• Ansell, T. J., , C. J. C. Reason, , I. N. Smith, , and K. Keay, 2000: Evidence for decadal variability in southern Australian rainfall and relationships with regional pressure and sea surface temperature. Int. J. Climatol., 20 , 11131129.

    • Search Google Scholar
    • Export Citation
  • Arblaster, J. M., , and G. A. Meehl, 2006: Contributions of external forcings to southern annular mode trends. J. Climate, 19 , 28962905.

    • Search Google Scholar
    • Export Citation
  • Cai, W. J., , P. H. Whetton, , and D. J. Karoly, 2003: The response of the Antarctic Oscillation to increasing and stabilized atmospheric CO2. J. Climate, 16 , 15251538.

    • Search Google Scholar
    • Export Citation
  • Fyfe, J. C., , G. J. Boer, , and G. M. Flato, 1999: The Arctic and Antarctic Oscillations and their projected changes under global warming. Geophys. Res. Lett., 26 , 16011604.

    • Search Google Scholar
    • Export Citation
  • Gillett, N. P., , and D. W. J. Thompson, 2003: Simulation of recent Southern Hemisphere climate change. Science, 302 , 273275.

  • Gillett, N. P., , T. D. Kell, , and P. D. Jones, 2006: Regional climate impacts of the Southern Annular Mode. Geophys. Res. Lett., 33 .L23704, doi:10.1029/2006GL027721.

    • Search Google Scholar
    • Export Citation
  • Hall, A., , and M. Visbeck, 2002: Synchronous variability in the Southern Hemisphere atmosphere, sea ice, and ocean resulting from the annular mode. J. Climate, 15 , 30433057.

    • Search Google Scholar
    • Export Citation
  • Haylock, M. R., and Coauthors, 2006: Trends in total and extreme South American rainfall in 1960–2000 and links with sea surface temperature. J. Climate, 19 , 14901512.

    • Search Google Scholar
    • Export Citation
  • Hendon, H. H., , and B. Liebmann, 1990: A composite study of onset of the Australian summer monsoon. J. Atmos. Sci., 47 , 22272240.

  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Karoly, D. J., 1990: The role of transient eddies in the low-frequency zonal variations in the Southern Hemisphere circulation. Tellus, 42A , 4150.

    • Search Google Scholar
    • Export Citation
  • Karoly, D. J., , and K. Braganza, 2005: Attribution of recent temperature changes in the Australian region. J. Climate, 18 , 457464.

  • Kidson, J. W., 1988: Indices of the Southern Hemisphere zonal wind. J. Climate, 1 , 183194.

  • Kushner, P. J., , I. M. Held, , and T. L. Delworth, 2001: Southern Hemisphere atmospheric circulation response to global warming. J. Climate, 14 , 22382249.

    • Search Google Scholar
    • Export Citation
  • Lavery, B., , G. Joung, , and N. Nicholls, 1997: An extended high-quality historical rainfall dataset for Australia. Aust. Meteor. Mag., 46 , 2738.

    • Search Google Scholar
    • Export Citation
  • L’Heureux, M. L., , and D. W. J. Thompson, 2006: Observed relationships between the El Niño–Southern Oscillation and the extratropical zonal-mean circulation. J. Climate, 19 , 276287.

    • Search Google Scholar
    • Export Citation
  • Li, Y., , W. Cai, , and E. P. Campbell, 2005: Statistical modeling of extreme rainfall in southwest Western Australia. J. Climate, 18 , 852863.

    • Search Google Scholar
    • Export Citation
  • Lorenz, D. J., , and D. L. Hartmann, 2001: Eddy–zonal flow feedback in the Southern Hemisphere. J. Atmos. Sci., 58 , 33123327.

  • Lorenz, D. J., , and D. L. Hartmann, 2003: Eddy–zonal flow feedback in the Northern Hemisphere winter. J. Climate, 16 , 12121227.

  • Marshall, G. J., 2003: Trends in the Southern annular mode from observations and reanalyses. J. Climate, 16 , 41344143.

  • McBride, J. L., , and N. Nicholls, 1983: Seasonal relationships between Australian rainfall and the Southern Oscillation. Mon. Wea. Rev., 111 , 19982004.

    • Search Google Scholar
    • Export Citation
  • Meneghini, B., , I. Simmonds, , and I. N. Smith, 2007: Association between Australian rainfall and the Southern Annular Mode. Int. J. Climatol., 27 , 109121.

    • Search Google Scholar
    • Export Citation
  • Miller, R. L., , G. A. Schmidt, , and D. T. Shindell, 2006: Forced annular variations in the 20th century Intergovernmental Panel on Climate Change Fourth Assessment Report models. J. Geophys. Res., 111 .D18101, doi:10.1029/2005JD006323.

    • Search Google Scholar
    • Export Citation
  • Mills, G. A., , G. Weymouth, , J. Lorkin, , M. Manton, , E. Ebert, , J. Kelly, , D. Jones, , and G. deHoedt, 1997: A national objective rainfall analysis system. BMRC Techniques Development Rep. 1, Bureau of Meteorology, Melbourne, Australia, 30 pp.

  • Mo, K. C., , and G. H. White, 1985: Teleconnections in the Southern Hemisphere. Mon. Wea. Rev., 113 , 2237.

  • Nicholls, N., 2003: Continued anomalous warming in Australia. Geophys. Res. Lett., 30 .1370, doi:10.1029/2003GL017037.

  • Power, S., , F. Tseitkin, , S. Torok, , B. Lavery, , R. Dahni, , and B. McAvaney, 1998: Australian temperature, Australian rainfall and the Southern Oscillation, 1910–1992: Coherent variability and recent changes. Aust. Meteor. Mag., 47 , 85101.

    • Search Google Scholar
    • Export Citation
  • Reason, C. J. C., , and M. Rouault, 2005: Links between Antarctic Oscillation and winter rainfall over western South Africa. Geophys. Res. Lett., 32 .L07705, doi:10.1029/2005GL022419.

    • Search Google Scholar
    • Export Citation
  • Rogers, J. C., , and H. van Loon, 1982: Spatial variability of sea level pressure and 500 mb height anomalies over the Southern Hemisphere. Mon. Wea. Rev., 110 , 13751392.

    • Search Google Scholar
    • Export Citation
  • Shindell, D. T., , and G. A. Schmidt, 2004: Southern Hemisphere climate response to ozone changes and greenhouse gas increases. Geophys. Res. Lett., 31 .L18209, doi:10.1029/2004GL020724.

    • Search Google Scholar
    • Export Citation
  • Shindell, D. T., , R. L. Miller, , G. Schmidt, , and L. Pandolfo, 1999: Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature, 399 , 452455.

    • Search Google Scholar
    • Export Citation
  • Silvestri, G. E., , and C. S. Vera, 2003: Antarctic Oscillation signal on precipitation anomalies over southeastern South America. Geophys. Res. Lett., 30 .2115, doi:10.1029/2003GL018277.

