Random Forest modelling and evaluation of the performance of a full-scale subsurface constructed wetland plant in Egypt
Madleen Salem, Mohamed Elsayed Gabr, Mohamed Mossad, Hani Mahanna
Abstract
This article presents a methodology to evaluate and model the performance of subsurface horizontal constructed wetland (SHCW) in Samaha village, Egypt's Delta. The influent and effluent biological oxygen demand (BOD), chemical oxygen demand (COD), total suspended solids (TSS), total phosphorus (TP), and total nitrogen (TN) were measured and analyzed for the period 2018–2020. The Random Forest process (RF) was applied to predict the removal efficiency of BOD, COD, and TSS using 112 samples, 90% of samples were used as training and 10% as a test. The results show removal efficiency of 72, 77, 78, 49, and 31%, for BOD, COD, TSS, TP, and TN respectively. The Random Forest modeling fit the results well with coefficient of determination (R2) for removal efficiency of BOD, COD, and TSS equal to 0.66, 0.68, and 0.79, respectively. The Random Forest process is a good way to come up with effective management and mitigation strategies.