Evaluation of systematic errors for the continuous-wave NO2 differential absorption lidar employing a multimode laser diode.

Abstract

The NO2-differential absorption lidar (NO2-DIAL) technique has been of great interest for atmospheric NO2 profiling. Comprehensive studies on measurement errors in the NO2-DIAL technique are vital for the accurate retrieval of the NO2 concentration. This work investigates the systematic errors of the recently developed continuous-wave (CW) NO2-DIAL technique based on the Scheimpflug principle and a high-power CW multimode laser diode. Systematic errors introduced by various factors, e.g., uncertainty of the NO2 differential absorption cross-section, differential absorption due to other gases, spectral drifting of the λon and λoff wavelengths, wavelength-dependent extinction and backscattering effect, have been theoretically and experimentally studied for the CW-DIAL technique. By performing real-time spectral monitoring on the emission spectrum of the laser diode, the effect of spectral drifting on the NO2 differential absorption cross-section is negligible. The temperature-dependent NO2 absorption cross-section in the region of 220-294 K can be interpolated by employing a linear fitting method based on high-precision absorption spectra at 220, 240, and 294 K. The relative error for the retrieval of the NO2 concentration is estimated to be less than 0.34% when employing the interpolated spectrum. The primary interference molecule is found to be the glyoxal (CHOCHO), which should be carefully evaluated according to its relative concentration in respect to NO2. The systematic error introduced by the backscattering effect is subjected to the spatial variation of the aerosol load, while the extinction-induced systematic error is primarily determined by the difference between the aerosol extinction coefficients at λon and λoff wavelengths. A case study has been carried out to demonstrate the evaluation of systematic errors for practical NO2 monitoring. The comprehensive investigation on systematic errors in this work can be of great value for future NO2 monitoring using the DIAL technique.