Benchmarking Defmod, an open source FEM code for modeling episodic fault rupture

Meng, Chunfang

doi:10.1016/j.cageo.2016.11.014

Cited by 30 publications

(22 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Defmod 19 , an open source finite element code for modeling crustal deformation, was benchmarked and validated by Meng 31 and has been successfully used to model earthquakes induced by fluid withdrawal and/or injection 31 and to investigate deformation in a geothermal field 32 . In this study, we used the poroelastic module of Defmod to model the surface uplift due to wastewater injection.…”

Section: Methodsmentioning

confidence: 99%

Wastewater leakage in West Texas revealed by satellite radar imagery and numerical modeling

Zheng

Kim

Ali

et al. 2019

Sci Rep

View full text Add to dashboard Cite

Wastewater, a byproduct of oil and gas production, is injected into disposal wells. Using Interferometric Synthetic Aperture Radar (InSAR) to observe ground deformation in the Ken Regan field, West Texas, we detected surface uplift that occurred near a wastewater disposal well from 2007 to 2011. High correlation between the observed deformation and the injection volume suggests that the uplift was caused by wastewater disposal in the well. Inverse elastic models were first used to calculate the injection depth and volume. Given the initial estimates of wastewater injection, forward poroelastic finite element models were applied to simulate stress/strain and displacement fields and to estimate the effective injection volume and depth, so as to ultimately understand the subsurface geomechanical processes and provide insight into the local hydrologic properties of the strata in the well location. Results from both elastic and poroelastic models indicate that the effective injection depth is much shallower than the depth reported to the Texas Railroad Commission (RRC). The most reasonable explanation is that the well was experiencing leakage due to casing failures and/or sealing problem(s). The Rustler Aquifer, within the zone of the effective injection depth, has been used as a source of freshwater for irrigation and livestock; wastewater leaked into this aquifer may possibly contaminate that freshwater. Our analysis that exploits remote sensing data and numerical models provides a clue as to understanding the subsurface hydrogeological process responding to the oil and gas activities and an indirect leakage monitoring method to supplement current infrequent leakage detection.

show abstract

Section: Methodsmentioning

confidence: 99%

Wastewater leakage in West Texas revealed by satellite radar imagery and numerical modeling

Zheng

Kim

Ali

et al. 2019

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The numerical part in this study is derived from an open source finite element code Defmod, https://bitbucket.org/stali/defmod, introduced by Ali (). The domain truncation effect is discussed by Meng (, Appendix C).…”

Section: Traction Cancellation To Obtain Half Space Solutionsmentioning

confidence: 99%

Esh3D, an Analytical and Numerical Hybrid Code for Full Space and Half‐Space Eshelby's Inclusion Problems

Meng

2019

Earth and Space Science

View full text Add to dashboard Cite

I present an open source code Esh3D that evaluates the quasi‐static elastic displacement and stress fields perturbed by ellipsoidal inclusions, known as the Eshelby solution. The improvement over my previous work includes obtaining solutions in half‐space by numerically canceling the traction to stage a free surface. The discrete numerical domain allows nonplanar surface. I first briefly introduce the ellipsoidal inclusion problem and solution evaluation. Then, I describe the traction cancellation method in forming the half‐space solutions. Next, I emulate a planar fault with a low aspect ratio ellipsoid and compare the half‐space solution against the well‐known Okada solution for flat and topographic surfaces. Finally, I provide comparisons between the full space and half‐space solutions to illustrate the free surface effects on the displacement and stress fields.

show abstract

“…In quasi-static simulations [12][13][14][15][16][17][18][19][20][21][22][23][24], wave propagation is ignored and the solution is a time series of solutions to static problems with potentially time-varying physical properties and boundary conditions. Such quasi-static simulations cannot accurately model dynamic stress changes resulting from an earthquake, whether they are natural [25] or induced [26,27]. This often means poor accuracy in the predicted event magnitude or the post-earthquake stress state of the fault and the surrounding rock, which in turn means poor prediction of the timing and location of subsequent events.…”

Section: Introductionmentioning

confidence: 99%

“…An important fallout of the quasi-static assumption in the context of induced seismicity modeling is that we do not fully understand the impact of pressure field variations on the seismic velocity and stress changes. Some of the prior studies [26,27] attempted to address this by combining a quasi-static formulation for the fault loading period with an elastodynamic formulation for the coseismic period. However, this required an ad-hoc definition of a transition time between the two formulations and sharp changes in the numerical solution of the problem, which can create numerical oscillations and affect the velocity of stress waves.…”

Section: Introductionmentioning

confidence: 99%

Towards a poroelastodynamics framework for induced earthquakes

Dana¹

2021

Preprint

View full text Add to dashboard Cite

Earthquakes can be triggered after pore pressure perturbations activate critically stressed seismogenic faults, where the perturbations can originate from natural causes like earth tides, rainfall, snowfall or anthropogenic causes like wastewater disposal, CO$_2$ injection, oil production, or groundwater extraction. As the faults slip under the action of the induced stress field, seismic waves are spawned from the hypocenter location. The waves propagate through the domain with a velocity that evolves with the evolving pressure and stress fields. The effect of these waves on the surrounding rock and the seismic velocity recorded on the seismograph can be modeled accurately only by incorporating elastodynamics in the deformation model coupled with flow-induced pressure perturbations. Hitherto, most of the literature in the realm has been limited to elastostatics coupled with flow within a prescribed/kinematic or quasi-dynamic fault slip framework. In this work, we provide a framework for coupling of wave propagation with pore pressure perturbations using one-way coupled poroelastodynamics in the presence of faults in which the pore pressure is specified apriori as a spatiotemporal function.We present results from analysis of displacement and velocity fields in the domain and tractions and slip evolution on the fault. The rendition of two-way coupled poroelastodynamics in which the flow problem is also solved is proposed as future work.

show abstract

Benchmarking Defmod, an open source FEM code for modeling episodic fault rupture

Cited by 30 publications

References 16 publications

Wastewater leakage in West Texas revealed by satellite radar imagery and numerical modeling

Wastewater leakage in West Texas revealed by satellite radar imagery and numerical modeling

Esh3D, an Analytical and Numerical Hybrid Code for Full Space and Half‐Space Eshelby's Inclusion Problems

Towards a poroelastodynamics framework for induced earthquakes

Contact Info

Product

Resources

About