Currently Active Research Projects

GMG Projects

GMG-4 (2018) - Understanding conditions for stable/unstable fault slip induced by fluid injections

Experimental plan:

Use the existing codes to model a field experiment on fluid injection.
Develop codes for coupling between evolving compaction/dilation of the fault gouge and fluid flow.

Publications:

Larochelle, S., Lapusta, N., Ampuero, J.-P., & Cappa, F. (2021). Constraining fault friction and stability with fluid-injection field experiments. Geophysical Research Letters, 48, e2020GL091188. https://doi.org/10.1029/2020GL091188. [PDF] *GMGPUB2

Heimisson, Elías Rafn and Rudnicki, John and Lapusta, Nadia (2021) Dilatancy and Compaction of a Rate-and-State Fault in a Poroelastic Medium: Linearized Stability Analysis. Journal of Geophysical Research. Solid Earth, 126 (8). Art. No. e2021JB022071. ISSN 2169-9313. doi:10.1029/2021jb022071. [PDF] *GMGPUB9

Sirorattanakul, K., Larochelle, S., Rubino, V. et al. Sliding and healing of frictional interfaces that appear stationary. Nature (2025). https://doi.org/10.1038/s41586-025-08673-0 [PDF]

GMG-6 (2018) - Experimental investigation of the interaction between fluids and failure of rock faults in shear

Experimental plan:

Conduct controlled and highly instrumented laboratory experiments with fluid injection into a pre-existing fault to study evolution in friction/pore pressure and triggering of fast/slow slip under various conditions.
Measure slip, slip rate, and shear stress evolution along the fault during the injection process.
Compare measurements with existing theories on the stability of fault slip.

Publications:

Gori, Marcello and Rubino, Vito and Rosakis, Ares J. et al. (2021) Dynamic rupture initiation and propagation in a fluid-injection laboratory setup with diagnostics across multiple temporal scales. Proceedings of the National Academy of Sciences of the United States of America, 118 (51). Art. No. e2023433118. ISSN 0027-8424. doi:10.1073/pnas.2023433118. [PDF] [SUP] *GMGPUB8

GMG-9 (2018) - Application of DAS in monitoring microseismicity and subsurface structure changes

Experimental plan:

Use the DAS instrument contributed by OptaSense through GMG to collect data in the Pasadena area.
Analyze the DAS data and develop new methods in microseismicity detection, and structure monitoring.
Optimize the data collection and processing procedures to improve the monitoring accuracy and efficiency.

Publications:

Wang, Xin and Williams, Ethan F. and Karrenbach, Martin et al. (2020) Rose Parade Seismology: Signatures of Floats and Bands on Optical Fiber. Seismological Research Letters. ISSN 0895-0695. (In Press) https://resolver.caltech.edu/CaltechAUTHORS:20200506-105533707

Zhan, Zhongwen (2020) Distributed Acoustic Sensing Turns Fiber‐Optic Cables into Sensitive Seismic Antennas. Seismological Research Letters, 91 (1). pp. 1-15. ISSN 0895-0695. https://resolver.caltech.edu/CaltechAUTHORS:20200116-083302517

GMG-11 (2022) - Forecast and control of injection-induced seismicity

Experimental plan:

WP1: Develop and test a probabilistic method to disentangle direct and indirect triggering of injection induced earthquakes. The method will provide an estimate of the probability that any particular earthquake was caused by an injection or a previous earthquake.
WP2: Test the method on a selection of examples of injection-induced seismicity, in particular from Oklahoma or the Montney Basin (British Columbia).
WP3: Compare the empirical spatio-temporal kernel functions with predictions from stress-based simulations (combining poroelastic stress calculation an earthquake nucleation based on rate-and-state friction). Assess the possibility of seismicity control through numerical experiments.

Publications:

Wang, G.L. & Avouac, J.P., 2025. Inferring the Causal Structure Among Injection-Induced Seismicity with Linear Intensity Models, Bulletin of the Seismological Society of America, 115, 1406-1434.

Kim, T., Im, K. & Avouac, J.P., 2025. Finite Size Effects on Seismicity Induced by Fluid Injection in a Discrete Fault Network With Rate-and-State Friction, Journal of Geophysical Research-Solid Earth, 130.

Kim, Taeho, Diego Gutiérrez-Oribio, Ioannis Stefanou, Mateo Acosta, and Jean-Philippe Avouac. 2025. "Single-well Based Control and Optimization of Hydraulic Stimulation and Induced Seismicity: Application to the Otaniemi Geothermal Project". Geothermics 132 (November): 103396. https://doi.org/10.1016/j.geothermics.2025.103396.

