A Method for Asking Experts about the Biological Status of Salmon Conservation Units in British Columbia

As with most fisheries, the assessment and management of salmon fisheries are highly dependent on the use of multiple indicators, yet methods for combining indicators have generally not been well developed or tested for consistency. This situation is exemplified well by the implementation of Fisheries and Oceans Canada’s (DFO’s) Wild Salmon Policy which requires the annual assessment of the biological status of Conservation Units (CUs) by combining the status (red, amber or green) of several indicators or metrics (spawner abundance, trends in spawner abundance, harvest rate, and spatial distribution). Existing methods for aggregating such metrics into an overall CU status assume either that all metrics are equally important or that they have some other pre-determined weighting. We developed a choice modeling questionnaire to quantitatively determine (1) the status of CUs, and (2) the relative importance of the various components contributing to that overall assessment. The questionnaire presented experts across British Columbia with a series of hypothetical CU scenarios, each one simultaneously presenting the biological metrics, the metric status, and the data quality and amount. The experts were asked to (1) rate the biological status of each CU scenario, and (2) identify which combination (of metric/metric status/data quality and amount) in each CU scenario pulled their CU status rating most towards a red or green rating. Metric status, regardless of the type of metric, was the single most important component contributing to the experts' assessment of CU status. Yet not surprisingly, considering the historical emphasis by salmon fisheries management agencies on spawner abundance estimates, we found the spawner abundance metric to be more important than the other three metrics. Data quality and amount became statistically significant only as an interaction effect with metric status. When data quality and amount was high, a red or green metric status resulted in a more extreme (more red or green) CU status rating. Overall, this more holistic quantitative evaluation of expert opinion may provide more consistent CU ratings across experts involved in such assessments. While not directly applicable to all CUs, the project is a case study of a novel method in fisheries management (use of choice modeling), which in the future could be modified for different regional or species contexts.