96-6 Using Release-Recapture Methods to Estimate Salmon Survival in a Complex Environment

Rebecca Buchanan , School of Aquatic & Fishery Sciences, University of Washington, Seattle, WA
The advent of small acoustic telemetry tags has greatly expanded our ability to track movements of juvenile salmonids as they migrate to the ocean.  We are now able to track tagged fish as they move through complex environments such as river deltas and estuaries, and can begin to address such questions as: What routes do salmon smolts take through deltas and estuaries?  Which route has the highest survival?  Which has the lowest?  What factors influence route selection?  Along with these questions arises the issue of the appropriate spatial scale for study and analysis.  Even with active tags, the spatial scale on which biological processes occur may not match the scale on which the data are collected.

The Vernalis Adaptive Management Program (VAMP) has used acoustic telemetry studies to address questions of route selection and survival for several years for juvenile Chinook salmon as they migrate through the San Joaquin River Delta in the Central Valley of California.  The Delta’s complex river structure requires a statistical release-recapture model that reflects the multiple routes that fish may take through the Delta.  A complicating factor in analysis of the VAMP data is the predation of tagged salmon smolts by predatory fish that then move past one or more detection arrays.  Without detections assumed to come from predators, survival estimates through the Delta were low in both 2009 and 2010, ranging from 1% to 10%.  The average probability of leaving the San Joaquin River for Old River was estimated to be 53% (SE = 2% to 3%) in both study years.