High Spatial and Temporal Resolution Assessment of Methane Emissions in Major Coal-Producing Regions of China Driving In Situ and Column Observations by WRF-GHG

Methane (CH₄), due to its strong global warming potential and roughly decade lifetime, plays a pivotal role in near-term climate mitigation strategies. Coal mine methane (CMM) is a major source of CH₄ globally, and particularly so in coal-producing regions. Effective control of CMM emissions is crucial for mitigating climate change impacts. This study integrates multi-scale observational datasets including in-situ observations, column FTIR observations, and daily satellite remote sensing together with high-resolution atmospheric modeling to investigate CH₄ transport and source contributions. The steps are hoped to lead to future development of a quantitative inverse modeling system flexible enough to provide for spatially-targeted, high-frequency mitigation strategies and interventions.

The Weather Research and Forecasting model is configured with its greenhouse gas tracer option (WRF-GHG) to separate sources of CH₄ on a grid-by-grid basis. This work employs a nested three-domain structure with the central region covering China’s major coal-producing regions, including the Qinshui Coalfield and 139 coal mines at 3km resolution. Ground-based in situ measurements from eddy covariance flux tower, mobile measurements using portable LGR analyzers, and upward looking EM27 FTIR previously deployed by the AERSC team at around 950 m elevation in Changzhi in Shanxi Province, provide high-frequency a priori emissions to drive the model, as well as in situ data to refine the simulations. Observations from TROPOMI, TCCON (in the 9km region) and the GAW WLG station (in the 27km region), provide additional datapoints for comparison and quantification of the spatial, temporal, and goodness of representation of fit. This study resolves CH₄ concentration dynamics across scales, from regional to individual coal mines. It is hoped that the results can offer a quantitative means to identify and attribute emissions from these major emitting regions at high spatial and temporal frequency.

A few interesting scientific points are explained in detail. (1) The WRF-GHG model shows improved agreement with observational datasets, especially so in terms of capturing more extreme events, when the AERSC team’s emissions datasets are used. Quantitative differences between WRF-GHG and TROPOMI_L3 demonstrate that while some areas are robust, other areas have significant differences, explained in part by the improved emissions inventories used herein. (2) Existing inventories lead to average values of XCH₄ simulated across 139 individual coal mines being lower than the observations made by the AERSC group, while at the same time leading to average values of XCH₄ in a subset of other regions not measured as being much higher than available observations. These regions and differences are detailed, and rationales for these differences are proposed. (3) Near-surface CH₄ concentrations show that anthropogenic CH₄ emissions contributes only 17.1% of total CH₄ within a 12 km radius of coal mining sites which is far too low, although the diurnal variations are closely linked to coal production activities, highlighting the model's robustness in capturing these dynamics but the problems with the spatial and magnitude aspects of current emission inventories.