{"Bibliographic":{"Title":"Alternative barging strategies to improve survival of transported juvenile salmonids, 2006","Authors":"","Publication date":"2007","Publisher":""},"Administrative":{"Date created":"08-16-2023","Language":"English","Rights":"CC 0","Size":"0000068581"},"Pages":["SH 153\n.Un54\nA48\nAlternative barging\n2007\nstrategies to improve\nsurvival of transported\njuvenile salmonids,\n2006\nFish Ecology\nDivision\nNorthwest Fisheries\nScience Center\nby\nBrad A. Ryan, Melissa Carper, Douglas M. Marsh,\nNational Marine\nFisheries Service\nDiane Elliott, Tony Murray, LynnMarie Applegate,\nConnie McKibben, and Sacha Mosterd\nSeattle, Washington\nNovember 2007","Alternative Barging Strategies to Improve Survival of Transported\nJuvenile Salmonids, 2006\nReport of research by\nBrad A. Ryan, 1 Melissa Carper,\n1\nDouglas M. Marsh, 1 Diane Elliott, 2 Tony Murray, 2\n2\nConnie McKibben, 2 and Sacha Mosterd2\nLynnMarie Applegate,\nReport of research by\n1 Fish Ecology Division\nNorthwest Fisheries Science Center\nNational Marine Fisheries Service\n2725 Montlake Boulevard East\nSeattle, WA 98112\nand\nWestern Fisheries Research Center\n2 United States Geological Survey\n6506 NE 65th St\nSeattle, WA 98115\nto\nWalla Walla District\nU.S. Army Corps of Engineers\n201 North 3rd\nWalla Walla, WA 99362-1876\nContract W68SBV60307671 and W68SBV60418618\nNovember 2007","ii","EXECUTIVE SUMMARY\nDuring spring 2006, we conducted a study to test the hypothesis that releasing\ntransported juvenile Pacific salmonids Oncorhynchus. spp to the lower Columbia River\nestuary at river kilometer (rkm) 10 would produced higher smolt-to-adult return rates\n(SARs) than releasing them just below Bonneville Dam at rkm 225. We speculated that\nreleasing transported fish an additional 215 km downstream from the location presently\nused could decrease smolt mortality due to predation by piscivorous fish and birds.\nAdults returning over the next several years will provide data to test this hypothesis.\nIn addition to evaluating a release location for transported fish, we used new,\nnon-lethal techniques to collect fish pathogen data. We determined pathogen loads in\nstudy fish to evaluate whether pathogens in individual fish affect vulnerability to avian\npredators as well as SARs.\nOn six consecutive Sundays, starting in April 2006 and running through May,\nrun-of-the-river yearling Chinook salmon O. tshawytscha and steelhead O. mykiss were\ncollected and tagged with passive integrated transponder (PIT) tags at the Lower Granite\nDam juvenile fish facility. Following tagging, fish were transferred to raceways and held\nuntil the following day, when they were loaded on barges for transport. A total of 13,729\nhatchery and 2,435 wild yearling Chinook salmon were tagged and released downstream\nfrom Astoria at rkm 10, while 20,488 hatchery and 3,707 wild yearling Chinook salmon\nwere tagged and released at Skamania Landing (rkm 225). In total, we released 25,726\nhatchery and 3,4045 wild steelhead at rkm 10 and 36,210 hatchery and 5,612 wild\nsteelhead at rkm 225.\nDuring each tagging day, 300 non-lethal gill clip samples were collected for\npathogen analyses (Renibacterium salmoninarum and Nucleospora salmonis), for a total\nof 1,800 samples over the season. All Astoria releases were made after dark on an\noutgoing tide to reduce avian predation by Caspian terns Hydroprogne caspia and\ndouble-crested cormorants Phalacrocorax auritas from the nearby nesting colonies on\nEast Sand Island.\nAbandoned bird colonies were scanned to detect PIT tags from fish released from\nthis and other studies, and these data were used to estimate the number of fish from each\nrelease group preyed upon by piscivorous birds. These data also allowed us to determine\nwhether infection with R. salmoninarum, N. salmonis, or both pathogens was correlated\nwith predation vulnerability. There was no evidence from the 2006 study that infection\nof fish with one or both pathogens influenced rates of predation, but R. salmoninarum\niii","infection levels in the majority of tested fish were low. Current methodologies for\nassaying N. salmonis can only provide numbers of fish infected, not infection levels.\nBoth of these pathogens are associated with chronic, slowly progressing\ninfections that may not always cause outright mortality, but may make fish vulnerable to\nsecondary infections. Therefore, the pathogens may have more effect on smolt-to-adult\nreturns (SARs) than on short-term mortality, including mortality by avian predation.\nOverall, R. salmoninarum was detected by one or both PCR assays in 77% of wild\nChinook salmon, 72% of hatchery Chinook salmon, 73% of wild steelhead, and 70% of\nhatchery steelhead sampled. N. salmonis was detected by polymerase chain reaction\n(PCR) in 4% of wild Chinook salmon, 1% of hatchery Chinook salmon, 6% of wild\nsteelhead, and 23% of hatchery steelhead tested.\nWe will need to wait several years for adult returns to determine the efficacy of\nreleasing transported salmonids at rkm 10 instead of the traditional release site at\nrkm 225. However, we do know that the new release location affected vulnerability to\navian predators; mean avian predation rates were 3.0% for yearling Chinook salmon\nreleased from Skamania Landing at rkm 225, but only 0.4% for those released near\nAstoria at rkm 10. Avian predation rates were 13.8% for steelhead released at Skamania\nLanding, but only 1.7% for their cohort released at Astoria. These are minimum\nestimates of the impact of avian predation, as not all tags consumed by birds are\ndeposited or found on colonies. Our results show that releasing fish farther downstream,\nat night, and on an outgoing tide will reduce avian predation by up to seven-fold on\naverage. This finding is relevant for management actions related to recovery of juvenile\nsalmonids that pass the world's largest Caspian tern and double-crested cormorant\ncolonies during their downstream migration.\niv","CONTENTS\nEXECUTIVE SUMMARY\niii\nINTRODUCTION\n1\nMETHODS\n3\nFish Acquisition and Tagging\n3\nFish Releases\n4\nPathogen Sampling\n5\nBird Colony Sampling\n6\nPathogen Analyses\n6\nRESULTS\n9\nTagging\n9\nAvian Predation\n10\nDate-Specific Predation Results\n11\nPathogen Analyses\n12\nDISCUSSION\n19\nACKNOWLEDGEMENTS\n23\nREFERENCES\n25\nV","vi","INTRODUCTION\nAt collector dams on the Snake and Columbia Rivers, migrating salmonid smolts\nare guided away from turbine intakes and collected for transport by truck or barge to a\nrelease site below Bonneville Dam. The purpose of transporting fish is to avoid mortality\ncaused by dam passage, but the benefit provided by transportation has varied for different\nfish stocks and with the timing of transport within the migration season (Muir et al. 2006;\nWilliams et al. 2005).\nTypically, about 50% of Snake River migrant smolts survive downstream\nmigration to below Bonneville Dam (Williams et al. 2005), while about 98% of\ntransported smolts survive (Budy et al. 2002). Therefore, one would expect about twice\nas many transported adults as inriver migrant fish to return as adults. Nevertheless, on an\nannual basis, the ratio of transported to in-river migrant adult returns is usually lower\nthan expected. This indicates that higher mortality is experienced for transported smolts\nafter release than for inriver migrants that survived migration. The difference in survival\nbetween inriver migrants and transported fish has been termed differential delayed\nmortality or \"D.\" The purpose of this study is to determine if transporting juvenile fish\nfarther downstream will increase smolt-to-adult return (SARs) and reduce D of\ntransported fish.\nFish condition and health have been assessed prior to and after transport in\nprevious studies (Pascho and Elliott 1989; Elliott and Pascho 1991, 1992, 1993, 1994;\nElliott et al., 1997; Congleton et al. 2000, 2005; Kelsey et al. 2002; Schreck et al. 2005).\nAlthough stress and stressors have been examined in detail in these studies, and\nmodification to the collection and transportation system have been made to reduce stress\n(Williams and Matthews 1995), transportation has not provided the benefit expected,\nparticularly for wild Chinook salmon Oncorhynchus tshawytscha (Williams et al. 2005).\nThis research continues an ongoing effort by the U.S. Army Corps of Engineers\nAnadromous Fish Evaluation Program (AFEP) to evaluate modifications to the existing\ntransportation program to improve post-release survival of transported fish.