Airborne Remote-Sensing Technologies Detect, Quantify Hydrocarbon Releases
Airborne imaging spectroscopy has evolved dramatically since the 1980s as a robust remote-sensing technique used to generate 2D maps of surface properties over large areas. Two recent applications are particularly relevant to the needs of the oil and gas sector and government: quantification of surficial hydrocarbon thickness in aquatic environments and mapping atmospheric greenhouse-gas components. These techniques provide valuable capabilities for monitoring petroleum seepage and for detection and quantification of fugitive emissions.
The Jet Propulsion Laboratory (JPL), a National Aeronautics and Space Administration (NASA) federally funded research-and-development center operated by the California Institute of Technology, has been a pioneer in optical remote sensing since the 1980s. JPL capabilities include expertise across all project phases, including sensor design and construction, airborne experiment execution, and data generation driven by science and customer needs. JPL has particular expertise in imaging spectroscopy, a passive method to interrogate objects or surfaces without physical contact. Such remote sensing has traditionally been applied to investigation of surface composition in terrestrial environments. These surface compositions are characterized by use of a spectral library that includes the surface-reflectance or emissivity fingerprints of constituent materials. Airborne imaging spectrometers provide a powerful method to survey wide spatial extents with high-performance surface characterization because of the wide contiguous spectral range at moderate spectral resolution. Novel quantitative methods have emerged recently for both atmospheric gases and surficial oil on water.
Airborne pushbroom imaging spectrometers incorporate a 2D focal-plane array to collect data over a wide swath beneath the aircraft by use of a nadir-mounted sensor (Fig. 1). The areal coverage and spatial resolution depend on the sensor design characteristics and altitude. The crosstrack sensor characteristics include the sensor field of view (FOV), which determines the swath coverage as a function of altitude, while instantaneous FOV (IFOV) defines the across-track resolution, or pixel size as projected on the ground. On the basis of these sensor characteristics, a simple geometric relationship links sensor characteristics to crosstrack performance parameters.
Contemporary JPL pushbroom airborne imaging spectrometers include two major types: Offner and Dyson spectrometers. Offner spectrometers operate by collecting light through a narrow optical slit and, by use of a dispersive grating and multiple mirrors, focusing light onto the focal-plane array (FPA) with high spectral uniformity. Thus, during flight, pushbroom sensors simultaneously image pixels beneath the aircraft across the entire sensor swath width. The FPA images discrete spectral channels across the entire contiguous spectral range while crosstrack spatial information is captured across the second axis. Pushbroom approaches eliminate any moving optical subsystems by implementing a fixed optical train. In order to optimize sensor performance with respect to the signal/noise ratio, it helps to fly slowly with these systems (80–100 knots) to enhance oversampling. The second type of spectrometer in which JPL specializes is the Dyson spectrometer. The main difference in Dyson-spectrometer designs compared with Offner types is that the dispersion is accomplished by an arsenic-doped silicon block. These Dyson designs often result in a smaller form factor, particularly in the thermal infrared region of the spectrum, while still maintaining excellent spectral uniformity.
Application 1—Imaging-Spectrometer Applications for Investigation of Oil on Water
The Deepwater Horizon oil spill began on 20 April 2010. One of the NASA remote-sensing instruments was deployed less than a month later: the airborne visible/infrared imaging spectrometer (AVIRIS). The surveys were conducted from high altitudes (approximately 20 km) to maximize spatial coverage (i.e., 12.2-km swath width).
The results from these experiments revealed the suitability of optical remote sensing for oil-slick assessment in the visible (0.4–0.7 μm), near-infrared (1.2–1.7 μm), and shortwave infrared regions (2.3 μm). It was demonstrated through correlation with laboratory measurements that the depth of the 1.2-μm hydrocarbon absorption feature provided quantitative oil-thickness information.
The collection of these Deepwater Horizon data was the first time that optical imaging spectrometry demonstrated quantitative capability for oil-slick-thickness determination. Thus, the suitability of this technique for disaster response and estimates of net surface oil has been recognized.
