Comparative Performance of Acoustic-Tagged and Passive Integrated Transponder-Tagged Juvenile Salmon in the Columbia and Snake Rivers, 2008
Total Page:16
File Type:pdf, Size:1020Kb
Comparative performance of acoustic-tagged and passive integrated transponder-tagged juvenile salmon in the Columbia and Snake Rivers, 2008 A. Michelle Wargo Rub, Benjamin P. Sandford, Lyle G. Gilbreath, Mark S. Myers, Mark E. Peterson, Lila L. Charlton, Steven G. Smith, and Gene M. Matthews July 2011 Comparative Performance of Acoustic-Tagged and Passive Integrated Transponder-Tagged Juvenile Chinook Salmon in the Columbia and Snake Rivers, 2008 A. Michelle Wargo-Rub, Benjamin P. Sandford, Lyle G. Gilbreath, Mark S. Myers, Mark E. Peterson, Lila L. Charlton, Steven G. Smith, and Gene M. Matthews Report of research by Fish Ecology Division, Northwest Fisheries Science Center National Marine Fisheries Service National Oceanic and Atmospheric Administration 2725 Montlake Blvd. East Seattle, Washington 98112 for Environmental Resources Branch, Planning and Engineering Division Portland District, U.S. Army Corps of Engineers Robert Duncan Plaza 333 S.W. 1st Avenue Portland, Oregon 97208-2946 Contract W66QKZ60441152 July 2011 ii EXECUTIVE SUMMARY Evaluation of Acoustic Tags in Migrating Yearling Chinook Salmon Migration rates, detection and survival probabilities, and avian predation rates were compared between fish tagged with a passive integrated transponder (PIT) tag vs. those tagged with both a PIT-tag and Juvenile Salmonid Acoustic Telemetry System (JSATS) tag. During spring 2008, we collected migrating hatchery yearling Chinook salmon at Lower Granite Dam. We tagged 4,139 of these fish with both a JSATS tag and a PIT tag (JSATS-tagged fish) and 50,814 with a PIT tag only (PIT-tagged fish). Samples were designed to be of sufficient size to determine a minimum difference of 5% between tag groups in detection and survival over a distance of 348 km, and to provide statistical power of 80% (α = 0.05). Fish were released to the tailrace of Lower Granite Dam on 10 d, from 24 April through 17 May. Acoustic-tagged fish were implanted with the 2008 model JSATS acoustic tag, which weighed 0.42 g in air. Average tag burden experienced by JSATS-tagged fish was 2.3% of body weight (range 0.8-7.2%). For both tag treatments, travel times, detection probabilities, and survival were estimated from individual PIT-tag detections at Little Goose, Lower Monumental, McNary, John Day, and Bonneville Dam. For estimates of detection probability and survival, we also utilized detections of JSATS-tagged fish from acoustic arrays at multiple locations within the study area. Mean detection probabilities were estimated for each PIT-tag detection site. Mean detection probability was higher for JSATS- than PIT-tagged fish at Little Goose, Lower Monumental, and John Day Dams. The difference in detection probability was 0.04 (P = 0.005) at Little Goose Dam, 0.05 (P = 0.002) at Lower Monumental, and 0.06 (P = 0.006) at John Day Dam. Mean detection probability at Ice Harbor Dam was 0.03 higher for JSATS- than PIT-tagged fish, and the difference approached significance (P = 0.067). There was no significant difference in detection probability between tag-treatment groups at McNary or Bonneville Dam (P = 0.242, and 0.174, respectively). In the Snake River, relative survival (ratio of survival estimates for JSATS-tagged/PIT-tagged groups) was not significantly different than one from release to Little Goose or Ice Harbor Dam (P = 0.107 and 0.336 respectively). Relative survival was 0.95 from release to Lower Monumental Dam and approached significance (P = 0.096). In the Columbia River, relative survival was 0.91 (P = 0.095) to McNary Dam, 0.72 (P = 0.001) to John Day Dam, and 0.69 (P = 0.021) to Bonneville Dam. A significant difference in travel time to John Day Dam was observed (P = 0.019), with JSATS-tagged fish arriving 0.81 d (19.44 h) after PIT-tagged fish, but significant differences in travel time were not observed at any other detection site. iii Overall mean PIT-tag recovery from upper river bird colonies was 2.0% for JSATS-tagged and 1.0% for PIT-tagged fish. Although this 1% difference in tag-recovery rate was statistically significant (P = 0.016), it was not likely to have been biologically meaningful. From estuarine bird colonies, the overall mean PIT-tag recovery was rate 3.0% for JSATS-tagged and 4.0% for PIT-tagged fish, and the difference was not significant (P = 0.881). Gross Necropsy and Histological Evaluation of Migrating Yearling Chinook Salmon To provide insight into the mechanism responsible for any tag effects observed, we subsampled study fish at two downstream sites for necropsy and histological evaluation. Up to 10 yearling Chinook salmon from each tag treatment and each release group were recaptured during migration using the separation-by-code (SbyC) systems in the juvenile bypass facilities at McNary and Bonneville Dams. Midway through the study, sampling