Assessing the performance of an in-river backwards run reconstruction of Fraser River sockeye under biological uncertainty.

This project is the first to assess reliability of run reconstruction methods for estimating sockeye salmon abundance and exploitation rates in the Fraser River. Run reconstruction is a model-based approach to estimating fishery harvest rates and catch of up to 35 individual sockeye salmon stocks. Fishery managers typically require these estimates to design pre-season fishing plans aimed at meeting stock-specific escapement and catch objectives defined under Canada’s Wild Salmon Policy, which stipulates management of individual sockeye stock (i.e., conservation units). This project will test the reliability of run reconstruction under certainty through a process referred to as simulation-estimation, in which a stochastic simulation model of sockeye migration and fisheries represents the “truth” that a run reconstruction attempts to estimate. We will use repeated simulation-estimation cycles to evaluate reconstruction performance by comparing exploitation and abundance estimates to known values from the simulator. The intern will partner with LGL Limited to develop and test a run reconstruction of Fraser River sockeye from 2002-2008. LGL has extensive run reconstruction experience, and consequently is interested in model development and performance testing.

Faculty Supervisor:

Dr. Sean Cox

Student:

Cameron Noble

Partner:

LGL Limted

Discipline:

Environmental sciences

Sector:

Fisheries and wildlife

University:

Simon Fraser University