Reduction in vegetation cover caused by human activities has a great impact on soil temperature. It is important to assess how soiltemperature responds to reduction of vegetation height and density. In this paper we first report the trends of mean annual soilsurface and air temperatures recorded at the meteorological stations near the Ecological Research Station for Grassland Farming(ERSGF) from 1961 to 2007, then we setup an experiment using reed (Phragmites australis) stalks with different heights anddensities to simulate effects of different vegetation height and density on soil and air temperatures. The warming rates of the meanannual soil and air temperatures were 0.043 and 0.041Ã‚°C aÃ¯Â€Â1, respectively. Changes of soil temperature were characterized by bothincreased mean annual maximum and minimum soil temperatures. At the experimental site, mean daily temperature, mean dailymaximum soil and air temperatures increased significantly. In contrast, mean daily minimum soil temperature increased significantlywhile mean daily minimum air temperature decreased significantly as the height and density of reed stalks reduced duringthe experimental period. Mean diurnal soil temperature ranges were smaller than mean diurnal air temperature ranges. These resultshighlight that the importance of vegetation cover on soil and air temperatures.