Acquisition of onboard CH4 detection data and high-precision CH4 satellite remote sensing inversion algorithm

To obtain information on the spatiotemporal distribution and variation of near-surface CH4 concentration, shortwave infrared CH4 spaceborne detection technology has developed rapidly in recent years. Multiple shortwave infrared detection satellites have been launched both domestically and internationally, conducting a series of CH4 detection and related application studies, achieving space-based CH4 detection. The scientific experimental satellites launched during this period have promoted the development of CH4 satellite remote sensing inversion algorithms and flux inversion algorithms, laying a technological foundation for improving space detection capabilities, producing data products, and directly applying them. Furthermore, they have demonstrated the application potential of shortwave infrared detection in global CH4 source-sink balance estimation from the perspectives of detection data accuracy and flux estimation.

Acquisition of spaceborne CH4 detection data

Accurate acquisition of satellite-borne CH4 detection data requires the support of high-precision CH4 satellite remote sensing inversion algorithms. The quality of satellite-detected CH4 data products typically depends on the selection of detection instruments and inversion parameters. Short-wave infrared detection has the advantage of being sensitive to near-surface CH4 concentration changes, but short-wave infrared absorption spectra are easily affected by atmospheric and surface parameters such as interfering gases, temperature, clouds, and aerosols, introducing significant errors. The goal of high-precision inversion algorithms is to reduce the influence of interfering factors and obtain the concentration information of the target gas from the effective spectral information. Internationally, various greenhouse gas satellite remote sensing inversion methods have been established using the spectral detection methods of the SCIAMACHY and GOSAT satellites. Among these, the most widely used are the all-physical model inversion algorithm, the proxy inversion method, and the WFM-DOAS inversion method.

Proxy algorithm

The Proxy algorithm typically uses stable atmospheric components such as CO2 as representatives to correct the influence of clouds and aerosols on the optical path in the CH4 detection band, thereby extracting effective CH4 concentration information. This requires that the influence of clouds and aerosols on the optical path in the bands where the target gas and the representative gas are located be comparable. Therefore, it is usually only applicable to satellite detection that simultaneously covers the 1.6 μm CO2 and CH4 absorption bands. The GOSAT data product obtained by the UoL-Proxy algorithm developed by the University of Leicester, UK, has been verified by TCCON, with a single-point detection accuracy of 13.72 ppb and a global overall bias of approximately 9.06 ppb. Studies show that the accuracy of XCH4 retrieved using the Proxy method is significantly affected by the accuracy of the representative gas XCO2, the optical path, and systematic error corrections, with the retrieval accuracy ranging from approximately 0.6% to 2%.

WFM-DOAS algorithm

The WFM-DOAS algorithm is based on classic differential absorption spectroscopy, using a weighting function of the total amount of the target gas column instead of the absorption cross section to fit the differential absorption spectrum. Initially built upon SCIAMACHY's detection spectroscopy, the WFM-DOAS algorithm enabled the inversion of various gas column concentrations; it has now been applied to the XCH4 and XCO inversion in Tropomi, yielding data products with high consistency with the results of the all-physical algorithm. Ground-based validation results show that the random error of XCH4 obtained by the WFM-DOAS algorithm is 14.0 ppb (0.8%), and the systematic error is 4.3 ppb (0.2%). The WFMDOAS inversion algorithm has relatively low requirements for spectral resolution, but high requirements for the accuracy of spectral calibration and auxiliary parameters such as meteorological fields (e.g., temperature and humidity profiles, pressure), and is significantly affected by cloud cover.