\nStudies conducted with Coho salmon O. kisutch found that smolts transported to a\nrelease point near Tongue Point in the Columbia River returned at 1.6 times greater rate\nthan those released upriver (Solazzi et al. 1991). Similarly, Gunnerod et al. (1988) found\nthat Atlantic salmon Salmo salar released in salt water returned at a higher rate. Marsh\net al. (1996, 1998, 2000) compared the Skamania Landing release site with a release site\nat Tongue Point (rkm 29) in the Columbia River estuary, but too few adult steelhead\nO. mykiss returned from either release point for a meaningful evaluation.\nThe primary objective of the 2006 alternate barge release site study was to\ndetermine whether releasing barged fish farther downstream near Astoria at rkm 10","(approximately 10 km downstream from the Astoria Bridge) would improve the SAR rate\nof spring Chinook salmon and steelhead (Figure 1). The strategy was to minimize the\ntime spent moving into and through the estuary, while documenting fish condition to\nprovide insight into the vulnerability of smolts to predators. Our approach was to tag\ntransported smolts with passive integrated transponder (PIT) tags (Prentice et al. 1990),\ncollect samples for pathogen analysis, and release fish at the current barge release site\ndownstream from Bonneville Dam near Skamania Landing (rkm 224) and at Astoria.\nSince complete adult returns are needed to calculate SARs, this objective will require\nseveral years to complete.\nOur second objective was to determine Renibacterium salmoninarum prevalence\nand severity, along with the presence of Nucleospora salmonis, within each release\ngroup. The infection profiles of R. salmoninarum and N. salmonis reported here can then\nbe correlated with avian predation rates and eventually with SARs. Our third and final\nobjective was to compare avian predation rates between Skamania Landing and Astoria\nrelease groups.\nChief Joseph\nWells\nGrand Coulee\nRocky Reach\nLower Granite\nRock Island\nDam\nWanapum\nPriest Rapids\nSkamania\nLanding\nAstoria\nHells Canyon\nOxbow\nBrownlee\nFigure 1. Study area showing Lower Granite Dam, where fish were collected and PIT\ntagged, the Skamania Landing barge release site (rkm 224), and the Astoria\nbarge release site (rkm 10) during 2006.\n2","METHODS\nFish Acquisition and Tagging\nDuring spring 2006, we collected and PIT-tagged two groups of steelhead and\ntwo groups of yearling Chinook salmon smolts at Lower Granite Dam. Fish were tagged\nat the NOAA tagging facility on six consecutive Sundays from April to May. Tagging\nfollowed the protocols and standards outlined in the PIT Tag Marking Procedures Manual\n(CBFWA 1999) for mass marking using simple PIT-tag injectors (see Marsh et al. 2001\nfor description of tagging methods used at this facility). After each tagging session, fish\nwere transferred to the east bank transport raceways for 24-h recovery.\nThe following day, one group of each species was loaded on an 8000-series\ntransportation barge for release at rkm 224 (Skamania Landing), where transported fish\nare typically released. A second group of each species was loaded on a 2000-series barge\nand released at rkm 10 (Astoria release) in the lower estuary.\nWe attempted to tag sufficient numbers of both yearling Chinook salmon and\nsteelhead to test a ratio of 1.3 (i.e., an SAR at least 30% higher) for adults returning from\ntransported groups releases at Astoria (TA) vs. those released Skamania Landing (Ts).\nThis ratio (TA/Ts) was based on an expected SAR of 1.0% at Lower Granite Dam for the\nAstoria releases (Table 1). For both yearling Chinook salmon and steelhead, we tagged\nhatchery and wild fish in proportion to those entering the juvenile bypass facility. While\nthe expected ratio required us to tag 53,000 fish of each species, the actual number tagged\nvaried in accordance with numbers of fish arriving at the dam.\nTable 1. Required sample sizes based on expected SAR (a = 0.05, = 0.20) and TA/Ts.\nTA/Ts\nExpected Astoria\nFish PIT tagged at Lower Granite Dam (n)\nratio\nSAR\nAstoria (rkm 10)\nSkamania Landing (rkm 225)\n1.2\n1.00\n48,000\n57,000\n1.2\n0.75\n64,000\n76,000\n1.2\n0.50\n95,000\n114,000\n1.3\n1.00\n23,000\n30,000\n1.3\n0.75\n31,000\n40,000\n1.3\n0.50\n46,000\n58,000\n3","Lower Granite Dam will serve as the principal recovery site for adults. Data\nacquired from other areas will be considered ancillary. To analyze results, statistical tests\nwill be applied when adult returns for the study are complete. Confidence intervals for\nthe TA/Ts ratio will be calculated using the ratio (survival) estimate (Burnham et al. 1987)\nand its associated empirical variance. The study will produce an overall, statistically\nbound TA/Ts estimate for fish returning to Lower Granite Dam.\nFish Releases\nThe Skamania Landing release groups were transported and released with normal\ntransportation fish. We attempted to keep loading density and water volume replacement\ntimes as close as possible between the Skamania Landing (8000 series) and Astoria (2000\nseries) barges, and did not exceed loading density and replacement rates set by the U.S.\nArmy Corps of Engineers. However, due to the unpredictable nature of fish arrival and\ncollection at the dam, keeping loading densities equal proved to be difficult.\nThe barge used for Astoria releases was towed with a separate vessel, which\nmirrored the path of the Skamania Landing barge until after it passed Bonneville Dam\nand continued downstream to rkm 10. Astoria releases were timed to occur at night on an\nebb tide to minimize predation by Caspian terns Hydroprogne caspia and double-crested\ncormorants Phalacrocorax auritas from the nearby nesting colonies on East Sand Island\nin the Columbia River estuary (Table 2). Dissolved oxygen levels, water temperatures,\nand mortalities were monitored on the 2000 series barge using the same standard\nprocedures used on the 8000 series barge.\nTable 2. Release dates, times, and locations for PIT-tagged juvenile steelhead and\nyearling Chinook salmon smolts released at Skamania Landing and near Astoria\nduring 2006. High tides for the Astoria releases are noted.\nAstoria releases (rkm 10)\nSkamania releases (rkm 224)\nHigh tide at\nRelease date\nTime\nrkm 10\nRelease date\nTime\n26 Apr\n3:15\n00:18\n26 Apr\n0:35\n3 May\n21:15\n19:12\n2 May\n21:15\n11 May\n1:45\n00:09\n9 May\n19:25\n17 May\n21:15\n17:54\n16 May\n19:55\n26 May\n2:30\n00:18\n25 May\n19:05\n2 June\n21:00\n19:22\n1 June\n19:10\n4","Pathogen Sampling\nFish were analyzed for the presence of two salmonid pathogens known to occur in\nthe Snake and Columbia River basins: R. salmoninarum, the causative agent of bacterial\nkidney disease (BKD), and N. salmonis, an intranuclear microsporidian parasite that\nprimarily infects lymphoblast cells and can cause a chronic, severe lymphoblastosis and a\nleukemic-like condition. Gill filament samples for determining the presence and levels of\nR. salmoninarum and the presence of N. salmonis were collected from fish in every\nrelease group during tagging. The goal was to sample 75 fish each of wild and hatchery\nChinook salmon and of wild and hatchery steelhead on each tagging date, for a total of\n300 fish per replicate. The total number of fish sampled over the season was close to the\ngoal of 1,800, but proportions of fish by species and origin varied depending on their\navailability at the dam on each tagging date (Table 3).\nTable 3. Release numbers (mortalities removed) of PIT-tagged hatchery (H) and wild\n(W) steelhead and yearling Chinook salmon smolts that were gill-clipped,\ntransported, and released at Skamania Landing by release date during 2006.\nChinook salmon\nSteelhead\nRelease date\nH\nW\nH\nW\nTotal\n26 Apr\n74\n69\n75\n75\n293\n2 May\n75\n75\n75\n75\n300\n9 May\n75\n73\n75\n74\n297\n16 May\n74\n74\n75\n75\n298\n25 May\n73\n67\n75\n74\n289\n1 Jun\n8\n19\n141\n131\n299\nTotal\n379\n377\n516\n504\n1,776\nSample collection methodology followed the protocol for non-lethal gill filament\nsampling described by Schrock et al. (1994). Briefly, a 2- X 3-mm gill sample\n(approximately 10 mg) was removed from each fish using surgical scissors. Samples\nwere placed in individual pre-weighed and labeled tubes, frozen immediately on dry ice,\nand transported to the USGS Western Fisheries Research Center for analysis. The use of\npre-weighed tubes and transport of undiluted samples on dry ice instead of dilution in\nethanol before transport enabled accurate weighing of samples. The PIT-tag code\nassociated with each gill filament sample number was recorded.