Atmos. Meas. Tech., 8, 4383–4397, 2015 www.atmos-meas-tech.net/8/4383/2015/ doi:10.5194/amt-8-4383-2015 © Author(s) 2015. CC Attribution 3.0 License. Real-time remote detection and measurement for airborne imaging spectroscopy: a case study with methane D. R. Thompson1, I. Leifer2, H. Bovensmann3, M. Eastwood1, M. Fladeland4, C. Frankenberg1, K. Gerilowski3, R. O. Green1, S. Kratwurst3, T. Krings3, B. Luna4, and A. K. Thorpe1 1Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA 2Bubbleology Research International, Solvang, CA, USA 3University of Bremen, Institute of Environmental Physics, P.O. Box 330440, 28334 Bremen, Germany. 4NASA Ames Research Center, Moffett Field, CA, USA Correspondence to: D. R. Thompson ([email protected]) Received: 15 March 2015 – Published in Atmos. Meas. Tech. Discuss.: 22 June 2015 Revised: 10 September 2015 – Accepted: 10 September 2015 – Published: 19 October 2015 Abstract. Localized anthropogenic sources of atmospheric 1 Introduction CH4 are highly uncertain and temporally variable. Airborne remote measurement is an effective method to detect and Airborne imaging spectrometers have been deployed for a quantify these emissions. In a campaign context, the science wide range of scientific, regulatory, and disaster response yield can be dramatically increased by real-time retrievals objectives. Traditionally these campaigns wait for favorable that allow operators to coordinate multiple measurements of environmental conditions and then fly pre-arranged survey the most active areas. This can improve science outcomes patterns (typically “mowing the lawn”), recording data for for both single- and multiple-platform missions. We describe post-flight radiometric calibration and geolocation. Signif- a case study of the NASA/ESA CO2 and MEthane eXperi- icant time can pass before data are analyzed fully, and re- ment (COMEX) campaign in California during June and Au- sults often arrive too late for mid-course corrections during gust/September 2014. COMEX was a multi-platform cam- the campaign. However, improvements in computing power, paign to measure CH4 plumes released from anthropogenic communication, and telemetry are changing this situation. sources including oil and gas infrastructure. We discuss prin- Tactical remote measurement generates in-flight calibrated ciples for real-time spectral signature detection and mea- data products to inform a real-time adaptive survey strategy. surement, and report performance on the NASA Next Gen- This can be coordinated to direct other platforms in multi- eration Airborne Visible Infrared Spectrometer (AVIRIS- platform campaigns. We use the term “tactical” to empha- NG). AVIRIS-NG successfully detected CH4 plumes in real- size environmental awareness and real-time decision making, 1 time at Gb s− data rates, characterizing fugitive releases with no military connotation. Its applications include the fol- in concert with other in situ and remote instruments. The lowing. teams used these real-time CH4 detections to coordinate measurements across multiple platforms, including airborne i. Detection of transient or rare targets – Many airborne in situ, airborne non-imaging remote sensing, and ground- missions hunt isolated or nonstationary phenomena. based in situ instruments. To our knowledge this is the first Examples include trace-gas emissions (Aubrey et al., reported use of real-time trace-gas signature detection in an 2015; Gerilowski et al., 2015), algal blooms (Karaska airborne science campaign, and presages many future ap- et al., 2004), invasive species (Ustin et al., 2002), iso- plications. Post-analysis demonstrates matched filter meth- lated microhabitats (Thompson et al., 2013b), and hur- ods providing noise-equivalent (1σ ) detection sensitivity for ricane intensity (Braun et al., 2013). Aircraft use radar 1.0 % CH4 column enhancements equal to 141 ppm m. to hunt extreme weather, and lidar to find cirrus, thun- derstorms or biomass burning (Rolph, 2003). In each case, tactical remote measurement can identify desired Published by Copernicus Publications on behalf of the European Geosciences Union. 4384 D. R. Thompson et al.: Real-time remote detection and measurement features (and equally importantly, their absence) dur- Google Earth (Google Earth, 2015), is common in sur- ing flight, permitting flight plan adjustments to improve face and airborne in situ applications. Many systems coverage (Davis et al., 2010). This reveals features’ allow the scientist to visualize spatial relationships be- temporal evolution and improves measurement confi- tween measured parameters, forming hypotheses on the dence. During multi-platform campaigns, real-time en- fly for immediate testing. Adaptive surveying can ad- vironmental awareness can guide teams acquiring com- dress new hypotheses during the campaign, while in- plementary in situ measurements. struments are deployed and environmental conditions are favorable. Telemetering live data allows remote in- ii. Disaster response – Remote measurements play a criti- vestigators to observe and participate in operational de- cal role in disaster response to oil spills (Leifer et al., cisions (Leifer et al., 2014). 