates, detection probabilities, to estimate their contribution to detection and sur- and travel time vival probability, we took this opportunistic approach Survival estimates and detection probabilities for adult to understand how various geographic, morphological, Okanagan Sockeye migrating upstream through the and phenological factors impacted a decent estimate of Columbia to the Okanagan River Basin were calculated survival. using a mark–recapture model in MARK [67]. Adults Adult return data through the Columbia and Okanagan pass through several dams with detection arrays on the rivers were examined to determine whether median day way back to their spawning grounds. In this analysis, of arrival at each dam and travel time from Bonneville adult survival rates and detection probabilities were to McNary, McNary to Wells, and Wells to OKC varied calculated using detections from Bonneville Dam to across years (2013–2019); basin of origin (Skaha Lake and McNary Dam to Wells Dam to OKC to OKP. Survival Osoyoos Lake) was also examined. Data were queried rates were calculated by migration reach and year that from PTAGIS using an interrogation summary. Adult adults returned, so results did not examine effect of and juvenile data were returned from the interrogation juvenile release site, year-class strength of juveniles, summary; all juvenile data were eliminated from analysis. or age-specific survival (juvenile or adult). In addition, Travel time was standardized to distance by calculating survival rates did not account for differences in detec - kilometer per day traveled for each PIT-tagged fish and tion probability estimates that manifested across years averaging across each reach: Bonneville to McNary Dam at various sites. OKP was installed in 2018, so data were (236  km), McNary to Wells Dam (360  km), and Wells only available for 2018 and 2019 for this site. Because Dam to OKC (177 km). Travel time was calculated from survival estimates rely on detecting an individual at date of last detection at downstream detection site to the site beyond the site of interest, survival estimates date of last detection at upstream detection site. To test to OKC were only possible in 2018 and 2019. Data whether arrival date at each dam detection site (Bonnev- obtained from PTAGIS were examined for errone- ille, McNary, Wells, OKC) differed by year and by lake of ous records, inconsistencies, and anomalies. All adult origin, a separate Wilcoxon test was conducted for year records where individuals were detected only at Wells and lake. Median migration speed (km/day) between Dam, OKC, and OKP (and not at Bonneville or McNary each detection site was calculated for all returning adults. dams) were eliminated from analysis (N = 33) because A Wilcoxon Rank Sum test was conducted to examine Bonneville and McNary dams collectively have near- whether all returning adults traveled at different speeds perfect detection rates and these individuals are almost M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 7 of 16 Table 2 Regression coefficients, standard errors, and associated Results P values for the full model estimating juvenile survival against Juvenile survival and travel time release year, release site, average length, and release date Following model selection, the model containing year, release site, and average length was the most Predictor Coefficient Standard error p parsimonious in explaining juvenile survival from Intercept − 4.85 0.48 < 0.01 release site to McNary Dam, while release date was 2013 − 0.88 0.10 < 0.01 not found to be a significant predictor of survival. 2014 − 1.38 0.10 < 0.01 This model was selected from the next best candidate 2015 − 1.15 0.10 < 0.01 model with ΔAICc = 21.72, and was found to signifi - 2016 − 0.54 0.12 < 0.01 cantly reduce deviance compared to the null model 2017 − 0.87 0.11 < 0.01 ( χ = 4, 698.6, p < 0.01) . Regression coefficients were 2018 − 0.46 0.11 < 0.01 found to be largely significant at the α = 0.01 level 2019 − 1.14 0.11 < 0.01 (Table  2). Coefficients to the log odds suggested vari - SKA + SKATAL 0.01 0.09 0.90 able juvenile survival success across years and sites OSOYBR 1.06 0.06 < 0.01 (Fig. 2). Smolts emigrating in 2014 and 2015 performed OSOYHA 1.19 0.07 < 0.01 poorer than those emigrating in 2017 and 2018. In addi- OSOYOL 0.89 0.05 < 0.01 tion, juveniles released from the OSOYBR site fared far Average Length 0.02 < 0.01 < 0.01 better than those from other sites—individuals released Release Date 0.03 < 0.01 < 0.01 from OSOYBR had odds of about 4.77-to-1 of surviving Release date was omitted in the final model based on AIC. Coefficients are given (95% CI [4.37, 5.21]). Juveniles tagged in Osoyoos Lake as raw values in log-odds space. P values represent the significance test result had higher survival from release site to McNary Dam in using the corresponding Wald’s statistic each year compared to fish tagged in Skaha Lake except for 2018 (Table  3; Fig.  3). Greater lengths were associ- between each detection site and to determine whether ated with greater outmigration success, with a 25  mm fish originating from Skaha or Osoyoos lakes exhibited increase in length corresponding with a doubling of the different migration speeds between detection sites. Dif - odds of survival (Fig.  2). Except in 2013, when travel ferences were considered significant at p = 0.05. time was equal, juvenile Okanagan Sockeye released from Skaha Lake traveled faster to McNary Dam than those released from Osoyoos Lake sites in all years, despite traveling a farther distance (migration distance Fig. 2 Regression model coefficients for the selected final model summarizing relationships of observed juvenile survival to predictors. All results are transformed to the logarithmic odds scale. Lines indicate the 95% confidence interval for the regression coefficient, as calculated through the transformed Wald’s statistic. The intercept of the model represents estimated juvenile survival for individuals released from Skaha Lake in 2012. Coefficients for year and release site are offsets to this base survival rate. The vertical line at 1 represents the inflection point between a decrease (less than 1) or an increase (greater than 1) to the default odds of return Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 8 of 16 Table 3 Survival estimates (and standard error) of groups of Survival to maturity PIT-tagged juvenile Okanagan Sockeye from release site in the Smolt-to-adult return rates (SARs) to Bonneville Dam Okanagan River Basin to McNary Dam by release year (2012– ranged from 0.4 to 6.1% across all release years for fish 2019) originating from the Okanagan River Basin. SARs were higher for fish tagged in Skaha Lake than for those tagged Year Release site (by distance) in Osoyoos Lake for all years (Table  5). SARs varied by OSOYOL OSOYHA OSOYBR SKA + SKATAL SKAHAL year and were highest in release-year 2013 (Skaha = 6.1%, 2012 NA NA NA 0.64 (0.23) NA Osoyoos = 3.1%) and lowest in 2015 (Skaha = 0.6%, 2013 NA NA 0.50 (0.12) 0.45 (0.16) NA Osoyoos = 0.4%). Proportion of ocean age-1, age-2, and 2014 NA NA 0.44 (0.05) 0.25 (0.06) NA age-3 fish returning varied by year (X = 16.1, p < 0.01) 2015 NA NA 0.54 (0.22) 0.29 (0.05) NA and by tagging basin (X = 32.3, p < 0.01; Fig.  4). Overall, 2016 0.51 (0.09) NA 0.51 (0.10) 0.40 (0.06) 0.39 (0.08) the mean age structure of adult Okanagan Sockeye at 2017 0.59 (0.07) NA 0.97 (0.29) NA NA Bonneville Dam included 27% ocean age-1, 69% ocean 2018 0.35 (0.08) 0.61 (0.09) NA NA 0.53 (0.07) age-2, and 3% ocean age-3. Returning adults were pre- 2019 0.54 (0.09) NA NA NA 0.28 (0.07) dominantly ocean age-2 for both Skaha- and Osoyoos- origin fish (Fig.  4). There was a higher percentage of OSOYBR Osoyoos Lake Narrows Highway 3 Bridge, OSOYHA Osoyoos Lake at Haynes Point Campground, OSOYOL Osoyoos Lake, SKA + SKATAL Tailrace of ocean age-1 fish returning from Skaha-origin smolts Skaha Dam, SKAHAL Skaha Lake (34%) than from Osoyoos-origin smolts (18%) across all years. Of all years, 2014 had the lowest percentage of ocean age-1 fish returning from either Skaha-origin for Skaha Lake sites is approximately 45 km longer than smolts (11%) or Osoyoos-origin smolts when no ocean Osoyoos sites). Travel time in 2015 was slower than age-1 fish returned. Numbers of ocean age-1 fish from in other years for smolts originating from both lakes Skaha Lake remained low in 2015 (6%), by contrast with (Table 4). age-1 Osoyoos-origin fish that had the highest number of all years (43%). Fig. 3 Relationships between juvenile survival and predictors. Boxplots are shown for categorical variables (release year and site) and scatter plots for continuous variables (average length and release date) M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 9 of 16 Adult migration Table 4 Mean harmonic travel time in days (and standard error) of groups of PIT-tagged juveniles from release site in the Adult survival of Okanagan Sockeye was estimated from Okanagan River Basin to McNary Dam by release year (2012– Bonneville Dam to McNary Dam and from McNary Dam 2019) to Wells Dam for all adult return years (2013–2019). Due to lack of detections at lower sites of fish also detected at Year Release site (by distance) OKP, sample sizes were only sufficient to calculate sur - OSOYOL OSOYHA OSOYBR SKA + SKATAL SKAHAL vival from Wells Dam to OKC in 2018. Detection prob- 2012 NA NA NA 11.4 (0.3) NA abilities ranged from 92 to 100% at Bonneville, McNary, 2013 NA NA 22.8 (0.6) 22.7 (0.6) NA and Wells Dams (Table  6). In most years, as expected, 2014 NA NA 20.2 (0.3) 13.3 (0.2) NA adults traveling from Bonneville to McNary Dam had 2015 NA NA 28.1 (1.0) 20.9 (0.4) NA higher survival than those traveling from McNary to 2016 26.6 (0.8) NA 22.1 (0.7) 17.1 (0.2) 15.9 (0.4) Wells Dam, except in 2015 when McNary to Wells Dam 2017 14.7 (0.2) NA 14.8 (0.4) NA NA had higher survival and in 2019 when survival was equal between these two reaches. Survival in 2015 was low- 2018 24.7 (0.5) 21.2 (0.2) NA NA 13.2 (0.2) est for both reaches. In 2018, Wells Dam to OKC had 2019 25.8 (0.4) NA NA NA 24.2 (0.6) lower survival than the other two downstream reaches OSOYBR Osoyoos Lake Narrows Highway 3 Bridge, OSOYHA Osoyoos Lake at Haynes Point Campground, OSOYOL Osoyoos Lake, SKA + SKATAL Tailrace of (Table 6). Skaha Dam, SKAHAL Skaha Lake Arrival dates of adult Okanagan Sockeye to Bonneville Dam, McNary Dam, and Wells Dam were approximately normal (Fig.  5). At these three lower dams, no differ - Table 5 Smolt-to-adult return rates (SARs) and standard error ences were observed between mean arrival date of fish (SE) in percentages, number of juvenile Okanagan Sockeye originating from Osoyoos- and Skaha-origin tag groups released, and number of adult Okanagan Sockeye detected at (Z < -0.37, p > 0.25). However, arrival date of adults to Bonneville Dam (BON) for each release year (2012–2016) and OKC exhibited a bimodal distribution with arrival date juvenile release site in the Okanagan River Basin peaks occurring on July 14 and October 2. Importantly, Year Release site # Juveniles # Total adults SAR SE differences were observed between fish from Skaha ver - released detected at BON sus Osoyoos tag groups at OKC detection site (Z = 4.84, 2012 Skaha 534 26 4.9% 0.93% p < 0.01). Cumulative probability indicates that a por- 2013 Osoyoos 2783 85 3.1% 0.33% tion of Skaha-origin fish arrives at OKC in July, followed 2013 Skaha 1203 73 6.1% 0.69% by few fish arriving in August and another big pulse in 2014 Osoyoos 3706 35 0.9% 0.16% September to early October (Fig.  6). Whereas Osoyoos- 2014 Skaha 1348 28 2.1% 0.39% origin fish don’t start arriving at OKC en masse until 2015 Osoyoos 1741 7 0.4% 0.15% mid-September, even though they arrive at similar times 2015 Skaha 5435 35 0.6% 0.11% to Wells Dam as Skaha-origin fish. Nonparametric pair - 2016 Osoyoos 4798 24 0.5% 0.10% wise comparisons using the Wilcoxon method indicated 2016 Skaha 5439 55 1.0% 0.14% that arrival timing to Bonneville, McNary, and Wells dams was not statistically different among years (p > 0.10) with exception of 2015 at McNary Dam and 2013 and 2014 at all three detection sites (p ≤ 0.02). At OKC site, differences were observed among some years; however, representation by tagged Okanagan Sockeye from both lakes of origin was not evenly distributed in those years where differences were observed. Okanagan Sockeye traveled at different migration speeds between detection sites on the upstream migra- tion from Bonneville Dam (X = 406.0, p < 0.01). Median migration speed (and 10% to 90% range) decreased as fish moved farther up the Columbia–Okanagan system from Bonneville Dam to McNary Dam, McNary Dam to Wells Dam, and Wells Dam to OKC at 52.0  km/day Fig. 4 Proportion of ocean age-1, age-2, and age-3 Okanagan (39.0 to 63.0  km/day), 36.4  km/day (27.0 to 45.4  km/ Sockeye adults returning to Bonneville Dam by release year (2012– day), and 2.6  km/day (1.8 to 36.4  km/day), respec- 2016) and juvenile tagging basin (Osoyoos Lake and Skaha Lake) in the Okanagan River Basin tively. Because of the large variation in migration speed, Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 10 of 16 Table 6 Survival estimates with standard error (SE) and detection probabilities with standard error for adult Okanagan Sockeye upstream migration from Bonneville to McNary Dam, from McNary to Wells Dam, and from Wells Dam to the Okanagan detection site (OKC) for 2013–2019 Year Survival estimates (SE) Detection probabilities (SE) Bonneville to McNary McNary to wells Wells to OKC Bonneville McNary Wells 2013 1.00 (0.00) 0.92 (0.08) NA 1.00 (0.00) 1.00 (0.01) NA 2014 0.95 (0.03) 0.79 (0.05) NA 0.96 (0.02) 1.00 (0.01) NA 2015 0.60 (0.59) 0.69 (0.06) NA 0.95 (0.03) 1.00 (0.01) NA 2016 0.88 (0.04) 0.85 (0.05) NA 0.98 (0.02) 1.00 (0.00) NA 2017 0.95 (0.03) 0.80 (0.05) NA 0.98 (0.02) 1.00 (0.01) NA 2018 0.85 (0.05) 0.83 (0.05) 0.64 (0.09) 0.95 (0.04) 1.00 (0.00) 0.92 (0.08) 2019 0.95 (0.05) 0.95 (0.05) NA 1.00 (0.01) 1.00 (0.01) NA Survival from Wells to OKC was unavailable prior to installation of the upstream array (OKP) in 2017 and otherwise unestimable due to sample size Fig. 5 Frequency of arrival dates of adult Okanagan Sockeye to Bonneville Dam (A), McNary Dam (B), Wells Dam (C), and OKC PIT array (D) across all years and tagging basins specifically between Wells Dam and OKC, data were slower and with less variation than Skaha-origin fish further compared to examine migration speed among with median migration speed of 2.0  km/day (1.6 to reaches based on which lake of origin (Osoyoos or 3.6 km/day); whereas Skaha-origin fish overall traveled Skaha Lake) fish were tagged. No difference in migra - slightly faster with much larger variation with median tion speed between Osoyoos- and Skaha-origin adults migration speed of 4.7  km/day (2.0 to 40.8  km/day; was observed in the Bonneville to McNary Dam reach Fig. 7). (Osoyoos N = 130, Skaha N = 182; Z = −  0.84, p = 0.40) or in the McNary to Wells Dam reach (Osoyoos N = 99, Discussion Skaha N = 159; Z = −  0.13, p = 0.89). However, there Results from PIT-tagging efforts in the Okanagan River were significant differences in migration speed between Basin over the 8-year study period provide key insights Osoyoos- and Skaha-origin adults traveling from Wells into survival and behavior of the largest population of Dam to OKC (Osoyoos N = 31, Skaha N = 64; Z = -5.38, Sockeye Salmon in the Columbia River Basin. Estimates p < 0.01; Fig .  