Spatiotemporal Dynamics and Driving Factors Analysis of Vegetation Net Primary Productivity in China Using Remote Sensing and Machine Learning

Abstract

Net primary productivity (NPP) is a key indicator of ecohydrological processes and vegetation responses—including in croplands—to hydrological and carbon dynamics under environmental change. However, existing studies rarely jointly analyze seasonal, interannual, and extreme-event NPP responses or account for spatial heterogeneity and autocorrelation, limiting ecohydrological attribution. This study integrates remote sensing and climate data (2001–2020) with time-series, statistical, machine-learning, and spatial analyses to assess NPP dynamics across China’s nine major river basins (SLR, HHR, HER, YWR, YER, PLR, SEB, SWB, and ILB). Spatial patterns show a southeast-to-northwest gradient, with peak NPP in humid PLR/SEB and minima in arid ILB; growing seasons account for 59.5%–98.9% of annual totals. Temperate basins show significant increases (3.60–5.98 g C m⁻² yr⁻¹, <italic>p</italic> < 0.05) from restoration, while ILB declines (−0.78 g C m⁻² yr⁻¹, <italic>p</italic> < 0.05) under drought stress; the 2011 drought reduced temperate NPP by 4.6%–12.0%, with humid basins showing greater resilience. SHapley Additive Explanations analysis identifies vegetation metrics (fractional vegetation cover and LAI) and hydrological factors (precipitation, temperature, and ET) as dominant, topography-modulated drivers. Among four candidate models, light gradient boosting machine performs best (<italic>R</italic>² = 0.87 and RMSE = 78.76 g C m⁻² yr⁻¹), significantly outperforming EXtreme Gradient Boosting (<italic>R</italic>² = 0.85), random forest (<italic>R</italic>² = 0.80), and a linear regression baseline (<italic>R</italic>² = 0.62; <italic>p</italic> < 0.05); incorporating spatial autocorrelation (Moran’s I = 0.32–0.90, <italic>p</italic> < 0.001) via a spatial lag model further improves accuracy (<italic>R</italic>² = 0.89 and RMSE = 70.63 g C m⁻² yr⁻¹), and ablation experiments attribute the largest gains to spatial dependency and hydrological features. These findings support adaptive ecohydrological strategies for ecosystem resilience and inform crop monitoring in intensive-agriculture basins, such as YWR and YER.