\n5","At the same PIT-tagging stations where fish were collected for pathogen analyses,\nwater samples were taken for quantification of R. salmoninarum in water at the juvenile\nfish facility. Water samples were taken four times during each tagging day: twice before\nthe recirculating water in the tagging system was changed, and twice after the water was\nchanged. The samples were preserved by addition of 0.01% thimerosal (final\nconcentration) to each 500-mL water sample.\nBird Colony Sampling\nUsing PIT tags allowed us to use avian predation data from the NOAA Fisheries\navian predation project (Ryan et al. 2007) to estimate predation rates of the fish released\nin this study. The avian predation project evaluates the impacts of predation by Caspian\nterns and double-crested cormorants on juvenile salmonids by detecting PIT tags on\npiscivorous water bird colonies in the Columbia River Basin (Ryan et al. 2001, 2003).\nComparing the rates of predation of PIT-tagged salmonids allowed us to determine\nwhether fish released at Skamania Landing were more susceptible to predation by\npiscivorous birds than fish released at Astoria. The data also allowed us to observe any\ndifferences in predation rate that may be due to R. salmoninarum or N. salmonis\ninfection. We used paired t-tests (P <0.05) to compare predation rates between release\nlocations.\nPathogen Analyses\nGill samples were weighed, processed and tested for R. salmoninarum by two\nPCR procedures: nested PCR (nPCR) and real-time quantitative PCR (qPCR). The\nnPCR was done according to the method of Chase and Pascho (1998). For the qPCR, the\nprocedure of Chase et al. (2006) was followed, except that a non-fluorescent quencher\nwas substituted for the fluorescent quencher dye (TAMRA) on the 3' end of the internal\nprobe RS1262. The use of this modified probe, MGBRS1262, was intended to increase\nthe sensitivity of the qPCR. Some previous work had indicated that the original qPCR\nmethod of Chase et al. (2006) had a lower sensitivity than the nPCR (Elliott and Pascho\n2004; McMichael et al. 2006), but only the qPCR can provide a measure of the infection\nlevels in fish. Thus, testing a single sample by both PCR techniques was desirable to\nprovide the most information.\nFor detection of N. salmonis in gill samples, the nested PCR method of Barlough\net al. (1995) was followed, with several modifications. A commercially available PCR\n6","buffer (Qiagen, Inc.1 was included in the master mix, and the use of gelatin was omitted.\nA 2-uL aliquot of the first round product was used in the nested round of amplification.\nThe cycling conditions for both PCR rounds were changed to the following: 94°C for\n5 min, followed by 35 cycles of 94°C for 30 sec, 65°C for 60 sec, and 72°C for 60 sec,\nwith a final extension of 72°C for 7 min. The current assay methodology can only\ndetermine the presence or absence of N. salmonis. A new methodology is being\ndeveloped that may allow quantitative assaying in the near future.\nFor enumeration of R. salmoninarum in water samples, a procedure modified\nfrom that of Elliott and McKibben (1997) was used. Water samples were shaken to mix\nthe contents, and large debris was allowed to settle for 5 min. Triplicate sub-samples\nwere prepared from each water sample. For each sub-sample, a 5-mL aliquot of the\nsample was combined with 3 mL of phosphate-buffered saline (PBS, 0.01 M phosphate,\npH 7.1) with 0.5% (by volume) Triton X-100 added (PBS-Triton). After vortex mixing,\neach sub-sample was triturated through a 22-gauge needle, and then filtered through a\nNuclepore 0.2-um pore diameter filter.\nAfter each filter was rinsed with 1-3 mL PBS Triton, 100 uL of a 1:40 dilution\n(by volume) of fluorescein isothiocyanate-labeled anti-R. salmoninarum polyclonal\nantiserum (Kirkegaard and Perry Laboratories) was pipetted onto each filter. The filters\nwere incubated in a humid chamber for 1 h at room temperature, then rinsed with 1-3 mL\nPBS-Triton, and counterstained with 1 mL Eriochrome Black T (Sigma; diluted 1:2000\nwt.:vol in PBS). Filters were air dried, and cover glasses were mounted with pH 9\nglycerol-DABCO mounting medium (Johnson et al. 1982). Each filter was examined by\nepifluorescence microscopy at 1000x magnification with a Zeiss Axiophot microscope.\nA total of 150 microscope fields were examined on each of three filters per water sample,\nand R. salmoninarum cells were counted.\nSeveral statistical methods were used for pathogen analyses among groups of\nsteelhead and Chinook salmon (Motulsky 1995; InStat 3, Graph Pad). Contingency\ntables were used to compare relative proportions of uninfected and infected fish for either\nR. salmoninarum, N. salmonis, or both pathogens. Fisher's exact test was used for\nanalysis of 2 x 2 tables, and a chi-square test was used to analyze larger contingency\ntables. For fish testing positive for R. salmoninarum by qPCR, R. salmoninarum level\ndata (both raw data and log-transformed data) were first tested for normality by the\nKolmogorov-Smirnov method. Because at least one data set in each comparison failed\n1 Use of trade names does not imply endorsement by the National Marine Fisheries Service, NOAA.\n7","the normality test (P <0.05) even after log transformation, the nonparametric\nMann-Whitney test was used for comparison of two groups, and the Kruskal-Wallis test\n(single factor analysis of variance by ranks) was used to compare R. salmoninarum levels\namong three or more groups. Dunn's multiple comparison test was applied when a\nsignificant result (P < 0.05) was observed using the Kruskal-Wallis test.\nFor each species, fork lengths of fish infected with one or both pathogens were\ncompared with fork lengths of fish in which neither pathogen was detected. Because\nlength data were not normally distributed, the Mann-Whitney test or Kruskal-Wallis test\nwas used for these comparisons as previously described. Similar procedures were used\nfor comparison of R. salmoninarum concentrations detected in water samples on different\ndays. Correlation of R. salmoninarum level with fork length were evaluated using the\nnonparametric Spearman rank correlation test, and this test was also used to evaluate\ncorrelations of R. salmoninarum level with concentration of the bacterium in water\nsamples.\n8","RESULTS\nTagging\nOn six consecutive Sundays from April through May, river-run yearling Chinook\nsalmon and steelhead were collected and tagged with PIT tags at the Lower Granite Dam\njuvenile fish facility (Table 4). A total of 13,729 hatchery and 2,435 wild yearling\nChinook salmon were tagged, loaded on a transport barge, and released at Astoria, while\n20,488 hatchery and 3,707 wild yearling Chinook salmon were tagged, transported, and\nreleased at Skamania Landing. In total, 25,726 hatchery and 3,445 wild steelhead were\nreleased at Astoria, and 36,210 hatchery and 5,612 wild steelhead were released at\nSkamania Landing. Additional fish were added to the holds of both barges in an attempt\nto equalize densities. Due to the unpredictable nature of fish arrival and collection at the\ndam, equalizing densities proved to be difficult (Table 5). However, final fish loading\ndensities were far below maximum capacities on all barges.\nTable 4. Release numbers of PIT-tagged wild and hatchery juvenile steelhead and\nyearling Chinook salmon by date at the Astoria and Skamania Landing release\nsites during 2006.\nSkamania Landing\nAstoria\nChinook salmon\nSteelhead\nChinook salmon\nSteelhead\nRelease\ndate\nHatchery\nWild\nHatchery\nWild\nHatchery\nWild\nHatchery\nWild\nTotal\n26 Apr\n2,199\n950\n5,657\n527\n2,837\n897\n6,695\n456\n20,218\n2-3 May\n5,769\n1,310\n4,917\n808\n3,556\n731\n3,242\n439\n20,772\n9 and 11 May\n5,569\n495\n5,364\n914\n3,310\n262\n3,899\n554\n20,367\n16-17 May\n5,456\n287\n6,504\n609\n3,309\n184\n4,295\n290\n20,934\n25-26 May\n1,449\n562\n9,704\n1,978\n702\n329\n5,504\n1,227\n21,455\n1-2 June\n46\n103\n4,064\n776\n15\n32\n2,091\n479\n7,606\nTotal\n20,488\n3,707\n36,210\n5,612\n13,729\n2,435\n25,726\n3,445\n111,352\n9","Table 5. Numbers of PIT-tagged fish, and the number of untagged fish added to increase\nbarge hold densities for the Skamania Landing (8000 series) and Astoria\n(2000 series) release barges, 2006. The total number pounds of fish in the barge\nholds are also shown.