    • Search Google Scholar
    • Export Citation
  • Smith, C. A., , and P. Sardeshmukh, 2000: The effect of ENSO on the intraseasonal variance of surface temperature in winter. Int. J. Climatol., 20 , 15431557.

    • Search Google Scholar
    • Export Citation
  • Smith, I. N., 2004: An assessment of recent trends in Australian rainfall. Aust. Meteor. Mag., 53 , 163173.

  • Thompson, D. W. J., , and J. M. Wallace, 2000: Annular modes in the extratropical circulation. Part I: Month-to-month variability. J. Climate, 13 , 10001016.

    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., , and S. Solomon, 2002: Interpretation of recent Southern Hemisphere climate change. Science, 296 , 895899.

  • Thompson, D. W. J., , M. P. Baldwin, , and S. Solomon, 2005: Stratosphere–troposphere coupling in the Southern Hemisphere. J. Atmos. Sci., 62 , 708715.

    • Search Google Scholar
    • Export Citation
  • Torok, S. J., , and N. Nicholls, 1996: A historical annual temperature data set for Australia. Aust. Meteor. Mag., 45 , 251260.

  • Trenberth, K. E., 1979: Interannual variability of the 500 mb zonal-mean flow in the Southern Hemisphere. Mon. Wea. Rev., 107 , 15151524.

    • Search Google Scholar
    • Export Citation
  • Trewin, B. C., , and A. C. F. Trevitt, 1996: The development of composite temperature records. Int. J. Climatol., 16 , 12271242.

  • Watterson, I. G., 2001: Zonal wind vacillation and its interaction with the ocean: Implications for interannual variability and predictability. J. Geophys. Res., 106 , 2396523976.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Composite daily 850-hPa winds [maximum vector (m s−1) shown to right of each panel] and sea level pressure for the (top) low, (middle) high, and (bottom) high-minus-low polarity of the daily SAM index in the December–February season, 1979–2005. The contour interval (CI) in the top two panels is 3 hPa, and in the bottom panel is 2 hPa. The number of days in each index polarity is indicated in the upper right of the top two panels. Vector winds are plotted heavy where they are deemed to significantly differ from 0 at the 90% level based on a t test. Positive (negative) contours of sea level pressure differences are solid (dashed) and the zero contour is heavy in bottom panel. The vector wind scale in the bottom panel is ½ that in the top two panels, but the maximum plotted vector is different as indicated.

  • View in gallery

    Same as in Fig. 1, but for the June–August season.

  • View in gallery

    Composite daily rainfall (contours and shading) and 850-hPa winds (maximum vector shown in lower left of each panel) for high–low polarity of the SAM index for March–May, June–August, September–November, and December–February. CI for rainfall differences is 0.5 mm day−1 with negative difference dashed. Differences that are deemed to be significantly different from zero based on a resampled Monte Carlo test are shaded. The number of days in the high and low index polarities of the SAM is listed in the upper right of each panel. The vector wind scale is the same in all panels, but the maximum plotted vector is different as indicated.

  • View in gallery

    Rate of the daily occurrence of easterly flow at 850 hPa for the (top) low, (middle) high, and (bottom) high-minus-low polarity of the daily SAM index in the December–February season 1979–2005. CI is 4%.

  • View in gallery

    Std dev of low-pass-filtered (7-day running mean) daily rainfall (mm day−1) for MAM, JJA, SON, and DJF for the period 1979–2005. Data-void regions are unshaded.

  • View in gallery

    Ratio of the frequency of occurrence of exceeding the highest weekly quintile rainfall accumulation in the high index polarity to the low index polarity of the SAM. Solid contours are for ratios 1.5:1, 2:1, 2.5:1, 3:1, and 3.5:1. Dashed contours are ratios 1:1.5, 1:2, 1:2.5, 1:3, and 1:3.5. Shading indicates regions where the ratio is significantly different than 1 based on a resampled Monte Carlo test. The number of days in the high and low index polarities of the SAM is indicated in the upper right of each panel.

  • View in gallery

    Composite daily maximum temperature differences (°C) between the high and low index polarities of the SAM. Differences are plotted solid (open) where they are deemed to be significantly different from 0 at the 95% (90%) level based on a t test. Positive (negative) differences are indicated by circles (triangles).

  • View in gallery

    Same as in Fig. 7, but for daily minimum temperature.

  • View in gallery

    Composite daily rainfall and 850-hPa wind differences between high and low index polarities of the SAM index during DJF in non-ENSO years. Plotting convention is same as in Fig. 3.

  • View in gallery

    (top) Observed rainfall trend for the DJF season 1979–2005 based. (bottom) Rainfall trend that can be attributed to the SAM. CI is 0.2 mm day−1 (27 yr)−1.

  • View in gallery

    Maximum temperature trend [°C (25 yr)−1] in DJF for (top) 1960–2005 and (middle) 1979–2005. (bottom) Maximum temperature trend that can be attributed to the SAM in the 1979–2005 period.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1022 1023 87
PDF Downloads 640 640 63

Australian Rainfall and Surface Temperature Variations Associated with the Southern Hemisphere Annular Mode

View More View Less
  • 1 Bureau of Meteorology Research Centre, Melbourne, Australia
  • | 2 Department of Atmospheric Science, Colorado State University, Fort Collins, Colorado
  • | 3 Bureau of Meteorology Research Centre, Melbourne, Australia
© Get Permissions
Full access

Abstract

Daily variations in Australian rainfall and surface temperature associated with the Southern Hemisphere annular mode (SAM) are documented using observations for the period 1979–2005. The high index polarity of the SAM is characterized by a poleward contraction of the midlatitude westerlies. During winter, the high index polarity of the SAM is associated with decreased daily rainfall over southeast and southwest Australia, but during summer it is associated with increased daily rainfall on the southern east coast of Australia and decreased rainfall in western Tasmania. Variations in the SAM explain up to ∼15% of the weekly rainfall variance in these regions, which is comparable to the variance accounted for by the El Niño–Southern Oscillation, especially during winter. The most widespread temperature anomalies associated with the SAM occur during the spring and summer seasons, when the high index polarity of the SAM is associated with anomalously low maximum temperature over most of central/eastern subtropical Australia. The regions of decreased maximum temperature are also associated with increased rainfall. Implications for recent trends in Australian rainfall and temperature are discussed.

