Does fragmentation increase extinction thresholds? A European-wide test with seven forest birds

AimTheory predicts that fragmentation aggravates habitat loss, increasing the extinction threshold of habitat specialists. However, contradictory empirical results have fuelled claims that fragmentation has been overemphasized, and more attention should be given to habitat loss for preserving species. We assess variation in species sensitivity to forest amount and fragmentation and evaluate if fragmentation is related to extinction thresholds in seven forest bird species. LocationEurope. MethodsWe use the percentage of forest cover and the proportion of cover occurring in the largest patch to partition effects of forest amount versus fragmentation, and apply logistic regression to model the presence-absence of 17 forest bird species. For seven species showing robust models, we define two fragmentation scenarios, low and maximum, across the forest cover gradient and quantify species' sensitivity to forest contraction with no fragmentation, and to fragmentation under constant forest cover. Finally, we develop two tests of the extinction threshold hypothesis by comparing the occurrence probability of each species under the two fragmentation scenarios at different forest covers. ResultsAs expected, forest contraction had negative impacts on the occurrence probability of all seven species modelled, but - in line with theory - fragmentation also led to a higher extinction threshold for three (Western capercaillie, Hazel grouse and Eurasian pygmy-owl). One species (Black woodpecker) exhibited the opposite pattern indicating that it probably benefits from fragmentation. Differences among species responses may reflect dispersal abilities, specializations in resources/habitat characteristics and/or sensitivity to potential modifications of interspecific interactions. Main conclusionsAlthough forest amount is of primary importance for the persistence of forest specialist birds, fragmentation is also relevant for some, and neglecting forest fragmentation would be a mistake for these species. Species-specific traits can be helpful for interpreting species' reactions to fragmentation, and it should not be assumed that it always, or never, matters.