Injection-induced earthquakes pose a serious seismic hazard but also offer an opportunity to gain insight into earthquake physics. Currently used models relating the maximum magnitude of injection-induced earthquakes to injection parameters do not incorporate rupture physics. We develop theoretical estimates, validated by simulations, of the size of ruptures induced by localized pore-pressure perturbations and propagating on prestressed faults. Our model accounts for ruptures growing beyond the perturbed area and distinguishes self-arrested from runaway ruptures. We develop a theoretical scaling relation between the largest magnitude of self-arrested earthquakes and the injected volume and find it consistent with observed maximum magnitudes of injection-induced earthquakes over a broad range of injected volumes, suggesting that, although runaway ruptures are possible, most injection-induced events so far have been self-arrested ruptures.
Bibliographical noteKAUST Repository Item: Exported on 2020-10-01
Acknowledged KAUST grant number(s): BAS/1339-01-01, URF/1/2160-01-01
Acknowledgements: Research presented in this paper is supported by King Abdullah University of Science and Technology (KAUST) in Thuwal, Saudi Arabia (grants BAS/1339-01-01 and URF/1/2160-01-01) and by the NSF (CAREER award EAR-1151926). Some of the 3D dynamic rupture simulations for verification of our model have been carried out using the KAUST Supercomputing Laboratory. We thank the Agence Nationale de la Recherche through the HYDROSEIS project (Role of fluids and fault HYDROmechanics on SEISmic rupture) under contract ANR-13-JS06-0004-01 for supporting the in situ experiments providing the data (Duboeuf et al.) used in Fig. 4. We also thank S. Goodfellow and L. De Barros for providing their laboratory and in situ data used in Fig. 4. J.P.A. and F.C. thank the Observatoire de la Côte d’Azur for supporting this research.