4

Observations, Monitoring, and Technology Evolution and Revolution for Legacy Issues

The second panel session of the workshop provided an overview of observations and monitoring from ecological perspectives, challenges, and breakthroughs in technologies and strategies for managing risk. These three presentations were followed by a panel discussion moderated by Douglas Hollett (U.S. Department of Energy). Hollett noted that this panel would delve into detail on the ways in which to anticipate future legacy issues in an environment where technology and science are advancing rapidly.

PANEL PRESENTATIONS

Observations and Monitoring: Reclaiming the Surface and Measuring Progress

Peter Stahl, University of Wyoming

As Director of the Wyoming Reclamation and Restoration Center, Stahl said he has been involved in revegetation of surface coal mine landscapes since the 1980s, as well as oil and gas well sites. He noted that revegetation or reclamation has now turned into habitat restoration with five important areas to consider: baseline data and reference ecosystems; differences and similarities between reclamation or restoration of unconventional well pads compared to conventional well pads; restoration progress; monitoring; and the long-term legacy impacts of surface disturbance.

Stahl began his discussion of baseline data and reference ecosystems with a photo of a Wyoming landscape showing little vegetation between sagebrush plants and large areas of bare ground (Figure 4.1). Wyoming has compiled significant amounts of baseline ecological data on sagebrush grassland communities, which vary in their health and management approaches. He showed an example of a site in poor condition in a well-known energy extraction area. The poor condition, which was represented by the lack of grasses and forbs, was due to historical grazing in the area, he said. Many groups such as the Bureau of Land Management, the National Resource Conservation Service, and rangeland scientists at the

University of Wyoming have tracked sagebrush grassland vegetation for many years; therefore, the understanding of these ecosystems in the state is relatively good. They have a system of grazing enclosures to understand the influence of grazing on the vegetation, for example.

To help judge the effect of surface restoration efforts for oil and gas wells, native vegetation adjacent to well pads are often used as reference ecosystem sites. Stahl noted, however, that many reference sites are often degraded and do not represent baseline reference conditions. He advocated using multiple reference or baseline sites to avoid the overreliance on any one given site.

Stahl emphasized that from a surface disturbance and restoration perspective, little difference exists between a site producing conventional or unconventional hydrocarbons. Unconventional well pads may be larger in surface area due to the hydraulic fracturing equipment that is placed on the site during operation. Unconventional well pads, however, allow for multiple wells to be drilled from one well pad, which can minimize surface landscape disturbance relative to multiple well pads. He noted that in site preparations, oil and gas operations remove six inches of top soil, which is replaced and revegetated after the operations are completed. He emphasized that the example he showed was from the first energy boom—the oil and gas boom in Wyoming—where replacing soil and revegetation has consistently been done.

Stahl noted progress in restoration efforts of the disturbed sites, with greater success in revegetation and habitat restoration. Operators are improving with regard to restoring the surface landscape. He noted that Wyoming is a difficult place to do this kind of work

because restoring native plant communities in this landscape is unique. Unlike Wyoming, most places do not have undisturbed landscapes. Revegetating cropland is a much simpler effort relative to restoring native plant communities, he said. As an example, he described the low success rate of restoring sagebrush grasslands and the need for improvement in how this kind of restoration is done. Wyoming has an extremely stressful environment for plants, he said, because of the semiarid climate, salty arid soils, and big weed problems. Cheat grass, an invasive species, is another big issue; the native plant seed that operators buy to use for revegetation is often contaminated with cheat grass seed, he said.

According to Stahl the University of Wyoming is collecting data on restoration methods, environmental conditions, and restoration success which are then put into a database to help support best management practices. The database has been used in mapping disturbance and determining reclamation success around the state and was very useful to the conservation efforts of sage grouse, for example.

Monitoring restoration and reclamation is basically a question of measurements, Stahl said. Improvements in restoration and reclamation cannot be made without monitoring, he stressed. Few operators want to spend time monitoring, which is primarily conducted by crews of college students in the summertime and most of these students are not botanists. He also noted that monitoring data quality is a problem. Monitoring crews work all summer long, but Wyoming vegetation is ephemeral. Green vegetation exists when soil moisture conditions are sufficient; by the end of July that moisture is often gone. The ephemeral plant species are only observable when sufficient moisture exists in the soil at the right time of year. Thus, if one is monitoring in August or later in the year, plant species may not be present. The seasonality and ephemeral nature of certain plant species therefore have implications for the accuracy of monitoring data.

Stahl also emphasized the lack of permanent monitoring records. The University of Wyoming collects data for their own database. If researchers see conflicting data on data sheets that get completed by hand in the field, they have no way to determine which data point is correct, as one cannot go back in time to the site. They are advocating the use of new digital photography-based monitoring methods to have permanent records.

