Abstract

For many years lidar systems have proved to be valuable tools for the measurement of atmospheric data with high spatial and temporal resolution. The lidar variant used so far for ozone height profiling has been known as the differential absorption and scattering, or DIAL, technique.¹ In DIAL two light pulses of different wavelengths are transmitted into the atmosphere, the elastically backscattered return signals are registered, and the ozone concentration is calculated from the difference of optical absorption by ozone at the two wavelengths. However, due to elastic particle scattering DIAL measurements are hardly possible in altitudes with enhanced particle content. The influence of particle scattering on the retrieval of ozone concentrations from measured lidar profiles can be reduced substantially if inelastic molecular return signals of a gas such as nitrogen are used instead of the elastic lidar returns. This so-called Raman-DIAL technique² has been implemented in the GKSS Raman lidar in spring 1994.³ A XeCl excimer laser and a frequency-tripled Nd:YAG laser serve as light sources, and the return signals at the nitrogen Raman wavelengths are used for the calculation of ozone concentration profiles. In addition, tropospheric water vapor and backscatter and extinction properties of atmospheric particles are obtained with the Raman-lidar technique. Elastic depolarization ratios are measured as well. Whereas the conventional DIAL and the Raman lidar techniques have undergone an extensive error analysis,^4,5 an error treatment of the Raman-DIAL technique has not been published so far. In the present contribution an outline of a Raman-DIAL error analysis is given. As space is limited, only ozone measurements in the free troposphere in the presence of cirrus clouds are considered.

© 1997 Optical Society of America