7). Osoyoos-origin fish traveled much of smolt survival during the downstream migration, M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 11 of 16 tshawytscha have been estimated at 21.5  km/day in the mainstem Columbia River [26]. Travel speed of yearling Chinook Salmon in natal tributaries has been measured at less than 6 km/day [51]. In comparison, travel speed of Okanagan Sockeye smolts through 130 km of Okanagan River and 388  km of Columbia River is generally above 25  km/day. Smolts released in Skaha Lake migrate even faster than those released in Osoyoos Lake, averaging 34 km/day and ranging up to 50 km/day. These observa - tions are comparable or faster than migration speeds in natural and modified systems reported elsewhere [26, 49, 65] and represent an impressive speed greater than 3.0 body lengths per second over an extended distance [3]. Our results are consistent with longstanding characteri- zations of Sockeye Salmon: smolts leave nursery areas in Fig. 6 Cumulative probability of adult arrival date to OKC site from Wells Dam for PIT-tagged adults originating from Skaha Lake (SKA) high numbers over a short period and travel quickly to and Osoyoos lakes (OSO) for all years combined (2013–2019) the ocean [30]. Perhaps more noteworthy is the estimated survival of smolts from release sites in the Okanagan River Basin to SAR and age structure, and upstream conversion rates McNary Dam. Survival averages 49% across all release of adults provide critical information on the viability of groups in our analyses, despite traversing over 130  km the population. The results also highlight extreme vari - in the highly modified Okanagan River and another ation in survival over a short timeframe in response to 388  km of the Columbia River that includes five large- changes in ocean conditions and river temperatures dur- scale hydroelectric projects. This translates to an average ing the spawning migration. Each aspect of the analyses survival rate of 87.0% per 100 km (95% CI 84.5 to 89.5%). is discussed below, including how shifts in environmental In comparison, mean survival of Sockeye Salmon smolts conditions affect Okanagan Sockeye. We then compare from the Sawtooth Valley in Idaho has been estimated Okanagan Sockeye originating from Osoyoos and Skaha between 87.5 and 91.9% per 100  km to the uppermost lakes to underscore the importance of spatial structure, Snake River hydroelectric project with fish passage (from diversity, productivity, and numerical abundance of Alturas, Pettit, and Redfish lakes to Lower Granite Dam, Okanagan Sockeye. Finally, we recommend minor modi- distances of 774, 767, and 750 km, respectively) [28]. Sur- fications to monitoring and evaluation efforts that will vival of these same populations in the mainstem Snake improve the management of Okanagan Sockeye. and Columbia rivers are more comparable (21-year The downstream migration of Okanagan Sockeye average of 81.1% per 100  km) [68], though lower than smolts is relatively efficient compared to other juve - Okanagan Sockeye. Sockeye Salmon smolts emigrat- nile salmonids in the Columbia River Basin. For exam- ing from Cultus Lake in the unimpounded Fraser River ple, travel speeds of yearling Chinook Salmon O. Fig. 7 Box and whisker plot showing migration speed (km/day) of Osoyoos-origin and Skaha-origin adult Okanagan Sockeye traveling from Bonneville Dam to McNary Dam, from McNary Dam to Wells Dam, and from Wells Dam to OKC site combining all years (2013–2019) Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 12 of 16 have considerably lower survival rates of ≈ 50 to 70% extreme variation in ocean conditions [42] and freshwa- per 100  km over 4  years of monitoring, though differ - ter environments [17], we suspect that the diversity in ent technologies were used in the evaluation [65]. Other age structure of Okanagan Sockeye will be a key aspect salmonids in the region also have lower smolt survival in of maintaining population viability in the future [48, 59]. mainstem [68] or tributary habitats [6]. While these esti- Survival of adult Okanagan Sockeye during their mates are not directly comparable, they do suggest that upstream migration in freshwater was highly variable Okanagan Sockeye have relatively high rates of survival and could become the most significant threat to the during the juvenile outmigration. species if increasing temperature trends continue [38, Like most Sockeye Salmon populations, fish originat - 46]. Conversion rates of Okanagan Sockeye through the ing from the Okanagan River Basin had a highly vari- 595  km-reach between Bonneville and Wells dams (i.e., able SAR. Release groups from this study, spanning just the product of the two reach conversion rates presented 5  years, vary by more than an order of magnitude, from in Table  6) ranged from a high of 92% in 2013 to a low 0.4% (2015 Osoyoos Lake releases) to 6.1% (2013 Skaha of 41% in 2015. This indicates that pre-spawn mortal - Lake releases). These estimates (mean of 2.2%) and pre - ity in the mainstem Columbia River during the 2015 vious estimates of Okanagan Sockeye SARs ranging to drought exceeded 250,000 adult Okanagan Sockeye. For more than 12% [Kim Hyatt, Department of Fisheries and added context, this loss is roughly equivalent to 11 mil- Oceans Canada, personal communication] are high rela- lion smolts over the previous 3  years (based on an aver- tive to other populations in the Columbia River Basin. age SAR of 2.2%); 360 million eggs in the 2015 brood Snake River Sockeye Salmon also traverse over 1,000 km year; and more than double the number of adult Sock- and eight hydroelectric projects during the smolt migra- eye Salmon observed returning to the Columbia River tion, though SARs have been found to be much lower, at between 1995 and 1999 [11]. More concerning is the less than 1.0% from 2005 to 2009 [28]. Sockeye Salmon additional losses that occurred above the hydroelectric SARs from the Wenatchee River Basin—88  km and two system in 2015: only 10,400 of the 187,055 (5.6%) adult hydroelectric projects downstream of the Okanagan Okanagan Sockeye observed passing Wells Dam ulti- River confluence—ranged from 0.02 to 2.55%, though mately reached spawning grounds in the Okanagan River these are likely biased low based on recent PIT-tag analy- Basin [38]. While survival of adult Okanagan Sockeye ses and observations of adults obstructed at trapping during the upstream spawning migration is generally facilities below the spawning grounds [31, 52]. In con- favorable relative to other Columbia River Basin stocks trast, all Sockeye Salmon originating from the Columbia (e.g., Snake River Sockeye Salmon) [16] or other salmo- River Basin (the southern extent of the species) generally nids [5], it is clear that droughts and higher water tem- have lower SARs compared to stocks in British Colum- peratures pose a tremendous risk to the population. bia or Alaska [39, 43]. However, Sockeye Salmon in the The most influential condition of variation in spawn - Columbia River Basin are similar to northern counter- ing escapement of Okanagan Sockeye is the thermal parts in that ocean conditions combined with broader- barrier that develops each summer at the confluence of scale climate indices are by far the strongest predictors of the Okanagan and Columbia rivers near Brewster, Wash- SARs [64, 69]. The final detection point used in SAR cal - ington (rkm 859), where water temperatures often dif- culations varies among these populations so corrections fer by 10–15 °C in July and August. The barrier was first for distance traveled, harvest, or natural mortality would reported in the 1960s by researchers that were evaluating provide more comparable estimates. Sockeye Salmon passage at the newly constructed Rocky Age structure of Okanagan Sockeye based on this Reach Hydroelectric Project (rkm 761) in 1961. Once study was mostly consistent with other populations, with water temperatures in the Okanagan River exceed 21 °C, most (69%) fish returning to spawn after 2  years at sea Okanagan Sockeye will not enter the tributary, creating and a measurable proportion of jacks (27%) and a few delays that can last several weeks [45]. The consequence (< 4%) adults that spend 3 years at sea. Age structure was of this delay was highlighted in 2015, where conversion highly variable among years, consistent with previous rates from Bonneville Dam to spawning grounds in 2015 observations of up to six different age combinations of dropped to 4.3% as water temperatures in the Okanagan Okanagan River stocks [22]. In comparison, Wenatchee River remained at or above 21.9  °C for 67 consecutive River Sockeye Salmon have lower diversity in age struc- days [22]. The bimodal distribution of arrival timing for ture and are comprised entirely of age 4 and 5 adults [22, Skaha Lake Okanagan Sockeye (Fig. 6) could prove criti- 31]. Nutrient levels in nursery lakes [62] and conditions cal for population viability since a greater proportion of during early marine entry are known to influence age at adults arrive prior to the summer thermal barrier. maturity in Sockeye Salmon [55] and likely influenced Recreational, tribal, and commercial harvest oppor- the years included in our analyses. Given the recent tunities further complicate management of Okanagan M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 13 of 16 Sockeye during summer months. An average of 19% of Lake [17]. While the younger age structure may increase Okanagan Sockeye are harvested between Wells Dam ocean survival and spawning escapement, the higher (rkm 830) and spawning grounds each year, often when proportion of jacks originating from Skaha Lake may fish are congregated in large numbers at the thermal bar - translate to reduced reproductive success [10]. Run tim- rier [37]. Mortality of fish that evade capture compounds ing is potentially the most important difference in adult this issue: some fishery managers apply a 40–60% mortal - returns of Okanagan Sockeye observed in this study: ity rate for salmon released from commercial gillnets and arrival date to the Okanagan River Basin (OKC site) was 10% for salmon release from recreational fisheries, rates earlier and more protracted for Skaha Lake adults with a that increase with increasing water temperatures [57]. large pulse arriving in July and then again in late Septem- A more detailed analysis that accounts for harvest and ber compared to cohorts tagged in Osoyoos Lake, which water temperatures in conversion rates would provide arrive in mid–late September. Delays in the spawning important insights to managers as droughts in the West- migration of Okanagan Sockeye often translate to higher ern United States increase in frequency and duration rates of mortality, particularly during higher water tem- [30]. Additional research on adult holding and harvest in peratures. Adults with a tendency to migrate earlier may the Canadian portions of Osoyoos Lake would also pro- fare better in future years by avoiding thermal barriers, vide important insights. greater pre-spawn mortality, or inadvertent mortality Key differences in physical attributes and post-release related to commercial and recreational fisheries [33, 45]. performance were observed between Okanagan Sockeye Conversely, later run timing may translate to avoidance released in Skaha Lake versus those released in Osoyoos of higher water temperatures in spawning tributaries, a Lake. Smolts tagged in Skaha Lake were typically larger well-documented occurrence in the Fraser River [32]. In (~ 10–20 mm in median total length) and migrated faster, either case, maintaining diversity in life history strategies though had lower downstream smolt survival com- of Okanagan Sockeye will be critical to sustaining the pared to those tagged in Osoyoos Lake. Survival was population in a changing climate. still lower for smolts tagged in Skaha Lake after adjust- These analyses highlight variability in survival and ing for distance, so some biological or ecological mecha- behavior of Okanagan Sockeye that can inform key nism likely affects smolts derived from hatchery-origin aspects of population viability (i.e., VSP) [48]. First, sur- fry introduced into Skaha Lake. Greater rates of preda- vival during the smolt migration, in the ocean, and dur- tion due to size differences [66], behavioral patterns [58], ing the spawning migration has a profound influence or migration past a specific location of high predation on productivity and, therefore, abundance of Okanagan [25] present possible explanations. Vaseux Lake, located Sockeye. While ocean survival [69] and increased escape- downstream of Skaha Lake, is relatively shallow and hosts ments of Okanagan Sockeye [33, 36] have generally large populations of Northern Pikeminnow Ptychocheilus driven productivity and abundance of Columbia River oregonensis and Smallmouth Bass Micropterus dolomieu Sockeye Salmon, the 2015 drought clearly illustrates how that likely influence juvenile survival. Cessation of migra - above-average water temperatures during the spawn- tion or alternative life history patterns (i.e., non-migrants ing migration can negate the production of millions of are perceived as mortality in mark–recapture modeling) fish during the last months of life. Second, the increased once Skaha Lake smolts encounter the productive Osoy- spatial structure (through reintroduction into historical oos Lake is another possibility [37]. Finally, differences spawning and rearing locations) and diversity (includ- between hatchery- and wild-origin juvenile Okanagan ing migratory timing, smolt size, and adult age structure) Sockeye could explain differences in survival during the highlighted in our analyses demonstrate how manage- smolt emigration [35, 56]. ment efforts have directly improved biodiversity of Despite lower smolt survival in freshwater, fish origi - Okanagan Sockeye. nating from Skaha Lake had higher SARs than cohorts The importance of rebuilding diverse wild popula - tagged in Osoyoos Lake. This may be a result of size or tions, or functioning portfolios, has been demonstrated time at ocean entry [18], their differing age structure [54], in other Sockeye Salmon populations [57]. The findings size-selective harvest [41], or another mechanism that reported here and impressive recovery of Okanagan we are unable to account for in these analyses. Adults Sockeye through the early 2010s demonstrate their resil- originating from Skaha Lake had a younger age struc- ience to a highly modified environment, whereas the ture compared to cohorts tagged in Osoyoos Lake, with drought in 2015 reminded managers how years of suc- a significantly greater proportion of jacks (34% versus cess can be erased by changing river conditions. Routine 18% of adults, respectively). Research has demonstrated monitoring to inform management efforts that improve how selective pressures can increase and perpetuate the aspects of population viability will be critical in pre- prevalence of jacks, which could be a factor in Skaha serving Okanagan Sockeye over the next century. The Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 14 of 16 Availability of supporting data current monitoring framework and analyses presented The data sets used and/or analyzed during the current study are available here could be improved by (1) additional PIT-tagging and from the corresponding author on reasonable request. improved detection arrays; (2) more detailed analyses of metrics presented herein; and (3) consideration of addi- Declarations tional monitoring tools, such as life cycle modeling [45], Ethics approval and consent to participate that can be used to understand which components of the Data analyzed in this manuscript were obtained under an existing fish tagging population warrant the greatest level of protection. program. All tagging followed regionally accepted protocols that minimize stress and injury to PIT-tagged salmonids. The document is available at http:// www. bioma rk. com/ Docum ents% 20and% 20Set tings/ 67/ Site% 20Doc uments/ Abbreviations PDFs/ Fish% 20Tag ging% 20Met hods. pdf BO1: Bonneville Bradford Island Ladder; BO2: Bonneville Cascades Island Lad- der; BO3: Bonneville Washington Shore Ladder; BO4: Bonneville Washington Consent for publication Ladder slots; MC1: McNary Oregon Shore Ladder; MC2: McNary Washington Not applicable; data are not derived from any individual person. Shore Ladder; WEA: Wells Dam adult fishways; BON: Bonneville Dam; CJS: Cormack–Jolly–Seber; CRITFC: Columbia River Inter-Tribal Fish Commission; Competing interests DART : Data Acquisition in Real Time; DFO: Fisheries and Oceans Canada; OKC: The authors declare that they have no competing interests. Okanagan Channel instream detection site; OKP: Penticton Channel instream detection array; OSOYOL: Osoyoos Lake; OSOYHA: Osoyoos Lake at Haynes Author details Point Campground; OSOYBR: Osoyoos Lake Narrows Highway 3 Bridge; PIT: Four Peaks Environmental Science & Data Solutions, Wenatchee, USA. Passive integrated transponder; PTAGIS: Columbia Basin PIT Tag Information Fisheries and Oceans Canada, Pacific Biological Station, Nanaimo, Canada. 3 4 System (www. ptagis. org); SKA: Skaha; SKATAL: Skaha Tailrace; SKAHAL: Skaha Okanagan Nation Alliance, Westbank, Canada. Columbia River Inter-Tribal Lake; SAR: Smolt-to-adult return; SE: Standard error; VSP: Viable Salmonid Fish Commission, Portland, USA. Population. Received: 21 May 2021 Accepted: 21 August 2021 Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40317- 021- 00262-y. References 1. Benson RD. Ongoing actions, ongoing issues: trying again to free federal Additional file 1: Appendix S1. Tag Files from PTAGIS of juvenile Sockeye dams from the ESA. Envtl L Rep. 2019;49:11019. tagged in the Okanagan basin from 2012 to 2019 used in juvenile survival 2. Biomark. Fish tagging methods. http:// www. bioma rk. com/ Docum ents% and travel time queries and estimates. 