Assessment of Seismic Risk in Geothermal and Hydrocarbon Reservoirs Using an Exact Analytical Solution of Stress Change

Both for the oil & gas and geothermal industry, induced seismicity caused by field development and operation can pose a risk, in particular when the reservoir (or overburden / underburden) is intersected by faults. The mechanisms by which faults can be reactivated (potentially leading to seismicity) include pressure effects (reservoir depletion, or pressure rise over large areas as a result of injection) or thermal effects (cooling such as in geothermal operations or heating such as in steam flooding). Earlier, we proposed a simple methodology to assess seismic risk for geothermal reservoirs that can also be applied to hydrocarbon reservoirs. This methodology uses an elastoplastic finite element model of the reservoir in question. However, its application turned out to be laborious. Therefore, we developed an exact analytical solution for the stress changes induced by cooling, depletion and /or pressurization along (a) representative fault(s). This solution is a generalisation of the Goodier analytical solution for the situation of non-vertical faults. The analytical solution can be used to quickly evaluate a number of different scenarios related to temperature and /or pressure distributions in the reservoir. In the case of fault activation, maximum fault displacements (slip) can be computed by linking the results to elastic finite element calculations for similar load conditions. Using published standard correlations, the seismic magnitude can subsequently be estimated from the computed fault displacements. The analytical model was applied to different fault geometries, reservoir temperature distributions and depletions. It turns out that certain fault geometries (dip angles, offsets) are far more prone to activation than other fault geometries. An explanation of this result is provided. Furthermore, for non-critically stressed faults, the risk of activation is far less for geothermal operations than for situations where large parts of the reservoir are depleted or pressurized. This can be explained by the fact that the extent of the cooled zone in geothermal operations is generally limited, even after 30 years of operation. Consequently, cooling-induced stress changes along the fault are significantly reduced because of arching by the adjacent non-cooled areas. Finally, one geothermal field example in The Netherlands is presented where the above methodology was applied to demonstrate that there exists no seismic risk over the entire field life.