Data on microtopography-vegetation effects in the Yangguan sample plot in China

Vegetation distribution patterns result from the combined effects of various ecological factors. Traditional research has primarily focused on macro-climate scales (e.g., satellite remote sensing), with limited attention to the heterogeneity of vegetation spatial distribution under the same climatic background at micro-scales (e.g., Unmanned Aerial Vehicle/UAV scale). This heterogeneity is particularly pronounced in hyper-arid regions, making it crucial for understanding microtopography-mediated water redistribution and complementing micro-scale details of macro-vegetation patterns. This study: (1) took the Yangguan plot (94.21°E, 40.00°N; 650×750 m in dimension) in the hyper-arid Gobi Desert of East Asia as a case study. Based on UAV photogrammetry, we employed Structure from Motion and Multi-View Stereo (SfM-MVS) techniques to achieve decimeter-resolution microtopography reconstruction. (2) Based on 1,000,000 random sampling points and 3,001 shrub centroids, we extracted 17 microtopographic, hydrological, and vegetation indicators for statistical modeling. In univariate models, hydrological indicators such as Flow Accumulation (FA) and Topographic Wetness Index (TWI) were identified as the primary indicators influencing vegetation distribution (R² > 0.1). Among multivariate models, the random forest model demonstrated the best predictive performance (maximum R² = 0.88). (3) We preliminarily revealed the mechanisms by which microtopography regulates water and vegetation distribution patterns: windward and leeward slopes affect moisture supply, while topographic relief shapes surface runoff to redistribute water resources. In catchment areas with enriched water, vegetation coverage increases significantly. This study provides new micro-scale perspective for the conservation of vegetation in hyper-arid regions under global change.