This thesis with an important aspect of the construction of numerical models for simulating groundwater flow and mass transport in heterogeneous aquifers: the integration of all available information, taking into account their degree of confidence and they come to characterize the physical environment. This integration is aimed at quantifying yreducción of uncertainty in predictions made from numerical models. One of the most important sources of information are the geophysical surveys. This thesis describes the nature and princiios of geophysical methods with emphasis on those applications able to collect information useful for characterization didrogeológica terrain. We review the relations established by various researchers between the main parameters geoeléctiros with hydrogeological and geossismicos. Subsequently, geostatistical techniques to combine information are reviewed in some detail. Two examples illustrate its behavior to be applied to a real case. Among the algorithms geostatistical stochastic representation of the probability field simulation is reviewed in detail, to develop its implementation in multivariate. Finally has a new sumulación stochastic algorithm that permits the integration of information whose support is different. The fields generated by this new technique has the property that the values and patterns of continuity of representation variables are linearly related.