4.2.2.3 Southern Hemisphere

Outside of Antarctica (see Section 4.6), very little land area in the SH experiences snow cover. Long-term records of snow cover, snowfall, snow depth or SWE are scarce. In some cases, proxies for snow line can be used, but the quality of data is much lower than for most NH areas.

4.2.2.3.1 South America

Estimates from microwave satellite observations for mid-latitude alpine regions of South America for the period of record 1979 to 2002 show substantial interannual variability with little or no long-term trend. A long-term increasing trend in the number of snow days was found in the eastern side of the central Andes region (33°S) from 1885 to 1996, derived from newspaper reports of Mendoza City (Prieto et al., 2001).

Other approaches suggest some response of snow line to warming in South America. The 0°C isotherm altitude (ZIA), an indication of snow line, has been derived from the daily temperature profile obtained from radiosonde data located at Quintero (32°47’S, 71°33’W, 8 m above sea level; Carrasco et al., 2005), which represents the snow line behaviour in the western Andes from about 30°S to 36°S. Over the 1975 to 2001 period of record, the linear change in winter ZIA was 121.9 ± 7.7 m, and the positive trend was dominated by atmospheric conditions on dry days (enhancing melt) with no trend on wet days (accumulation zone unchanged).

4.2.2.3.2 Australia and New Zealand

For the mountainous south-eastern area of Australia, studies of late winter (August–September) snow depth have shown some significant declines (as much as 40%) since 1962. Trends in maximum snow depth were more modest. The stronger declines in late winter are attributed to spring season warming, while maximum snow depth is largely determined by winter precipitation, which has declined only slightly (Hennessy et al., 2003; Nicholls, 2005).

In New Zealand, annual observations of end-of-summer snow line on 47 glaciers have been made by airplane since 1977, and reveal large interannual variability primarily associated with atmospheric circulation anomalies (Clare et al., 2002); it is noteworthy, however, that the four years with highest snow line occurred in the 1990s. The only study of seasonal snow cover in the Southern Alps found no trend over the 1930 to 1985 period (Fitzharris and Garr, 1995) and has not been updated.