Mobile atmospheric measurements and local-scale inverse estimation of the location and rates of brief CH<sub>4</sub> and CO<sub>2</sub> releases from point sources

Abstract

Abstract. We present a local-scale atmospheric inversion framework to estimate the location and rate of methane (CH4) and carbon dioxide (CO2) releases from point sources. It relies on mobile near-ground atmospheric CH4 and CO2 mole fraction measurements across the corresponding atmospheric plumes downwind of these sources, on high-frequency meteorological measurements, and on a Gaussian plume dispersion model. The framework exploits the scatter of the positions of the individual plume cross sections, the integrals of the gas mole fractions above the background within these plume cross sections, and the variations of these integrals from one cross section to the other to infer the position and rate of the releases. It has been developed and applied to provide estimates of brief controlled CH4 and CO2 point source releases during a 1-week campaign in October 2018 at the TOTAL experimental platform TADI in Lacq, France. These releases typically lasted 4 to 8 min and covered a wide range of rates (0.3 to 200 g CH4/s and 0.2 to 150 g CO2/s) to test the capability of atmospheric monitoring systems to react fast to emergency situations in industrial facilities. It also allowed testing of their capability to provide precise emission estimates for the application of climate change mitigation strategies. However, the low and highly varying wind conditions during the releases added difficulties to the challenge of characterizing the atmospheric transport over the very short duration of the releases. We present our series of CH4 and CO2 mole fraction measurements using instruments on board a car that drove along roads ∼50 to 150 m downwind of the 40 m × 60 m area for controlled releases along with the estimates of the release locations and rates. The comparisons of these results to the actual position and rate of the controlled releases indicate ∼10 %–40 % average errors (depending on the inversion configuration or on the series of tests) in the estimates of the release rates and ∼30–40 m errors in the estimates of the release locations. These results are shown to be promising, especially since better results could be expected for longer releases and under meteorological conditions more favorable to local-scale dispersion modeling. However, the analysis also highlights the need for methodological improvements to increase the skill for estimating the source locations.

Near‐field atmospheric inversions for the localization and quantification of controlled methane releases using stationary and mobile measurements

“…We expect to triangulate the source location based on the change of mole fraction depending on the change of wind direction (Luhar et al, 2014;Kumar et al, 2021).…”

“…The model error for the inversions using mobile measurements is defined consistently with the formulation of J p (Equation 10) (Kumar et al, 2021) as the average over all plume cross-sections during a release of the absolute value of the relative differences between the amplitudes of the observed and the simulated plume cross-sections when forcing the Gaussian model with a known release rate and location. This model error varies from approximately 27% (release-3) to approximately 100% (release-26) with an average value of approximately 54% for all the 25 releases.…”

“…Atmospheric transport inverse modelling strategies relying on atmospheric dispersion modelling and on its simple or statistical inversion, are often used to tackle both the localization and quantification of sources (Ars et al, 2017;Kumar et al, 2021). Controlled-release experiments with known location and emission rates have been conducted to support the development and evaluation of different measurement and modelling approaches and to improve the accuracy, reliability and applicability of anthropogenic methane emission monitoring at the facility-level scale (Loh et al, 2009;Brantley et al, 2014;Luhar et al, 2014;Albertson et al, 2016;Ars et al, 2017;Feitz et al, 2018;Ravikumar et al, 2019;Edie et al, 2020;Shah et al, 2020;Kumar et al, 2021). Such experiments reveal that the current level of uncertainties in the emission estimates hardly reach values lower than 30% (Brantley et al, 2014;Foster-Wittig et al, 2015;Albertson et al, 2016;FprEN-17628, 2021;Kumar et al, 2021).…”

“…Regarding (a), the inversion technique used by Kumar et al (2021) to estimate the single-point release location and rates from the near-surface mobile measurements, relied on the fit between the modelled and observed areas of individual methane mole fraction plume cross-sections and between the measured winds and the directions from the source location estimate to these individual plume cross-sections. Longer releases allow for increasing the number of plume cross-sections, which should thus increase the precision of such a method.…”

“…To our knowledge, there have been only a very limited number of experiments and studies of CH 4 emission estimation using such fixed-point measurements (Luhar et al, 2014;Feitz et al, 2018). Here, we propose a specific inversion technique, which, similar to Kumar et al (2021), takes advantage of the variation of the wind to increase the constraint on the inversion of the locations and rates of single CH 4 release points. It thus compares modelled and observed averages of the concentrations at the measurement locations binned over relatively short periods of time or within sectors of wind directions (with two versions of this inversion technique being tested).…”