Matthieu Authier (Research Engineer from the Delmoges Project – CIBBRiNA)
Abstract
Managing human activities, which can result in additional mortality on many marine Protected, Endangered or Threatened Species (PETS) is key to reach the ambitious set out by the EU 2030 Biodiversity Strategy. By-catch, the undesirable and non-intentional catch of non-target species in marine fisheries, is one of the main causes of mortality of marine mammals (which are often PETS) worldwide. Data on anthropogenic removals (including by-catch) of PETS (including marine mammals) are often unreliable because of, among others, inadequate sampling design, lack of enforcement, non-representative samples and undereporting (due for example to social desirability bias). All cetacean species benefit from some legal protection whether at the national or international level. The common dolphin (\textit{Delphinus delphis}) in the Bay of Biscay epitomized the current challenges : the failure to enforce its strict protection earned (i) France an infringement procedure from the European Commission ; and (ii) the French Government the ordnance from the highest administrative court (‘The Conseil d’Etat’) of a one month spatio-temporal closure of all high-risk fisheries operating in the Bay of Biscay in the winter 2024.
Managing by-catch hinges on the computation of so-called biological reference points, also known as removals limits/thresholds : these represent an upper limit to the number of animals that can be removed from a population without compromising the long-term viability of said population with unacceptably high probability. Methods to compute removals limits for cetaceans originate from scientific work carried out in the 1990s by the International Whaling Commission (IWC) whereby computer simulations are harnessed to investigate the likelihood outcomes of different management schemes. The framework outlined by the IWC used so-called ‘harvest control rule’ (or just control rule) that take data from current monitoring as inputs to output a threshold (or a quota in case of a commercial species). Importantly, the framework allows to assess the effect of knowledge gaps and data biases to devise control rules that are robust against these. Two rules are commonly in use : the Potential Biological Removal (PBR) from the US Marine Mammal Protection Act ; and the Removals Limit Algorithm, a child of the Catch Limit Algorithm devised to set quotas on the hunt of large whales.
We developed a stochastic surplus production model, a kind of parameter-lean model common in fishery sciences, and proposed a new control rule derived from this 4-parameters model. The model assumes (i) a simple proportional relationship between true abundance and removals, and (ii) stationarity (time-invariance) in removal rate. This assumption is untenable if management is to be effective as the very purpose managing anthropogenic removals of PETS is to minimize them. To preserve parameter-leaness, we resorted to a weighted-likehood approach for estimation (in a Bayesian framework) with time-dependent weights chosen such that older removal data are progressively and smoothly down-weighted. Using simulations, we benchmarked our new control rule relying on the stochastic surplus production model in a case study which revealed the competitiveness of the new rule to meet current conservation policy desiderata such as minimizing removals over time.