Corresponding author address: Harry Hendon, Bureau of Meteorology Research Centre, GPO Box 1289, Melbourne 3001, Australia. Email: hhh@bom.gov.au

Abstract

Daily variations in Australian rainfall and surface temperature associated with the Southern Hemisphere annular mode (SAM) are documented using observations for the period 1979–2005. The high index polarity of the SAM is characterized by a poleward contraction of the midlatitude westerlies. During winter, the high index polarity of the SAM is associated with decreased daily rainfall over southeast and southwest Australia, but during summer it is associated with increased daily rainfall on the southern east coast of Australia and decreased rainfall in western Tasmania. Variations in the SAM explain up to ∼15% of the weekly rainfall variance in these regions, which is comparable to the variance accounted for by the El Niño–Southern Oscillation, especially during winter. The most widespread temperature anomalies associated with the SAM occur during the spring and summer seasons, when the high index polarity of the SAM is associated with anomalously low maximum temperature over most of central/eastern subtropical Australia. The regions of decreased maximum temperature are also associated with increased rainfall. Implications for recent trends in Australian rainfall and temperature are discussed.

Corresponding author address: Harry Hendon, Bureau of Meteorology Research Centre, GPO Box 1289, Melbourne 3001, Australia. Email: hhh@bom.gov.au

1. Introduction

The Northern and Southern Hemisphere annular modes play a prominent role in the climate of their respective hemispheres. Both modes are characterized by approximately zonally symmetric, equivalent barotropic seesaws in the strength of the zonal flow between ∼55°–60° and ∼35°–40° latitude. The structure of the Southern Hemisphere annular mode (SAM; also referred to as the Antarctic Oscillation or High Latitude Mode) is documented in, for example, Trenberth (1979), Rogers and van Loon (1982), Mo and White (1985), Kidson (1988), Karoly (1990) and Thompson and Wallace (2000). The strong similarity between the SAM and the Northern Hemisphere annular mode (NAM) is documented in Thompson and Wallace (2000).

The annular modes are naturally occurring patterns of variability in the climate system and have a typical decorrelation time scale of ∼2 weeks (Lorenz and Hartmann 2001, 2003). However, the annular modes also appear to be sensitive to increasing greenhouse gases in model simulations (e.g., Shindell et al. 1999; Fyfe et al. 1999; Kushner et al. 2001; Cai et al. 2003; Miller et al. 2006; Arblaster and Meehl 2006), and over the past few decades, the SAM has exhibited trends during austral summer that are consistent with forcing by the Antarctic ozone hole (Thompson and Solomon 2002; Gillett and Thompson 2003; Shindell and Schmidt 2004). As such, the climate impacts of the annular modes have implications not only for the current climate, but for the interpretation of climate change as well.

The climate impacts of the NAM have been extensively examined in numerous studies. However, relatively little work has been done on documenting regional climate variations associated with the SAM. This is partly explained because much of the inhabitable landmasses in the Southern Hemisphere are equatorward of the region where the SAM produces its largest changes in circulation. Nonetheless, rainfall variations associated with the SAM have been suggested, for instance, in southern South America (Silvestri and Vera 2003; Haylock et al. 2006), western South Africa (Reason and Rouault 2005) and southwestern Australia (Meneghini et al. 2007). Recent and projected climate change in southwestern Australia (decreased wintertime rainfall) has been attributed to a positive trend in the SAM (e.g., Ansell et al. 2000; Cai et al. 2003; Li et al. 2005). In many of these studies, decreased wintertime rainfall during the high polarity of the SAM has been attributed to a poleward shift of the midlatitude storm track.

The purpose of the present study is to comprehensively document the impact of the SAM on daily rainfall and surface temperature variations in station-based data throughout Australia. The results clarify the role of the SAM for variations of Australian climate and also provide a sounder basis for attribution of recent Australian climate change to observed trends in the SAM.

2. Data and analysis method

Our primary approach is to composite daily variations in rainfall and temperature during days corresponding to the high and low index polarities of the SAM. Daily data are used throughout the study in order to exploit the fact that the typical decorrelation time for the SAM is 1–2 weeks. Thus, use of daily data increases the sample sizes used in the analyses relative to composites based on, say, monthly mean data. The analyses are restricted to the period post-1979, for which assimilation of satellite data provides a reliable estimate of daily variations in the SAM in global reanalyses (e.g., Marshall 2003).

The SAM is defined using daily index values provided by the U.S. National Weather Service Climate Prediction Center (CPC; http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/aao/aao_index.html).

The CPC daily SAM index was constructed by projecting daily 700-hPa height anomalies onto the leading empirical orthogonal function (EOF) of monthly mean 700-hPa height poleward of 20°S. The height data are from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis (Kalnay et al. 1996). The EOF analysis was conducted on the base period 1979 to 2000. Before computing the EOFs, the seasonal cycle was removed from the data, and grid points were weighted by the square root of the cosine of latitude. The EOFs are based on the covariance matrix. We use daily values of the SAM index for the period January 1979–February 2005. By convention, the high index polarity of the SAM (positive index) is associated with anomalous westerly flow along ∼60°S juxtaposed against anomalous easterly flow along ∼40°S.

Daily Australian rainfall variations are assessed using an objective analysis (Mills et al. 1997) of daily gauge reports (Lavery et al. 1997) that span the entire country. The daily analyses are available on a 1° grid for the same period as the SAM index (January 1979–February 2005). Daily maximum and minimum surface temperature variations are assessed using approximately 100 high-quality stations that have nearly continuous records for the period 1979–2005 (Trewin and Trevitt 1996; Torok and Nicholls 1996). Circulation variations in the Australian region associated with the SAM are assessed with daily mean analyses of sea level pressure and winds at 850 hPa from the NCEP–NCAR reanalysis (Kalnay et al. 1996).

Composite anomalies are formed for each season: December–February (DJF), March–May (MAM), June–August (JJA), and September–November (SON). During all seasons, high and low index days are defined as days on which anomalies in the daily SAM index exceed one standard deviation in absolute value based on the full 1979–2005 daily record. The magnitude of the daily SAM index exceeds one standard deviation in absolute value about one-third of the time, hence for each season roughly 400 days (out of a total of ∼2400 days) correspond to the high and low index polarity, respectively. We examine all four seasons separately because 1) southwestern Australia receives the bulk of its rainfall during winter, but much of the rest of Australia receives significant rainfall in spring, summer, and autumn; and 2) the mechanisms for rainfall in southeast Australia vary from winter (primarily midlatitude frontal systems) to summer (e.g., moist northeasterly tropical intrusions associated with a poleward shift of the subtropical ridge and the development of the monsoon trough). Hence, the regional impacts of the SAM are expected to vary with season.

The significance of the composited anomalies is judged in two ways. For temperature and winds, which are more normally distributed than rainfall, the composite differences are estimated to be significantly different from zero at the 95% (90%) level using a t test applied to the difference of two means:
i1520-0442-20-11-2452-eq1
Here, X1 and X2 are the sample means for the high and low phases and
i1520-0442-20-11-2452-eq2
where s1 and s2 are the standard deviations of daily winds or temperatures for the high and low phases, and N1 and N2 are the sample sizes for each phase. Here, Neff1 and Neff2 are the effective sample sizes taking into account the autocorrelation of the daily SAM index. For temperature and wind data, N1 and N2 typically range 350–400 days, and NeffN/3.

