Rainfall monitoring networks provide fundamental information for water resources management and hydraulic engineering. In this study, we developed a multi-criteria approach for rainfall monitoring network optimization. The criteria comprise two transinformation based objectives accounting for overlapped information among selected stations and transferable information from selected to unselected stations, and a rescaled Nash-Sutcliffe Coefficient (NSC) objective measuring the total residual error. The approach was applied to a rainfall monitoring network in the Taihu Lake Basin of China. Pareto solutions under different scenarios were selected from massive candidate solutions for comparison.