Abstract:

Land surface temperature (LST) is an important variable in the surface energy budget,significantly affecting the global agriculture，natural disasters，hydrology,and ecological environment.With global climate change,quantitative research on LST and its related factors become critical in both regional and global scales.The Qilian Mountains in Qinghai and Gansu Province,China,as the boundary between the Qinghai-Tibetan Plateau and Northern Hinterland,is influenced by the continental climate and the climate of the QinghaiTibetan Plateau.It belongs to the temperate semi-arid climate zone and is an important geographical and climatic boundary line in China.The study of LST in the Qilian Mountains is essential for understanding the hydrothermal cycle in the northwest and improving the fragile ecological environment.The climate pattern,vegetation,and different land-use of the Qilian Mountains have been studied comprehensively.However,the spatiotemporal variability characteristics of LST and its response mode to terrain are not yet clear.Besides,the impact of different vegetation types on LST changes need to be discussed.Based on the MODIS LST data from 2000 to 2017,this study used a linear trend method to analyze the temporal and spatial variation of LST in the Qilian Mountains and used Pearson correlation analysis to discuss the regulation of LST by different types of vegetation and the influence of altitude on its change.The results are as follows: (1) the accuracy of MODIS LST products can meet the requirements of temporalspatial variation analysis of surface temperature in the Qilian Mountains.(2) LST of the Qilian Mountains increased at a rate of 0.17 ℃·(10 a)^-1，with two rising periods in 2000-2003 (1.83 ℃·(10 a)^-1) and 2004-2017 (0.14 ℃·(10 a)^-1).The annual average 〖WTBX〗LST〖WTBZ〗 increased from SouthWest to NorthEast,and the significant rising area accounts for 77.26% of the total area.The extremely significant rising areas were mainly distributed in the middle and low altitudes below 3 500 m.In addition,the winter change rate was the highest (0.22 ℃·(10 a)^-1),and the warmer area reached 98.23%.The winter warming had the greatest contribution to the increase of LST in the Qilian Mountains.(3) vegetation had a significant impact on LST changes,cultivated lands had a better inhibitory effect on LST rise,and vegetation in the desert had a stronger regulatory effect on LST.Except for summer，the NDVI and LST were positively correlated in the rest of the season，and the positively correlated areas were mainly concentrated in the highaltitude forest land.(4) the effect of altitude on LST was significantly weak during the summer due to conditions such as vegetation cover changes.The LST in the desert showed the strongest response to altitude,with strong annual fluctuations between 2 500 and 3 500 m.Meanwhile,vegetation degradation and human activities at low and medium altitudes may be the main factors accelerating the increase in LST.

Key words: MODIS LST, MODIS NDVI, vegetation type, Qilian Mountains