Poyang Lake, as the largest freshwater lake in China and a typical floodplain lake, plays a vital role in maintaining the hydrological and ecological balance of the middle and lower reaches of the Yangtze River. In recent decades, under the combined influence of climate change and human activities, water area of Poyang Lake has undergone significant spatiotemporal changes. However, owing to the difficulty of traditional methods in effectively decomposing complex non-stationary signals across time scales, the multi-scale characteristics and driving mechanisms of these changes remain unclear. This study integrated multi-source remote sensing images, hydrological and meteorological observations data, and employed the empirical mode decomposition method to systematically analyze the dynamic changes of the water area of Poyang Lake from 2000 to 2022. This study constructed a comprehensive analysis framework consisting of “multi-temporal monitoring–multi-scale decomposition–multi-factor correlation” to quantitatively reveal the shrinking trend of the water area of Poyang Lake from 2000 to 2022 (with an average annual reduction of 28.02 km 2 ) and the phased transformation that occurred around 2011. The study clarifies the scale-dependent nature of driving factors, revealing the dominant role of different meteorological elements in lake water area changes across various temporal scales. Spatially, the representativeness of hydrological stations shifts systematically with scale, reflecting the spatiotemporally differentiated impacts of upstream basin processes and Yangtze River interactions on lake hydrology. Furthermore, the sharp decline in sediment discharge at Hukou Station serves as an integrated indicator of human activities, revealing profound alterations in river-lake material exchange processes caused by sand mining and reservoir operations, thereby providing critical process-based evidence for understanding lake shrinkage. Simultaneously, the research methodology offers a transferable analytical paradigm for understanding the eco-hydrological processes of floodplain lakes worldwide facing similar climatic and anthropogenic pressures. • Introduces a comprehensive multi-scale framework integrating EMD to decode floodplain lake dynamics across time. • Reveals scale-dependent of climatic drivers, shifting from precipitation to evapotranspiration with scale. • Demonstrates spatially shifting hydrological station representativeness, uncovering inherent spatial heterogeneity. • Quantifies human-induced sediment changes as a proxy indicator of lake shrinkage.
Wang et al. (Thu,) studied this question.