Steps Toward a “Best Practices” Protocol
THE IMPORTANCE OF CONSIDERING THE ADOPTION OF BEST PRACTICES
This report has shown that induced seismicity may be associated with the development of different energy technologies involving fluid injection and sometimes fluid withdrawal (see, e.g., Chapter 3). Furthermore, despite an increased understanding of the basic causes of induced seismicity (Chapter 2), these kinds of energy development projects will retain a certain level of risk for inducing seismic events that will be felt by members of the public (see Chapter 5). While the events themselves are not likely to be very large or result in any significant damage, they will be of concern to the affected communities and thus require both attention before an energy project involving fluid injection gets under way in areas of known seismic activity (whether tectonic or induced) and management and mitigation of the effects of any felt seismic events that occur during operation.
This chapter outlines specific practices that consider induced seismicity both before and during the actual operation of an energy project and that could be employed in the development of a “best practices” protocol specific to each energy technology. The aim of any eventual best practices protocol would be to diminish the possibility of a felt seismic event from occurring, and to mitigate the effects of an event if one should occur. The committee views the ultimate successes of any such protocol as being fundamentally tied to the strength of the collaborative relationships and dialogue among operators, regulators, the research community, and the public (see also Chapter 4). Indeed, protocols, when properly developed and understood, can serve to protect and benefit the various parties involved both directly and indirectly in energy project development.
The chapter begins with a few examples of induced seismicity “checklists” and protocols in the literature that have been developed for the purpose of management of induced seismicity for specific energy projects. The chapter then discusses some of the key components of these checklists and protocols and develops two induced seismicity protocol “templates,” one for enhanced geothermal systems and another for wastewater injection wells. The chapter includes discussion of the incorporation of a “traffic light” system to manage fluid injection and concludes with a discussion of the role and importance of public outreach and engagement prior to and during development of energy projects involving fluid injection. The committee acknowledges that this kind of preemptive management approach
embodied in any best practices protocol for induced seismicity can be complicated by the challenges of determining whether any seismicity felt in a region with injection wells is induced or is due to natural, geologic causes (see Chapter 1). However, we suggest that the benefit of the collective dialogue and establishing best practices in the event of a felt seismic event is in itself constructive, with few or no negative consequences.
EXISTING INDUCED SEISMICITY CHECKLISTS AND PROTOCOLS
Induced seismicity does not fall squarely in the sole purview of any single government agency and, in fact, requires input and cooperation among several local, state, and federal entities, as well as operators, researchers, and the public (see Chapter 4). Because of these shared interests and potential responsibilities, the committee suggests that the agency with authority to issue a new injection permit or the authority to revise an existing injection permit is the most appropriate agency to oversee decisions made with respect to induced seismic events, whether before, during, or after an event has occurred. In many cases this responsibility would fall to state agencies that permit injection wells. In areas that are known by experience to be susceptible to induced seismicity, a best practices protocol could be incorporated into the approval process for any proposed (new) injection permit. In areas where induced seismicity occurs, but was not anticipated in a particular area, existing injection permits relevant to that area could be revised to include a best practices protocol.
Two Checklists to Evaluate the Potential for Induced Seismicity and the Probable Cause of Observed Events
Checklists can be convenient tools for government authorities and operators to discuss and assess the potential to trigger seismic events through injection, and to aid in determining if a seismic event is or was induced. Two checklists, one to address each of these two circumstances—the potential for induced seismicity and the determination of the cause of a felt event—were developed nearly two decades ago by Davis and Frohlich (1993) to address each of these circumstances (summarized in the sections that follow). Their work recommends a list of ten “yes” or “no” questions to quantify “whether a proposed injection project is likely to induce a nearby earthquake” and a list of seven similar questions to quantify “whether an ongoing injection project has induced an earthquake.”
