Preliminary Modelling for the Stock Assessment of Shortfin Mako Shark, Isurus oxyrinchus using CMSY and JABBA

The shortfin mako shark, Isurus oxyrinchus (SMA), is a highly migratory pelagic species found globally. It is particularly vulnerable as bycatch in longline fisheries, and has a vulnerable status according to the IUCN. SMA is considered a data-limited stock as there is incomplete catch information, limited information on the catch composition (size frequencies), and few abundance indices (e.g., standardised CPUE series). A preliminary stock assessment was performed by Brunel et al. in 2018 for the IOTC convention area using CMSY, a catch-only method, and a built-in Bayesian surplus production model (BSM), based on reconstructed catch data and standardised CPUEs from the EU longline fleet of Spain (2006-2016), and Portugal (2000-2016). This preliminary assessment found that SMA had been experiencing overfishing from the 1990s (F/Fmsy =2.57), but that the biomass of the stock was decreasing but not overfished (B2015/Bsmy close to 1). Here, we perform preliminary assessments using updated catch and CPUE indices for SMA. Contrary to Brunel et al, we use nominal catch of SMA (1964-2018) and scaled CPUEs from Japan (1993-2018), Spain (2001-2018), Taiwan (2005-2018), and Portugal (2000-2018). Catch ratio between SMA catch and target species catch as used in Brunel et al. were not available at the time of the stock assessment. We develop a demographic analysis based on Leslie Matrices to determine a prior for r (resilience, or intrinsic growth rate). We run JABBA (Winker et al. 2018) and CMSY, two surplus production models and compare these results to those of Brunel et al. 2018. We compare the outputs of CMSY to those of JABBA, providing background and advice in parameterising the model. We find that JABBA and CMSY give consistent results in terms of reference points, which indicate that SMA is experiencing overfishing (F/Fmsy well above 1) but is not overfished (most results indicate B/Bmsy > 1). As JABBA can take into account all CPUEs to inform the stock assessment, we define these model outputs as our best case scenario. The projections are run and provided for this model.