Kaveh, Hojjat, Jean-Philippe Avouac, and Andrew M. Stuart. 2025. "Spatiotemporal Forecast of Extreme Events in a Chaotic Model of Slow Slip Events". Geophysical Journal International 240 (2): 870–85. https://doi.org/10.1093/gji/ggae417.

Kyungjae Im, Jean‐Philippe Avouac; Maximum Magnitude of Induced Earthquakes in Rate and State Friction Framework. Seismological Research Letters 2024; doi: https://doi.org/10.1785/0220240382

Kim, Taeho;Avouac, Jean‐Philippe (2023) Stress-Based and Convolutional Forecasting of Injection-Induced Seismicity: Application to the Otaniemi Geothermal Reservoir Stimulation. https://doi.org/10.1029/2022jb024960.

GMG-13 (2023) - A Discrete Fault Network model to forecast induced earthquakes, creep and tremors

Project Description:

The project aims at the development of a new modeling capability to simulate fault reactivation (seismic slip, aseismic and tremors) due to a fluid injection for a network of faults. The faults will be assumed to have known geometry and governed by rate-and-state friction. The model will allow taking into account pore pressure diffusion, poro-elastic and thermos-elastic stress changes. The model will make use of the lumped-mass method. This approximation speeds up the computation time making it possible to model a 3-D fault network with non-planar geometries. The model will be benchmarked against dynamic simulations done with the Boundary Integral Method and compared with simple simulations amenable to analytical approximations.

Work plan:

WP1: Benchmarking with simulations run with the Boundary Integral Method.
WP2: Comparison with semi-analytical approximations (cf, Segall and Lu, 2016).
WP3: Comparison with case examples of injection induced seismicity (Coso, Otaniemi).

Publications:

Im, K. & Avouac, J.P. (2025). Maximum Magnitude of Induced Earthquakes in Rate and State Friction Framework, Seismological Research Letters, 96, 1654-1664.

S.L. Antoine, R. Shrestha, C. Milliner, K. Im, C. Rollins, K. Wang, K. Chen, & J. Avouac, The 2025 Mw7.7 Mandalay, Myanmar, earthquake reveals a complex earthquake cycle with clustering and variable segmentation on the Sagaing Fault, Proc. Natl. Acad. Sci. U.S.A. 122 (33) e2514378122, https://doi.org/10.1073/pnas.2514378122 (2025).

Im, K., and J.-P. Avouac (2024). Quake-DFN: A Software for Simulating Sequences of Induced Earthquakes in a Discrete Fault Network, Bull. Seismol. Soc. Am. XX, 1-18, [PDF] DOI: 10.1785/0120230299

Kyungjae Im , Jean-Philippe Avouac, Cascading foreshocks, aftershocks and earthquake swarms in a discrete fault network, Geophysical Journal International, Volume 235, Issue 1, October 2023, Pages 831–852. https://doi.org/10.1093/gji/ggad278

GMG-15 (2023) - Efficient non-ergodic ground motion models using physics-based simulations and machine learning: Methodology and application for hazard assessment at the Groningen O&G field

Project Description:

We propose to develop a computationally efficient methodology for non-ergodic ground motion models (NGMMs) tailored to induced earthquake characteristics. Using ground motion observations and geophysical data from O&G fields (for example, Groningen), the proposed methodology will model tepeatable effects related to the source, path and site, and remove them from the GMM aleatory variability, leading to better characterized site-specific hazard estimation. The proposed framework will also be transferable to regional natural earthquakes, leading to better constrained ground motion predictions and risk assessment for site-specific and for city-scale applications.

Experimental plan:

Use the database of ground motion recordings¹ at Groningen to identify repeatable effects of induced and natural earthquake source, path and site effects.
Combine high resolution subsurface characterization² with wave-propagation simulations of simulated induced event catalog to better identify repeatable path effects.
Remove the repeatable effects from established GMMs for O&G field³, compute reduction aleatory variability, quantify the epistemic uncertainty in the non-ergodic effects, and evaluate impact in site-specific probabilistic seismic hazard analysis.