WILL INJECTION INDUCE EARTHQUAKES: TEN-POINT CHECKLIST
The ten-question checklist evaluates four factors related to possible earthquake hazards: historical background seismicity, local geology, the regional state of stress, and the nature of the proposed injection. Table 6.1, modified from Davis and Frohlich (1993), compares
TABLE 6.1 Criteria to Determine if Injection May Cause Seismicity
|Question||NO APPARENT RISK||CLEAR RISK||Texas City, Texas||Tracy, Quebec||Denver RMA, Colorado|
|1a||Are large earthquakes (M ≥ 5.5) known
in the region (within several hundred km)?
|1b||Are earthquakes known near the injection
site (within 20 km)
|1c||Is rate of activity near the injection site
(within 20 km) high?
|2a||Are faults mapped within 20 km
of the site?
|2b||If so, are these faults known to
|2c||Is the site near (within several hundred
km of) tectonically active features?
|State of Stress|
|3||Do stress measurements in the region
suggest rock is close to failure?
|4a||Are (proposed) injection practices
sufficient for failure?
|4b||If injection has been ongoing at the
site, is injection correlated with
the occurrence of earthquakes?
|4c||Are nearby injection wells associated
|TOTAL “YES” ANSWERS||0||10||1||5||4|
aAssumes stress measurements completed prior to survey.
NOTE: RMA, Rocky Mountain Arsenal.
SOURCE: Davis and Frohlich (1993).
the answers of this ten-point criteria list for three injection wells. The wells listed include an existing injection well located in Texas, a proposed injection project in Quebec, and the injection well located at Rocky Mountain Arsenal in Denver with questions answered “as if injection had not yet taken place.”
The authors note, “In actuality, if one were to propose injection at a site near Denver today, the existence of the earthquake activity between 1962 and 1972 would alter the profile, and there would be six or more ‘yes’ answers” (p. 214). The authors go on to say, “At the Tracy, Quebec site we find five ‘yes’ answers…. We would thus conclude that the situation is more similar to Denver than the Texas Gulf Coast” (p. 214).
DID INJECTION INDUCE THE OBSERVED EARTHQUAKE(S): SEVEN-POINT CHECKLIST
The list of seven questions from Davis and Frohlich (1993) again evaluates four factors related to possible cause: background seismicity, temporal correlation, spatial correlation, and injection practices. In Table 6.2 the seven questions are listed and are specifically phrased so that a “yes” answer would indicate underground injection induced the earthquake(s) and a “no” answer would indicate the earthquake(s) were not caused by injection.
Two injection wells are evaluated in Table 6.2. The well in Denver, Colorado, was the injection well at the Rocky Mountain Arsenal, which was definitely shown to be the cause of induced earthquakes in the mid-1960s. The Painesville, Ohio, well, also known as the Calhio well, which was injecting liquid waste from agricultural manufacturing, was investigated as a cause of earthquakes and revealed ambiguous results; the scientists who examined the data could not make a certain correlation between the injection well and the earthquakes, in part due to historical (natural) seismic activity in the area.1
An Example Best Practices Protocol for Induced Seismicity Associated with Enhanced Geothermal Systems
As an example of a protocol used in projects expected to result in induced seismicity, the Department of Energy (DOE) has published a best practices protocol for addressing the potential of induced seismicity associated with the development of enhanced geothermal systems (EGS) (Majer et al., 2012). The steps that a developer might follow in that protocol are summarized in Box 6.1. The DOE states that this protocol is not intended as a proposed substitute to existing local, state, and/or federal regulations but instead is intended to serve as a guideline for the systematic evaluation and management of the anticipated effects of the induced seismicity that are expected to become related to the development of an EGS project.
1 For example, see www.dnr.state.oh.us/geosurvey/earthquakes/860131/860131/tabid/8365/Default.aspx.
TABLE 6.2 Seven Questions Forming a Profile of a Seismic Sequence
|Question||Earthquakes Clearly NOT Induced||Earthquakes Clearly Induced||I Denver, Colorado||II Painesville, Ohio|
|1||Are these events the first known
earthquakes of this character in the region?
|2||Is there a clear correlation between
injection and seismicity?
|3a||Are epicenters near wells (within 5 km)?||NO||YES||YES||YES?|
|3b||Do some earthquakes occur at or near
|3c||If not, are there known geologic
structures that may channel flow
to sites of earthquakes?
|4a||Are changes in fluid pressure at
well bottoms sufficient to
|4b||Are changes in fluid pressure at
hypocentral locations sufficient to
|TOTAL “YES” ANSWERS||0||7||6||3|
SOURCE: Davis and Frohlich (1993).
Using this protocol as a foundation, the committee has adapted the protocol’s set of seven steps in Table 6.3 to illustrate a set of parallel activities, with steps 2 through 7 undertaken essentially concurrently, as opposed to sequentially, to help manage and mitigate induced seismicity from injection associated with EGS. Viewing a protocol as a set of parallel activities is useful not only for general project management but also for the ability it provides to reassess the protocol through time as circumstances of an energy project change and more data are acquired. This resulting matrix form can be used as a template to develop an appropriate protocol to mitigate the potential to induce seismicity in other
The Department of Energy Protocol for Addressing Induced Seismicity Associated with Enhanced Geothermal Systems
The elevated downhole fluid pressures used in EGS induce fracturing that can result in a level of induced seismicity that is felt at the surface and that in some cases has caused serious concern among those living nearby (see Chapter 3). To attempt to avoid the repeated occurrence of such results, while encouraging the future use of geothermal resources, a protocol has evolved to serve as a guide for EGS developers within the United States as well as internationally. The most current protocol, developed by the Department of Energy (Majer et al., 2012), “outlines the suggested steps that a developer should follow to address induced seismicity issues, implement an outreach campaign and cooperate with regulatory authorities and local groups.” This sequence of seven steps can be summarized as follows:
STEP 1. Perform Preliminary Screening Evaluation. Assess the feasibility of the proposed project as to its technical, socioeconomic, and financial risks in order to provide an initial measure of the project’s potential acceptability and ultimate success. Review local regulatory conditions, the level of natural seismicity, and the probable impacts of the project on any nearby communities and sensitive facilities.
STEP 2. Implement an Outreach and Communication Program. Before operations begin, implement a public relations plan that describes the proposed operations, determine the resulting concerns, address those concerns, and then periodically meet with the locals to explain the upcoming operations and the results of the work done to date.
STEP 3. Review and Select Criteria for Ground Vibration and Noise. Identify and evaluate local environmental and regulatory standards for induced vibration and noise. Develop appropriate acceptance criteria for an EGS project.
STEP 4. Establish Local Seismic Monitoring. Collect baseline data on the regional seismicity that exists before operations begin. Install and operate a local seismometer array to monitor the project’s operations.
STEP 5. Quantify the Hazard from Natural and Induced Seismic Events. Estimate the ground shaking hazard from the natural seismicity to provide a baseline to evaluate the additional hazard from the induced seismicity.
STEP 6. Characterize the Risk of Induced Seismic Events. Characterize the expected induced ground motion and identify the assets and their vulnerability within the area likely to be influenced by the project.
STEP 7. Develop a Risk-Based Mitigation Plan. If the level of seismic impacts becomes unacceptable, direct mitigation measures are needed to further control the seismicity. A “traffic light” system can allow operations to continue as is (GREEN), or require changes in the operations to reduce the seismic impact (AMBER), or require a suspension of operations (RED) to allow time for further analysis. Indirect mitigation may include community support and compensation.
energy technologies. The committee has done this exercise for induced seismicity associated with injection wells used for oil and gas development (Environmental Protection Agency [EPA] Underground Injection Control [UIC] Class II wells) or with carbon storage (EPA UIC Class VI wells) and has developed an example of the primary elements that might be included in a best practices protocol matrix (Table 6.4).
THE USE OF A TRAFFIC LIGHT CONTROL SYSTEM
The protocols described in Box 6.1 and Tables 6.3 and 6.4 refer to a “traffic light” control system for responding to an instance of induced seismicity. Such a system, although rarely employed in energy technology projects with active cases of induced seismicity,2 allows for low levels of seismicity but adds additional monitoring and mitigation requirements when seismic events are of sufficient intensity to result in a concern for public health and safety. The preferred criterion to be used for such a control system has been the level of ground motion observed at the site of the sensitive receptor, be it a public or private facility. Seismic event magnitude alone is generally insufficient as the only criterion because of the nature of attenuation (absorption or loss of energy) with increasing distance from an event location to a sensitive receptor site. Zoback (2012) provides a summary of a traffic light system for the purpose of managing potential induced seismicity from wastewater disposal.
As an example, the Bureau of Land Management (BLM) recently issued as its “Conditions of Approval”3 for a proposed EGS project the specific procedures to be followed in the event that induced seismicity is observed to be caused by the proposed stimulation (hydraulic fracturing) operation. The specific procedures included the use of the traffic light control system that allows hydraulic fracturing to proceed as planned (green light) if it does not result in an intensity of ground motion in excess of Mercalli IV (“light” shaking with an acceleration of less than 3.9%g), as recorded by an instrument located at the site of public concern. However, if ground motion accelerations in the range of 3.9%g to 9.2%g are repeatedly recorded within one week, equivalent to Mercalli V (“moderate” shaking), then the operation is required to be scaled back (yellow light) to reduce the potential for the further occurrence of such events. And finally, if the operation results in a recorded acceleration of greater than 9.2%g, resulting in “strong” Mercalli VI or greater shaking, then the active operation is to immediately cease (red light).
