Application of explicit precautionary principles in data-limited fisheries management

Many management bodies require applying the precautionary approach when managing marine fisheries resources to achieve sustainability and avoid exceeding limits. For data-limited stocks, however, defining and achieving management objectives can be difficult. Management procedures can be optimized towards specific management objectives with genetic algorithms. We explored the feasibility of including an objective that limited the risk of a stock falling below various limit reference points in the optimization routine for an empirical data-limited control rule that uses a biomass index, mean catch length, and includes constraints (the "rfb-rule"). This was tested through management strategy evaluation on several fish stocks representing various life-history traits. We show that risk objectives could be met, but more restrictive risk limits can lead to a potential loss of yield. Outcomes were sensitive to simulation conditions such as observation uncertainty, which can be highly uncertain in data-limited situations. The rfb-rule outperforms the method currently applied by ICES, particularly when risk limitation objectives are considered. We conclude that the application of explicit precautionary levels is useful to avoid overfishing. However, we caution against the indiscriminate use of arbitrary risk limits without scientific evaluation to analyse their impact on stock yields and sustainability.