\nRelease date\nBarge\nNumber tagged\nNumber not tagged\nTotal pounds\n26 April\nSkamania Landing\n9,333\n7,092\n1,727\n26 April\nAstoria\n10,885\n14,129\n2,539\n2 May\nSkamania Landing\n12,804\n1,431\n1,585\n3 May\nAstoria\n7,968\n12,139\n2,068\n9 May\nSkamania Landing\n12,342\n35,252\n4,719\n11 May\nAstoria\n8,025\n16,318\n2,413\n16 May\nSkamania Landing\n12,856\n10,357\n3,196\n17 May\nAstoria\n8,078\n17,512\n3,480\n25 May\nSkamania Landing\n13,693\n8,886\n2,600\n26 May\nAstoria\n7,762\n4,000\n1,192\n1,517\n1 June\nSkamania Landing\n4,989\n14,427\n1,241\n2 June\nAstoria\n2,617\n13,867\nAvian Predation\nBased on PIT tag recoveries on East Sand Island, we estimated that avian\npredators consumed a significantly higher proportion of fish released at Skamania\nLanding than of those released at Astoria. For both steelhead and yearling Chinook\nsalmon, the avian predation rate was lower for fish released at Astoria during nighttime\nand on an ebb tide than for fish released 215 km upstream at Skamania Landing. On the\nEast Sand Island tern colony, 11.8% of the tags from steelhead released at Skamania\nLanding were recovered, while 1.6% of the tags from steelhead released at Astoria were\nrecovered. An additional 2.0% of tags from steelhead released at Skamania Landing and\n0.1% of tags from steelhead released at Astoria were detected on the colonies of\ndouble-crested cormorants. Losses to both colonies combined were of 13.8% for\nSkamania Landing and 1.7% for Astoria release locations. Smaller proportions of\nPIT tags from Chinook salmon were recovered on the colonies, but these recoveries still\nshowed the trend of significantly higher predation rates for fish released at Skamania\nLanding (Table 6).\n10","Table 6. Percentage of PIT tags detected on the East Sand Island Caspian tern and\ndouble-crested cormorant colonies for both steelhead and yearling Chinook\nsalmon smolts from 2006 releases at Skamania Landing and near Astoria. A\npaired t-test was used to compare release locations.\nTags detected (%)\nSteelhead\nChinook salmon\nBird colony\nAstoria\nSkamania\nAstoria\nSkamania\nCaspian tern\n1.61\n11.8\n0.32\n1.54\nPaired t-test\nt = 5.89, P = 0.002\nt = 7.24 , P = 0.001\nDouble-crested cormorant\n0.13\n1.98\n0.09\n1.44\nPaired t-test\nt = 4.21 P = 0.008\nt = 4.40 , P = 0.007\nDate-Specific Predation Results\nTagged fish were released on six separate occasions between April and June at\nboth the Skamania Landing and Astoria release sites (Table 4). The Astoria release\ngroups showed lower tag proportions detected on both tern and cormorant colonies than\ndid the Skamania Landing release groups for all replicates (Table 7).\nTable 7. Percentage of PIT tags detected by date on the East Sand Island Caspian tern\nand double-crested cormorant colonies for both steelhead and yearling Chinook\nsalmon released during 2006.\nTags detected (%)\nCaspian tern\nDouble-crested cormorant\nRelease date\nAstoria\nSkamania\nAstoria\nSkamania\nSteelhead\n26 April\n1.45\n12.39\n0.01\n1.63\n2-3 May\n0.35\n5.17\n0.38\n3.39\n9 and 11 May\n0.54\n10.07\n0.11\n2.84\n16-17 May\n2.55\n9.35\n0.17\n1.25\n25-26 May\n1.71\n13.57\n0.12\n2.16\n1-2 June\n3.81\n20.70\n0.08\n0.31\nChinook salmon\n26 April\n0.64\n2.45\n0.11\n1.11\n2-3 May\n0.14\n1.40\n0.12\n1.68\n9 and 11 May\n0.08\n1.75\n0.00\n1.42\n16-17 May\n0.52\n1.15\n0.09\n1.51\n25-26 May\n0.10\n1.09\n0.19\n1.04\n1-2 June\n0.00\n1.34\n0.00\n0.00\n11","Pathogen Analyses\nThe proportion of fish found infected with R. salmoninarum by qPCR was\nsignificantly higher than that found by nPCR in gill samples from wild and hatchery\nChinook salmon and wild and hatchery steelhead (P <0.0001; Table 8). However, the\nprevalence of R. salmoninarum did not differ among species between nPCR (P = 0.76)\nand qPCR testing (P = 0.81). Prevalence of N. salmonis was significantly higher in\nhatchery steelhead than in wild steelhead, wild Chinook salmon, or hatchery Chinook\nsalmon (Table 8; P <0.0001).\nTable 8. Pathogen detection in gill samples from all tested fish. Detection of\nRenibacterium salmoninarum by nested PCR (nPCR) and quantitative PCR\n(qPCR) and detection of Nucleospora salmonis by nPCR in gill tissues from\nhatchery and wild Chinook salmon smolts and hatchery and wild steelhead\nsmolts sampled non-lethally at the time of tagging from the six release groups\nof fish marked with PIT tags at Lower Granite Dam during 2006.\nNumber of positive fish/Number sampled (%)\nRenibacterium salmoninarum\nNucleospora salmonis\nFish species\nnPCR\nqPCR\nnPCR\n14/394 (4%)\nWild Chinook salmon\n138/394 (35%)\n250/394 (63%)\n237/383 (62%)\n2/383 (1%)\nHatchery Chinook salmon\n124/383 (32%)\n181/506 (36%)\n319/498 (64%)\n29/498 (6%)\nWild steelhead\n116/510 (23%)\nHatchery steelhead\n176/510 (35%)\n313/510 (61%)\nAmong fish testing positive for R. salmoninarum by qPCR, R. salmoninarum\nlevels were generally less than 100 bacteria per mg of gill sample; samples from only\n12 fish had R. salmoninarum concentrations exceeding this level. These fish included\nthree hatchery Chinook salmon (highest concentration 636 R. salmoninarum per mg),\nthree wild Chinook salmon (highest concentration 172 R. salmoninarum per mg), three\nhatchery steelhead (highest concentration 203 R. salmoninarum per mg), and three wild\nsteelhead (highest concentration 513 R. salmoninarum per mg). Nevertheless, slightly\nmore than half of the qPCR-positive fish of each species had R. salmoninarum levels\nabove the threshold for consistent detection by the assay (5 bacteria per q PCR reaction;\nTable 9).\n12","Table 9. Mean levels of Renibacterium salmoninarum detected by qPCR in smolts\nsampled non-lethally from the six release groups of fish marked with PIT tags at\nLower Granite Dam during 2006. Means not sharing a common letter are\nsignificantly different (P < 0.05).\nNumber above threshold\nfor consistent R. salmoninarum\nGeometric mean number\ndetection/Total number\nR. salmoninarum per mg\npositive by qPCR (%)\ngill sample\" (+SD)\nFish species\nWild Chinook salmon\n132/250 (53)\n11 (+3) y\nHatchery Chinook salmon\n124/237 (52)\n12 (+2) y\nWild steelhead\n177/319 (55)\n14 (+2) Z\nHatchery steelhead\n173/313 (55)\n14 (+2) Z\na\nFive bacteria per qPCR reaction.\nb\nNumber of R. salmoninarum per mg calculated according to the following formula:\n(Number of R. salmoninarum per reaction X 40) / Sample weight,\nwhere 40 is the dilution factor\nMean R. salmoninarum levels in wild Chinook salmon were not significantly\ndifferent from those in hatchery Chinook salmon (P >0.05), and mean R. salmoninarum\nlevels in wild steelhead were not significantly different from those in hatchery steelhead\n(P >0.05). However, R. salmoninarum levels in wild steelhead were significantly higher\nthan levels in both wild (P <0.01) and hatchery Chinook salmon (P <0.01).\nR. salmoninarum levels in hatchery steelhead were also significantly higher than levels in\nboth wild Chinook salmon (P <0.05) and hatchery Chinook salmon (P <0.01).\nThe proportions of fish testing positive for R. salmoninarum, N. salmonis, or both\npathogens varied by species and sample date (Table 10). For wild Chinook salmon, the\nproportions of fish positive for either or both pathogens did not differ significantly among\nsample dates (P = 0.24). However, for hatchery Chinook salmon, the proportions of fish\npositive for either or both pathogens were significantly higher (P <0.0001) for samples\ncollected on 7, 14, and 22 May than for samples collected on 23 or 30 April.\nFor wild steelhead, the proportions of fish positive for one or both pathogens were\nsignificantly higher (P <0.0001) for samples collected on 7, 14, and 30 May than for\nsamples collected on 23 and 30 April or 22 May, and were also significantly higher\n(P = 0.006) for samples collected on 23 and 30 April than for the samples collected on\n22 May. For hatchery steelhead, the proportions of fish positive for either or both\npathogens were significantly higher (P <0.0001) for samples collected on 7 and 30 May\nthan for those collected on 23 and 30 April and 14 May.\n13","Table 10. Proportions of fish testing positive for Renibacterium salmoninarum only, for\nNucleospora salmonis only, for both pathogens, or for neither pathogen among\nall fish sampled on each date from PIT-tagged groups at Lower Granite Dam\nin 2006.