Based on grass surface temperature (GST), air temperature (AT), ground temperature (GT), cloudiness and precipitation et al data from 38 stations in Tibet, China from 2007 to 2020, the spatial and temporal variation characteristics of GST and its influencing factors were analyzed using climate statistical diagnostic methods, for scientific study of local grassland ecosystems and professional meteorological services. The results showed that the annual mean GST in Tibet showed a decreasing distribution from southeast to northwest. There was a significant negative correlation between GST and altitude, with each 100 m increase in altitude, the seasonal mean GST decreased by 0.44-0.70 ℃, and the annual mean GST decreased by 0.58 ℃; there was also a significant curve relationship with latitude, and the annual and seasonal GST increased (decreased) with increasing latitude in the area south (north) of 29.3°N. The GST of each station showed a peak and a valley of daily variation, the daily minimum value occurred at 07:00—08:00 Beijing time (BJT), the daily maximum value occurs at 14:00 BJT; the monthly average minimum value happened in January, the monthly average maximum value appeared in June or July. The seasonal variation of GST, 76% of the stations showed the climatic characteristics of summer>spring>autumn>winter. The annual range of GST in Tibet was 21.4 ℃, which was 3.1 ℃ larger than the annual range of AT; the diurnal range of GST reached 35.7 ℃, much higher than the diurnal range of AT, which larger was 21.6 ℃. The difference value between GST and AT was the largest in summer, followed by spring, and the two were closer in winter; the difference value between GST and GT was the largest in spring, followed by summer, and the smallest in winter. In spatial distribution, the monthly mean GST was significantly positively correlated with AT and GT, and negatively correlated with mean wind speed and snow days. There was a significant negative correlation between the effect of snow depth on GST except in winter. In most months, the relationship between mean GST and total cloudiness, low cloudiness and precipitation was not significant. 86.8% of the stations had significant negative correlation between hourly GST and precipitation from May to September.