Based on the daily maximum and minimum temperature data from 6 meteorological stations in Northern Tibet from 1961 to 2010,the temporal and spatial variation characteristics of extreme temperature events at grass growing season were analyzed using the ten indices of extreme temperature. The methods of a 10-year smoothing average, linear regression, correlation analysis,and a Mann-Kendall model test were employed to reveal the rate of change,statistical significance of the trends, and mutation points of extreme temperature indices. The analysis showed that maximum value of daily maximum temperature (TXx), minimum value of daily maximum temperature (TXn),minimum value of daily minimum temperature (TNn) and maximum value of daily minimum temperature (TNx) significantly increased at grass growing season during recent 50 years with the maximum TNn increasing rate at 0.40 ℃? 10 a~(-1) Diurnal temperature range (DTR) significantly decreased at a rate of -0.17 ℃ ? 10 a~(-1) whereas the length of growing season (GSL) significantly increased with a rate of 1.81 d ? 10 a~(-1). The numbers of warm days (TX90p) and warm nights (TN90p) significantly increased with a rate of 4.11 and 11.81 d ? 10 a~(-1), respectively(P <0.01). Furthermore,the number of TN90p significantly increased with a rate of 23.03 d ? 10 a~(-1) during 1981 2010. The numbers of cool days (TX1Op) and cold nights (TN1Op) significantly decreased with a rate of 3.33 and 6.69 d ? 10 a~(-1),respectively(P<0.01). In terms of decadal variations, the warm indices (TNx, TX90p,TN90p and GSL) increased while the cold indices (TX1Op and TN1Op) and DTR decreased. It was also found that the mutation of the TXx was in 1984,the other extreme temperature indices (TNn, TNx, TX1Op, TN90p, TN1Op, DTR and GSL) occurred after the 1900s, but no mutation occurred for TXn and TX90p. Extreme temperature indices suggested that the climate change was beneficial for grass growing in Northern Tibet.