The flux-distribution that-a heliostat field-projects on-its-mceiver must-be-controlled carefully. It affects the performance of the system and its operational safety. Control strategies combine the use of real measurements with models of the system. This work proposes a new methodology to characterize the behavior of a given field for a whole year. It applies data modeling techniques to generalize the flux maps projected by a subset of heliostats at different solar positions. Upon success, it allows us to analytically estimate the flux map that every heliostat would form on the receiver at any time. The resulting field model is valuable for control purposes. It is also a convenient way of sharing representations of fields. The proposed methodology has been successfully applied to a model of field CESA-I, at the Plataforma Solar de Almeria (PSA). The average error obtained provides a high reduction in the variance of the estimated parameters. Thus, the field model is reliable enough to predict the flux maps generated by its heliostats at any solar position.