IAHS2022-89, updated on 22 Sep 2022
https://doi.org/10.5194/iahs2022-89
IAHS-AISH Scientific Assembly 2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Artificial Neural Networks (ANNs) and Hydrological Models to Simulate Streamflow in a context of climate change: Case of Davo river basin in Côte d’Ivoire

Yao Morton Kouamé1, N'diaye Edwige Hermann Meledje2, Tanina Idrissa Soro1, Gneneyougo Emile Soro3, Destyle Van Kombyla4, Kouakou Lazare Kouassi1, Bi Tié Albert Goula3, Jean Emmanuel Paturel5, Arona Diedhiou6,7, and Ernest Amoussou8
Yao Morton Kouamé et al.
  • 1Université Jean Lorougnon Guédé, Daloa, Côte d'Ivoire (mortonkouame@gmail.com)
  • 2Centre de Recherche en Ecologie, Laboratoire de Géologie Marine, Sédimentologie et Environnement, Abidjan, Côte d’Ivoire
  • 3Laboratoire Géosciences et Environnement, Université Nangui Abrogoua, Abidjan, Côte d’Ivoire
  • 4Université Marien NGOUABI, Brazzaville, Congo
  • 5HydroSciences Montpellier (HSM) – UM2, IRD, Case MSE – Place Eugène Batillon – 34095, Montpellier cedex 5 (France), Université Nangui ABROGOUA (Côte d’Ivoire
  • 6Université Grenoble Alpes, IRD, CNRS, Grenoble INP, IGE, F-38000 Grenoble, France, 7LAPA-MF; Centre d’Excellence Changement Climatique, Biodiversité et Agriculture durable (CEA CCBAD),
  • 7Université Houphouët-Boigny d’Abidjan, (Côte d'Ivoire).
  • 8Département de Géographie et Aménagement de Territoire (DGAT/FLASH), Université de Parakou, BP 123 Parakou, Bénin

The streamflow is important for planning et decision-making of development activities  related to water resource. The hydrological mathematical models are frequently used to simulate the transformation of rainfall into streamflow as well as Artificial Neural Networks (ANNs. In order to contribute to the establishment of an adequate framework for water resource management in the Davo river basin, this study presents an intercomparison between streamflow estimations from conventional hydrological modelling and ANN analysis. The Davo River basin is located in the southwestern forest of Côte d'Ivoire. The hydrological models used are the HEC-HMS model (Hydrologic Engineering Center-Hydrologic Modeling System) and the GR4J model (Rural Engineering with four daily parameters), while the data-driven ANN model is developed in MATLAB. The three tools receive the same data and their parameters are calibrated using the same objective functions. The quality of the simulations is measured in the control phase using several statistical criteria. For the assesment of the impacts of climate change on hydrological regime of the Davo river basin, the AFRICA-CORDEX (Coordinated Downscaling Experiment for Africa) data under two RCP (representative concentration pathway) scenarios   RCP4.5 and RCP8.5 are integrated in the three tools of modelling. The approach developed in this study allows a considerable improvement of the outils performances developed in calibration and validation. In addition, the results of the current study provide valuable feedback for water resources’ planners in Côte d’Ivoire and the developing regions.

Keywords: Climate change, Streamflow, Artificial Neural Networks, HEC-HMS, GR4J, Davo river basin, Côte d’Ivoire

How to cite: Kouamé, Y. M., Meledje, N. E. H., Soro, T. I., Soro, G. E., Kombyla, D. V., Kouassi, K. L., Goula, B. T. A., Paturel, J. E., Diedhiou, A., and Amoussou, E.: Artificial Neural Networks (ANNs) and Hydrological Models to Simulate Streamflow in a context of climate change: Case of Davo river basin in Côte d’Ivoire, IAHS-AISH Scientific Assembly 2022, Montpellier, France, 29 May–3 Jun 2022, IAHS2022-89, https://doi.org/10.5194/iahs2022-89, 2022.