Probability of detection for airborne sensing of methane plumes using long wavelength infrared hyperspectral imaging

Hyperspectral imaging over the long wavelength infrared (LWIR) or “thermal” spectrum is a promising means to both detect and quantify methane (CH 4 ) emissions, particularly those arising from upstream oil and gas activities. In contrast to systems that interrogate the 1.6 or 2.2 μm CH 4 absorption bands using an external light source, such as a laser or sunlight, passive systems rely on thermal emission of the CH 4 plume and the background surface over the 7.78 μm band. This approach offers potential advantages over systems based on the short wavelength infrared (SWIR) band, particularly in scenarios involving low solar intensity or over highly absorbing surfaces. This paper presents the outcomes of three measurement campaigns conducted under conditions relevant to the Canadian oil and gas industry, including probability of detection (POD) curves that may be used to identify the likelihood of identifying an emission of a given rate under prescribed environmental conditions. POD curves are then used to generate a conservative estimate of the availability of the technique at one location in southern Alberta, Canada. The results highlight that the thermal contrast between the plume and the background surface is crucial to the viability of this technique. To the authors' knowledge, this work presents the first POD curves for an airborne passive LWIR hyperspectral methane detection system and provides a quantitative comparison of detection performance over bare ground and snow-covered surfaces.