Climate is predicted to change rapidly in the current century, which may lead to shifts of species' ranges, reduced populations and extinctions. Predicting the responses of species abundance to climate change can provide valuable information to quantify climate change impacts and inform their management and conservation, but most studies have been limited to changes in habitat area due to a lack of abundance data. Here, we use generalized linear model and Bayesian information criteria to develop a predictive model based on the abundance of the grey-headed robin (GHR) and the data of climatic environmental variables. The model is validated by leave-one-out cross-validation and equivalence tests. The responses of GHR abundance, population size and habitat area by elevation are predicted under the current climate and 15 climate change scenarios. The model predicts that when temperature increases, abundance of GHR displays a positive response at high elevation, but a negative response at low elevation. High precipitation at the higher elevations is a limiting factor to GHR and any reduction in precipitation at high elevation creates a more suitable environment, leading to an increase in abundance of GHR, whereas changes in precipitation have little impact at low elevation. The loss of habitat is much more than would otherwise be assumed in response to climate change. Temperature increase is the predominant factor leading to habitat loss, whereas changes in precipitation play a secondary role. When climate changes, the species not only loses part of its habitat but also suffers a loss in its population size in the remaining habitat. Population size declines more than the habitat area under all considered climate change scenarios, which implies that the species might become extinct long before the complete loss of its habitat. This study suggests that some species might experience much more severe impacts from climate change than predicted from models of habitat area alone. Management policies based on predictions of habitat area decline using occurrence data need to be re-evaluated and alternative measures need to be developed to conserve species in the face of rapid climate change.