SPARC logo (45 Ko)
S P A R C

Stratospheric Processes And their Role in Climate
A project of the World Climate Research Programme

Home Initiatives Organisation Publications Meetings Acronyms and Abbreviations Useful Links

 

Long and short term variability in the dynamical attributes of the Arctic polar vortex and its effect on ozone

Hideaki NAKANE, National Institute for Environmental Studies, Tsukuba, Japan (nakane@nies.go.jp)

 

The meteorology of the Arctic polar vortex has been recognised as an important influence on the severity of ozone depletion in the Arctic region. The area (or volume) which is at temperatures below that of the formation of the Polar Stratospheric Clouds (PSC) and the duration of the period of the PSC formation control the amount of the active chlorine available to destroy ozone. Pawson and Naujokat [1999] showed the effect of decreasing polar temperature during the 1990s on active chlorine concentrations. The relationship between the temperature in the stratosphere and the quasi-biennial oscillation (QBO) and solar activity has been studied previously [e.g. Labitzke, 1987]. In this short report, trends and variability in intensity, duration, radius and stability of the Arctic polar vortex are reported [Nakane et al., 2000a]. In addition, we briefly report a preliminary estimation of the effect of the chlorine loading on the ozone depletion in the vortex based on ozone sonde observations over Japan under similar dynamical conditions but different chlorine loadings [Nakane et al., 2000b].

Long and short term variability in Arctic polar vortex characteristics

Maps of Ertel's potential vorticity (PV) distributions on isentropic surfaces are known to be useful for viewing the dynamical processes related to the polar vortex because PV behaves as a quasi-conserved tracer. PV can be calculated using temperature and wind fields from 3D global analysis data from ECMWF, NCEP/NCAR, etc. The edge of the polar vortex, the barrier to meridional transport, can be defined as the region where the gradient in PV is large. To quantify the intensity of this barrier, the first derivative of PV with respect to equivalent latitude fe is calculated, dPV/dfe [Nash et al., 1996]; the equivalent latitude is defined as the geometrical latitude enclosing the same area as the PV isoline. To eliminate the dependence on potential temperature and fe, the normalised gradient of PV is defined as NGPV=(dPV/dfe)/PV.

The NGPV as a function of equivalent latitude and time for the 39 winters from 1959 to 1997 in the Northern Hemisphere is shown in Figure 1. Here, "winter" refers to winter/spring and "1959" refers to 1958/1959. In each winter, the vortex onset, growth, and shrinking due to stratospheric warmings and dissipation can be seen in the maps of NGPV. Large NGPV corresponds to a strong vortex. The location of the meridional maximum in NGPV, femax, defines the edge of the polar vortex. Therefore, (90 - femax) is the radius of the polar vortex in units of degrees. Strong vortices with large NGPV were formed in 1976, 1989, 1990, 1993, 1995, 1996 and 1997. Stable vortices, characterised by smaller fluctuations in radius of the vortex, were in 1967, 1976, 1982 and 1997. The strong stratospheric warming in February 1963 is clearly identified by the sudden shrinking of the vortex. In most cases, polar vortices persisted after disappearance of PSCs, and low ozone concentrations continued as long as the vortices were stable.

In Figure 1, positive trends in the strength and stability of the polar vortex can be qualitatively seen. A quantitative analysis of the vortex characteristics [Nakane et al., 2000a] concluded that:

(1) There were positive trends in the strength, duration, radius and stability of the vortex.

(2) Short-term variability in the vortex indices was correlated with the QBO.

(3) A coupling of the 11-year solar cycle and the QBO generated long-term variability. Positive correlations were found between all four characteristics of the vortex (especially strong correlation in the case of vortex strength) and the QBO during solar inactive phases; during solar active phases, the correlations were weak or weakly negative.

Figure 1

Figure 1. The normalised gradient of potential vorticity (NGPV) at 475K as a function of equivalent latitude and time for the 39 winters from 1959 to 1997 from 45°N to 85°N calculated using the NCAR/NCEP re-analysis data.

A case study of the relative importance of chlorine loading and the dynamical condition of the Arctic polar vortex on ozone depletion

Vertical profiles of ozone have been measured at Moshiri (44°N, 142°E) in Hokkaido, Japan since 1996. Episodic low ozone events were found on April 14 and 23, 1996 when the station was inside the decaying Arctic polar vortex. Significant chemical ozone loss was found in the Arctic vortex in 1996 by the MATCH technique (Rex et al., 1997). The ozone profile obtained at Moshiri on April 23, 1996 was compared with the profile measured at Sapporo on April 13, 1972 when Moshiri and Sapporo (43°N, 121°E) were also thought to be inside the Arctic polar vortex (Figure 2). The dynamical conditions of the vortex in 1972 were similar to those of 1996 except that the vortex was a little weaker in 1972 (Figure 1) and the temperature inside the vortex was higher in 1972. The ozone profiles at Sapporo in 1972 and at Moshiri in 1996 were similar except between 150hPa and 20hPa; at around 70hPa, the ozone mixing ratio was 55% smaller in the 1996 Moshiri case compared to the 1972 Sapporo case (Figure 3).

Figure 2

Figure 2. The PV maps (q =475K) on April 13, 1972 (top) and on April 23, 1996 (bottom) calculated using the NCEP/NCAR reanalysis data. In 1972, Sapporo was covered by a part of the polar vortex, and in 1996 Moshiri was also covered by a part of the polar vortex just broken.

