Historical evolution and future trends of land cover patterns and structural ecological connectivity in the China-Russia Xingkai Lake Basin

Transboundary wetlands are essential for regional ecological security, yet their landscape structure and connectivity are increasingly altered by land use change. We quantified land cover dynamics, landscape reorganization, and structural ecological connectivity in the China-Russia Xingkai Lake Basin during 2000-2025 and projected changes to 2050. Landsat-based land cover maps were generated using a random forest classifier. Landscape configuration was characterized using PD, LPI, ED, and COHESION, whereas structural ecological connectivity was quantified using the integral index of connectivity (IIC); XGBoost and SHapley Additive exPlanations were used to examine environmental and socioeconomic correlates; and a Markov-cellular automata framework was used to simulate four future scenarios. Wetland area declined by 44.9% and became increasingly fragmented. In China, gross wetland-to-cropland and wetland-to-open surface water transitions were broadly comparable, while total cropland area declined slightly by 0.4% at the net level. In Russia, cropland increased by 48.8%, while gross wetland-to-open surface water conversion was more prominent, although this mapped pathway may partly reflect hydrological variability and classification uncertainty. At the 500-m link threshold, basin-scale IIC declined by 16.5%, from 0.1647 to 0.1376, alongside reduced core habitat area and increased patch number. Land cover transitions were associated with different combinations of topographic, climatic, and human-related variables across the border. By 2050, ecological protection retained the largest wetland and core habitat areas and produced the highest IIC (0.1528), 17.6% higher than under agricultural priority (0.1299). These findings highlight the need for coordinated wetland retention, core habitat protection, and sustainable land use management across the transboundary basin.