Atmospheric CO₂ Inversion Method Based On Mid-wave Infrared Spectral Information

The massive emission of greenhouse gases do great harm to the earth’s environment.One of the most important greenhouse gases is CO2,atmospheric CO2 is the main object of the climate changing research and the policy formulation of energy conservation and emission reduction,and thus the accurate detection of CO2 change is the basis of these work.Satellite remote sensing plays an irreplaceable role with its global detection capability.The CO2 content in the atmosphere is low and the gradient is small,while the research on climate change has high requirements on the accuracy of atmospheric CO2 content.Therefore,the key to satellite CO2 remote sensing is the high precision of detection,in which the inversion is an essential part of remote sensing,so that high precision becomes the key of this technology.At present,it is difficult to improve accuracy and the results are not stable enough when it comes to the satellite remote sensing inversion,and the atmospheric CO2 source is not concerned.In order to improve the accuracy of atmospheric CO2 concentration inversion and solve the problem of remote sensing of carbon sources,this paper discusses the problems of atmospheric CO2 inversion,starting with the characteristics of medium wave infrared spectrum,and explores simulation analysis and inversion experiment.Given the characteristic that the strong absorption of atmospheric CO2 in midinfrared band is sensitive to the change of CO2 content,this paper analyzes the spectral characteristics of mid-infrared radiation at 4.3 μm and the influence of environmental factors including aerosol,temperature,nitrous oxide and water vapor,etc.,finding that the band basic is not affected by aerosol and surface,the main spectral radiation energy comes from atmospheric thermal emission,and CO2 has very obvious absorption characteristics in the spectrum range.Also,through the comparison of the characteristics of different bands,it is found that,compared with the near-infrared band,the mid-infrared band has different sensitivity to the change of CO2 concentration,and the mid-infrared band is more sensitive at 5～15 km.And in comparison with the thermal infrared band,it is less affected by water vapor.In view of these characteristics of midinfrared band in atmospheric CO2 inversion,an atmospheric CO2 inversion algorithm based on mid-infrared band is designed.Due to the lack of satellite measured midinfrared hyperspectral data at the present stage,inversion experiments based on simulated spectra are adopted.The results show that the CO2 profile of the middle and upper atmosphere at 5～15 km is in good agreement with the true value profile.The content of 13CO2 in the atmosphere is an important basis to distinguish between the natural and man-made carbon sources,which is of great significance in climate research.It is found that the mid-wave infrared contains a strong absorption characteristic band of the carbon isotope 13CO2 which is separated from the 12CO2 band.Due to the low content of 13CO2 in the atmosphere,the strong absorption characteristics of mid-infrared can well reflect the change of 13CO2 in the atmosphere.Because of the atmospheric convection and circulation,there is a close relationship between the surface carbon emission and the 13CO2 in the middle and upper atmosphere between 5km and 15km.The sensitivity of mid-infrared to this high atmosphere gives it the potential to remote sensing the 13CO2.The research in this paper shows that the 13CO2 detection has high requirements.For example,the temperature profile needs to meet the accuracy higher than 0.03 K;the accuracy of nitrous oxide profile needs to be higher than 5ppb;the instrument needs to have a signal-to-noise ratio above 600[1].The 13CO2 concentration inversion method and simulation experiment show that the accuracy of environmental parameters affects the inversion results:the inversion error of about 0.4 ppm will be caused by the temperature increase of 1 K;30%of the water vapor profile error will bring about 0.05 ppm inversion error;an increase of 0.02 ppm in N2O content will lead to an increase of about 0.4 ppm in inversion results.These results are important bases for the design of a new generation of satellite remote sensing detection and inversion methods.When it comes to improving the application ability of satellite remote sensing monitoring results,the accuracy and stability of atmospheric CO2 inversion is a key problem.In this paper,two methods of algorithm collaboration and band joint are used to optimize the inversion algorithm.In the collaborative algorithm,taking the atmospheric environment in China as the research object,this paper calculates the CO2 profile samples according to the differences in China’s regional characteristics and constructed the representative sample set that is suitable for China’s regional characteristics,then substituted the CO2 profile obtained by statistical inversion as the initial value into the physical inversion method,finally formed a new algorithm for synergistic statistics and physical methods.The experimental analysis of GMI remote sensing data shows that the accuracy of the collaborative inversion algorithm is 47.7%higher than that of the single physical inversion algorithm.In terms of the band joint algorithm,near-infrared and mid-wave infrared are combined.By taking advantage of the characteristics of different sensitivity of near-infrared and mid-infrared bands to changes in atmospheric CO2 concentration at different heights,the algorithm experiment of combination inversion of CO2 concentration profile at 0-15km in nearinfrared and mid-wave infrared bands is carried out in a complementary way at different heights.The inversion results show that using the near-infrared and thermal infrared bands to constrain the CO2 profile at the same time can better reconstruct the vertical distribution of atmospheric CO2 and improve the inversion accuracy of atmospheric CO2.