Assessment and multi-scenario prediction of ecosystem services in the Yunnan-Guizhou Plateau based on machine learning and the PLUS model

Introduction Machine learning techniques, renowned for their ability to process complex datasets and uncover key ecological patterns, have become increasingly instrumental in assessing ecosystem services. Methods This study quantitatively evaluates individual services—such as water yield, carbon storage, habitat quality, and soil conservation—on the Yunnan-Guizhou Plateau for the years 2000, 2010, and 2020. A comprehensive ecosystem service index is employed to assess the overall ecological service capacity, revealing spatiotemporal variations in services and exploring the trade-offs and synergies among them. Additionally, machine learning models identify the key drivers influencing ecosystem services, informing the design of future scenarios. The PLUS model is used to project land use changes by 2035 under three scenarios—natural development, planning-oriented, and ecological priority. Based on the land use simulation results for these scenarios, the InVEST model is applied to evaluate various ecosystem services. Results During 2000-2020, ecosystem services on the Yunnan-Guizhou Plateau exhibited significant fluctuations, driven by complex trade-offs and synergies. Land use and vegetation cover were the primary factors affecting overall ecosystem services, with the ecological priority scenario demonstrating the best performance across all services. Discussion The research integrates machine learning with the PLUS model, providing more efficient data interpretation and more precise scenario design, offering new insights and methodologies for managing and optimizing ecosystem services on the Yunnan-Guizhou Plateau. These findings contribute to the development of more effective ecological protection and sustainable development strategies, applicable to both the plateau and similar regions.