Research Document - 2002/106

An evaluation of a recruitment forecasting procedure for Strait of Georgia herring

By R.W. Tanasichuk


This report describes an evaluation of an extension of the recruitment forecasting procedure which has been used since 1987 to forecast recruitment (number of age 3 spawners) for the West Coast Vancouver Island (WCVI) stock of Pacific herring (Clupea pallasi) (Ware and Tanasichuk 1987). The extension is to forecast recruitment for Strait of Georgia herring. The forecasting procedure is based on the age composition of samples of herring trawled along the southwest coast of Vancouver Island during summer fisheries oceanography surveys.

There are two components of the forecasting methodology. The first is a regression equation. It was developed to forecast the proportion of age 3 fish estimated to be in the prefishery biomass based on the proportion of age 2+ fish in the samples trawled the previous summer. Results of geometric mean regression analyses showed a significant linear relationship with parameter estimates that are stable over time. Residual analysis showed no time trend in the residuals, nor any effect of sampling time or the magnitude of the forecast. A retrospective analysis over 1995 - 2002 showed that observed proportions at age 3 were within the 95% confidence interval of the forecast for all 8 years. On average, the forecasted proportion of age 3 fish was 0.02 +/- 0.084 ( mean +/- 2 SE) less than observed proportion.

The second part consists of forecasting the recruitment category using the forecasted proportion age 3 and the assessment model's forecast of returning adults. By convention, recruitment is categorised as Poor, Average or Good based on the cumulative frequency distributions of the time series of number of recruits estimated by the assessment model. Also by convention, recruitment is generally anticipated to be Average. Results of a retrospective analysis showed that the forecast was accurate for 6 of the 8 years. The convention of assuming Average recruitment would have been accurate in 2 of the 8 years.

 Complete PDF document
26 pages (163K)

Accessibility Notice:

This document is available in PDF format. If the following 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.

Active offer to produce in both official languages:

This report uses scientific and technical terms and is published in the official language of the working group or scientific expert that produced the document. If this document is not accessible to you in the official language of your choice, please contact the Secretariat.

Date modified: