Drivers of mammal richness, diversity and occurrence in heterogeneous landscapes composed by plantation forests and natural environments

摘要

Agroecosystems dominate many landscapes throughout the world and it is crucial to understand the patterns of wildlife communities thriving in these areas, and the drivers shaping these patterns. Thus, we evaluated the effects of vegetation structure and landscape composition on the richness, diversity and occurrence patterns of medium and large mammal in a heterogeneous Amazonian landscape, composed by planted forest and native habitats. Based on line transects and camera trap surveys in three different environments (eucalyptus plantations, savannas and forests), we detected 17 mammal species, including two species threatened nationally and worldwide. Mammal richness and diversity were similar among the environments. Mammals were more recorded in eucalyptus plantations, at homogeneous areas (i.e. few edges), in sites with fewer roads, and with larger proportions of natural environments. The detection of Primates, Camivora, Rodentia and Pilosa was positively correlated with the distance to savannas, while Cingulata, Artiodactyla, and Perissodactyla show the same pattern towards forest patches. Primates, Carnivora, and Rodentia were also associated with taller vegetation. Didelphimorphia was registered mostly in heterogeneous areas (i.e. higher total edge amount) while five orders were mainly recorded in homogeneous areas. Eucalyptus plantations immersed in a landscape containing significant patches of savanna and forest can maintain part of the local biodiversity, although plantations alone may not be enough to support populations of most mammal species. Indeed, eucalyptus plantations, savannas, and forest patches have a complementary role in mammal richness and occurrence patterns. Since distinct drivers influence occurrence patterns of different taxonomic groups, the existence of areas with different characteristics in the same landscape will favor higher total biodiversity.