[Analysis of Spatiotemporal Variation Characteristics and Driving Forces of NPP in Shanxi Province from 2000 to 2020 Based on Geodetector].

The net primary productivity (NPP) of vegetation, as an indispensable element in carbon cycle studies, characterizes plant growth status. This study applied MODIS NPP products from 2000 to 2020 and multi-source data on elevation, slope, precipitation, temperature, land use, and population density in Shanxi province. We used trend analysis, correlation analysis, and geographic probes to explore the spatial and temporal evolution characteristics and driving factors of NPP in Shanxi province and its national planned coal-mining areas. The results showed that: ① the overall NPP exhibited a fluctuating upward trend from 2000 to 2020, with an average rate of increase (in terms of C) of 6.7 g·(m2·a)-1. The total NPP varied significantly among different land types, with arable land>forest land>grassland>construction land>water area>unused land. ② The spatial heterogeneity of NPP changes was obvious, with lower NPP values in the western and northern regions and higher average NPP values in the eastern and southern regions; the NPP comparison of three major coal bases showed that Jindong coal base>Jinzhong coal base>Jinbei coal base. ③ The correlation between NPP and precipitation was high, with 62.2% of regions having a significant correlation (P<0.05), mainly in central and eastern Shanxi province. The relationship between NPP changes and temperature was weak, with only 1.10% of regions having a significant correlation (P<0.05). ④ The comparison of the q-means of each factor in different years based on geographic probes showed that precipitation (0.165)>land use (0.124)>population density (0.085)>slope (0.080)>elevation (0.064)>air temperature (0.024), further indicating that precipitation was the dominant driver of NPP changes over the years. 5 The influence of the two-factor interaction was significantly higher than that of the single factor, and the influence of anthropogenic factors was gradually increasing. From 2000 to 2020, the interaction factor precipitation∩population density (0.275) with the highest explanatory power replaced precipitation∩temperature (0.385) as the interaction factor precipitation with the highest explanatory power.