Report of the Working Group on Recreational Fisheries Surveys (WGRFS), 12–16 June 2017, Azores, Portugal

The ICES Working Group on Recreational Fisheries Surveys (WGRFS) role is to sum-marise and quality assure recreational fishery data collected in European countries, and provide advice for ICES on recreational fishing issues. In 2017, 31 scientists from 17 countries attended the WGRFS to: share and evaluate current national surveys; as-sess the validity of new survey designs; provide support on the use of survey data in stock assessment; review national and regional data plans; review novel survey meth-ods; highlight new work on post-release mortality; assess the potential for research on human dimensions; and review the treatment of outliers. The terms of reference and agenda for the working group are provided in Section 1. WGRFS compiled and assessed the quality of recreational harvest and release data col-lected within Europe for use in stock assessment (Section 2). These were summarised by country for four major sea areas (Baltic Sea, North Sea and Eastern Arctic, North Atlantic, and Mediterranean and Black Seas) and species (European sea bass, cod, pol-lack, elasmobranchs, salmon, eels, and tuna). The design, quality and analysis of marine recreational fisheries surveys was investi-gated (Section 3). A summary of the experience of marine recreational fisheries surveys in New Zealand and Canada was provided (Section 3.1). The identification and treat-ment of outliers in the analyses of recreational fishing surveys was investigated, with several methodologies highlighted (e.g. hotdeck imputation, trimmed means) (Section 3.2). However, there was rarely evidence that datapoints are incorrect, so care was needed when deciding treatment of outliers. In general, where outlier or imputation procedures were used, the sensitivity of the results to the approach should be investi-gated. In addition, a more comprehensive assessment was needed of the methods and proposals for robust approaches, and should be done at a future WGRFS meeting. The quality of national recreational catch sampling schemes in Belgium, Norway, and Swe-den were assessed using the WGRFS Quality Assessment Tool (QAT) (Section 3.3). It was only possible to assess the design of these surveys, as the analysis was not com-plete, with the designs considered appropriate in all cases. The interactions between marine recreational fisheries surveys and data and the EU-MAP and regional coordination were discussed (Section 4). Several WGRFS members were involved in the review of the National Work Plans (NWP), so feedback was pro-vided to the STECF on the process. The main challenges were the mismatch between the evaluation criteria and template, lack of feedback to MSs on the evaluation, identi-fication of experts, timely requests for expert input, and evaluation of national surveys during the WGRFS. The European Parliament study, EURecFish that builds on the WGRFS analysis, and estimated numbers, participation, effort and expenditure by rec-reational fishers in Europe, was discussed. The storage of recreational fisheries data was assessed and it was agreed that processed recreational fisheries data should be stored in regional databases, and that WGRFS should work with ICES to develop a plan and time-scale for delivery of this solution. The use of recreational fishing catches in stock assessment was reviewed for western Baltic cod, European sea bass, and Baltic salmon and sea trout (Section 5). For sea bass, all recreational data were delivered through a data call and the WGRFS assessed how to provide appropriate data for use in stock assessment. Recreational catches and post-release mortality should be included in the assessment. The lack of survey data made removals after the introduction of management measures difficult to estimate, but in the short term should be based on data from the UK and Netherlands alongside ex-trapolation for France, with new data from France and Belgium included as soon as available. It is very important that time-series are collected for sea bass and that new methods for reconstructing time-series of catches are developed to improve the accu-racy of assessments of stocks with significant recreational components. There is an ur-gent need to include Danish and Swedish recreational catches of western Baltic cod in the assessment. This includes exploratory stock assessment runs to test the effect of including all recreational data on SSB and F, possibly by pooling total recreational catch. Regional cooperation and sharing of data (e.g. biological) across subdivisions should be further explored to fill data gaps. There is also a need to observe recreational fishing effort dynamics in response to the introduced management measures. Recrea-tional catches of salmon and sea trout were included in the assessments, but marine recreational catches were not well defined. In the case of Baltic salmon, expert judge-ment was used to develop understanding of the trolling fishery that demonstrated the importance of this fishery. Coverage of marine catches of sea trout were poor, so better data were needed to reduce the uncertainty in the assessments. There was an urgent need for the collection of more robust catch, effort, post-release mortality, and socio-economic data for recreational Baltic Sea trout and salmon fisheries. To estimate fishery-specific mortality, WGRFS collected information on recreational fishing practices in different European marine recreational fisheries during two work-shops conducted in 2015 and 2016. A practical implementation of this was shown and proposals made on how to proceed with post-release mortality (Section 6). It was agreed that recreational fishery characteristics for certain target species should be col-lected to enable extrapolation between stocks and fisheries, and sublethal impacts of catch and release investigated. Novel approaches for data collection were reviewed including smartphone apps and webcams (Section 7). WGRFS assessed the information that needs to be collected from apps and the challenges with using app-derived data. The potential of smartphone apps was clear, but so were the challenges in using the data. The extent and direction of biases could be addressed through comparison of app data with onsite data, such as creel or access point surveys. Due to the broad range of apps available, data collection standards should be developed by a collaboration of app companies and end-users. Two examples of the use of webcams in New Zealand and Germany were presented and opportunities discussed. The application opportunities of human dimension research in recreational fisheries were explored and several methods presented. Understanding anglers’ reactions to recreational fisheries regulations help to predict changes in fishing effort dynamics and welfare. Further human dimension research will provide recommendations for alloca-tion decisions between sectors and optimum co-management of commercial and rec-reational fisheries. To facilitate comparisons between different countries the experimental design (e.g. choice experiments) should be harmonized