20and% 20Set tings/ 67/ Site% 20Doc uments/ PDFs/ Fish% 20Tag ging% 20Met hods. pdf; 2012. Accessed Nov 2013. 3. Brett JR. Swimming performance of sockeye salmon (Oncorhynchus Acknowledgements nerka) in relation to fatigue time and temperature. J Fish Board Canada. Fisheries and Oceans Canada and Four Peaks Environmental Science & Data 1967;24(8):1731–41. Solutions provided funding for this manuscript. Okanagan Nation Alliance has 4. Buchanan RA, Skalski JR. A migratory life-cycle release-recapture led Sockeye Salmon recovery efforts in the upper Columbia River Basin and model for salmonid PIT-tag investigations. J Agric Biol Environ Stat. monitoring activities, including the management of PIT-tagging and detec- 2007;12(3):325–45. tion arrays. The Bonneville Power Administration and the public utility districts 5. Buchanan RA, Skalski JR. Using multistate mark-recapture methods to (PUDs) of Chelan, Douglas, and Grant counties provided biological support model adult salmonid migration in an industrialized river. Ecol Model. and programmatic funding for multiple aspects of the recovery efforts for 2010;221(4):582–9. Okanagan Sockeye Salmon. Chelan and Grant PUDs provided funding for PIT 6. Buchanan RA, Skalski JR, Mackey G, Snow C, Murdoch AR. Estimating infrastructure, tags, and the kł cp̓əlk̓ stim̓ Hatchery, which was instrumental cohort survival through tributaries for salmonid populations with variable in reintroducing Sockeye Salmon to historical spawning habitat in Skaha and ages at migration. North Am J Fish Manag. 2015;35(5):958–73. Okanagan lakes. Molly Steere, Carolyn Roper, and Nate Hough-Snee of Four 7. Burgner RL. Life history of sockeye salmon (Oncorhynchus nerka). Pacific Peaks Environmental Science & Data Solutions contributed to the manuscript salmon life histories. 1991, p. 3–117. through technical review, GIS support, and interpretation of results. Tom 8. Burnham KP, Anderson DR. A practical information-theoretic approach. Kahler of Douglas County PUD and two anonymous reviewers provided excel- Model selection and multimodel inference. 2002, p. 2. lent feedback to improve the content of this manuscript. We thank the late 9. Bussanich R, Hyatt K, Wright H. Proceedings of an expert’s workshop on Kim Hyatt (“Dr. Sockeye”) for his vision, expertise, leadership, and enthusiasm Columbia river and hydro-system impacts on migration success and for Sockeye Salmon. He is responsible for many efforts to recover Okanagan production variations of Anadromous Salmon 2017 Dec. 6–7, p. 28. Sockeye and with his monumental contributions he has left us with a priceless 10. Carlson SM, Rich HB Jr, Quinn TP. Reproductive life-span and sources of and enduring legacy that will require our continued dedication to maintain. mortality for alternative male life-history strategies in sockeye salmon, Oncorhynchus nerka. Can J Zool. 2004;82(12):1878–85. Authors’ contributions 11. CBR (Columbia Basin Research). Columbia Basin Research. www. cbr. washi JM led study design, analyses, data interpretation, and technical writing. KH ngton. edu. Accessed 18 Mar 2020. and JF supported technical writing and data interpretation. EK provided 12. Chapman DW. Salmon and steelhead abundance in the Columbia River statistical support and conducted regression analyses. SF and RB led program in the nineteenth century. Trans Am Fish Soc. 1986;115(5):662–70. development and data collection, also providing support on data interpre- 13. Columbia River DART (Columbia Basin Research, University of Washing- tation and technical writing. KS performed initial data queries and survival ton). DART PIT Tag Release and Observation Summary. 2020. http:// www. analyses. All authors read and approved the final manuscript. cbr. washi ngton. edu/ dart/ query/ pit_ sum_ tagfi les 14. Cormack RM. Estimates of survival from the sighting of marked animals. Funding Biometrika. 1964;51(3/4):429–38. This manuscript was funded by Fisheries and Oceans Canada with additional 15. Cosens B, Fremier A. Social-ecological resilience in the Columbia river support from Four Peaks Environmental Science & Data Solutions. basin: the role of law and governance. In: Practical panarchy for adaptive water governance. 2018, p. 47–64. M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 15 of 16 16. Crozier EG, Burke BJ, Sandford BP, Axel GA, Sanderson B. Passage and 36. Hyatt KD, Stockwell MM, Rankin DP. Impact and adaptation responses survival of adult Snake River sockeye salmon within and upstream from of Okanagan River sockeye salmon (Oncorhynchus nerka) to climate the Federal Columbia River Power System. 2014. variation and change effects during freshwater migration: stock res- 17. DeFilippo LB, Schindler DE, Ohlberger J, Schaberg KL, Birch Foster M, Ruhl toration and fisheries management implications. Can Water Resour J. D, Punt AE. Recruitment variation disrupts the stability of alternative life 2003;28(4):689–713. histories in an exploited salmon population. Evol Appl. 2019;12(2):214–29. 37. Hyatt KD, McQueen DJ, Ogden AD. Have invasive mysids (Mysis diluviana) 18. Farley EV, Murphy JM, Adkison MD, Eisner LB, Helle JH, Moss JH, Nielsen J. altered the capacity of Osoyoos Lake, British Columbia to produce sock- Early marine growth in relation to marine-stage survival rates for Alaska eye salmon (Oncorhynchus nerka)? Open Fish Sci J. 2018;11(1):1–26. sockeye salmon (Oncorhynchus nerka). Fish Bull. 2007;105(1):121–30. 38. Hyatt KD, Stiff HW, Stockwell MM. Historic water temperature (1924– 19. Folks S, Bussanich R, Stevens A, Teather M. q̓awsitk (Okanagan River) 2018), river discharge (1929–2018), and adult sockeye salmon migration sockeye smolt out-of-Basin survival: PIT tagging 2016. Westbank, BC: (1937–2018) observations in the Columbia, Okanogan, and Okanagan Okanagan Nation Aquatic Enterprises Ltd.; 2016. p. 12. Rivers. Fisheries and Oceans Canada= Pêches et Océans; 2020. 20. Folks S, Teather M, Benson R. q̓awsitk (Okanagan River) sockeye smolt 39. Irvine JR, Akenhead SA. Understanding smolt survival trends in sockeye out-of-Basin survival: purse seining and PIT tagging BY 2015. Westbank, salmon. Mar Coast Fish. 2013;5(1):303–28. BC: Okanagan Nation Aquatic Enterprises Ltd.; 2017. p. 16. 40. Jolly GM. Explicit estimates from capture-recapture data with both death 21. Fryer JK. Columbia Basin sockeye salmon: causes of their past decline, fac- and immigration-stochastic model. Biometrika. 1965;52(1/2):225–47. tors contribution to their present low abundance, and the outlook for the 41. Kendall NW, Quinn TP. Quantifying and comparing size selectivity among future [dissertation]. Seattle ( WA): University of Washington; 1995. p. 274. Alaskan sockeye salmon fisheries. Ecol Appl. 2012;22(3):804–16. 22. Fryer JK, Kelsey D, Wright H, Folks S, Bussanich R, Hyatt KD, Stockwell MM. 42. Kintisch E. ‘The Blob’ invades Pacific, flummoxing climate experts. Science. Studies into factors limiting the abundance of Okanagan and Wenatchee 2015;348(6230):17–8. sockeye salmon in 2015. Columbia River Inter-Tribal Fish Commission 43. Koenings JP, Geiger HJ, Hasbrouck JJ. Smolt-to-aduit survival pat- Technical Report 17-06; 2017. p. 217. terns of sockeye salmon (Oncorhynchus nerka): effects of smolt length 23. Fryer JK, Wright H, Folks S, Alliance ON, Hyatt KD, Stockwell MM. Limiting and geographic latitude when entering the sea. Can J Fish Aquat Sci. factors of the abundance of Okanagan and Wenatchee Sockeye Salmon 1993;50(3):600–11. in 2011 Columbia River Inter-Tribal Fish Commission Technical Report for 44. Ledgerwood RD, Ryan BA, Dawley EM, Nunnallee EP, Ferguson JW. A BPA Project 2008-503-00; 2012. surface trawl to detect migrating juvenile salmonids tagged with passive 24. Fulton LA. Spawning areas and abundance of steelhead trout and coho, integrated transponder tags. North Am J Fish Manag. 2004;24(2):440–51. sockeye, and chum salmon in the Columbia River basin—past and 45. Lessard RB, Hilborn R, Chasco BE. Escapement goal analysis and stock present. Spec Sci Rep Fish. 1970;618:37. reconstruction of sockeye salmon populations (Oncorhynchus nerka) 25. Furey NB, Hinch SG, Lotto AG, Beauchamp DA. Extensive feeding on using life-history models. Can J Fish Aquat Sci. 2008;65(10):2269–78. sockeye salmon Oncorhynchus nerka smolts by bull trout Salvelinus con- 46. Major RL, Mighell JL. Influence of Rocky Reach Dam and the temperature fluentus during initial outmigration into a small, unregulated and inland of the Okanogan River on the upstream migration of sockeye salmon. British Columbia river. J Fish Biol. 2015;86(1):392–401. Fish Bull. 1967;1(66):131–47. 26. Giorgi AE, Hillman TW, Stevenson JR, Hays SG, Peven CM. Factors that 47. Matala AP, Narum SR, Saluskin BP, Johnston MV, Newell JE, Fast DE, Gal- influence the downstream migration rates of juvenile salmon and steel- breath PF. Early observations from monitoring a reintroduction program: head through the hydroelectric system in the mid-Columbia River basin. return of Sockeye salmon to a nursery lake of historical importance. Trans North Am J Fish Manag. 1997;17(2):268–82. Am Fish Soc. 2019;148(2):271–88. 27. Gresh T, Lichatowich J, Schoonmaker P. An estimation of historic and 48. McElhany P, Rucklelshaus MH, Ford MJ, Wainwright TC, Bjorkstedt EP. Via- current levels of salmon production in the Northeast Pacific ecosystem: ble salmonid populations and the recovery of evolutionarily significant evidence of a nutrient deficit in the freshwater systems of the Pacific units. U.S. Dept. Commer., NOAA Tech. Memo. NMFS-NWFSC-42. 2000. p. Northwest. Fisheries. 2000;25(1):15–21. 156. 28. Griswold RG, Kohler AE, Taki D. Survival of endangered Snake River sock- 49. Melnychuk MC, Welch DW, Walters CJ. Spatio-temporal migration pat- eye salmon smolts from three Idaho lakes: relationships with parr size at terns of Pacific salmon smolts in rivers and coastal marine waters. PLoS release, parr growth rate, smolt size, discharge, and travel time. North Am ONE. 2010;5(9):e12916. J Fish Manag. 2011;31(5):813–25. 50. Merritt WS, Alila Y, Barton M, Taylor B, Cohen S, Neilsen D. Hydrologic 29. Gustafson RG, Wainwright TC, Winans GA, Waknitz FW, Parker LT, Waples response to scenarios of climate change in sub watersheds of the RS. Status review of sockeye salmon from Washington and Oregon. U.S. Okanagan basin, British Columbia. J Hydrol. 2006;326(1–4):79–108. Dept. Commer., NOAA Tech. Memo. NMFS-NWFSC-33; 1997. p. 282. 51. Monzyk FR, Jonasson BC, Hoffnagle TL, Keniry PJ, Carmichael RW, Cleary 30. Hao Z, Hao F, Singh VP, Zhang X. Changes in the severity of compound PJ. Migration characteristics of hatchery and wild spring Chinook salmon drought and hot extremes over global land areas. Environ Res Lett. smolts from the Grande Ronde River basin, Oregon, to Lower Granite 2018;13(12):124022. Dam on the Snake River. Trans Am Fish Soc. 2009;138(5):1093–108. 31. Hillman T, Miller M, Willard C, Hopkins S, Johnson M, Hughes M, et al. 52. Murauskas JG, Fryer JK, Nordlund B, Miller JL. Trapping effects and fisher - Monitoring and evaluation of the Chelan and Grant County PUDs ies research: a case study of sockeye salmon in the Wenatchee River. USA hatchery programs. HCP Hatchery Committees and the PRCC Hatchery Fisheries. 2014;39(9):408–14. Sub-Committee, Wenatchee and Ephrata, WA. 2018. 53. Osborn RP. Climate change and the Columbia river treaty. Washington J 32. Hinch SG, Cooke SJ, Farrell AP, Miller KM, Lapointe M, Patterson DA. Dead Environ Law Policy. 2012;2(1):75–123. fish swimming: a review of research on the early migration and high 54. Peterman RM. Model of salmon age structure and its use in presea- premature mortality in adult Fraser River sockeye salmon Oncorhynchus son forecasting and studies of marine survival. Can J Fish Aquat Sci. nerka. J Fish Biol. 2012;81(2):576–99. 1982;39(11):1444–52. 33. Hyatt KD, Stockwell MM. 2019. Chasing an illusion? Successful restoration 55. Peterman RM. Patterns of interannual variation in age at maturity of of Okanagan River Sockeye Salmon (Oncorhynchus nerka) in a sea of sockeye salmon (Oncorhynchus nerka) in Alaska and British Columbia. Can uncertainty. In: CC Krueger, WW Taylor, S Youn (eds.) From catastrophe to J Fish Aquat Sci. 1985;42(10):1595–607. recovery: stories of fisheries management successes. Bethesda: American 56. Powell MS, Hardy RW, Flagg TA, Kline PA. Proximate composition and fatty Fisheries Society. 2019, p. 65–100. acid differences in hatchery-reared and wild Snake River sockeye salmon 34. Hyatt K, Stockwell M, Stiff H, Ferguson R. Salmon responses to hydrocli- overwintering in nursery lakes. North Am J Fish Manag. 2010;30(2):530–7. matological conditions in British Columbia in 2015. State of the physical, 57. Price MHH, Moore JW, Connors BM, Wilson KL, Reynolds JD. Portfolio biological and selected fishery resources of pacific Canadian Marine simplification arising from a century of change in salmon population Ecosystems in. 2016, vol. 44, p. 198–205. diversity and artificial production. J Appl Ecol. 2021;58(7):1477–86. 35. Hyatt KD, Mathias KL, McQueen DJ, Mercer B, Milligan P, Rankin DP. Evalu- 58. Scheuerell MD, Schindler DE. Diel vertical migration by juvenile sockeye ation of hatchery versus wild sockeye salmon fry growth and survival in salmon: empirical evidence for the antipredation window. Ecology. two British Columbia lakes. North Am J Fish Manag. 2005;25(3):745–62. 2003;84(7):1713–20. Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 16 of 16 59. Schindler DE, Hilborn R, Chasco B, Boatright CP, Quinn TP, Rogers LA, 67. White GC, Burnham KP. Program MARK: survival rate estimation from both Webster MS. Population diversity and the portfolio effect in an exploited live and dead encounters. Bird Study. 1999;46(Supplement):S120–39. species. Nature. 2010;465(7298):609–12. 68. Widener DL, Faulkner JR, Smith SG, Marsh TM, Zabel RW. Survival esti- 60. Seber GA. A note on the multiple-recapture census. Biometrika. mates for the passage of spring-migrating juvenile salmonids through 1965;52(1/2):249–59. Snake and Columbia River dams and reservoirs, 2017. Fish Ecology Divi- 61. Simmatis B, Jeziorski A, Zemanek A, Selbie DT, Hyatt K, Fryer JK, Cumming sion, Northwest Fisheries Science Center 2018. BF, Smol JP. Long-term reconstruction of deep-water oxygen conditions 69. Williams JG, Smith SG, Fryer JK, Scheuerell MD, Muir WD, Flagg TA, Zabel in Osoyoos Lake (British Columbia, Canada): implications for Okanagan RW, Ferguson JW, Casillas E. Influence of ocean and freshwater conditions River sockeye salmon. Lake Reservoir Manage. 2018;34(4):392–400. on Columbia River sockeye salmon Oncorhynchus nerka adult return 62. Stockner JG, Macisaac EA. British Columbia lake enrichment programme: rates. Fish Oceanogr. 2014;23(3):210–24. two decades of habitat enhancement for sockeye salmon. Regul Rivers: 70. Yaniw N, Benson R. Okanagan River Sockeye Smolt Out-of-Basin Survival: Res Manage. 2021;12(4–5):547–61. Purse Seining and PIT Tagging BY 2016. Westbank, BC: Prepared by 63. Tenney J, Warf D, Tancreto N. Columbia basin PIT tag information system. Okanagan Nation Aquatic Enterprises Ltd.; 2018. p. 13. 2016 annual report. Report to the Bonneville Power Administration, 71. Zhang X, Li HY, Deng ZD, Leung LR, Skalski JR, Cooke SJ. On the Project. 1990, p. 080-0. variable effects of climate change on Pacific salmon. Ecol Model. 64. Tucker S, Thiess ME, Morris JF, Mackas D, Peterson WT, Candy JR, Beacham 2019;1(397):95–106. TD, Iwamoto EM, Teel DJ, Peterson M, Trudel M. Coastal distribution and 72. Zuur AF, Ieno EN, Elphick CS. A protocol for data exploration to avoid consequent factors influencing production of endangered Snake River common statistical problems. Methods Ecol Evol. 2010;1(1):3–14. sockeye salmon. Trans Am Fish Soc. 2015;144(1):107–23. 65. Welch DW, Melnychuk MC, Rechisky ER, Porter AD, Jacobs MC, Ladouceur Publisher’s Note A, McKinley RS, Jackson GD. Freshwater and marine migration and Springer Nature remains neutral with regard to jurisdictional claims in pub- survival of endangered Cultus Lake sockeye salmon (Oncorhynchus nerka) lished maps and institutional affiliations. smolts using POST, a large-scale acoustic telemetry array. Can J Fish Aquat Sci. 2009;66(5):736–50. 66. West CJ, Larkin PA. Evidence for size-selective mortality of juvenile sock- eye salmon (Oncorhynchus nerka) in Babine Lake, British Columbia. Can J Fish Aquat Sci. 1987;44(4):712–21. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Animal Biotelemetry Springer Journals

Migration and survival of Okanagan River Sockeye Salmon Oncorhynchus nerka, 2012–2019

Loading next page...
