Research Document - 2015/075

Hierarchical Bayesian modelling of Atlantic Salmon egg to smolt time series from monitored rivers of eastern Canada to define and transport reference points

By G. Chaput, E. Prévost, J.B. Dempson, M. Dionne, R. Jones, A. Levy, M. Robertson, and G. Veinott

Abstract

This paper presents analyses of the available time series of Atlantic salmon (Salmo salar) spawner to smolt data from fourteen monitored rivers in eastern Canada. A freshwater life history model is presented using a hierarchical Bayesian modelling framework to estimate and transfer reference points for Atlantic salmon. The results show that the stock and recruitment dynamic of Atlantic salmon within the freshwater portion of the life cycle is highly variable within and among rivers with consequence that the stock and recruitment parameters are uncertain. These uncertainties propagate into the estimation of reference points for management. Models examined for transport of reference points included three potential covariates; presence of lacustrine habitat, mean age of smolts, and proportion of the egg depositions coming from multi-sea-winter (MSW) salmon. Differences in the freshwater carrying capacity of salmon rivers attributed to the presence of lacustrine habitat which is used by salmon juveniles for rearing is confirmed. Density-independent survival rate is estimated to be higher for rivers with older smolt ages as well as for stocks with increasing proportions of the egg depositions from MSW salmon. The transfer of reference points from data-rich stocks to data-poor stocks, which are the great majority of stocks in eastern Canada, poses the greatest challenge. Analyses using hierarchical Bayesian frameworks are the favored approach for analyzing multiple stock and recruitment data sets and to address data-poor situations.

Complete document (PDF, 9760 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.