\nNumber positive\nNumber positive\nNumber positive\nR. salmoninarum\nNumber with\nR. salmoninarum\nSample date and fish\nNumber\nN. salmonis only\nand N. salmonis\nneither pathogen\nonly (%)\nspecies\ntested\n(%)\n(%)\ndetected (%)\n23 April\nWild Chinook\n75\n43 (57%)\n2 (3%)\n8 (11%)\n22 (29%)\nHatchery Chinook\n75\n39 (52%)\n1 (1%)\n0\n35 (47%)\nWild steelhead\n75\n41 (55%)\n3 (4%)\n2 (3%)\n29 (39%)\nHatchery steelhead\n75\n39 (52%)\n6 (8%)\n5 (7%)\n25 (33%)\n30 April\nWild Chinook\n75\n62 (83%)\n0\n0\n13 (17%)\nHatchery Chinook\n75\n43 (57%)\n1 (1%)\n0\n31 (41%)\nWild steelhead\n75\n51 (68%)\n1 (1%)\n3 (4%)\n20 (27%)\nHatchery steelhead\n68\n37 (54%)\n8 (12%)\n10 (13%)\n13 (19%)\n7 May\nWild Chinook\n75\n54 (72%)\n0\n2 (3%)\n19 (25%)\nHatchery Chinook\n75\n66 (88%)\n0\n0\n9 (12%)\nWild steelhead\n67\n54 (81%)\n1 (1%)\n3 (4%)\n9 (13%)\nHatchery steelhead\n75\n52 (69%)\n4 (5%)\n13 (17%)\n6 (8%)\n14 May\nWild Chinook\n75\n61 (81%)\n0\n1 (1%)\n13 (17%)\nHatchery Chinook\n75\n63 (84%)\n0\n0\n12 (16%)\nWild steelhead\n75\n59 (79%)\n0\n2 (3%)\n14 (19%)\nHatchery steelhead\n75\n33 (44%)\n10 (13%)\n12 (16%)\n20 (27%)\n22 May\nWild Chinook\n75\n53 (71%)\n0\n1 (1%)\n21 (28%)\nHatchery Chinook\n75\n56 (75%)\n0\n0\n19 (25%)\nWild steelhead\n75\n32 (43%)\n3 (4%)\n1 (1%)\n39 (52%)\nHatchery steelhead\n75\n29 (39%)\n11 (15%)\n11 (15%)\n24 (32%)\n30 May\nWild Chinook\n19\n17 (89%)\n0\n0\n2 (11%)\nHatchery Chinook\n8\n8 (100%)\n0\n0\n0\nWild steelhead\n131\n107 (82%)\n1 (<1%)\n9 (7%)\n14 (11%)\nHatchery steelhead\n142\n98 (69%)\n6 (4%)\n20 (14%)\n18 (13%)\nAll dates\nWild Chinook\n394\n290 (74%)\n2 (<1%)\n12 (3%)\n90 (23%)\nHatchery Chinook\n383\n275 (72%)\n2 (<1%)\n0\n106 (28%)\nWild steelhead\n498\n344 (69%)\n9 (2%)\n20 (4%)\n125 (25%)\nHatchery steelhead\n510\n288 (56%)\n45 (9%)\n71 (14%)\n106 (21%)\n* Positive for R. salmoninarum by nPCR, qPCR, or both PCRs.\n14","When data were combined for all sample dates, however, the proportions of fish\npositive for one or both pathogens did not differ significantly among species (P = 0.095).\nOverall, one or both pathogens were detected in 77% of the wild Chinook salmon, 72%\nof the hatchery Chinook salmon, 75% of the wild steelhead, and 79% of the hatchery\nsteelhead.\nFish that were infected with either R. salmoninarum or N. salmonis or both\npathogens did not differ significantly in length (P >0.26) from fish not infected with\neither pathogen. However, among wild Chinook salmon positive for R. salmoninarum by\nqPCR testing, there was a significant negative correlation (P = 0.01) between length and\nR. salmoninarum infection level. There was no significant correlation between length\nand R. salmoninarum infection level for hatchery Chinook salmon (P = 0.07), wild\nsteelhead (P = 0.14), or hatchery steelhead (P =0.17)\nAmong the PIT tags recovered from the East Sand Island tern and cormorant\ncolonies, 159 (20 Chinook salmon and 139 steelhead) were from fish that had been\nsampled at Lower Granite Dam for detection of R. salmoninarum and N. salmonis.\nPathogen prevalence and levels in these fish were similar to those of all fish tested for\npathogens. Of fish with PIT tags recovered on bird colonies, 67% of wild and 60% of\nhatchery Chinook salmon and 67% of wild and 77% of hatchery steelhead were positive\nfor one or both pathogens (Table 11).\nTable 11. Proportions of fish testing positive for Renibacterium salmoninarum only, for\nNucleospora salmonis only, for both pathogens, or for neither pathogen among\nfish sampled at Lower Granite Dam during 2006, with PIT tags subsequently\nrecovered on the East Sand Island piscivorous bird colonies.\nR. salmoninarum\nNeither\nR. salmoninarum\nN. salmonis\nand N. salmonis\npathogen\nNumber\npositive only*\npositive only\npositive\ndetected\nFish species\ntested\nn (%)\nn (%)\nn (%)\nn (%)\nWild Chinook\n15\n10 (67%)\n0\n0\n5 (33%)\nHatchery Chinook\n5\n3 (60%)\n0\n0\n2 (40%)\nWild steelhead\n52\n33 (63%)\n0\n2 (4%)\n17 (33%)\nHatchery steelhead\n87\n53 (61%)\n6 (7%)\n8 (9%)\n20 (23%)\n* Positive for R. salmoninarum by nPCR, qPCR, or both PCRs.\n15","Overall, 52% of the fish testing positive for R. salmoninarum by qPCR had levels\nof the bacterium above the threshold for consistent detection by the assay (5 bacteria per\nqPCR reaction; Table 12). R. salmoninarum concentrations for qPCR-positive fish\nranged from <1 to 148 bacteria/mg for wild Chinook salmon, from 5 to 37 bacteria/mg\nfor wild steelhead, and from 3 to 59 bacteria/mg for hatchery steelhead. The single\nqPCR-positive hatchery Chinook salmon had an R. salmoninarum concentration of\n13 bacteria/mg.\nTable 12. Mean levels of Renibacterium salmoninarum detected by qPCR in smolts\nsampled at Lower Granite Dam during 2006, with PIT tags subsequently\nrecovered on the East Sand Island piscivorous bird colonies.\nNumber above threshold for\nconsistent\nGeometric mean number\nR. salmoninarum detection by\nR. salmoninarum per mg\ngill sample\" b (+SD)\nFish species\nqPCR/Total positive (%)\nWild Chinook salmon\n4/9 (44%)\n9 (+8)\nHatchery Chinook salmon\n1/1 (100%)\nc\nWild steelhead\n16/29 (55%)\n14 (+2)\nHatchery steelhead\n29/58 (50%)\n13 (+2)\na\nFive bacteria per qPCR reaction.\nb\nNumber of R. salmoninarum per mg calculated according to the following formula:\n(Number of R. salmoninarum per reaction X 40) / Sample weight, where 40 is the dilution factor\nc\nSingle qPCR-positive fish had 13 R. salmoninarum per mg gill sample.\nThe highest mean R. salmoninarum concentrations detected in water samples\ntaken from the Lower Granite Dam tagging station were usually obtained from the last\nsample taken each day (Table 13). Mean R. salmoninarum concentrations in a given\nsample ranged from 8 to 133 bacteria per mL. There was no significant correlation\nbetween daily mean R. salmoninarum levels detected in gill samples of qPCR-positive\nfish, and daily mean R. salmoninarum concentrations detected in water samples\n(P = 0.56). The lowest and highest mean R. salmoninarum levels in gill samples were\ndetected on 14 and 30 May, respectively, and the lowest and highest R. salmoninarum\nconcentrations in water samples were detected on 22 May and 30 April, respectively\n(Table 14).\n16","Table 13. Mean Renibacterium salmoninarum concentrations in water samples taken\nfrom the PIT-tagging station from which fish were also sampled for pathogen\ntesting at Lower Granite Dam, 2006. After the first two water samples were\ntaken each day, the water in the recirculating system was changed.\nGeometric mean\nSample date and time\nR. salmoninarum/mL (+SD)\n23 April\n9:45 a.m.\n8 (+1)\n11:30 a.m.\n35 (+1)\n12:30 p.m.\n20 (+2)\n4:00 p.m.\n133 (+1)\n30 April\n10:20 a.m.\n29 (+2)\n11:00 a.m.\n30 (+1)\n1:00 p.m.\n44 (+1)\n4:00 p.m.\n54 (+2)\n7 May\n10:15 a.m.\n23 (+1)\n12:20 p.m.\n10 (+2)\n1:30 p.m.\n16 (+1)\n3:10 p.m.\n25 (+1)\n14 May\n9:45 a.m.\n16 (+1)\n11:30 a.m.\n16 (+2)\n1:40 p.m.\n16 (+2)\n3:30 p.m.\n37 (+2)\n22 May\n11:30 a.m.\n11 (+2)\n11:45 a.m.\n14 (+2)\n1:00 p.m.\n11 (+2)\n4:00 p.m.\n14 (+2)\n30 May\n10:00 a.m.\n12 (+1)\n11:00 a.m.\n21 (+1)\n12:30 p.m.\n11 (+2)\n2:05 p.m.\n28 (+1)\n17","Table 14. Mean levels of Renibacterium salmoninarum detected by qPCR in daily\nsamples of smolts (both species combined) sampled non-lethally during\nPIT-tagging at Lower Granite Dam in 2006. Also shown are mean daily\nconcentrations of R. salmoninarum detected by MF-FAT in water samples\ntaken at the tagging station from which fish were sampled. Within a column,\nmeans not sharing a common letter are significantly different (P <0.05).\nGeometric mean\nGeometric mean\nR. salmoninarum\nR. salmoninarum\nSample date\nper mg gill sample (+SD)\nper mL water sample (+SD)\n23 April\n10 (+2) X\n29 (+3) t\n30 April\n14 (+2) y\n38 (+1) tu\n7 May\n16 (+2) yz\n17 (+2) tv\n14 May\n9 (+3) X\n20 (+2) tuvw\n22 May\n11 (+2) X\n13 (+2) tvw\n30 May\n18 (+2) Z\n17 (+2) tvw\n18","DISCUSSION\nOne goal of this study was to evaluate whether releasing transported salmonids\ndownstream from the Astoria Bridge could increase survival to ocean entry. We\nhypothesized that using this release site would increase survival to ocean entry by\nallowing fish to avoid some avian predators in the Columbia River estuary. Steelhead are\nparticularly vulnerable to predation by piscivorous birds; Collis et al. (2001) reported that\nover 15% of the PIT tags from steelhead detected at Bonneville Dam in 1998 were later\nfound on estuarine bird colonies. In contrast, they found only 2% of the PIT tags from\nyearling Chinook detected at the dam that year.\nIn 1998 the major site of tag recovery was Rice Island, which was then home to\nthe largest Caspian tern colony in North America (Collis et al. 2002). Ryan et al. (2002,\n2003) and Glabek et al. (2003) reported similar results in subsequent years, as the tern\ncolony was relocated from Rice Island to East Sand Island. To optimize survival, smolts\nwere released at night on an ebb tide, during which time most would be expected to pass\nthe bird colonies and reach the ocean during one tidal cycle (Ledgerwood et al. 2001).