Exploring environmental drivers of spatiotemporal variability in net primary productivity increase across steppe ecosystems

Steppe ecosystems play a crucial role in maintaining the ecological balance of terrestrial environments. However, their spatiotemporal variability and the differential responses of their subtypes to environmental drivers remain insufficiently understood. This study integrated the Updated Vegetation Map of China (1,1,000,000) with remote sensing and statistical data, employing trend analysis and piecewise structural equation modeling (piecewiseSEM) to address this gap. Our results reveal that 69.44% of the steppe area experienced a significant increase in net primary productivity (NPP) between 2001 and 2023, with an average annual growth rate of 1.61 g C m −2 yr −1 . This increase followed a southwest-northeast gradient, with the four steppe subtypes exhibiting distinct response patterns consistent with this spatial trend. NPP in the meadow steppe was primarily driven by soil fertility under favorable hydrothermal conditions. In contrast, typical steppe productivity was influenced by the interaction between temperature and precipitation. Desert steppe NPP was constrained by limited water availability, whereas productivity in the alpine steppe increased due to the alleviation of temperature constraints. Overall, soil factors emerged as the dominant drivers of NPP variation. Climate change and human activities exerted significant indirect effects, moderated by soil properties that buffered their impacts and contributed to the stabilization of long-term NPP trends. These findings offer valuable insights for the management of China's steppe ecosystems, supporting sustainable land-use planning and enhancing the broader understanding of ecosystem resilience and productivity in arid regions. • Identified spatial-temporal variability across different steppe subtypes. • Compared response patterns of steppe subtypes to environmental drivers. • Found soil factors as dominant drivers of steppe NPP variation. • Quantified direct, indirect, and total effects of environmental drivers.