Inversion Of Aerosol Backscatter And Extinction Coefficients Based On Ground-based Raman-Mie Lidar

Pure rotational Raman-Mie lidar detects atmospheric aerosol with the pure rotational Raman backscatter signals of molecules and the elastic backscatter signals of aerosol particles.It's a useful system for the study of climatology and meteorology as with the advantages of providing accurate atmospheric aerosol profiles with high spatial and high temporal resolution.This dissertation focuses on the self-developed pure rotational Raman-Mie lidar system for data precision inversion and application research.The parametric model of the lidar system is developed by simulation to verify the feasibility of the proposed improved inversion methods.Besides,new methods(including de-noising scheme,geometric factor correction scheme,calibration scheme and lidar ratio inversion scheme)for determination of accurate aerosol optical properties are also proposed based on the influence factors of inversion precision.The main innovation content of this dissertation is summarized as follows.1.Proposed an improved de-noising method to improve the backscatter signals and improve the inversion accuracy of aerosol backscatter coefficient in the near range and low signal-to-noise ratio region.The numerical and experimental results show that the aerosol backscattering coefficient in the near range retrieved with improved de-noising algorithm is about 4-10 times accurate than that retrieved with traditional de-noising algorithm.Moreover,the effective detection altitude can be well improved by the improved de-noising algorithm.2.Focused on the geometric factor determination method for the Mie channel of the Raman-Mie lidar system and an improved calibration method for the reference parameter to exclude the error introduced in the existing method by the iteration and the aerosol retrieval with elastic signal.The accuracy can be improved by 2 times with the proposed geometric factor correction method because of the insensitivity to aerosol lidar ratio.The improved calibration method can improve the inversion accuracy by about 10-14 times owing to the less dependence on reference altitudes.3.Proposed an iterative method to invert the lidar ratio profile and the aerosol extinction coefficient.This method can obtain the coherent and accurate inversion results based on a derived inverse relationship between atmospheric backscatter coefficient and aerosol lidar ratio.Simulation and experiment results show the inversion accuracy of aerosol extinction coefficient for iterative method can be improved in the near-surface aerosol layer and the optical thick layer through avoiding the insufficiency error and incoherence error.4.Analyzed the trait of upper air visibility and the haze two-dimensional evolution characteristic in haze days.Firstly,a multi-functional aerosol inversion software interface is coded with MATLAB based on the proposed improved inversion methods.Then the hourly and daily variation traits of upper air visibility were investigated during the haze episode in the northwest of downtown Beijing.An opposite tendency can be found for upper air visibility compared with the changes of PM2.5 mass concentration.And the upper air visibility on non-haze days is about 3-5 times higher than that on haze days.Moreover,the vertical transport of pollutants can be drawn from the delayed response between upper air visibility at high altitude and at low altitude.In addition,the two-dimensional haze characteristics could be investigated using the correlation between vertical haze parameters(atmospheric boundary layer,haze thickness and aerosol optical thickness)and horizontal haze parameter(upper air visibility).Finally,the characteristics of multi-parameters have been analyzed and concluded for different haze levels.