Fig. 1.
Seasonal changes of the total sunlit area poleward of 60°N and corresponding area of the region with non-zero total ozone values in the version 2.7 NIWA database.
Fig. 2.
Proxies used in the statistical modelling of year-to-year total ozone variability: (a) Effective Equivalent Stratospheric Chlorine-EESC, (b) total number of days with the minimum temperature in the 50–90°N region <195 K (the threshold for the potential existence of PSCs), (c) eddy heat flux at 50 hPa averaged over 45–75°N region in the period 1 December–15 April of the next year, (d) the Arctic Oscillation Index. The horizontal lines show the 1979–2011 mean values.
Fig. 3.
The area of the polar (ϕ > 60°N) region with the total ozone deficit exceeding 30% of the 1979–2010 daily mean values in the period 1 January 2011–31 May 2011 from the original (Fig. 3a) and corrected (Fig. 3b) area data. The time series of the yearly maximum of the deficit area in the 1979–2011 period based on original (Fig. 3c) and corrected (Fig. 3d) area data. Straight lines (Fig. 3c, d) show an ordinary least-squares linear fit to the data subset for years with the large maximum deficit area (>3.5×106 km2) and to the remaining data for years with the low maximum of the deficit area.
Fig. 4.
Size of the polar (ϕ > 60°N) region with extremely low ozone mixing ratios, i.e., with the deficit larger than 30% of the long-term mean, from measurements by the Solar Backscatter Ultraviolet Radiometer-2 (SBUV/2) onboard of the NOAA satellites in the period 2005–2011 at various levels: (a) 50 hPa, (b) 30 hPa, (c) 10 hPa, and (d) 1 hPa.
Fig. 5.
Performance of the statistical model: (a) observed (version 7.2 NIWA total ozone) versus modelled values of the area-weighted total ozone averaged over the period 15 February–15 April, (b) corresponding time series (1979–2011) of the observed (full circles and solid curve) and modelled (open circles and dashed curve) late winter/early spring total ozone values. The curves represent the data smoothed by the LOWES regression.
Fig. 6.
Summer means (1 June–31 August) of the area weighted polar (ϕ > 60°N) total ozone versus the preceding March total ozone mean.