For the composite rainfall differences, we adopt a Monte Carlo resampling technique whereby 500 synthetic realizations of the composite high-minus-low difference are generated from daily data that have been randomized while retaining the original redness of the time series. The randomization is done by successively shifting the start date of the SAM index and reversing its time sequence relative to the rainfall time series, and then recomputing the composites. The shifting is done 500 times in 4-day increments, and the composites are recalculated using data only from the desired seasons. The technique maintains the autocorrelation (redness) of both the daily SAM index and the rainfall data. We then sort the 500 samples from lowest to highest, based on the magnitude of the composite difference. The 50th highest composite magnitude is the threshold for the highest decile in this randomized sample. This threshold is then used as the baseline for the 90% significance level in the original composite. Note that we sort based on the magnitude of the composite anomaly and use the same threshold for determining the significance of a positive or negative anomaly. Thus our significance test makes no assumption about the sign of the expected anomaly, which is equivalent to applying a two-tailed test.

3. Rainfall and temperature variations

a. Composite circulation anomalies

To provide context for the rainfall and surface temperature variations associated with the SAM, we first examine the associated changes in the lower-tropospheric flow in the Australian region. Composite maps of winds at 850 hPa and sea level pressure for low index days, high index days, and their differences are shown in Fig. 1 for summer (DJF) and in Fig. 2 for winter (JJA). The total composites for the high and low index polarities are provided so that the differences (e.g., the high-minus-low index maps) can be interpreted within the context of the mean state.

To first order, the high and low index polarities of the SAM have opposite signed but otherwise identical climate impacts. Thus in the discussion that follows we will refer to the high-minus-low index composite differences as anomalous conditions during the high index polarity of the SAM (similarly, anomalous conditions associated with the low index polarity of the SAM can be approximated as the opposite of the high-minus-low index composite maps). The high-minus-low index composite maps correspond to conditions associated with a roughly 3 standard deviation change in the daily SAM index (i.e., the mean of the SAM index averaged over the high and low index composites is ∼+1.5 and ∼−1.5 standard deviations, respectively).

In both summer and winter, the high index polarity of the SAM is associated with anomalously high surface pressure centered at about 45°S and enhanced circumpolar westerlies poleward of 55°S (Figs. 1 and 2, bottom). At Australian latitudes (15°–40°S) the anomalous flow is predominantly easterly with a magnitude of ∼3–5 m s−1. The lower-tropospheric zonal wind anomalies extend and strengthen upward through the troposphere (not shown), indicative of the equivalent barotropic nature of the SAM. The composite analysis also captures zonal asymmetries in the pressure and wind fields associated with the SAM, with centers of maximum surface pressure occurring near 90°E and east of 180°. The midlatitude asymmetries inherent in the SAM are much less prominent than the asymmetries associated with the NAM (e.g., Thompson and Wallace 2000), but nevertheless contribute to longitudinal variations in the climate impacts of the SAM.

In winter (JJA; Fig. 2), when the climatological mean westerlies and subtropical ridge extend equatorward into subtropical central Australia, the anomalous easterlies associated with the high index polarity of the SAM cover most of the continent (Fig. 2, bottom). In contrast, during summer (DJF; Fig. 1), when the climatological mean subtropical ridge is contracted poleward and the monsoon trough is developed over central and northern Australia (e.g., Hendon and Liebmann 1990), the anomalous easterlies associated with the high index polarity of the SAM are confined to southern Australia (Fig. 1, bottom). Wind anomalies associated with the high index polarity of the SAM during spring (highlighted for the Australian region in Fig. 3, bottom left) are similar to summer, though the easterly anomalies across southern Australia have a weaker meridional component. The easterly anomalies during the high index polarity of the SAM in autumn (highlighted for the Australian region in Fig. 3, top left) exhibit a pronounced southerly component over the western half of the continent.

b. Rainfall anomalies

Figure 3 shows composites maps of rainfall for the high-minus-low index polarities of the SAM for all four seasons superposed on the attendant differences in the 850-hPa flow (i.e., the wind anomalies for DJF and JJA in Fig. 3 are reproduced from Figs. 1 and 2, respectively). In all seasons, significant anomalies are mainly confined to the southern half of the continent, but the contrast between the pattern of rainfall anomalies in autumn/winter (MAM/JJA) and spring/summer (SON/DJF) is striking. In autumn/winter (Fig. 3, top), the high polarity index of the SAM is associated with decreased rainfall in the extreme southwest, which has been alluded to in previous studies (e.g., Ansell et al. 2000; Meneghini et al. 2007). Decreased rainfall also occurs in the southeast to the west of the Australian Alps during winter. In contrast, during spring/summer (Fig. 3, bottom), the high index polarity of the SAM is associated with increased rainfall on the southern third of the east coast of the mainland and on the east coast of Tasmania. Decreased rainfall is also apparent on the west coast of Tasmania, especially during spring.

The negative rainfall anomalies in the southwest and in the southeast to the west of the Australian Alps in autumn–winter during high index polarity of the SAM are consistent with the accompanying anomalous easterly winds. The easterly anomalies correspond to a weakening of the climatological mean westerlies, and presumably act to reduce or weaken the westward progression of rain-producing synoptic weather systems that develop in the midlatitude westerlies. Similarly, the negative rainfall anomalies on the west coast of Tasmania in spring–summer, which are also accompanied by anomalous easterly winds, appear to stem from reduced upslope flow (weakened westerlies) on the western slopes of the generally north–south-oriented orography.

The increased rainfall on the southern third of the east coast of the mainland in spring and summer (Fig. 3, bottom) is consistent with an anomalous upslope source of moisture from the Tasman Sea. However, the mean winds are westerly along the southeast cost, even during high polarity of the SAM (e.g., Fig. 1). Thus, it remains to be demonstrated that the high polarity of the SAM is associated with an actual increase in upslope (easterly) conditions. We demonstrate this by computing the rate of occurrence of easterlies at 850 hPa in the low index polarity (Fig. 4, top), high index polarity (Fig. 4, middle), and high-minus-low difference (Fig. 4, bottom). Upslope easterlies occur about 10% of the time on the southeast coast during the low index polarity of the SAM, increasing to 30% of the time during the high index polarity of the SAM. Hence, wet conditions on the southeast coast during the high index polarity of the SAM are consistent with a twofold increase in the occurrence of upslope easterly conditions over that during the low index polarity. However, a more thorough analysis of the relationship between rain episodes and upslope flow on the east coast is warranted in order to fully understand the impact of the SAM.