at Bonneville Dam was discontinued due to extremely high flows, which resulted in high debris loads on the fish guidance screens. We resumed sampling for the Bonneville target groups at John Day Dam within approximately 48 h of this disruption. Respective subsamples from JSATS- and PIT-tagged release groups totaled 98 and 92 at McNary Dam, 29 and 67 at Bonneville Dam, and 57 and 53 at John Day Dam. John Day and Bonneville fish were combined in a single group for analysis. Recaptured fish were euthanized and examined for tag loss, disease, and histological change due to tag implantation. Kidney tissue samples were also collected and examined for the antigen to Renibacterium salmoninarum (Rs), the causative agent of bacterial kidney disease (BKD). A group of 100 non-tagged reference fish was used to provide baseline data for comparisons of gross necropsy, histological evaluation, and assessments of Rs in JSATS- and PIT-tag treatment fish. Reference fish were hatchery yearling Chinook collected at Lower Granite Dam for evaluations of migrating fish. The majority of fish subsampled at McNary Dam from both tag treatments were described as having either opaque, frayed, or missing fins (79% of PIT-tagged fish and 90% of JSATS-tagged fish). In contrast, a large majority of reference fish, and fish subsampled further downstream at Bonneville Dam, were described as having normal fins. Additionally, 13% of JSATS-tagged fish collected at McNary Dam were described as having eyes that were hemorrhagic. By comparison, 99% of the Lower Granite reference fish were described as having normal eyes. These observations indicated that fish might have experienced a trauma sometime between release and recapture. We also noted that in the McNary subsamples, the percentage of fish with hemorrhagic eyes was higher in JSATS-tagged than in PIT-tagged fish, and the difference between treatment groups was significant (P = 0.003). iv For both tag treatments, gross necropsy revealed less caecal and mesenteric fat in fish collected at McNary and Bonneville Dam than in reference fish, although these metrics were rated similarly between treatment groups at both locations. Splenic engorgement/enlargement was more prevalent in Lower Granite reference fish than in tag-treatment fish of either type subsampled at either McNary or Bonneville Dam. Enlarged spleens were observed in a higher percentage of fish collected at McNary than at Bonneville Dam. In the Bonneville subsamples, the percentage of fish observed with food in the stomach was higher for PIT-tagged than JSATS-tagged fish, and the difference between treatment groups was significant (P = 0.038). However, there was no difference between treatments for this metric in the McNary subsamples. Greater than 99% of all reference fish and SbyC subsampled fish were rated as having normal kidneys on gross exam. Liver abnormalities were seen in both reference and SbyC fish, and were of similar prevalence among treatment groups. Comparative histopathology metrics varied by recapture site and were mixed with respect to nutritional indicators, with some being higher in JSATS-tagged and others in PIT-tagged fish. However, histological indicators of inflammation and healing showed a consistent pattern of higher inflammation and slower healing in JSATS-tagged fish. For example, chronic inflammation within the mesentery was more prevalent in JSATS-tagged than in PIT-tagged fish subsampled at both McNary and Bonneville Dam (P = 0.001), and when present in both treatment groups, it was rated as more severe in JSATS-tagged fish at both recapture sites (P < 0.001). In Bonneville subsamples, chronic peritonitis was higher in JSATS-tagged than PIT-tagged fish (P = 0.001), and poor apposition of the incision was more common (P = 0.006), as were internal adhesions at the incision site (P = 0.063). Microscopic evidence of the incision/injection site was less prevalent in PIT-tagged than JSATS-tagged fish from both McNary and Bonneville Dams (P = 0.001), and the epidermis was more often observed to be "retracted" away from the incision/injection site in the JSATS-tagged fish from Bonneville Dam (P = 0.015). Rs antigen levels were evaluated using enzyme-linked immunosorbent assay (ELISA). For hatchery yearling Chinook, the range of Rs antigen levels was 0.065-0.335 in reference fish (one outlier at 1.11), 0.069-0.289 in both JSATS and PIT treatments recaptured at McNary Dam, and 0.076-0.540 in both JSATS and PIT treatments recaptured at Bonneville Dam. Since ELISA values were considered low for all but two fish recaptured at McNary Dam (1 JSATS and 1 PIT), no statistical analyses were conducted to evaluate differences between treatments at this location. At Bonneville Dam, 13% of JSATS-tagged fish had ELISA values considered moderate, compared to 11% of PIT-tagged fish. However, there was no statistical difference in these values based on a Kruskal-Wallis test (P = 0.830; a nonparametric equivalent of analysis of variance).