Grant Bigg: Polar Mesocyclone variability and deep water formation

School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK

Abstract

This analysis re-examines the 2 year data set of polar mesocyclones compiled by Julie Harold from AVHRR images (Harold et al., 1999a), and speculates on the link of mesocyclone number and distribution with changes in deep water formation within the northern sub-polar waters during the mid-1990s.

The AVHRR data set covers the two years from October 1993 - September 1995. Each image was checked for cloud signatures associated with mesocyclones over the NE Atlantic region. Measures of mesocyclone size, location, maximum cloud height and nature were recorded. The climatology is discussed in Harold et al. (1999a) and the mechanisms responsible for the observed distribution are considered by Harold et al. (1999b). Figure 1 shows the data region and the density of cyclogenesis (per 106 km2 sighted more than once over the two year period in each sub-area). Note the high number of systems in the Norwegian and Greenland Seas. Many of these are small (< 600 km diameter) mesocyclones within 1000 km of the sea-ice edge.
Figure 1 (see also below): Cyclogenesis density (no. per 106 km2 sighted more than once, October 1993 - September 1995).

Mesocyclones can be associated with large air-sea heat fluxes. It is this property which we shall invoke to suggest a link between mesocyclones and pre-conditioning of surface water for deep water formation. Two examples were discussed. One, on December 17-18, 1994, had air-sea heat fluxes in excess of 500 Wm-2 along the ice edge in the Greenland Sea according to the daily average NCEP Reanalysis for the 18th December. This case was one with a circle of 4 mesocyclones contained within the cold air behind a vigorous occlusion (Fig. 2). The Reanalysis in this instance showed the broad character of this system, but also some detail of the individual systems. In contrast, the mesocyclone in a similar region, off the ice-edge of SE Greenland rather than being associated with a synoptic system, on March 7 (Fig. 3), 1995 was not apparent at all on the NCEP Reanalysis for the day.
Figure 2(see also below): Infra-red image at 0800 GMT, Dec. 18, 1994 and
Figure 3(see also below): Infra-red image at 0300 GMT, March 7, 1995

Variation in mesocyclone number and distribution may therefore be important for the evolution of upper ocean properties through the winter. There is a seasonal dependence, with maxima in mesocyclone numbers over the NE Atlantic in the winter and spring. However, over the two years of the climatology data set the most striking feature is the interannual trend. Figure 4 shows a large and almost monotonic decrease in mesocyclone numbers over this period, with only around 15% as many systems in summer 1995 as in autumn 1993. This trend is not linked directly to the larger scale circulation of the North Atlantic as there is no correlation with the North Atlantic Oscillation (NAO). However, in the second year of the climatology there were only half as many mesocyclones in the cold air behind synoptic-scale (> 1000 km diameter) storms as in the first. Air-sea interaction therefore decreased, despite both winters of the climatology having similar NAO states.
Figure 4(see also below): Monthly number of mesocyclones and the NAO, October 1993 - September 1995

During this period the depth of oceanic convection in the Labrador Sea reached its deepest level of recent years (Lazier et al., 2002). In following years convection was much shallower as the climate of the Labrador Sea warmed, while convection became much deeper in the Greenland Sea (Wadhams et al., 2002) as this region cooled. The number of small mesocyclones, near the ice-edge in the Greenland Sea, decreased much less dramatically than the overall number in the 2 year data set. Combined with the change in large scale atmospheric circulation over the North Atlantic since the abrupt but temporary switch to a negative NAO index in the winter of 1995-6, this change in mesocyclone distribution may be linked to the recent change in oceanic convection regime. This highlights the need to extend the NE Atlantic climatology to pursue this potential air-sea linkage furthe

REFERENCES

Harold, J. M., G. R. Bigg and J. Turner, 1999a, Mesocyclone activity over the Northeast Atlantic Ocean. Part 1: vortex distribution and variability, Int. J. Climatol., 19: 1187-1204.
Harold, J. M., G. R. Bigg and J. Turner, 1999b, Mesocyclone activity over the Northeast Atlantic Ocean. Part 2: an investigation of causal mechanisms, Int. J. Climatol., 19: 1283-1299.
Lazier, J., R. Hendry, A. Clarke, I. Yashayaev, P. Rhines, 2002. Convection and restratification in the Labrador Sea, 1990-2000. Deep-Sea Res. Part I49, 1819-1835.
Wadhams, P., J. Holfort, E. Hansen, J. P. Wilkinson, 2002. A deep convective chimney in the winter Greenland Sea. Geophys. Res. Lett.29, art. no. 1434.


Figure 1 Cyclogenesis density (no. per 106 km2 sighted more than once, October 1993 - September 1995).
Figure 2 Infra-red image at 0800 GMT, Dec. 18, 1994
Figure 3 Infra-red image at 0300 GMT, March 7, 1995
Figure 4 Monthly number of mesocyclones and the NAO, October 1993 - September 1995