Document Type : Research Paper
Water Engr. Dept. Faculty of Agriculture, University of Tabriz, Iran
Water Engr. Dept. Faculty of Agriculture, University of Tabriz
Mathematical Dept. Faculty of Mathematical Sciences. University of Tabriz
Hydrological modeling plays a valuable role in watershed management. In order to advance this important, in this study, a combination of linear programming, cross-wavelet transform and Kalman filter as a control model for the analysis of nine compound rainfall and runoff events were used in Sufi Chay Basin. The results were compared with modified time area unit hydrograph, and e geomorphologic unit hydrograph. Finally, using the evaluation criteria used in the research, the final performance of these methods was investigated and analyzed. it was revealed that the modified time area method has a relatively weaker performance than the other two methods, which is due to the assumptions used in drawing the isochrones lines. The LP-CW-KF method showed the best performance among the studied methods, which simulated the compound events in the calibration and validation stage with a mean squared error of 2.47 and 2.7, respectively. On average, in all the events and the three studied methods, the mean absolute relative error (MARE) was 0.069 in the time to peak, 0.131 in the peak discharge and 0.125 in the base time. Therefore, on average, all methods showed a better performance in estimating time to peak.