Incorporating parameter uncertainty into evaluation of spawning habitat limitations on Chinook salmon (Oncorhynchus tshawytscha) populations

Incorporating parameter uncertainty into a Monte Carlo procedure for estimating spawning habitat capacity helped determine that spawning habitat availability is unlikely to limit recovery of six populations of Chinook salmon (Oncorhynchus tshawytscha) in Puget Sound. Spawner capacity estimates spanned up to four orders of magnitude, yet there was virtually no overlap of distributions of capacity estimates with distributions of current spawner abundance (<0.2% overlap), except for the Suiattle River population (51% overlap). Empirical distributions of input parameters contained several important sources of uncertainty, insuring reasonably wide distributions of capacity estimates. The most defensible ranges of input parameters tended to produce conservative capacity estimates, indicating that increased model accuracy would only strengthen our conclusion that spawning habitat is not a constraint on these populations. There are insufficient data with which to develop parameter distributions that better represent historical capacity, which would certainly be higher than our estimates. Our results suggest that factors other than spawning capacity limit population size and that recovery efforts for Skagit River Chinook salmon need not focus on spawning habitat restoration.