Satellite imagery provides significant data reliable to study weather and climate elements through frequent observation for these elements using remote sensing systems that care about such researches, like NOAA satellite which is considered the oldest climate remote sensing system. Satellites resemble clues to mathematical models so as to define pressure distribution, temperature thickness and density of atmosphere layers, calculating wind movement indirectly by measuring cloud movement from fixed satellites, as well as using telescopes with high capacity in both visual and thermal fields, as it records temperature gradation within cloud layers, comparing them and conclude simple primary results. This research depended on data from The Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN) project under California University meteorological, hydrological and remote sensing observation center which acts upon extracting rainfall data from satellite imagery using Neural Network Models where data are available in four patterns among which the data type PERSIAN CCS has been relied upon being a highly accurate spatial data where cell dimensions reach 4 km × 4 km. Research period extended between (2015 – 2021). During this period these digital files have been analyzed and modelled for purpose of defining spatial and time variations in rainfall expected along the study period and its impacts in flash floods in the study area basins.