Report of the Workshop on Length-Based Indicators and Reference Points for Elasmobranchs (WKSHARK4)

The workshop WKSHARK4, Workshop on Length-Based Indicators and Reference Points for Elasmobranchs, chaired by Pascal Lorance (France) and Jan Jaap Poos (Netherlands), was held in Nantes, France 6–9 February 2018. Twelve participants contributed by presence and six more attended daily video-conference and participated by correspondence.
The overall objective was to analyse the appropriateness of length-based indicators (LBIs) for assessment of the status of elasmobranch stocks. The concerns raised by WGEF in 2017 regarding the use of LBIs were investigated (Chapter for ToR a), a simulation approach based on a Leslie matrix was taken to investi-gate the possible values of LBIs in unexploited and exploited populations, this was mostly applied to the lesser-spotted dogfish, (Chapter for ToR b and c), and LBIs were applied to a few case studies (Chapter 4 Chapter for ToR d).  
In the analyses of concerns raised by WGEF, it was shown that estimates of life-history parameters, in particular growth parameters, are uncertain. The asymptotic length, L∞, is often estimated at higher value than the larger fish in age-at-length. The lack of old fish in data implies that the size where growth levels off is uncertain. When fitting growth models to data impacts both estimates of the growth, coefficient (K) and the asymptotic length (L∞). Because both are used in the calculation of reference points (RPs) for LBIs the impact of these uncertainties on the stock status derived from LBIs may be large. Investigations of the possible impact of the fishing gear selection and spatial distributions from a number of case studies did not reveal problems that would undermine the ap-propriateness of LBIs for the studied stocks. However, in one case, the undulate ray in ICES Division 9a, the very inshore distribution of the species implies that sampling should cover properly small-scale coastal fleet to derive representative estimate of the length-frequency distribution of the commercial catch. In contrast, the four stocks investigated for constant recruitment, selection and fishing mortality in recent years were not in that situation and yearly variations in these stocks and fisheries parameters may lead to misinterpretation of LBIs (e.g. if larger recruitment is occurring in a stock, the first consequence will be a reduc-tion of the mean size observed in the catch, which could be interpreted as a decline of large fish). Extensive work was carried out with the Leslie matrix approach. This was mainly used to investigate the suitability of existing RPs. The model was further developed to simulate fisheries selectivity and calculate the expected values of LBIs in lesser-spotted dogfish stocks. LBIs were calculated from empirical data for 6 stocks exploited by UK fisheries, including spurdog, for which a quantitative assessment is available. These studies provided mixed results from the various LBIs. The LBIs based on length at first capture (Lc and L25%) invariably highlighted that length at first capture (i.e. length 50% selected) is smaller than length at maturity, and therefore are always below their RPs val-ues. The LBIs representing exploitation at MSY (Lmean = LF = M) were often above the corresponding RP. Therefore, contrasting diagnoses are obtained from the various LBIs, suggesting some inconsistencies in their RPs when applied to elasmobranchs. It was suggested that trend-based metrics should be considered un-til appropriate reference points are validated.
Results from WKSHARK4 should be considered by WGEF and it is recom-mended that LBIs are calculated for more stocks and that the Leslie matrix model explored during WKSHARK4 is applied to several simulated elasmobranch stocks with varied life-history traits to investigate the expected value of reference points for LBIs.