ICES, in consultation with the North Atlantic Salmon Conservation Organisation (NASCO), convened a series of workshops to explore how to use biological and environmental data in models to advance the conservation of wild Atlantic salmon (Salmo salar L.) at sea. This workshop set out to consider multiple candidate hypotheses contributing to changes in the temporal patterns of abundance, and agree the priority research questions. No agreement on the development of a set of priority marine mortality hypotheses was reached. This resulted from the recognition of the hierarchical nature of ecosystem controls, and important complexities introduced by evolutionary diversity. An integrated ecological-evolutionary framework was proposed for the evaluation of hypotheses, and to identify key points in space and time. There was an agreed need for the continuation of cooperative initiatives to examine drivers of marine growth change using standardised approaches, and in the evolutionary delineation of stock units. These were seen as productive pathways to significantly enhance understanding of the marine factors impacting species abundance. The workshop recognised that options for developing and testing hypotheses remain constrained by the availability and quality of data, and identified ways to mobilise existing knowledge resources on key aspects of salmon ocean ecology. These focused on the synthesis of physical ocean data and model outputs, involving ocean basin-wide evaluations of available energy from surveys of lower trophic levels, and updating of population-specific biological information. The workshop agreed on the need for a specific call for data from pelagic commercial fisheries, given the broad scale of this activity and potential overlap with salmon migrations. There was also the recognition that Atlantic salmon should be included in the ICES Working Group on Bycatch of Protected Species (WGBYC) Protected, Endangered and Threatened Species list. Much of the work required to mobilise useful data sources was recognised as being outside the scope of existing ICES data calls, or the constituted core work of ICES Working Group on North Atlantic Salmon (WGNAS). Recommendations for the third workshop are for 1. More detailed consideration of how to access the work needed for data mobilisation, and 2. The identification of well-defined, achievable outcomes.