Based on air temperature, ground temperature, snow cover duration and snow cover depth records from 61 weather stations in Liaoning Province from 1971 to 2016, the effect of snow cover on ground-air temperature difference were analyzed. The analyses show that the difference of ground temperature caused by different observation ways before and after replacement of automatic weather station plays a dominant role for ground-air temperature difference increasing significantly; ground-air temperature difference change was influenced by snow cover duration and snow cover depth obviously. In the area with long snow cover duration, snow cover increases the albedo, resulting in snow cover surface temperature decreasing and snow-air temperature difference increasing. Besides, due to the insulating effect of snow on the ground, ground-air temperature difference increases significantly. Therefore, ground (snow)-air temperature difference has a negative correlation with snow cover duration before replacement of automatic weather station, and ground-air temperature difference has a positive correlation with snow cover duration after the replacement. Variation coefficients of daily mean ground-air temperature difference with daily snow depth changes between 0.045-0.858 ℃·cm-1 from station to station. In the area where annual snow cover duration is less than 40 days and snow cover depth is below 10 cm, the heat preservation effect of snow increases significantly with the increase of snow depth. In the area where annual snow cover duration is more than 40 days and maximum snow cover depth is above 10 cm, when snow cover depth below 10 cm, the heat preservation effect of snow has increased significantly with the increase of snow depth, however, when snow cover depth above 10 cm, the heat preservation effect of snow has increased slowly with the increase of snow depth.
Because of complicated landform and harsh climate, the meteorological observation data were very scarce in northwestern Tibet, which is characterized by vulnerable ecological environment. Drought monitoring by remote sensing could obtain drought conditions with continuous changes on the surface, which was of great significance for the arrangement of agricultural and husbandry production in this area. In this paper, we try to reveal the drought in northwestern Tibet based on primary data, standard products (surface temperature, vegetation index) of FY-3A/VIRR and drought monitoring and TVDI (temperature vegetation drought index). Then, the monitoring results has been compared, respectively, the results of the EOS/MODIS data monitoring, field measured soil moisture data during this period, and the precipitation data from weather stations. The results were compared with those monitoring results based on MODIS data, measured soil moisture, the meteorological observation data. It is found that the TVDI calculated by FY-3A/VIRR showing a significant negative correlation with the measured soil moisture and accumulated precipitation, which had passed the remarkable test of the 0.01 level; they were basically the same that the spatial distribution characteristics of TVDI drought grades estimated by remote sensing using FY-3A/VIRR data and EOS/MODIS data. Therefore, the FY-3A/VIRR data could be used to carry out drought monitoring by remote sensing in northwestern Tibet instead of EOS/MODIS data, also available for providing data support for guiding agricultural and husbandry production in northwestern Tibet.