“…Here, we expand conventional cumulative effects assessment approaches (hereafter, species-scale assessment) [25][26][27] with recent Page 3/24 progress in theoretical ecology 14,28,29 to propose a novel method to capture the indirect propagation of the effects of stressors through species interactions (hereafter, network-scale assessment).We focus our network-scale assessment on the St. Lawrence marine ecosystem, in eastern Canada (see methods). This ecosystem is formed by one of the largest estuaries in the world and a vast interior sea.Together, they host diverse and productive ecological communities and provide a wealth of ecosystem services bene ting the Canadian economy: a rich commercial sheries industry, a seaway that grants access to one of the most densely populated regions in North-America and more than 40 ports, an expanding aquaculture production, and a thriving tourism industry 30,31 .We demonstrate our approach by assessing and mapping the cumulative effects of 18 stressors on 193 species between 2010 and 2015. We use data-based or theoretically-derived indicators to characterize the distribution and intensity of stressors 32 , the distribution of species, the network of species interactions, and species-speci c sensitivities to stressors; these are then combined through a theoretical framework 14 to predict a cumulative effect score for every species considered (see methods).…”