Temporal and spatial change of Net Primary Productivity of vegetation and its determinants in Hexi Corridor

The variation of vegetation Net Primary Productivity（NPP） has an important impact on the global carbon cycle in arid and semi-arid regions. The response of NPP to climate change in this region shows great spatial and temporal heterogeneity, while its driving mechanism is not clear. In this study, the Hexi Corridor in China was selected as study area, and the NPP from 2002 to 2018 was estimated using a random forest algorithm. The effects of climate and human activities on NPP were calculated based on the partial derivative method. The results showed that（1） random forest algorithms presented good performance in the NPP estimation in arid and semi-arid areas.（2） The average annual NPP of Hexi Corridor from 2002 to 2018 was 153.32 gC m-2 a-1, and the total amount was 37.468 Tg C/a, which indicated a decreasing distribution pattern from southeast to northwest. During the study period, the NPP showed an increasing trend of 2.37 gC m-2 a-1（P=0.09）.（3） 52.51% of the NPP variation in the Hexi Corridor was contributed by climate factors, and 47.49% was contributed by human activities.（4） For the impact of climate change on the NPP, precipitation dominated 72.21% of the area, and temperature contributed 73.71% of the amount of NPP variation. The former affected the NPP change pattern and the latter dominated the amount of NPP variation. Both warming and humidification are conducive to the increase of NPP in this area. With the warm and humid climate in the Northwest China, the vegetation in the Hexi Corridor will continue to increase. This study is helpful for understanding the response mechanisms of the NPP to climate change in arid and semi-arid areas, and provides a theoretical basis for decision making in adaptation to climate change.