In this study, daily temperature data from 44 synoptic stations were prepared over the vast territory of Iran during the observation period (1981-2010). Then, daily temperature data for two future periods (2041-2070 and 2071-2099) were simulated by using the statistical downscaling model of SDSM based on HadCM3 SRES A2, B2. After checking the simulation accuracy, 95th percentile threshold was used to identify the heat wave occurrences in the warm seasons of the year. Therefore, it is considered as a heat wave if the daily maximum temperature is higher than the 95th percentile threshold with at least three or more consecutive days. The results showed that the seasonal averages of heat waves intensity are expected to increase and the seasonal numbers of heat waves frequency are expected to decrease according to the output of all scenarios in coming decades. While the future duration of heat waves does not follow from stable patterns, however, long-lasting heat waves are more expected to happen in northern inland than southern coasts. Also, the lowland regions of southern coast are expected to have the lowest increasing changes in the heat waves frequency than northern highlands. While the southern coasts which experienced the strongest heat waves of the last decades are expected to have the worst conditions of future heat waves owing to the more potential availability of atmospheric humidity. Overall, the future occurrences of heat waves are expected to happen stronger and longer lasting along with lower frequencies in northern than southern latitudes.