WP 2.3

Model development for simulation of leakage into shallow aquifers and for seismic monitoring of deep storage formations and leakage

Leakage of stored gas from deep caverns or porous aquifer storage formations may induce geochemical reactions in shallow aquifers. Such reactions may be expected along the leakage pathways as well as in the near field of the shallow aquifers affected by the leakage. In order to allow detailed but still efficient simulation of these processes and geochemical reactions the THMC code OpenGeoSys (OGS) will be extended, in WP 2.3 for methods of local a-priori grid refinement with treatment of irregular grid nodes for the simulation of small scale processes in large model domains. Moreover, methods for geochemical front-tracking and geochemical steady-state will be developed, implemented and tested in order to increase the efficiency of geochemical reaction simulations. In cooperation with the experimental studies of WP 1.8 (WG Dahmke) functional relations for pressure and temperature dependencies of geochemical processes such as sorption and exchange reactions will be implemented in the model. Verification of the new methods and relations is achieved by benchmarks test simulations and model comparisons.

The injection of gas in the subsurface leads to small changes of the physical parameters (elastic moduli, density, specific resistivity) within the reservoir. Therefore the spatiotemporal mapping and monitoring of the gas propagation requires a highly resolving geophysical imaging approach. The central part of the monitoring concept which will be developed in WP 2.3 is the seismic waveform inversion. Classical traveltime based tomographic approaches are limited in resolution to the first Fresnel zone as shown by the reconstruction of the complex Marmousi model (cf. (Fig. 1a and 1c). The resolution can be significantly improved if not only traveltime information of specific waves but also phase and amplitude information of the whole recorded wavefield are incorporated in the inversion process. The result of such a waveform inversion (Fig. 1b) can show complex geological structures present in the true subsurface model (Fig. 1c).

Figure 1

The key for an efficient waveform inversion are accurate, highly parallelized seismic forward modeling codes. Beside the optimization and development of the 2D finite-difference (FD) code DENISE for modeling and inversion of waves in visco- and poro-elastic media, the main emphasis is the development of a 2D discontinuous Galerkin finite-element (DG-FEM) code GERMAINE. The geophysical monitoring strategies are evaluated using synthetic injection scenarios based on multiphase flow simulations (Fig. 2a-c). The resulting distribution of the gas saturation are linked with the changes of seismic velocities and density via an appropriate rock model (Fig. 2d). For realistic reflection seismic acquisition geometries synthetic datasets can be generated with seismic modeling codes (Fig. 2e). This data can be used for a resolution analysis and further optimizations of the elastic waveform inversion approach. The inverted seismic velocity and density changes (Fig. 2f) can be directly compared with the true changes of the elastic material parameters (Fig. 2d) or, after an additional inversion of the rock model, converted into a gas saturation distribution (Fig. 2g) and compared with the results of the multi-phase flow simulations. The geophysical monitoring concept is complemented by the combination of seismic waveform inversion with geoelectric, electromagnetic, gravimetric and microseismic methods (Hagrey et al. 2014).

Figure 2

The working groups Geohydromodelling (CAU) and Aplied Geophyscs (CAU) participate in the reseach activities connected with WP 2.3. The implementation of temperature dependencies of specific geochemical reactions requires an identification and quantifications of these relations in the experimental studies of WP 1.8. The extended code OGS then will be used for the simulation of gas leakage-scenarios in WP 3.2 and WP 3.3. Based on these simulations the combined geophysical monitoring concept can be tested and optimized.

References:
Hagrey, S.A.al, Köhn, D., Rabbel, R., 2014, Geophysical assessment of renewable gas energy compressed in geologic pore storage reservoirs, SpringerPlus 2014, 3:267, 16p. Doi:10.1186/2193-1801-3-267.

Martin, G.S., Marfurt, K.J. and Larsen, S., 2002, Marmousi-2: An Updated Model for the Investigation of AVO in Structurally Complex Areas, 72nd Annual International Meeting, Society of Exploration Geophysicists, Expanded Abstracts, 1979-1982.