The climate of Australia is a mix of tropical and extratropical influences. Northern Australia lies in the tropics and is strongly affected by the Australian monsoon circulation (Section 14.2.2) and ENSO (Section 14.4). Southern Australia extends into the extratropical westerly circulation and is also affected by the middle latitude storm track (Section 14.6.2), the SAM (Section 14.5.2), mid-latitude transient wave propagation, and remotely by the IOD (Section 14.3.3) and ENSO.
Eastern–northeastern Australian rainfall is strongly influenced by the ENSO cycle, with La Niña years typically associated with wet conditions and more frequent and intense tropical cyclones in summer, and El Niño years with drier than normal conditions, most notably in spring. The SAM plays a significant role in modulating southern Australian rainfall, the positive SAM being associated with generally above-normal rainfall during summer (Hendon et al., 2007; Thompson et al., 2011), but in winter with reduced rainfall, particularly in Southwest Western Australia (Hendon et al., 2007; Meneghini et al., 2007; Pezza et al., 2008; Risbey et al., 2009; Cai et al., 2011c). Rossby wavetrains induced by tropical convective anomalies associated with the IOD (Cai et al., 2009), and associated with ENSO through its coherence with the IOD (Cai et al., 2011b) also have a strong impact, leading to lower winter and spring rainfall particularly over Southeastern Australia during positive IOD and El Niño events. Along the eastern seaboard, ETCs (Section 14.6.2) exert a strong influence on the regional climate, while ENSO and other teleconnections play a lesser role (Risbey et al., 2009; Dowdy et al., 2012).
Significant trends have been observed in Australian rainfall over recent decades (Figure 14.25), varying vastly by region and season. Increasing summer rainfall and decreasing temperature trends over northwest Australia have raised the question of whether aerosols originating in the NH play a role (Rotstayn et al., 2007; Shi et al., 2008b; Smith et al., 2008; Rotstayn et al., 2009; Cai et al., 2011d), but there is no consensus at present. By contrast, a prominent rainfall decline has been experienced in austral winter over southwest Western Australia (Cai and Cowan, 2006; Bates et al., 2008) and in mid-to-late autumn over southeastern Australia (Murphy and Timbal, 2008). Over southwest Western Australia, the decrease in winter rainfall since the late 1960s of about 20% have led to an even bigger (~50%) drop in inflow into dams. The rainfall decline has been linked to changes in large-scale mean sea level pressure (Bates et al., 2008), shifts in synoptic systems (Hope et al., 2006), changes in baroclinicity (Frederiksen and Frederiksen, 2007), the SAM (Cai and Cowan, 2006; Meneghini et al., 2007), land cover changes (Timbal and Arblaster, 2006), anthropogenic forcing (Timbal et al., 2006), Indian Ocean warming (England et al., 2006) and teleconnection to Antarctic precipitation (van Ommen and Morgan, 2010).
Over southeastern Australia, the decreasing rainfall trend is largest in autumn with sustained declines during the drought of 1997–2009, especially in May (Cai and Cowan, 2008; Murphy and Timbal, 2008; Cai et al., 2012a). The exact causes remain contentious, and for the decrease in May, may include ENSO variability and long-term Indian Ocean warming (Cai and Cowan, 2008; Ummenhofer et al., 2009b), a weakening of the subtropical storm track due to decreasing baroclinic instability of the subtropical jet (Frederiksen et al., 2010; Frederiksen et al., 2011a, 2011b) and a poleward shift the ocean–atmosphere circulation (Smith and Timbal, 2012; Cai and Cowan, 2013). The well-documented poleward expansion of the subtropical dry zone (Seidel et al., 2008; Johanson and Fu, 2009; Lucas et al., 2012), particularly in April and May, is shown to account for much of the April–May reduction (Cai et al., 2012a). Rainfall trends over southeastern Australia in spring, far weaker but with a signature in the subtropical ridge (Cai et al., 2011a; Timbal and Drosdowsky, 2012), have been shown to be linked with trends and variability in the IOD (Cai et al., 2009; Ummenhofer et al., 2009b). Antarctic proxy data that capture both eastern Australian rainfall and ENSO variability (Vance et al., 2012) show a predominance of El Niño/drier conditions in the 20th century than was the average over the last millennium.
