Abstract: Objectives: Net primary productivity (NPP) of vegetation is an important factor to monitor the carbon cycle process of regional terrestrial ecosystem and regional carbon sources and sinks. Estimating NPP of vegetation and revealing its evolution characteristics and its response to natural and human factors are of great significance to the sustainable development of regional ecosystem. Methods: Taking Yinchuan as an example, based on Carnegie-Ames-Stanford Approach(CASA) model combined with sub-meter high-resolution remote sensing images, climate, land use and other data, the NPP of vegetation in different seasons from 2015 to 2021 in the study area are estimated and its temporal and spatial evolution characteristics are analyzed. Gradient Boosting Decision Tree (GBDT) model is used to reveal the nonlinear response and threshold effect of natural factors, human activities and land use on NPP of vegetation. Results: 1) The NPP of vegetation in Yinchuan was 0-753.951 g C·m^-2·a^-1, and the high value areas were concentrated in the agricultural irrigation areas along the Yellow River. The detailed spatial distribution maps of NPP of field scale and urban block scale were obtained with high-resolution images used in the estimation of NPP, which also improves the estimation accuracy. 2) The seasonal variation trend of vegetation NPP in Yinchuan was consistent with vegetation phenology and crop growth cycle. In the third quarter of 2015-2021, the NPP of vegetation showed an overall increasing trend, while the local NPP decreased significantly due to the occupation of cultivated land by urban expansion, while the NPP of vegetation in ecological protection areas increased significantly. 3) Natural factors such as NDVI soil organic carbon density, elevation, rainfall, etc. had significant threshold effects on NPP, and the relative contribution of NDVI is the largest, when it exceeded 0.6, it had no obvious promotion effect on NPP of vegetation. The threshold of NPP of dry land was higher than that of paddy field, and construction land played a positive role in NPP when urban green foundation in the construction land reaches a certain proportion. When human activities are in a certain range, they can promote NPP of vegetation to a certain extent. When the population exceeds the threshold, the larger the population, the lower the ecological carrying capacity, and the lower the NPP of vegetation. Conclusions: In this study, high-resolution images are used to estimate vegetation NPP, which improves the refine degree of spatial distribution of vegetation NPP. The seasonal characteristics of vegetation NPP in Yinchuan city are significant, and the changing trend of NPP with seasons is consistent with vegetation phenology and crop growth cycle. Natural factors, land use and human activities in Yinchuan city have significant effects on vegetation carbon sink with significant threshold effects. Natural factors have a greater relative contribution to vegetation NPP than the other two factors.