Figure 3

Figure 3. Vertical profiles of ozone and temperature observed by ozone sondes at Moshiri (44°N, 142°E) by NIES and STEL of the Nagoya University on April 23, 1996 (black ) and by Japan Meteorological Agency (JMA) at Sapporo (43°N, 121°E) on April, 13, 1972 (gray). The parts of the polar vortex covered the measurement sites in the middle and lower stratosphere in both cases.

To understand the effects of the differences in chlorine loading and in the dynamical features of the Arctic polar vortices on chemical ozone loss, preliminary photochemical multi-trajectory-box model simulations inside the vortex were carried out for both cases. The chemical box-model includes gas phase reactions, heterogeneous reactions and diabatic descent in the polar vortices. Thirty-six trajectories were placed inside the vortex and were run from December 1, 1995 to April 23, 1996 using the zonal, meridional and vertical winds from the ECMWF data. The diabatic descent rate was calculated by averaging the potential temperature levels of all trajectories, and its uncertainty was given by the standard deviation. The potential temperature of the airmass descended from 625K on 1 December, 1995 to 460K on 23 April, 1996. The same descent rate was assumed for the 1972 case because the behaviour of the polar vortex in 1972 was similar to that in 1996 (see Figure 1); the uncertainty in the descent rate seems to be larger than its variability. First, backward trajectories from Moshiri on April 23, 1996 and from Sapporo on April 13, 1972 were run and it was confirmed that both series of trajectories came from inside the polar vortices. Then, the forward trajectories were run, with the photochemical model, from December 1, 1971 to April 13, 1972 and from December 1, 1995 to April 23, 1996, with the descent rate estimated above. A natural chlorine loading of 0.8ppbv and a total chlorine loading of 1.34ppbv for 1972 and 3.35ppbv for 1996 were obtained from the Equivalent Effective Chlorine Loading (EESC) estimated in "Scientific Assessment of Ozone Depletion:1998" (Madronich and Velders, 1998) which represent the stratospheric chlorine loading below 25km in middle and high latitudes.

The model estimated ozone mixing ratio was 1.6ppmv on 23 April, 1996 while observations showed 1.4ppmv. For 1972, the calculated ozone mixing ratio was 3.1ppmv and the observed ratio was 3.0ppmv on 13 April. Thus, the agreement between the model and observation is good.

The calculated amounts of ozone depletion in the decaying Arctic polar vortex when it covered Sapporo and Moshiri are summarised in Table 1. It is concluded that the increase in chlorine loading is primarily responsible for the increase in ozone loss in the Arctic polar vortex from 1972 to 1996. Ozone loss would have changed from 2.0 to 0.7ppmv if the chlorine loading had been 1.34ppbv in the 1996 vortex. The results also suggest that ozone depletion should be considerably enhanced by strong and cold Arctic polar vortices under high chlorine loading. Such a scenario is quite likely since there are positive trends in vortex intensity, stability, duration and size as seen in Figure 1, and the chlorine loading is predicted to decrease slowly.

EESC (ppbv)
0.8 (natural abundance) 1.34 (1972) 3.35 (1996)
1972 vortex
0.35 0.5 1.3
1996 vortex
0.35 0.7 2.0

Bold values are the actual cases.

Table 1. Calculated ozone loss (ppmv) from December 1, 1971 to April 13, 1972 and from December 1, 1995 to April 23, 1996.

Acknowledgements

This work was carried out as a part of the project in the framework of the Global Environment Research Fund financed by the Environment Agency of Japan. The study on the Arctic polar vortices was done with M. Ninomiya, GEF and G. Bodeker, NIWA. The ozone sonde observations were carried out in co-operation with STEL of Nagoya University. The trajectory box model was developed in NIES by A. Lukyanov, CAO.

References

Madronich, S. and G.J.M. Velders, Halocarbon scenarios for the future ozone layer and related consequences, Scientific Assessment of Ozone Depletion: 1998. WMO Report N 44, 1998. Chapter 11.

Labitzke, K., Sunspots, the QBO, and the stratospheric temperature in the north polar region, Geophys. Res. Lett., 14, 535-537, 1987.

Nakane, H., M. Ninomiya and G. Bodeker, Trends and interannual variability in the Arctic polar vortex, to be submitted to Geophys. Res. Lett., 2000a.

Nakane, H., H. Akiyoshi, Y. Kondo, A. Lukyanov, V. Yushkov, V. Dorokhov, K. Saigo and M. Ninomiya, Effects of chlorine loading and Arctic polar vortex behaviors on vertical profiles of ozone in mid- and high-latitude region in the Northern Hemisphere, Extended abstracts of the Quadrennial Ozone Symposium 2000, in press, 2000b.

Nash, E. R., P. A. Newman, J. Rosenfield, and M. Schoeberl, An objective determination of the polar vortex using Ertel’s potential vorticity, J. Geophys. Res., 101, 9471-9478, 1996.

Pawson, S., and B. Naujokat, The cold winters of the middle 1990s in the northern low stratosphere, J. Geophys. Res., 104, 14,209-14,222, 1999.

Rex, M., et al., Prolonged stratospheric ozone loss in the 1995 - 96 Arctic winter, Nature, 389, 835 - 838, 1997.

 

Back to SPARC Newsletter 15 Homepage