In this paper,a genetic algorithm(GA)-based four-dimensional variational data assimilation(4DVAR) is applied to the MM5 model.In order to verify the performance of the GA assimilation system,the output of the GA based 4DVAR model has been compared with that of an adjoint 4DVAR model.The results indicate that the new method of GA-based 4DVAR can be considered as one of the effective methods to improve numerical weather prediction initial field and raise the prediction accuracy of physical field and rainfall.The new method shows a better performance than the adjoint counterpart based on the present experimental case.