Algorithm improvement and validation of National Institute for Environmental Studies ozone differential absorption lidar at the Tsukuba Network for Detection of Stratospheric Change complementary station.

Recently, a data processing and retrieval algorithm (version 2) for ozone, aerosol, and temperature lidar measurements was developed for an ozone lidar system at the National Institute for Environmental Studies (NIES) in Tsukuba (36 degrees N,140 degrees E), Japan. A method for obtaining the aerosol boundary altitude and the aerosol extinction-to-backscatter ratio in the version 2 algorithm enables a more accurate determination of the vertical profiles of aerosols and a more accurate correction of the systematic errors caused by aerosols in the vertical profile of ozone. Improvements in signal processing are incorporated for the correction of systematic errors such as the signal-induced noise and the dead-time effect. The mean vertical ozone profiles of the NIES ozone lidar were compared with those of the Stratospheric Aerosol and Gas Experiment II (SAGE II); they agreed well within a 5% relative difference in the 20-40 km altitude range and within 10% up to 45 km. The long-term variations in the NIES ozone lidar also showed good coincidence with the ozonesonde and SAGE II at 20, 25, 30, and 35 km. The temperatures retrieved from the NIES ozone lidar and those given by the National Center for Environmental Prediction agreed within 7 K in the 35-50 km range.

Review of ozone and temperature lidar validations performed within the framework of the Network for the Detection of Stratospheric Change.

The use of assimilation tools for satellite validation requires true estimates of the accuracy of the reference data. Since its inception, the Network for Detection of Stratospheric Change (NDSC) has provided systematic lidar measurements of ozone and temperature at several places around the world that are well adapted for satellite validations. Regular exercises have been organised to ensure the data quality at each individual site. These exercises can be separated into three categories: large scale intercomparisons using multiple instruments, including a mobile lidar; using satellite observations as a geographic transfer standards to compare measurements at different sites; and comparative investigations of the analysis software. NDSC is a research network, so each system has its own history, design, and analysis, and has participated differently in validation campaigns. There are still some technological differences that may explain different accuracies. However, the comparison campaigns performed over the last decade have always proved to be very helpful in improving the measurements. To date, more efforts have been devoted to characterising ozone measurements than to temperature observations. The synthesis of the published works shows that the network can potentially be considered as homogeneous within +/-2% between 20-35 km for ozone and +/-1 K between 35-60 km for temperature. Outside this altitude range, larger biases are reported and more efforts are required. In the lower stratosphere, Raman channels seem to improve comparisons but such capabilities were not systematically compared. At the top of the profiles, more investigations on analysis methodologies are still probably needed. SAGE II and GOMOS appear to be excellent tools for future ozone lidar validations but need to be better coordinated and take more advantage of assimilation tools. Also, temperature validations face major difficulties caused by atmospheric tides and therefore require intercomparisons with the mobile systems, at all sites.