Application 2—Imaging-Spectrometer Applications for Remote Sensing of Atmospheric Methane
Contemporary demonstrations of advanced NASA airborne imaging spectrometers for detection of fugitive methane emissions yield impressive results. These imaging techniques use sensors with wide spectral ranges in the visible to shortwave infrared (VSWIR) or the long wave infrared (LWIR). The NASA sensors offer much greater signal/noise ratios and greater spectral resolution than the few imaging spectrometers available commercially. Thus, these JPL applications reap the benefits of the most advanced imaging spectrometers in the VSWIR and LWIR regions that have been built. JPL and colleagues have begun flights over conventional oil fields and unconventional production areas to help constrain natural and anthropogenic methane emissions, including quantification of fugitive-emission sources by use of highly mature algorithms. These airborne spectrometers have demonstrated sensitivities at flux rates as low as <250 scf/hr when flown at low altitudes (approximately 1000 m) using VSWIR or LWIR sensors. These results were demonstrated with existing NASA spectrometers that were not designed specifically for methane detection.
Imaging spectrometers provide a unique solution for noninvasive investigation of large areas. The feasible spatial coverage for a daily survey at low altitude is on the order of hundreds of square kilometers (flight-plan dependent) while flying at relatively low altitudes (1–3 km).
One need that has resulted in wide adoption of imaging spectroscopy is that production of data products is typically labor intensive, resulting in significant delay in results because of the vast amount of data generated by these imaging spectrometers. One solution is to implement real-time algorithms as part of an onboard flight data system. A real-time detection system for methane point-source visualization currently exists as part of the AVIRIS onboard data system. This successful implementation results in real-time data analysis during collection and allows for an adaptive flight planning approach using the heads-up display.
Imaging Spectroscopy in the Shortwave Infrared (SWIR) Using AVIRIS
The JPL next-generation AVIRIS is a passive imaging spectrometer that operates by collecting the upwelling (reflected) solar radiation in discrete bands across the range of the visible (0.4 μm) through the shortwave infrared (2.5 μm). Using this technique, characteristics of surface features can be diagnosed by use of the detected spectral signatures or fingerprints. AVIRIS provides high spectral resolution for a visible/infrared imaging spectrometer (5-nm bandwidth), exceeding those of other flight systems by at least a factor of two. Increased spectral resolution allows for more-detailed discrimination between surface features.
In September 2014, six AVIRIS scenes were acquired over Garfield County, Colorado, a region with considerable gas and oil extraction. Flights were made approximately 1.4 km above ground level, which resulted in images approximately 0.8 km wide and 8 km long, with a ground resolution of 1.3 m per pixel. Quantitative methane retrievals were performed on all images, and a number of plumes were clearly visible emanating from multiple well pads.
Fig. 2 clearly indicates a plume consistent with the local wind direction (white arrow) that extends 200 m downwind of the emission source. Google Earth imagery obtained from June 2014 indicates that the likely source is tanks located on the edge of the well pad. Five wells are located at the center of this well pad, and all use horizontal drilling to produce mostly gas.
Conclusions and Path Forward
The results demonstrate the utility of existing advanced NASA imaging spectrometers for detection of oil on water and quantitative mapping of methane plumes. While existing data sets for both applications are currently quite small, future opportunities to demonstrate these capabilities further are a high priority for the program.
The optimal solution for wide adoption of methane monitoring is to build an imaging spectrometer sensor fit for purpose. None of the technologies used was designed specifically for quantitative methane detection; however, sensitivities in the range of 250 scf/hr remain impressive. A new sensor would improve the achievable sensitivity (<10 scf/hr) and increase specificity for small point-source emissions sources. This is the optimal solution from a science perspective to help understand the spatio-temporal variability of natural and anthropogenic methane emissions. The major improvements of this spectrometer design include a narrower spectral range with enhanced spectral resolution. These factors will increase the sensitivity, specificity, and spatial resolution, while virtually eliminating any false positives. This sensor has been designed to be accommodated on a fixed-wing aircraft or helicopter for more-flexible flight implementation.
This article, written by Special Publications Editor Adam Wilson, contains highlights of paper OTC 25984, “Crosscutting Airborne Remote-Sensing Technologies for Oil and Gas and Earth Science Applications,” by A.D. Aubrey, C. Frankenberg, R.O. Green, M.L. Eastwood, and D.R. Thompson, National Aeronautics and Space Administration Jet Propulsion Laboratory, California Institute of Technology, and A.K. Thorpe, University of California, Santa Barbara, prepared for the 2015 Offshore Technology Conference, Houston, 4–7 May. The paper has not been peer reviewed. Copyright 2015 Offshore Technology Conference. Reproduced by permission.