Analysis of Area Changes and Driving Factors in Chirui Lake

Changes in lake area directly reflect the state of regional hydrological cycles and ecological balance. However, the lack of long-term monitoring data and the complexity of driving factors make it difficult to formulate effective management strategies. This study used landsat imagery and the random forest (RF) algorithm to construct a time series of the lake area of Chirui Lake from 1990 to 2024, revealing a counter-seasonal phenomenon in which the lake area during the low-flow period was significantly larger than that during the high-flow period. To clarify the driving mechanisms behind this phenomenon, a comprehensive analysis of climatic and hydrological factors and lake area was conducted using methods such as the Pettitt test, Morlet wavelet analysis, principal component analysis (PCA), structural equation modeling (SEM), and long short-term memory (LSTM) neural networks. The results indicate that, from the perspectives of climate and hydrology, sub surface runoff is likely to have played a significant role in the area changes of Chirui Lake, and that between 2025 and 2029, the lake’s surface area is likely to continue to fluctuate significantly between 0.4 km2 and 0.6 km2. This study explored the influence of climate and hydrology on the changes in the area of Chirui Lake and provides five-year forecast of lake surface area, however, given the inherent uncertainties in the forecast, these findings should be viewed as preliminary references rather than direct decision-making tools for lake management.