The rainfall anomalies associated with variations of the SAM approach 2 mm day−1 in magnitude along the southeast coast in spring–summer and in the southwest and southeast during autumn–winter. The importance of these SAM-related rainfall anomalies can be assessed by comparing their amplitudes to the standard deviation of daily rainfall that has been low-pass filtered with application of a 7-day running mean (Fig. 5). The standard deviation of 7-day running mean, as opposed to raw daily data, is considered to account for the persistence of the SAM, (i.e., the daily rainfall anomalies shown in Fig. 3 are representative of conditions that persist for ∼1 week). Figure 5 indicates that the greatest rainfall variability occurs in northern Australia during the summer monsoon, but this is a region that experiences little significant impact from the SAM. The southeast experiences near-constant rainfall variability year-round of ∼2–3 mm day−1, while the southwest experiences a wintertime maximum in variability of ∼3 mm day−1.

Comparison of Figs. 3 and 5 indicates that the SAM anomalies along the southeastern coast in spring–summer and in the southwest and southeast during autumn–winter approach 1 standard deviation of the 7-day running mean rainfall. Thus, in regions where the SAM has its greatest impact, a ∼3 standard deviation anomaly of the SAM (as implied by the high-minus-low composites) is associated with a ∼1 standard deviation change in weekly rainfall. In terms of variance, the SAM accounts for up to ∼15% of the week-to-week rainfall variance in these regions.

Another way to quantify the impact of the SAM on precipitation is to determine its impact on the occurrence of significant rainfall events, defined here as an accumulation above a specified threshold. We consider weekly rainfall accumulation in the highest quintile, which we define based on observed daily rainfall for 1950–2005. Because of the relatively short record with which to form composites (1979–2005), we use weekly as opposed to daily accumulation and we consider the upper quintile as opposed to a more extreme threshold such as the upper decile in order to reduce noise and thus produce more reliable estimates of the changes in probabilities.

At every continental grid point an upper quintile for the weekly rainfall rate is established, which by definition is exceeded 20% of the time. The climatology of the upper quintile threshold (not shown) varies spatially and with season in a similar fashion as the weekly standard deviation (Fig. 5). As for the weekly standard deviation, the upper quintile threshold shows a general lack of seasonality in southeast Australia, maximum in the southwest during winter, and overall highest values in the north during summer. Threshold upper quintile weekly rainfall accumulations range from in excess of 100 mm in the north during summer to 25–55 mm in the southeast and southwest during winter (with maximum exceeding 70 mm on the west coast of Tasmania in winter).

The frequency of occurrence of days that exceed the weekly quintile rainfall rate for periods when the SAM is in the high-versus-low index polarities is shown in Fig. 6. Ratios greater (less) than one indicate an increased (decreased) likelihood of a significant weekly rainfall event when the SAM is in its high index polarity. The ratio is shaded only when it is significantly different from 1 at the 90% level. Significance is assessed using a resampled Monte Carlo test similar to that for the mean rainfall anomalies (Fig. 3). Here, we compute and sort 500 synthetic realizations of the ratio of the frequency of occurrence of the weekly rainfall accumulation in the upper quintile in the high-to-low polarity index of the SAM. The 25th highest and 25th lowest composite ratios are used as the thresholds for statistical significance of the high and low values of the actual composite ratios, respectively. This is equivalent to 90% statistical significance with a two-sided test.

For the most part, the impact of the SAM on the incidence of an upper quintile rainfall event is consistent with its impact on mean precipitation. During winter (Fig. 6, top right), the likelihood of a rainfall accumulation in the upper quintile is ∼1.5 times larger during the low index polarity in southeast Australia, and up to 2 times larger during the low index polarity in southwest Australia. These regions coincide with locations where the mean rainfall is increased (reduced) during the low (high) index polarity of the SAM (Fig. 3, top right). A region of modest increase in the incidence of an upper quintile event during the high index polarity is evident in southern Queensland/northern New South Wales.

During spring/summer (Fig. 6, bottom), the likelihood of an upper quintile event is up to 2 times greater in the high index polarity throughout much of southern Australia, with the largest difference of ∼3 observed on the southeast coast during summer. Interestingly, the region of increased probability extends into south-central Australia during spring, which is a region where the difference in mean rainfall is not as pronounced (Fig. 3, bottom). The mechanism for this increased occurrence of significant rainfall in south-central Australia is not known but is possibly related to an elevated occurrence of cutoff lows on the equatorward side of the enhanced ridge along ∼45°S (cf. Figs. 2 and 3) during the high phase of the SAM. During spring and summer, an upper quintile event is also twice as likely during the low index polarity of the SAM in western Tasmania, but more likely during the high index polarity in eastern Tasmania.

c. Temperature anomalies

Composites of daily maximum (Fig. 7) and minimum (Fig. 8) temperature for the high-minus-low index polarities of the SAM are made in a similar fashion as for rainfall. The greatest temperature anomalies occur during spring and summer, when much of the south-central and -eastern portions of the continent experience reduced maximum temperature during the high phase of the SAM (Fig. 7, bottom). These are regions that experience increased rainfall during the high phase of the SAM. In autumn (MAM; Fig. 7, top left) reduced maxima are confined to the western portion of the continent, where the flow anomalies associated with the SAM have a pronounced southerly component (i.e., implying cold advection) in this season (Fig. 3).

In general, the results for minimum temperature are weaker than their maximum temperature counterparts, but consistent with the distribution of rainfall anomalies. For instance, in spring and summer minimum temperature (Fig. 8, bottom) is reduced in the south-central portions of the continent, which are regions that experience increased rainfall and reduced maximum temperatures (Fig. 7) during the high index polarity of the SAM. Presumably, these reduced minimum temperatures follow from reduced maximums as a result of reduced daytime warming in regions of increased rainfall, which is associated with increased cloudiness (reduced insolation) and enhanced surface evaporation from moist soil (e.g., Power et al. 1998). During winter in the southeast and southwest, where rainfall is reduced during the high phase of the SAM, minimum temperatures are reduced (Fig. 8, top). If, as in summer, minimum temperatures are largely controlled by the follow-on effect from changes in the daytime maximum temperature as a result of variations in rainfall, then in these regions of reduced rainfall in winter we would expect increased minimum temperatures. But the regions in the southeast and southwest during winter show little signal in maximum temperature (Fig. 7, top), so the reduced minimum is not a result of reduced daytime warming. Rather, the reduced minimums presumably stem from enhanced clear-sky cooling in regions of reduced cloud cover associated with reduced rainfall. This clear-sky cooling effect on minimum temperature is only prominent if the effect of rainfall on daytime maximums is absent or removed (e.g., Power et al. 1998).

