At the landscape level, birds respond strongly to habitat amount but weakly to fragmentation

Abstract Aim It is usually thought that habitat fragmentation acts negatively on species survival, and consequently, on biodiversity. Recent literature challenges whether habitat fragmentation per se affects species richness, beyond the effect of habitat area. Theoretical studies have suggested that fragmentation may matter most when the amount of available habitat is small or at intermediate levels. However, a recent review suggests that the effect of fragmentation on species richness is usually positive. Here, we dissect the richness–fragmentation relationship. What is the effect size? Does it depend upon the amount of habitat cover? How do individual species respond to fragmentation? Methods Applying a macroecological approach, we empirically related avian richness and the probability of occurrence ( p occ ) of individual species to fragmentation (number of patches), after controlling for habitat amount in 991 landscapes, each 100‐km 2 , in southern Ontario, Canada. Results Species richness was strongly related to total habitat amount, but habitat fragmentation had no detectable additional effect. Individual species’ p occ related strongly to habitat amount. For some species, p occ also related secondarily to habitat fragmentation within landscapes. Logistic models revealed that p occ related significantly negatively to fragmentation after controlling for habitat amount for only ~13% of forest‐ and 18% of open‐habitat species bird species. However, p occ related significantly positively to fragmentation for even greater proportions of species, including some red‐listed species. Fragmentation effects were not stronger at low or intermediate levels of habitat amount within landscapes. Conclusion In earlier studies, negative effects of isolation were observed at the patch level in experimental manipulations. However, at the landscape level , avian species richness in southern Ontario apparently responds primarily to habitat amount and negligibly to fragmentation. We argue that the evidence is inconsistent with the hypothesis that reducing habitat fragmentation per se would be an effective conservation strategy for birds at the landscape level.