Development of an Automatic Calibration Tool Using Genetic Algorithm for the ARNO Conceptual Rainfall-Runoff Model

Rainfall-runoff simulation is one of the key steps in hydrology. Conceptual models are frequently used in rainfall-runoff simulation. However, a major difficulty in practice remains on how to optimize the parameters of the model. This is often a time-consuming and labor-intensive task for the modeler when manual calibration is adopted together with employing the knowledge of the model structure and parameters. In this study, an automatic calibration tool was developed to calibrate the ARNO conceptual rainfall-runoff model using the simple genetic algorithm (SGA). SGA is a simple, powerful, and popular optimization method, which explores the search space for the global optimum and has been successfully employed in many optimizations problems. The ARNO model was calibrated automatically for rainfall-runoff simulation of the Pataveh basin, which is a sub-basin of Karun River basin in Iran. The simulation performance of the model was evaluated on the basis of various performance criteria. Efficiency coefficient and coefficient of determination reached values higher than 0.80 during calibration and validation. The values of the remaining performance statistics were acceptable. The results show that this model with employed automatic calibration tool can successfully be used for continuous rainfall-runoff simulation.