JPT logo
SPE logo

Tech Bits

SPE/SEG Workshop on Injection-Induced Seismicity

In September, SPE and the Society of Exploration Geophysicists (SEG) cohosted a workshop on injection-induced seismicity in Broomfield, Colorado. The goal of the workshop was to determine industry knowledge about seismicity accompanying injection activities, identify the gaps in knowledge, and to discuss ways to improve understanding of such seismicity. Approximately 115 participants gathered to share knowledge from diverse regions and experiences. Attendees were evenly split between service companies and research organizations, and while a majority of the participants described themselves as geophysicists, about 20% were engineers. Encouraging an open discussion between engineers and geophysicists was one of the reasons for holding the workshop.

The opening session set the stage for the workshop, and included presentations about case studies of induced seismicity associated with several types of human activities. The benefits and risks of induced seismicity to subsurface and surface facilities were discussed as a basis for understanding what information is needed to better understand and mitigate potential risks associated with the seismicity.

A poster session (chaired by Jamie Rich) highlighted topics related to characterizing and mitigating fluid-induced seismicity. Other  sessions included:

Underground Mines–In Situ Laboratories for Injection-Induced Seismicity (cochaired by Mark Willis)

This session focused on investigations into the established field of mining-induced seismicity, including case studies from fluid injections in underground mines and the opportunity to perform controlled studies.

Induced Seismicity Associated with Enhanced Geothermal Systems (EGS)—From Understanding to Hazard Control (chaired by Serge Shapiro and Ahmed Abou-Sayed)

This session included discussions about seismicity accompanying the development of EGS reservoirs and how public and regulatory concerns about induced seismicity risk have been addressed particularly through the United States Department of Energy EGS-induced seismicity protocol.

Looking Back at Induced Seismicity – Case Histories Examining Wastewater Disposal and Induced Seismicity – What is Known (chaired by Julie Shemeta and Dicky Hall)

This session consisted of presentations about wastewater disposal operations, observations of the comparison of seismicity with various injections, and potential interaction with existing faults.

What Do We Know About Fraccing-Induced Seismicity (co-chaired by Jamie Rich)

This included presentations about characterizing hydraulic fractures through microseismicity, the types of measurements that are made to characterize induced seismicity, and applications of those measurements. Statistical analysis methods for investigating the attribution of induced seismicity were discussed.

Geomechanics of Injection Induced Seismicity – Faulting and Fracturing (chaired by Mark Mack and Ahmad Ghassemi)

This session provided a geomechanical background about the conditions that are more likely to induce seismicity and the mechanics of faulting and hydraulic fracturing. The stress and pressure changes that are induced by hydraulic fracturing and the stimulation of fracture networks were examined to understand the reservoir perturbations that are expected to develop during fracturing.

What are We Lacking for Predictive Geomechanics (chaired by Norm Warpinski and Mark Houston)

It provided details of the geomechanical conditions that occur during longer term oil, gas, and geothermal injection operations, both with and without associated production. The importance of considering both poroelastic and thermoelastic behavior was described.

Site Characterization – Needs and Challenges for Predicting the Response to Injection (chaired by Jim Rutledge and Ernie Majer).

This addressed three key parts of site characterization:

  • Characterizing and measuring in-situ stress and natural fractures
  • Characterizing faults and faulting in terms of potential risks
  • The use of site characterization and historical seismicity for ground motion prediction and seismic risk assessment

Monitoring the Induced Seismicity (chaired by Leo Eisner and Peter Duncan).

This session covered the role of seismic monitoring in risk assessment, prediction, detection and mitigation. Issues included what needs to be monitored, when the monitoring should take place, how best to implement monitoring (array design, equipment selection and analysis strategies) and where the gaps exist in current monitoring practice.

Risk Mitigation – Panel Discussion (chaired by Bob Siegfried and Werner Heigl)

A panel of industry and academia experts provided comments on reducing risk to the public and infrastructure associated with induced seismicity, and engaged in a mostly interactive discussion with the participants.

Technical discussions at the workshop covered a broad range of topics, and raised a number of relevant and interesting factors specific to injection-induced seismicity:

  • The vast majority of injections do not result in seismicity that can be felt on the surface and it appears to be a very small proportion of all the operations that may have resulted in such induced seismicity.
  • The geomechanical conditions conducive to the occurrence of induced seismicity include a preinjection state of stress close to the level required to cause preferentially oriented, pre-existing faults to slip. The perturbations in stress from increased pore pressure or stress changes associated with fluid injection can lead to either increased or decreased likelihood of slip.
  • Distinctions can be made between “induced seismicity” (might have otherwise not occurred) vs. “triggered seismicity” (accelerated timing of tectonic activity). The causative mechanism between natural tectonic forces and impact of human activity could potentially distinguish these two types of seismicity, although the terms induced and triggered are often used interchangeably.
  • Often it is difficult to uniquely determine the cause and effect mechanisms in order to confidently identify an induced, triggered, or natural tectonic earthquake. Nevertheless a series of seven questions (Davis and Frohlich, 1993) is commonly used to identify induced seismicity.
  • Induced seismicity can occur for some period of time after injection has stopped.
  • Seismograms of induced seismicity are largely similar to tectonic seismicity, and can be analyzed using spectral or waveform analysis to determine the slip area, displacement, and source strength. The source strength is generally reported on the earthquake magnitude scale, and induced seismicity also tends to follow the Gutenberg-Richter frequency-magnitude power law relation originally established for tectonic earthquakes.
  • Fault slip can occur both seismically (signal strength and frequency detectable by seismometers) and aseismically (undetected and possibly not felt due to slow slip).
  • A distinction can be made between seismic hazard (the probability that a given seismic level will occur) and seismic risk (undesired consequences of that hazard). Seismic hazard is often evaluated by estimating the potential source strength or magnitude, but possible risk to infrastructure is better quantified from estimates of the resulting peak ground velocity or acceleration of the associated shaking.
  • A number of distinct techniques may be able to quantify the maximum magnitude that may occur for a certain injection.
  • A “traffic light system” type of scenario can be imagined in which different thresholds of observed seismic hazard are associated with a green, amber, and ultimately red “stop operation” protocol. Operational procedures can then potentially be set up with different actions controlled by the different color light levels.

For further reading