Analysing potential inputs to the 2025 stock assessment of Western and Central Pacific oceanic whitetip shark (Carcharhinus longimanus)

Oceanic whitetip sharks (Carcharhinus longimanus) are caught as bycatch in longline and purse seine fisheries in the Western and Central Pacific Ocean (WCPO). Previous assessments indicated the stock was overfished and overfishing was occurring, and the recent assessment highlighted issues associated with some of the inputs.

This report documents Phase 1 of the current project which investigates the potential methods for catch reconstruction, CPUE and length compositions for use in the stock assessment for oceanic whitetip shark in 2025 (Phase 2). This study used model developments made in the context of other shark assessments to develop a set of potential inputs for the upcoming assessment.

Different subsets of longline CPUE observer data showed similar trends, with the exception of the distant water fleet index. Overall, the indices showed a consistent decline from a high in 1999 that flaĴens around 2005. The indices remained fairly low post-2005, but seemed to increase slightly in the most recent years. Purse seine CPUE indices showed different standardisation effects between set-types, but both free-school and associated indices showed an increase in CPUE in recent years (2009–2020). These indices were corrected for climatic conditions expressed by the NINA4 index.

Predicting longline interactions across the WCPO indicated that the total predicted interactions have declined since 1999, a trend that is consistent with previous analyses. Overall, for both purse seine fisheries, the predicted total interactions declined since the late 1990s and have remained comparably low since then.

Length compositions trends were inconsistent among years and months for models associated with catch and CPUE. However, area effects in longline length data were observed, with smaller fish in the equatorial regions and larger fish at higher latitudes.

The following recommendations are made:

• There are likely to be sufficient data and a sufficiently consistent signal in the different datasets, especially from longline, to conduct a stock assessment.

• We suggest that a fully integrated assessment could be aĴempted, based on the consistency of datasets developed herein. We suggest that our updated analyses of input data should be compared with previous assessment models in a stepwise fashion.

• As year effects are relatively minor in the longline fisheries, future catch reconstruction attempts could extrapolate interactions further back in time to avoid complications with assuming or estimating initial fishing mortalities in the assessments.

• Alternative assessment methods (for example surplus production assessments) should be run in parallel with an integrated assessment. Length-based or hybrid length-based spatial assessments provide an alternative approach that is independent of recent longline data, and allows for multi-model inference that can strengthen conclusions and potential management advice from an integrated approach.