The authority for the permitting of Class II injection well location varies by state and is discussed in Chapter 4. Well permits of Class II injection wells in Colorado, for example, are reviewed by the Colorado Geological Survey (COGCC, 2011). During a geologic review,
2 To the committee’s knowledge, the traffic light system has been applied only at the Berlin geothermal field in El Salvador (Majer et al., 2007) and at Basel, Switzerland.
3 R.M. Estabrook, BLM, Conditions of Approval for GSN-340-09-06, Work Authorized: Hydroshear, The Geysers, January 31, 2012.
TABLE 6.3 Primary Elements of a Protocol for Addressing Induced Seismicity in EGS Technologies Adapted as a Series of Parallel Activities Extending over the Lifetime of the Operation
|Initial Screening to Determine the Feasibility of the EGS Project||Assess the local hazard potential from natural seismicity; the local, state, and federal regulations; the nearness of the project to population centers; the probable magnitude of induced events; and the probable risks of potential damage from both natural and induced events. If the proposed EGS project appears to be feasible based on this initial screening assessment, then the Essential Activities of the EGS project as listed below are recommended to proceed in the manner described within each of the five sequential stages of project development as identified herein.|
|Category of Essential Activities||PREPARATION STAGE||DRILLING STAGE||STIMULATION STAGE||OPERATIONS STAGE||COMPLETION STAGE|
|Public and Regulatory Communications||Identify the local people and organizations to be met with. Hold an initial public meeting, explain the planned project, identify their concerns.||Meet with and inform the public, regulators, and media as to the drilling schedule. Upon completion meet and explain the drilling results.||Meet with and inform the public, regulators, and media as to the stimulation schedule and results.||Meet with and inform the public, regulators, and media as to the operations schedule and results.||Meet with and inform the public, regulators, and media as to the project completion.|
|Criteria for Ground Vibration and Noise||Install ground motion and noise monitoring instrumentations.||Report to the public, regulators, and media the monitoring results.||Report to the public, regulators, and media the monitoring results.||Report to the public, regulators, and media the monitoring results.||Report to the public, regulators, and media the monitoring results.|
|Seismic Monitoring||Determine areal size and sensitivity needed for local array. Install and operate the seismic recording array and allow timely public access to results.||Continue to monitor the seismicity recorded and publically report the results.||Add and/or reposition array’s seismometers as needed to follow and characterize the induced events.||Add and/or reposition array’s seismometers as needed to follow and characterize the induced events.||Continue to record and report on the induced seismicity as long as needed to describe the local conditions.|
|Hazard Assessment||Evaluate the potential additional hazard to be expected from the locally induced seismicity.||Review and reassess the potential for damage based on local observations.||Review and reassess the potential for damage based on local observations.||Review and reassess the potential for damage based on local observations.||Report to the public, regulators, and media on any actual hazards observed.|
|Risk Assessment||Develop a probabilistic risk analysis to estimate the probability of risk (monetary loss) to be expected.||Revise the Risk Assessment as appropriate, based on any physical damage, nuisance, and/or economic losses attributed to the project operations.||Report to the public, regulators, and media on the actual results experienced.|
|Direct Mitigation Plans||Develop a plan to control the level and impact of locally induced seismicity.||If needed, implement the control system to cause the drilling, stimulation, or continuing operations to be temporarily reduced or suspended until the level of the locally induced seismicity has been returned to an acceptable level, as determined by the regulatory agencies.||Report to the public, regulators, and media on the actual results experienced.|
|Indirect Mitigation Plans||Provide local jobs, support local community facilities, and provide compensation if appropriate. Continue indirect mitigation activities as long as needed.|
TABLE 6.4 Summary of the Primary Elements of a Protocol for Addressing Induced Seismicity Associated with Injection Wells Used for Oil and Gas Development (EPA UIC Class II wells) or Associated with Carbon Sequestration (EPA UIC Class VI wells)
|Additional UIC Permitting Requirements||After Drilling and Prior to Injection (A Second Look)||Monitoring Requirements During Injection|
|Public and Regulatory Communications||Operator should identify local residents and cities and counties that could be affected by induced seismicity and hold public meetings to explain project and identify concerns.||Operator should notify appropriate regulatory agencies and the local public and provide updated information and analysis based on any new information obtained during drilling operations.||Operator should provide periodic updates to appropriate regulatory agencies and the local public on the locations and extent of their injection operations and the locally observed seismic activity.|
