Estimating Passage Abundance of Natural and Hatchery Fall Chinook Salmon Smolts at Lower Granite Dam

Since ESA-listing, fall Chinook salmon in the lower Snake River, Idaho have undergone marked increases in abundance from near-extirpation levels.  This increase has been associated with management efforts including reduced harvest, stable minimum spawning flows, summer flow augmentation, predator control, increased hatchery production and supplementation; improved dam passage structures, and summer spill operations.  Modeling the effects of those efforts on natural production requires estimates of passage abundance for natural and hatchery smolts at Lower Granite Dam while accounting for multiple data sources such as radio- and PIT-tagged fish detections and the counts of visual marks on fish recaptured in the sample tank at the dam. We employ a hierarchical Bayesian model to estimate the daily fractions of hatchery and natural juveniles at the dam using these various mark-recapture data.  Our Bayesian approach provided abundance estimates with measures of precision on both a daily and annual scale. These estimates will be incorporated into a two-stage life cycle model designed to assess the factors affecting changes in adult abundance.