Calibration method of Fernald inversion for aerosol backscattering coefficient profiles via multi-wavelength Raman–Mie lidar

Abstract

Mie scattering lidar is widely used in the detection of atmospheric aerosol optical properties, however, the inversion algorithm is lack of accuracy, which is mainly caused by the inaccurate calibration of reference condition in Fernald inversion. Here, we propose a multi-wavelength aerosol backscattering ratio parameterized-calibration method (MABP-CM) based on parameterized equations for reference condition calibration of Fernald inversion. The aerosol backscatter ratios at 355 nm, 532 nm, and 1064 nm wavelengths are respectively calculated from the aerosol size distribution of aerial survey data. We find that there is a positive correlation between the backscatter ratios at different wavelengths, and the parameterized equations are established. The parameterized equations can be used as conversion factors of aerosol backscatter ratio at different wavelengths, which is beneficial to improving the calibration accuracy of multi-wavelength backscatter ratios at reference height in Fernald inversion. The Raman inversion is used to retrieve the 355 nm aerosol backscattering coefficient and aerosol backscatter ratio. Then, combined with MABP-CM, the 532 nm and 1064 nm aerosol backscatter ratios at reference height are respectively derived, to obtain the backscattering coefficients with Fernald backward integration equation. With the improved method, the reference height is allowed not to be at clear atmosphere, and the reference condition can still be accurately calibrated. Simulation verification of multi-wavelength backscattering coefficients is carried out with aerial survey data. Our simulation suggests that MABP-CM can reduce the error of 532 nm backscattering coefficient to ±20%, and reduce the error of 1064 nm backscattering coefficient to ±40%. Multi-wavelength aerosol backscattering coefficients can be simultaneously retrieved with high accuracy on cloudy and heavily haze days with MABP-CM. Finally, the multi-wavelength Raman–Mie lidar is used to observe the short-term and long-term changes of aerosol backscattering coefficient on clear and cloudy days.