The changes in mean maximum and minimum temperature associated with the SAM are also accompanied by changes in the incidence of extreme temperature events. Here, we define an extreme event as a temperature exceeding or dropping below the climatological threshold for the highest or lowest decile. We focus on maximum temperature exceeding the upper decile in summer and minimum temperature dropping below the lower decile in winter. Selective stations across south-central Australia (from west to east) are summarized in Table 1 for summer, when this region experiences reduced maximum temperature during the high index polarity of the SAM. For instance, at Kalgoorlie in south-central Western Australia where the high-minus-low maximum temperature anomaly is –2.9°C, an extreme maximum temperature (>39.5°C) occurs in 45 out of 348 days in the low phase of the SAM but only in 23 out of 332 days in the high phase of the SAM. Thus, an extreme maximum temperature occurs about twice as often in the low phase as during the high phase of the SAM. Note that the decile threshold (39.5°C) is about 1.5 standard deviations greater than the mean (32.6°C), which is an equivalent method of defining extremes. Similarly, at Sydney on the southeast coast, the summertime maximum decile threshold of 30.0°C is exceeded twice as often during the low index (53 out of 348 days) as during the high index (22 out of 333 days) polarity of the SAM.

In winter, stations in the southwest and southeast that exhibit a reduction in minimum temperature in the high index polarity of the SAM (Fig. 8) also exhibit a higher rate of occurrence of extreme minimum temperature (Table 2). For instance at Wandering, which is in the important wheat belt of Western Australia, wintertime frost is almost 3 times less likely during the low index (17 out of 373 days) as during the high index (44 out of 410 days) polarity of the SAM. Similarly, in southeastern Australia, hard frost at Rutherglen (minimum temperatures less than –2.5°C) is almost twice as likely in the high index as compared to the low index of the SAM.

Note that the impact of the SAM on the incidence of extreme temperatures is consistent with the changes in the mean temperature (as indicated in Figs. 7 and 8) but may also reflect changes in the shape (e.g., variance and skewness) of the temperature frequency distribution at individual stations.

4. Discussion

a. Relationship with ENSO

Two issues regarding the regional impacts of the SAM in Australia need further consideration. The first is the impact of the El Niño–Southern Oscillation (ENSO) phenomenon on the composite anomalies developed in the previous section. The SAM is uncorrelated with ENSO (as measured by popular indices of eastern equatorial Pacific sea surface temperature such as the Niño-3.4 SST index) in autumn through spring (Table 3). However, during summer (DJF), the warm phase of the ENSO cycle is significantly associated with the low index polarity of the SAM (e.g., L’Heureux and Thompson 2006). To ensure that we have extracted the SAM signal in rainfall in summer and not just a residual of the ENSO signal, we recomputed the high–low rainfall composite by excluding the summers during El Niño (1982/83, 1986/87, 1991/92, 1997/98, and 2002/03) and La Niña [1988/89, 1994/95, and 2000/01; the dates are identified by Smith and Sardeshmukh (2000) and are updated online at http://www.cdc.noaa.gov/people/cathy.smith/best/#years]. The high–low rainfall and wind composite differences for summer in the non-ENSO years are remarkably similar to that when the ENSO years are included (i.e., compare Fig. 9 with the bottom-right panel in Fig. 3). This suggests that the results in this study are dominated by variations in the SAM and are not heavily contaminated by the impact of ENSO. However, this does not necessarily imply that we can separate the contributions of ENSO and the SAM to the summertime rainfall variations in the southeast.

b. Trends

The other issue that warrants attention is the role of the SAM in recent trends in Australian rainfall (e.g., Smith 2004) and temperature (e.g., Nicholls 2003). The SAM has exhibited a trend toward its high index polarity over the period 1979–2005, but the trend is restricted primarily to the summer and, to a lesser extent, autumn months (Thompson and Solomon 2002; Marshall 2003; Table 4). The positive trend in the SAM corresponds to an increase of about ∼½ of the daily standard deviation of the SAM index during summer for the period 1979–2005. The trend during autumn is a more modest increase of ∼⅓ of the daily standard deviation. As there has been no significant trend in the SAM during winter for this period, it is difficult to ascribe any observed wintertime rainfall or temperature trends to a trend in the SAM. This does not preclude, however, the possible contribution of the SAM to the wintertime rainfall decline in the southwest prior to 1979 (e.g., Smith 2004), which extends prior to the start of the analysis presented here.

On the other hand, the SAM has likely contributed to the observed summertime rainfall and temperature trend in the period 1979–2005, at least in the southern and central part of the country where the impacts of the SAM are most pronounced. The top panel in Fig. 10 shows the observed trend in rainfall during DJF for the period 1979–2005. The trend is dominated by wetter conditions in the tropical northwest, drier conditions in the northeastern coastal region, and increased rainfall extending into the southwest and southeast. Note that while the trend in the southern part of the country is small relative to that in the north, it nevertheless accounts for a comparable fraction of the total variance in their respective regions (i.e., cf. Fig. 10 to Fig. 5).

The bottom panel in Fig. 10 shows the contribution of the SAM to the observed trends. The contribution of the SAM to the summertime trends is found by first recomputing the high–low summertime composite using detrended data. We use detrended data in order to obtain an unbiased estimate of contribution of the SAM to rainfall variability, but the high–low summertime rainfall composite based on detrended data is nearly identical to that based on raw data (Fig. 3, bottom right). The SAM contribution to the trend (Fig. 10, bottom) is then estimated by 1) normalizing the detrended high–low composite by the composite high–low amplitude of the SAM index (=∼3), and then 2) scaling this normalized high–low detrended composite by the observed trend in SAM for the period 1979–2005. Only those locations where the SAM signal in daily rainfall is determined to be statistically significant at the 90% level, as in Fig. 3, are shaded. As expected, the SAM does not account for the large positive rainfall trend in the northern and central portions of the continent where there is little signal of the SAM in rainfall. However, the SAM does account for 50%–75% of the more modest positive trends in southeast Australia, and the east–west dipole in precipitation trends across Tasmania. A similar analysis for the autumn season suggests that the modest positive trend in the SAM accounts for a portion of the observed drying trend in southwest Australia (not shown).

The contribution of the SAM to trends in surface temperature is assessed in a similar fashion. We concentrate on maximum temperature just in summer because the SAM signature in maximum temperature in autumn and in minimum temperature in both summer and autumn is small. Australia as a whole has experienced a warming trend since the middle of the last century that has been attributed to greenhouse gas warming (e.g., Karoly and Braganza 2005). During summer, this warming is most pronounced in the eastern portion of the continent, where maximum temperature has increased at a rate of ∼1°C (25 yr)−1 for the period 1960–2005 (Fig. 11, top). During the more recent period considered in this study (1979–2005), much of north-central Australia has cooled (consistent with this being a region of increased rainfall but one that is unrelated to the SAM; Fig. 10) and the long-term warming in the southeast has largely been mitigated. This region of mitigated warming in the south and east coincides with where the SAM contribution to the trend (1979–2005) has a cooling of up to 0.5°C (25 yr)−1 (Fig. 11, bottom). Thus, it is tempting to infer that the recent upward trend in the SAM during summer has acted to mitigate some of the longer-term greenhouse gas warming across central-east Australia.

