A Bayesian Analysis of Daily Weir Counts for Estimating Chinook Salmon Escapement

• Activity Number: 645
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics and the Environment
• Author(s): Margaret Short*+ and Jim Jasper
• Companies: University of Alaska, Fairbanks and Alaska Department of Fish and Game
• Keywords: salmon escapement ; weir ; Bayesian model ; Markov chain Monte Carlo

Abstract

Estimation of salmon escapement is important in managing this vital component of the Alaskan economy. Fish that ``escape'' the fishery upstream to spawn are counted as they pass a weir (barrier) on their way upstream. On days when the weir is operational, the fish may spend several days before passing; on days when it is inoperational, they pass freely. We model the daily counts, incorporating missingness due to breaches in the weir, using a Bayesian hierarchical model, fitted using Markov chain Monte Carlo. Model fitting presents a substantial challenge and we discuss it at length. We illustrate using Kogrukluk River data for the years 1976-2007. This represents joint work with Jim Jasper, at the Alaska Department of Fish and Game.