With respect to the long-term legacy of surface disturbance in Wyoming, Stahl noted the establishment of sage grouse conservation areas due to the many different kinds of surface landscape disturbance in the state and the concern for sage grouse. The conservation areas in which disturbance is minimized amount to a substantial portion of Wyoming’s landscape. Stahl suggested that about 25 percent of the state is now protected to maintain habitat to prevent the listing of the sage grouse. In 2015, sage grouse was recommended not to be placed on the endangered species list by the U.S. Fish and Wildlife Service. This decision will be reviewed in 2020. The concern for the protection of the sage grouse was an issue that brought regulators, industry, academia, and the public together to prevent the listing of the bird as an endangered species and to maintain the bird’s habitat. The success of these efforts, in part, depends on the resiliency of the sagebrush grassland ecosystem in Wyoming.

Challenges and Breakthroughs in Downhole Technologies: From Plugging to Long-Term Monitoring

Barry Freifeld, Lawrence Berkeley National Laboratory

Freifield has spent the last 25 years of his career at the Lawrence Berkeley National Laboratory examining subsurface technology with more recent focus on the natural gas storage industry. He noted the importance of understanding the legacy concerns of natural

gas storage and that he took part in an interagency report led by the Department of Energy, which released recommendations based on what was learned from the gas leakage at the gas storage facility in Aliso Canyon, Los Angeles County, California.

Natural gas storage began in the United States around 1920, he said. Since that time, operators of gas storage fields stored gas underground in depleted wells in their fields with the intent to remove the gas at a later time to sell. The gas storage industry thus evolved based on legacy infrastructure, Freifeld said. Currently, about 17,500 wells that are spread out over 400 different sites with about 200 different operators in the country. The average well age is about 50 or 60 years, with most built in the 1950s and 1960s. Because operators wanted to move gas in and out of these wells, they removed the tubing from the wells and put the wells on casing injection and production. Under this scenario, the barrier to leakage is removed and there is high pressure on the well casing. The operators also evolved their own best practices for these wells as gas storage fields over time.

Although the Aliso Canyon incident in California was the most recent failure in terms of a major leak from an underground gas storage facility (it emitted approximately 5 billion cubic feet of gas over more than 100 days in 2015), Freifeld also noted gas leakage of about 6 billion cubic feet with an accompanying fire and explosion at Moss Bluff gas storage facility in Texas in 2004 (Clark, 2004) and the gas leak from the Yaggy underground gas storage facility in Kansas which contributed to an explosion and fire (Allison, 2001).

Following the Aliso Canyon leak, the state of California issued emergency regulations in February 2016 indicating that operators had to examine all of their gas storage fields. Operators identified wells that needed maintenance and roughly two-thirds of the gas storage operations in the state were temporarily shut down. For example, Pacific Gas & Electric, which is northern California’s biggest natural gas operator, temporarily shut their McDonald’s Island storage field. The lesson from this experience, Freifeld suggested, is the need to consider whether the practices being undertaken are indeed best practices or whether practices can be improved to prevent future problems.

Freifeld then described the three prerequisites for a leak from a well: a source of fluid; a gradient to drive the fluid; and a pathway for the fluid to move (Figure 4.2). Focusing on the subsurface, he noted that the casing strings and the casing cement act as a system. The source of the fluid could be from deep in the production zone or from shallower zones.

In addition to a source, a pressure gradient is needed to drive the fluid to the surface. He mentioned that it is very difficult to drive high-density, heavy brine up the well; however, a pressure gradient can drive gases such as methane, ethane, and propane to the surface more readily. Potential migration pathways to move the fluid to the surface can be found in several locations along a well (Figure 4.3).

The leakage of gas can occur via different pathways and he said that while cementing is important, it is never a perfect system. Cement can shrink; it has compressive strength, but it does not have any tensile strength, which can result in tensile fractures, he said. Freifeld then outlined leakage pathways between the casing and cement (a) in Figure 4.3, leakage on the other side of the casing (b) in Figure 4.3, and leakage through the porosity of the cement itself (c) in Figure 4.3.

Freifeld investigated leaking wells and proposed the use of technology such as fiber optics to look at fluid flow and well integrity in the surface and intermediate casing. Operators had been interested in the technology to look at fluid flow, but had not applied fiber optics regularly to examine the casing. He noted that operators pay attention to what is taking place in the production zone, but they do not have an incentive to examine surface and intermediate casing and what is taking place in the cement around the casing adjacent to the rock formation. They do monitor sustained casing pressure, and a lot of states regulate

that aspect because it is an indicator of well integrity and potential failures. An additional indicator of well integrity is leakage at the surface.