References:

¹ Ntinalexis M, Kruiver PP, Bommer JJ, et al. A database of ground motion recordings, site profiles, and amplification factors from the Groningen gas field in the Netherlands. Earthquake Spectra. 2023;39(1):687-701. doi:10.1177/87552930221140926

² Kruiver, P., Pefkos, M., Rodriguez-Marek, A., Campman, X., Ooms-Asshoff, K., Chmiel, M., . . . Van Elk, J. (2022). Capturing spatial variability in the regional Ground Motion Model of Groningen, the Netherlands. Netherlands Journal of Geosciences, 101, E16. doi:10.1017/njg.2022.13

³ Bommer, J.J., Stafford, P.J., Ruigrok, E. et al. Ground-motion prediction models for induced earthquakes in the Groningen gas field, the Netherlands. J Seismol 26, 1157–1184 (2022). https://doi.org/10.1007/s10950-022-10120-w

GMG-18 (2025) - Coupling fault rupture and multiphase flow

Project Description: We propose to develop a coupled model of multiphase flow around a fault and fault rupture. Faults often serve as barriers for fluids in the fault-normal direction while being more permeable in the along-fault direction; the permeability of faults and nearby bulk evolves with slip, damage, and healing. We will start by studying water injections into water-saturated medium, with the eventual goal of considering supercritical CO2 injection displacing water. The model will allow us to study fault slip triggered by fluid pressure due to injection away from the fault. We will explore the question of whether the details of fluid injection/extraction dynamics become imprinted in the observable fault rupture dynamics, first in relatively homogeneous models, and then in increasingly heterogeneous ones.

Work plan:

We will couple the boundary integral code for dynamic fault rupture (BiCyclE) with MATLAB Reservoir Simulation Toolbox (MRST)--a finite volume and finite element MATLAB software for multiphase flow and poromechanics. As the first step, the two models will be one-way coupled in space and time via the fluid pore pressure.

Task 1: Implement the coupling framework between MRST and BiCyclE for single-phase injection. Task 2: Compare the rupture dynamics due to fluid injection away from the fault between models with analytical solutions for poroelastic fluid pressure changes and the MRST-generated pressure due to heterogeneous fault permeability structure. Task 3: Exploring the use of our modeling framework as a diagnostic tool of fault stability. Task 4 (time-permitting): Implement the coupling framework between MRST and BiCyclE for two-phase injection and compare the rupture dynamics produced by the singe-phase versus two-phase injection.

Publications: Acosta, Mateo, Thomas Ledevin, Guillaume Salha, Charles Forestier, Lucie Michelin, Xiaojing Fu, and Jean-Philippe Avouac. 2025. "Flow2Quake, an Integrated Multiphase Flow, Geomechanical and Seismicity Model for Efficient Forecasting of Injection and Extraction Induced Earthquakes". International Journal of Greenhouse Gas Control 145 (July): 104388. https://doi.org/10.1016/j.ijggc.2025.104388.

GMG-19 (2025) - Microseismicity-based assessment of stimulated volumen during EGS stimulation: application to the NewBerry Geothermal Reservoir

Project Description: The goal of this project is to develop a framework for the rapid assessment of fracture propagation during EGS stimulation. This framework consists of two modules. The first integrates modern earthquake detection and location techniques—including template matching, machine learning-based detection methods, and double-difference relocation—to enhance earthquake catalogs beyond standard approaches. The second module incorporates a fracture propagation model based on poroelasticity theory and established analytical solutions for hydraulic fractures. We will benchmark our framework using existing datasets from hydraulic fracturing operations, in particular from the Newberry EGS 2014&2025 stimulations.

Work plan:

WP #1: – Building microearthquake templates for template matching analysis to enhance earthquake detection. The goal is to generate two refined earthquake catalogs for the Newberry EGS stimulations in 2014 and 2025.
WP #2: - Relocating the generated earthquake catalogs using double-difference techniques to refine event locations. The goal is to accurately estimate the stimulated volumes during the 2014 and 2025 Newberry EGS stimulations.
WP #3: Applying our analytical model, informed by mechanical and hydraulic data from the stimulations, to simulate fracture propagation during the Newberry EGS stimulations. The results will be benchmarked against fracture geometry and distributions inferred from microseismicity.
WP #4: Open sourcing the framework's components in a Python environment on the GMG website, ensuring accessibility and reproducibility

Publications:

Research Related to GMG but not funded by the IAB

Fukushima, R., Kano, M., Hirahara, K., Ohtani, M., Im, K. & Avouac, J.P., 2025. Physics-Informed Deep Learning for Estimating the Spatial Distribution of Frictional Parameters in Slow Slip Regions, Journal of Geophysical Research-Solid Earth, 130

Milliner, C., Avouac, J.P., Dolan, J.F. & Hollingsworth, J., 2025. Localization of inelastic strain with fault maturity and effects on earthquake characteristics, Nature Geoscience, 18.