As the urgency to evaluate the impacts of climate change on marine ecosystems increases, there is a need to develop robust projections and improve the uptake of ecosystem model outputs in policy and planning. Standardising input and output data is a crucial step in evaluating and communicating results, but can be challenging when using models with diverse structures, assumptions, and outputs that address region-specific issues. We developed an implementation framework and workflow to standardise the climate and fishing forcings used by regional models contributing to the Fisheries and Marine Ecosystem Model Intercomparison Project (FishMIP) and to facilitate comparative analyses across models and a wide range of regions, in line with the FishMIP 3a protocol. We applied our workflow to three case study areas-models: the Baltic Sea Mizer, Hawai’i-based Longline fisheries therMizer, and the southern Benguela ecosystem Atlantis marine ecosystem models. We then selected the most challenging steps of the workflow and illustrated their implementation in different model types and regions. Our workflow is adaptable across a wide range of regional models, from non-spatially explicit to spatially explicit and fully-depth resolved models and models that include one or several fishing fleets. This workflow will facilitate the development of regional marine ecosystem model ensembles and enhance future research on marine ecosystem model development and applications, model evaluation and benchmarking, and global-to-regional model comparisons.

Climate change is already impacting ecosystem composition and species distributions. A question remains as to how future climate change will alter ecosystem structure. Here we study two distinctly different, but equally valuable New Zealand fisheries (Tasman Bay and Golden Bay, and Chatham Rise), and the potential impacts that climate change could cause. We use mizer, a size-based multispecies modelling package, to simulate a high number of interacting fish species in each ecosystem. By utilising therMizer, an extension of mizer which incorporates temperature effects on species’ metabolic rate and aerobic scope, we implement historical climate data from the Fisheries and Marine Ecosystem Model Intercomparison Project (FishMIP). This enables us to recreate the historical time period of 1961–2010, deriving reasonable steady state with biomasses closely matching those observed in the past. We then simulate forwards to 2060, allowing for temperature to remain steady or to increase, and for fishing to occur, and look at the resulting changes in species biomasses. It was found that while specific species responses differ, both ecosystems generally decrease in abundance under climate change. Chatham Rise is much more impacted by fishing effects. Species with higher thermal tolerance ranges tended to fare better under climate change. Issues raised during the incorporation of temperature effects include knowledge of species’ thermal tolerances and the importance of how the initial steady state is tuned, as this has significant flow-on effects on species dynamics. This study will help advance other ecosystem models aiming to account for future climate change impacts.