Ozone and UV Radiation
G. Bernhard1, G. Manney2,3, V. Fioletov4, J.-U. Grooß5, A. Heikkilä6, B. Johnsen7,
T. Koskela6, K. Lakkala8, R. Müller5, C. Lund Myhre9, M. Rex10
1Biospherical Instruments, San Diego, CA
2Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA
3New Mexico Institute of Mining and Technology, Socorro, NM
4Environment Canada, Toronto, Ontario, Canada
5Forschungszentrum Jülich, Jülich, Germany
6Finnish Meteorological Institute, Helsinki, Finland
7Norwegian Radiation Protection Authority, Østerås, Norway
8Finnish Meteorological Institute, Arctic Research Centre, Sodankylä, Finland
9Norwegian Institute for Air Research, Kjeller, Norway
10Alfred Wegener Institute for Polar and Marine Research, Potsdam, Germany
November 7, 2012
- Temperatures in the Arctic stratosphere in early December 2011 were among the lowest on record. Strong dynamical activity in late December 2011 and January 2012 caused the temperature to rise rapidly and led to conditions unfavorable to sustaining chemical ozone loss.
- Ozone concentrations in the Arctic stratosphere and UV radiation levels at Arctic and sub-Arctic locations during the spring of 2012 were generally within the range of values observed during the first decade of this century.
- Below-average ozone concentrations at several sites in southern Scandinavia led to increases in the UV Index of about 12% during January, February and March of 2012.
Ozone molecules in the Earth's atmosphere greatly attenuate the part of the Sun's ultraviolet (UV) radiation that is harmful to life. Reductions in the atmospheric ozone amount will always lead to increased UV levels, but other factors such as the height of the Sun above the horizon, cloud cover and aerosols also play important roles. This essay compares ozone and UV radiation measurements performed in the Arctic in 2012 with historical records.
Stratospheric ozone concentrations measured during the spring of 2012 in the Arctic were, by and large, within the typical range observed during the first decade of this century. The 2012 ozone levels were considerably higher than those in the spring of 2011, when unprecedented chemical ozone losses occurred (Manney et al., 2011). The minimum total ozone column1 for March 2012, averaged over the "equivalent latitude2" band 63°-90° N, was 372 Dobson Units (DU3). The 2011 record-low was 302 DU (Fig. 1.8). The average for 2000-2010 is 359 DU, 13 DU below the value for 2012.
The monthly mean total ozone columns for February through May 2012 are compared with baseline data from the 1978-1988 period in Fig. 1.9. During February, total ozone was more than 30% below the baseline value at Svalbard. Regions with monthly mean ozone levels 10% and more below the historical reference encompassed the North Pole, the North Sea, northern Siberia, northern Greenland, Scandinavia, Iceland, the British Isles, Denmark, the Netherlands and northern Germany. Above-average ozone levels were observed over the Aleutian Islands in the north Pacific Ocean. In March, the area with total ozone 10% below the baseline was centered on the North Sea and extended towards southern Scandinavia, the British Isles, France and central Europe. Much of eastern Canada, the eastern United States and southern Alaska were also affected by below-average total ozone columns. In April, Arctic regions with lower-than-normal ozone included the northern part of Canada (Victoria Island) and southern Greenland. Extended areas with large deviations from the historical measurements were not observed in the Arctic during May and through the summer.
The above discussion refers to monthly mean values. Departures from the baseline (either up or down) were larger for individual days. For certain regions and days, the ozone layer was 30% thinner than the long term mean. Deviations exceeding -35% were observed in the southwestern part of Russia as late as the second half of April. Deviations above the reference tended to be smaller in magnitude and less frequent.
The distribution of total ozone column over the Arctic on 3 April of the years 1981 (a year with a long-lasting and cold Arctic vortex, and relatively low stratospheric chlorine concentrations), 2002 (long-lasting warm vortex, high total chlorine loading), 2011 (long-lasting cold vortex, high chlorine), and 2012 (warm vortex, high chlorine) are illustrated in Fig. 1.10. The figure emphasizes that chemical ozone loss resulting from chlorine activation is most effective in years when there is a long-lasting, cold vortex, such as 2011. Years with a warm vortex, such as 2002 and 2012, result in little ozone loss.
Chemical ozone loss
Arctic stratospheric temperatures in December 2011 were among the lowest on record but rose to near average temperatures after strong dynamical activity in late December. Low temperatures facilitate the formation of polar stratospheric clouds (PSC), which provide surfaces for heterogeneous reactions (i.e., reactions between gases and liquid or solid matter) that activate stratospheric chlorine. The activated chlorine, in turn, destroys ozone rapidly in catalytic cycles. Sudden stratospheric warming in January 2012 halted PSC formation and hence the activation of chlorine, and led to conditions that were unfavorable to sustaining chemical ozone loss. The temporal evolution of several parameters which are crucial for stratospheric ozone chemistry are illustrated in the following with data from the Microwave Limb Sounder (MLS) on the Aura satellite (Fig 1.11).
