Best Practice Methodological Appraisal For Geothermal Systems In Clastic Reservoirs: A Revision

Abstract

In 2011, when the Dutch geothermal industry was pioneering, we presented our first attempt for an appraisal methodology for direct heat utilization of deep geothermal energy systems in sedimentary reservoirs (van Leeuwen et al., 2011). Today, the Dutch geothermal industry is more mature. An appraisal system for stimulation renewable energy (SDE) subsidy applications is in place in the form of DoubletCalc (Mijnlief et al., 2014; Veldkamp et al., 2015). DoubletCalc employs uncertainties in subsurface parameter estimates, but challenges remain on how to quantify them. As over time, new insights changed most of the steps applied in our methodology, a thorough revision is necessary. In this revision of our 2011 methodology, we introduce our current best practices for the evaluation and quantification of subsurface uncertainties with respect to the prediction of the geothermal capacity of clastic sedimentary geothermal reservoirs in the Netherlands. Whereas the guidelines for the subsidy application describe the information the document should contain, it does not explicitly state which research should be done. In general, but dependent of the quality and the quantity of the available data, the following activities are carried out to quantify the subsurface uncertainties. To estimate depth, thickness, temperature and permeability of the geothermal reservoir, seismic data is interpreted and well log data is analysed. Although well test data is ideal for this purpose, the reservoir permeability is often predicted using a porosity-permeability relation including uncertainties that is derived from core samples. Following, a Monte Carlo simulation is run to obtain probability distributions of the relevant reservoir properties in the analysed wells. Depth of the geothermal reservoir is predicted based on the interpreted seismic data. If the base of the geothermal reservoir cannot be picked on the seismic data, well data is interpolated. The uncertainty range here is dictated by the quality of the seismic pick and the time-depth inversion. Next, the estimated reservoir properties have to be spatially extrapolated from the wells to the desired project location. This is done by means of a porosity-depth and a second porosity-permeability relation. Both relations are based on the average values of the reservoir properties in the wells analysed. A temperature gradient is derived based on corrected bottom hole temperatures. Its uncertainty is incorporated in the uncertainty in the depth of the reservoir. Now the reservoir properties are estimated. The subsurface well locations can be determined and a conceptual well design can be drafted. At this point, the optimum between the geothermal capacity and the Coefficient of Performance (COP) of the geothermal system is determined by an iterative process. When this optimum is found, a second Monte Carlo simulation is run to determine the probability distributions of the capacity, flow and COP of the geothermal system under development. Following this recipe to determine the reservoir parameters, well design and corresponding performance of a geothermal system, either for use in DoubletCalc or using another approach, a sensible appraisal of the expected geothermal performance, and its uncertainty, is achieved. It is believed that this is, given the data available and the budget and time constraints, the best practice covering the most significant uncertainties for geothermal feasibility studies in reservoirs in clastic sedimentary basins.

AAPG Datapages/Search and Discovery Article #90345 © 2018 AAPG European Region, Geothermal Cross Over Technology Workshop, Part II, Utrecht, The Netherlands, April 17-18, 2018