Improving Classical Approach for Flare Parameters Estimation from VIIRS NtL Remote Sensing Data by Linear and Nonlinear Spectral Unmixing Methods

During oil extraction process, natural gases escape from wells, and the procedure for recovering these gases requires heavy investments from oil and gas companies. That is why, most often, they prefer to burn them with flares. This practice, which is very common by oil and gas companies in oil-producing countries, is highly emitting greenhouse gases. Under increasing pressure from the World Bank and environmental defenders, several producer countries are committed to reducing gas flaring. To this end, researchers in the oil and gas industry, academia and governments are working to develop approaches for determining ways to measure gas flaring and its emissions. Among the most widely used approaches, for local and global monitoring of gas flares, are those that exploit remote sensing data, particularly Night time Light (NtL) ones. Indeed, it is possible to extract, from such data, some physical parameters of flames produced by gas flares, and these parameters can be used to estimate annual volumes of flared gas. In this investigation, three spectral analysis-based approaches are tested to estimate flare physical parameters from Visible Infrared Imaging Radiometer Suite (VIIRS) NtL data. Linear and nonlinear spectral unmixing methods applied on NtL VIIRS data, are anticipated to improve, by using a pure spectrum of flames, results provided by the classical one that considers the modeling of the Planck law curve applied on the manipulated observed mixed data. Experiments, based on realistic synthetic VIIRS NtL data, are conducted and obtained results confirm the expected improvements.