Runoff mapping using the SCS‑CN method and artificial neural network algorithm, Ratga Basin, Iraq

Abstract

"The traditional Soil Conservation Service )SCS( process for calculating the runoff depth tends to be a very tedious and time time-consuming hydrological modeling process. Therefore, a geographic information system (GIS) is now being utilized as a tool alongside the common SCS-CN method for runoff calculations. This research aims to estimate the spatial distribution of runoff depth from Ratga, an agricultural watershed from the Iraqi Western Desert, using the SCS-CN method, GIS, generalized regression neural network (GRNN), field observation dataset, and remote sensing data. The GRNN model was used to predict the soil type based on spectral reflectance data. The results refer to an excellent performance of this model with the maximum absolute error was 8.44%, 14.11%, and 4.15% for sand, silt, and clay soil, respectively, and the sandy soil has the highest correlation coefficient (0.83). The outcome of the SCS method showed the CN value ranged from 70 to 85 of normal conditions. This investigation outline that the maximum volume of surface runoff of the 2018 to 2020 years was 4,324,528 m3. This paper proves that incorporating GIS with the SCS-CN model and ANN provides a robust tool for calculation runoff depth in the Iraqi Western Desert, representing barren catchments of Iraq."