5. Conclusions

The high index polarity of the SAM is associated with a poleward contraction of the midlatitude storm track and thus easterly anomalies across much of southern and central Australia. During winter, the easterly anomalies across southern Australia are associated with decreased daily rainfall in the southwest and in the southeast to the west of the Australian Alps, which are regions that receive much of their wintertime rainfall from synoptic-scale disturbances in the midlatitude westerlies. During spring and summer, the easterly anomalies during the high index polarity of the SAM are associated with increased daily rainfall on the southeast coast of the mainland, which appears to result from an increased occurrence of moist upslope flow from the Tasman Sea. The SAM explains 10%–15% of the weekly rainfall variability in the southwest and southeast during winter and on the southeast coast during spring–summer, which is comparable to or larger than the variability in these regions associated with ENSO (e.g., McBride and Nicholls 1983). Thus, the SAM is an important contributor to rainfall variability even in regions where the ENSO signal is prominent.

Variations in the SAM also impact Australian surface temperatures. Maximum surface temperatures tend to be decreased (increased) in the regions of enhanced (decreased) rainfall. The strongest signals are in spring and summer across much of southern and eastern Australia, where maximum temperatures are reduced in regions of increased rainfall during the high index polarity of the SAM. The relationships between the SAM and daily minimum temperatures are weaker than their maximum temperature counterparts, but wintertime minimums tend to be decreased in regions of reduced rainfall, indicative of increased clear-sky cooling. The cooling of much of Australia during the high index polarity of the SAM is consistent with new calculations by Gillett et al. (2006) based on year-round station-based data.

The usefulness of the present results for deterministic prediction of rainfall and temperature is limited by both the signal strength of the climate impacts of the SAM and by the ability to predict the SAM in the first place. Skillful prediction of the SAM beyond ∼10 days has not yet been demonstrated. However, it is likely that the SAM is predictable on seasonal time scales via its apparent relationship to ENSO (L’Heureux and Thompson 2006). Additionally, there is growing evidence that low-frequency variations in the Southern Hemisphere stratosphere—driven either by internal atmospheric dynamics or polar ozone depletion—are dynamically linked to changes in the tropospheric flow that resemble the SAM (Thompson et al. 2005). Although the SAM is thought to interact with the Southern Ocean (e.g., Hall and Visbeck 2002) and that this interaction may increase the persistence of the SAM (Watterson 2001), useful enhancement of predictability arising from the interaction with the Southern Ocean has yet to be demonstrated.

The SAM has contributed to much of the observed increases in rainfall over southeastern Australia and eastern Tasmania and the decreases in rainfall over western Tasmania during the austral summer season for the period 1979–2005. However, there is little evidence that the SAM has contributed to seasonal precipitation changes over Australia during other seasons in the past 25 yr. It would be interesting to examine the possible impacts of the SAM on Australian climate changes that occurred prior to the start of our analysis in 1979, but such an analysis is hampered by the availability of reliable indices of the SAM in the presatellite era (e.g., Marshall 2003).

The SAM also appears to have mitigated some of the longer-term increase in temperature across Australia, which has been attributed to increasing greenhouse gases and other anthropogenic forcing (Karoly and Braganza 2005). In conjunction with the recent upward trend of the SAM during summer, maximum summertime temperatures across southern and eastern Australia have warmed less or even leveled relative to the longer-term upward trend. As much of the positive summertime trend in the SAM has been attributed to the Antarctic ozone hole (Thompson and Solomon 2002; Gillett and Thompson 2003), it would be interesting to see if 1) climate model simulations of recent climate change exhibit a mitigating effect of stratospheric ozone depletion on greenhouse gas warming in southern and eastern Australia during summer, and 2) the predicted future recovery of the Antarctic ozone hole minimizes the effect of the SAM on Australian temperature trends.

Acknowledgments

Comments on an earlier version of the manuscript by B. Timbal and P. Hope are appreciated. This work was initiated when DT visited BMRC in 2004. DT is grateful for support provided by BMRC for that visit and for funding provided by the National Science Foundation Climate Dynamics Program.

REFERENCES

  • Ansell, T. J., , C. J. C. Reason, , I. N. Smith, , and K. Keay, 2000: Evidence for decadal variability in southern Australian rainfall and relationships with regional pressure and sea surface temperature. Int. J. Climatol., 20 , 11131129.

    • Search Google Scholar
    • Export Citation
  • Arblaster, J. M., , and G. A. Meehl, 2006: Contributions of external forcings to southern annular mode trends. J. Climate, 19 , 28962905.

    • Search Google Scholar
    • Export Citation
  • Cai, W. J., , P. H. Whetton, , and D. J. Karoly, 2003: The response of the Antarctic Oscillation to increasing and stabilized atmospheric CO2. J. Climate, 16 , 15251538.

    • Search Google Scholar
    • Export Citation
  • Fyfe, J. C., , G. J. Boer, , and G. M. Flato, 1999: The Arctic and Antarctic Oscillations and their projected changes under global warming. Geophys. Res. Lett., 26 , 16011604.

    • Search Google Scholar
    • Export Citation
  • Gillett, N. P., , and D. W. J. Thompson, 2003: Simulation of recent Southern Hemisphere climate change. Science, 302 , 273275.

  • Gillett, N. P., , T. D. Kell, , and P. D. Jones, 2006: Regional climate impacts of the Southern Annular Mode. Geophys. Res. Lett., 33 .L23704, doi:10.1029/2006GL027721.

    • Search Google Scholar
    • Export Citation
  • Hall, A., , and M. Visbeck, 2002: Synchronous variability in the Southern Hemisphere atmosphere, sea ice, and ocean resulting from the annular mode. J. Climate, 15 , 30433057.

    • Search Google Scholar
    • Export Citation
  • Haylock, M. R., and Coauthors, 2006: Trends in total and extreme South American rainfall in 1960–2000 and links with sea surface temperature. J. Climate, 19 , 14901512.

    • Search Google Scholar
    • Export Citation
  • Hendon, H. H., , and B. Liebmann, 1990: A composite study of onset of the Australian summer monsoon. J. Atmos. Sci., 47 , 22272240.

  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Karoly, D. J., 1990: The role of transient eddies in the low-frequency zonal variations in the Southern Hemisphere circulation. Tellus, 42A , 4150.

    • Search Google Scholar
    • Export Citation
  • Karoly, D. J., , and K. Braganza, 2005: Attribution of recent temperature changes in the Australian region. J. Climate, 18 , 457464.

  • Kidson, J. W., 1988: Indices of the Southern Hemisphere zonal wind. J. Climate, 1 , 183194.

  • Kushner, P. J., , I. M. Held, , and T. L. Delworth, 2001: Southern Hemisphere atmospheric circulation response to global warming. J. Climate, 14 , 22382249.

