International Journal of Greenhouse Gas Control 103 (2020) 103188 Contents lists available at ScienceDirect International Journal of Greenhouse Gas Control journal homepage: www.elsevier.com/locate/ijggc Application of emerging monitoring techniques at the Illinois Basin – Decatur Project Curt Blakley a, *, Carl Carman a, Chris Korose a, Don Luman a, Joseph Zimmerman a, Michael Frish b, Jeremy Dobler c, Nathan Blume c, Scott Zaccheo d a Illinois State Geological Survey, Prairie Research Institute, University of Illinois, 615 East Peabody Drive, Champaign, IL, USA Physical Sciences Inc., 20 New England Business Center, Andover, MA, USA Harris Corporation, 1919 W Cook Road, Fort Wayne, IN, USA d Atmospheric and Environmental Research, Inc., 131 Hartwell Avenue, Lexington, MA, USA b c A R T I C L E I N F O A B S T R A C T Keywords: Midwest geological sequestration consortium Illinois Basin – Decatur Project Tunable laser spectroscopy GreenLITE Aerial imagery Remote sensing The Illinois Basin – Decatur Project is a large-scale carbon capture and storage demonstration project located in Decatur, Illinois, USA. In this project, one million metric tons of carbon dioxide (CO2) was captured from an ethanol production facility and successfully injected into a deep saline reservoir over a period of three years. The scale of this project presented an opportunity to explore emerging technologies for effective long-term monitoring of the carbon capture and storage site. This research documents the application of three emerging monitoring techniques: (1) a prototype “open-path” sensor, a method of continuously monitoring atmospheric CO2 by applying tunable diode laser absorption spectroscopy to provide a warning system for personnel safety along pipelines and at wellheads; (2) the Greenhouse Gas Laser Imaging Tomography Experiment (GreenLITE), an automated system for measuring two-dimensional spatial distribution of atmospheric CO2 concentrations; and (3) periodic aerial imagery, a method of documenting surface-vegetation dynamics to detect vegetative responses to CO2 leaks. The goal of this work was to assess the advantages and limitations of these monitoring techniques and quantify their reliability over an extended deployment at an active industrial site. These results will aid in determining viability of long-term monitoring on a commercial scale. 1. Introduction The Midwest Geological Sequestration Consortium (MGSC), led by the Illinois State Geological Survey at the University of Illinois, is one of seven Regional Carbon Sequestration Partnerships funded by the U.S. Department of Energy (DOE), has been exploring the long-term potential of geological carbon dioxide (CO2) sequestration in the Illinois Basin since 2003. In addition to investigating the regional feasibility of this process, the MGSC has completed smaller scale pilot projects to assess potential CO2 enhanced oil recovery (Frailey et al., 2012a, 2012c; Midwest Geological Sequestration Consortium (MGSC, 2009) and CO2 enhanced coalbed methane recovery (Frailey et al., 2012b). The MGSC has conducted a large-scale carbon capture and storage (CCS) demonstration project, the Illinois Basin – Decatur Project (IBDP), at the Archer Daniels Midland Company (ADM) ethanol production facility in Decatur, Illinois, USA (Fig. 1). In this project, one million metric tons of CO2 was successfully injected into the Mt. Simon Sandstone reservoir from November 17, 2011, to November 26, 2014, at a rate of approximately 1000 metric tons per day. The success of CO2 storage activities at the IBDP site was a significant factor in establishing a second demonstration project, the Illinois Industrial Carbon Capture and Storage (IL-ICCS) project. These unique projects have laid the groundwork for commercial-scale storage in the Illinois Basin and nationwide. However, CCS activities at these and future sites must be carefully monitored to demonstrate they are protective of human health and the environment. The monitoring, verification, and accounting (MVA) program for the 0.65 km2 project site at the IBDP was led by the Illinois State Geological Survey (ISGS), with cooperation from Schlumberger Carbon Services and ADM. Near-surface and subsurface monitoring began in 2008 and is scheduled to continue through at least 2020 in accordance with the terms of the U.S. Environmental Protection Agency-issued injection permit. Details of recent MVA efforts at the IBDP site can be found in Locke et al. (2011, 2013), Carman et al. (2014, 2019), and Korose et al. (2014). The MVA at the IBDP was a collaborative effort in which more * Corresponding author. E-mail address: cblakley@illinois.edu (C. Blakley). https://doi.org/10.1016/j.ijggc.2020.103188 Received 26 March 2020; Received in revised form 3 October 2020; Accepted 5 October 2020 Available online 6 November 2020 1750-5836/Published by Elsevier Ltd. C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 than 20 different near-surface and subsurface monitoring methods were tested for their efficacy in monitoring large-scale CCS projects. This research explores the application of three prototype, or less frequently used, technologies in an industrial setting to determine their long-term viability and potential application for a future full-scale commercial CCS project: site storage efficiency and protect human health and the environment. 3 Aerial photography captured through color infrared (CIR) imaging, which documents plant stresses resulting from potential CO2 interaction. Because of significant surface-changing site activities at the IBDP site, CIR imaging was terminated; however, aerial photography was continued to document substantial changes to the project site caused by operational and industrial activities. These visualizations helped researchers interpret near-surface monitoring data and communicate with regulators and other stakeholders. 1 A prototype “open-path” sensor (OPS), which uses tunable diode laser absorption spectroscopy (TDLAS) to continuously monitor atmospheric CO2 along a laser beam’s line of sight approximately 100 m long. The OPS was intended as a warning system (not an absolute concentration measurement) to be used along pipelines and at wellheads for personnel safety and to help verify site storage efficiency. 2 The Greenhouse Gas Laser Imaging Tomography Experiment (GreenLITE), which is an automated system for measuring the twodimensional (2-D) spatial distribution of atmospheric CO2 levels over a large area at a field site. The system that was deployed at the IBDP site consisted of two laser-based transceivers, 30 retroreflectors, and cloud-based software tools for data processing, storage, and dissemination, from which 2-D maps of CO2 concentrations could be generated in near real time. This system was intended to help verify 2. Methods 2.1. Open-path sensor Deployment of the OPS at the IBDP site offered a unique opportunity to evaluate TDLAS diagnostic tools at an active CCS site with an extensive environmental monitoring program already in place. Tunable diode laser absorption spectroscopy may be a cost-effective method for identifying anomalous temporal or spatial changes in atmospheric CO2 concentrations that indicate leakage, especially along pipelines and near the wellhead infrastructure (Frish, 2017; Frish et al., 2014). Fig. 1. Regional location of the Illinois Basin – Decatur Project (IBDP) site. 2 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 of the optical components affecting alignment. The measured values of F2/F1 may be converted to the path-integrated CO2 concentration (i.e. ppm-m) based on a set of calibration constants generally determined by inserting a reference cell into the laser path. The reference cells contain a known amount of CO2 within a closed tube that has laser-transmissive windows at both ends. However, such calibration is unnecessary for the leak detection application described in this study. As illustrated in data presented in Fig. 9 below, leaks are indicated by characteristic temporal fluctuations in the CO2 signal, not by any specific increase in CO2 concentration. 