Carnegie Mellon University
October 15, 2013

Assessing the Risk of Carbon Storage

Assessing the Risk of Carbon Storage
This article is Part 3 of a three-part series that explores CEE research on carbon capture, utilization, and storage (CCUS), an innovative technique being considered for its potential to slow climate change. Part 3 focues on CEE efforts to develop a risk assessment framework for use in CCUS operations.
Part One  Part Two

As levels of atmospheric carbon dioxide (CO2) continue to rise at a sobering rate, CEE researchers are exploring a bold technique that could play a key role in global greenhouse gas reduction. The technique is known as carbon capture, utilization, and storage, or CCUS, and has a straightforward objective: capturing CO2 and storing it underground before it makes its way to the atmosphere. Through the collaborative efforts of the NETL-RUA, a partnership between the National Energy Technology Laboratory and five nationally recognized universities (CMU, Penn State, Pitt, Virginia Tech, and WVU), a team of CEE researchers and NETL scientists are researching the potential benefits and risks associated with CCUS.

Assessing monitoring techniques

When CO2 is stored in reservoirs more than a kilometer underground, how do you find out whether it’s leaking? Ask CEE H. John Heinz Professor Mitchell Small. For the past six years, Small has studied ways to assess the performance of monitoring networks that would be used to detect leakage of stored CO2 at CCUS sites. “There is a whole suite of monitoring techniques that can be used for monitoring CO2 leakage,” Small explained. “And all of them are complicated by the fact that CO2 occurs naturally in the environment.” Small has collaborated on the project with Yale’s Cheryl Yang (CEE ‘11), NETL’s Grant Bromhal, Brian Strazisar, and Arthur Wells, and the University of West Virginia’s William Gray and Egemen Ogretim

Small and his collaborators are researching a number of monitoring methods to determine the limits of each method. One of these methods is surface CO2 flux monitoring, in which a structure resembling a cake dish cover is placed on the ground at a site and the CO2 levels underneath it are measured periodically. “The challenge is whether the CO2 levels become high enough so that they look different from what might naturally occur,” Small explained. “For that reason, surface CO2 flux monitoring is most effective for detecting big leaks.” The team is also looking at the use of isotopic ratios to distinguish between naturally occurring CO2 and CO2 leaking from the subsurface. 

Other techniques in the project include groundwater sampling and the addition to injected CO2 of tracers (such as perfluorocarbons) to injected CO2, which act as red flags for scientists taking samples from a CCUS site. While the team is mainly studying monitoring that can be conducted near the surface, Small explained that they are working toward methods used to detect the leak at its root – in the storage reservoirs, which lie more than a kilometer underground. One such method is seismic surveying, in which scientists use induced seismic waves to produce images of the subsurface that can reveal cracks in the caprock or the movement of CO2 and other fluids towards the surface.

While all of these methods are valid ways to detect escaping CO2, Small noted that they are most effective when combined strategically. “We’re confident that we can detect big leaks, but we want to make sure that we can also identify the small leaks, which over time could negate the greenhouse gas abatement benefits of carbon storage,” he said. “We found that combining monitoring methods is the best approach, both to detect leakage at different stages and to detect different types and sizes of leaks.”

As Small and his collaborators continue their research on various monitoring methods, they grow closer to achieving the project’s chief objective: linking the monitoring research with risk assessment methods to form an adaptive management risk model. “Monitoring is critical,” he said. “If you have a strong monitoring system, you can detect issues and respond to them in a way that limits damage and keeps a site viable for a longer period of time.”

Using simple monitoring methods to detect escaping CO2

To get a good idea of what is happening underground in a carbon storage site, site operators must select from a wide array of monitoring techniques. Satellite-based measurements, 4D seismic surveys, and other methods discussed above are just a few of the many ways to detect CO2 leakage. However, these techniques vary widely in cost; conducting annual seismic surveys of a CCUS site may cost on the order of 5 million dollars, which over a period of thirty years could have a price tag of more than 150 million dollars. CEE PhD candidate Nick Azzolina thinks some of the more economical methods that are routinely collected as part of standard industrial practice have much to offer.

Azzolina, who is advised by Small and by Professor of the Practice Dave Nakles, is evaluating the possibility of using underground pressure monitoring to detect CO2 leakage at CCUS sites. “Pressure measurements are made routinely as part of subsurface reservoir management,” Azzolina explained. “So if they’re already collecting the information, the added cost for monitoring is negligible. If you can get similar detection methods using these lower-cost methods, then you’re potentially saving millions of dollars.”

