%0 Journal Article %A James A. Orr %A Jeremy J. Piggott %A Andrew L. Jackson %A Michelle C. Jackson %A Jean-François Arnoldi %T Stressor interactions at the community level: explaining qualitative mismatches between observations %D 2022 %R 10.1101/2022.06.20.496833 %J bioRxiv %P 2022.06.20.496833 %X Understanding the ecological impacts of global change forces us to consider the antagonistic and synergistic interactions between the multiple stressors that ecosystems may face. At the community-level, such interactions are quantified based on the responses of various ecosystem-functioning or diversity metrics. Worryingly, in empirical data, we find that community metrics often observe opposite interactions between the same two stressors - sometimes even systematically. Here, we investigate this puzzling pattern via a series of geometrical abstractions. By representing stressors and their interactions as displacement vectors in community state-space, and community metrics as directions in this space, we show that the angle between two directions determines the probability, over random stressor combinations, that the metrics will observe opposite interactions. We find that diversity and functioning can easily be associated to opposing directions, which explains the systematic mismatches seen in empirical data.Competing Interest StatementThe authors have declared no competing interest. %U https://www.biorxiv.org/content/biorxiv/early/2022/06/21/2022.06.20.496833.full.pdf