 
/lp/springer-journals/migration-and-survival-of-okanagan-river-sockeye-salmon-oncorhynchus-gj5E1y1ESo
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2050-3385
DOI
10.1186/s40317-021-00262-y
Publisher site
See Article on Publisher Site

Abstract

Background: Okanagan River Sockeye Salmon Oncorhynchus nerka (Okanagan Sockeye) are one of two remain- ing self-sustaining Sockeye Salmon populations in the Columbia River Basin. We used detection histories of smolts implanted with passive integrated transponder (PIT ) tags between 2012 and 2019 to estimate survival and behavioral metrics during reintroduction efforts and changing environmental conditions over the monitoring period. Results: Smolts migrating to McNary Dam, whose route includes 130 km of the Okanagan River and 388 km of the Columbia River, generally had high survival (mean of 87.0% per 100 km) and fast migration speeds (up to 50 km/day) relative to other salmonids in the region. Smolt-to-adult returns (SARs) ranged from 0.4 to 6.1% and were greater for fish originating from Skaha Lake compared to cohorts tagged in Osoyoos Lake. Most adults returned after 2 years in the ocean (69%), followed by jacks (27%), and adults that spent 3 years at sea (4%), though Skaha Lake adults had a significantly younger age structure than cohorts from Osoyoos Lake. Survival of adults from Bonneville Dam (rkm 235) upstream to Wells Dam (rkm 830) was generally high (80–92%), and migration speed decreased in upstream reaches. Survival from Wells Dam to the Okanagan River was only estimable in 2018, where 64% of adults survived to the spawning grounds. The upstream migration of adult Okanagan Sockeye was significantly compromised during the drought of 2015 when less than 5% of Okanagan Sockeye that returned to the Columbia River reached spawning grounds. Conclusions: Our results indicate that Okanagan Sockeye have exceptional survival and migratory ability relative to other salmonids, though poor ocean conditions combined with warming water temperatures in freshwater habitats in recent years have the potential to devastate the population. The success of reintroduction efforts to increase spatial structure and diversity of Okanagan Sockeye is, therefore, critical to maintaining the population in years to come. Keywords: Sockeye Salmon, PIT tags, Columbia River Basin, Survival inhabited the Columbia River Basin. Sockeye Salmon Background once existed in eight Columbia River tributaries, includ- Okanagan River Sockeye Salmon Oncorhynchus nerka ing at least 12 nursery lakes above the Snake River con- (Okanagan Sockeye) originating from Osoyoos Lake fluence [21, 24, 47], and peak returns likely ranged from (Canada), and Lake Wenatchee Sockeye Salmon (United 2.5 to 3.2 million adults [12, 21]. Degradation of and States) comprise the last two self-sustaining populations blocked access to habitat, urban and industrial devel- of anadromous O. nerka among several that formerly opment, and overexploitation have greatly diminished returns [27, 36], and by the late 1900s, Columbia River *Correspondence: jmurauskas@fourpeaksenv.com Sockeye Salmon had been virtually extirpated except for Four Peaks Environmental Science & Data Solutions, Wenatchee, USA populations in the Okanagan (spelled Okanogan in the Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 2 of 16 U.S.) and Wenatchee River basins [21, 29, 47]. Despite ongoing threats, Okanagan Sockeye have demonstrated perhaps the greatest recovery of any salmonid population in the Pacific Northwest as returns have increased by an order of magnitude over the last 30  years. Mean counts at Wells Dam increased from 17,083 adults during the 1990s to 199,907 adults during the 2010s [13]. The recov - ery is largely a response to water resource management in the Okanagan Basin, increased smolt survival through hydroelectric projects, a period of favorable ocean con- ditions, and the Skaha Lake reintroduction program, including the kł cp̓əlk̓ stim̓ Hatchery in Penticton, British Columbia [33, 69]. While Okanagan Sockeye have demonstrated excep- tional resilience, anthropogenic actions continue to influence the population at all life stages. Urbanization, agricultural operations, and water quality and quantity in spawning and early rearing habitats still dominate management discussions; although, anticipating risks to the species has become increasingly complicated under increasing water temperatures [50]. For example, more than 90% of Okanagan Sockeye that entered the Colum- bia River did not survive to the spawning grounds dur- ing record high water temperatures observed in 2015 [22, 34, 38]. The effects of the 2015 drought persisted through Fig. 1 Overview of the Columbia River migratory corridor (top right); 2019 when the dominant year class of that brood year the Okanagan River Basin (bottom right); and juvenile release sites (open circles), PIT arrays (open triangles), and dams (closed circles) returned at roughly 25% of the recent 10-year average in the Okanagan River Basin (left). Dams passed during migration [11]. With temperatures expected to continue increasing include (1) Bonneville, (2) The Dalles, (3) John Day, (4) McNary, (5) in the Columbia River Basin [71], supporting the recov- Priest Rapids, (6) Wanapum, (7) Rock Island, (8) Rocky Reach, (9) Wells, ery of imperiled Pacific Salmon stocks is a critical goal (11) Zosel, (12) McIntyre, (13) Skaha, and (14) Penticton dams. Chief of water management agreements between the United Joseph Dam (10) is shown for reference but does not have upstream passage facilities States and Canada [15, 53] and litigation concerning hydroelectric development in the Columbia River [1]. Within the commonly referenced Viable Salmonid Population (VSP) framework, the persistence of salmo- and decreasing hypolimnetic oxygen concentrations nid populations is characterized in terms of their abun- resulting from anthropogenic eutrophication greatly dance, productivity, spatial structure, and diversity [48]. reduce suitable nursery habitat for juveniles during late Okanagan Sockeye have reached record levels of abun- summer each year [61]. This situation emphasizes the dance and productivity, but the lack of spatial structure importance of interventions to increase the diversity and and diversity presents risk to the population, especially spatial structure for Okanagan Sockeye. with increases in water temperatures observed in recent Beginning in the late 1990s, efforts to reintroduce years. Habitat in the upper Okanagan River Basin Okanagan Sockeye to historical spawning and rearing became inaccessible to anadromous fishes following habitat in the upper Okanagan River Basin was initiated construction of several dams in the 1900s that blocked by a diverse group of stakeholders led by the Okanagan access to Vaseux Lake (McIntyre Dam, 1921), Skaha Nation Alliance, a First Nations Council in Canada, Lake (Okanagan Falls Dam, 1958), and Okanagan Lake with critical support from Fisheries and Oceans Canada (Penticton Dam, 1915; Fig.  1). Most of the 20th-century (DFO), public utility districts in the United States, and spawning and rearing habitat of Okanagan Sockeye in the the Habitat Conservation Trust Foundation. Reintroduc- Okanagan River Basin [57] was, therefore, restricted to tion of Okanagan Sockeye into Skaha Lake commenced approximately 20 km of the Okanagan River near Oliver, in 2004 [9]. By the 2010s, up to 10% of adult returns to British Columbia, and Osoyoos Lake [33]. Most juvenile the Okanagan River Basin comprised hatchery-origin Okanagan Sockeye rear in portions of Osoyoos Lake, fish released as fry from the newly constructed the kł where increasing water temperatures of the epilimnion M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 3 of 16 cp̓ lk̓ stim̓ Hatchery and upstream passage was improved downstream and pass through the Osoyoos Lake Nar- or reestablished at McIntyre Dam below Vaseux Lake rows, which connects the central and north basins of (2009) and Okanagan Falls Dam below Skaha Lake (2011) the lake. From Zosel Dam at the outlet of Osoyoos Lake, [33]. To track the success of these efforts, a monitor - the Okanagan River flows south through the United ing and evaluation program was implemented in 2005. States, converging with the Columbia River near Brew- Okanagan Nation Alliance staff later worked to create ster, Washington, in the reservoir created by Wells Dam. an extensive passive integrated transponder (PIT) tag Smolts then migrate through nine hydroelectric projects monitoring network through the freshwater migration along the Columbia River to reach the Pacific Ocean corridor. The combined efforts to protect, enhance, and (Fig.  1). Okanagan Sockeye then spend anywhere from 1 monitor Okanagan Sockeye have proven critical to suc- to 3 years in the ocean before returning to migrate up the cessful management of one of the most culturally, eco- Columbia River to spawning grounds in the Okanagan logically, and economically important fisheries in the River. Interior Columbia River Basin. The purpose of this manuscript is to provide the first Tagging methods and release sites comprehensive analysis of the migration behaviors and Okanagan Sockeye smolts were captured and PIT-tagged population dynamics of Okanagan Sockeye based on data at five sites from 2012 to 2019 within the Okanagan from juveniles that were PIT-tagged and released in the River Basin to measure juvenile outmigration sur- Okanagan Basin. These data may be used to provide esti - vival and travel time from release site to McNary Dam, mates of smolt survival during the downstream migra- SAR from the release site back to Bonneville Dam, and tion, age-at-maturity, smolt-to-adult returns (SARs), adult survival and travel time upstream from Bonnev- conversion rates of adults during the upstream migra- ille Dam through the Columbia and Okanagan rivers. tion, and a variety of related behavioral metrics, such as Tagging and release sites were identified by codes des - travel time and date of arrival [4, 9, 52]. Results presented ignated by the Columbia Basin PIT Tag Information Sys- here will provide survival estimates and other data across tem (PTAGIS; www. ptagis. org), and included: Osoyoos critical life stages that can be used to inform recovery Lake (OSOYOL), Osoyoos Lake at Haynes Point Camp- strategies that optimize measures in the VSP framework. ground (OSOYHA), Osoyoos Lake Narrows Highway Specifically, smolt survival, SARs, and adult conversion 3 Bridge (OSOYBR), the tailrace approximately 0.5  km rates to spawning habitat exert profound influences on downstream of Skaha Dam (SKATAL), and Skaha Lake salmon abundance and productivity. In addition, behav- (SKAHAL) (Fig.  1). Fish labeled as released at SKA and ioral information and survival of Okanagan Sockeye orig- SKATAL sites were pooled in each year as this was the inating from re-established habitat are critical to plans same release site. Smolts were captured using either fyke to improve spatial structure and diversity of the popula- nets, rotary screw traps, purse seines, or a combination tion. Finally, recommendations for expanding the current at each site over the study years (Table  1). From 2013 to monitoring framework are provided so managers can 2017, all or a portion of smolts from Osoyoos Lake were understand how to best allocate limited resources as con- captured using a floating fyke net at the OSOYBR site servation efforts evolve in coming years. (see methods in 2). Starting in 2017 purse seine methods were primarily used in all Osoyoos sites [20]. An 8.5  m Methods purse seiner fishing with 183 m seine net (1.27 cm knot - Study area ted mesh) was used to capture smolts at all Osoyoos Lake The Okanagan River is a major tributary to the Columbia and SKAHAL sites from 2016 to 2019. Purse seines fished River and has an approximate length of 185  km (37  km to a depth of 12 m, concentrating in the central basin of Canadian section, 148  km United States section). Wild- Osoyoos Lake and in the southern region of Skaha Lake origin Okanagan Sockeye spawn primarily in the river where most Okanagan Sockeye smolts congregate. From between Osoyoos and Skaha lakes (Fig.  1) during Octo- 2012 to 2016, smolts emigrating from Skaha Lake were ber [62]. Fry emerge from late April to early May and captured using two rotary screw traps that operated generally spend 1  year rearing in Osoyoos Lake, and/ intermittently throughout the outmigration period [19] or the Okanagan River before emigrating as smolts [7, downstream of Okanagan Falls Dam. The success of the 33]. Hatchery- and, more recently, wild-origin smolts program resulted in the increase of target number of PIT- that rear in Skaha Lake pass through Okanagan Falls tagged smolts from a combined total of 5,000 for Osoy- Dam, continuing their journey down the Okanagan oos and Skaha lakes to 5,000 for each lake (Table 1) [70]. River through McIntyre Dam and Osoyoos Lake (Fig. 1). Smolts were anesthetized using a 40  mg/l solution of Okanagan Sockeye smolts from both lakes begin their tricaine methanesulfonate (MS-222) and tagged with outmigration at similar times (April–May). Smolts travel Biomark HPT 12 PIT tags (134.2  kHz; 12  mm length) Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 4 of 16 Table 1 Number of PIT-TAGGED Okanagan Sockeye by PTAGIS release site code, capture method, and release year (2012–2019) Year OSOYOL OSOYHA OSOYBR SKA + SKATAL SKAHAL TOTAL Purse seine Purse seine Fyke net Purse seine Screw trap Purse seine Combined 2012 534 534 2013 2783 1203 3986 2014 3706 1348 5054 2015 1741 5435 7176 2016 3044 1754 3101 2338 10,237 2017 8794 152 2,642 11,588 2018 1521 3562 5860 10,943 2019 4968 4114 9082 OSOYBR Osoyoos Lake Narrows Highway 3 Bridge, OSOYHA Osoyoos Lake at Haynes Point Campground, OSOYOL Osoyoos Lake, SKA + SKATAL Tailrace of Skaha Dam, SKAHAL Skaha Lake according to procedures outlined by PTAGIS and Bio- detections of PIT-tagged adults at various points. The mark [2]. Tags were implanted with an MK-25 Rapid first detection site through which all adult Okanagan Implant Gun along with APT12 pre-loaded needles (Bio- Sockeye must pass along their return migration are the mark, Boise, Idaho). In addition to receiving a PIT tag, fish ladders at Bonneville Dam. These detection points a proportion of smolts were measured for fork length (PTAGIS Site Code) include arrays in the Bradford Island (mm) each year. Following tagging, all mortalities were Ladder (BO1), Bonneville Cascades Island Ladder (BO2), removed from the tagged population. Visible determi- Bonneville WA Shore Ladder (BO3), and Bonneville WA nation of hatchery versus wild origin was not possible Ladder slots (BO4). The next upstream detection site because hatchery Okanagan Sockeye were released as used in this study is McNary Dam where PIT-tagged fry when adipose clips were not possible. Therefore, fish adults are detected at the McNary Oregon Shore Lad- origin was not considered in estimation of survival and der (MC1) or McNary Washington Shore Ladder (MC2). travel times. Continuing into the upper Columbia