1. Cumulative impact assessment (CIA) is a promising approach to guide marine spatial planning (MSP) and management. One limitation of CIA is the neglect of seascape connectivity, which may spread the impact of localised pressures to ambient areas, e.g. through lost dispersal and recruitment of organisms.

2. We here, for the first time, incorporate seascape connectivity into a traditional CIA model using a connectivity matrix, exemplified by dispersal of propagules estimated through biophysical modelling. Two connectivity impactsare identified: the source impact represents downstream areas losing recruits because of reduced larval dispersal from sites affected by the pressure, and the sink impact represents loss of recruits originating from upstream areas prevented from settlement in the site affected by the local pressure.

3. By including seascape connectivity into the Swedish MSP-guiding CIA tool Symphony we demonstrate how to practically account for remote effects of local environmental impact. Our example on blue mussel shows how reducing mussel fitness in a given area may have impacts on mussels far from the acting pressures. Overall, results indicate that connectivity impact for blue mussels plays a minor role in most areas, less than 10% of the ordinary cumulative impact. However, in some smaller areas, e.g. on offshore banks and the Danish Straits, seascape connectivity may increase ordinary cumulative impact with 20-30%. In an example of scenario-based CIA analyses of MSP projections, we demonstrate how impacts of particular management actions, e.g. shipping rerouting and wind power developments, can be tracked far from the original area of influence.

4. Depending on the dispersal ability of ecosystem components, a local pressure may impact a considerable area through seascape connectivity, transgressing management units and national borders. Although the mean connectivity impact may be modest for a single ecosystem component, the consideration of seascape connectivity across multiple ecosystem components may significantly alter the mapping of cumulative impact and the assessment of different MSP scenarios.

5. Synthesis and applications. Our extension of Cumulative Impact Assessment offers a new method for mapping and practically integrating seascape connectivity with ecosystem-based MSP and other spatial instruments for policy making, such as marine protected areas.

The article by Jonsson et al. in Journal of Applied Ecology presents an extended model for the estimate of marine Cumulative Impact Assessment (CIA) that includes seascape connectivity. We use a new CIA model (Symphony) based on spatial data layers of human pressures and ecosystem components (e.g. mussel reefs, seagrass beds, and photic soft sediments), which covers the Swedish Exclusive Economic Zone in the North Sea and the Baltic Sea. This CIA model is combined with modelled seascape connectivity to show how estimates of connectivity can be implemented into a model of CIA. As a demonstration we here consider the connectivity of blue mussel (Mytilus edulis/trossulus) habitats represented by the dispersal of planktonic mussel larvae. Dispersal was estimated with a biophysical model, which simulates the dispersal trajectories of larvae in the seascape from release to settlement. Dispersal probability between all gridded locations in the seascape was summarised into a connectivity matrix. With a set of Python and Cuda scripts the ordinary cumulative impact was compared to the cumulative impact when also seascape connectivity was considered.