|Hazard Assessment||Evaluate the potential additional hazard to be expected from locally induced seismicity.||Review and reassess the potential for induced seismicity based on any additional information obtained during drilling and completion of the injection well.||Report to the appropriate regulatory agencies and the public on any actual hazards observed during injection activity.|
|Risk Assessment||Develop a probabilistic risk analysis to estimate the probability of risk to be expected.||Revise the risk assessment as appropriate based on any additional information obtained during the drilling and completion of the injection well.||Revise the risk assessment as appropriate based on additional information obtained during injection activity.|
|Criteria for Ground Vibration||Determine areal size, sensitivity, and appropriate instrumentation needed for local array.|
|Seismic Monitoring||Install and operate the seismic recording array to obtain baseline seismic data and record seismic events due to injection activity.|
|Mitigation Plans||Develop a plan to control the level and impact of locally induced seismicity based on the hazard and risk assessment and baseline seismic data.||Revise mitigation plan as appropriate based on any additional information obtained during the drilling and completion of the injection well.||Continuously review and assess mitigation plan to determine effectiveness.|
NOTE: The entire protocol would apply to injection wells proposed in areas where induced seismicity has actually occurred. In areas where induced seismicity was not expected but later occurred, the shaded requirements would apply as revisions to the original injection permit.
the historical earthquake data near the well are closely examined, along with any published fault maps in the area. Additional data regarding fault information, such as that available from three-dimensional (3D) seismic images or other geological information from the well operator may be requested if the well appears to be sited in a high-risk area.
MITIGATING THE EFFECTS OF INDUCED SEISMICITY ON PUBLIC AND PRIVATE FACILITIES
The best practices protocols appropriately include an emphasis on establishing a public relations plan to inform the public as well as the appropriate regulatory agencies of the purpose of the proposed or existing project, the intended operations, and the expected impacts on the nearby communities and/or facilities. Public acceptance begins with an understanding of what is expected to transpire and what contingencies exist for dealing with the unexpected. Inherent in any public information and communication plan is the idea that a developer regularly meets with the local public to explain the schedule and activities of each upcoming stage of operations, as well as the results of the operations performed to date. During the committee’s information gathering session in The Geysers in Northern California and at the associated workshop in Berkeley, we had an opportunity to discuss the 50-year history of induced seismicity at The Geysers geothermal field and meet with the operators, regulatory authorities, researchers, and the local residents from Anderson Springs and Cobb, nearest to The Geysers operations, and subject to the effects of ground shaking due to induced seismicity (see Appendix B—meeting agenda). The discussions we had with these individuals provided some interesting lessons (Box 6.2) regarding the value and potential success of constructive public engagement, for all parties, when induced seismicity may be or becomes an issue in an energy development project. The committee found several very important points to consider regarding the value of successful public outreach, using this example from The Geysers:
1. Time. Public engagement, even if begun early in a project’s planning processes, is a process that occurs over a long time and not a goal in itself. As a process, public engagement requires dedicated and frequent communications among industry, the public, government officials, and researchers.
2. Information and education. Although the initial burden to supply information and to educate local residents lies with the operator and government authorities, residents, too, have a responsibility to become informed and to be constructive purveyors of data and information back to those responsible for operations to allow constructive dialogue to take place.
3. Managed expectations through transparency. Coupled to the sharing of information and education is the idea of managing expectations. Each group involved
The Geysers: Toward Mitigating the Effects of Induced Seismicity
About 40 years ago researchers at the U.S. Geological Survey (USGS) and elsewhere began reporting that induced seismicity was associated with the geothermal production and injection operation at The Geysers (e.g., Hamilton and Muffler, 1972). At first, the causes of the seismicity in this area, where natural seismic activity has a long history, were unclear to the seismologists and to the local operators. Following the installation of additional seismometers to increase the accuracy of locating the events, it became evident that the earthquakes were primarily associated with the injection wells associated with The Geysers and, indeed, essential for continued operation of the field to produce electricity (see Chapter 3; Box 3.1). Consequently, when a pipeline project was proposed 15 years ago to deliver wastewater for increased injection at The Geysers to maintain and enhance power generation, the Environmental Impact Report required the establishment of a Seismic Monitoring Advisory Committee (SMAC) to monitor and report on the production and injection, and seismic activities.