Using mathematical models to simulate a hypothetical CCUS site, Azzolina designed a series of plausible CO2 leakage scenarios and analyzed model outputs of pressure taken both within and above the storage reservoir to determine whether the pressure measurements were enough to alert site managers to leakage. “As you inject CO2 into a reservoir, the pressure goes up in the reservoir, whereas you don’t expect to see pressure rise above the overlying caprock,” he said. “So if you’re monitoring pressure underground and it’s not behaving as your model anticipated, you need to be asking why it’s happening.” 

Azzolina found that measuring pressure in only one area—the storage reservoir itself—was not enough to detect small CO2 leakage rates. However, another tactic showed encouraging results. “When we coupled that routine measurement with pressure measurements taken in other areas above the caprock, we could detect much smaller leakage rates,” he said. “In other words, combining measurements improved our ability to detect when CO2 is leaking.”

In addition to studying the potential of pressure monitoring, Azzolina is exploring the use of other practical monitoring methods to detect leakage. He plans to expand his focus to include soil gas measurements and groundwater salinity monitoring, with the ultimate goal of helping site owners design more cost-effective monitoring plans. And he noted that his findings on low-cost monitoring methods don’t have to be limited to CCUS. “Monitoring research can be broadly applied to many different processes, from carbon storage to shale gas drilling and extraction,” said Azzolina. “I’m hoping that some of these lessons that we’re learning can also be useful elsewhere.” 

Putting the pieces together

In the vast, deeply complex process of carbon capture and storage, site operators can find themselves swimming in information: monitoring data, samples, geologic history, regional trends, and more. They need a way to integrate this information into a unified, adaptive site management plan so that carbon storage sites can be managed safely and effectively. That’s where CEE Professor of the Practice Dave Nakles enters the picture. Nakles is working to develop an integrated risk assessment framework that can be used to predict and, if necessary, mitigate environmental and economic risks throughout the life cycle of a CCUS site. The framework, which involves an overarching DOE modeling framework known as CO2 PENS (Predicting Engineered Natural Systems), is the focus of a collaboration between CEE and NRAP, the National Risk Assessment Program of NETL. 

“A carbon storage site is made up of many parts – the reservoir where the CO2 is stored, the seal above it, and the various groundwater aquifers closer to the surface – and each of these parts can be described using mathematical models that are based on the fundamental physics and chemistry of these subsurface zones,” Nakles explained. “We’re working to assist NRAP by tying all of these models together and using them to describe the various ways that a molecule of CO2 could work its way up from the storage reservoir to a drinking water aquifer or the atmosphere.”

So how does a risk assessment actually work? First, researchers identify the potential risks of a process; in the case of CCUS, one such risk might be the escape of CO2 through the seal (a layer of rock overlying the storage reservoir) followed by its contact with a drinking water aquifer. The researchers then assign a probability to each of these risk pathways that describes the likelihood that each risk could become reality. Those probabilities, combined with the potential environmental or human health impact of such an event, tell the site manager which areas need to be carefully monitored during the site’s life cycle, from planning and operation to closure and post-closure monitoring.

Nakles noted that risk assessment can be used to support adaptive management strategies for a carbon dioxide storage site. Adaptive site management strategies utilize targeted, risk-based monitoring data from a site to modify the site operations so as to minimize the potential risk associated with its operations. “Risk assessment allows for more intelligent operation of the system,” he said. “The sooner one can apply this risk assessment framework and guide the development of the storage operations, the better.” He cited an example of a potential CCUS site where his team’s preliminary studies of the site revealed a high risk that injected CO2 would contaminate a nearby natural gas field. To minimize this risk, the site managers moved the injection point to a neighboring geologic formation that was separated from the gas field by a natural geologic barrier. “By addressing the issue at the start, we were able to avoid the problem rather than discovering it after operations were initiated and being forced to apply expensive remediation strategies to mitigate it.”

For Nakles, the most gratifying part of the project is integrating multiple data sources into a comprehensive tool that will avoid the potential risks as a carbon storage site. “The excitement and the creativity is integrating these models into a functioning system that is capable of supporting a risk assessment,” he explained. “We’re connecting the pieces.”

Top graphic: Larry Scott, Colorado Geological Survey

In carbon capture, utilization, and storage (CCUS), carbon dioxide emitted from industrial sites such as power plants is captured, compressed, and injected into underground reservoirs. Potential storage sites range from depleted oil and gas reservoirs to saline aquifers and unmineable coal deposits.