Spatiotemporal dynamics and driving mechanisms of wetlands in arid regions of Xinjiang

Wetlands in arid regions provide essential economic, cultural, and climate change mitigation services. However, studies that comprehensively consider the overall wetland dynamics of Xinjiang, a typical inland arid region, are limited. We analyzed the spatiotemporal dynamics and driving factors in wetlands in Xinjiang from 1990 to 2020. Wetland maps were generated at 10-year intervals using the Google Earth Engine cloud computing platform. The complex relationships between natural environmental factors and human activities influencing wetland change were quantitatively analyzed using the partial least squares-path modeling (PLS-PM). The overall accuracy of wetland mapping exceeded 0.94, effectively classifying wetlands into five categories: water, marsh, swamp, saline wetland, and flooded flat. During the study period, the total area of wetlands increased by 1,958.58 km2. Specifically, the area of water bodies expanded by 2,988.40 km2, while the areas of marshes, saline wetlands, and flooded flats significantly decreased. PLS-PM analysis revealed that the driving factors had four direct impact paths and nine indirect impact paths on wetland changes. Notably, anthropogenic factors emerged as the primary drivers of wetland loss. The influence of topography on wetland distribution shifted from negative to positive due to interactions with climatic and soil factors. These findings not only deepen our understanding of the spatiotemporal dynamics of wetlands in arid regions but also provide a scientific basis for developing effective conservation and restoration strategies.