The committee includes representatives of the Bureau of Land Management and California state regulatory agencies, county government, the USGS and Lawrence Berkeley National Laboratory, the local communities, and the operators of the geothermal facilities. Real-time results of the seismic monitoring are continuously available to all at the Northern California Seismic website, and the semiannual meetings of this committee provide a forum for all the stakeholders to compare the locations and magnitudes of the reported seismic events to the locations of the reported production and injection activities.
Despite the benefits of establishing the SMAC, the geothermal operators were still viewed by some local residents as not having taken sufficient responsibility for mitigating the effects of the clearly increased numbers of induced seismic events being felt within the local communities (see Box 3.1), and a petition was filed to declare the situation as being a public nuisance. The county government established two subcommittees to deal directly with the residents of the two local communities of Anderson Springs and Cobb. Each subcommittee has representatives of its local community, the local operators, and the local county supervisor. Ground motion recording instruments were installed in each community, and the resulting information is available in near real time at an independently controlled website. This information allows anyone with Internet access to compare the recorded time of an observed ground motion with the reported times of the separately reported local seismic events in order to determine the location of the apparent source that caused the observed ground motion.
The members of each subcommittee have developed a system of receiving, reviewing, and approving damage claims attributed to the local induced seismicity. Over the past 6 years the geothermal operators have reimbursed the homeowners for their costs to have their home damages repaired, at a total expense of less than $100,000 while contributing funds far in excess of this for improvements to the common facilities in the local communities. In addition the county government has continued to contribute to these communities part of the mitigation funds it receives as redistributions of the royalty payments made to the federal government by the local geothermal operators. This system of coordinating the use of the combined resources of both industry and local government has much improved the mitigation of the effects of the locally induced seismicity, and it is now resulting in much improved and mutually satisfactory relationships among the parties.
SOURCES: DOE (2009); J. Gospe, Anderson Springs Community Alliance, 2011, “Man-Made Earthquakes & Anderson Springs,” DVD, June 30; see also www.andersonsprings.org/.
in an energy development project has different goals and expectations. Mutual understanding of other groups’ goals and expectations is fundamental to developing strong and constructive communication. Transparency regarding these goals and expectations is important to their management.
COGCC (Colorado Oil and Gas Conservation Commission). 2011. COGCC Underground Injection Control and Seismicity in Colorado. January 19. Denver, CO: Department of Natural Resources. Available at cogcc.state.co.us/Library/InducedSeismicityReview.pdf (accessed February 2012).
Davis, S.D., and C. Frohlich. 1993. Did (or will) fluid injection cause earthquakes? Seismological Research Letters 64(3-4):207-224.
DOE (U.S. Department of Energy). 2009. Appendix J—Statement of Compliance with DOE Seismicity Protocol. Geysers Power Company’s Enhanced Geothermal System Demonstration Project, Northwest Geysers Geothermal Field, Sonoma County, California (DOE/EA 1733). Available at www.eere.energy.gov/golden/NEPA_FEA_FONSI.aspx (accessed February 2012).
Hamilton, R.M., and L.J.P. Muffler. 1972. Microearthquakes at The Geysers geothermal area, California. Journal of Geophysical Research 77:2081-2086.
Majer, E.L., R. Baria, M. Stark, S. Oates, J. Bonner, B. Smith, and H. Asanuma. 2007. Induced seismicity associated with enhanced geothermal systems. Geothermics 36:185-222.
Majer, E.L., J. Nelson, A. Robertson-Tait, J. Savy, and I. Wong. 2012. Protocol for Addressing Induced Seismicity Associated with Enhanced Geothermal Systems. DOE/EE-0662. U.S. Department of Energy. Available at www1.eere.energy.gov/geothermal/pdfs/geothermal_seismicity_protocol_012012.pdf (accessed April 2012).
Zoback, M.D. 2012 (April 2). Managing the seismic risk posed by wastewater disposal. Earth Magazine 38-43. Available at nodrilling.files.wordpress.com/2012/02/zoback-earth.pdf (accessed April 2012).