ESTIMATING GEOLOGIC ENERGY STORAGE IN DEPLETED HYDROCARBON RESERVOIRS USING MATERIAL BALANCE EQUATIONS EMBEDDED IN A BAYESIAN ERRORS-IN-VARIABLES MODEL

The subsurface storage potential for gas in geologic settings, such as depleted hydrocarbon reservoirs and solution-mined salt caverns, is becoming salient to future energy infrastructure planning. Technologies such as carbon capture utilization and storage (CCUS), CO₂-enhanced oil recovery (CO₂-EOR), H₂ storage, and natural gas storage help to meet growing energy demands, reduce emissions, and meet climate goals, providing energy security amid geopolitical uncertainties. Therefore, estimates of underground gas storage capacity could be useful across spatial scales for future use in the energy transition. Material balance is a fundamental method in reservoir engineering for estimating gas in place and potential storage capacity at the spatial scale of national assessments. We propose embedding material balance equations within a Bayesian errors-in-variables model, combining data from multiple sources and allowing estimation of distributions of reservoir properties needed for assessments. Uncertainties associated with these reservoir properties have traditionally been expert elicited, whereas the uncertainty estimates from our model are data driven. Our estimates can be used either directly in an assessment (model distribution via simulations), replacing this expert elicitation step, or be used as impartial data-driven information presented to assessors during elicitation. We demonstrate our method in a case study in the Michigan Basin, a large contributor to the United States’ current natural gas storage capacity. We calibrate the proposed errors-in-variables model using data from operational gas storage facilities in Michigan, allowing estimation of the uncertainty associated with reservoir parameters that define the material balance relationship between reservoir pressure and gas volume. We validate our model predictions against reported capacity and storage pressures using cross validation and continuous rank probability scoring. Incorporating a statistical framework into the material balance approach provides a model that can utilize often conflicting data from multiple sources to estimate input and output uncertainties. This method brings rigor to uncertainty quantification as part of a larger effort by the U.S. Geological Survey to assess domestic energy gas storage resources in depleted hydrocarbon reservoirs.