The eastern Bering Sea shelf supports a large biomass of several groundfish populations, particularly walleye pollock (Theragra chalcogramma). The main trophic interactions between dominant groundfish populations in this region involve walleye pollock as prey. A multispecies virtual population analysis model (MSVPA) covering the period 1979-1995 has been parameterized for the eastern Bering Sea in order to move closer to providing multispecies management advice for this region. The MSVPA model is based on extensive diet data from 70 predator/prey/year/quarter combinations and currently includes the following species as predators: walleye pollock (Theragra chalcogramma), Pacific cod (Gadus macrocephalus), Greenland turbot (Reinhardtius hippoglossoides), yellowfin sole (Pleuronectes asper), arrowtooth flounder (Atheresthes stomias), and northern fur seal (Callorhinus ursinus). Arrowtooth flounder and northern fur seals are entered as ''other predators'', which means that population and mortality estimates are not directly made for these species. Estimates of their consumption rates, diet, and population abundance are input, however, so that their predation on prey species in the model can be calculated. Prey species are walleye pollock, Pacific cod, Greenland turbot, yellowfin sole, rock sole (Lepidopsetta bilineatus), and Pacific herring (Clupea pallasi). Results show that large numbers of walleye pollock, particularly age-0 and age-1 fish, are consumed and cannibalism by adult pollock constitutes the largest source of predation mortality for age-0 fish. Predation plays an important role in explaining the recruitment dynamics of pollock. Further advances may be made when these estimates are linked to models that examine the effect of climate-related factors influencing larval survival. 2000 International Council for the Exploration of the Sea

Virtual population analysis and the statistical catch-at-age methods are common stock assessment models used for management advice. The difference between them is the statistical assumptions allowing the fitting of parameters by considering how errors enter into the models and the data sources for the estimation. Fishery managers are being asked to consider multispecies interactions in their decisions. One option to achieve this goal is the multispecies virtual population analysis (MSVPA); however, its lack of statistical assumptions does not allow the use of tools used in single-species stock assessment. We chose to use a two-species system, walleye pollock (Theragra chalcogramma) and Pacific cod (Gadus macrocephalus), to incorporate the predation equations from MSVPA into an age-structured multispecies statistical model (MSM). Results suggest that both models produced similar estimates of suitability coefficients and predation mortalities. The adult population estimates from the single-species stock assessment and MSM were also comparable. MSM provides a measure of parameter uncertainty, which is not available with the MSVPA technologies. MSM is an important advancement in providing advice to fisheries managers because it incorporates the standard tools such as Bayesian methods and decision analysis into a multispecies context, helping to establish useful scenarios for management in the Bering Sea.Résumé : L'analyse des populations virtuelles et les méthodes statistiques reliées aux captures en fonction de l'âge sont des modèles courants d'évaluation des stocks utilisés pour générer des avis sur l'aménagement. La différence entre ces deux approches réside dans les présuppositions statistiques qui permettent l'ajustement des paramètres en considérant comment les erreurs entrent dans les modèles et les sources de données utilisés pour les estimations. Les gestionnaires de la pêche doivent considérer les interactions plurispécifiques dans leurs décisions. Une façon d'atteindre cet objectif est d'utiliser l'analyse des populations virtuelles plurispécifiques (MSVPA); cependant, l'absence de présuppositions statistiques ne permet pas l'emploi des outils qui servent dans l'évaluation des stocks monospécifiques. Nous avons choisi un système comprenant deux espèces, la goberge de l'Alaska (Theragra chalcogramma) et la morue du Pacifique (Gadus macrocephalus), afin d'incorporer les équations de prédation de MSVPA dans un modèle statistique plurispécifique (MSM) structuré en fonction de l'âge. Les résultats montrent que les deux modèles produisent des estimations semblables des coefficients d'adéquation et des mortalités dues à la prédation. Les estimations des populations adultes réalisées à partir des évaluations de stocks monospécifiques et de MSM sont aussi comparables. MSM fournit une mesure de l'incertitude des paramètres, ce qui n'est pas disponible dans les méthodologies MSVPA. MSM représente un progrès important dans la production d'avis pour les gestionnaires, parce qu'il incorpore les outils habituels, tels qu...

