The winter atmospheric circulation over Australia and New Zealand, 1911-1993

Abstract

General circulation models predict greater changes will occur in winter than in other seasons under enhanced global warming. This thesis uses the Jones (1989,1991) gridded mean sea level pressure (MSLF) data set, for the period 1911-1993, to analyse the winter (JJA) circulation over Australia and New Zealand to see if any climate change has occurred. The mean winter circulation in this region is dominated by persistent anticyclonic activity over the Australian continent, and to the south and south-east of Australia and New Zealand by the Southern Westerlies and the circumpolar trough.

Evidence for climate change was found in winter, with the greatest changes occurring in July. An increase in the strength of the westerly winds was found over the south Tasman Sea. This has been caused by an increase in MSLF over Australia and a decrease in MSLF over the Southern Ocean. Using the gridded data, synoptic maps of the July and winter circulation were constructed and analysed to determine the mean latitude and longitude of the Australian Winter Anticyclone (A WA), and its longitudinal extent for the period 1911-1993. It was shown that the size and therefore mass of the A WA has increased, while there has been no long term change in its latitude.

These circulation changes are corroborated by significantly increased rainfall on the West Coast of the South Island of New Zealand, increased lake levels in Otago, and a stronger 850 hPa wind over Invercargill. Changes observed in the winter rainfall patterns over Australia are also consistent with the observed circulation changes.

The interannual fluctuations in the MSLF at Alice Springs (used as a good surrogate for the intensity of the A WA) are shown to be largely explained by the Southern Oscillation Index, and seasonal variations in the latitude of the anticyclones. Removing these short term variations, a residual trend for increased MSLF remains. This follows changes in the mean annual global temperature pattern, but with a lag of about 7 years. Three possible theories for this are discussed. It is concluded that the most probable is a change in the longitudinal temperature gradient across the southern Indian Ocean, where temperatures have warmed faster in the south-west compared to the south east. It is proposed that this has displaced the Indian Ocean anticyclone eastward, closer to Australia, with a significant shift occurring in the early 1940's. Some evidence for this is discussed.