高分五号02星高光谱成像仪中美典型甲烷超级排放源遥感反演与分析

Mapping methane super-emitters in China and United States with GF-5 02 hyperspectral imaging spectrometer

检测和估算化石燃料生产活动中的甲烷泄漏有助于甲烷减排。星载高光谱成像仪是甲烷排放点源遥感监测的重要技术手段。本研究利用国产高分五号02星(GF-5 02)获取的高光谱成像仪(AHSI)与欧空局哨兵五号星(Sentinal-5P)的对流层观测仪(TROPOMI)遥感数据,基于优化的甲烷柱浓度反演算法和中尺度气象模式,开展了对中美两国煤矿和油气设施的甲烷排放点源检测、量化以及不确定性评估。实验结果表明:(1) GF-5 02星AHSI载荷在中美两国甲烷排放热点区域内探测到了4处显著的甲烷点源泄漏排放,排放速率均大于0.5 t/h。其中,在二叠纪盆地监测到一处超级排放源,甲烷排放量高达11.7±4.4 t/h;(2)甲烷点源排放通量速率的估算会受到背景气象场的影响,点源处风速的不确定性贡献最大。研究结果表明,GF-5 02星高光谱成像仪在全球甲烷点源遥感识别和排放量估算中的应用潜力,可以为未来全球能源行业的甲烷泄漏检测工作提供重要数据支撑。

Rapid identification of anomalous methane sources in the fossil fuel industry would enable action to reduce greenhouse gas emissions. Spaceborne hyperspectral imaging spectrometers have recently been shown to be instrumental for this mission. In this study, we utilize the rapid development of spaceborne imaging spectroscopy technology and data processing methods to perform a satellite-based large-scale and high-resolution survey of methane super-emitters in China and the United States. Our dataset is acquired by the Advanced Hyperspectral Imager(AHSI) onboard domestic GF-5 02 satellite(i.e., GF-5 02AHSI) and TROPOspheric Monitoring Instrument(TROPOMI)onboard European Space Agency's Sentinel-5 Precursor satellite(Sentinel-5P). Our core objective is to identify, quantify, and assess uncertainty of methane point emissions from coal mines and oil/gas facilities in China and the United States, with the overarching motivation of assisting future emission reduction efforts.Major steps include the following:(1) We retrieve methane concentration enhancements(i.e., increments above background levels in the amount of methane present in the atmospheric column, ? XCH4) using the optimized matched-filter algorithm applied to GF-5 02AHSI spectra in the 2300 nm shortwave infrared spectral region.(2) Emission plumes in the ?XCH4 maps are detected based on a semi-automatic method.(3) We estimate the source rate(Q) for individual methane plume using the integrated mass enhancement(IME) method.(4) We estimate uncertainties in Q by propagating random errors in IME and U10 to a 1-σ precision error in Q. A 50% random error in wind speed is assumed for U10 data, which is consistent with the approximately 1.5 m/s error standard deviation in wind speed.(5) We further assess the magnitude of our estimated plume-level emission rates through a simulation-based study with the weather research forecast model coupled with large eddy simulation.Major findings include the following:(1) GF-5 02AHSI has detected four significant methane point source leak emissions in methane hotspot regions of China and the United States, with the emission rates greater than 0.5 tons per hour. A super-emitter is detected in the Permian Basin, and the emission amount is 11.7 ± 4.4 tons per hour.(2) The estimation of methane point source emission flux rate is affected by the background meteorological field, and the uncertainty of wind speed at the point source has the largest contribution.Findings demonstrate the potential of GF-5 02AHSI in remote sensing identification and estimation of global methane point emissions,which can provide important data support for future methane leak detection in the global energy industry.