[Spatiotemporal Evolution of Carbon Storage and Multi-scenario Prediction in the Yunnan-Guizhou Plateau Based on the InVSET-Ridge Regression-PLUS Model].
摘要
As an important indicator for measuring the carbon sequestration capacity of ecosystems，carbon storage is of great significance for alleviating global climate change. By taking advantage of machine learning and ecosystem service models，an integrated analysis framework based on the InVEST-Ridge Regression-PLUS model was constructed to conduct a quantitative analysis of the spatio-temporal evolution characteristics and driving mechanisms of carbon storage in the Yunnan-Guizhou Plateau from 2000 to 2020，and future scenarios were designed to predict the changing trends of regional carbon storage under different land use paths. The results show that：Firstly，from 2000 to 2020，the carbon storage in the Yunnan-Guizhou Plateau generally presented a slow growth trend，and the growth rate continued to decline，showing a distribution pattern of "higher in the south and lower in the north." Secondly，vegetation coverage was a crucial determining factor for carbon storage in this area，and the conversion between different land use types affected the spatial distribution of carbon storage. Thirdly，in the future scenario simulation，the carbon storage under the carbon sink enhancement scenario performed best，effectively verifying the effects of ecological projects such as the conversion of farmland to forest and grassland restoration，providing a scientific basis for the dynamic assessment and optimization of carbon storage in the Yunnan-Guizhou Plateau and similar karst areas.