2.1.1. Theory of operation The OPS system uses the near-infrared (NIR) absorption spectra of the CO2 molecule to monitor the CO2 column-density (i.e. pathintegrated concentration) along its optical path (Fig. 2a). Components of the OPS consist of (1) a tunable diode laser source that transmits a beam along the optical path toward (2) a passive directionally reflective target, which returns laser radiation back to (3) the receiver optics, which collect the returned laser radiation onto (4) an infrared (IR) detector. Additional components include a solar power system with a 19.3 A-h lithium-ion battery and a 65-watt, 12 × 6 m solar panel, a laser control and signal processing circuit board, a wireless radio modem that transmits the board data to a remote workstation, and a computer workstation that displays output and logs data. The OPS system applies wavelength modulation spectroscopy (Fig. 2b) to measure the absorption over an infrared CO2 feature of interest. The modulated laser current at a determined frequency (fm) results in intensity modulation at the detector, creating a detector signal component at frequency fm, called F1. In addition, the laser wavelength is swept over the CO2 absorption feature of interest. During a single current modulation cycle, the laser wavelength is scanned across the band of the selected absorption feature twice, resulting in a component of the detector signal amplitude modulation at 2*fm, called F2. The signal processor demodulates the periodic signal to measure the root mean square amplitudes of F1 and F2, which are used to calculate the relative CO2 amount. Frequency F1 is a measure of the laser power received at the detector; it is generally free of influences from other sources of optical power (e.g., sunlight) that are also sensed by the detector. Frequency F2 is proportional to the product of the received laser power and path-integrated CO2 concentration. Thus, the ratio F2/F1 provides a measurement that is directly proportional to the CO2 abundance and is independent of the received laser power. This feature enables the sensors to track relative CO2 levels despite changes in laser power transmitted across the optical path. Such changes can occur, for example, because of variability in reflectance of the illuminated backscatter surfaces, precipitation in the optical path, or thermally induced movements 2.1.2. TDLAS application Wavelength modulation spectroscopy takes advantage of the fact that CO2 molecules absorb IR radiation on well-known, narrow-wavelength bands centered on specific wavelengths in the IR electromagnetic spectrum (Miller and Brown, 2004). The OPS sensors operate in the NIR (1–2.5 μm) spectral regions. Frish et al. (2011) discussed the criteria for designing TDLAS sampling instruments optimized for precision and accuracy. Further details related to the configuration and deployment of TDLAS-based CO2 sensors for CCS monitoring have recently been reported (Zimmerman et al., 2014a, 2014b). Similar techniques have also been demonstrated for monitoring methane at industrial sites by using both handheld sensors (Frish et al., 2005) and fixed open-path configurations (Frish and Cummings, 2013). 2.1.3. Deployment description and duration An OPS prototype (Fig. 3) was deployed at the IBDP site from June 16, 2012, to December 22, 2013, to demonstrate its use as a warning system for personnel safety along pipelines and at wellheads and to verify storage containment for CO2 accounting purposes. It was designed for simple installation and easy operation with limited maintenance. The OPS was situated immediately north of the IBDP CO2 injection well (CCS1) and supply pipeline, with the 100 m monitoring path running east (laser transceiver) to west (reflector). The unit continuously transmitted CO2 absorption data at 0.1 Hz to a workstation located in a trailer Fig. 2. Operation of the prototype CO2 open-path sensor (OPS) (a) tunable diode laser absorption spectroscopy transceiver sensor principles; (b) wavelength modulation spectroscopy (WMS) signal processing: laser wavenumber (v) modulates across a spectral absorption line with a modulation depth (δ) creating a modulated transmitted signal at the detector (Frish et al., 2011). Reprinted from Energy Procedia 63, Zimmerman, J.W., Locke, R.A., II, Blakley, C.S., Frish, M.B., Laderer, M.C., Wainner, R.T., Tunable diode laser absorption spectrometers for CO2 wellhead and pipeline leakage monitoring: Experiences from prototype testing at the Illinois Basin – Decatur Project, USA, p. 4083–4094. Copyright 2014, with permission from Elsevier. 3 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Fig. 3. Prototype CO2 open-path sensor (OPS) installation at the IBDP CO2 injection wellhead in Decatur, Illinois: (a) aerial view of the IBDP injection site, and (b) OPS transceiver with a solar panel supply. Reprinted from Energy Procedia 63, Zimmerman, J.W., Locke, R.A., II, Blakley, C.S., Frish, M.B., Laderer, M.C., Wainner, R.T., Tunable diode laser absorption spectrometers for CO2 wellhead and pipeline leakage monitoring: Experiences from prototype testing at the Illinois Basin – Decatur Project, USA, p. 4083–4094. Copyright 2014, with permission from Elsevier. south of the injection well. of atmospheric CO2 with the atmospheric CO2 and soil CO2 flux measurements the ISGS obtained by IR gas analysis-based techniques at the IBDP site (Carman et al., 2019). 2.2. GreenLITE 2.2.1. Theory of operation GreenLITE is a laser absorption spectrometer that uses an intensitymodulated continuous wave (IM-CW) technique to measure the differential optical depths between two fixed wavelengths selected at specific points on a CO2 absorption feature, one near the peak of the feature and one in an area of relatively low absorption. From this differential optical depth and these measured local weather data, the CO2 concentration can be determined (Measures, 1984). The system is arranged in such a manner that both transceivers can point to each retroreflector in series (Fig. 4a) and measure the concentration along each path, or “chord,” in The GreenLITE system measures the 2-D spatial distribution of atmospheric CO2 concentrations over areas between 1 and 30 km2 for use in MVA programs at CCS sites. The system consists of a pair of laserbased transceivers, 30 retroreflectors (for this installation), and a set of cloud-based software tools for data processing, storage, and dissemination, which can generate 2-D maps of CO2 concentrations in near real time. The GreenLITE system was deployed at the IBDP site to demonstrate autonomous operation of the system in a real-world CCS environment over an extended period and under a wide range of operating