Evaluating the suitability of vegetation natural regeneration in the Yellow River Basin based on a coupled PLUS and Monte-Carlo simulation model

Vegetation natural regeneration (VNR) constitutes a prioritized pathway for ecological restoration, yet its spatial implementation is often impeded by landscape-scale stochasticity and environmental heterogeneity. To evaluate the suitability of VNR across the Yellow River Basin (YRB), China, a framework coupling the Patch-generating Land Use Simulation (PLUS) model with Monte Carlo stochastic simulations was developed at a 1 km resolution. Utilizing multi-temporal land-use data, the Vegetation Natural Regeneration Suitability Index ( S i ) and its prediction uncertainty for the YRB in 2030 were quantified to guide adaptive ecological zoning, and were also analyzed via Geodetector for detecting driving factors. Results showed that the S i of areas with confirmed regeneration ( S i = 0.5775±0.389) exhibited nearly three-fold higher suitability than those with non-regeneration ( S i = 0.1939±0.260). The YRB was classified into three zones according to ( S i ), the low potential zone ( S i =0), medium potential zone (0< S i ≤0.5), and high potential zone (0.5< S i ≤1), which account for 90.45%, 7.11%, and 2.44% of the total area, respectively, with high-potential hotspots (covering 1.58% of the YRB) clustered in middle-reach of the YRB. Spatial differentiation of S i was dominated by vapor pressure deficit, potential evapotranspiration (PET) and mean annual temperature, as well as actual evapotranspiration (AET). with the coupling effect between AET and PET ( q =0.63) exerting the strongest interactive explanatory power. This study provides an approach for quantitatively evaluating suitability of vegetation natural regeneration, and can serve as a decision-support tool for implementing differentiated natural restoration strategies across complex and ecologically fragile basins.