Vegetation Dynamics and Driving Mechanisms in the Western Sichuan Basin Revealed by An Integrated Analytical Framework Based on Trend Analysis and Machine Learning Algorithms
Year: 2025
Type: preprint
Abstract: Identifying vegetation trends and their driving mechanisms under the combined influence of natural and anthropogenic factors is essential for understanding ecosystem dynamics, particularly in ecologically sensitive mountainous regions. This study developed an integrated framework combining trend analysis and interpretable machine learning to evaluate vegetation dynamics in the Western Sichuan Basin from 2000 to 2020. The results revealed a significant improvement in regional vegetation cover, with a growth rate of 0.00152 yr -1 and 81.00% of the area showing positive trends. The selected environmental and anthropogenic variables exhibited distinct nonlinear response patterns and threshold effects. Elevation and temperature were identified as the dominant drivers of vegetation change, while the impact of grazing intensity became increasingly prominent over time. SHAP interaction analysis further revealed that favorable topographic and climatic conditions can amplify positive vegetation responses to human activities. These findings highlight the value of interpretable machine learning for uncovering complex ecosystem processes and provide a scientific basis for guiding ecological restoration and sustainable land-use policies in mountainous landscapes.
Institution: Southwest Jiaotong University
Language: English
Subfield: Ecology
Field: Environmental Science
Domain: Physical Sciences
SDGs: Life on land, Climate action
Open Access status: gold
OpenAlex ID: W4415023416
Cites:
Cited by:
Related to: