Re-Analysis of Hydroacoustic Fish-Passage Data from Bonneville Dam after Spill-Discharge Corrections iii
SummaryThe Portland District of the U.S. Army Corps of Engineers requested that the Pacific Northwest National Laboratory (PNNL) re-analyze four years of fixed-aspect hydroacoustic data after the District made adjustments to spill discharge estimates that may have been sufficient to alter previous results and conclusions. Results of the 2005 study also are presented in this report. This study supports the Portland District and its effort to maximize survival of juvenile salmon passing Bonneville Dam. Major passage routes through Bonneville Dam include 10 turbines and a sluiceway at Powerhouse 1 (B1), an 18-bay spillway, and eight turbines and a sluiceway at Powerhouse 2 (B2).In this report, we present new estimates of all major fish-passage metrics for study years 2000, 2001, 2002, and 2004, and reiterate estimates for 2005. No study was conducted in 2003. We also describe relations between spill and sluiceway effectiveness and percent passage by route versus percent flow through the same route. Routes of interest included the spillway, Powerhouse 1 (B1) turbines, Powerhouse 2 (B2) turbines, B1 sluiceway relative to B1, B2 Corner Collector (B2CC) relative to B2, and all non-turbine routes combined (i.e., fish-passage efficiency). For a detailed description of sampling methods by year, we refer readers to the original technical reports (Ploskey et al. 2002a(Ploskey et al. , 2002b(Ploskey et al. , 2003(Ploskey et al. , 2005(Ploskey et al. , and 2006. During the re-analysis, we took the opportunity to adjust spatial expansions of spillway-passed fish to compensate for reduced hydroacoustic detectability that occurs when spill discharge through individual bays increases, something that was done for the first time in the 2006 report on 2005 data.The original reports and all associated results, discussion, and conclusions for non flow-related metrics remain valid and useful, but effectiveness measures for study years 2000years , 2001years , 2002years , and 2004years as reported in Ploskey et al. (2002a should be superseded with the new estimates for spring (Table S.1) and summer (Table S.2). The only fish-passage metrics that changed more than 3% from original estimates were related to effectiveness. Re-analysis produced spill effectiveness estimates that ranged from 12% to 21% higher than previous estimates in spring and 16.7% to 27.5% higher in summer, but the mean spill effectiveness over all years was only slightly above 1:1 (1.17 for spring and 1.29 for summer). While re-analysis increased spill effectiveness, it decreased surface-passage effectiveness in the years that this metric was measured (by 10.1% in spring and 10.7% in summer of 2002 and 9.5% in spring and 10.2% in summer of 2004). Tight and non-overlapping 95% confidence limits suggest that original and revised estimates of spill-passage and surface-passage effectiveness were significantly different. Among the smallest changes were in Project ...