conditions. A secondary goal was to compare GreenLITE measurements Fig. 4. GreenLITE system at the IBDP project site (a) layout showing chords from transceivers (T1 and T2) to reflectors, (b) transceiver installed at the IBDP site, and (c) retroreflector installed at the IBDP site. 4 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 the measurement field. The set of retrieved concentration values for all the chords becomes input for a sparsely sampled tomographic reconstruction algorithm to generate 2-D models of the field concentrations. GreenLITE computes dry-air CO2 mole fractions along the chords of interest, and then combines these values to create 2-D georeferenced maps of the distribution of CO2 over the field of interest. These algorithmic approaches are described in depth in Dobler et al. (2015) and are summarized here. The dry-air mole fractions are computed based on the standard steepest descent search technique to find the radiative transfer model based the optical depths (Δτ) that best match the observed Δτ given the atmospheric state (surface temperature, relative humidity, and pressure) along the chord of interest. The line-by-line radiative transfer model (Clough et al., 2005) computes optical depths based on Voigt line shape functions at user-defined atmospheric levels and a continuum model that includes self- and foreign-broadened water vapor as well as continua for CO2, O2, N2, O3, and extinction attributable to Rayleigh scattering. The version used in this work included 2012 updates to the CO2 line parameters and coupling coefficients based on the work of Devi et al. (2007a, 2007b). The 2-D estimates of CO2 distributions are defined by the best fit of a 2-D analytical plume-based model to a set of retrieved chord concentration values. In this approach, the field of interest is defined as FCO2 (x, y) = a + bx + cy + dxy + N ∑ 2 2 αn e−βn xr e−γn yr , leveraged both the airborne and the point-to-point IM-CW approaches to address the need for continuous monitoring of CCS sites. The retrieval and reconstruction algorithms leveraged prior work by Giuli et al. (1991, 1999) to develop an approach that integrated point-to-point chord measurements with a sparsely sampled tomographic method to reconstruct the underlying 2-D distribution of CO2 over an extended open-air site. Dobler et al., 2017 provides discussion of instrument optimization for precision and accuracy over the series of four deployment applications. The detection and mapping capability of GreenLITE was first demonstrated at the Zero Emissions Research and Technology field site maintained by Montana State University in Bozeman, where a 70-m pipeline is being used to diffuse CO2 into the soil approximately 2 m below the surface at rates up to 0.3 metric tons/day. The GreenLITE system was tested over a 4-week period in the fall of 2014 and demonstrated the ability to detect not only the controlled release of CO2 but also the CO2 signature from a manure pile located near the test site. Additionally, GreenLITE was shown to track diurnal trends extremely well by comparison with atmospheric point measurements made with a LI-COR-based West Systems surface flux meter. Without an underground release of CO2, the IBDP did not provide the ideal scenario for testing the ability of GreenLITE to reliability detect and accurately map the release of a CO2 plume. (1) 2.2.3. Deployment The primary purpose for deploying GreenLITE at the IBDP site was to demonstrate the ability of the system to operate autonomously for an extended period in a wide range of environmental conditions. GreenLITE was deployed at the IBDP CCS site in February 2015. The configuration of the installation allowed for monitoring an area of approximately 0.2 km2, which was largely determined by obstructions and site topography. Although this area did not include the main injection well, which had been shut in since November 2014 when injection operations at the IBDP site were concluded, it did include the IBDP deep monitoring well known as Verification Well 1 (VW1) and much of the area covered by the IBDP soil flux monitoring network grid. Preoperational site preparation included mowing the site to provide a clear line of sight and installing power receptacles for the transceivers. After a preoperational evaluation, the system became operational and continuously and autonomously collected data from April 1, 2015, until August 17, 2015. Fig. 4 shows a diagram of the site layout and images of the system components. As installed, the GreenLITE system measured CO2 column concentrations for 60 chords ranging in length from 157 to 595 m approximately 1.5–3 m above the surface, depending on the surface topography. Four on-site maintenance visits were made at roughly one-month intervals to clean the transceiver optical windows and the retroreflectors, adjust the positioning of retroreflectors that had been struck by mowing crews, and remove vegetation obstructing the chord lines of sight. n=0 where, F(x, y) represents the CO2 concentration as a function of the x, y location, and the linear model on the right-hand side describes the field elements. The first four terms in the equation describe a basic background that consists of a constant offset and a gradient term that varies linearly as a function of x and y. The variable a describes the average background concentration, and the remaining elements describe a simple linear gradient across the field. The summation represents a set of simple Gaussian plumes to describe potential localized CO2 sources. Each 2-D Gaussian plume is a function of a normalized set of xr and yr parameters, defined as √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ xr = (x − xn ) + (y − yn ) cos(θn ) (2) and √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ yr = (x − xn ) + (y − yn )sin(θn ), (3) where (xn, yn) is the plume center and θn is the angle of rotation. The number of modeled plumes is limited by the number of observed chords. The upper limit is defined as the number of independent chord measurements minus the four background parameters, divided by the number of plume parameters. For example, a site with 40 intersecting chords limits the number of plumes to a maximum of 6 ([40 – 4]/6 = 6). In practice, the number of supportable plumes is on the order of 3–4 because of the complexity of the search space and the independent nature of the chord values. The best fit F(x, y) is constructed by minimizing the root mean square error between the retrieved CO2 part per millions (ppm) values and the modeled values constructed through discrete integration of F(x, y) by using a sequential least squares programming optimization algorithm. 2.3. Aerial imagery The primary objective of airborne remote sensing at the IBDP was to provide documentation of carbon sequestration and industrial activities at the site by using high-resolution, digitized aerial photography. Once the IBDP site was selected, equipment and other infrastructure were quickly established (Fig. 5), and it was deemed desirable to establish a record of surface changes during the active life of the site. Although high-resolution photographs in both CIR and natural color were obtained for the IBDP site, CIR was not fully implemented to detect plant stresses related to potential CO2 leakage because of limitations imposed by substantial landscape modification resulting from industrial activities (rather than sequestration activities) and because CIR film processing was more costly than natural color. 