Results

Figure 7. Total column ozone over Canada for 2002, 2003, 2004,
and all these years together

Figure 8. Total column ozone over Canada from 01.01.2002 until 31.12.2004

        OSIRIS measures ozone over Canada not during the whole year, but from the end of February and until the beginning of October. This is because OSIRIS operates by measuring the scattered sunlight near the Earth twilight region (also called the  ‘terminator), and the polar night at high latitudes prevents it from making measurements during October-February. Similarly, OSIRIS does not measure the ozone in the Southern hemisphere during local summer from June until September. In addition, the orbit of Odin is such that it devotes more time observing the atmosphere in the Northern hemisphere. Thus, more measurements are usually taken in the Northern hemisphere that in the Southern.

        Figure 7 shows TCO over Canada during 2002-2004 and Figure 8 – same data, but plotted for three years in a row.

        OSIRIS TCO over Canada is highest in spring (March-April) and lowest in fall (September-October). During the summer time, most popular for all kinds of outdoor activity, TCO over Canada is quite high and varies approximately from 270 to 330 DU.
There is no significant difference in TCO over Canada between 2002, 2003 and 2004 except for early spring of 2004 when TCO were about 10-20 DU lower than during previous 2 years.

Figure 9. Total column ozone over Region #3 for 2002, 2003, 2004,
and all these years together

Figure 10. Total column ozone over Region #3 from 01.01.2002 until 31.12.2004

        Figure 9 shows TCO over Region #3 (a mirror projection of Canada’s location over equator into the Southern hemisphere) during 2002-2004 and Figure 8 – same data, but plotted for three years in a row.

        As mentioned before, Odin is more focused on the Northern hemisphere and its measurements in the Southern hemisphere do not last as many months as in the Northern.  Consequently, OSIRIS ozone data in the Southern hemisphere are limited to the period from October until March at higher latitudes and from September until April at lower latitudes. That is why the gaps in the data for Region #3 are much larger than for Canada and all other chosen locations.
OSIRIS TCO over Region #3 is different from that over Canada:

        1. It is generally 20-40 DU lower than over Canada and
        2. It is very low (up to 180 DU) in September-October because of the large ozone destruction         and formation of the ozone hole.

        During the local summer (outdoor) time, TCO over Region #3 is between 260 and 300 DU, which is 20-40 DU lower, than over Canada.

        TCO over Region #3 did not change much from 2002 to 2003, but in the local summer of 2004 (November-December) it was 10-30 DU lower than in both 2002 and 2003. I think that this is a very alarming signal that the ozone in the Southern hemisphere (at higher latitudes) may still be decreasing!

Figure 11. Total column ozone over Australia for 2002, 2003, 2004,
and all these years together



Figure 12. Total column ozone over Australia from 01.01.2002 until 31.12.2004

        Figure 11 shows TCO over Australia during 2002-2004 and Figure 12 – same data, but plotted for three years in a row.

        OSIRIS TCO over Australia varies between 230 and 300 DU. This is much lower than TCO levels over both Canada and Region #3 (except for the ozone hole period in Region #3). Even during the local spring, when the ozone amount should be largest, TCO over Australia is still less than 300 DU whereas it reaches 380 DU for Region #3 and 450 DU for Canada.
During the local summer in December-February (outdoor time) TCO over Australia is in the range from 240 to 270 DU. This is much lower compared to both Canada (270 to 330 DU) and Region #3 (260-300 DU).

        TCO over Australia did not change much from 2002 to 2003. However, it became 10-30 DU lower in September-December, 2004. This is again a very alarming sign that the ozone depletion in the Southern hemisphere may still be in progress!

Figure 13. Total column ozone over Region #4 for 2002, 2003, 2004,
and all these years together

Figure 14. Total column ozone for Region #4 from 01.01.2002 until 31.12.2004

        Figure 13 shows TCO over Region #4 (a mirror projection of Australia’s location over equator into the Northern hemisphere) during 2002-2004 and Figure 14 – same data, but plotted for three years in a row.

        OSIRIS TCO over Region #4 varies between 200 and 280 DU. Lowest values correspond to late winter and late fall periods, which are not present on the plots for Australia. As mentioned before, OSIRIS does not measure the ozone over the Southern hemisphere as much as over the northern. Therefore, it is more appropriate to analyze and compare the summer period only, which is available for both Australia and Region #4.

        During the local summer in June-August TCO over Region #4 TCO is in the range from 240 to 280 DU. This is about 10 DU higher compared to Australia’s local summer (240-270 DU), but 30-50 DU lower than in Canada (270-330 DU).

        Similar to Canada, TCO over Region #4 did not show any noticeable trend between 2002 and 2004. This could be a sign that the ozone in the Northern hemisphere is much more stable that in the Southern hemisphere.

5. CENTRAL AFRICA

Figure 15. Total column ozone over Central Africa for 2002, 2003, 2004,
and all these years together

Figure 16.Total column ozone over Central Africa from 01.01.2002 until 31.12.2004

        Figure 15 shows TCO over Central Africa during 2002-2004 and Figure 16 – same data, but plotted for three years in a row.
OSIRIS TCO over Central Africa varies from 240 to 300 DU in 2002 and from 230 to 280 DU in 2003-2004.

        The local summer in Central Africa, which is around equator, is hard to define. So, I use corresponding months for comparison with other locations during their local summer.

        In June-August TCO over Central Africa varied from 240 to 270 DU. This is very similar to Australia’s summer (240-270 DU) and about 10 DU lower than over Region #4 (240-280 DU).
Overall TCO over Central Africa did not show significant differences between 2002 and 2004, except for 1) December 2004, when TCO was up to 20 DU lower than in 2002 and 2003: 2) January-February 2002, when TCO was 10-20 DU lower than in 2002 and 2003.

Errors
        These results can potentially be affected by such errors as:

        1. Random errors in OSIRIS ozone retrievals. These errors may be up to 10% for each individual measurement. However, they should average out when daily average total column ozone values are calculated from the numerous individual profiles.

        2. Systematic errors in OSIRIS ozone retrievals. Although systematic errors are present at higher altitudes (30-48 km), they affect all retrievals in the same way and, therefore, should not influence comparisons between different countries.

        3. Variable tropopause height. The tropopause height varies from 12 km near the equator to 8 km near the poles. Therefore, OSIRIS stratospheric column ozone, which is always calculated from 10 to 48 km, can be underestimated over the equator and overestimated at very high latitudes (probably, for the northern part of Canada).

        4. Limited time scale. Only 3 years of ozone data from OSIRIS are long enough for comparisons between different countries. However, they are still too short to study the general ozone decrease or recovery. Continuing ozone decrease in the Southern hemisphere from OSIRIS data must, therefore, be compared with other ozone datasets.