Abstract
The aim of this paper is to study the association between the extratropical Southern Hemisphere and the decadal variability in the Pacific Ocean (PO). We discuss a pattern of coherent large‐scale anomalies and trends in cyclone and anticyclone behaviour in light of the climate variability in the PO over the ERA40 reanalysis period (1957–2002). The two representative PO indices are the Pacific Decadal and Interdecadal Oscillations (PDO and IPO), and here the PDO is chosen owing to it being less associated with the southern oscillation index (SOI). Composites of the indicators of the density and intensity of cyclones/anticyclones given by an automatic tracking scheme were calculated for the years when the PDOI was more than one standard deviation above or below its mean. Although the ERA40 is not free from noise and assimilation changes, the results show a large‐scale feature, which seems to be robust and agrees with earlier studies using different data sets. The sea‐level pressure shows a strong annular structure related to the PDO, which is not seen for the SOI, with lower pressure around Antarctica during the positive phase and vice versa. More intense (and fewer) cyclones and anticyclones were observed during the positive PDO. This is less consistent for the SOI, particularly during the summer when a different PDO/SOI pattern arises at high latitudes. The trends project a pattern coincident with the positive PDO phase and seem to be linked with the main climate shift in the late seventies. Trends observed over the Tasman Sea are consistent with declining winter rainfall over southeastern Australia. Most patterns are statistically significant and seem robust, but random changes in ENSO may play a part, to a certain degree, in modulating the results, and a physical mechanism of causality has not been demonstrated. Although global warming and related changes in the Southern Annular Mode (SAM) may also help explain the observed behaviour, the large‐scale response presented here provides a new insight and would be of considerable interest for further modelling studies. Copyright © 2007 Royal Meteorological Society