River from McNary Dam, the next upstream site used in this study is Wells Detection sites for juveniles and adults Dam where adults are detected at the Adult Ladders Juvenile survival, detection probabilities, and travel time (WEA). After passing Wells Dam, Okanagan Sockeye estimates were based on detections of Okanagan Sock- travel up the Okanagan River where the first adult PIT- eye from release sites to McNary Dam (Table 1). McNary tag detection site used in this evaluation is at Okanagan Dam was used as the first point of detection for travel Channel at Vertical Drop Structure-3 (OKC), installed time and survival estimates because it is most capable of in 2009 (Fig. 1). The last upstream detection point in the providing reasonably precise estimates of survival over Okanagan River Basin, the Penticton Channel PIT array the greatest distance of the juvenile freshwater migration (OKP), was installed and operational in 2018. corridor. During their downstream migration, smolts pass a maximum of four diversion dams in the Okanagan Juvenile survival, detection probabilities, and travel time River before reaching the Columbia River, where they analyses then pass five hydroelectric projects (Wells [river kilom - Survival, detection probabilities, and travel time of eter, rkm 830], Rocky Reach [rkm 762], Rock Island [rkm Okanagan Sockeye smolts were estimated for each PIT- 730], Wanapum [rkm 669], and Priest Rapids [rkm 639]) tagged release group to McNary Dam (2012–2019; before reaching McNary Dam (rkm 470). After McNary Table  1). Estimates of juvenile survival and travel time Dam, juveniles pass through John Day (rkm 347), The were generated from the Columbia River Data Acqui- Dalles (rkm 308), and Bonneville (rkm 235) dams before sition in Real Time (DART) query program [13]. The reaching the Pacific Ocean (rkm 0). The farthest down - DART query retrieves PIT-tag data files of individual- stream detection site was in the Columbia River estuary, based capture history from PTAGIS (for tag files used where the National Marine Fisheries Service operated a see Additional file  1: Appendix S1) and uses these files paired-trawl detection system [44]. to generate survival estimates, detection probabilities, Survival and travel time of adult Okanagan Sockeye and travel times for each release site in each year. Sur- migrating through the Columbia River and Okanagan vival estimates, detection probabilities, and associated River back to their spawning grounds was based on standard error for each release group in each year were M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 5 of 16 estimated in DART using the Cormack–Jolly–Seber Skaha Lake (SKAHAL) in 2019. Because these omissions (CJS) model [14, 40, 60]. Assumptions for the CJS model appeared random across years for each release site, we include (1) individuals marked for the study are a repre- interpolated average length for these releases groups by sentative sample from the population of interest; (2) sur- averaging values across years where length data were pre- vival and capture probabilities are not affected by tagging sent from the same site (SKA + SKATAL: n = 4, standard or detection; (3) all detections are instantaneous; (4) the deviation 3.21  mm; SKAHAL: n = 2, standard deviation fate of each tagged individual is independent of the fate of 5.21 mm.) Exploratory analyses were conducted prior to all others; (5) all tagged individuals alive at the beginning model fitting to ensure distributional assumptions were of a reach have the same probability of surviving until the met and to avoid issues in multi-collinearity [72]. All end of that reach; (6) all tagged individuals alive at the modeling and validation were conducted in R (v. 3.6.1, R beginning of a detection site have the same probability Core Team 2019). of being detected at that site; (7) each individual detected Travel-time estimates were based on the release date at a particular detection site has the same probability for each PIT-tagged fish and the first detection date at of being removed, and (8) the probability of removal is McNary Dam. Harmonic mean travel time was used to independent of the survival process. Data files contain - describe each group’s (by release site and year) rate of ing records from all Okanagan release sites from PTAGIS travel from release site to McNary Dam. This statistical were examined for erroneous records, inconsistencies, summary of central tendency is more robust to the pres- and anomalies. Data files of releases of juvenile Okanagan ence of outliers (i.e., a very fast or slow fish) than arith - Sockeye in Osoyoos Lake (OSOYOL, OSOYOHA, and metic mean and is computed following Eq. (2): OSOYBR) and Skaha Lake and Skaha Dam tailrace (SKA, t = SKATAL, and SKAHAL) since 2012 were analyzed. Sur- 1 (2) i=1 t vival estimates were not adjusted for juveniles that did i not emigrate, tag failure, tag loss, or other factors, which where t is the computed harmonic mean travel time; n could result in fish surviving but not being detected at a is the observed number of unique fish detected at the downstream location. Due to these factors, actual sur- detection sites; t is the observed reach travel time for vival may be higher than the reported estimates. each fish i through n. Survival estimates across all years and sites were used in a regression-model framework to better understand the factors associated with juvenile survival. Multiple Smolt to adult survival rates and proportion of ocean‑age weighted logistic regression was used to explore how adults returning variation in juvenile survival is explained by outmigration Smolt-to-adult survival rates (SARs) and associated year, release site, release date, and smolt size (i.e., average standard errors were calculated as the number of PIT- length). The modeling framework followed Eq. (1): tagged smolts released from each lake of origin (Skaha Lake or Osoyoos Lake) in each year (2012–2016) that logit(p ) ∼ β + β x + ε , ε ∼ N 0, σ survived to return as adults to Bonneville Dam as ocean i 0 k k i i (1) age-1, age-2, or age-3 adults in 2013–2019. This approach k=1 assumes that smolts emigrate to the ocean the year they where p is the probability of juvenile survival observed are tagged and is supported by the negligible proportion across all individuals from group i (i = 1, …, 18); β is the of smolts that hold over an additional year following tag- intercept; β is the weighted slope term for each covariate ging [23]. PIT-tagged fish released in years 2017–2019 x ; ε is the normally distributed random error associated were excluded from the analysis due to incomplete age- with group i. structure data sets (i.e., return data only includes through Each observation was weighted using the number of 2019; therefore, no age-2 and age-3 data are available individuals within the release group (i.e.,  w = n ) , and i i for 2018 and 2017, respectively) and small sample sizes these weights were used to adjust the covariance matrix in those years. Data used in the analysis were queried to give higher weights to those site and year combinations using the PTAGIS interrogation summary tool. A query where the number of smolts was highest. Forward selec- reporting all instances of Okanagan Sockeye detected at tion based on AICc [8] was used to find the most parsi - the Bonneville adult ladders (BO1, BO2, BO3, and BO4) monious model amongst a suite of candidate models, from each release site and each year was generated. Bon- including outmigration year, release site, release date, and neville Dam provides the first points of detection for all smolt length. Two out of 18 release groups were initially adults as they begin their upriver migration to spawn- lacking length data: individuals released from Skaha Dam ing grounds and has nearly 100% detection efficiency (SKA + SKATAL) in 2015 and individuals released from when pooling data from the four PIT-tag interrogation Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 6 of 16 sites [63]. Duplicate tag numbers were deleted from each certainly resident O. nerka (i.e., kokanee). Included in data set to eliminate duplicate detections within each these 33 samples were eight fish detected only at OKP detection site. All data from the PTAGIS interrogation in 2019; these fish were not included in calculations of summaries were examined for erroneous records, incon- survival estimates. Survival estimates of adults return- sistencies, and anomalies. The equation used to calculate ing to OKC were, therefore, only possible in 2018 due SAR for each basin release group for each year across all to lack of data or exclusion of individuals that were not ages of adults returning (SAR ) is provided in the follow- confirmed to emigrate to the ocean. An additional 180 ij ing equation: individuals were eliminated from the analysis because they were juveniles detected in adult fishways dur - ij ing the year of release as smolts. A total of 395 adult SAR = ij (3) ij Okanagan Sockeye were ultimately used in the mark– recapture analysis to estimate adult survival and detec- where A is the total number of adults originating from ij tion probabilities upstream through the Columbia either Skaha or Osoyoos lakes and detected at Bonneville River to the Okanagan River Basin. Dam (i) in each year  (j); R is the total number of juve- ij Our analysis relied on survival rates derived in a mark– niles released from each lake (i) in each year (j). recapture framework to understand important factors Data were further analyzed to determine length of driving juvenile survival. Parameter estimation proceeds time of ocean residence to calculate proportion of ocean by first estimating detection probabilities, and then using age-1, age-2, and age-3 Okanagan Sockeye that return to these estimates to provide an estimate of survival proba- Bonneville Dam. Ocean age was determined by subtract- bility based of apparent detection. Therefore, our survival ing year of release from year of return. A Chi-squared probabilities were sensitive to differences in detection test was conducted to determine whether proportions efficiency across years and sites and were thus likely differed by return year and tagging basin. affected by factors that can influence detection efficiency (e.g., flow). Although a more robust approach would include covariates within a mark–recapture framework Adult upstream survival estimates, detection probabilities, to estimate their contribution to detection and sur- and travel time vival probability, we took this opportunistic approach Survival estimates and detection probabilities for adult to understand how various geographic, morphological, Okanagan Sockeye migrating upstream through the and phenological factors impacted a decent estimate of Columbia to the Okanagan River Basin were calculated survival. using a mark–recapture model in MARK [67]. Adults Adult return data through the Columbia and Okanagan pass through several dams with detection arrays on the rivers were examined to determine whether median day way back to their spawning grounds. In this analysis, of arrival at each dam and travel time from Bonneville adult survival rates and detection probabilities were to McNary, McNary to Wells, and Wells to OKC varied calculated using detections from Bonneville Dam to across years (2013–2019); basin of origin (Skaha Lake and McNary Dam to Wells Dam to OKC to OKP. Survival Osoyoos Lake) was also examined. Data were queried rates were calculated by migration reach and year that from PTAGIS using an interrogation summary. Adult adults returned, so results did not examine effect of and juvenile data were returned from the interrogation juvenile release site, year-class strength of juveniles, summary; all juvenile data were eliminated from analysis. or age-specific survival (juvenile or adult). In addition, Travel time was standardized to distance by calculating survival rates did not account for differences in detec - kilometer per day traveled for each PIT-tagged fish and tion probability estimates that manifested across years averaging across each reach: Bonneville to McNary Dam at various sites. OKP was installed in 2018, so data were (236  km), McNary to Wells Dam (360  km), and Wells only available for 2018 and 2019 for this site. Because Dam to OKC (177 km). Travel time was calculated from survival estimates rely on detecting an individual at date of last detection at downstream detection site to the site beyond the site of interest, survival estimates date of last detection at upstream detection site. To test to OKC were only possible in 2018 and 2019. Data whether arrival date at each dam detection site (Bonnev- obtained from PTAGIS were examined for errone- ille, McNary, Wells, OKC) differed by year and by lake of ous records, inconsistencies, and anomalies. All adult origin, a separate Wilcoxon test was conducted for year records where individuals were detected only at Wells and lake. Median migration speed (km/day) between Dam, OKC, and OKP (and not at Bonneville or McNary each detection site was calculated for all returning adults. dams) were eliminated from analysis (N = 33) because A Wilcoxon Rank Sum test was conducted to examine Bonneville and McNary dams collectively have near- whether all returning adults traveled at different speeds perfect detection rates and these individuals are almost M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 7 of 16 Table 2 Regression coefficients, standard errors, and associated Results P values for the full model estimating juvenile survival against Juvenile survival and travel time release year, release site, average length, and release date Following model selection, the model containing year, release site, and average length was the most Predictor Coefficient Standard error p parsimonious in explaining juvenile survival from Intercept − 4.85 0.48 < 0.01 release site to McNary Dam, while release date was 2013 − 0.88 0.10 < 0.01 not found to be a significant predictor of survival. 