Objective As a typical ecologically fragile area, the evolution law and core driving mechanism of vegetation net primary productivity in the Loess Plateau are not yet clear against the backdrop of global climate change and the implementation of the Grain for Green Project for 20 years. This study focuses on the spatiotemporal dynamics, climate response, and future trends of NPP, aiming to provide scientific support for precise regulation of regional ecological restoration.

Method Based on the Carnegie Ames Stanford Approach model, the NPP of the Loess Plateau from 2000 to 2020 was estimated. Combining Theil Sen trend analysis, Mann Kendall significance test, Hurst exponent, and Pearson correlation analysis, the long-term spatiotemporal evolution characteristics of NPP and its response mechanism to climate factors were systematically explored.

Result (1) The annual average value of NPP on the Loess Plateau is 318.49 g/(m²·a), which shows a spatial distribution pattern of high in the southeast and low in the northwest. The high-value areas are concentrated in areas such as southern Shanxi and the Guanzhong Plain with superior water and heat, while the low value areas are distributed in areas such as the arid and rainless Ordos Plateau. (2) From 2000 to 2020, the overall NPP showed a significant upward trend, with an average annual growth rate of 6.34 g/(m²·a). Among them, forest land had the fastest growth rate, at 7.63 g/(m²·a), with a highly significant increase in the proportion of regions accounting for 60.77%, concentrated in key implementation areas such as Yulin and Yan'an in northern Shaanxi, which confirms the significant effectiveness of ecological engineering. (3) The Hurst exponent prediction shows that 61.08% of the regions' NPP may shift from an upward trend to a downward trend in the future, and 35.28% of the regions may continue to rise in the future. (4) Precipitation is the core driving factor for NPP increase, with a positive correlation in 95% of regions, while temperature has an inhibitory effect on vegetation growth in arid areas, clarifying the regional differentiation patterns driven by climate in semi-arid regions.

Conclusion The vegetation restoration on the Loess Plateau has achieved significant results in the implementation of the Grain for Green and Grassland Project for 20 years. In the future, special attention should be paid to the ecological reversal risk in 61.08% of the region. This study provides a scientific basis for the evaluation of regional ecological restoration effectiveness and adaptive management, and is of great significance for improving the global carbon cycle research system and promoting regional sustainable development.