The problem of the optimization of ocean monitoring networks is tackled through the implementation of data assimilation techniques in a numerical model. The methodology has been applied to the tide gauge network in the Lagoon of Venice (Italy). The data assimilation methods allow identifying the minimum number of stations and their distribution that correctly represent the lagoon's dynamics. The methodology is easily exportable to other environments and can be extended to other variables.