Thunder storms or lightning storms are kind of storm that generally associated with convective clouds and usually associated with floods, hail, strong winds, and flash flood. These characteristics, together with turbulence and icing make this one of the most destructive atmospheric phenomenon in flight operations. The purpose of this research is the detection of deep convective clouds (DCC) and severe thunderstorm systems in winter rainfall in the South-East of Iran. To achieve this objective, we used different data, including TRMM / PR and LIS, NOAA / AVHRR satellite images and Synoptic stations data. To determine the exact location of storm, the infrared brightness temperature of NOAA/AVHRR images were used. DCC pixels are identified by their cold infrared (IR) brightness temperatures. The result showed that the vast majority of the clouds had temperatures less than 260 ° K in the time of storm occurrence, but all of these clouds had not been occurrence of severe lightning activity. With coincidence of brightness temperature and LIS data, it was found that the clouds which had been temperature below 240 ° K, are more appropriate for lightning occurrence. There are two reasons that there is no lightning in these clouds. First, these clouds had been lightning activities, but it had not seen with LIS sensor because of short view time (about 90s). Second, these clouds in the upper levels had been adequate extension but in middle or low levels did not have adequate extension. If the second reason was true, the brightness temperature and reflectance data at the same time can be used to correct this problem. Therefore, with using brightness temperature and reflectance data we obtained better results. Thus, by applying the threshold of brightness temperature less than 240 degrees Kelvin and reflection more than 50%, the areas that had been deep convection and intense lightning, were identified. Also, the results showed that the regions that had been intense lightning activity, had been maximum rainfall.