Increasing the value of bird-habitat studies in tropical forests: choice of approach and habitat measures

Abstract

AbstractAn important component of many conservation studies is the assessment of bird-habitat relationships, but limited resources often lead to constraints on study design, quality and quantity of bird data, and restrict the number and types of habitat variables gathered. The aim of this study was to identify habitat features that were both relatively easy and quick to collect and powerful in identifying bird-habitat relationships. We also discuss some issues with our study and alternative approaches that may help in future bird-habitat studies in tropical forests. Twenty-four habitat measures representing geographical (e.g. altitude, topography, X and Y coordinates), vegetation structure (e.g. tree sizes), and tree floristics (abundance of 28 indicator tree species) features were collected in association with bird presence/absence data from point transects within a 1,500 ha Philippine lowland forest. We used hierarchical partitioning of regression analyses to assess which of these geographical and structural variables along with four floristics axes derived from DECORANA were the most important variables for explaining the occurrence of individual bird species and guilds. The ten most powerful variables for a range of bird species included seven geographical and three floristic variables, while the ten least important were all structural variables. There were differences in importance of individual variables across guilds, with, for example, floristics very important in canopy frugivores, and geographical variables more important for upperstorey gleaning insectivores. We stress the importance of geographical variables in linking birds to habitat at this local scale, but also suggest that efforts are made to collect some floristics data, perhaps a subset of species that represent resources for birds (e.g. Ficus spp.), people's use of the forest (e.g. dipterocarps), and indicators of forest type. While the habitat variables and approach in this study adequately identified bird-habitat relationships for most species, we suggest improvements and alternative methods that may improve results in other studies.