2.2.2. Intensity-modulated continuous wave application The IM-CW approach was initially developed as a potential mechanism for actively measuring CO2 sources and sinks from space. This concept was first demonstrated by Harris in 2004 and then by the ITT Corp. (now ITT Inc.) from an airborne platform, and it has been used extensively for evaluation by NASA (Dobbs et al., 2007, 2008; Dobler et al., 2013). In 2011–2012, Dobler (2013) evaluated a novel implementation of this approach in which a separate transmitter and receiver were used to enable point-to-point measurements, potentially from a geostationary orbit or between ground stations. The GreenLITE system 5 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Fig. 5. Map showing the IBDP wells and the original locations of selected near-surface monitoring installations (before loss or abandonment). Aerial imagery was acquired March 29, 2016, from the Illinois Department of Transportation. camouflage detection, contains an emulsion layer sensitive to the lower end of the NIR, at approximately 0.74 to 0.85 μm. Wartime applications of CIR-based aerial photointerpretation led directly to its widespread use by governmental agencies and the private sector for many succeeding decades, particularly in the area of natural resource applications. Approximately 15 years ago, digital mapping camera (DMC) systems that incorporated four spectral bands (the visible spectrum and NIR) became commercially available and have since become commonplace. 2.3.1. Theory Surface vegetation is commonly used in remote sensing studies as an indicator of external stressors, wherein both the plant color and the internal cell structure of plants can undergo changes in vigor because of abnormal stressors such as drought, disease, damage, or other factors. Although photosynthesis in green plants converts water and CO2 into O2 in the presence of sunlight, abnormally high levels of CO2 can be deleterious to many plants, eventually resulting in changes to the plant cell structure. Terashima et al. (2014), highlights the impact that plants may face in response to the prolonged exposure to elevated CO2 and Jones et al. (2014), provides a compilation of free-air CO2 enrichment (FACE) experiments that study of the effects of elevated CO2 on plants and ecosystems. These changes can be detected by and are enhanced in the reflected NIR portion of the electromagnetic spectrum, which extends from wavelengths of approximately 0.74–2.5 μm. Aerial CIR film, originally developed in the United States during World War II for enemy 2.3.2. Earlier MGSC pilot project experience From June 11, 2007, to October 2, 2009, CIR aerial imagery was periodically acquired over the MGSC’s ECBM pilot project at the Tanquary Farms site in Wabash County, Illinois (Frailey et al., 2012b) by using a Zeiss/Intergraph Imaging digital mapping camera system (Leica Geosystems, 2016). The Tanquary project imagery was digitally remotely sensed and calibrated to assess any potential vegetative 6 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 response to subsurface CO2 injection. When aerial imagery collection was implemented in 2007 at the ECBM Tanquary Farms site, the intent was to monitor potential impacts of CO2 leakage primarily detected from stress on local vegetation. Although no adverse impacts of injection were detected, which was a favorable outcome for the injection project, the process and results also helped in reevaluating aerial imagery needs for future CCS projects. The combination of (1) the low frequency of aerial overflights (biannual), (2) the lack of purposeful CO2 release testing to evaluate the potential of NIR/CIR imagery to detect stresses on local vegetation, (3) the inability to detect any adverse impacts at the Tanquary Farms site, and (4) the high cost of DMC imagery (~$13,000 for the deliverables from each overflight) relegated the acquisition of aerial imagery to the role of site documentation instead of monitoring. aerial framing camera was used for the acquisitions. Because the photographs were collected with traditional film, the exposed film was first processed by an outside contractor. When the processed film was returned to IDOT, the photogrammetry staff digitized each frame with a high-precision Leica aerial film scanning system. As an MSGC partner, IDOT contributed the aircraft mobilization, aerial photography processing, and digitization costs to the IBDP. The only cost incurred by the IBDP for this effort was for the purchase of the aerial film. The digitized images were subsequently forwarded to the ISGS for postprocessing. In contrast to the DMC geographic information system (GIS)-ready orthorectified imagery used at the ECBM pilot project, geometric distortions inherent within the IDOT aerial camera had to be corrected (georeferenced) for each frame to fit a standard cartographic map projection to utilize the digitized images for mapping or GIS applications. 2.3.3. Deployment description and duration at the IBDP Aerial photography was used as part of the MVA program at the IBDP site to document changes in land use from the beginning of project activities through the post injection phase of the project. The Illinois Department of Transportation (IDOT) Aerial Survey Division conducted the CIR aerial photography acquisitions at the IBDP site beginning in September 2008. Film-based CIR aerial photography was chosen over DMC-based collection because it was a more cost-effective method of site documentation and was preferable for the development of GIS base maps of the IBDP site. Two overflights of the IBDP site were conducted each year, one during the spring season and a second during the late summer or early fall. Additionally, two separate photograph acquisitions were conducted for each overflight: (1) a higher altitude collection to capture the entire IBDP site and surrounding area (Fig. 6), and (2) a lower altitude collection to capture the injection well and immediate environs at an increased ground resolution (Fig. 7). A large-format (22.86 × 22.86 cm) 3. Results 3.1. Open-path sensor The OPS operated near the injection wellhead at the IBDP field site for nearly a year with limited maintenance, and it rapidly detected variations in path-integrated CO2 concentrations, including those related to routine injection operations. Fig. 8 shows typical outputs from the OPS sensor from August 4, 2013, to August 11, 2013. The detector’s DC voltage, a measure of the total light (laser plus solar) received, varied as expected, with time of day and cloud cover affecting the solar power received (Fig. 8c). Short-lived spikes often occurred before sunset when the sun was aligned most directly with the path of the OPS transceiver. However, sunlight was assumed to have no effect on instrument performance unless it saturated the detector, which did not occur during this test. Even with these responses to sunlight, the ability of the system Fig. 6. Color IR image overflight of the IBDP site and surrounding area, acquired on September 9, 2008. The original image scale is representative fraction (RF) 1:19,200, or 1 cm = 192 m. The ADM facility is at the lower left center, and the IBDP injection site is situated adjacent to the plant within the yellow outlined area. 7 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Fig. 