Injection-induced Seismic Moment Release in Stratified Sedimentary Rocks

Induced earthquakes during deep geo-energy extraction in stratified sedimentary rocks indicate that the size of these earthquakes depends not only on fluid injection volume but also on host rock type. Existing models may not be appropriate to predict the maximum seismic moment for the case the fluid injection point and induced earthquake epicenter located in different rock layers. Here we propose an analytical model to predict the maximum seismic moment based on the hypothesis that a far-field dynamic rupture is promoted by a near-field, injection-induced aseismic slip through shear stress transfer. We establish partial differential equations and quasi-dynamic elasticity equations based on the one-dimensional plane strain crack model and solve these equations with the boundary condition at the interface between two adjacent layers. We validate this model using the field data from numerous injection-induced earthquakes and compare the maximum seismic moment with those obtained from McGarrs and Galis models. Our results show the maximum seismic moment can be underestimated if fluid is injected into a lower layer with a larger shear modulus and the seismic event is induced in an upper layer as a high strain energy can be stored in the stronger, deeper layer. Our study highlights that the aseismic and seismic slip may dominate different fault patches, controlling seismic energy release from different rock layers. For simplification, an equivalent shear modulus of multiple rock layers can be derived to predict the maximum seismic moment based on the injected volume. However, this is not reliable because the maximum seismic moment can also be influenced by fault and interface permeabilities as well as heterogeneous frictional properties.