2014 − 1.38 0.10 < 0.01 This model was selected from the next best candidate 2015 − 1.15 0.10 < 0.01 model with ΔAICc = 21.72, and was found to signifi - 2016 − 0.54 0.12 < 0.01 cantly reduce deviance compared to the null model 2017 − 0.87 0.11 < 0.01 ( χ = 4, 698.6, p < 0.01) . Regression coefficients were 2018 − 0.46 0.11 < 0.01 found to be largely significant at the α = 0.01 level 2019 − 1.14 0.11 < 0.01 (Table  2). Coefficients to the log odds suggested vari - SKA + SKATAL 0.01 0.09 0.90 able juvenile survival success across years and sites OSOYBR 1.06 0.06 < 0.01 (Fig. 2). Smolts emigrating in 2014 and 2015 performed OSOYHA 1.19 0.07 < 0.01 poorer than those emigrating in 2017 and 2018. In addi- OSOYOL 0.89 0.05 < 0.01 tion, juveniles released from the OSOYBR site fared far Average Length 0.02 < 0.01 < 0.01 better than those from other sites—individuals released Release Date 0.03 < 0.01 < 0.01 from OSOYBR had odds of about 4.77-to-1 of surviving Release date was omitted in the final model based on AIC. Coefficients are given (95% CI [4.37, 5.21]). Juveniles tagged in Osoyoos Lake as raw values in log-odds space. P values represent the significance test result had higher survival from release site to McNary Dam in using the corresponding Wald’s statistic each year compared to fish tagged in Skaha Lake except for 2018 (Table  3; Fig.  3). Greater lengths were associ- between each detection site and to determine whether ated with greater outmigration success, with a 25  mm fish originating from Skaha or Osoyoos lakes exhibited increase in length corresponding with a doubling of the different migration speeds between detection sites. Dif - odds of survival (Fig.  2). Except in 2013, when travel ferences were considered significant at p = 0.05. time was equal, juvenile Okanagan Sockeye released from Skaha Lake traveled faster to McNary Dam than those released from Osoyoos Lake sites in all years, despite traveling a farther distance (migration distance Fig. 2 Regression model coefficients for the selected final model summarizing relationships of observed juvenile survival to predictors. All results are transformed to the logarithmic odds scale. Lines indicate the 95% confidence interval for the regression coefficient, as calculated through the transformed Wald’s statistic. The intercept of the model represents estimated juvenile survival for individuals released from Skaha Lake in 2012. Coefficients for year and release site are offsets to this base survival rate. The vertical line at 1 represents the inflection point between a decrease (less than 1) or an increase (greater than 1) to the default odds of return Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 8 of 16 Table 3 Survival estimates (and standard error) of groups of Survival to maturity PIT-tagged juvenile Okanagan Sockeye from release site in the Smolt-to-adult return rates (SARs) to Bonneville Dam Okanagan River Basin to McNary Dam by release year (2012– ranged from 0.4 to 6.1% across all release years for fish 2019) originating from the Okanagan River Basin. SARs were higher for fish tagged in Skaha Lake than for those tagged Year Release site (by distance) in Osoyoos Lake for all years (Table  5). SARs varied by OSOYOL OSOYHA OSOYBR SKA + SKATAL SKAHAL year and were highest in release-year 2013 (Skaha = 6.1%, 2012 NA NA NA 0.64 (0.23) NA Osoyoos = 3.1%) and lowest in 2015 (Skaha = 0.6%, 2013 NA NA 0.50 (0.12) 0.45 (0.16) NA Osoyoos = 0.4%). Proportion of ocean age-1, age-2, and 2014 NA NA 0.44 (0.05) 0.25 (0.06) NA age-3 fish returning varied by year (X = 16.1, p < 0.01) 2015 NA NA 0.54 (0.22) 0.29 (0.05) NA and by tagging basin (X = 32.3, p < 0.01; Fig.  4). Overall, 2016 0.51 (0.09) NA 0.51 (0.10) 0.40 (0.06) 0.39 (0.08) the mean age structure of adult Okanagan Sockeye at 2017 0.59 (0.07) NA 0.97 (0.29) NA NA Bonneville Dam included 27% ocean age-1, 69% ocean 2018 0.35 (0.08) 0.61 (0.09) NA NA 0.53 (0.07) age-2, and 3% ocean age-3. Returning adults were pre- 2019 0.54 (0.09) NA NA NA 0.28 (0.07) dominantly ocean age-2 for both Skaha- and Osoyoos- origin fish (Fig.  4). There was a higher percentage of OSOYBR Osoyoos Lake Narrows Highway 3 Bridge, OSOYHA Osoyoos Lake at Haynes Point Campground, OSOYOL Osoyoos Lake, SKA + SKATAL Tailrace of ocean age-1 fish returning from Skaha-origin smolts Skaha Dam, SKAHAL Skaha Lake (34%) than from Osoyoos-origin smolts (18%) across all years. Of all years, 2014 had the lowest percentage of ocean age-1 fish returning from either Skaha-origin for Skaha Lake sites is approximately 45 km longer than smolts (11%) or Osoyoos-origin smolts when no ocean Osoyoos sites). Travel time in 2015 was slower than age-1 fish returned. Numbers of ocean age-1 fish from in other years for smolts originating from both lakes Skaha Lake remained low in 2015 (6%), by contrast with (Table 4). age-1 Osoyoos-origin fish that had the highest number of all years (43%). Fig. 3 Relationships between juvenile survival and predictors. Boxplots are shown for categorical variables (release year and site) and scatter plots for continuous variables (average length and release date) M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 9 of 16 Adult migration Table 4 Mean harmonic travel time in days (and standard error) of groups of PIT-tagged juveniles from release site in the Adult survival of Okanagan Sockeye was estimated from Okanagan River Basin to McNary Dam by release year (2012– Bonneville Dam to McNary Dam and from McNary Dam 2019) to Wells Dam for all adult return years (2013–2019). Due to lack of detections at lower sites of fish also detected at Year Release site (by distance) OKP, sample sizes were only sufficient to calculate sur - OSOYOL OSOYHA OSOYBR SKA + SKATAL SKAHAL vival from Wells Dam to OKC in 2018. Detection prob- 2012 NA NA NA 11.4 (0.3) NA abilities ranged from 92 to 100% at Bonneville, McNary, 2013 NA NA 22.8 (0.6) 22.7 (0.6) NA and Wells Dams (Table  6). In most years, as expected, 2014 NA NA 20.2 (0.3) 13.3 (0.2) NA adults traveling from Bonneville to McNary Dam had 2015 NA NA 28.1 (1.0) 20.9 (0.4) NA higher survival than those traveling from McNary to 2016 26.6 (0.8) NA 22.1 (0.7) 17.1 (0.2) 15.9 (0.4) Wells Dam, except in 2015 when McNary to Wells Dam 2017 14.7 (0.2) NA 14.8 (0.4) NA NA had higher survival and in 2019 when survival was equal between these two reaches. Survival in 2015 was low- 2018 24.7 (0.5) 21.2 (0.2) NA NA 13.2 (0.2) est for both reaches. In 2018, Wells Dam to OKC had 2019 25.8 (0.4) NA NA NA 24.2 (0.6) lower survival than the other two downstream reaches OSOYBR Osoyoos Lake Narrows Highway 3 Bridge, OSOYHA Osoyoos Lake at Haynes Point Campground, OSOYOL Osoyoos Lake, SKA + SKATAL Tailrace of (Table 6). Skaha Dam, SKAHAL Skaha Lake Arrival dates of adult Okanagan Sockeye to Bonneville Dam, McNary Dam, and Wells Dam were approximately normal (Fig.  5). At these three lower dams, no differ - Table 5 Smolt-to-adult return rates (SARs) and standard error ences were observed between mean arrival date of fish (SE) in percentages, number of juvenile Okanagan Sockeye originating from Osoyoos- and Skaha-origin tag groups released, and number of adult Okanagan Sockeye detected at (Z < -0.37, p > 0.25). However, arrival date of adults to Bonneville Dam (BON) for each release year (2012–2016) and OKC exhibited a bimodal distribution with arrival date juvenile release site in the Okanagan River Basin peaks occurring on July 14 and October 2. Importantly, Year Release site # Juveniles # Total adults SAR SE differences were observed between fish from Skaha ver - released detected at BON sus Osoyoos tag groups at OKC detection site (Z = 4.84, 2012 Skaha 534 26 4.9% 0.93% p < 0.01). Cumulative probability indicates that a por- 2013 Osoyoos 2783 85 3.1% 0.33% tion of Skaha-origin fish arrives at OKC in July, followed 2013 Skaha 1203 73 6.1% 0.69% by few fish arriving in August and another big pulse in 2014 Osoyoos 3706 35 0.9% 0.16% September to early October (Fig.  6). Whereas Osoyoos- 2014 Skaha 1348 28 2.1% 0.39% origin fish don’t start arriving at OKC en masse until 2015 Osoyoos 1741 7 0.4% 0.15% mid-September, even though they arrive at similar times 2015 Skaha 5435 35 0.6% 0.11% to Wells Dam as Skaha-origin fish. Nonparametric pair - 2016 Osoyoos 4798 24 0.5% 0.10% wise comparisons using the Wilcoxon method indicated 2016 Skaha 5439 55 1.0% 0.14% that arrival timing to Bonneville, McNary, and Wells dams was not statistically different among years (p > 0.10) with exception of 2015 at McNary Dam and 2013 and 2014 at all three detection sites (p ≤ 0.02). At OKC site, differences were observed among some years; however, representation by tagged Okanagan Sockeye from both lakes of origin was not evenly distributed in those years where differences were observed. Okanagan Sockeye traveled at different migration speeds between detection sites on the upstream migra- tion from Bonneville Dam (X = 406.0, p < 0.01). Median migration speed (and 10% to 90% range) decreased as fish moved farther up the Columbia–Okanagan system from Bonneville Dam to McNary Dam, McNary Dam to Wells Dam, and Wells Dam to OKC at 52.0  km/day Fig. 4 Proportion of ocean age-1, age-2, and age-3 Okanagan (39.0 to 63.0  km/day), 36.4  km/day (27.0 to 45.4  km/ Sockeye adults returning to Bonneville Dam by release year (2012– day), and 2.6  km/day (1.8 to 36.4  km/day), respec- 2016) and juvenile tagging basin (Osoyoos Lake and Skaha Lake) in the Okanagan River Basin tively. Because of the large variation in migration speed, Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 10 of 16 Table 6 Survival estimates with standard error (SE) and detection probabilities with standard error for adult Okanagan Sockeye upstream migration from Bonneville to McNary Dam, from McNary to Wells Dam, and from Wells Dam to the Okanagan detection site (OKC) for 2013–2019 Year Survival estimates (SE) Detection probabilities (SE) Bonneville to McNary McNary to wells Wells to OKC Bonneville McNary Wells 2013 1.00 (0.00) 0.92 (0.08) NA 1.00 (0.00) 1.00 (0.01) NA 2014 0.95 (0.03) 0.79 (0.05) NA 0.96 (0.02) 1.00 (0.01) NA 2015 0.60 (0.59) 0.69 (0.06) NA 0.95 (0.03) 1.00 (0.01) NA 2016 0.88 (0.04) 0.85 (0.05) NA 0.98 (0.02) 1.00 (0.00) NA 2017 0.95 (0.03) 0.80 (0.05) NA 0.98 (0.02) 1.00 (0.01) NA 2018 0.85 (0.05) 0.83 (0.05) 0.64 (0.09) 0.95 (0.04) 1.00 (0.00) 0.92 (0.08) 2019 0.95 (0.05) 0.95 (0.05) NA 1.00 (0.01) 1.00 (0.01) NA Survival from Wells to OKC was unavailable prior to installation of the upstream array (OKP) in 2017 and otherwise unestimable due to sample size Fig. 5 Frequency of arrival dates of adult Okanagan Sockeye to Bonneville Dam (A), McNary Dam (B), Wells Dam (C), and OKC PIT array (D) across all years and tagging basins specifically between Wells Dam and OKC, data were slower and with less variation than Skaha-origin fish further compared to examine migration speed among with median migration speed of 2.0  km/day (1.6 to reaches based on which lake of origin (Osoyoos or 3.6 km/day); whereas Skaha-origin fish overall traveled Skaha Lake) fish were tagged. No difference in migra - slightly faster with much larger variation with median tion speed between Osoyoos- and Skaha-origin adults migration speed of 4.7  km/day (2.0 to 40.8  km/day; was observed in the Bonneville to McNary Dam reach Fig. 7). (Osoyoos N = 130, Skaha N = 182; Z = −  0.84, p = 0.40) or in the McNary to Wells Dam reach (Osoyoos N = 99, Discussion Skaha N = 159; Z = −  0.13, p = 0.89). However, there Results from PIT-tagging efforts in the Okanagan River were significant differences in migration speed between Basin over the 8-year study period provide key insights Osoyoos- and Skaha-origin adults traveling from Wells into survival and behavior of the largest population of Dam to OKC (Osoyoos N = 31, Skaha N = 64; Z = -5.38, Sockeye Salmon in the Columbia River Basin. Estimates p < 0.01; Fig .  7). Osoyoos-origin fish traveled much of smolt survival during the downstream migration, M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 11 of 16 tshawytscha have been estimated at 21.5  km/day in the mainstem Columbia River [26]. Travel speed of yearling Chinook Salmon in natal tributaries has been measured at less than 6 km/day [51]. In comparison, travel speed of Okanagan Sockeye smolts through 130 km of Okanagan River and 388  km of Columbia River is generally above 25  km/day. Smolts released in Skaha Lake migrate even faster than those released in Osoyoos Lake, averaging 34 km/day and ranging up to 50 km/day. These observa - tions are comparable or faster than migration speeds in natural and modified systems reported elsewhere [26, 49, 65] and represent an impressive speed greater than 3.0 body lengths per second over an extended distance [3]. Our results are consistent with longstanding characteri- zations of Sockeye Salmon: smolts leave nursery areas in Fig. 6 Cumulative probability of adult arrival date to OKC site from Wells Dam for PIT-tagged adults originating from Skaha Lake (SKA) high numbers over a short period and travel quickly to and Osoyoos lakes (OSO) for all years combined (2013–2019) the ocean [30]. Perhaps more noteworthy is the estimated survival of smolts from release sites in the Okanagan River Basin to SAR and age structure, and upstream conversion rates McNary Dam. Survival averages 49% across all release of adults provide critical information on the viability of groups in our analyses, despite traversing over 130  km the population. The results also highlight extreme vari - in the highly modified Okanagan River and another ation in survival over a short timeframe in response to 388  km of the Columbia River that includes five large- changes in ocean conditions and river temperatures dur- scale hydroelectric projects. This translates to an average ing the spawning migration. Each aspect of the analyses survival rate of 87.0% per 100 km (95% CI 84.5 to 89.5%). is discussed below, including how shifts in environmental In comparison, mean survival of Sockeye Salmon smolts conditions affect Okanagan Sockeye. We then compare from the Sawtooth Valley in Idaho has been estimated Okanagan Sockeye originating from Osoyoos and Skaha between 87.5 and 91.9% per 100  km to the uppermost lakes to underscore the importance of spatial structure, Snake River hydroelectric project with fish passage (from diversity, productivity, and numerical abundance of Alturas, Pettit, and Redfish lakes to Lower Granite Dam, Okanagan Sockeye. Finally, we recommend minor modi- distances of 774, 767, and 750 km, respectively) [28]. Sur- fications to monitoring and evaluation efforts that will vival of these same populations in the mainstem Snake improve the management of Okanagan Sockeye. and Columbia rivers are more comparable (21-year The downstream migration of Okanagan Sockeye average of 81.1% per 100  km) [68], though lower than smolts is relatively efficient compared to other juve - Okanagan Sockeye. Sockeye Salmon smolts emigrat- nile salmonids in the Columbia River Basin. For exam- ing from Cultus Lake in the unimpounded Fraser River ple, travel speeds of yearling Chinook Salmon O. Fig. 7 Box and whisker plot showing migration speed (km/day) of Osoyoos-origin and Skaha-origin adult Okanagan Sockeye traveling from Bonneville Dam to McNary Dam, from McNary Dam to Wells Dam, and from Wells Dam to OKC site combining all years (2013–2019) Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 12 of 16 have considerably lower survival rates of ≈ 50 to 70% extreme variation in ocean conditions [42] and freshwa- per 100  km over 4  years of monitoring, though differ - ter environments [17], we suspect that the diversity in ent technologies were used in the evaluation [65]. Other age structure of Okanagan Sockeye will be a key aspect salmonids in the region also have lower smolt survival in of maintaining population viability in the future [48, 59]. mainstem [68] or tributary habitats [6]. While these esti- Survival of adult Okanagan Sockeye during their mates are not directly comparable, they do suggest that upstream migration in freshwater was highly variable Okanagan Sockeye have relatively high rates of survival and could become the most