J. 2005. A framework for ecosystem impacts assessment using an indicator approach. e ICES Journal of Marine Science, 62: 592e597.Assessment of the historical, present, and future states of marine ecosystems and the effects that humans and climate have on the state of an ecosystem are crucial to the scientific advice required to implement an ecosystem-based fishery management system. Management of federal groundfish fisheries in Alaska considers not just the target fishery, but also the possible impact those fisheries might have on other species and the ecosystem. Management actions have ranged from providing protection of endangered species in the region to preventing new fisheries from starting on key foodweb components such as forage fish. A scientific framework for providing ecosystem-based advice that puts the ecosystem first has been evolving over the past few years. This framework provides a way of assessing ecosystem factors that influence target species, the impact the target fishery may have on associated species, and ecosystem-level impacts of fishing. An indicator approach that describes ecosystem status, and trends and measures of human and climate influence has been developed to provide advice to fishery managers. This approach is now being expanded to utilize a variety of models to predict possible future trends in various ecosystem indicators. Future implementation challenges include the refinement of these predictive models, and the inclusion of climate into the models. Identification of sensitive and meaningful ecosystem indicators is also required before a more formalized decisionmaking process, one that includes ecosystem considerations, can be developed. Most important, the culture of fishery management and research organizations needs to change to embrace the ecosystem-based protections already mandated by various laws.

The need to understand the multispecies implications of various harvesting regimes is becoming more important as fishery managers move toward ecosystem‐based management. We explore the possible effects that different exploitation rates may have on eastern Bering Sea groundfish using a multispecies simulation context that incorporates predator–prey relationships. At present, some groundfish species in the eastern Bering Sea are exploited up to the recommended levels of allowable biological catch, whereas others, for economic or bycatch limitation reasons, are only lightly exploited. We explore the possible long‐term multispecies implications of different exploitation patterns on the biomass and yield of several groundfish species using predator−prey suitability estimates derived from multispecies virtual population analysis (MSVPA) in a multispecies simulation modeling context and compare those predictions with those from single‐species forecasting models. Three different fishing scenarios that included eight species in the eastern Bering Sea were implemented in these models. In one scenario, the present exploitation rates were used for the model simulations; in the second scenario, all exploited species were more evenly exploited by fishing each species at its recommended allowable biological catch levels; and in the third scenario, there was no fishing for all the species. Results from the single‐species and multispecies model simulations mostly showed the same direction of population trends. However, the magnitude of change was different for some species, which can be ascribed mainly to predation interactions. Greater differences were seen between the simulations that used the present exploitation rates and those in which no fishing occurred. The multispecies simulations that included predation interactions predicted much lower equilibrium population sizes for prey species populations under conditions of no fishing than did single‐species simulations that did not take predator–prey relationships into account. These results show that multispecies models provide new insights into the implications of single‐species harvesting strategies, particularly for prey species that are also the target of commercial fisheries.

The Western and Central Pacific Ocean sustains the highest tuna production in the world. This province is also characterized by many islands and a complex bathymetry that induces specific current circulation patterns with the potential to create a high degree of interaction between coastal and oceanic ecosystems. Based on a large dataset of oceanic predator stomach contents, our study used generalized linear models to explore the coastal-oceanic system interaction by analyzing predator-prey relationship. We show that reef organisms are a frequent prey of oceanic predators. Predator species such as albacore (Thunnus alalunga) and yellowfin tuna (Thunnus albacares) frequently consume reef prey with higher probability of consumption closer to land and in the western part of the Pacific Ocean. For surface-caught-predators consuming reef prey, this prey type represents about one third of the diet of predators smaller than 50 cm. The proportion decreases with increasing fish size. For predators caught at depth and consuming reef prey, the proportion varies with predator species but generally represents less than 10%. The annual consumption of reef prey by the yellowfin tuna population was estimated at 0.8±0.40CV million tonnes or 2.17×1012±0.40CV individuals. This represents 6.1%±0.17CV in weight of their diet. Our analyses identify some of the patterns of coastal-oceanic ecosystem interactions at a large scale and provides an estimate of annual consumption of reef prey by oceanic predators.