7. Higher resolution CIR image of the IBDP injection site acquired on September 9, 2008. The original image scale is RF 1:7,200, or 1 cm = 72 m. to identify anomalous signals caused by CO2 leakage was not expected to be substantially affected. The relative atmospheric CO2 concentration (Fig. 8d) calculated from F2/F1 varied diurnally (as is typical of CO2 levels around vegetation during the summer), reaching peak levels in the early evening and decreasing to minimum values at night. The F1 and F2 varied substantially as the laser power collected by the detector varied. A notable event occurred on the morning of August 8, 2013, when the received laser power was substantially reduced by absorption in fog and mist, although the typical CO2 diurnal concentration change was still observed despite the diminished F1 and F2 magnitudes. During maintenance operations performed on the IBDP injection well, it was possible to test the OPS responses to exposure of CO2 concentrations greater than atmospheric background. On September 6, 2013, routine pipeline blowdown procedure elevated CO2 concentrations for approximately 1 h near the OPS installation before CO2 injection operations resumed (Fig. 9 a). A similar procedure on October 25, 2013, resulted in a lower relative change in monitored concentration for roughly 3.5 h (Fig. 9b). The response to the plume released during maintenance was much larger during September than during October. This was attributed to a smaller CO2 release rate and the influence of the wind, which was stronger (3.6–4.5 m/s) and out of the west-southwest, encouraging less transport of the CO2 plume across the OPS beam path during the October measurement period. In contrast, winds were light (<2.2 m/s) and out of the south, causing the exhaust plume to be carried toward and across the OPS beam path during the September measurement period. Nevertheless, the distinct rise in average concentration relative to the background combined with a high degree of skewness in the CO2 signal characterizes the leak and distinguishes it from sensor noise. By discerning these temporal changes in CO2, which can be as small as system noise and significantly less than the diurnal CO2 variations, this method provides a very sensitive leak detector without need for accurate calibration of CO2 concentration (Frish, 2017). 3.2. GreenLITE system During the approximately 5 months that the GreenLITE system was operational at the IBDP site, we obtained a data set containing more than 2 million raw samples of chord optical depth, 1.8 million column CO2 concentration values, and approximately 72,000 2-D reconstructions. The system collected data for approximately 3,800 h, with an uptime duty cycle of greater than 95 %. The GreenLITE system operated in a wide range of environmental conditions, with temperatures ranging from –20 to 33 ◦ C and wind gusts up to 27 m/s. In situ data were collected by the ISGS with a LI-COR Biosciences LI8100A infrared CO2/H2O analyzer with a LI-8150 multiplexer, enabling comparisons with the GreenLITE measurements. Measurements from the multiplexer were collected every 30 min from eight ports: four closed accumulation chambers (two bare shallow and two natural shallow rings) that measured soil CO2 fluxes and four sampling ports at 9, 55, 168, and 243 cm above the ground surface that measured atmospheric CO2 concentrations (Carman et al., 2014, 2019). Comparisons of the GreenLITE data with in-situ measurements collected with the LI-COR Biosciences-based multiplexer showed that general atmospheric CO2 concentration trends were tracked extremely well (Fig. 10). The GreenLITE data typically measured concentrations that were 5–10 ppm smaller than the in situ atmospheric concentration measurements made by the IBDP multiplexer. A number of factors may have contributed to this difference: (1) the multiplexer was measuring at a single location and was unable to capture variations along the length of the chord that was captured by GreenLITE; (2) the height of the highest multiplexer measurement was 2.43 m, whereas the GreenLITE chords nearest the multiplexer ranged from approximately 2–3.5 m above the 8 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Fig. 8. Open-path sensor readings for the week of August 4, 2013: (a) F1, (b) F2, (c) detector DC voltage level, (d) path-integrated CO2 concentration (uncalibrated). Reprinted from Energy Procedia 63, Zimmerman, J.W., Locke, R.A., II, Blakley, C.S., Frish, M.B., Laderer, M.C., Wainner, R.T., Tunable diode laser absorption spectrometers for CO2 wellhead and pipeline leakage monitoring: Experiences from prototype testing at the Illinois Basin – Decatur Project, USA, p. 4083–4094. Copyright 2014, with permission from Elsevier. Fig. 9. Open-path sensor-integrated CO2 concentration monitored during routine pipeline blowdown after well maintenance on (a) September 6, 2013 CST, and (b) October 25, 2013 CST. Reprinted from Energy Procedia 63, Zimmerman, J.W., Locke, R.A., II, Blakley, C.S., Frish, M.B., Laderer, M.C., Wainner, R.T., Tunable diode laser absorption spectrometers for CO2 wellhead and pipeline leakage monitoring: Experiences from prototype testing at the Illinois Basin – Decatur Project, USA, p. 4083–4094. Copyright 2014, with permission from Elsevier. surface at the multiplexer location and varied along the length of the path; (3) the multiplexer was calibrated by the manufacturer but was not regularly calibrated with a gas standard, so absolute accuracy was unknown. Overall, the agreement was very good regarding measured trends, and an offset was not unexpected. Additional comparisons were made between the GreenLITE atmospheric CO2 concentration data and point measurements of CO2 soil flux concentrations measured by the ISGS. Spatial analysis of flux variability was performed with GIS software (Korose et al., 2014). Fig. 11 shows an example comparison of these data. It is important to note that the flux measurements were made point by point over a 2-day period, whereas the GreenLITE system generated a 2-D concentration map approximately every 10 min. Concentration maps were generated by the IBDP flux chamber monitoring grid, and an average of all 2-D reconstructions generated over the same 2-day time period during which the chamber measurements were being taken was used for comparison. Even though the two methods were not measuring the same CO2 parameters, the GreenLITE measurements showed good agreement with the flux chamber measurements when estimating the CO2 spatial distribution across the site. Note that the GreenLITE system did not have any paths directly over the injection well, which may have contributed to the absence of the larger concentrations shown in the lower right of the flux map but not in the GreenLITE reconstruction. 