Study on stomach content of fish to support the assessment of good environmental status of marine food webs and the prediction of MSY after stock restoration

The current reform of the Common Fisheries Policy anticipates more extensive use of long-term management plans which are consistent with the ecosystem approach to fishery management. Both long term management plans and estimates of the fishing mortality providing MSY are particularly sensitive to changes in natural mortality, and a prerequisite for estimating natural mortality correctly is the accurate knowledge of species interactions for application in multispecies models. The use of historic data in the multispecies models has so far been limited by the need for data to represent a full spatial coverage. However, the recent model developments have made it possible to integrate regional samples of stomach content data into the multispecies and ecosystem models. It is hence no longer necessary to have complete spatial coverage in a given year, before new stomach data can be included into the models. This provided a unique opportunity to utilize the vast amount of historic data available at individual fisheries research institutes. DTU Aqua together with 8 partners from the Baltic and North Sea has conducted a stomach collection and analysis project in order to (i) include all appropriate historical stomach content information into the Baltic and North Sea stomach content databases, (ii) conduct stomach content analyses of new cod stomachs collected in the Baltic Sea, to support our knowledge of the spatial and temporal stability of cod preferences, and (iii) conduct stomach content analyses of whiting stomachs collected in the Baltic Sea and grey gurnard, mackerel and hake collected in the North Sea to support our knowledge of potentially important predators for which the diet is presently poorly known or is expected to have changed significantly since the last sampling efforts in this area. All existing and new samples and data are included in the final product of this tender, i.e. a common stomach database in ICES exchange format, which is going to be made available to the scientific community via ICES. This database can then be used to re-estimate multispecies reference points such as FMSY of the different fish species. Within the Baltic Sea, efforts were focused on stomach content analyses of Baltic cod as the most abundant piscivorous fish in this ecosystem. The latest data in the cod stomach content database are from 1993. Since then, the Baltic has changed markedly. Both abundance and spatial distribution of cod and its major fish prey, herring and sprat have changed. Furthermore, due to the occurrence of extended hypoxic areas on the sea-bed, the availability of benthic food may also have changed dramatically. In addition to cod, a limited number of whiting stomach samples from the westernmost areas of the Baltic were collected and analyzed, as this species is potentially another important piscivorous predator in these regions. Before the start of the project, stomach content data for 49476 Baltic cod from the period 1977-1993 were available. The initial estimate in the proposal for the present stomach tender was that this number could be increased by ca. 170% during the course of the project. However, considerably more data have been made available during the project, especially due to the initial underestimation of the available number of historic data. The new data that have been made available increased the number of data to 255% of the initially available data. Since the focus of this tender was on the Baltic Sea, only relatively limited resources were allocated for work in the North Sea. Thus, in the North Sea, a ‘most benefits for least cost’ strategy was employed. This involved (1) the collection and analysis of new stomachs from selected species which were judged to provide the most relevant new knowledge for multispecies modeling and (2) the transformation of existing historic data into ICES exchange format. Stomachs of 1907 grey gurnard, 517 hake and 702 mackerel have been analyzed, corresponding to 119, 65 and 44 % of target values, respectively. In contrast to the Baltic Sea part of the project, that already had a large number of samples available from the beginning, the North Sea partners had to wait with initiating the stomach analyses until sufficient numbers of samples had become available, and hence the number of analyzed stomachs was not as predictable as for the Baltic Sea. Furthermore, the conversion of historic data from the DAPSTOM database has provided a large number of additional data for these three target species as well as a multitude of other species (> 207.000 data records for 184 species). Both for the Baltic and the North Sea regions, the new data have already been used and generated great interest in the scientific community. In order to facilitate the application of these data, DTU Aqua will beyond this project make data extracts available on request and collaborate with ICES to make the database available on the ICES data portal as soon as possible.