Temperatures below the threshold temperature for PSC occurrence and chlorine activation of about 196 K (-77°C) existed locally until late January 2012 (Fig. 1.11a). When PSCs are formed, gas-phase nitric acid (HNO3) molecules occurring in the stratosphere are partly converted to solid particles such as nitric acid dihydrate (NAD), which then activate chlorine. The formation of PSCs at the beginning of the 2011/2012 winter is indicated by the large decrease of gaseous nitric acid in early December 2011 (Fig. 1.11b). The conversion of chlorine from inactive "reservoir chemicals" such as hydrogen chloride (HCl) to active forms such as chlorine monoxide (ClO) commenced at about the same time and is indicated by the decrease in HCl (Fig. 1.11c) and the increase in ClO (Fig. 1.11d). Of note, active chlorine in the form of ClO occurs only in sunlit regions of the vortex. Hence the decrease in HCl appears larger and earlier than the increase in ClO.
ClO is the primary ozone-destroying form of chlorine, so its presence is a sign for the potential for chemical ozone destruction. However, the catalytic cycles that enable ClO to destroy large amounts of ozone also require sunlight. So, even with chlorine activated, ozone destruction is typically small in January when much of the Arctic is still dark. The small drop in ozone in late January 2012 indicated in Fig. 1.11e suggests that a small amount of ozone was destroyed when the polar vortex was positioned so that substantial portions of it received sunlight.
In late January, a very strong and prolonged "stratospheric sudden warming" (SSW) event resulted in temperatures rising above the threshold below which chlorine can be activated, and chlorine was thus converted back to inactive forms by mid-February. No further ozone destruction occurred, and ozone increased slightly through mid-March as vertical motions transported higher ozone down from above.
SSWs are a common dynamical event in the Arctic winter, during which the strong westerly winds that encircle the polar vortex reverse to easterly and polar stratospheric temperatures rise abruptly, sometimes increasing by more than 30 K over 2-3 days. Such events have historically occurred on average about once every two winters, but are irregular, with periods of many years without one occurring (e.g., in most of the 1990s) and other periods such as the past decade having many more than average. The contrast between the meteorological conditions in 2011/2012 with those in 2010/2011 highlights the large range of interannual variability in Arctic winter conditions, and hence in Arctic ozone loss.
UV levels measured at Arctic terrestrial locations during the first half of 2012 were generally within the typical range of values observed during the last two decades, with notable exceptions discussed below.
Figure 1.12 compares measurements of the UV Index for 12 Arctic and sub-Arctic sites performed in 2012 and 2011 with historical measurements. The UV Index is a measure of the ability of UV radiation to cause erythema (sunburn) in human skin. It is calculated by weighting UV spectra with the CIE action spectrum for erythema (McKinlay and Diffey, 1987) and multiplying the result by 40 m2/W. Changes in the UV Index tend to anti-correlate with changes in total ozone. This can be seen by comparing the center panels of Fig. 1.12, which show UV Index measurements of 2011 and 2012 relative to the climatological average, with the bottom panels, which show a similar analysis for total ozone. The anti-correlation is most obvious for periods not affected by clouds.
Closer inspection of Fig. 1.12 reveals that total ozone at the southern Scandinavian sites (last row of Fig. 1.12) was significantly below the long-term mean for much of January, February and March 2012, consistent with the ozone maps shown in Fig. 1.9. On average, ozone was reduced by 16% over Trondheim, 11% over Finse, 6% over Jokioinen and 10% over Østerås. These reductions led to increases of the UV Index by 11% at Finse and 13% at Østerås. UV levels at Trondheim and Jokioinen were not notably affected, likely because of the dominance of cloud effects. While it is unusual that total ozone remains below the climatological average for three consecutive months, reductions for individual days remained, by and large, within historical limits.
Total ozone at Barrow, Alaska, was 257 DU on 9 June 2012 and 267 DU on 10 June 2012. The long-term mean for the two days is 345 DU and the standard deviation of the year-to-year variability is 23 DU. Thus, total ozone on these two days was 3.8 and 3.4 standard deviations below the climatology. Satellite images (e.g., http://www.temis.nl/protocols/o3field/o3month_omi.php?Year=2012&Month=06&View=np) indicate that the low-ozone event was caused by advection of ozone-poor air from lower latitudes originating from above the United States. The transport of ozone-poor air from lower to higher latitudes is well documented (e.g., Bojkov and Balis, 2001), but advection from sub-tropical to polar latitudes is less common. As a consequence of low ozone, the UV Index at Barrow on 10 June 2012 was 40% above the mean value for this day.
Figure 1.12 also highlights measurements made in 2011. The abnormally low stratospheric ozone concentrations in spring of that year (see Arctic Report Card 2011) led to large increases in UV radiation during March and April. These increases were considerably larger than any enhancement observed in 2012.