9 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Fig. 10. Comparison during a 3-day period of GreenLITE measured atmospheric CO2 concentrations (blue) and collocated in situ IR gas analysis measurements (green) during deployment at the IBPD. The 3-day comparison window was used as a representative subset of data for the period of deployment. (a) The solid blue line denotes the average GreenLITE concentration, whereas the dotted lines denote ±1 standard deviation from the mean of the 5-min average maximum and minimum of all GreenLITE chords. (b) Average difference between GreenLITE and in situ data. Fig. 11. Comparison of GreenLITE measured atmospheric CO2 concentrations and soil CO2 flux measurements. (a) Estimated concentration map generated from flux chamber measurements. Small black dots indicate the sampling locations. (b) Average of all GreenLITE reconstructed concentration maps generated during the fluxsampling period. response quantitatively when monitoring the site for potential CO2 leakage. More important, remote-sensing monitoring of a purposeful subsurface release of CO2 was not conducted to measure the potential impacts to the vegetative cover of the IBDP site. In addition, the high degree of industrial activity at the IBDP (Table 1) created substantial changes to the ground cover throughout the MVA field area. The lack of an established and consistent vegetated surface over much of the MVA field during the project (Fig. 7) further hindered our ability to use plant health to observe any potential CO2 leakage at the IBDP site. Therefore, CIR acquisitions were discontinued in 2012 because of the combination of increased film-processing costs, significant changes in site conditions resulting from industrial activities, changes in project personnel, and a reevaluation of remote imagery needs for the project. However, natural 3.3. Aerial imagery for IBDP site documentation The temporal series of high-resolution digitized aerial photographs collected as part of the MVA program at the IBDP was an important means of documenting previous activities at the site as well as site activities and land surface changes over the life of the project. For selected years, digitized CIR aerial photography was used by the IBDP to supplement natural-color photography. The NIR component of the photographs highlighted vegetated areas and could be used to describe qualitatively the land surface changes pertinent to the monitoring of near-surface environmental conditions at the site. However, the processed CIR film was not calibrated from one overflight to the next; thus, the digitized photographs could not be used to measure the vegetation 10 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 environmental conditions. Further work is needed to validate the precision and accuracy of the system in estimating atmospheric CO2 concentrations, but the results from the IBDP testing are promising and show that, at a minimum, GreenLITE can provide meaningful information over large areas and should be considered as a monitoring component in MVA programs for future CCS projects. Table 1 Significant surface disturbances documented at the IBDP monitoring, verification, and accounting field site.a Year Season Disturbance 2007 Spring 2010 2010 Fall Fall 2011 2013 Fall Fall 2016 2016 Spring Summer 2017 Spring Before site monitoring and CO2 injection. Circles are remnants of a center pivot irrigation system A. Northern berm construction B. Verification well (VW1) access road and well and pad construction C. Surface grading and organic materials spread on field D. Power substation expansion, drainage ditch, and southern berm extension E. Excavation for surface drainage impoundment F. Increase in land surface elevation by build-up of earthen material and site grading During post injection monitoring a 4.3. Integration of aerial imagery with GIS data The high-resolution digitized aerial photographs captured at the IBDP were georeferenced and used as an imagery layer in the project GIS database. The GIS software enabled an on-screen comparison of the multitemporal photography and allowed the IBDP site area to be visualized along with map overlays, such as the local infrastructure, project wells, field sampling locations and measured values, and other related spatial information (Figs. 3a, 4 a, ). The digitized aerial photographs of the IBDP are of sufficiently high resolution to provide feature detail for an assessment of land surface changes and supplement the additional data available for the IBDP site for the corresponding time frames. Table 1 and Fig. 12 show selected examples of observed and documented earthmoving activity, construction, and monitoring equipment installation (and loss) at the IBDP site. The natural-color aerial images were the primary GIS map base for the IBDP site and were used for a variety of purposes. Digital and paper maps of the site supported internal project communication, the scheduling of field work and site activities, and the planning and placement of new field installations. In several instances, digitized aerial photographs of the site were combined with associated GIS data and engineering CAD drawings to plan for upcoming industrial construction and the eventual loss and abandonment of selected nearsurface monitoring installations. Having the digitized aerial photographs of the IBDP was fundamental to developing a suite of maps and graphics to communicate project progress and results. These were used both internally among project staff and scientists and externally for reports to funding agencies and when providing information to the general public. The letters A–F correspond to the locations shown in Fig. 12. color overflights were continued on a biannual basis throughout the remainder of the IBDP. 4. Discussion 4.1. Open-path sensor We anticipate, based on results from the prototype testing, that the OPS design deployed at the IBDP site can be refined and used at other CCS sites as part of an integrated environmental and safety monitoring strategy. Most OPS system components proved effective for use as a warning system to ensure personnel safety. The deployment and installation of the transceiver, reflector, and solar panel mounting were straightforward. The transceiver housing kept the electronics, radio, and optics free of moisture while allowing easy access to components for inspection. Communication between the transceiver and the remotecontrol computer was easily established when the radio modems were properly configured, and data logging with the supplied software was manageable. One key issue encountered was that thermal stresses degraded optical alignment, likely because the laser launch assembly was mounted on the transceiver. Thermal expansion of the telescope body apparently caused the laser to aim off center of the target reflector, resulting in a diminished and occasionally lost return signal. This issue caused some loss of data during deployment at the IBDP and has since been corrected in a similar open-path CH4 monitor installed at a natural gas leak detection test site (Frish and Cummings, 2013). In addition, to make the OPS a more robust and accurate monitoring tool, we recommend calibrating the instrument signal (and recorded data) to relatively high-accuracy measurements made with other equipment and incorporating this step into standard procedures for future installations and operation. 5. Conclusions Three relatively new techniques or applications of established techniques were evaluated as potential tools for use in an MVA program at commercial scale CCS operations. Successful application of these techniques at the IBDP project site was demonstrated by the following factors: (1) operational status for the duration of deployment or observation, (2) application to an active industrial site (IBDP), and (3) potential application to a commercial-scale injection project. Note that a planned release of CO2 within the measurement area(s) would have benefited all the methods tested at the IBDP by allowing for the observation of site-specific responses to an actual instance of elevated CO2 detected. 