Assessment of blue shark in the southwestern Pacific

This paper presents the 2016 stock assessment of blue shark (Prionace glauca) covering the southern hemisphere component of the Western and Central Pacific Fisheries Commission Convention Area (WCPFC-CA) and fisheries for the period 1994–2014. This represents the first attempt to assess this stock, noting that the species has previously been assessed within the North Pacific (Kleiber et al., 2009; ISC, 2014; Rice et al., 2014). This assessment relies on the stock assessment software MULTIFAN-CL (Fournier et al., 1998; Hampton and Fournier, 2001; Kleiber et al., 2014), which fits size-based, age- and spatially-structured population models to data from multiple sources.
A number of challenges were experienced in the development of this assessment. While the stock of blue shark within the southern WCPFC-CA may be considered ‘data rich’ for a shark species, it is considered data poor in comparison to assessments performed for tuna and most billfish. Catch data are generally of poor quality and have to be reconstructed (see Tremblay-Boyer and Takeuchi, 2016), such that both catch inputs and the resulting CPUE time-series are uncertain. A major objective of this assessment is therefore to establish and examine key areas of uncertainty, and the impacts on estimates of stock status.
An initial grid of 36 model runs across four different axes of structural uncertainty was examined. These used two assumed catch time series, three CPUE time series, three assumptions of the stock recruitment relationship steepness parameter, and two alternative penalties for deviations from the stock recruitment relationship. Comparison of the objective function indicated that uncertainty in the catch time series estimates was influential for the results. There was also a weak tendency for support of higher steepness.
Stock assessments conducted using MULTIFAN-CL and other similar methods usually report stock status relative to the spawning biomass and fishing mortality at maximum sustainable yield (MSY) and spawning biomass relative to unfished levels (so-called depletion estimates). In both cases, predictions of equilibrium recruitment are required from an assumed or estimated stock recruitment relationship (SRR). Early attempts to estimate the SRR, even when the steepness parameter was specified, resulted in very large estimates of unexploited equilibrium recruitment and spawning biomass that were considered to be unrealistic. Therefore, we do not present any results in this assessment that were dependent on SRR estimates. Instead, we computed estimates of unexploited spawning biomass and depletion based on the estimated recruitments without adjustment for the SRR. No estimates of MSY-related quantities were possible under these circumstances.
For model runs conditioned on the Pacific-wide CPUE-based catch estimation, biomass declines moderately over the period of the assessment; however, for the runs conditioned on the observer-based blue shark/general shark ratio catch estimation, biomass is stable in the first half of the time series and tends to increase thereafter. For all of the runs, spawning biomass depletion is estimated to be 0.08 to 0.10, inferring very strong impacts of fishing from unexploited conditions. However, these impacts have been fairly stable over the period of the assessment. Recruitment is variable from year to year for the Pacific-wide CPUE-based catch estimation runs, but is higher in the second half of the time series for the observer-based ratio catch estimation runs.
Given the encountered issues with available data, generally poor fits to CPUE time series by the model, and uncertainty in the estimated stock recruitment relationship, we view this assessment as a work-in-progress. We do not recommend that the derived stock status estimates be used as the basis for management advice at this time.
Based on the uncertainties in data inputs, recommendations for future work and assessments are as follows:
•an investigation into catch, effort and length data prior to 1994 should be undertaken, particularly for the high seas driftnet fishery that was active in the South Pacific until the early 1990s;
•future catch reconstructions should utilise data sources additional to observer data, such as
trade data;
• SC should develop a strategy to ensure observer deployments are spatially and temporally representative of fishing effort; to inform this strategy, it is recommended that spatial and temporal observer coverage for each fleet is reported and evaluated in an annual observer data gaps paper;
•pending availability of resources, SPC should undertake a systematic, fleet-specific evaluation of the nature and extent of these data quality issues as a specific project and report the findings to SC13;
•pending availability of resources, SPC should undertake an analysis of the statistical power of WCPO observer coverage configurations to detect changes in spatio-temporal abundance of bycatch species;
•the use of Electronic Monitoring and Electronic Reporting approaches should be pursued to
supplement observer coverage;
•approaches to ensure the full enumeration of key shark catches in logsheets (both retained and discarded) should be pursued;
•an investigation should occur into whether there is a size bias in the individual lengths recorded by the observers;
•observers should report the sex of sharks for all fish handled;
• the project “Develop proposed target and limit reference points for elasmobranchs” specified as part of the 2015 Shark Research Plan should continue to be prioritized so that planned stock assessments for sharks in the WCPO be effectively translated into management advice;
•further work focused on growth, mortality, reproduction and movement for South Pacific blue shark should be prioritised to overcome the paucity of biological data for this stock;
•further work on stock structure should be undertaken for blue shark in the WCPO south of the equator to confirm the assumption of independent North and South Pacific blue shark stocks;
•an investigation of the potential for modern genetic techniques, such as gene tagging and close-kin genetic analysis, to provide fishery independent indicators of population size should be undertaken;
•a careful consideration of the availability and quality of additional sources of data existing east of the WCPFC-CA should be performed before future stock assessments for South Pacific blue shark occur;
•active collaborations between interested CCMs should be pursued for future South Pacific blue shark assessments, e.g., through expert workshops.