Keywords
geothermic, Markov Chain Monte Carlo, multiple-point statistics, gravity residuals, Geneva basin
Start Date
16-9-2020 8:40 AM
End Date
16-9-2020 9:00 AM
Abstract
Geothermal resources are of increasing interest for industry, public authorities and academic community thanks to the wide variety of applications covering individual to industrial uses for heat storage and production, and power generation. Geothermal reservoirs in sedimentary settings are often compartmentalized due to lithological heterogeneities and fault zones, which can act either as preferential flow path or impermeable barrier. Geophysical methods, such as reflection seismic and gravity, are useful to explore geothermal reservoir potential, by reducing the risks (e.g. costs, induced seismicity) associated with drilling. Indeed, reflection seismic allows the delineation of geological formations and fault zones geometry. In addition, gravity data, sensitive to density contrasts, may allow the detection of large open fracture networks, that locally decreases the density of the geological formation. Thus analysis of density contrasts can be used to predict the porosity distribution of the reservoir and identify drilling location candidates. In order to characterize the fractured zones and their density, relatively to the geological formation density, we invert the profiles and grid networks of gravity data acquired in the Geneva basin over a zone of 150x70km. We used a Markov chain Monte Carlo algorithm, which allows us to estimate the probability density function of an ensemble of solutions. The fractured zone models are generated with graphcuts, a multiple-point statistics technique that enables fast model parameter simulations. Several scenarios describing the heterogeneity of subsurface density are investigated and the inversion results are compared and contrasted to previous fault and geological interpretations. Ultimately, this work provides useful information to identify and constrain the subsurface uncertainties associated with the definition of the structural setting and reservoir distribution.
Uncertainty quantification and porosity estimation of geological structures associated to sedimentary geothermal reservoirs, based on Bayesian inversion of gravity anomalies with graphcuts
Geothermal resources are of increasing interest for industry, public authorities and academic community thanks to the wide variety of applications covering individual to industrial uses for heat storage and production, and power generation. Geothermal reservoirs in sedimentary settings are often compartmentalized due to lithological heterogeneities and fault zones, which can act either as preferential flow path or impermeable barrier. Geophysical methods, such as reflection seismic and gravity, are useful to explore geothermal reservoir potential, by reducing the risks (e.g. costs, induced seismicity) associated with drilling. Indeed, reflection seismic allows the delineation of geological formations and fault zones geometry. In addition, gravity data, sensitive to density contrasts, may allow the detection of large open fracture networks, that locally decreases the density of the geological formation. Thus analysis of density contrasts can be used to predict the porosity distribution of the reservoir and identify drilling location candidates. In order to characterize the fractured zones and their density, relatively to the geological formation density, we invert the profiles and grid networks of gravity data acquired in the Geneva basin over a zone of 150x70km. We used a Markov chain Monte Carlo algorithm, which allows us to estimate the probability density function of an ensemble of solutions. The fractured zone models are generated with graphcuts, a multiple-point statistics technique that enables fast model parameter simulations. Several scenarios describing the heterogeneity of subsurface density are investigated and the inversion results are compared and contrasted to previous fault and geological interpretations. Ultimately, this work provides useful information to identify and constrain the subsurface uncertainties associated with the definition of the structural setting and reservoir distribution.
Stream and Session
false