|Paper title||Review of point-source methane detection and quantification methods using hyperspectral data|
|Form of presentation||Poster|
Methane is the most important anthropogenic greenhouse gas after carbon dioxide. In fact, it is responsable for about one quarter of the climate warming experienced since preindustrial times. A considerable amount of these emissions comes from methane point-sources, typically linked to fuel production installations. Thus, detection and elimination of these emissions represents a key means to reduce the concentration of greenhouse gases in the atmosphere.
A functional global monitoring of methane emissions is possible because of satellites, which capture the upwelling radiance at the top-of-atmosphere level in different spectral bands. One example of this technology is the Sentinel-5P TROPOMI mission which monitors methane emissions at a global scale and daily revisit. However, its relatively low spatial resolution cannot pinpoint methane point-source emissions with high accuracy. In contrast, the Italian PRISMA mission presents a lower temporal revisit but a larger spatial resolution of 30 m and measures the top-of-atmosphere radiance in the 400–2500 nm spectral range, where significant methane absorption features are well characterized. Therefore, PRISMA mission can largely complement the capabilities of TROPOMI for the detection and quantification of methane at a global scale.
In this study, different methodologies for point-source methane retrieval detection and quantification using PRISMA data have been reviewed in order to determine the most accurate procedure. The review goes from multitemporal methods that compare data from different days with methane emission to days with no emission to target detection algorithms such as the simple matched-filter based algorithm applied to the ~2300 nm methane absorption window in shortwave infrared spectral region. The accuracy of the different methodologies has been assessed under different scenarios that consider the most relevant error sources in the retrieval such as the surface brightness and homogeneity. This assessment has flagged the main areas of potential improvement of the retrieval methodologies and, consequently, several techniques have been developed that include the detection of false positives (e.g. the identification of plastic and hydrocarbons) and the minimisation of the surface heterogeneity impact.