Twenty-two years observations of stratospheric ozone concentration, temperature and aerosol over Tsukuba, Japan

Differential absorption lidar (DIAL) at the National Institute for Environmental Studies (NIES) in Tsukuba (36°N, 140°E), Japan has been making routine observation for almost 22 years. Since 1988, more than 650 vertical profiles of stratospheric ozone and temperature have been obtained. We compared the lidar data with satellite data from the Stratospheric Aerosol and Gas Experiment (SAGE II) and assimilation data from the National Center for Environmental Prediction (NCEP). The lidar and SAGE II ozone profiles agreed within 5% in altitude range from 18 km to 40 km and within 10% up to 45 km. The lidar and NCEP temperatures agreed within 7 K in the 35to 50-km range. Ozone levels were highest in spring at altitudes below 20 km. Above 30 km, the ozone maximum occurred during summer. The annual cycle of temperature is observed with a spring maximum for all altitudes in the 35to 50-km range. Ozone variations caused by the Quasi-Biennial Oscillation (QBO) and the 11-year solar cycle are discussed, along with ozone trends observed after subtraction of the natural variations. INTRODUCTION Over the last several decades, increasing attention has been paid to stratospheric species and parameters related to ozone depletion and climate change. To understand the characteristics and their changes in the stratosphere, long-term measurements of ozone concentrations, aerosols and temperature have been conducted by means of ground-based sensors. In this paper, we present an overview of ozone and temperature data obtained with the lidar system at the National Institute for Environmental Studies (NIES) in Tsukuba, Japan, in the context of long-term observation of ozone concentration and temperature for period from 1988 to 2010. The lidar data are compared with data obtained from the Stratospheric Aerosol and Gas Experiment (SAGE II) and the National Center for Environmental Prediction (NCEP). METHODS AND APPARATUS The NIES ozone lidar system (located at 36°N, 140°E) is a typical UV differential absorption lidar (DIAL) system for low stratospheric ozone measurements. The lidar system has been in operation since 1988, and several replacements and improvements have been made since it was first installed (Nakane et al. 1992a; Nakane et al. 1992b). The system parameters and improvements were presented in detail by Park et al. 2006. Currently, the NIES ozone lidar system uses wavelength channels of 308/355-nm (for the Mie/Rayleigh scattering DIAL mode) and 332/386-nm (for the Raman scattering DIAL mode). The 355-nm signal is used for aerosol and temperature measurements. The SAGE II is a satellite instrument (McCormick et al. 1989), that measured ozone absorption at various altitudes by comparing the visible part of solar spectra obtained with different absorption cross-sections through the atmosphere. The SAGE II data were obtained from ftp://ftp-rab.larc.nasa.gov/pub/sage2/v6.20 using retrieval algorithm version 6.20. The vertical profiles were restricted to data obtained within ±3 latitude and ±20 longitude from the NIES lidar location. In this study, we used only values of ozone concentration with statistical errors smaller than 20% for both instruments. The NCEP vertical profiles of temperature were obtained from ftp://ftp.cpc.ncep.noaa.gov/wd53rl/ndscdata/ncep/temp. The NIES lidar vertical profiles of temperature were restricted to data obtained with statistical errors smaller than 15%. RESULTS AND DISCUSSIONS Figure 1 plots point-to-point long-term temporal variations of ozone concentration measured at altitudes of 15, 20, 25, 30, 35, and 40 km. The solid circles present results of measurements by NIES ozone DIAL during period from 1988 to 2007. The open squares show results from SAGE II up to July 2005. The time series plotted on the Fig.1 are based on 549 vertical profiles of the NIES ozone DIAL and 374 profiles of SAGE II. In general, the variations in ozone observed with the NIES ozone lidar agree well with those measured with the SAGE II, at all altitudes and over the entire period. The two instruments did not obtain measurements at exactly the same place and or time, and therefore there were small differences between individual data points for the two instruments. Systematic differences due to diurnal variations of ozone concentration have to be considered because the SAGE II measurements were taken at sunrise and sunset, whereas the lidar measurements were taken only at night. The fact that most of the profiles were not obtained during the same day, might result in overestimation of the disagreement between the DIAL and SAGE II data. To minimize these effects, we used monthly mean values of ozone concentration for the following analysis. The time series of ozone exhibited not only longterm trends but also shorter time scale variations, in which annual cycle, Quasi-Biennial Oscillation (QBO) and the 11-year solar cycle are significant. Among these, only annual variations of ozone can be clearly distinguished for all the altitude in Fig. 1. Therefore, to accurately analyze other variations and the long-term trends, subtraction of annual cycle is required. Figure 2 plots point-to-point long-term temporal variations in temperature measured at altitudes of 35, 40, 45, 50, 55, and 60 km. The solid circles present results of measurements by NIES DIAL during period from 1988 to 2007. On the same figure by open red squares are also shown the temperature time series obtained by interpolation from the three-dimensional grid data (NCEP upper atmosphere data) on the same day and at the same time. The figure plots 574 temperature profiles retrieved from lidar signals and 6791 NCEP temperature profiles. In general, seasonal variations and year-to-year variations in temperature exhibit good agreement between two data sets. The variations in temperatures measured with the NIES lidar and the NCEP temperatures agreed within 7 K in the 35to 50-km altitude range, which corresponding to relative difference smaller than 2% The climatological mean vertical profiles obtained by lidar and SAGE II, and the relative differences between them, are shown in Figure 3. These profiles were obtained by averaging of all available monthly mean profiles from 1988 to 2005, separately for the two instruments. The climatology mean ozone profiles measured by lidar and SAGE II agreed within ±5% in the altitude range from 16 to 37 km and within 20% up to 44 km. At 15 km, the climatological mean obtained by lidar was markedly smaller then the SAGE II mean (relative 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 0.2 0.4 0.6 0.8 1.0 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 0.5 1.0 1.5 2.0 2.5 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 2.0 3.0 4.0 5.0 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 3.0 3.5 4.0 4.5 5.0 5.5 6.0 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 1.0 2.0 3.0 4.0 5.0 6.0 7.0 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 0.0 1.0 2.0 3.0 4.0 5.0 15 km 20 km 25 km 30 km 35 km 40 km O zo ne c on ce nt ra tio n (x 10 18 m -3 ) Lidar SAGE II