Petroleum-resource appraisal and discovery rate forecasting in partially explored regions: an application to supply modeling

This study examines the temporal properties and determinants of petroleum exploration for firms operating in the Denver basin. Expectations associated with the favorability of a specific areas are modeled by using distributed lag proxy variables (of previous discoveries) and predictions from a discovery process model. In the second part of the study, a discovery process model is linked with a behavioral well-drilling model in order to predict the supply of new reserves. Results of the study indicate that the positive effects of new discoveries on drilling increase for several periods and then diminish to zero within 2-1/2 yr after the deposit discovery date. Tests of alternative specifications of the argument of the distributed lag function using alternative minimum size classes of deposits produced little change in the model's explanatory power. This result suggests that, once an exploration play is underway, favorable operator expectations are substained by the quantity of oil found per time period rather than by the discovery of specific size deposits. 39 references.