significant threat to the during the juvenile outmigration. species if increasing temperature trends continue [38, Like most Sockeye Salmon populations, fish originat - 46]. Conversion rates of Okanagan Sockeye through the ing from the Okanagan River Basin had a highly vari- 595  km-reach between Bonneville and Wells dams (i.e., able SAR. Release groups from this study, spanning just the product of the two reach conversion rates presented 5  years, vary by more than an order of magnitude, from in Table  6) ranged from a high of 92% in 2013 to a low 0.4% (2015 Osoyoos Lake releases) to 6.1% (2013 Skaha of 41% in 2015. This indicates that pre-spawn mortal - Lake releases). These estimates (mean of 2.2%) and pre - ity in the mainstem Columbia River during the 2015 vious estimates of Okanagan Sockeye SARs ranging to drought exceeded 250,000 adult Okanagan Sockeye. For more than 12% [Kim Hyatt, Department of Fisheries and added context, this loss is roughly equivalent to 11 mil- Oceans Canada, personal communication] are high rela- lion smolts over the previous 3  years (based on an aver- tive to other populations in the Columbia River Basin. age SAR of 2.2%); 360 million eggs in the 2015 brood Snake River Sockeye Salmon also traverse over 1,000 km year; and more than double the number of adult Sock- and eight hydroelectric projects during the smolt migra- eye Salmon observed returning to the Columbia River tion, though SARs have been found to be much lower, at between 1995 and 1999 [11]. More concerning is the less than 1.0% from 2005 to 2009 [28]. Sockeye Salmon additional losses that occurred above the hydroelectric SARs from the Wenatchee River Basin—88  km and two system in 2015: only 10,400 of the 187,055 (5.6%) adult hydroelectric projects downstream of the Okanagan Okanagan Sockeye observed passing Wells Dam ulti- River confluence—ranged from 0.02 to 2.55%, though mately reached spawning grounds in the Okanagan River these are likely biased low based on recent PIT-tag analy- Basin [38]. While survival of adult Okanagan Sockeye ses and observations of adults obstructed at trapping during the upstream spawning migration is generally facilities below the spawning grounds [31, 52]. In con- favorable relative to other Columbia River Basin stocks trast, all Sockeye Salmon originating from the Columbia (e.g., Snake River Sockeye Salmon) [16] or other salmo- River Basin (the southern extent of the species) generally nids [5], it is clear that droughts and higher water tem- have lower SARs compared to stocks in British Colum- peratures pose a tremendous risk to the population. bia or Alaska [39, 43]. However, Sockeye Salmon in the The most influential condition of variation in spawn - Columbia River Basin are similar to northern counter- ing escapement of Okanagan Sockeye is the thermal parts in that ocean conditions combined with broader- barrier that develops each summer at the confluence of scale climate indices are by far the strongest predictors of the Okanagan and Columbia rivers near Brewster, Wash- SARs [64, 69]. The final detection point used in SAR cal - ington (rkm 859), where water temperatures often dif- culations varies among these populations so corrections fer by 10–15 °C in July and August. The barrier was first for distance traveled, harvest, or natural mortality would reported in the 1960s by researchers that were evaluating provide more comparable estimates. Sockeye Salmon passage at the newly constructed Rocky Age structure of Okanagan Sockeye based on this Reach Hydroelectric Project (rkm 761) in 1961. Once study was mostly consistent with other populations, with water temperatures in the Okanagan River exceed 21 °C, most (69%) fish returning to spawn after 2  years at sea Okanagan Sockeye will not enter the tributary, creating and a measurable proportion of jacks (27%) and a few delays that can last several weeks [45]. The consequence (< 4%) adults that spend 3 years at sea. Age structure was of this delay was highlighted in 2015, where conversion highly variable among years, consistent with previous rates from Bonneville Dam to spawning grounds in 2015 observations of up to six different age combinations of dropped to 4.3% as water temperatures in the Okanagan Okanagan River stocks [22]. In comparison, Wenatchee River remained at or above 21.9  °C for 67 consecutive River Sockeye Salmon have lower diversity in age struc- days [22]. The bimodal distribution of arrival timing for ture and are comprised entirely of age 4 and 5 adults [22, Skaha Lake Okanagan Sockeye (Fig. 6) could prove criti- 31]. Nutrient levels in nursery lakes [62] and conditions cal for population viability since a greater proportion of during early marine entry are known to influence age at adults arrive prior to the summer thermal barrier. maturity in Sockeye Salmon [55] and likely influenced Recreational, tribal, and commercial harvest oppor- the years included in our analyses. Given the recent tunities further complicate management of Okanagan M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 13 of 16 Sockeye during summer months. An average of 19% of Lake [17]. While the younger age structure may increase Okanagan Sockeye are harvested between Wells Dam ocean survival and spawning escapement, the higher (rkm 830) and spawning grounds each year, often when proportion of jacks originating from Skaha Lake may fish are congregated in large numbers at the thermal bar - translate to reduced reproductive success [10]. Run tim- rier [37]. Mortality of fish that evade capture compounds ing is potentially the most important difference in adult this issue: some fishery managers apply a 40–60% mortal - returns of Okanagan Sockeye observed in this study: ity rate for salmon released from commercial gillnets and arrival date to the Okanagan River Basin (OKC site) was 10% for salmon release from recreational fisheries, rates earlier and more protracted for Skaha Lake adults with a that increase with increasing water temperatures [57]. large pulse arriving in July and then again in late Septem- A more detailed analysis that accounts for harvest and ber compared to cohorts tagged in Osoyoos Lake, which water temperatures in conversion rates would provide arrive in mid–late September. Delays in the spawning important insights to managers as droughts in the West- migration of Okanagan Sockeye often translate to higher ern United States increase in frequency and duration rates of mortality, particularly during higher water tem- [30]. Additional research on adult holding and harvest in peratures. Adults with a tendency to migrate earlier may the Canadian portions of Osoyoos Lake would also pro- fare better in future years by avoiding thermal barriers, vide important insights. greater pre-spawn mortality, or inadvertent mortality Key differences in physical attributes and post-release related to commercial and recreational fisheries [33, 45]. performance were observed between Okanagan Sockeye Conversely, later run timing may translate to avoidance released in Skaha Lake versus those released in Osoyoos of higher water temperatures in spawning tributaries, a Lake. Smolts tagged in Skaha Lake were typically larger well-documented occurrence in the Fraser River [32]. In (~ 10–20 mm in median total length) and migrated faster, either case, maintaining diversity in life history strategies though had lower downstream smolt survival com- of Okanagan Sockeye will be critical to sustaining the pared to those tagged in Osoyoos Lake. Survival was population in a changing climate. still lower for smolts tagged in Skaha Lake after adjust- These analyses highlight variability in survival and ing for distance, so some biological or ecological mecha- behavior of Okanagan Sockeye that can inform key nism likely affects smolts derived from hatchery-origin aspects of population viability (i.e., VSP) [48]. First, sur- fry introduced into Skaha Lake. Greater rates of preda- vival during the smolt migration, in the ocean, and dur- tion due to size differences [66], behavioral patterns [58], ing the spawning migration has a profound influence or migration past a specific location of high predation on productivity and, therefore, abundance of Okanagan [25] present possible explanations. Vaseux Lake, located Sockeye. While ocean survival [69] and increased escape- downstream of Skaha Lake, is relatively shallow and hosts ments of Okanagan Sockeye [33, 36] have generally large populations of Northern Pikeminnow Ptychocheilus driven productivity and abundance of Columbia River oregonensis and Smallmouth Bass Micropterus dolomieu Sockeye Salmon, the 2015 drought clearly illustrates how that likely influence juvenile survival. Cessation of migra - above-average water temperatures during the spawn- tion or alternative life history patterns (i.e., non-migrants ing migration can negate the production of millions of are perceived as mortality in mark–recapture modeling) fish during the last months of life. Second, the increased once Skaha Lake smolts encounter the productive Osoy- spatial structure (through reintroduction into historical oos Lake is another possibility [37]. Finally, differences spawning and rearing locations) and diversity (includ- between hatchery- and wild-origin juvenile Okanagan ing migratory timing, smolt size, and adult age structure) Sockeye could explain differences in survival during the highlighted in our analyses demonstrate how manage- smolt emigration [35, 56]. ment efforts have directly improved biodiversity of Despite lower smolt survival in freshwater, fish origi - Okanagan Sockeye. nating from Skaha Lake had higher SARs than cohorts The importance of rebuilding diverse wild popula - tagged in Osoyoos Lake. This may be a result of size or tions, or functioning portfolios, has been demonstrated time at ocean entry [18], their differing age structure [54], in other Sockeye Salmon populations [57]. The findings size-selective harvest [41], or another mechanism that reported here and impressive recovery of Okanagan we are unable to account for in these analyses. Adults Sockeye through the early 2010s demonstrate their resil- originating from Skaha Lake had a younger age struc- ience to a highly modified environment, whereas the ture compared to cohorts tagged in Osoyoos Lake, with drought in 2015 reminded managers how years of suc- a significantly greater proportion of jacks (34% versus cess can be erased by changing river conditions. Routine 18% of adults, respectively). Research has demonstrated monitoring to inform management efforts that improve how selective pressures can increase and perpetuate the aspects of population viability will be critical in pre- prevalence of jacks, which could be a factor in Skaha serving Okanagan Sockeye over the next century. The Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 14 of 16 Availability of supporting data current monitoring framework and analyses presented The data sets used and/or analyzed during the current study are available here could be improved by (1) additional PIT-tagging and from the corresponding author on reasonable request. improved detection arrays; (2) more detailed analyses of metrics presented herein; and (3) consideration of addi- Declarations tional monitoring tools, such as life cycle modeling [45], Ethics approval and consent to participate that can be used to understand which components of the Data analyzed in this manuscript were obtained under an existing fish tagging population warrant the greatest level of protection. program. All tagging followed regionally accepted protocols that minimize stress and injury to PIT-tagged salmonids. The document is available at http:// www. bioma rk. com/ Docum ents% 20and% 20Set tings/ 67/ Site% 20Doc uments/ Abbreviations PDFs/ Fish% 20Tag ging% 20Met hods. pdf BO1: Bonneville Bradford Island Ladder; BO2: Bonneville Cascades Island Lad- der; BO3: Bonneville Washington Shore Ladder; BO4: Bonneville Washington Consent for publication Ladder slots; MC1: McNary Oregon Shore Ladder; MC2: McNary Washington Not applicable; data are not derived from any individual person. Shore Ladder; WEA: Wells Dam adult fishways; BON: Bonneville Dam; CJS: Cormack–Jolly–Seber; CRITFC: Columbia River Inter-Tribal Fish Commission; Competing interests DART : Data Acquisition in Real Time; DFO: Fisheries and Oceans Canada; OKC: The authors declare that they have no competing interests. Okanagan Channel instream detection site; OKP: Penticton Channel instream detection array; OSOYOL: Osoyoos Lake; OSOYHA: Osoyoos Lake at Haynes Author details Point Campground; OSOYBR: Osoyoos Lake Narrows Highway 3 Bridge; PIT: Four Peaks Environmental Science & Data Solutions, Wenatchee, USA. Passive integrated transponder; PTAGIS: Columbia Basin PIT Tag Information Fisheries and Oceans Canada, Pacific Biological Station, Nanaimo, Canada. 3 4 System (www. ptagis. org); SKA: Skaha; SKATAL: Skaha Tailrace; SKAHAL: Skaha Okanagan Nation Alliance, Westbank, Canada. Columbia River Inter-Tribal Lake; SAR: Smolt-to-adult return; SE: Standard error; VSP: Viable Salmonid Fish Commission, Portland, USA. Population. Received: 21 May 2021 Accepted: 21 August 2021 Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40317- 021- 00262-y. References 1. Benson RD. Ongoing actions, ongoing issues: trying again to free federal Additional file 1: Appendix S1. Tag Files from PTAGIS of juvenile Sockeye dams from the ESA. Envtl L Rep. 2019;49:11019. tagged in the Okanagan basin from 2012 to 2019 used in juvenile survival 2. Biomark. Fish tagging methods. http:// www. bioma rk. com/ Docum ents% and travel time queries and estimates. 20and% 20Set tings/ 67/ Site% 20Doc uments/ PDFs/ Fish% 20Tag ging% 20Met hods. pdf; 2012. Accessed Nov 2013. 3. Brett JR. Swimming performance of sockeye salmon (Oncorhynchus Acknowledgements nerka) in relation to fatigue time and temperature. J Fish Board Canada. Fisheries and Oceans Canada and Four Peaks Environmental Science & Data 1967;24(8):1731–41. Solutions provided funding for this manuscript. Okanagan Nation Alliance has 4. Buchanan RA, Skalski JR. A migratory life-cycle release-recapture led Sockeye Salmon recovery efforts in the upper Columbia River Basin and model for salmonid PIT-tag investigations. J Agric Biol Environ Stat. monitoring activities, including the management of PIT-tagging and detec- 2007;12(3):325–45. tion arrays. The Bonneville Power Administration and the public utility districts 5. Buchanan RA, Skalski JR. Using multistate mark-recapture methods to (PUDs) of Chelan, Douglas, and Grant counties provided biological support model adult salmonid migration in an industrialized river. Ecol Model. and programmatic funding for multiple aspects of the recovery efforts for 2010;221(4):582–9. Okanagan Sockeye Salmon. Chelan and Grant PUDs provided funding for PIT 6. Buchanan RA, Skalski JR, Mackey G, Snow C, Murdoch AR. Estimating infrastructure, tags, and the kł cp̓əlk̓ stim̓ Hatchery, which was instrumental cohort survival through tributaries for salmonid populations with variable in reintroducing Sockeye Salmon to historical spawning habitat in Skaha and ages at migration. North Am J Fish Manag. 