Information Gain in Rainfall-Runoff Modeling (Tunisia)
- 89 Downloads
Rainfall runoff conceptual models are used mainly to help make decisions in surface water resources and water quality management. The model calibration is an important step in selecting suitable sets of parameters. The model validation helps to demonstrate the model performance outside the calibration period, in extrapolation. Often, the calibration/validation of the model are carried out using the Generalized Split Sample Test (GSST) method. It is proposed to depict the validation analysis in view of the information gain between observed rainfall (mm day−1) and runoff (mm day−1). For illustration, we adopt the bucket bottom hole (BBH) model with a daily step. The model is lumped. The calibration is achieved by accepting the sets of parameters (solutions) that guarantee a relative error <20%. In addition, a Nash-Sutcliffe coefficient (NSH) condition is established with NSH >0.75. The evaluation of model performance in validation is based on NSH criterion at the monthly and decadal scales. The case study is the basin of the Sejnane River (376 km2) now controlled by a dam in northern Tunisia. It is found that the period with the highest information gain between rainfall and runoff observations constitutes a good “receiver” which means a good ability to reproduce runoff data regardless of the calibration period.
KeywordsRainfall Runoff NSH Information gain GSST
- 1.Bargaoui, Z., Houcine, A.: Sensitivity to calibration data of simulated soil moisture related drought indices. Revue Sécheresse. 21(4), 1–7 (2010)Google Scholar
- 6.Gray, R.M.: Entropy and Information Theory First Edition, Corrected. Springer, New York (2013). https://ee.stanford.edu/~gray/it.pdf
- 7.Kobayashi, T., Matsuda, S., Nagai, H., Tesima, J.: A bucket with a bottom hole (BBH) model of soil hydrology. 5p. IAHS Publ. No. 270 (2001)Google Scholar
- 8.Moriasi, D.N., Arnold, J.G., Van Liew, M.W., Bingner, R.L., Harmel, R.D., Veith, T.L.: Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Trans. Am. Soc. Agric. Biol. Eng. 50(3), 885–900 (2007)Google Scholar