LAUSR

The scientific community remains divided on the most effective way to design landscapes for biodiversity conservation or restoration. Although there is a consensus that habitat loss is the main cause of biodiversity decline worldwide, the extent to which fragmentation (i.e. the division of remaining habitats into smaller areas) contributes to this decline is a subject of ongoing debate. The spatial arrangement of remaining patches and the nature and permeability of the intermediate matrix (i.e. how easily animals can move through it) are other elements related to habitat loss that are little considered. A better understanding of the effects of these factors on populations could help the community move forward. Here, we conducted a multigenerational, landscape‐scale experiment with the microarthropod Folsomia candida and quantified the respective effects of matrix resistance and inter‐patch distance on colonization rate, population size and extinction, at fixed habitat amount. We found that the amount of reachable habitat in the landscape, encompassing both the quantity of habitat and the matrix resistance, is a good predictor of population size and extinction rate. Survival of individuals while crossing different matrix types was the key underlying mechanism, as it determined both colonization rate and demography, preventing individuals from reaching and using remote or difficult‐to‐access patches. Our study shows that an explicit consideration of matrix resistance considerably improves both our understanding and our predictive ability of populations fate at landscape‐scale. It also opens new avenues for landscape ecology theory as well as long‐awaited perspectives for applied conservation.