\nOur results supported this hypothesis, showing that a significantly higher proportion of\nsalmon released at Skamania were found on both large avian colonies in the estuary\ncompared to their cohorts released at Astoria.\nIn addition to facilitating the avoidance of avian predators, the release of fish near\nAstoria could improve survival in other ways. The release site downstream from\nBonneville Dam near Skamania Landing has the highest rate of predation by northern\npikeminnow Ptychocheilus oregonensis in the Columbia River (Ward et al. 1995). By\ntransporting smolts farther downstream to Astoria, this source of potential mortality can\nbe avoided. Release in the lower estuary also allows smolts to avoid migrating through\nthe Willamette River confluence area, where high levels of toxic chemicals have been\nfound (Spromberg et al. in press). Barged fish would still pass through this area (and be\nexposed to the same water as migrating fish), but their duration of exposure to any toxic\nchemicals would be shorter. Finally, during years with low flow/high water temperature,\nsteelhead migrants often residualize in reservoirs late in the migration. Few of these\nresiduals survive to migrate the following spring (Williams et al. 2005). Releasing them\nnear the mouth of the river in strong current during an ebb tide might encourage these\nfish to migrate rather than overwinter in reservoirs, thus improving overall SARs.\nIn the end, the important question is not which release site allows juvenile\nsalmonids to survive at greater rates to ocean entry, but rather which release site produces\nthe greatest SARs. It is conceivable that survival to ocean entry could be higher for fish\n19","released at Astoria, but that even with higher short-term survival, these fish could return\nat the same or lower rates than fish released at Skamania Landing. This could occur if\nfish released at Astoria were less physiologically prepared to enter seawater. Conversely,\nthe group released at Astoria could be in better condition due to avoiding migration\nthrough the lower Columbia River, and could produce higher SARs. Ultimately, the\nsuccess of either release site will be determined by examining differences in SARs\namong release groups.\nModification of the qPCR procedure for R. salmoninarum improved the\nsensitivity of the assay compared to that used in the 2005 study (McMichael et al. 2006).\nIn that study, for which only hatchery steelhead were sampled, 14% of the 1,002 fish\nwere positive for R. salmoninarum by qPCR, whereas 61% of the 510 hatchery steelhead\ntested in 2006 were positive by qPCR. In comparison, similar proportions of hatchery\nsteelhead were positive for R. salmoninarum by nPCR in 2005 (33%) and 2006 (35%).\nFurthermore, in 2005, only 2% of the qPCR-positive hatchery steelhead had R.\nsalmoninarum levels at or above the threshold for consistent detection of R.\nsalmoninarum (5 bacteria per PCR reaction), while 55% of the qPCR-positive hatchery\nsteelhead in 2006 had R. salmoninarum levels at or above this threshold. The biological\nsignificance of the increased detection sensitivity of the modified qPCR remains to be\ndetermined.\nThe present study using PCR for R. salmoninarum detection found no significant\ndifferences in prevalence of the bacterium between steelhead and Chinook salmon, but\nshowed higher infection levels in steelhead. Previous comparisons of Chinook salmon\nand steelhead at Lower Granite Dam used the enzyme-linked immunosorbent assay\n(ELISA) to detect R. salmoninarum (Pascho and Elliott 1989; Elliott and Pascho 1991,\n1993). In those studies, both levels and prevalence of R. salmoninarum in Chinook\nsalmon were found to be higher than or equal to those in steelhead. In addition, these\nearlier studies consistently found significantly higher R. salmoninarum levels by ELISA\nin wild steelhead than in hatchery steelhead, a result that was not observed in the present\nPCR study. (Wild and hatchery stocks of yearling Chinook salmon could not be\ndistinguished during 1988-1990, with the exception of a few tagged fish, SO these were\nnot separated in analyses.)\nBecause the ELISA detects soluble antigen produced by R. salmoninarum and\nPCR detects bacterial DNA, the results of the studies are not directly comparable. R.\nsalmoninarum antigen can persist in fish in the absence of live bacteria (Pascho et al.\n1997); the correlation between bacterial viability and detection by PCR has not been\ndetermined. Some studies suggest that changes in hatchery practices may have\ncontributed to reductions in R. salmoninarum levels (as determined by ELISA testing) in\n20","certain hatchery populations of Chinook salmon in the Snake and Columbia River basins\n(Maule et al. 1996; Vanderkooi and Maule 1999; Munson and Johnson 2005), although it\nis unknown whether these changes have resulted in increased long-term survival of fish.\nYear-to-year variation in prevalence and severity of R. salmoninarum infections\nalso occurs among salmonid smolts (Pascho and Elliott 1989; Elliott and Pascho 1991,\n1993; Elliott et al. 1997; Vanderkooi and Maule 1999) and may have influenced the\nresults of our study. Although PCR results of the present research cannot be directly\ncompared to ELISA results of past studies, limited evidence suggests that the majority of\nfish tested by PCR in the present study would have shown negative or low R.\nsalmoninarum antigen levels by ELISA testing (Chase et al. 2006). Therefore, these fish\nwould not have been considered clinically diseased (e.g., showing grossly visible kidney\nlesions had they been sacrificed) at the time of sampling.\nThe influence of low R. salmoninarum infection levels on long-term survival of\nfish is not well understood. However, in a previous study with yearling hatchery\nChinook salmon, significantly higher survival was observed during downriver migration\nand a 3-month seawater holding period among groups of smolts with low average\nR. salmoninarum antigen levels (by ELISA testing of kidney tissue) than among groups\nwith medium to high average R. salmoninarum antigen levels (Pascho et al. 1993; Elliott\net al. 1995). Similar studies have not yet been conducted with groups of fish from which\ntissues have been tested by PCR. Laboratory research is currently underway to better\ndefine the relation between concentrations of R. salmoninarum DNA detected by PCR in\nkidney and gill samples and the infection status of fish. In addition, field studies such as\nthis one are intended to provide further information on possible correlations between\nR. salmoninarum levels in salmonid populations and fish survival.\nA surprising result was the high prevalence of N. salmonis in hatchery steelhead\ncompared with wild steelhead and hatchery and wild Chinook salmon. The prevalence of\nN. salmonis detected in hatchery steelhead in 2006 (23%) was similar to the 25%\nprevalence detected in hatchery steelhead in 2005, when this was the only group of fish\nsampled (McMichael et al. 2006). Although N. salmonis mortality has been primarily\nreported in juvenile and adult Chinook salmon in freshwater and seawater (see Gresoviac\net al. 2000 for summary), the pathogen also causes mortality in certain stocks of steelhead\nreared in hatcheries in Idaho (Kathy Clemens, Idaho Fish Health Center, U.S. Fish and\nWildlife Service, Orofino Idaho, personal communication).\nRegular monitoring of N. salmonis is not done for many hatchery steelhead\npopulations because of the difficulty of detection prior to the development of PCR assays.\nTherefore, the possible contribution of the pathogen to delayed mortality of steelhead\nsmolts has not been determined. Similar to R. salmoninarum infections, N. salmonis\n21","infections can result in reduced immune function, allowing for other opportunistic\ninfections, but can also be directly fatal (Hedrick et al. 1990).\nThe presence of one or both pathogens did not appear to influence the\nsusceptibility of steelhead and Chinook salmon to predation by piscivorous birds,\nalthough the low number of PIT tags recovered on East Sand Island from infected\nChinook salmon (20 fish total, including 15 wild fish and 5 hatchery fish) precluded\nmeaningful analyses for this species. A lack of preferential predation on infected fish\nwas not surprising, considering the relatively low R. salmoninarum levels in the majority\nof fish tested. The N. salmonis levels were not determined.