Incorporating intraspecific variation into species responses reveals both their resilience and vulnerability to future climate change

Abstract

Incorporating intraspecific variation into species responses can improve our understanding of the effects of climate change. However, most studies overlook such variation or model intraspecific groups independently, leading to widely varying estimates at the species level. In this study, we used a generalized additive framework to model the climate responses of 25 bird species in the tropical Andes, allowing for the estimation of nonlinear responses between subspecies. We measured the effects of environmental covariates (climate, topography, and primary productivity) on relative abundances and projected population trends for near-future (2050) scenarios. Compared with models that ignored subspecies, the subspecies-level models performed better in terms of accuracy, discrimination, and goodness-of-fit, while projecting fewer species to decline in relative abundance. Thus, species could be more resilient than estimated using species-level models that ignore intraspecific variation. Nevertheless, our modeling approach also revealed that intraspecific groups may be vulnerable to future threats. We found that subspecies can have low relative abundances and small population sizes. Additionally, based on a preliminary IUCN Red List assessment, one in four subspecies could be threatened due to a small geographical range or a declining population trend. Most of these subspecies at risk inhabit centers of endemism in the northern Andes, such as the Santa Marta Mountains and the Serranía de Perijá. We suggest that protecting subspecies and other intraspecific groups (e.g. populations) could be a critical conservation strategy to buffer against the impacts of climate change, especially in biologically diverse regions.