Abstract:In order to study the impacts of aquifer heterogeneity on the results of Monte Carlo simulation of groundwater, two methods, i. e. , the one-to-one filling method and the arrangement filling method, were used to obtain the permeability coefficient random field, and the results of Monte Carlo simulation using the two methods were compared. The study suggests that, for the aquifer with strong heterogeneity, the necessary number of realizations of Monte Carlo simulation should be at least 2 500 for the one-to-one filling method and 10 000 for the arrangement filling method, in order to obtain reasonable results. In the case of consistent absolute number of realizations, the one-to-one filling method performs better than the arrangement filling method. However, when the number of realizations reaches 22 500 or above, there is no significant difference between the two methods.