New technologies exist today that can be used to assess well integrity and reservoir protection, Freifeld said (Figure 4.4). Monitoring behind the casing for pressure and fluid chemistry, which multiple vendors use to measure pressure in the production zone, could also be used to measure flow along the surface casing. The fiber optic technology of distributed temperature sensing (DTS) and distributed acoustic sensing (DAS) may be used to assess leakage. They are also used by operators to monitor what is taking place in the production zone.

To conclude, Freifeld emphasized four points:

• The role of field studies and the role government could play in partnering with industry to measure leakage rates and where leakage is occurring behind the well casing;
• Monitoring of groundwater wells and surface monitoring to help determine baseline conditions prior to and after a well has begun operating;
• Before wells are plugged and abandoned, establish well integrity. Regulatory bodies do not currently require this. He emphasized there is no benefit to plugging a well that is leaking up to the surface behind the well casing; and
• Risk management plans, as part of plugging and abandoning a well, could help avoid future negative legacy impacts.

Strategies for Managing Risks and Minimizing Negative Legacy Issues

Mark Boling, Southwestern Energy

Boling opened his remarks by taking the approach of general risk management to minimize negative legacy impacts. Risk management, he said, can be represented in three steps:

• Identify the risk;
• Assess the probability of occurrence, and the potential impact of each of the risks; and
• Implement strategies to mitigate the risk.

He noted two types of risk: actual risk and perceived risk. He emphasized that public policy

and regulation are not always based on actual risk, but often on the perception of the risk. Closing the information gap between actual and perceived risk is critical for effective risk management and achieving public trust on a given activity, for example, unconventional hydrocarbon development. Assessing the risk and potential impacts from the risk is also key to successful implementation of strategies to mitigate the risk, he said.

Boling discussed three different categories of regulatory influence to mitigate risk through influencing behavior:

• Command and control regulations. Regulations that dictate specific technologies or practices for conducting operations. He gave the example of pollution control and the use of certain scrubbers in coal production.
• Performance-based regulations. Regulations that establish performance standards and allow industry some operational flexibility to meet standards. Particularly during times of rapid technology advancement, a performance-based approach can generate innovation.
• Incentive-based regulations. Regulations that encourage private investment to achieve public policy objectives unrelated to mitigating risk.

A fourth type of regulatory influence may also deserve consideration, he said: a hybrid approach between the incentive-based, and either command and control or performance-based regulation. Boling then suggested that operators may be incentivized to go above and beyond regulations to mitigate risk assuming they are not penalized for trying a new technology or practice that may prove beneficial to minimizing negative environmental legacy.

In terms of mitigating risk by influencing behavior, Boling then discussed the importance of stakeholder opinions from investors, consumers, and the public. Public companies listen very intently if shareholders are concerned about issues such as methane emissions, he said. Consumers, such as large-scale purchasers (e.g., commercial and industrial) have some influence over how public companies conduct their business. He noted the critical role public pressure can play, by actions or opinions expressed, that can influence a company’s behavior in mitigating risk. Boling then commented on the scientific community and the critically important work that they do. He suggested that the scientific community could take additional steps to consider how best to communicate the science to the target audience. Other types of influencing behavior exist within companies themselves. These internal influences can include the tone at the top of a company and how that tone is transferred to all employees. Another internal influence on behavior is incentive compensation. Southwestern Energy used incentive compensation with methane emissions, for example. A bonus plan was put in place where people in the company would get a bonus if they were able to reduce the company’s emission intensities rate. Another example he shared was an energy conservation water challenge that Southwestern Energy began a few years ago. The challenge involved a pledge to become fresh water neutral in their operations by the end of a particular calendar year.

To conclude, Boling shared some thoughts on mitigating well integrity risks (Figure 4.5). He asked the audience to consider what they thought might be the best way to influence behavior to mitigate the risks.

MODERATED DISCUSSION

Hollett began the discussion by noting that Freifeld talked about wellbore systems in gas storage systems and wellbore systems in shale systems. He asked Freifeld to expand upon issues related to leakage from shale wells, for example, versus leaks from full casing productions out of gas storage systems. Freifeld said that California instituted surface leak detection requirements as part of its emergency regulations and other more intensive well integrity monitoring. He suggested that there was little oversight of the gas industry before the Aliso Canyon gas leak event. With the new regulations, leaking wells have been detected prompting several operators to respond and fix the well leaks. Freifeld noted that a Proceedings of the National Academy of Sciences paper (Osborn et al., 2011) shows an increase in incidents of methane gas detected within about one kilometer of unconventional hydrocarbon resource development projects. Other statistical evidence indicates that methane is migrating out of these wells. Gas is buoyant and tends to dissipate and will find the pathway of least resistance. When discussing either gas storage or unconventional development, the mechanisms of gas migration are similar, Freifeld said. The corrosion and failure of cement systems is analogous between gas storage and unconventional systems. He emphasized a holistic risk perspective of well design and integrity and asked how wells could be designed to ensure that there are no legacy impacts 20 or 30 years into the future.

