Uncertainties in the input data needed to run a basin model lead to uncertainties in the predictions. Comprehensive reaction, transport, mechanical (RTM) models require the use of extensive data sets for calibration in order to generate a realistic synthetic response.

An automated approach to integrate comprehensive basin modeling with seismic, well-log and other types of data is presented. The approach takes advantage of our comprehensive RTM basin model to complete an algorithm based on information theory (IT) that places basin modeling on a rigorous foundation.

This IT method can be used to calibrate the transport, mechanical or other laws underlying the model. Incompleteness in a model can self-consistently be compensated by increasing the amount of observed data used. As the procedure is fully automated, the predictions can be continuously updated as new observed data becomes available.

Finally, the procedure makes it possible to augment the model itself as new processes are added in a way that is dictated by the available data. In summary, the automated data/model integration places basin simulation in a novel context of informatics that allows for data to be used to assess and minimize risk in such practical applications as the prediction of petroleum reservoir location and characteristics.