In order to simplify the complex calibration process of parameters, a modified Muskingum routing model suitable for the simulation of ice thawing flood during river ice tranquil breakup period is proposed, in which the flow generated by thawing ice in the reach is considered as a part of the outflow of the lower section of the reach. Moreover, the optimization algorithm is applied to the parameter calibration process of the model. Taking the Ningxia-Inner Mongolia reach of the Yellow River as an example, the parameter K and x of the proposed model are calibrated by trial and error method, nonlinear programming method and genetic algorithm. The simulation results show that the outflow hydrograph calculated by the three methods are all qualified. Overall, the nonlinear programming method has the highest accuracy, with the determination coefficient (Dc)of the flood being 0.980 and the average relative error (RE) of the process being 4.840 %. However, the peak flow simulated by the trial and error method is more accurate. Moreover, its simulation accuracy of ice covered period is higher than that of open flow period. The proposed model provides a new approach for the study of ice thawing flood routing.