The annual and seasonal trend analysis of different surface temperature parameters (average, maximum, minimum and diurnal temperature range) has been done for historical (1971–2005) and future periods (2011–2099) in the middle catchment of Sutlej river basin, India. The future time series of temperature data has been generated through statistical downscaling from large scale predictors of CGCM3 and HadCM3 models under A2 scenario. Modified Mann–Kendall test and Cumulative Sum (CUSUM) chart have been used for detecting trend and sequential shift in time series of temperature parameters. The results of annual trend analysis for period of 1971–2005 show increasing as well as decreasing trends in average ($T_{\text{Mean}}$), maximum ($T_{\text{Max}}$), minimum ($T_{\text{Min}}$) temperature and increasing trends in Diurnal Temperature Range (DTR) at different stations. But the annual trend analysis of downscaled data has revealed statistically significant (95% confidence level) rising trends in $T_{\text{Mean}}$, $T_{\text{Max}}$, $T_{\text{Min}}$ and falling trend in DTR for the period 2011–2099. The decreasing trend in DTR is due to higher rate of increase in $T_{\text{Min}}$ compared to $T_{\text{Max}}$.