In addition to atmospheric ozone concentrations, UV radiation is affected by the height of the Sun above the horizon, clouds, aerosols (liquid and solid particles suspended in air), the reflectivity of the surface (high, when snow or ice covered), and other factors (Weatherhead et al., 2005). The main driver of the annual cycle is the solar elevation. Sites closest to the North Pole (Alert, Eureka and Ny-Alesund in Fig. 1.12) have the smallest peak radiation, with UV Index remaining below 4 all year. Although UV Indices below 5 are considered "low" or "moderate" (WHO, 2002), people involved in certain outdoor activities may receive higher-than-expected UV doses if their faces and eyes are oriented perpendicular to the low Sun or if they are exposed to UV radiation reflected off snow.
Clouds lead to a large variability in UV levels on time scales from minutes to days, but their effect is largely reduced when the ground is covered by fresh snow (Bernhard et al., 2008). Measurements at Barrow, and to a lesser extent at Alert and Eureka, show a large asymmetry between spring (low variability) and fall (high variability) because the surface at these sites is covered by snow until about June and free of snow thereafter until the beginning of winter. During summer and fall, the variability introduced by clouds is substantially larger than that related to ozone variations (compare shaded ranges in center and bottom panels of Fig. 1.12).
1Total ozone column is the height of a hypothetical layer that would result if all ozone molecules in a vertical column above the Earth's surface were brought to standard pressure (1013.25 hPa) and temperature (273.15 K).
2Equivalent latitude is a latitude-like coordinate aligned with the polar vortex (Butchart and Remsberg, 1986).
3Dobson Unit, the standard unit for measuring the total ozone column. 1 DU equals a column height of 0.01 mm and corresponds to 2.69x1016 molecules / cm2.
Bernhard, G., C. R. Booth and J. C. Ehramjian. 2008. Comparison of UV irradiance measurements at Summit, Greenland; Barrow, Alaska; and South Pole, Antarctica. Atmos. Chem. Phys., 8, 4799-4810. Available at: http://www.atmos-chem-phys.net/8/4799/2008/acp-8-4799-2008.html.
Bojkov, R. D. and D. S. Balis. 2001. Characteristics of episodes with extremely low ozone values in the northern middle latitudes 1957-2000. Ann. Geophys., 19, 797-807. Available at: http://www.ann-geophys.net/19/797/2001/angeo-19-797-2001.html.
Butchart, N. and E. E. Remsberg. 1986. The area of the stratospheric polar vortex as a diagnostic for tracer transport on an isentropic surface. J. Atmos. Sci., 43, 1319-1339.
Manney, G. L., M. L. Santee, M. Rex., N. J. Livesey, M. C. Pitts, P. Veefkind, E. R. Nash, I. Wohltmann, R. Lehmann, L. Froidevaux, L. R. Poole, M. R. Schoeberl, D. P. Haffner, J. Davies, V. Dorokhov, H. Gernandt, B. Johnson, R. Kivi, E. Kyrö, N. Larsen, P. F. Levelt, A. Makshtas, C. T. McElroy, H. Nakajima, M. C. Parrondo, D. W. Tarasick, P. von der Gathen, K. A. Walker and N. S. Zinoviev. 2011. Unprecedented Arctic ozone loss in 2011 echoed the Antarctic ozone hole. Nature, 478, 469-475. Available at: http://www.nature.com/nature/journal/v478/n7370/full/nature10556.html.
McKinlay, A. F. and B. L. Diffey. 1987. A reference action spectrum for ultraviolet induced erythema in human skin. CIE Res. Note, 6(1), 17- 22.
Müller, R., J.-U. Grooß, C. Lemmen, D. Heinze, M. Dameris and G. Bodeker. 2008. Simple measures of ozone depletion in the polar stratosphere. Atmos. Chem. Phys., 8, 251-264. Available at: http://www.atmos-chem-phys.org/8/251/2008/acp-8-251-2008.pdf.
Rex, M., R. J. Salawitch, P. von der Gathen, N. R. P. Harris, M. P. Chipperfield and B. Naujokat. 2004. Arctic ozone loss and climate change. Geophys. Res. Lett., 31, L04116, doi:10.1029/2003GL018844.
Weatherhead B., A. Tanskanen and A. Stevermer. 2005. Ozone and Ultraviolet Radiation, Chapter 5 in Arctic Climate Impact Assessment, 1042 pp, Cambridge University Press, New York. Available at: http://www.acia.uaf.edu/PDFs/ACIA_Science_Chapters_Final/ACIA_Ch05_Final.pdf.
WHO. 2002. Global Solar UV Index: A Practical Guide, 28 pp., published by World Health Organization (WHO), World Meteorological Organization (WMO), United Nations Environment Programme (UNEP) and the International Commission on Non-Ionizing Radiation Protection (ICNIRP), WHO, Geneva, Switzerland, ISBN 9241590076. Available at: http://www.who.int/uv/publications/en/GlobalUVI.pdf.