Multispecies lanscape functional connectivity enhances local bird species' diversity in a highly fragmented landscape

Abstract

Local species assemblages are likely the result of habitat and landscape filtering. However, there is still limited knowledge on how landscape functional connectivity complements habitat attributes in mediating local species assemblages in real-world fragmented landscapes. In this study, we set up a non-manipulative experimental design in a standard production forest to demonstrate how functional connectivity determines the spatial distribution of a bird community. We test single- and multispecies spatially explicit, landscape functional connectivity models framed within the circuit theory, considering also patch attributes describing habitat size and quality, to weight their effects on species occurrence and community assemblage. We found that single-species functional connectivity effects contributed positively for occurrence of each species. However, they rarely provided competing alternatives in predicting community parameters when compared to multispecies connectivity models. Incorporating multispecies connectivity showed more consistent effects for all community parameters, than single-species models, since the overlap between species’ dispersal abilities in the landscape shows poor agreement. Habitat size and quality, though less important, were also determinant in explaining community parameters while possibly relating to the provision of suitable nesting and foraging conditions. Both habitat and landscape filters concur to govern community assembly, though likely influencing different processes: while landscape connectivity determines which species can reach a patch, habitat quality determines which species settle in the patch. Our results also suggest that surrogating multispecies connectivity from single species has potential to source bias by assuming species perceive landscape and its barriers similarly. Inference on this issue must be gathered from as much species as possible.