4.2. GreenLITE system 5.1. Open-path sensor Based on its deployment and performance at the IBDP, GreenLITE has a high potential for value-added integration into a comprehensive CCS MVA program. This system has the unique capabilities of (1) offering operators real-time feedback on atmospheric concentrations over large areas via a web-based interface, and (2) operating autonomously. This type of real-time information is especially important for site safety and the potential detection of an atmospheric expression of CO2 leakage from the storage reservoir. During deployment at the IBDP, the system also provided operators with text message or e-mail notifications for user-defined thresholds and other relevant instrument information, which allowed system performance to be monitored remotely. Installation of the system was straightforward, and the maintenance requirements for GreenLITE after installation were approximately monthly or as needed. These tests demonstrated the ability of GreenLITE to operate in a real-world CCS environment over a wide range of Tunable diode laser absorption spectroscopy offers a flexible and effective sensing technology for measuring target gases in complex mixtures associated with evolving CCS efforts. The OPS was deployed for one year at the IBDP site, demonstrating long-term operation of a long-path sensor for pipeline and wellhead monitoring and showing responses to changes in ambient CO2 during maintenance procedures. With refinements, TDLAS systems such as these could be integrated into environmental monitoring strategies at CCS sites by following procedures like those applied with methane sensor technology. The TDLAS sensors have potential benefits for CCS operators by reducing the cost of leak surveying and verifying well and site integrity. 11 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Fig. 12. Significant surface disturbances documented at the IBDP monitoring, verification, and accounting field site. Single-letter labels correspond to the activities listed in Table 1. Aerial photography acquisition dates: (left to right, top row) March 16, 2007, and November 8, 2010; (middle row) October 21, 2011, and November 27, 2013; (bottom row) March 29, 2016, and April 18, 2017. 12 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 National Laboratory (LBNL), Ivan Krapac of the ISGS for his work with LBNL and PSI to begin this study, initial sensor deployment ideas, and paper review, and Randy Locke of the ISGS for supporting testing of these monitoring techniques and paper review. The Illinois Department of Transportation (IDOT) was a Midwest Geological Sequestration Consortium partner from September 2008, until November 2017; the IDOT Aerial Survey Division performed aerial photography overflights and film processing from the beginning of IBDP activities through the post injection phase of the project. The Midwest Geological Sequestration Consortium is funded by the U.S. Department of Energy through the National Energy Technology Laboratory (NETL) via the Regional Carbon Sequestration Partnership Program(contract number DE-FC2605NT42588) and by a cost share agreement with the Illinois Department of Commerce and Economic Opportunity, Office of Coal Development through the Illinois Clean Coal Institute. Physical Sciences Inc. work was supported by Department of Energy STTR Grant #DESC0001575, “Networked Sensors for Sequestration MVA”. 5.2. GreenLITE system The GreenLITE system offered a near real-time assessment of variable atmospheric CO2 concentrations across a 0.2 km2 project area. Over the 5 months of deployment at the IBDP site, minimal hands-on adjustments were made to the system. With modifications in the power and internal laser source components, GreenLITE would also have the potential to provide automated data acquisition that could be adjusted to measure other greenhouse gases over an extended time. Potential refinement to the deployment design in the future could enable better characterization of external sources of measured greenhouse constituents to aid in interpretation of the 2-D reconstruction images. The nearly continuous monitoring and good agreement between the spatial distribution of atmospheric CO2 concentrations measured by the GreenLITE system and the distribution of periodic soil CO2 flux measurements at the IBDP site suggest that the GreenLITE system may be a more worksparing and cost-effective method of monitoring CO2 leakage at largescale CCS sites than the soil CO2 flux monitoring technique. Appendix A. Supplementary data 5.3. Aerial imagery Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.ijggc.2020.103188. Monitoring the plant response to potential environmental changes (e.g., increased CO2 at the ground surface) would require, at a minimum, digitally acquired CIR imagery, spectral calibration for a consistent measurement reference over time, and a flight or acquisition revisit frequency close enough to detect any potential plant stress events. Industrial activities unrelated to a CCS project have the potential to cause major land disturbances, and typical agricultural practices in the Midwest of the United States, such field cultivation and crop rotation, result in land changes that can complicate the large-scale interpretation of CIR imagery to detect potential CO2 leakage, should it occur at an industrialscale CCS project. Although CIR-based aerial photography could not be applied for long-term injection monitoring at the IBDP, the acquisition of aerial photography was invaluable for MVA environmental data analysis and validation and for documenting project activities. References Carman, C.H., Locke, R.A., Blakley, C.S., 2014. Update on soil CO2 flux monitoring at the Illinois Basin – decatur Project. Energy Procedia 63, 3869–3880. Carman, C.H., Krapac, I.G., Blakley, C.S., Korose, C.P., Locke II, R.A., 2019. Illinois Basin – Decatur Project: Soil Carbon Dioxide Flux Monitoring. Circular 599. Illinois State Geological Survey, Champaign. In press. Clough, S.A., Shephard, M.W., Mlawer, E.J., Delamere, J.S., Iacono, M.J., CadyPereira, K., Boukabara, S., Brown, P.D., 2005. Atmospheric radiative transfer modeling: a summary of the AER codes. J. Quant. Spectrosc. Radiat. Transf. 91, 233–244. Devi, V.M., Benner, D.C., Brown, L.R., Miller, C.E., Toth, R.A., 2007a. Line mixing and speed dependence in CO2 at 6227.9 cm–1: constrained multispectrum analysis of intensities and line shapes in the 30013 ⟵ 00001 band. J. Mol. Spectrosc. 245, 52–80. Devi, V.M., Benner, D.C., Brown, L.R., Miller, C.E., Toth, R.A., 2007b. Line mixing and speed dependence in CO2 at 6348 cm–1: Positions, intensities, and air- and selfbroadening derived with constrained multispectrum analysis. J. Mol. Spectrosc. 