Research And Application Of NO₂ Retrieval Algorithm For Environmental Trace Gases Monitoring Instrument

Optical remote sensing technology,as an important monitoring method,has been used in various earth observation and meteorological observation.Especially based on the development and application of optical remote sensing technology on different observation platforms,many important discoveries and understandings have been made.Since the late 1970s,optical remote sensing technology based on differential optical absorption spectroscopy(DOAS)has been proposed,which has promoted the widespread application of spectroscopy-based analysis methods in atmospheric monitoring.At the same time,it also promotes the development of environmental science and atmospheric science,such as the monitoring of pollutant gases and trace gases in the atmosphere,to make people aware of the sources and transmission of these polluting gases,and to provide reference and basis for pollution control and policy formulation.At present,the DOAS-based spectral telemetry method has been successfully applied to satellites,airborne and ground-based observation platforms.It is of great significance in the establishment of the earth's atmospheric pollution model,the summary of the distribution and change of polluted gases,and the monitoring of gas columns.Compared with other air pollution analysis methods,this method has its unique advantages.First,the optical telemetry method uses the transmission characteristics of light in the atmosphere,which is not available by other physical and chemical methods.Therefore,it can be used for various space monitoring including outer space,atmosphere,and ground surface.Second,the DOAS algorithm uses the natural light as the analysis medium,compared with the chemiluminescence method and other analysis methods with greater applicability and limitations,it has the characteristics of fast inversion speed and wide observation range.With the demand for pollution traceability and large-area monitoring,DOAS-based satellite telemetry platforms have become one of the important ways for humans to understand the distribution and changing of atmospheric pollutants such as NO2.The multiple hyperspectral detection payloads based on the ultraviolet-visible band launched by ESA and NASA have made many achievements in the field of satellite telemetry,and china also successfully launched the first ultraviolet-visible hyperspectral payload-Environmental trace gases Monitoring Instrument(EMI)in May 2018,to make up for the gap in the monitoring of pollutant gases based on spectral analysis on china's satellite-borne platform.However,due to insufficient payload calibration and the limitation of key components of the instrument,as well as the introduction of unknown spectral structure in the Fraunhofer reference spectrum acquisition,obvious striping occurred during the NO2 inversion process,which reduces the accuracy of concentration retrieved.Therefore,the research is mainly based on the ultraviolet-visible band of the EMI,and the overall process of obtaining the tropospheric NO2 amounts is discussed and studied,including the optimization of EMI NO2 inversion algorithm,the processing of NO2 remote sensing strips,the separation of tropospheric and stratospheric NO2 columns,and the verification and application of data products.The main work of the research is as follows:Firstly,the EMI NO2 inversion algorithm has been studied and optimized.This paper discusses the selection of various parameters in the EMI NO2 retrieval algorithm,and then analyzes the accuracy and striping characteristics of the NO2 inversion results.The optimal spectral band selection scheme based on the analytic hierarchy process is proposed,and the AHP-DOAS inversion algorithm is constructed.In addition,the research found that there are obvious fixed fitting residuals in the inversion process of pixels in different spatial dimensions.From this,the principal component analysis method is used to extract the residud structure for analysis.After the above corrections,accurate NO2 columns products were obtained,which provided effective data support for further analysis and verification of data products.The second is the correction of the concentration striping of EMI NO2 slant column.After the optimization of the EMI NO2 inversion algorithm,the strip effect is obviously improved,but there is still a certain degree of striping.The paper evaluated the characteristics and effects of the moment matching method and the Fourier transform method in removing NO2 strips.Finally,the strip removal method of Fourier transform filtering was selected to realize the correction of the NO2 columns.The third is the separation of tropospheric-stratospheric NO2 columns based on EMI-based data products.Using tropospheric-stratospheric separation technology,the feasibility of STREAM tropospheric-stratospheric layering method in the application of EMI NO2 products is analyzed.According to the global distribution characteristics of NO2,the effect of STREAM method and reference sector method(RSM)on stratospheric-tropospheric NO2 stratification was compared.The fourth is the verification and application of EMI NO2 columns resultsIn order to further verify the reliability of the NO2 L2 data product,the EMI data product was verified by using the NO2 columns inversion results of the multi-axis DOAS installed in the Beijing-Tianj in-Hebei region.The comparison between TROPOMI and ground-based data products and EMI shows the applicability and reliability of the optimization of the NO2 inversion algorithm for EMI instruments.In addition,based on EMI NO2 data products and ground-based NO2 telemetry data,the NO2 distribution and changes in the Beijing-Tianj in-Hebei region were analyzed,and the NO2 pollution situation in Beijing from 2018 to 2019 was evaluated.Using statistical methods and meteorological data information,the transmission characteristics of NO2 pollution in this area were studied,which once again proved the importance of EMI in monitoring NO2 pollution.