\nPrevious work has indicated higher vulnerability to predation by piscivorous fish\namong juvenile Chinook salmon with moderate to high R. salmoninarum infection levels\nas determined by ELISA (Mesa et al. 1998), but the influence of similar levels of this\npathogen or N. salmonis on vulnerability to avian predation has not been determined.\nChanges in R. salmoninarum concentrations in the water of the tagging trough\nfrom which fish were sampled for pathogen analyses did not correspond to changes in\nmean infection levels in fish on different sample days. This may suggest that R.\nsalmoninarum levels in gill samples were not significantly affected by surface\ncontamination with R. salmoninarum in the water. The consistent detection of the\nhighest R. salmoninarum concentrations at the end of the day suggested that bacteria may\ncontinue to concentrate in the trough, despite the midday water change. On most days,\nhowever, the increase in bacteria in the last sample was minimal, and overall\nconcentrations were similar to those observed in other recirculating systems in\nconventional tagging trailers (D. G. Elliott, USGS, unpublished data). The MF-FAT\ncannot distinguish live from dead bacteria, SO the viability of R. salmoninarum in the\ntroughs was unknown.\nPrevious studies with coho and Atlantic salmon found that release of transported\nsmolts to the estuary or ocean resulted in higher adult return rates than release to\nfreshwater (Solazzi et al. 1991; Gunnerod et al. 1988). Studies to evaluate the release of\ntransported steelhead in the Columbia River estuary (Tongue Point) vs. Skamania\nLanding were conducted from 1992 to 1994 (smolt release years). For the 1994 release\nyear, the ratio of Tongue Point to Skamania Landing adult returns was 3.0, while for the\nother two release years it was near 1.0. However, these results were inconclusive\nbecause too few adults returned from all three release years (Marsh et al. 1996, 1998,\n2000). The return of PIT-tagged adults over the next several years will be required to\ndetermine whether adult returns are improved by releasing transported yearling Chinook\nsalmon and steelhead smolts to a site lower in the Columbia River estuary (rkm 10)\ncompared to the traditional release site at Skamania Landing.\n22","ACKNOWLEDGEMENTS\nWe thank Ann Setter, Mike Halter, Kent Blevins, and staff from the U.S. Army\nCorps of Engineers and the Washington Department of Fish and Wildlife Smolt\nMonitoring Program for their assistance at Lower Granite Dam. We thank Ken McIntyre\nand Neil Passach (National Marine Fisheries Service) for assistance with fish tagging,\nand Mike Brim (Pacific States Marine Fisheries Commission) for his assistance on the\nbarge. Dave Marvin, Albert Giorgi, and Tim Wik provided helpful comments on our\ndraft report that improved this final version.\n23","24","REFERENCES\nBarlough J.E., T. S. McDowell, A. Milani, L. Bigornia, S. B. Slemenda, N. J. Pieniazek,\nand R. P. Hedrick. 1995. Nested polymerase chain reaction for detection of\nEnterocytozoon salmonis genomic DNA in chinook salmon Oncorhynchus\ntshawytscha. Diseases of Aquatic Organisms 23:17-23.\nBudy, P., G. P. Thiede, N. Bouwes, C. E. Petrosky, and H. Schaller. 2002. Evidence\nlinking delayed mortality of Snake River salmon to their earlier hydropower\nsystem experience. North American Journal of Fisheries Management 22:35-51.\nBurnham, K. P., D. R. Anderson, G. C. White, C. Brownie, and K. H. Pollock. 1987.\nDesign and analysis methods for fish survival experiments based on release-\nrecapture. American Fisheries Society Monograph 5:1-437.\nCBFWA (Columbia Basin Fish and Wildlife Authority). 1999. PIT Tag Marking\nProcedures Manual. Version 2.0. Available at www.psmfc.org. (August 2007).\nChase, D. M., D. G. Elliott, R. J. Pascho. 2006. Detection and quantification of\nRenibacterium salmoninarum DNA in salmonid tissues by real-time quantitative\npolymerase chain reaction. Journal of Veterinary Diagnosis and Investigation\n18:375-380.\nChase, D. M., and R. J. Pascho. 1998. Development of a nested polymerase chain\nreaction for amplification of a sequence of the p57 gene of Renibacterium\nsalmoninarum that provides a highly sensitive method for detection of the\nbacterium in salmonid kidney. Diseases of Aquatic Organisms 34:223-229.\nCollis, K. D., D. D. Roby, D. P. Craig, S. Adamany, J. Y. Adkins, and D. E. Lyons.\n2002. Colony size and diet composition of piscivorous waterbirds on the lower\nColumbia River: Implications for losses of juvenile salmonids to avian predation.\nTransactions of the American Fisheries Society 131:537-550.\nCollis, K., D. D. Roby, D. P. Craig, B. R. Ryan, and R. D. Ledgerwood. 2001. Colonial\nwaterbird predation on juvenile salmonids tagged with passive integrated\ntransponders in the Columbia River Estuary: Vulnerability of different salmonid\nspecies, stocks, and rearing types. Transactions of the American Fisheries Society\n130:385-396.\n25","Congleton, J. L., J. Evavold, D. Jones, M. Santora, B. Sun, and T. Wagner. 2005.\nEvlauation of physiological condition of transported and inriver migrating\njuvenile salmonids and effects on survival (DACW68-00-C-0030). Annual\nReport, 2003. Report of Idaho Cooperative Fish and Wildlife Research Unit to\nthe U.S. Army Corps of Engineers, Walla Walla District, Walla Walla, WA.\nCongleton, J. L., W. J. LaVoie, C. B. Schreck, and L. E. Davis. 2000. Stress indices in\nmigrating juvenile Chinook salmon and steelhead of wild and hatchery origin\nbefore and after barge transportation. Transactions of the American Fisheries\nSociety 129:946-961.\nElliott, D. G., and C. L. McKibben. 1997. Comparison of two fluorescent antibody\ntechniques (FATs) for detection and quantification of Renibacterium\nsalmoninarum in coelomic fluid of spawning Chinook salmon Oncorhynchus\ntshawytscha. Diseases of Aquatic Organisms 30:37-43.\nElliott, D. G., and R. J. Pascho. 1991. Juvenile fish transportation: impact of bacterial\nkidney disease on survival of spring/summer chinook salmon stocks. Annual\nReport, 1989 (Contract E86880047) prepared by the U.S. Fish and Wildlife\nService, Seattle, WA, for the U.S. Army Corps of Engineers, Walla Walla, WA.\nElliott, D. G., and R. J. Pascho. 1992. Juvenile fish transportation: impact of bacterial\nkidney disease on survival of spring/summer chinook salmon stocks. Annual\nReport, 1990 (Contract E86890043) prepared by the U.S. Fish and Wildlife\nService, Seattle, WA, for the U.S. Army Corps of Engineers, Walla Walla, WA.\nElliott, D. G., and R. J. Pascho. 1993. Juvenile fish transportation: impact of bacterial\nkidney disease on survival of spring/summer chinook salmon stocks. Annual\nReport, 1991 (Contract E86910058) prepared by the U.S. Fish and Wildlife\nService, Seattle, WA, for the U.S. Army Corps of Engineers, Walla Walla, WA.\nElliott, D. G., and R. J. Pascho. 1994. Juvenile fish transportation: impact of bacterial\nkidney disease on survival of spring/summer chinook salmon stocks. Annual\nReport, 1992 (Contract E86920048) prepared by the U.S. Fish and Wildlife\nService, Seattle, WA, for the U.S. Army Corps of Engineers, Walla Walla, WA.\nElliott, D. G., and R. J. Pascho. 2004. Studies on the detection, transmission, and\ndevelopment of Renibacterium salmoninarum infections in Great Lakes salmonid\nfishes. Final report, Project No. 1999.51 (1999.12). Great Lakes Fishery Trust,\nLansing, MI. Available at www.glft.org\nElliott D.G., R. J. Pascho, and A. N. Palmisano. 1995. Brood stock segregation for the\ncontrol of bacterial kidney disease can affect mortality of progeny Chinook\nsalmon (Oncorhynchus tshawytscha) in seawater. Aquaculture 132:133-144.\n26","Elliott, D. G., R. J. Pascho, L. M. Jackson, G .M. Matthews, and J. R. Harmon. 1997.\nPrevalence and levels of Renibacterium salmoninarum infection in spring/summer\nchinook salmon (Oncorhynchus tshawytscha) smolts at dams on the Columbia\nand Snake rivers. Journal of Aquatic Animal Health 9:114-126.\nGlabek, J. H., B. A. Ryan, E. P. Nunnallee, and J. W. Ferguson. 2003. Detection of\npassive integrated transponder (PIT) tags on piscivorous bird colonies in the\nColumbia River Basin, 2001. Report of the National Marine Fisheries Service to\nthe U.S. Army Corps of Engineers, Walla Walla, Washington. Journal of\nEukaryotic Microbiology 47:379-387.\nGunnerod, T.B., N.A. Hvidsten, and T.G. Heggberget. 