Research Document - 2015/048

Estimates of a Biologically-Based Spawning Goal and Biological Benchmarks for the Canadian-Origin Taku River Coho Stock Aggregate

By Gottfried Pestal and Sandy Johnston

Abstract

This paper establishes a biological frame of reference for spawner abundances of the Coho Salmon (Oncorhynchus kisutch) stock aggregate originating in the Taku River in northwestern British Columbia. This transboundary stock aggregate is managed cooperatively by Canada and the US under the Pacific Salmon Treaty.

The project focused on fitting alternative Spawner-Recruit (SR) models and estimating biological benchmarks for each model. A long time series of good-quality data are available for Taku River Coho Salmon at the aggregate level including: spawner estimates based on a consistent mark-recapture survey covering the period 1987-2013; recruitment estimates for the 1987 to 2009 brood years based on catch and escapement data and coded-wire tag recoveries in Alaskan and Canadian fisheries and surveys; and annual age and sex composition estimates.

Estimates of biological benchmarks for the Taku River Coho Salmon aggregate were robust to alternative assumptions. For example, estimates of the number of spawners that maximizes sustainable yield under long-term average conditions (S_MSY) are remarkably consistent across 3 variations of the Ricker SR model for the base-case data set and all available years of data (1987 -2009 brood years). Median estimates range from 62,000 to 79,000 spawners, differing by less than 30% from each other. Based on statistical and practical considerations, the DFO’s regional peer review process recommended one of the model-data combinations as the most appropriate basis for advice regarding management goals for Taku River Coho Salmon: the Ricker AR1 model, which corrects for observed time-series patterns in residuals, fitted to estimates of total spawners and total adult recruits based on the age composition in the Canyon Island survey.

In addition, we present various reference points derived from the SR model fits to address policy considerations under Canada’s Wild Salmon Policy (e.g. upper and lower benchmarks for the relative abundance metric in a status assessment) and Alaska’s Sustainable Salmon Policy (e.g. range of spawner abundance with given probability of achieving some specified proportion of maximum sustainable yield).

As a consistency check, we also used 2 methods recently applied in data-poor settings: one based on percentiles of observed spawner abundance; and the other based on average smolt abundance observed for large brood years. These are more widely applicable and account for the majority of spawning goals for Pacific Salmon currently in use by DFO and ADFG. The percentile method produced a cautionary approximation of spawners that maximizes sustainable yield (S_MSY), but the smolt capacity method was highly sensitive to the productivity assumption, giving approximate S_MSY estimates that were either much higher or much lower than the SR-based estimates.

Complete document (PDF, 4,011 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.