Monitoring CO2 containment, part 1
Atmospheric monitoring
The global awakening to the urgency of emissions reduction and the march towards new energy is transforming the energy sector. With energy companies, industrial corporations and governments motivated with a vigor, the likes of which we have not seen before, it appears that this revolution will succeed in a significantly lower carbon future.
With ever increasing momentum, new(ish) energy sources such as solar, wind, geothermal and hydrogen, are all attracting a dizzying amount of subsidy and interest.
However, as clean, or low carbon, as these sources of energy may be, the need to increase the sequestration of produced carbon dioxide (CO2) will continue to grow. Only a few weeks ago ExxonMobil announced it is planning for more than 20 new carbon capture and storage (CCS) opportunities around the world to enable large-scale emission reductions.
Further motivation was provided by the US Treasury Department in June 2020 who issued regulations to implement Section 45Q of the Tax Code, which provides tax credits for capturing and sequestering CO2. Those involved with a broad range of carbon capture projects and technology can now claim tax credits under Section 45Q of up to $50 per ton of CO2 captured and placed in secure geological storage. ‘Secure’ is important. Under ‘Special Rules’, note F, point 2 of the code, it states,
“… establish regulations for determining adequate security measures for the geological storage of qualified carbon oxide under subsection (a) such that the qualified carbon oxide does not escape into the atmosphere.”
This sets the stage for the requirement of comprehensive long term monitoring programs for all projects claiming a 45Q tax credit which is currently available for the next 12 years (at least).
This credit is hugely significant to the economics of CCS projects. Schmelz et al (2020), modeled the total cost of CCS projects in the northeastern and midwestern US and estimate that more than 8 Gt of the total CO2 emissions from that region can be stored for less $60 per ton. In otherwords, at a post credit cost of just $10 per ton.
Bottom line, no monitoring, no tax credit!
Global CCUS Projects 2021
From the learnings and knowledge sharing of many of the early large scale CCS demonstration projects, the broad framework of CCS monitoring programs has been established. Over the years, I have been fortunate to have been involved, in some form or another, in many of these (Aquistore, Decatur, Otway, Quest, Sleipner, Weyburn-Midale, Leismer, Citronelle, Farnsworth).
These programs are commonly referred to as monitoring, verification and accounting (MVA) projects or sometimes, measurement, monitoring and verification (MMV). Either way you get the point!
Whichever acronym you choose, these (MVA) programs are designed to monitor the movement of the CO2 in the subsurface over time, to ensure it stays within the desired geological formation and to make sure there is no leakage at the surface.
An MVA program has three main objectives.
Satisfy all legal monitoring requirements (i.e., 45Q to secure the tax credit)
Optimize operational storage efficiency
Reassure the public that containment has, and continues to be, achieved.
An additional regulatory requirement to consider is that since 2010, in the US, the Environmental Protection Agency (EPA) have identified those wells used for the geologic sequestration of CO2 as class VI wells which have been subject to specific construction and monitoring requirements.
An MVA program is designed according to a site-specific risk assessment, this is not a one size fits all solution. With that in mind the IEA Greenhouse Gas R&D program has a nice online tool summarizing most of the available monitoring tools and will make program recommendations based on user input on geological setting (https://ieaghg.org/ccs-resources/monitoring-selection-tool).
MVA technologies can be categorized into three main areas based on where the injected CO2 is being measured.
Atmospheric
Surface / Near-surface
Subsurface
Cartoon illustrating three monitoring areas of a typical CCUS MVA program
All MVA programs should contain elements from each category to demonstrate containment at the required depth within the designated geological formation(s).
This blog post will talk about atmospheric monitoring, subsequent posts will cover the rest.
Part 1 : Atmospheric Monitoring
The atmospheric monitoring component of an MVA aims to demonstrate that the injected CO2 does not leak into the atmosphere and involves sampling the air a few feet above the ground surface for elevated levels of CO2 or trace gases present in the injected gas. These tools need to be very sensitive to measure concentration levels above the natural background levels and the techniques need to address the fact that leakage may not occur near the point of injection. The gas could leak along fault planes or aquifers and reach the surface some distance from the planned storage site.
