Rainfall is one of the essential climatic elements that affects the instability of geostructures. In recent years, extreme events, such as rainstorms in semiarid to arid regions subjected to long drought periods, have exacerbated the failure of natural and artificial earth fill slopes. Such severe instabilities were reported in parts of a railway embankment in southeastern Iran during the 2020-spring rainstorm. Most hazards, such as landslides due to floods, are specific extreme events in the north of Iran. However, most of the analyses have been focused on the maximum rainfall amount obtained from the rainfall station output data as the worst-case scenario, which has been proved that may not always be the case. More importantly, the distribution of rain stations in remote areas is very sparse in the order of tens to hundreds of kilometers. In other words, the uncertainty of rainfall events becomes even more pronounced. Therefore, the primary purpose of this study is to explore the application of an alternative method based on evenly distributed satellite data. To achieve this goal, monthly precipitation from 2015 to 2020 is compared between Global Precipitation Measurement (GPM) data and rain gauge data recorded in the southern and northern regions. Afterward, rainfall data are examined based on the quantitative statistical criteria, and the most appropriate probability distributions are estimated and proposed accordingly. The Generalized Extreme Value (GEV) distribution is the most suitable one for low-to-intermediate rainfall. The Logistic and Frechet distributions are presented for heavy rain. The satellite precipitation distribution of the southern region is eventually modified based on the more reliable and robust correlations for the northern region. Results suggest that the underestimated rainfall amount obtained from traditional rain station records could be a potential reason for the unconventional design and consequential damage to some infrastructure in the southern region.