Drivers of the spatial scale that best predict primate responses to landscape structure
Understanding the effect of landscape structure on biodiversity is critically needed to improve management strategies. To accurately evaluate such effect, landscape metrics need to be assessed at the correct scale, i.e. considering the spatial extent at which species–landscape relationship is strongest (scale of effect, SE). Although SE is highly variable, its drivers are poorly known, but of key relevance to understand the way species use the landscape. In this study, we evaluate whether and how species traits, biological responses, landscape variables and the regional context of the study drive SE in Mexican primates. We estimated the relative abundance and immature‐to‐female ratio (a proxy of reproductive success) of howler monkeys Alouatta palliata and A. pigra and spider monkeys Ateles geoffroyi in 48 forest patches from four rainforest regions (12 patches per region) with different land‐use intensity. We then assessed the composition (forest cover, matrix functionality) and configuration (forest patch density, connectors’ density, forest edge density) of local landscapes considering 13 scales (100 to 1300‐m radius) to identify the spatial extent at which each landscape variable best predict each response variable in each species and region. We found that SE did not differ significantly among the drivers evaluated. However, it tended to be lower for connectors’ density than for forest patch density and forest edge density, probably because connectors’ density is associated with local‐scale processes such as supplementary dynamics. Surprisingly, SE also tended to be higher in the more disturbed region than in the rest of the regions, probably because primates in the more disturbed region used larger areas of the landscape. Our findings therefore suggest that SE depends more strongly on landscape variables and regional context than on species traits and biological responses, and hence, especial caution should be taken when attempting to generalize SE to different explanatory variables and regions.
No Supplementary Data
No Article Media