An online participant asked if Wyoming has a Surface Damages Act or some other requirement that drives industry to minimize or mitigate surface impacts and whether a regulatory framework is necessary to address these impacts or if a voluntary industry action be adequate? Stahl said that he does not think Wyoming has a Surface Damages Act and that he personally favors a regulatory approach.

Another online participant asked if soil compaction is an issue in site reclamation and whether soil compaction affects long-term revegetation success in Wyoming. Stahl indicated that soil compaction can be an issue if, for example, soils are re-applied to an area in a wet condition; in Wyoming this would not be common. One participant noted that if the cement is leaking between the casing and the rock, then conventional well plugging is not useful. Freifeld suggested one way to address the issue would be to mill out the casing, although this is expensive and not usually used. He elaborated that an operator should not be allowed to plug and abandon a well until it is demonstrated that the well has integrity and the plugging is functioning as intended. Hollett noted a lot of interest in adaptive materials and DOE’s funding of some research in this area. The materials, which would be used outside of the well casing environment, have the capacity to respond to both stress and chemical changes, and help address leakage.

Another participant raised a question related to well integrity and monitoring in tectonically active areas. Freifeld said that cases of well failure during earthquakes are known in Southern California. A lot of these wells are used to inject steam for low permeability resource development. In general, Freifeld noted that one has to look at the risk of well failure which is not addressed in regulations. Boling noted that Southwestern has a lot of wells around the country and has not had any experience with seismicity affecting their wells to date.

Hollett asked Boling to comment on how to communicate risk so that it is understood. Boling said one way is to use animations to tell the story of an issue such as well integrity. He emphasized the importance of going out into the community and interacting with people in ways that they can understand. Freifeld followed by suggesting that communication is a lot more complicated when dealing with legacy issues. The public understands risks that may happen tomorrow or next week, but legacy impacts are something that may happen in the next generation. Technologists and scientists do not have a good mathematical understanding of that future risk and can only talk in general terms. They still have not done the hard work of figuring out the real risk, Freifeld said.

A participant commented on the apparent policy failure with respect to surface reclamation and asked if it would be helpful to contextualize the Wyoming experience from which others could learn. The participant also asked about influencing behavior change within industry, beyond the use of regulation, and was especially interested in relationships with the investment community. Do particular examples in oil and gas development or other sectors exist where there are opportunities to learn from past experiences? Boling responded by emphasizing the importance of explaining risk. In the context of hydraulic fracturing, he noted that the way hydraulic fracturing was explained to the public did not help to focus on the real problem. Further, there was little understanding of what the public understood about the hydraulic fracturing process. He also mentioned methane emissions and said that the investment community is looking at natural gas relative to coal in the context of global warming. Investors see methane emissions as a problem, the scope of which has yet to be defined.

A participant asked Stahl how the jurisdictional complexity of Bureau of Land Management field offices in Wyoming contributes to or detracts from reclamation successes. Stahl noted that this complexity is a real concern. Each field office in Wyoming has its own re-

quirements for successful reclamation which creates significant concern for operators as they have to meet different standards wherever they are working. These approaches then lead to different reclamation outcomes. Some of these standards are not stringent and may not lead to ecosystem restoration. Others standards are stringent and are difficult to meet. The University of Wyoming has investigated reclamation success and sees very few success stories.

Hollett posed a question to all panelists regarding wells that were constructed about 40 years ago and managed since that time, versus wells constructed and managed in more recent times. He emphasized the dramatic difference in how wells are managed now (e.g., new drill rig design, well pads are smaller, and changes in truck traffic) relative to the past. He asked if panelists had thoughts on game changers in terms of surface footprint, environmental impact, and operation. Stahl indicated that in the future, he would like to see better regulations on monitoring of surface reclamation, to get insights into successful reclamation. Freifeld noted that the path to a well pad with no legacy issues is to have no well pad at all. His presentation focused on environmental issues such as leakage, but he said that if one examines unconventional resource recoveries, the percentage recovery from the reservoir is quite low. If this percentage can be increased due to better access to production zones, negative impacts to the environmental legacy may be decreased. Although not a game changer in terms of reducing future negative environmental legacy, Boling said that an important game changer in their operations was getting smarter about logistics. For example, moving water by pipe rather than trucking water led to decreased impacts to community infrastructure. He emphasized that to achieve success, an integrated approach of innovation and regulation might be helpful.

A participant raised a final question on the apparent lack of comprehensive federal approaches to well site remediation and management. Boling responded saying that most of the regulations associated with what is happening at the well site are regulated at the state level.

This page intentionally left blank.