Automating procedures for the forecasting of terminal run and escapement of Chinook, coho and chum salmon stocks using open-source statistical software

The annual exercise of forecasting terminal run or escapement is a critical aspect of management and conservation of salmonids coastwide. This project involves the completion of an automated computer program (henceforth called “ForecastR”) relying on the open-source statistical software R to generate age-specific forecasts using a variety of generic models including (i) simple and complex sibling regressions with the ability to include environmental covariates, (ii) time series models such as ARIMA, exponential smoothing, and naïve models, and (iii) mechanistic models such as average return rate models that depend on auxiliary data such as the number of outmigrant juveniles, the number of hatchery fish released or the number of spawners.