Stochastic inversion of the tracer experiment of the enhanced geothermal system demonstration reservoir in Soultz-sous-Forêts — Revealing pathways and estimating permeability distribution

In this study, we identify a suite of fluid flow pathways and estimate permeability distribution for the engineered geothermal reservoir at the European enhanced geothermal system test site at Soultz-sous-Forêts, France. The work is based on the 2005 hydraulic connectivity experiment between the injection borehole GPK3 and the production boreholes GPK2 and GPK4. We use a massive Monte Carlo approach, together with a geostatistically based assignment of permeability distributions using a stochastic technique, in order to investigate the observed tracer concentration curves. By applying different fitting criteria, 49 single models were selected, providing a good fit to the observation data from a model ensemble of 10,000 realizations. These models are classified in three groups with different flow geometries and permeability distributions, and were analyzed for their Darcy velocities. This approach allows the limits for possible permeability distributions and geometries of the stimulated zones to be quantitatively approximated.

► We model the tracer experiment of 2005 at the Soultz EGS stochastically.
► A Monte Carlo approach is used to create random permeability fields.
► We identify realizations with a good fit of the GPK2 tracer curve.
► Characteristics of the corresponding permeability field are extracted.
► Three groups of possible tracer pathways and their permeabilities are identified.