These atmospheric changes are typically small, the U.S. Department of Energy has adopted a loss rate of approximately 1% of the injected CO2 over a period of 100 years as an acceptable rate (Flude et al, 2016). One of the main challenges with this technique is differentiating between CO2 which is naturally occurring and that which has leaked from the subsurface.
There are three main techniques for atmospheric monitoring of CO2 related to confirming subsurface confinement.
eddy covariance (EC) flux measurement techniques
tracking of atmospheric tracers
the use of optical CO2 sensors
The eddy covariance (EC) method is a micrometeorological technique which directly measures gas transport (or flux) between the surface and atmosphere. In (very) simple terms, an appropriately placed 3D anemometer measures the interaction of surface airflow (wind) with eddies caused by the movement of CO2 from the surface, into the atmosphere. Burba et al (2013) give a very good overview of the technique and describe the use of this method at the Midwest Geological Sequestration Consortium, Decatur, Illinois site.
Tower mounted 3D anemometer and gas analyzer (top). Air flow over an area with no flux (a), air flow over an area with flux (b), bottom. Burba et al, 2013.
Solar powered eddy covariance station at CCUS site near Decatur, Illinois (left). Setting up the station, right. Burba et al, 2013.
Another, more compelling, IMHO, atmospheric technique used to monitor for the potential surface leakage of injected CO2 is tracking geochemical tracers which, in some cases, have been added to the CO2 prior to injection (for example sulfur hexafluoride, SF6). An advantage of using geochemical tracers is they are detectable at very low concentrations. Etheridge et al (2011), describe an interesting case study from the Otway Basin of Victoria, Australia.
In addition to the CO2 flux measurements using eddy covariance, described above, the Otway project utilized two types of gas tracers. SF6 was added to the CO2 prior to injection and Methane (CH4) comprised about 20% of the injected gas but occurs at relatively low atmospheric concentrations. These gas tracers were monitored from a tower located 700m northeast of the injection well. Reassuringly no evidence of leakage of the injected CO2 was found using the tracer data which was supported by results from sub surface monitoring techniques used at the site, over the same time interval. In the absence of leaked CO2, Etheridge et al (2011) use data from planned surface events to demonstrate the effectiveness and sensitivity of the atmospheric measurements. One such event was the scheduled venting of gas from the observation well (Naylor-1) related to a geochemical sampling campaign. The timing of the event was selected when the wind was blowing toward the air monitoring tower. The atmospheric concentrations measured at the tower over that time period are shown below.
CO2CRC Otway project location (left), schematic of well locations and monitoring types (right), in particular note, ‘Airmonitoring’ site.
Courtesy UKCCS Research Center Blog post
Atmospheric concentrations during scheduled venting of well fluid. The venting period occurs at about 10:00 and is represented as a solid bar. The upper chart shows two blue curves representing a continuous CO2 measurement and a flux type measurement. Each measurement shows clear diurnal variation. The lower chart shows concentration changes in tracers SF6 and CH4. Both show positive correlation with the venting period. The baseline curves are from an air pollution meteorological station in Tasmania. After Etheridge et al (2011).
The upper chart shows the typical diurnal variation of atmospheric CO2; high during the night associated with CO2 emission, consistent with respiration and low during the day associated with CO2 uptake, consistent with photosynthesis. There is no obvious correlation to the venting period discernable within the diurnal range fluctuations. The lower chart shows that concentrations of both SF6 and CH4 were significantly higher during venting suggesting that these tracers could prove beneficial in the presence of leaked CO2 from the subsurface.
