|Document Type:||Contract Report|
|Title:||Evaluation of juvenile salmonid condition (descaling) under different turbine operating conditions at McNary Dam, 2010|
|Author/Editor:||Gordon A. Axel, Michael H. Gessel, Eric E. Hockersmith, Matthew G. Nesbit, Benjamin P. Sandford|
|Publisher:||National Marine Fisheries Service|
|Contracting Agency:||U.S. Army Corps of Engineers. Walla Walla, Washington|
|Keywords:||descaling,dam passage,McNary Dam,turbine passage,|
The objective of this study was to test for significant differences in descaling rates of steelhead and yearling and subyearling Chinook, coho, and sockeye salmon exposed to ESBSs (extended length bar screens) and gatewells during two turbine operations. The two turbine operations compared were a "target" operating range, which was higher than 1% of peak operating efficiency (>1%, or 13,300-15,000 ft3/s) vs. the upper 1% of peak operational efficiency (1%, or 11,600-12,400 ft3/s).
Descaling rates at dams can be influenced by a number of factors such as stock differences, smoltification levels, and previous migration history as well as by external or environmental factors such as turbine unit location, turbine operating condition, debris load, and experimental handling. To provide comparable conditions for comparison of descaling rates between two operational treatments, we measured descaling simultaneously in two turbine unit intake slots (4A and 5A). During testing, one unit was operated within the "target" operating range (> 1% of peak efficiency), while the other was operated within the upper 1% of peak efficiency. To account for a possible unit effect, we switched operational treatments between units every other night. Therefore, each 2 d, two-unit "block" resulted in a paired test of descaling rates between the two operating conditions.
With data from these tests, we modeled descaling through time using logistic regression. Factors included in the model were turbine unit, operational treatment, date, and head differential. We also examined two-way interactions between these four variables. We used quasi-likelihood Akaike information criterion (QAIC) to rank the models. Models differing by less than 2.0 from the best-fitting model were averaged across predicted values using the respective Akaike weights. Candidate models ranged from the most complex, with all factors and all two-way interactions, to the most simple, with only unit as a factor explaining differences in descaling. Since unit was essentially a nuisance factor in this analysis, we included it in all models.
|Theme:||Recovery and rebuilding of marine and coastal species|
Describe the relationships between human activities and species recovery, rebuilding and sustainability.