On seasonal to decadal time scales, New Zealand precipitation is modulated by the SAM (Kidston et al., 2009; Thompson et al., 2011), ENSO (Kidson and Renwick, 2002; Ummenhofer and England, 2007) and the IPO (Griffiths, 2007). Increased westerly flow across New Zealand, associated with negative SAM and with El Niño events, leads to increased rainfall and generally lower than normal temperatures in western regions. The positive SAM and La Niña conditions are generally associated with increased rainfall in the north and east of the country, and warmer than normal conditions. On longer time scales, a drying trend since 1979 across much of New Zealand during austral summer is consistent with recent trends in the SAM and to a lesser extent ENSO and the IPO (Griffiths, 2007; Ummenhofer et al., 2009a). In western regions, however, the drying is accompanied by a trend towards increased heavy rainfall (Griffiths, 2007). Temperatures over New Zealand have risen by just under 1°C over the past century (Dean and Stott, 2009). The upward trend has been modulated by an increase in the frequency of cool southerly wind flows over the country since the 1950s, without which the observed warming is consistent with largescale anthropogenic forcing (Dean and Stott, 2009).
A recent analysis (Irving et al., 2012; their Figure 9) shows that climate projections over Australia using CMIP5 models, which generally simulate the climate of Australia well (Watterson et al., 2013), are highly consistent with existing CMIP3-derived projections. The projected changes include a further 1.0 to 5.0°C temperature rise by the year 2070 (relative to 1990); a long-term drying over southern areas during winter, particularly in the southwest (Figure 14.27), that is consistent with an upward trend of the SAM (Pitman and Perkins, 2008; Shi et al., 2008a; Cai et al., 2011c); a long-term rainfall decline over southern and eastern areas during spring, in part consistent with a upward trend of the IOD index (Smith and Chandler, 2010; Zheng et al., 2010; Weller and Cai, 2013; Zheng et al., 2013). Precipitation change in northeast Australia remains uncertain (Moise et al., 2012), related to the lack of consensus over how ENSO may change (Collins et al., 2010; Section 14.4). In terms of climate extremes, more frequent hot days and nights and less frequent cold days and nights are projected (Alexander and Arblaster, 2009). Changes in the intensity and frequency of extreme rainfall events generally follow the mean rainfall change (Kharin et al., 2007), although there is an increase in most regions in the intensity of short duration extremes (e.g., Alexander and Arblaster, 2009).
For New Zealand, future climate projections suggest further increases in the westerlies in winter and spring, though model biases in jet latitude in the present climate reduce confidence in the detail of future projections (Barnes et al., 2010). The influence of poleward expansion of the subtropical high-pressure belt is projected to lead to drier conditions in parts of the country (Figure 14.27; Table 14.1), and a decrease in westerly wind strength in northern regions. Such projections imply increased seasonality of rainfall in many regions of New Zealand (Reisinger et al., 2010). Both flood and drought occurrence is projected to approximately double over New Zealand during the 21st century, under the SRES A1B scenario. Temperatures are projected to rise at about 70% of the global rate, because of the buffering effect of the oceans around New Zealand. Temperature rises are projected to be smallest in spring (SON) while the season of greatest warming varies by region around the country. Continued decreases in frost frequency, and increases in the frequency of high-temperature extremes, are expected, but have not been quantified (Reisinger et al., 2010).
In summary, based on understanding of recent trends and on CMIP5 results, it is likely that cool season precipitation will decrease over southern Australia associated in part with trends in the SAM, the IOD and a poleward shift and expansion of the subtropical dry zone. It is very likely that Australia will continue to warm through the 21st century, at a rate similar to the global land surface mean. The frequency of very warm days is very likely to increase through this century, across the whole country.
It is very likely that temperatures will continue to rise over New Zealand. Precipitation is likely to increase in western regions in winter and spring, but the magnitude of change is likely to remain comparable to that of natural climate variability through the rest of the century. In summer and autumn, it is as likely as not that precipitation amounts will change.