| The effects of habitat fragmentation have been explored in detail in many different systems, with great variation in results. Despite methodological variation, much of this variance is real, and reflects fundamental differences in the responses of faunal assemblages to habitat perturbation. Most studies have been carried out in relatively young landscapes, in which interactions may not have stabilized after fragmentation. Based on these ideas, I carried out several separate studies, all focused on the long-term effects of habitat fragmentation, treating Pleistocene-age montane patches as old (equilibrial) habitat fragments. Reviewing existing studies of Pleistocene-age fragmentation revealed striking similarity in responses, with patch area and vegetation having consistently strong positive relationships with richness, and latitude important at larger scales. Testing this generality in Mesoamerican humid montane forest patches validated these ideas, with regional scale variables (notably latitude) explaining most variation in species richness. Interestingly, diversities were concentrated in the mountains on either side of the two lowland isthmuses, lending support to a historic component of regional diversity patterns. To explore the smaller-scale basis of these patterns, a landscape level inquiry was conducted in humid pine-oak forests in Oaxaca, at a comparable spatial scale to many studies of anthropogenic fragmentation. Area, vegetation, and elevation were all closely inter-related, with a combined positive influence on avian richness. The fauna exhibited a highly nested arrangement, suggesting that this is a system in which extinction has been a major structuring force. These patterns appear to have been driven by a combination of patch-level factors (especially area and habitat heterogeneity) and species-specific attributes (notably density and vagility). |