2015;35(5):958–73. Okanagan lakes. Molly Steere, Carolyn Roper, and Nate Hough-Snee of Four 7. Burgner RL. Life history of sockeye salmon (Oncorhynchus nerka). Pacific Peaks Environmental Science & Data Solutions contributed to the manuscript salmon life histories. 1991, p. 3–117. through technical review, GIS support, and interpretation of results. Tom 8. Burnham KP, Anderson DR. A practical information-theoretic approach. Kahler of Douglas County PUD and two anonymous reviewers provided excel- Model selection and multimodel inference. 2002, p. 2. lent feedback to improve the content of this manuscript. We thank the late 9. Bussanich R, Hyatt K, Wright H. Proceedings of an expert’s workshop on Kim Hyatt (“Dr. Sockeye”) for his vision, expertise, leadership, and enthusiasm Columbia river and hydro-system impacts on migration success and for Sockeye Salmon. He is responsible for many efforts to recover Okanagan production variations of Anadromous Salmon 2017 Dec. 6–7, p. 28. Sockeye and with his monumental contributions he has left us with a priceless 10. Carlson SM, Rich HB Jr, Quinn TP. Reproductive life-span and sources of and enduring legacy that will require our continued dedication to maintain. mortality for alternative male life-history strategies in sockeye salmon, Oncorhynchus nerka. Can J Zool. 2004;82(12):1878–85. Authors’ contributions 11. CBR (Columbia Basin Research). Columbia Basin Research. www. cbr. washi JM led study design, analyses, data interpretation, and technical writing. KH ngton. edu. Accessed 18 Mar 2020. and JF supported technical writing and data interpretation. EK provided 12. Chapman DW. Salmon and steelhead abundance in the Columbia River statistical support and conducted regression analyses. SF and RB led program in the nineteenth century. Trans Am Fish Soc. 1986;115(5):662–70. development and data collection, also providing support on data interpre- 13. Columbia River DART (Columbia Basin Research, University of Washing- tation and technical writing. KS performed initial data queries and survival ton). DART PIT Tag Release and Observation Summary. 2020. http:// www. analyses. All authors read and approved the final manuscript. cbr. washi ngton. edu/ dart/ query/ pit_ sum_ tagfi les 14. Cormack RM. Estimates of survival from the sighting of marked animals. Funding Biometrika. 1964;51(3/4):429–38. This manuscript was funded by Fisheries and Oceans Canada with additional 15. Cosens B, Fremier A. Social-ecological resilience in the Columbia river support from Four Peaks Environmental Science & Data Solutions. basin: the role of law and governance. In: Practical panarchy for adaptive water governance. 2018, p. 47–64. M urauskas et al. Anim Biotelemetry (2021) 9:37 Page 15 of 16 16. Crozier EG, Burke BJ, Sandford BP, Axel GA, Sanderson B. Passage and 36. Hyatt KD, Stockwell MM, Rankin DP. Impact and adaptation responses survival of adult Snake River sockeye salmon within and upstream from of Okanagan River sockeye salmon (Oncorhynchus nerka) to climate the Federal Columbia River Power System. 2014. variation and change effects during freshwater migration: stock res- 17. DeFilippo LB, Schindler DE, Ohlberger J, Schaberg KL, Birch Foster M, Ruhl toration and fisheries management implications. Can Water Resour J. D, Punt AE. Recruitment variation disrupts the stability of alternative life 2003;28(4):689–713. histories in an exploited salmon population. Evol Appl. 2019;12(2):214–29. 37. Hyatt KD, McQueen DJ, Ogden AD. Have invasive mysids (Mysis diluviana) 18. Farley EV, Murphy JM, Adkison MD, Eisner LB, Helle JH, Moss JH, Nielsen J. altered the capacity of Osoyoos Lake, British Columbia to produce sock- Early marine growth in relation to marine-stage survival rates for Alaska eye salmon (Oncorhynchus nerka)? Open Fish Sci J. 2018;11(1):1–26. sockeye salmon (Oncorhynchus nerka). Fish Bull. 2007;105(1):121–30. 38. Hyatt KD, Stiff HW, Stockwell MM. Historic water temperature (1924– 19. Folks S, Bussanich R, Stevens A, Teather M. q̓awsitk (Okanagan River) 2018), river discharge (1929–2018), and adult sockeye salmon migration sockeye smolt out-of-Basin survival: PIT tagging 2016. Westbank, BC: (1937–2018) observations in the Columbia, Okanogan, and Okanagan Okanagan Nation Aquatic Enterprises Ltd.; 2016. p. 12. Rivers. Fisheries and Oceans Canada= Pêches et Océans; 2020. 20. Folks S, Teather M, Benson R. q̓awsitk (Okanagan River) sockeye smolt 39. Irvine JR, Akenhead SA. Understanding smolt survival trends in sockeye out-of-Basin survival: purse seining and PIT tagging BY 2015. Westbank, salmon. Mar Coast Fish. 2013;5(1):303–28. BC: Okanagan Nation Aquatic Enterprises Ltd.; 2017. p. 16. 40. Jolly GM. Explicit estimates from capture-recapture data with both death 21. Fryer JK. Columbia Basin sockeye salmon: causes of their past decline, fac- and immigration-stochastic model. Biometrika. 1965;52(1/2):225–47. tors contribution to their present low abundance, and the outlook for the 41. Kendall NW, Quinn TP. Quantifying and comparing size selectivity among future [dissertation]. Seattle ( WA): University of Washington; 1995. p. 274. Alaskan sockeye salmon fisheries. Ecol Appl. 2012;22(3):804–16. 22. Fryer JK, Kelsey D, Wright H, Folks S, Bussanich R, Hyatt KD, Stockwell MM. 42. Kintisch E. ‘The Blob’ invades Pacific, flummoxing climate experts. Science. Studies into factors limiting the abundance of Okanagan and Wenatchee 2015;348(6230):17–8. sockeye salmon in 2015. Columbia River Inter-Tribal Fish Commission 43. Koenings JP, Geiger HJ, Hasbrouck JJ. Smolt-to-aduit survival pat- Technical Report 17-06; 2017. p. 217. terns of sockeye salmon (Oncorhynchus nerka): effects of smolt length 23. Fryer JK, Wright H, Folks S, Alliance ON, Hyatt KD, Stockwell MM. Limiting and geographic latitude when entering the sea. Can J Fish Aquat Sci. factors of the abundance of Okanagan and Wenatchee Sockeye Salmon 1993;50(3):600–11. in 2011 Columbia River Inter-Tribal Fish Commission Technical Report for 44. Ledgerwood RD, Ryan BA, Dawley EM, Nunnallee EP, Ferguson JW. A BPA Project 2008-503-00; 2012. surface trawl to detect migrating juvenile salmonids tagged with passive 24. Fulton LA. Spawning areas and abundance of steelhead trout and coho, integrated transponder tags. North Am J Fish Manag. 2004;24(2):440–51. sockeye, and chum salmon in the Columbia River basin—past and 45. Lessard RB, Hilborn R, Chasco BE. Escapement goal analysis and stock present. Spec Sci Rep Fish. 1970;618:37. reconstruction of sockeye salmon populations (Oncorhynchus nerka) 25. Furey NB, Hinch SG, Lotto AG, Beauchamp DA. Extensive feeding on using life-history models. Can J Fish Aquat Sci. 2008;65(10):2269–78. sockeye salmon Oncorhynchus nerka smolts by bull trout Salvelinus con- 46. Major RL, Mighell JL. Influence of Rocky Reach Dam and the temperature fluentus during initial outmigration into a small, unregulated and inland of the Okanogan River on the upstream migration of sockeye salmon. British Columbia river. J Fish Biol. 2015;86(1):392–401. Fish Bull. 1967;1(66):131–47. 26. Giorgi AE, Hillman TW, Stevenson JR, Hays SG, Peven CM. Factors that 47. Matala AP, Narum SR, Saluskin BP, Johnston MV, Newell JE, Fast DE, Gal- influence the downstream migration rates of juvenile salmon and steel- breath PF. Early observations from monitoring a reintroduction program: head through the hydroelectric system in the mid-Columbia River basin. return of Sockeye salmon to a nursery lake of historical importance. Trans North Am J Fish Manag. 1997;17(2):268–82. Am Fish Soc. 2019;148(2):271–88. 27. Gresh T, Lichatowich J, Schoonmaker P. An estimation of historic and 48. McElhany P, Rucklelshaus MH, Ford MJ, Wainwright TC, Bjorkstedt EP. Via- current levels of salmon production in the Northeast Pacific ecosystem: ble salmonid populations and the recovery of evolutionarily significant evidence of a nutrient deficit in the freshwater systems of the Pacific units. U.S. Dept. Commer., NOAA Tech. Memo. NMFS-NWFSC-42. 2000. p. Northwest. Fisheries. 2000;25(1):15–21. 156. 28. Griswold RG, Kohler AE, Taki D. Survival of endangered Snake River sock- 49. Melnychuk MC, Welch DW, Walters CJ. Spatio-temporal migration pat- eye salmon smolts from three Idaho lakes: relationships with parr size at terns of Pacific salmon smolts in rivers and coastal marine waters. PLoS release, parr growth rate, smolt size, discharge, and travel time. North Am ONE. 2010;5(9):e12916. J Fish Manag. 2011;31(5):813–25. 50. Merritt WS, Alila Y, Barton M, Taylor B, Cohen S, Neilsen D. Hydrologic 29. Gustafson RG, Wainwright TC, Winans GA, Waknitz FW, Parker LT, Waples response to scenarios of climate change in sub watersheds of the RS. Status review of sockeye salmon from Washington and Oregon. U.S. Okanagan basin, British Columbia. J Hydrol. 2006;326(1–4):79–108. Dept. Commer., NOAA Tech. Memo. NMFS-NWFSC-33; 1997. p. 282. 51. Monzyk FR, Jonasson BC, Hoffnagle TL, Keniry PJ, Carmichael RW, Cleary 30. Hao Z, Hao F, Singh VP, Zhang X. Changes in the severity of compound PJ. Migration characteristics of hatchery and wild spring Chinook salmon drought and hot extremes over global land areas. Environ Res Lett. smolts from the Grande Ronde River basin, Oregon, to Lower Granite 2018;13(12):124022. Dam on the Snake River. Trans Am Fish Soc. 2009;138(5):1093–108. 31. Hillman T, Miller M, Willard C, Hopkins S, Johnson M, Hughes M, et al. 52. Murauskas JG, Fryer JK, Nordlund B, Miller JL. Trapping effects and fisher - Monitoring and evaluation of the Chelan and Grant County PUDs ies research: a case study of sockeye salmon in the Wenatchee River. USA hatchery programs. HCP Hatchery Committees and the PRCC Hatchery Fisheries. 2014;39(9):408–14. Sub-Committee, Wenatchee and Ephrata, WA. 2018. 53. Osborn RP. Climate change and the Columbia river treaty. Washington J 32. Hinch SG, Cooke SJ, Farrell AP, Miller KM, Lapointe M, Patterson DA. Dead Environ Law Policy. 2012;2(1):75–123. fish swimming: a review of research on the early migration and high 54. Peterman RM. Model of salmon age structure and its use in presea- premature mortality in adult Fraser River sockeye salmon Oncorhynchus son forecasting and studies of marine survival. Can J Fish Aquat Sci. nerka. J Fish Biol. 2012;81(2):576–99. 1982;39(11):1444–52. 33. Hyatt KD, Stockwell MM. 2019. Chasing an illusion? Successful restoration 55. Peterman RM. Patterns of interannual variation in age at maturity of of Okanagan River Sockeye Salmon (Oncorhynchus nerka) in a sea of sockeye salmon (Oncorhynchus nerka) in Alaska and British Columbia. Can uncertainty. In: CC Krueger, WW Taylor, S Youn (eds.) From catastrophe to J Fish Aquat Sci. 1985;42(10):1595–607. recovery: stories of fisheries management successes. Bethesda: American 56. Powell MS, Hardy RW, Flagg TA, Kline PA. Proximate composition and fatty Fisheries Society. 2019, p. 65–100. acid differences in hatchery-reared and wild Snake River sockeye salmon 34. Hyatt K, Stockwell M, Stiff H, Ferguson R. Salmon responses to hydrocli- overwintering in nursery lakes. North Am J Fish Manag. 2010;30(2):530–7. matological conditions in British Columbia in 2015. State of the physical, 57. Price MHH, Moore JW, Connors BM, Wilson KL, Reynolds JD. Portfolio biological and selected fishery resources of pacific Canadian Marine simplification arising from a century of change in salmon population Ecosystems in. 2016, vol. 44, p. 198–205. diversity and artificial production. J Appl Ecol. 2021;58(7):1477–86. 35. Hyatt KD, Mathias KL, McQueen DJ, Mercer B, Milligan P, Rankin DP. Evalu- 58. Scheuerell MD, Schindler DE. Diel vertical migration by juvenile sockeye ation of hatchery versus wild sockeye salmon fry growth and survival in salmon: empirical evidence for the antipredation window. Ecology. two British Columbia lakes. North Am J Fish Manag. 2005;25(3):745–62. 2003;84(7):1713–20. Murauskas et al. Anim Biotelemetry (2021) 9:37 Page 16 of 16 59. Schindler DE, Hilborn R, Chasco B, Boatright CP, Quinn TP, Rogers LA, 67. White GC, Burnham KP. Program MARK: survival rate estimation from both Webster MS. Population diversity and the portfolio effect in an exploited live and dead encounters. Bird Study. 1999;46(Supplement):S120–39. species. Nature. 2010;465(7298):609–12. 68. Widener DL, Faulkner JR, Smith SG, Marsh TM, Zabel RW. Survival esti- 60. Seber GA. A note on the multiple-recapture census. Biometrika. mates for the passage of spring-migrating juvenile salmonids through 1965;52(1/2):249–59. Snake and Columbia River dams and reservoirs, 2017. Fish Ecology Divi- 61. Simmatis B, Jeziorski A, Zemanek A, Selbie DT, Hyatt K, Fryer JK, Cumming sion, Northwest Fisheries Science Center 2018. BF, Smol JP. Long-term reconstruction of deep-water oxygen conditions 69. Williams JG, Smith SG, Fryer JK, Scheuerell MD, Muir WD, Flagg TA, Zabel in Osoyoos Lake (British Columbia, Canada): implications for Okanagan RW, Ferguson JW, Casillas E. Influence of ocean and freshwater conditions River sockeye salmon. Lake Reservoir Manage. 2018;34(4):392–400. on Columbia River sockeye salmon Oncorhynchus nerka adult return 62. Stockner JG, Macisaac EA. British Columbia lake enrichment programme: rates. Fish Oceanogr. 2014;23(3):210–24. two decades of habitat enhancement for sockeye salmon. Regul Rivers: 70. Yaniw N, Benson R. Okanagan River Sockeye Smolt Out-of-Basin Survival: Res Manage. 2021;12(4–5):547–61. Purse Seining and PIT Tagging BY 2016. Westbank, BC: Prepared by 63. Tenney J, Warf D, Tancreto N. Columbia basin PIT tag information system. Okanagan Nation Aquatic Enterprises Ltd.; 2018. p. 13. 2016 annual report. Report to the Bonneville Power Administration, 71. Zhang X, Li HY, Deng ZD, Leung LR, Skalski JR, Cooke SJ. On the Project. 1990, p. 080-0. variable effects of climate change on Pacific salmon. Ecol Model. 64. Tucker S, Thiess ME, Morris JF, Mackas D, Peterson WT, Candy JR, Beacham 2019;1(397):95–106. TD, Iwamoto EM, Teel DJ, Peterson M, Trudel M. Coastal distribution and 72. Zuur AF, Ieno EN, Elphick CS. A protocol for data exploration to avoid consequent factors influencing production of endangered Snake River common statistical problems. Methods Ecol Evol. 2010;1(1):3–14. sockeye salmon. Trans Am Fish Soc. 2015;144(1):107–23. 65. Welch DW, Melnychuk MC, Rechisky ER, Porter AD, Jacobs MC, Ladouceur Publisher’s Note A, McKinley RS, Jackson GD. Freshwater and marine migration and Springer Nature remains neutral with regard to jurisdictional claims in pub- survival of endangered Cultus Lake sockeye salmon (Oncorhynchus nerka) lished maps and institutional affiliations. smolts using POST, a large-scale acoustic telemetry array. Can J Fish Aquat Sci. 2009;66(5):736–50. 66. West CJ, Larkin PA. Evidence for size-selective mortality of juvenile sock- eye salmon (Oncorhynchus nerka) in Babine Lake, British Columbia. Can J Fish Aquat Sci. 1987;44(4):712–21. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Animal BiotelemetrySpringer Journals

Published: Sep 3, 2021

Keywords: Sockeye Salmon; PIT tags; Columbia River Basin; Survival

References