AI for better stock assessment – An innovative update in the CMSY methodology promises a more accurate assessment of fish stocks. Experts confirm that this new approach offers a realistic perspective on sustainable catches.
The international team of researchers that developed the CMSY++ published the results in Acta Ichthyologica et Piscatoria. The study showed that the improved model offers a more accurate perspective on maximum sustainable yield over the long term, as it takes into account stable environmental conditions.
With the help of an artificial neural network, trained with catch and biomass data from 400 fish stocks, the CMSY++ now allows managers and scientists to accurately estimate how much fish remain in a given stock and how much fishing effort can be applied, based solely on catch data.
The concept of Maximum Sustainable Catch Yield (MSY), devised in the 1950s by MB Schaefer, proposes that preserving at least 50 per cent of the unexploited biomass is essential to ensure maximum long-term yield.
Dr Rainer Froese, lead author of the study and senior researcher at the GEOMAR Helmholtz Centre for Ocean Research, points out that the MSY++ fits the observed data better than traditional models, which often overestimate catch capacity after excessive fishing periods.
Daniel Pauly, co-author of the study and principal investigator of the Sea Around Us initiative at the University of British Columbia, adds that ‘gold standard’ models underestimate the biomass needed to achieve sustainable yields. This finding may explain the difficulties in restoring depleted stocks despite adopting the predictions of such models.
Ultimately, the CMSY++ stands to be a breakthrough in fish stock monitoring, offering a more accurate and realistic perspective on sustainable catches, crucial for responsible management of marine resources.
AI for better stock assessment