Monitoring changes in the net primary productivity (NPP) of vegetation and its response to climatic changes is integral to gain insight into the carbon cycle mechanism and facilitate human well-being. Nonetheless, past studies lacked a month-scaled exploration and largely ignored spatial heterogeneity of the time lag effect. In this study, we focused on the Hengduan Mountain region situated in southwest China and analyzed the variation of NPP and its main driving mechanisms during 2000–2016. Our results revealed that: (1) The annual NPP signified a slightly increasing trend. On the grid-scale, the variation region of NPP was mainly characterized by increasing trends in most of the growth stages, except for July. (2) The response of NPP to climate change presented obvious spatial heterogeneity and complexity. The increasing NPP of the growing season in the northeast part and areas around Xichang can be attributed to temperature rising, while the increase of temperature in the central valleys led to a decline in NPP. Furthermore, decreasing precipitation led to the decrease in NPP in the western valleys. Thus, adequate water should be provided for the vegetation in western valley, and drought-tolerant plants should be choice in the central valley. (3) Changes in NPP were more influenced by temperature than precipitation, which was more evident on monthly scale, especially during April and May, and for the southeast and northeast part in June and August. (4) The lag time of NPP variation in relation to temperature changes was shorter than that of precipitation in most growth stages, indicating that NPP is more sensitive to temperature. In the colder region or growth stages, the lagged time of the response of NPP to temperature was shorter, which was more obvious on the monthly scale. This research will have important implications for ecosystem management. © 2021