Full Physical Inversion Algorithm

All-physical inversion algorithms utilize forward modeling to simulate the atmospheric radiative transfer process of signals, significantly reducing errors caused by optical path uncertainties due to scattering during radiative transfer, and typically achieving high inversion accuracy. Major international greenhouse gas satellite remote sensing inversion algorithms include the GOSAT standard algorithm NIES-FP, NASA's OCO-2 standard algorithm ACOS, the University of Leicester's UoL-FP, the TROPOMI standard algorithm RemoTeC, and the Chinese carbon satellite standard algorithm IAPCAS. These algorithms can be used to obtain high-precision atmospheric XCH4 and XCO2 data products from satellite shortwave infrared spectra. With in-depth analysis of inversion results errors and increasing demands for product accuracy, these algorithms are continuously updated and optimized. Currently, the accuracy of XCH4 inversions using TROPOMI and GOSAT has been greatly improved. The inversion accuracy of TROPOMI XCH4 obtained using the improved RemoTeC algorithm has reached −3.4 ± 5.6 ppb. The acquisition of high-precision XCH4 data products lays a data foundation for atmospheric CH4 budget estimation. The IAPCAS algorithm was developed for data applications from China's TanSat satellite. Using this algorithm, a high-precision XCO2 data product of 1.47 ppm can be obtained from the TanSat probe spectrum, and it also has the ability to obtain XCH4 and XCO2 data from the GOSAT spectrum. The establishment of this algorithm and its successful application in multi-satellite inversion have laid the algorithmic foundation for the practice of subsequent satellite missions in my country.

Flux estimation methods

Estimating global and regional CH4 budgets can help improve our understanding of CH4 source-sink distribution and provide a reference for addressing future climate change. Current flux estimation methods are mainly divided into two approaches: bottom-up and top-down. The bottom-up approach, based on emission inventories, station-observed fluxes, or grid-scale simulated flux data, performs spatial extrapolation and is a traditional carbon budget estimation method. However, due to the low timeliness of data, the calculation results are usually lagging and cannot capture dynamic changes in sources and sinks in a timely manner. The top-down approach mainly relies on atmospheric component concentration data and emission inventory data obtained through observation, utilizing atmospheric chemical transport models and data assimilation methods to estimate fluxes. Existing research shows that using satellite observation data for flux estimation can reduce uncertainty by about 85% compared to using only ground-based observations. However, this method has strict requirements on the accuracy of the observation data; even a 1% regionally or temporally correlated systematic bias can affect the validity of the flux inversion results. Based on atmospheric chemical transport models and high-precision satellite XCH4 data products, multiple teams have conducted research on CH4 flux estimation. Numerous studies have shown that using satellite data in flux retrieval can significantly reduce the uncertainty of flux retrieval in different regions, providing important validation data for bottom-up CH4 emission inventories and wetland emission models. Currently, GOSAT data is widely used in surface CH4 flux estimation, aiding in the analysis of CH4 source and sink distribution and changes at global and regional scales. It provides reliable evidence for assessing the contributions of different sectors to CH4 emissions and for gaining a deeper understanding of changes in CH4 source and sink distribution and their impact on climate. TROPOMI has high spatiotemporal resolution, and its data volume and accuracy are far superior to GOSAT, enabling it to excel in fine-scale flux estimation. These research results provide scientific data support for global methane budgeting and further understanding of CH4 source and sink changes.

High spatiotemporal resolution satellite-borne CH4 data products can directly reflect anthropogenic CH4 emission processes. TROPOMI has demonstrated its advantage in continuous spatiotemporal detection for CH4 emission monitoring and estimation, and its high-precision detection capabilities also create conditions for the monitoring and quantification of regional area-source CH4 emissions. Combined with synchronously detected atmospheric CO concentration data, the stability of local CH4 emissions can be independently assessed without atmospheric transport models. The GHGSat series of commercial satellites, with high spatial resolution, are designed to acquire CH4 point-source emissions from industrial production. By identifying fine-scale CH4 emission plumes, they have been widely used in local emission monitoring and emission estimation.

Source of reference material:

Analysis of the Current Status and Development Trends of Atmospheric Methane Observation Satellites Aiming at Carbon Neutrality and Carbon Peak in my country