How do we determine sustainable levels of harvest for fish stocks that are data-limited?
Approximately 80% of fisheries around the world are unassessed. They have limited information about their population status and trends, making it difficult to manage these stocks sustainably. Often the only data available for many of these species are time series of catch (i.e. tonnes of fish removed by fisheries per year).
Several mathematical models have been developed to assess stock status of data-limited fisheries using only this catch data and basic life history parameters. Not surprisingly, these models often produce inaccurate or biased estimates.
I identified whether we can use these catch-only fisheries stock assessment methods can be used to set robust, sustainable exploitation strategies for data-limited fisheries. I also assessed the value of new information for improving the selection of management strategies for these fisheries.
This work was in collaboration with an international team of fisheries scientists, led by Dr. Andrew Cooper (Simon Fraser University), Dr. Andrew Rosenberg (Environmental Defense Fund) and Dr. Liz Selig (Conservation International). The project was funded by the Gordon and Betty Moore Foundation and managed through Conservation International. During this time I was part of the Quantitative Fisheries Research Group, School of Resource and Environmental Management, Simon Fraser University.
The catch-only stock assessment methods are described in the FAO report (Rosenberg et al. 2014).
Publications and resources
Walsh JC et al. 2018. Trade-offs for data-limited fisheries when using harvest strategies based on catch-only models. Fish and Fisheries 19:1130–1146.
Anderson SC et al. 2017. Improving estimates of population status and trajectory with superensemble models. Fish and Fisheries 18:732–741.
Rosenberg AA et al. 2017. Applying a new ensemble approach to estimating stock status of marine fisheries around the world. Conservation Letters 11:e12363 (open access).
datalimited Github repository – R package to run 4 catch-only models and the superensemble