Another class of atmospheric CO2 monitoring technique uses optical sensors such as infra‐red diode lasers or non‐dispersive infra‐red gas analyzers. These sensors can be combined with 3D anemometers used to measure EC flux (described above) to build a tomographic image of the atmospheric gas concentration above a CO2 leakage site. Bayesian inversion can then be used to determine the location of the CO2 source and the size of the gas plume. A challenge with the approach is that these sensors are not able to differentiate between CO2 released from storage and natural variations in ambient CO2. Also, surface leaks may not occur near where the CO2 was injected so a large array of these sensors would be required.
Jones et al (2009) describe an early example of using both systems to identify naturally occurring gas vents in the Laacher See, a flooded caldera, in Germany and the Latera caldera in central Italy. The image below (left) shows results from a tripod mounted infra‐red diode laser. An aerial image of the Caldera with the CO2 concentrations derived from two days of acquisition are overlain. The two areas of elevated readings (red) correspond to the location of the gas vents. The data on the right shows results from a non‐dispersive infra‐red gas analyzer. The transects were walked with the analyzer at various heights from the ground. The results clearly indicate the sensitivity of the measurement to sensor height which, when more than 10cm above ground level, is unable to resolve the second, weaker vent.
(Left) Image of mobile laser data from the western shore of the Laacher See. Data from consecutive days from the central portion are both displayed and show that the two main vents were clearly identified on both occasions. (Right) Walking traverses over gas vents at Latera with the ground surface measurement system (infrared analyzer) measuring. CO2 concentrations at different heights show the fall-off in response with increasing height; the weaker vent is not seen at heights greater than 10 cm.
A potential large scale airborne optical system which leverages differential absorption LIDAR (DIAL) is one of the newest technologies in this field (https://www.nist.gov/programs-projects/differential-absorption-lidar-detection-and-quantification-greenhouse-gases). The system operates at two infrared wavelengths; one resonant with the gas being investigated and one not. The resonant wavelength is more strongly absorbed so the difference between the two signals is proportional to the density of gas present. The system is designed for quantifying greenhouse gas emissions from a number of sources including industrial and natural sources.
The cartoon illustrates the DIAL system measuring two plumes of CO2 emissions separated by 500 meters. The panel on the right is a simulation of the raw signal return, the log of the range corrected signal, and the final DIAL result which clearly shows the concentration of CO2 in each plume of gas.
The techniques described above for measuring released or leaked CO2 into the atmosphere, are very much the last line of defense. Other (subsurface) monitoring methods should provide early indications of gas migration upward toward the surface and allow time for mitigation measures to be implemented. It is therefore crucially important that these atmospheric surveillance measures can demonstrate ongoing stable background levels of CO2 and potentially utilize scheduled operational events, such as venting, to confirm their required sensitivity in the event of unplanned subsurface leakage.
The next blog update will look at surface and near surface monitoring methods including geochemical monitoring of soils and vadose zone, and of course my new favorite subject, surface displacement monitoring with InSAR.
References
Burba, George & Madsen, Rod & Feese, Kristin. (2013). Eddy Covariance Method for CO2 Emission Measurements in CCUS Applications: Principles, Instrumentation and Software. Energy Procedia. 40. 329-336
Stephanie Flude et al. (2016) Inherent Tracers for Carbon Capture and Storage in Sedimentary Formations: Composition and Applications. Environmental Science & Technology 50 (15), 7939-7955
Charles R. Jenkins et al. (2012) Safe storage and effective monitoring of CO2 in depleted gas fields. Proc Natl Acad Sci U S A. Jan 10; 109(2): E35-E41
Etheridge, David et al. (2011). Atmospheric monitoring of the CO2CRC Otway Project for large scale CO2 storage projects. Energy Procedia. 4. 3666-3675
D.G. Jones et al. (2009) New and established techniques for surface gas monitoring at onshore CO2 storage sites, Energy Procedia, Volume 1, Issue 1, Pages 2127-2134
William J. Schmelz et al (2020) Total cost of carbon capture and storage implemented at a regional scale: northeastern and midwestern United States. Interface Focus.102019006520190065http://doi.org/10.1098/rsfs.2019.0065