ANNUAL RUNOFF MODELLING OF KIZILIRMAK BASIN BY ARTIFICIAL INTELLIGENT TECHNIQUES.

Estimation and modelling of meteorological parameters are very important while management and planning of water resources are being design. In this study, the equations which can be used for modeling the annual values of the meteorological data of Kizilirmak Basin is derived from ANN (Artificial Neural Network), GEP (Gene Expression Programming) and Regression analysis program (Datafit). In this regard, measuring data is used to associate on flow for a better understanding of the agreement among testing model fit. The aim of this study is to acquire the formulations which can be used in the flow estimation under influence of different meteorological parameters for Kizilirmak Catchment. The figures were considered from the nonlinear regression analysis. During the analysis, precipitation, humidity and temperature were used as input parameters and discharge was used as output parameter. Also Mean Square Error (MSE), Root mean square error (RMSE), Coefficient of Determination (R² ) and Adjusted coefficient of Determination (AdjR²) parameters were calculated for each methods (ANN, GEP and Datafit). The obtained equations were evaluated for each models respect to Meteorological and flow data. Overall, the study demonstrated a good capturing skill of GEP driven flow estimations relative to observation data and model results. The applied approaches developed in this study can motivate future studies over basins study storm event analysis beyond hydrological modelling. [ABSTRACT FROM AUTHOR]