Spatiotemporal patterns and regional differences of lake ice phenology across mainland China
摘要
Lake ice phenology (LIP) is one of the most sensitive and direct surface indicators of climate change, exerting substantial impacts on regional hydrological processes, ecosystems, and human activities. However, existing studies in China have primarily focused on regional scales such as the Tibetan Plateau, Xinjiang, and the Northeast China, lacking a systematic spatiotemporal analysis based on representative lakes across the entire country. In this study, 20 typical lakes across mainland China were selected, and key LIP parameters were extracted from CETB passive microwave remote sensing (PMRS) data using the Bayesian Ensemble Change Detection (BECD) algorithm, with validation performed using MODIS snow-cover products. Based on these datasets, the spatiotemporal variations and dominant driving mechanisms of LIP from 1978 to 2022 were systematically investigated. The results show that over the past four decades, 85% of the studied lakes exhibited delayed freeze-up, advanced break-up, and consequently shortened ice-cover duration. Significant regional differences were observed: the Tibetan Plateau showed the strongest changes, followed by Xinjiang, while Northeast China displayed the weakest responses. Temperature was identified as the primary factor influencing the timing of ice melt, whereas wind speed played a key regulatory role during freeze-up. Precipitation and other factors contributed to regional variability in lake ice responses across 85% of the studied lakes. By integrating multisource remote sensing with a consistent methodology, this study provides a comprehensive nationwide assessment of LIP dynamics and their climatic drivers, offering important scientific insights into how Chinese lakes respond to climate warming and supporting environmental monitoring in sensitive cold regions.