Evaluation of factors affecting juvenile chinook salmon fish guidance efficiency (DRAFT)

Publication Date



Report of Research funded by U.S. Army Corps of Engineers and National Marine Fisheries Service


Army Corps of Engineers, barriers, behavior, chinook, Columbia River, day/night samples, design, fish behavior, fish guidance, fish movement, flume, Fyke nets, guidance, guidance efficiency, hydroelectric, hydroelectric dams, intake, intake screens, juvenile, mortality, porosity, predation, prototype, prototype test, salmon, salmonids, screens, turbines, upstream, water velocity


Juvenile salmonids that pass through turbines at hydroelectric dams have mortalities that range from 8-19%. As one means to decrease the number of fish that pass through turbines and thus decrease the mortality, the U.S. Army Corps of Engineers has installed screens in turbine intakes at most of their dams on the Snake and Columbia Rivers. Although turbine intake screens have successfully diverted a large percentage of fish from turbines at some dams, at others they have not. This study looked at three aspects related to turbine screens to provide insight into why screens may or may not work under different conditions. Firstly, fish behavioral studies were conducted in an oval flume at the NMFS Field Station at Pasco. The goal was to determine how changes in water velocities affected fish movement. Since turbine intake screens decrease water velocities, we hypothesized that in some cases fish avoided conditions created by the screens. For the study, fish were released, one at a time, into a moving body of water into which different porosity barriers were placed. Fish did not avoid barriers as long as water velocities upstream from them did not decrease too abruptly. However, when a barrier caused water velocities to decrease by approximately 10 cm/s over a distance of approximately 10 cm, fish avoided the area. Design of screens to divert fish should take this fish behavior into account. Secondly, we considered a number of physical factors that potentially affect FGE to look for correlations between the factors and FGE. However, there was too little data to conduct an analysis with most of the proposed factors. Sufficient data existed to evaluate the following six factors: (1) River temperature on the day of each FGE test, (2) turbidity (measured by Secchi disk) on the day of each FGE test, (3) test duration (number of hours), (4) number of fish (all species) collected during each FGE test, (5) the number of fish (all species) collected per hour of FGE test, and (6) the proportion of yearling chinook salmon in the total collection. No consistent correlations were found between fish guidance efficiency and any of these factors. It is unlikely that sufficient data will ever become available to predict changes in fish guidance related to changes in physical factors that may affect FGEs. Thirdly, a review of past FGE test designs was conducted to explain differences between prototype results and actual performance after the final installation of equipment. In the early years, the techniques used for prototype FGE measurements did not include fyke nets under the screens to recover unguided fish. In these cases, FGEs were often overestimated. In latter years, techniques for prototype tests and evaluations after final installation were the same. We speculate that changes in physical conditions at dams, or populations of predators may account for the differences found.

This document is currently not available here.