Lidar observations of polar stratospheric clouds in Dumont d'Urville

As is well known, human-produced chlorine and bromine gases have contributed to a worrisome decrease in stratospheric ozone. This depletion is expected to gradually disappear as a result of reductions in the emission and production of these gases imposed by the Montreal Protocol of 1987. By the same token, since 1980 the amount of major greenhouse gases has increased, which not only affects climate but makes ozone recovery highly problematic. ORACLE-O3 is a collaborative project associated with the third International Polar Year (IPY). Its main purpose is to extensively study the amplitude and time extension of ozone recovery. Owing to their key role in ozone depletion, polar stratospheric cloud (PSC) studies are part of the project. PSCs form in the lower polar winter stratosphere when temperatures fall far enough. They are classified into two types, type 2 (large ice particles) and type 1 (small particles containing nitric acid), which in turn is subdivided into type 1a (solid) and type 1b (liquid). Particle type is largely influenced by temperature and the thermal history of the air masses in which PSCs occur.^1,2 However, substantial uncertainties about these processes remain. Understanding how PSCs form and evolve is a critical issue in quantifying the impact of climate change on their frequency and, further, on chlorine activation and subsequent ozone depletion.

Figure 1. Calculations of forward air-mass trajectories using the Goddard Space Flight Center model, starting over Davis at different altitude levels on 25 June 2006. DDU: Dumont d'Urville. DAV: Davis. MCM: McMurdo. TNB and so on represent other locations not considered in this article. (Courtesy of A. Klekociuk.)

Ground-based lidar (light detection and ranging) observations at Dumont d'Urville (DDU, located at 66^°40'S, 140^°01'E, ) started in 1989. The old system was replaced in 2005. The new lidar, LOANA (lidar ozone aerosols of NDACC—Network for Detection of Atmospheric Composition Changes—in Antarctica), allows stratospheric particle measurements at 532nm, and observation of depolarization, stratospheric temperature, and ozone concentrations. The station is a primary site of the NDACC.

Figure 2. Lidar observations (left) over Davis on 25–26 June 2006, showing PSC layers between 14 and 24km, corresponding to the origin of the trajectories above, and (right) over Dumont d'Urville on 3 July 2006, showing the same layers, corresponding to air masses originating over Davis, passing over Dumont d'Urville. UT: Universal time. MCS: Maultichannel scale. UTC: Coordinated universal time.

Figure 3. Simulations performed by the MIMOSA-μφ model of 532nm lidar backscatter coefficient fields at 410K on (left) 25 June 2006 and (right) 3 July 2006. The speckled white zone corresponds to ice crystals (type 2 PSCs). ECMWF: European Centre for Medium-Range Weather Forecasts.

Within ORACLE-O3, PSC lidar observations are being coordinated between Dumont d'Urville, Davis (68.00^°S, 78.50^°E), and McMurdo (77.86^°S, 166.48^°E) during IPY winters. Space lidar observations are made by CALIPSO. The originality of the approach is that, for the first time, it applies to lidar observations of PSCs the ‘MATCH’ method previously developed for ozonesonde measurements. This enables combination of lidar (both ground-based and space-borne) observations with trajectory calculations to infer information regarding how each PSC type forms and to assess our ability to predict these clouds for various environmental conditions. The analysis is carried out using microphysical model calculations. The main campaigns took place during the winters of 2007 and 2008, and the test MATCH-PSC was performed during winter 2006.

Here we present the first qualitative results from the 2006 pre-campaign. Trajectories were calculated by the Australian Antarctic Division using the Goddard Space Flight Center (GSFC) model. Figure 1 shows forward trajectories starting above Davis at different altitude levels from 13.9 to 29km. Trajectories from 23.8 and 26.3km arrive over Dumont d'Urville between 27 June and 3 July. The thick PSC layers (between 14 and 24km) observed over Davis are detected over Dumont d'Urville 7–8 days later (see Figure 2). The high advection contour model MIMOSA³ is coupled to a microphysical module that is able to calculate lidar optical properties at 532nm. Figure 3 displays simulations from 25 June and 3 July 2006 at 410K. They show that the PSCs observed over Davis and Dumont d'Urville are part of a large PSC field, rotating and evolving within the vortex.

Next steps include further quantitative studies of the evolution of observed backscatter coefficient values for several MATCH cases along the trajectories between three or at least two lidar stations. These values will then be compared with the corresponding CALIPSO observations. Our aim is to reveal correlations between evolution of the thermodynamical conditions of the air masses and the types of PSCs formed.