Modeling the Spatio-Temporal Evolution of Chlorophyll-a in Three Tropical Rivers Comoé, Bandama, and Bia Rivers (Côte d’Ivoire) by Artificial Neural Network

Rivers and their watersheds play a key role in the global biogeochemical cycle of nitrogen and phosphorus in the biosphere. They are also important economic resources for humans. However, little information is available on eutrophication of West African coastal rivers due to costly analytical instruments and socio-economic difficulties. In this study, the spatial distributions of chlorophyll-a biomass in the Comoe, Bandama, and Bia Rivers (Cote d’Ivoire) were mapped during the dry, rainy and flood seasons, and chlorophyll-a dynamics were simulated using Artificial Neural Network (ANN) models. The results showed a state of advanced eutrophication during the three sampling seasons. The best generalizable models obtained from the data collected during 2 years and covering three hydrological seasons of the rivers forecasted between 76% and 85% of present chlorophyll-a concentrations (static approach), and between 73% and 84% of future chlorophyll-a concentrations (both dynamic t and t + 1 approaches). These models achieved satisfactory accuracy with low relative mean errors (MRE) ranging from 3.22% to 7.71%. The results of this study suggest that ANN model could be an original and less expensive tool for monitoring river water eutrophication in developing countries.