The following article pertains to use the first observed hourly precipitation records and estimated by satellites, such as: global precipitation measurement—GPM (0.1°×0.1°) and the remotely sensed information using artificial neural networks—PERSIANN (0.04°×0.04°) for the flow simulation. For the modelling process, the GR4H included in airGR package was used, which is a free code developed in R. The calibration processes were carried out considering hydrological temporary simulation frame (1h, 3h, …, 24h; where: h=hour). The study area was the humid Rimac River basin, located 50 km from the city of Lima. This basin was selected because it concentrates the largest meteorological and hydrological hourly information around the country. The obtained results suggest that the maximum efficiency is located in modelling scales from 3 to 9 hours with efficiencies within the range [0.5–0.7] from the Nash coefficient. Moreover, it was determined that the product GPM shows a higher efficiency in the representation of the hourly flood hydrographs in the basin.