2012; Clark et al., 2010; Reuter et al., 1995), search and rescue (Eismann et al., 2009), fires (Ambrosia et al., These techniques require high-performance data telemetry 2003, 2011; Mandl et al., 2008; Dennison and Roberts, and communication. As the technologies proliferate, unantic- 2009), and earthquakes (Kruse et al., 2014). In any dis- ipated applications are likely to appear – just as instant results aster, information arrives at the incident command cen- from the digital CCD transformed chemical photography in ter from a range of sources of differing reliability. Re- dramatic and unforeseen ways. mote measurements can contribute repeatable and ob- This study demonstrates tactical remote measurement with jective analysis, allowing more efficient, confident allo- imaging spectroscopy during a multi-aircraft, multi-platform cation of ground and airborne assets while keeping re- campaign, CO2 and MEthane eXperiment (COMEX). The sponders safe. The immediate risks to human life de- COMEX campaign was funded by NASA and ESA to ex- mand short response times, for which tactical measure- plore synergies between NASA’s proposed HyspIRI (Hy- ment can provide situational awareness. perspectral Infrared Imager) mission and ESA’s CarbonSat Earth Explorer 8 candidate mission. Greenhouse gas emis- iii. Data quality assurance – Tactical remote measurement sions were measured from a range of important anthro- adds flexibility and confidence to flight management de- pogenic sources. Investigators surveyed landfills, husbandry, cisions. Currently, mid-campaign flight planning often and fossil-fuel production sites in southern California during occurs without knowing the quality of data already col- summer and fall, 2014. A multi-scale experimental design lected. This risks wasting resources if, for example, the combined airborne and surface measurements to character- mission continues under marginal environmental con- ize CH sources on scales of meters to tens of kilometers. ditions. On the other hand, conservative planning can 4 Ground-validated airborne imaging spectroscopy identified miss opportunities. Tactical science products can inform sources and their heterogeneity. This was followed by down- flight plans and mid-day scrub decisions to avoid spend- wind surface surveys together with airborne sounding and ing flight hours on low-value or redundant data. For ex- in situ observations transecting plumes at different upwind ample, it may reveal interference such as cirrus clouds and downwind distances. Surface mobile survey teams car- (Gao et al., 1993a), sun glint (Kay et al., 2009), and ried sensors to specific locations of interest. Finally, repeated unacceptable aerosol scattering (Bojinski et al., 2002). surface in situ surveys studied longer term temporal variabil- This also allows instrument subsystem failures to be ity and larger spatial context. recognized and addressed immediately. COMEX exploited tactical remote measurements from iv. Robotic exploration – Real-time analysis can improve multiple platforms. We focus on one participating instru- autonomous operations when communication opportu- ment, the Airborne Visible Infrared Spectrometer - Next nities are rare and bandwidth is limited, such as in space Generation (AVIRIS-NG) (Hamlin et al., 2011; Green et al., exploration. Remote spacecraft that are out of touch 1998), which mapped CH4 enhancements in real time. A with ground control can autonomously detect high- simple detection method based on a band ratio (BR) was suf- value spectral signatures that guide prioritized downlink ficient to detect several sources and enhance the COMEX or trigger additional measurements (Thompson et al., campaign. These initial results motivated the development of 2013a). Operators can generate compact map products a more sophisticated matched filter detection approach, de- onboard the spacecraft and downlink them to supple- scribed in this paper, which was developed after COMEX ment raw spectra, expanding spatial coverage at a low and has been adopted by subsequent CH4 monitoring cam- bandwidth cost. Onboard cloud screening is one exam- paigns. Although prior studies quantified CH4 anomalies us- ple of data volume reduction; it can improve yields by ing Visible Shortwave Infrared (VSWIR) imaging spectrom- a factor of 2 or more for Earth orbiting instruments eters