COMPARISON OF ESTIMATION METHODS TO QUANTIFY METHANE PLUME CONCENTRATION AT HIGH SPATIAL RESOLUTION FROM HYPERSPECTRAL IMAGES

The paper discusses two very distinct methane products from the IASI instrument aboard the MetOp-A satellite. One (referred to as LMD NLISv8.3) uses a machine-learning approach, while the other (RALv2.0) uses a more conventional optimal estimation approach. We used a variety of model and independent reference measurement data to assess both products' overall quality, their differences, and specific aspects of each product that would benefit from further analysis by the product development teams.

We propose analyzing the aerosol composition of plumes emitted by different industrial stacks using PRISMA satellite hyperspectral observations. Three industrial sites have been observed: a coal-fired power plant in South Africa, a steel plant in China, and gas flaring at an oil extraction site in Algeria. Aerosol optical thickness and particle radius are retrieved within the plumes. The mass flow rate of particulate matter is estimated in the plume using the integrated mass enhancement method.

With its near-real-time observations and good horizontal coverage, the Infrared Atmospheric Sounding Interferometer (IASI) instrument can contribute to the monitoring systems for a systematic and continuous detection of exceptional atmospheric events such as fires, anthropogenic pollution episodes, volcanic eruptions, or industrial releases. In this paper, a new approach is described for the detection and characterization of unexpected events in terms of trace gases using IASI radiance spectra.