The Total Ozone Column as a Climate Variability Indicator: The South American Case
Historically trend analyses of the atmosphere have been carried out within fixed geographical bands in both Northern and Southern Hemisphere (WMO, 1999). However, alternative ways to classify the atmosphere by the use of the tropopause height, total column ozone or geopotential height were proposed almost twenty years ago (Shalamyanskiy and Romashkina, 1980; Karol et al., 1987; Shapiro et al., 1987). Experimental data were used to break down the atmosphere in regions where the variability of several meteorological quantities is greatly reduced (Karol et al, 1987). In a recent study for the Northern Hemisphere, Hudson et al. (2003) have shown that in fact the atmosphere can be classified in at least three permanent regions named as tropical, mid-latitude, and polar regimes. The boundaries of these regions are the upper-level subtropical and polar jets, and their associated fronts. In their work, total ozone from the Total Ozone Mapping Spectrometer (TOMS) is used to delineate the position of these upper-level jets and therefore the boundaries of each regime.

The importance of this work lies in the fact that trends of any atmospheric constituent that depends on the regimes characteristics can be analyzed separately within each regime. Even further, due to a dramatic reduction of artificially introduced noise (by mixing all regimes) it seems possible to interpret and associate more precisely the temporal behavior of ozone within each regime with climatic phenomena such as El Niño-Southern Oscillation (ENSO), the Arctic Oscillation (AO), and the Quasi-biennial Oscillation (QBO).

In the specific case of the Southern Hemisphere (and South America in particular) the interactions ocean-atmosphere can be studied more accurately when the total ozone column is analyzed by regime. In this sense, the main goal of this work is to analyze the behavior of the total ozone column within each regime for relatively long period. A much “cleaner” signal for every regime could lead to a better understanding of the climatic variability by associating the ozone anomalies to ocean/atmosphere events. Since there is almost continuous daily total ozone data from the TOMS instruments for about 24 years, the proposed analysis will take into account at least two solar cycles, many El-Niño events, and several QBOs as well. Furthermore, the comparison of the temporal evolution of precise experimental ozone data with the behavior of other meteorological quantities (for instance, SST and winds) could contribute to the understanding of the interactions between ocean and atmosphere in the South American region.

In order to classify the atmosphere in the Southern Hemisphere similar techniques as described by Hudson et al. (2003) will be applied. This is, total ozone from TOMS will be used to determine the location of the subtropical and polar fronts. There are some complications in the Southern Hemisphere that prevent following this approach directly. First, the scarcity of ground-based data (rawin and ozone sondes) makes the validation on a daily basis almost impossible. Second, larger differences between ground-based and satellite ozone column were reported for the Southern Hemisphere that reduces the reliability of the total ozone retrieved from TOMS (Piacentini et al., 2000). Third, the use of data from the re-analysis helps to confirm the general location of the boundaries but do not solve the problem completely because the re-analysis is also affected by the lack of ground-based information as discussed by Kistler et al. (2001). Clearly this fact worsens for the Southern Hemisphere case where only a small fraction of the surface is covered by land.

For these reasons, a combination of all the information available, which include the use total ozone column from TOMS (McPeters et al., 1996), reanalysis data from NCEP/NCAR (Kalnay et al., 1996), satellite-derived profiles of ozone and water vapor, and ground-based rawinsondes and ozonesondes will be used to identify the upper-level jets and the corresponding fronts.