The use of fuzzy approach is becoming increasingly common in the analysis of hydrology and water resources problems. In this research, a fuzzy inference system was developed and used to model the rainfall-runoff relationship on a semi-arid catchment, namely the Kurukavak catchment, which is a sub-basin of the Sakarya basin in Turkey. An adaptive neuro fuzzy inference system (ANFIS) was chosen for use in the current study. The results and comparative study indicate that the fuzzy method is more suitable to predict runoff than classical regression model.

Massachusetts Institute of Technology (MIT) Media Laboratory, MIT;Harvard Univ., Dep. Stat., Stat. Genomics Comput. Biol. Lab.;Univ. Texas Austin, Texas Adv. Comput. Cent.;Stat. Comput. Intell. Lab. Purdue Univ.;Indiana Univ.

2007 International Conference on Artificial Intelligence, ICAI 2007 -- 25 June 2007 through 28 June 2007 -- Las Vegas, NV -- 92744