Ground-based remote sensing of methane - estimating emissions on facility and regional scales

Kurzfassung

Atmospheric methane (CH4) causes the second largest radiative forcing of the long living greenhouse gases. The methane concentration in the Earth's atmosphere increased by a factor of roughly 2.5 since 1750. The quantication of methane sources is crucial to understand the underlying carbon cycle and hence, the impact of anthropogenic emissions on the global changing climate. Methane emissions from coal production are one of the main sources of anthropogenic CH4 in the atmosphere. Poland is the largest hard coal producer in the European Union with the Polish area of the Upper Silesian Coal Basin (USCB) as the main part of it. During the coal mining process, methane is emitted from the coal bed and vented through exhaust shafts to keep the mine safe for workers. Different inventories estimate the emission of the USCB between 344 kt a-1 (EUROSTAT, 2020b) and 720 kt a-1 (Janssens-Maenhout et al., 2017). However, recent studies (Luther et al., 2019; Kostinek et al., 2020; Fiehn et al., 2020) show a general agreement with the E-PRTR inventory, which suggests 466 kt a-1 (E-PRTR 2014) for the USCB. During the Carbon dioxide and Methane Mission 2018 (CoMet), five portable, ground-based, direct sun-viewing Fourier transform infrared spectrometers (FTS) are deployed in the USCB. One instrument is mounted on a truck to perform stop-and-go measurements downwind of single facilities by crossing the emitted methane plumes in 1 to 10 km distance. With a mass balance approach making use of wind information from three co-deployed 3D wind lidars, the emissions of the coal mine ventilation shafts are estimated ranging from 6 ± 1 kt a-1 for a single shaft to 109±33 kt a-1 for a small group of shafts. Wind-related relative errors on the emission estimates typically amount to 20% for the mobile instrument approach. The other four FTS are deployed in the four cardinal directions around the USCB in approx. 50 km distance to the center of the basin. The upwind instrument measures the background methane information from which the downwind observations are deducted to receive regional methane enhancements. WRF (Weather Research and Forecast) model runs with assimilated 3D wind lidar data feed a Lagrangian particle dispersion model (FLEXPART) to simulate the methane distribution. The residuals between simulated and measured enhancements are minimized with a Phillips-Tikhonov regularized, non-negative least squares approach using the E-PRTR inventory data as a-priori information. The regularization parameters are graphically chosen via L-curve determination. Atmospheric variability is expressed through an ensemble of different model runs, each with altered, basic meteorological parameters. One of six case studies agree with the E-PRTR estimates. The other five case studies suggest 1.4 to 3 times higher emissions than reported by the E-PRTR. The errors introduced by the model ensemble range between 10% and 32%. The functional principle of the mobile mass balance method and the model approach based on stationary network observations could thus be demonstrated. With general errors amounting to 20%, the two methods may be applied to verify instantaneous emissions on facility scale as well as on regional scale.