We studied the influence of the semi-annual oscillation (SAO) on near-surface temperatures in Antarctica, using observations of 27 stations that were operational during (part of) the period 1957–79. For the annual cycle of surface pressure, the second harmonic explains 17–36% of the total variance on the Antarctic Plateau, 36–68% along the East Antarctic coast and almost 80% on the west coast of the Peninsula, and decreases further to the north. As a result of the amplification of the wave-3 structure of the circulation around Antarctica, a significant modification of the seasonal cooling is observed at many stations. The magnitude of this modification is largely determined by the strength of the temperature inversion at the surface: the percentage of the variance explained by the second harmonic of the annual temperature cycle is then largest on the Antarctic Plateau (11–18%), followed by the large ice shelves and coastal East Antarctica (6–12%) and stations at or close to the Peninsula (0–5%). A significant coupling between the half-yearly wave in surface pressure and that in surface temperature is found for coastal East Antarctica, which can be directly explained by the changes in meridional circulation brought about by the SAO. We show that the coupling of Antarctic temperatures to the meridional circulation is not only valid on the seasonal time scale of the SAO, but probably also on daily and interannual time scales. This has important implications for the interpretation of time series of Antarctic temperatures, a problem that will be addressed in part 2 of this paper.

Following a weakening of the semi-annual oscillation (SAO) since the mid-1970s, the half-yearly pressure wave in the Southern Hemisphere has become less significant. As a result, May/June temperatures have decreased in East Antarctica, which has moderated Antarctic warming. Spectral analysis of 87 years of pressure data at Orcadas suggest that the recent weakening of the SAO is part of the natural variability of the Southern Hemisphere circulation on decadal timescales. We interpret the time series of composite Antarctic temperature in terms of the historical strengthening and weakening of the SAO. If the dominant oscillations that occurred in the past prove to be persistent, an accelerated East Antarctic warming trend is expected for the coming decades. There are indications that the strength of the SAO is linked to the Southern Oscillation, in the sense that warm phases of the Southern Oscillation coincide with strong westerlies, a weakly developed SAO and below-average temperatures in East Antarctica. Temperatures on the west coast of the Antarctic Peninsula show strongly deviant patterns, which can not be explained by the same mechanism that applies to East Antarctica.