Abstract: Species distribution models are widely used to assess the effects of climate change on species distribution. Random forest (RF) model is a relatively novel machine learning method among species distribution models, quite high in accuracy. To use the method to assess effects of climate change on potential habitats of animal species, three animal species, Syrmaticus reevesii, Manis pentadactyla and Macaca thibetana, were selected as research subjects and the meteorological data of 2050 and 2080 were figured out with the aid of the three atmospheric global circulation models (GCM), i.e. MIROC32-medress, CCCMA-CGCM2 and BCCR-BCM2.0 and a greenhouse gas emission default scenario (SRES-A2). The climate dataset of each period consists of 19 bio-climatic factors, which were all cited as environment variables for the random forest model to predict potential habitats for the three species of animals under the current climate conditions and in the two future time periods (2050 and 2080), characterize shifts of the potential habitats of the species with movement of the potential habitat centroids in position, and analyze changes in area and altitude of the potential suitable habitats of the species. At the end, prediction accuracy of the model was evaluated with the receiver operating characteristic (ROC) curve and the true skill statistics (TSS) methods. Results show that the potential habitats of the three species would move gradually northwards and upwards in altitude. Among them, Manis pentadactyla’s would move the fastest and in 2080 it would reach as far as 133km up to the north, while Syrmaticus reevesii‘s would move upward in altitude the fastest and in 2080, it would be 152m higher than the current. Besides, the potential habitats of all the three species would gradually expand in area, with Syrmaticus reevesii‘s expanding by the largest margin. It is, therefore, suggested that the impact of global climate change on species should be taken into account in develop long-term protection strategies for wild life.