Seasonally frozen ground (SFG) is a critical component of the Earth’s surface that affects energy exchange and the water cycle in cold regions. The estimation of SFG depth has generally required intensive parameterization which has limited estimates in data-scarce regions such as the Qinghai-Tibet Plateau (QTP). We propose a simple yet robust modeling framework employing ground surface temperatures as major model inputs to assess the spatiotemporal patterns of the SFG depth in the Yarlung Zangbo River Basin (YZRB) on the QTP. The model was calibrated using SFG depth measurements throughout the YZRB from 1980 to 2010. Results suggest that the SFG depth in the YZRB has decreased at a rate of 2.50 cm · a−1 from 1980 to 2010. Future projections indicate that the SFG depth in the YZRB will continue to decrease in response to future warming. The present SFG may no longer exist by 2180 under the RCP 8.5 scenario (if not considering the transition of permafrost to SFG). The proposed modeling framework provides an important basis for the evaluation of the hydrological cycles (e.g. surface water-groundwater interactions) in cold regions under changing climatic conditions.