242 (2), 90–117. Dobbs, M.E., Sharp, W.E., Browell, E.V., Zaccheo, T.S., Moore III, B., 2007. A sinusoidal modulated CW integrated path differential absorption lidar for mapping sources and sinks of carbon dioxide from space. In: 14th Coherent Laser Radar Conference (CLRC XIV). Snowmass, Colorado. Curran, Red Hook, NY. Dobbs, M.E., Dobler, J.T., Braun, M.G., McGregor, D.P., Overbeck, J.A., Moore III, B., Browell, E.V., Zaccheo, T.S., 2008. A modulated CW fiber laser-lidar suite for the ASCENDS mission. In: Proceedings of the 24th International Laser Radar Conference (ILRC 24). Boulder, Colorado. Curran, Red Hook, NY. Dobler, J.T., 2013. A novel approach for active measurement of atmospheric greenhouse gases from a geostationary orbit. Presented at 93rd American Meteorological Society Annual Meeting, Austin, TX. Dobler, J.T., Harrison, F.W., Browell, E.V., Lin, B., McGregor, D.P., Kooi, S., 2013. Atmospheric CO2 column measurements with an airborne intensity-modulated continuous wave 1.57 μm fiber laser lidar. Appl. Opt. 52 (12), 2874–2892. Dobler, J.T., Zaccheo, T.S., Blume, N., Braun, M., Botos, C., Pernini, T.G., 2015. Spatial mapping of greenhouse gases using laser absorption spectrometers at local scales of interest. In: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing XI: Proceedings of SPIE, v. 9645, Singh, U.N., and Nicolae, D.N. (Eds.). SPIE. Bellingham, WA. https://doi.org/10.1117/12.2197713. Dobler, J.T., Zaccheo, T.S., Pernini, T.G., Blume, N., Broquet, G., Vogel, F., Ramonet, M., Braun, M., Staufer, J., Ciais, P., Botos, C., 2017. Demonstration of spatial greenhouse gas mapping using laser absorption spectrometers on local scales. J. Appl. Remote Sens. 11. https://doi.org/10.1117/1.JRS.11.014002. Frailey, S.M., Krapac, I.G., Damico, J.R., Okwen, R.T., McKaskle, R.W., 2012a. CO2 Storage and Enhanced Oil Recovery: Bald Unit Test Site, Mumford Hills Oil Field, Posey County, Indiana. Open File Series 2012-2015, Illinois State Geological Survey, Champaign. Frailey, S.M., Morse, D., Krapac, I.G., McKaskle, R.W., 2012b. Sequestration and Enhanced Coal Bed Methane: Tanquary Farms Test Site, Wabash County, Illinois. Open File Series 2012-2, Illinois State Geological Survey, Champaign. Frailey, S.M., Parris, T.M., Damico, J.R., Okwen, R.T., McKaskle, R.W., 2012c. CO2 Storage and Enhanced Oil Recovery: Sugar Creek Oil Field Test Site, Hopkins County, Kentucky. Open File Series 2012-2014, Illinois State Geological Survey. Champaign. Frish, M., 2017, “Systems and Methods for Sensitive Open-Path Gas Leak and Detection Alarm,” US Patent No. 9,797,798 B2. CRediT authorship contribution statement Curt Blakley: Conceptualization, Methodology, Investigation, Writing - original draft, Writing - review & editing, Supervision, Project administration. Carl Carman: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing - review & editing. Chris Korose: Data curation, Visualization, Writing - review & editing. Don Luman: Resources, Formal analysis, Investigation, Writing - original draft. Joseph Zimmerman: Investigation, Data curation, Methodology, Formal analysis, Writing - original draft, Writing - review & editing. Michael Frish: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Project administration, Resources, Software, Supervision, Visualization, Writing - review & editing. Jeremy Dobler: Investigation, Resources, Writing - original draft, Writing - review & editing, Supervision, Project administration, Funding acquisition. Nathan Blume: Investigation, Writing - original draft, Writing - review & editing. Scott Zaccheo: Conceptualization, Formal analysis, Methodology, Software, Writing - original draft, Writing - review & editing. Declaration of Competing Interest The authors report no declarations of interest. Acknowledgments The authors would like to thank Billy C. Bruns and Stephen M. Picek of the ISGS for providing support for field deployment, testing, and maintenance of the prototype equipment, Jen Lewicki for her early contributions to this project as a researcher at the Lawrence Berkeley 13 C. Blakley et al. International Journal of Greenhouse Gas Control 103 (2020) 103188 Frish, M., Cummings, B., 2013. Laser sensors for gas pipeline explosion protection. PSI1802/TR-2926, Final Report for the California energy Commission Under Grant no. 57238A/11-01G. Physical Sciences Inc., Andover, MA. Frish, M., Wainner, R., Stafford-Evans, J., Green, B., Allen, M., Chancey, S., Rutherford, J., Midgley, G., Wehnert, P., 2005. Standoff sensing of natural gas leaks: evolution of the remote methane leak detector (RMLD). In: Conference on Lasers and Electro-optics/Quantum Electronics and Laser Science and Photonic Applications. Systems and Technologies. https://doi.org/10.1117/12.884526. Frish, M., Wainner, R., Laderer, M., Parameswaran, K., Sonnerfroh, D., Druy, M., 2011. Precision and accuracy of miniature tunable diode laser absorption spectrometers. In: Druy, M.A., Crocombe, R.A. (Eds.), Next-Generation Spectroscopic Technologies IV: Proceedings of SPIE, v. 8032. SPIE, Bellingham, WA no. 803209-9. Frish, M., Wehnert, P., Rutherford, J., Rongere, F., 2014. “Continual 24/7 Pipeline Monitoring System for Leakage,” Presentation to American Gas Association Operations Conference, Pittsburgh, PA. Giuli, D., Toccafondi, A., Biffi Gentili, G., Freni, A., 1991. Tomographic reconstruction of rainfall fields through microwave attenuation measurements. J. Appl. Meteorol. Climatol. 30, 1323–1340. Giuli, D., Facheris, L., Tanelli, S., 1999. Microwave tomographic inversion technique based on a stochastic approach for rainfall fields monitoring. IEEE Trans. Geosci. Remote Sens. 37, 2536–2555. Jones, A., Scullion, J., Ostle, N., Levy, P., Gwynn-Jones, D., 2014. Completing the FACE of elevated CO2 research. Environ. Int. 73C, 252–258. https://doi.org/10.1016/j. envint.2014.07.021. Korose, C.P., Locke II, R.A., Blakley, C.S., Carman, C.H., 2014. Integration of near-surface monitoring information using ArcGIS at the Illinois Basin – Decatur Project, USA. Energy Procedia 63, 3945–3955. Leica Geosystems, 2016. Leica DMC III Airborne Digital Camera (accessed 22 June 2016). http://leica-geosystems.com/products/airborne-systems/imaging-sensors/lei ca-dmciii. Locke II, R.A., Krapac, I.G., Lewicki, J.L., Curtis-Robinson, E., 2011. Characterizing nearsurface CO2 conditions before injection—perspectives from a CCS project in the Illinois Basin, USA. Energy Procedia 4, 3306–3313. Locke II, R.A., Larssen, D., Salden, W., Patterson, C., Kirksey, J., Iranmanesh, A., Wimmer, B.T., Krapac, I.G., 2013. Pre-injection reservoir fluid characterization at a CCS demonstration site: Illinois Basin – Decatur Project, USA. Energy Procedia 37, 6424–6433. Measures, R.M., 1984. Laser Remote Sensing: Fundamentals and Applications. Wiley, New York. Midwest Geological Sequestration Consortium (MGSC), 2009. Enhanced Oil Recovery I: Loudon Single-well Huff’ n’ Puff. U.S. Department of Energy Contract No. DE-FC2603NT41994. Illinois State Geological Survey, Champaign. Miller, C., Brown, L., 2004. Near Infrared spectroscopy of carbon dioxide I. 16O12C16O line positions. J. Mol. Spectrosc. 228, 329–354. Terashima, I., Yanagisawa, S., Sakakibara, H., 2014. Plant responses to CO2: background and perspectives. Plant Cell Physiol. 55, 237–240. https://doi.org/10.1093/pcp/ pcu022. Zimmerman, J., Locke II, R., Blakley, C., Frish, M., Laderer, M., Wainner, R., 2014a. Initial testing of prototype tunable diode laser absorption spectrometers for CO2monitoring applications at the Illinois Basin – Decatur Project, USA. In: Presented at 13th Annual Carbon Capture, Utilization and Storage Conference. Pittsburgh, PA. Zimmerman, J., Locke II, R., Blakley, C., Frish, M., Laderer, M., Wainner, R., 2014b. Tunable diode laser absorption spectrometers for CO2wellhead and pipeline leakage monitoring: Experiences from prototype testing at the Illinois Basin – Decatur Project, USA. Energy Procedia 63, 4083–4094. 14