1988. Open sea releases of\nAtlantic salmon smolts, Salmo salar, in central Norway, 1973-1983. Canadian\nJournal of Fisheries and Aquatic Sciences 45:1340-1345.\nGresoviac, S. J., J. S. Khattra, S.A. Nadler, M. L. Kent, R. H. Devlin, C. P. Vivares, E.\nde la Fuente, and R. P. Hedrick. 2000. Comparison of small subunit ribosomal\nRNA gene and internal transcribed spacer sequences among isolates of the\nintranuclear microsporidian Nucleospora salmonis. Journal of Eukaryotic\nMicrobiology 47:379-387.\nHedrick, R.P., J. M. Groff, T. S. McDowell, M. Willis, and W. T. Cox. 1990.\nHematopoietic intranuclear microsporidian infection with features of leukemia in\nchinook salmon Oncorhynchus tshawytscha. Diseases of Aquatic Organisms\n8:189-197.\nJohnson, G. D., R. S. Davidson, K. C. McNamee, G. Russell, D. Goodwin, and E. J.\nHolborow. 1982. Fading of immunofluorescence during microscopy: a study of\nthe phenomenon and its remedy. Journal of Immunological Methods 55:231-242.\nKelsey, D. A., C. B. Schreck, J. L. Congleton, and L.E. Davis. 2002. Effects of juvenile\nsteelhead on juvenile Chinook salmon behavior and physiology. Transactions of\nthe American Fisheries Society 131:676-689.\nLedgerwood, R. D., B. Ryan, and R. N. Iwamoto. 2001. Estuarine and nearshore-ocean\nacoustic tracking of juvenile spring chinook salmon from the Columbia River.\nPages 245-255 in Moore, A., I. Russell. (Eds.) Advances in fish telemetry:\nproceedings of the 3rd conference on fish telemetry in Europe, June 20-25, 1999,\nNorwich, UK. Centre for Environment, Fisheries, and Aquaculture Science,\nLowestoft, UK.\nMarsh, D. M., J. R. Harmon, K. W. McIntyre, K. L. Thomas, N. N. Paasch, B. P.\nSandford, D. J. Kamikawa, and G. M. Matthews. 1996. Research related to\ntransportation of juvenile salmonids on the Columbia and Snake Rivers, 1995.\nReport of the U.S. National Marine Fisheries Service to the U.S. Army Corps of\nEngineers, Walla Walla District.\n27","Marsh, D. M., J. R. Harmon, N. N. Paasch, K. L. Thomas, K. W. McIntyre, B. P.\nSandford, and G. M. Mathews. 1998. Research related to transportation of\njuvenile salmonids on the Columbia and Snake Rivers, 1997. Report of the U.S.\nNational Marine Fisheries Service to the U.S. Army Corps of Engineers, Walla\nWalla District.\nMarsh, D. M., J. R. Harmon, N. N. Paasch, K. L. Thomas, K. W. McIntyre, B. P.\nSandford, and G. M. Mathews. 2000. Research related to transportation of\njuvenile salmonids on the Columbia and Snake Rivers, 1998. Report of the U.S.\nNational Marine Fisheries Service to the U.S. Army Corps of Engineers, Walla\nWalla District.\nMarsh, D. M., J. R. Harmon, N. N. Paasch, K. L. Thomas, K. W. McIntyre, B. P.\nSandford, and G. M. Matthews. 2001. Research related to transportation of\njuvenile salmonids on the Columbia and Snake Rivers, 2000. Report of the\nNational Marine Fisheries Service to the U.S. Army Corps of Engineers, Walla\nWalla, Washington.\nMaule, A.G., D. W. Rondorf, J. W. Beeman, and P. V. Haner. 1996. Incidence of\nRenibacterium salmoninarum infections in juvenile hatchery spring Chinook\nsalmon in the Columbia and Snake rivers. Journal of Aquatic Animal Health\n8:37-46.\nMcMichael, G.A., J. A. Vucelick, B. Ryan, R. L. McComad, S. Smith, M. Carper, D.\nElliott, T. Murray, L. Applegate, and C. McKibben. 2006. Alternate barging\nstrategies to improve survival of transported juvenile salmonids-2006. Annual\nReport, 2005 (Contract DACW57-00-D-0009 #11) prepared by Battelle-Pacific\nNorthwest Division, Richland, WA, for the U.S. Army Corps of Engineers, Walla\nWalla, WA. 40 p.\nMesa, M. G., T. P. Poe, A. G. Maule, and C. B. Schreck. 1998. Vulnerability to\npredation and physiological stress responses in juvenile Chinook salmon\n(Oncorhynchus tshawytscha) experimentally infected with Renibacterium\nsalmoninarum. Canadian Journal of Fisheries and Aquatic Sciences\n55:1599-1606.\nMotulsky, H. 1995. Intuitive Biostatistics. Oxford University Press, New York. 386 p.\nMuir, W. D., D. M. Marsh, B. P. Sandford, S. G. Smith, and J. G. Williams. 2006.\nPost-hydropower system delayed mortality of transported Snake River\nstream-type Chinook salmon: Unraveling the mystery. Transactions of the\nAmerican Fisheries Society 135:1523-1534.\nMunson, A.D., and K. A. Johnson. 2005. Integrated management of BKD using\nELISA-based culling. Bacterial Kidney Disease: Challenges for the 21st Century.\nSymposium, November 15-17, 2005, Seattle, WA.\n28","Pascho, R. J., and D. G. Elliott. 1989. Juvenile fish transportation: impact of bacterial\nkidney disease on survival of spring/summer chinook salmon stocks. Annual\nReport, 1988 (Contract E86880047) prepared by the U.S. Fish and Wildlife\nService, Seattle, WA, for the U.S. Army Corps of Engineers, Walla Walla, WA.\n319 p.\nPascho, R.J., D.G. Elliott, and S. Achord. 1993. Monitoring of the in-river migration of\nsmolts from two groups of spring Chinook salmon, Oncorhynchus tshawytscha\n(Walbaum), with different profiles of renibacterium salmoninarum infection.\nAquaculture and Fisheries Management 24:163-169.\nPascho, R.J., T. D. Goodrich, and C. L. McKibben. 1997. Evaluation by the enzyme-\nlinked immunosorbent assay (ELISA) of Renibacterium salmoninarum bacterins\naffected by persistence of bacterial antigen. Journal of Aquatic Animal Health\n9:99-107.\nPrentice, E. F., T. A. Flagg, and C. S. McCutcheon. 1990. Feasibility of using\nimplantable passive integrated transponder (PIT) tags in salmonids. American\nFisheries Society Symposium 7:317-322.\nRyan, B. A., M. C. Carper, B. P. Sandford, and G. M. Matthews. 2007. Detection of\npassive integrated transponder (PIT) tags on piscivorous bird colonies in the\nColumbia River Basin, 2006. Report of the National Marine Fisheries Service to\nthe U.S. Army Corps of Engineers, Walla Walla, Washington.\nRyan, B. A., J. W. Ferguson, R. D. Ledgerwood, and E. P. Nunnallee. 2001. Detection\nof passive integrated transponder tags from juvenile salmonids on piscivorous\nbird colonies in the Columbia River Basin. North American Journal of Fisheries\nManagement 21:417-421.\nRyan, B. A., J. H. Glabek, J. W. Ferguson, E. P. Nunnallee, and R. D. Ledgerwood.\n2002. Detection of passive integrated transponder (PIT) tags on piscivorous bird\ncolonies in the Columbia River Basin, 2000. Report of the National Marine\nFisheries Service to the U.S. Army Corps of Engineers, Walla Walla,\nWashington.\nRyan, B. A., S. G. Smith, J. M. Butzerin, and J. W. Ferguson. 2003. Relative\nvulnerability to avian predation of juvenile salmonids tagged with passive\nintegrated transponders in the Columbia River estuary, 1998-2000. Transactions\nof the American Fisheries Society 132:275-288.\nSchreck, C. B., M. D. Karnowski, and B. J. Clemens. 2005. Evaluation of postrelease\nlosses and barging strategies that minimize postrelease mortality. Final report of\nthe Oregon Cooperative Fish and Wildlife Research Unit, Oregon State Univ., to\nthe U.S. Army Corps of Engineers, Walla Walla District.\n29","Schrock, R. M., J. W. Beeman, D. W. Rondorf, and P. V. Haner. 1994. A microassay for\ngill sodium, potassium-activated ATPase in juvenile Pacific salmonids.\nTransactions of the American Fisheries Society 123:223-229.\nSolazzi, M.F., T.E. Nickelson, and S.L. Johnson. 1991. Survival, contribution, and\nreturn of hatchery coho Salmon (Oncorhynchus kisutch) released into freshwater,\nestuarine, and marine environments. Canadian Journal of Fisheries and Aquatic\nSciences 48:248-253.\nSpromberg, J. A., and L.L. Johnson. In press. Potential effects of freshwater and\nestuarine contaminant exposure on Lower Columbia River Chinook Salmon\n(Oncorhynchus tshawytscha) populations. Population-level Ecotoxicological\nRisk Assessment.\nVanderkooi, S.P., and A. G. Maule. 1999. Prevalence of Renibacterium salmoninarum\nin juvenile Chinook salmon at Columbia and Snake River hatcheries, 1993-1996.\nJournal of Aquatic Animal Health 11:162-169.\nWard, D. L., J. H. Petersen, and J.J. Loch. 1995. Index of predation on juvenile\nsalmonids by northern squawfish in the lower and middle Columbia River and\nLower Snake River. Transactions of the American Fisheries Society\n124:321-334.\nWilliams, J. G., and G. M. Matthews. 1995. A review of flow and survival relationships\nfor spring and summer Chinook salmon, Oncorhynchus-tshawytscha, from the\nSnake-River Basin. Fishery Bulletin 93(4):732-740.\nWilliams, J. G., S. G. Smith, R. W. Zabel, W. D. Muir, M.D. Scheuerell, B. P. Sandford,\nD. M. Marsh, R. McNatt, and S. Achord. 2005. Effects of the federal Columbia\nRiver Power system on salmon populations. NOAA Technical Memorandum\nNMFS-NWFSC-63\n30"]}