Evaluation of Several Empirical Models in Estimating the Monthly Runoff in Urban Region. Case Study: Kafr Kela Al Bab Region, Egypt

Abstract

 For flood risk management in a particular location, precise runoff prediction is crucial, although it's still a challenging process because it's unexpected and imprecise. Accordingly, the main goal of this study is to 
predict urban runoff in the Kafr Kela Al Bab region of Egypt using the quadratic model (QM), Elman neural network (ENN), cascade forward neural network (CFNN), and Poisson regression (PR). To accomplish this 
purpose, global meteorological data are collected for analysis in this study between 1985 and 2023. The results showed that the QM model outperforms the other models, with R^2 = 0.9906 and 0.9902 during the validation 
phase. Furthermore, the RSME and MAE demonstrate that the observed values and the QM models' derived values strongly agree with each other in terms of statistical parameters.