In this study,the spatial and temporal variation of snow cover in the Tianshan Mountains and its relation to temperature and precipitation were analyzed by using linear tendency estimate and correlation analysis, based on the Moderate Resolution Imaging Spectroradiometer ( MODIS) snow cover,MODIS land surface temperature data and China Meteorological Forcing Dataset during 2001 - 2015. The results demonstrated that: ( 1) There was obvious difference in monthly distribution of the snow cover percentage ( SCP) for different elevation zones; the annual distribution of snow cover percentage presented a unimodal pattern below 4 000 m and over 5 000 m altitude; the maximum and minimum snow cover percentages occurred in winter and summer below 4 000 m,while the maximum and minimum values appeared in summer and winter over 5 000 m,respectively; the seasonal cycle of snow cover percentage exhibited a bimodal distribution between 4 000 m and 5 000 m altitude, with a peak in spring and autumn; during 2001 - 2015,there was a slight decreasing trend in annual mean snow cover percentage in Tianshan Mountains. However,there were differences in change trend of snow cover percentage in various seasons: there was a slight increasing trend in autumn,while a decreasing trend had been found in other seasons with more obvious in winter and summer. ( 2) The snow cover frequency ( SCF) presented a spatial pattern from west high to east low and north high to south low due to water vapor sources and topography, and showed a positive correlation with elevation; the SCF had showed a decreasing trend during 2001 - 2015 in the Tianshan Mountains with more obvious in the altitude range between 3 600 m and 4 600 m. ( 3) In spring and summer,the temperature was the main cause of the change of snow cover percentage,and showed a negative correlation with snow; while in autumn and winter,precipitation was the main cause of the snow change,and had a positive correlation with snow. ( 4) Overall,the snow cover frequency showed a negative correlation with annual mean temperature and a positive correlation with precipitation; although there were great spatial differences in the degree of correlation and the significance level,the impact of temperature was larger than that of precipitation on SCF change.