    • Search Google Scholar
    • Export Citation
  • Lavery, B., , G. Joung, , and N. Nicholls, 1997: An extended high-quality historical rainfall dataset for Australia. Aust. Meteor. Mag., 46 , 2738.

    • Search Google Scholar
    • Export Citation
  • L’Heureux, M. L., , and D. W. J. Thompson, 2006: Observed relationships between the El Niño–Southern Oscillation and the extratropical zonal-mean circulation. J. Climate, 19 , 276287.

    • Search Google Scholar
    • Export Citation
  • Li, Y., , W. Cai, , and E. P. Campbell, 2005: Statistical modeling of extreme rainfall in southwest Western Australia. J. Climate, 18 , 852863.

    • Search Google Scholar
    • Export Citation
  • Lorenz, D. J., , and D. L. Hartmann, 2001: Eddy–zonal flow feedback in the Southern Hemisphere. J. Atmos. Sci., 58 , 33123327.

  • Lorenz, D. J., , and D. L. Hartmann, 2003: Eddy–zonal flow feedback in the Northern Hemisphere winter. J. Climate, 16 , 12121227.

  • Marshall, G. J., 2003: Trends in the Southern annular mode from observations and reanalyses. J. Climate, 16 , 41344143.

  • McBride, J. L., , and N. Nicholls, 1983: Seasonal relationships between Australian rainfall and the Southern Oscillation. Mon. Wea. Rev., 111 , 19982004.

    • Search Google Scholar
    • Export Citation
  • Meneghini, B., , I. Simmonds, , and I. N. Smith, 2007: Association between Australian rainfall and the Southern Annular Mode. Int. J. Climatol., 27 , 109121.

    • Search Google Scholar
    • Export Citation
  • Miller, R. L., , G. A. Schmidt, , and D. T. Shindell, 2006: Forced annular variations in the 20th century Intergovernmental Panel on Climate Change Fourth Assessment Report models. J. Geophys. Res., 111 .D18101, doi:10.1029/2005JD006323.

    • Search Google Scholar
    • Export Citation
  • Mills, G. A., , G. Weymouth, , J. Lorkin, , M. Manton, , E. Ebert, , J. Kelly, , D. Jones, , and G. deHoedt, 1997: A national objective rainfall analysis system. BMRC Techniques Development Rep. 1, Bureau of Meteorology, Melbourne, Australia, 30 pp.

  • Mo, K. C., , and G. H. White, 1985: Teleconnections in the Southern Hemisphere. Mon. Wea. Rev., 113 , 2237.

  • Nicholls, N., 2003: Continued anomalous warming in Australia. Geophys. Res. Lett., 30 .1370, doi:10.1029/2003GL017037.

  • Power, S., , F. Tseitkin, , S. Torok, , B. Lavery, , R. Dahni, , and B. McAvaney, 1998: Australian temperature, Australian rainfall and the Southern Oscillation, 1910–1992: Coherent variability and recent changes. Aust. Meteor. Mag., 47 , 85101.

    • Search Google Scholar
    • Export Citation
  • Reason, C. J. C., , and M. Rouault, 2005: Links between Antarctic Oscillation and winter rainfall over western South Africa. Geophys. Res. Lett., 32 .L07705, doi:10.1029/2005GL022419.

    • Search Google Scholar
    • Export Citation
  • Rogers, J. C., , and H. van Loon, 1982: Spatial variability of sea level pressure and 500 mb height anomalies over the Southern Hemisphere. Mon. Wea. Rev., 110 , 13751392.

    • Search Google Scholar
    • Export Citation
  • Shindell, D. T., , and G. A. Schmidt, 2004: Southern Hemisphere climate response to ozone changes and greenhouse gas increases. Geophys. Res. Lett., 31 .L18209, doi:10.1029/2004GL020724.

    • Search Google Scholar
    • Export Citation
  • Shindell, D. T., , R. L. Miller, , G. Schmidt, , and L. Pandolfo, 1999: Simulation of recent northern winter climate trends by greenhouse-gas forcing. Nature, 399 , 452455.

    • Search Google Scholar
    • Export Citation
  • Silvestri, G. E., , and C. S. Vera, 2003: Antarctic Oscillation signal on precipitation anomalies over southeastern South America. Geophys. Res. Lett., 30 .2115, doi:10.1029/2003GL018277.

    • Search Google Scholar
    • Export Citation
  • Smith, C. A., , and P. Sardeshmukh, 2000: The effect of ENSO on the intraseasonal variance of surface temperature in winter. Int. J. Climatol., 20 , 15431557.

    • Search Google Scholar
    • Export Citation
  • Smith, I. N., 2004: An assessment of recent trends in Australian rainfall. Aust. Meteor. Mag., 53 , 163173.

  • Thompson, D. W. J., , and J. M. Wallace, 2000: Annular modes in the extratropical circulation. Part I: Month-to-month variability. J. Climate, 13 , 10001016.

    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., , and S. Solomon, 2002: Interpretation of recent Southern Hemisphere climate change. Science, 296 , 895899.

  • Thompson, D. W. J., , M. P. Baldwin, , and S. Solomon, 2005: Stratosphere–troposphere coupling in the Southern Hemisphere. J. Atmos. Sci., 62 , 708715.

    • Search Google Scholar
    • Export Citation
  • Torok, S. J., , and N. Nicholls, 1996: A historical annual temperature data set for Australia. Aust. Meteor. Mag., 45 , 251260.

  • Trenberth, K. E., 1979: Interannual variability of the 500 mb zonal-mean flow in the Southern Hemisphere. Mon. Wea. Rev., 107 , 15151524.

    • Search Google Scholar
    • Export Citation
  • Trewin, B. C., , and A. C. F. Trevitt, 1996: The development of composite temperature records. Int. J. Climatol., 16 , 12271242.

  • Watterson, I. G., 2001: Zonal wind vacillation and its interaction with the ocean: Implications for interannual variability and predictability. J. Geophys. Res., 106 , 2396523976.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Composite daily 850-hPa winds [maximum vector (m s−1) shown to right of each panel] and sea level pressure for the (top) low, (middle) high, and (bottom) high-minus-low polarity of the daily SAM index in the December–February season, 1979–2005. The contour interval (CI) in the top two panels is 3 hPa, and in the bottom panel is 2 hPa. The number of days in each index polarity is indicated in the upper right of the top two panels. Vector winds are plotted heavy where they are deemed to significantly differ from 0 at the 90% level based on a t test. Positive (negative) contours of sea level pressure differences are solid (dashed) and the zero contour is heavy in bottom panel. The vector wind scale in the bottom panel is ½ that in the top two panels, but the maximum plotted vector is different as indicated.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 2.
Fig. 2.

Same as in Fig. 1, but for the June–August season.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 3.
Fig. 3.

Composite daily rainfall (contours and shading) and 850-hPa winds (maximum vector shown in lower left of each panel) for high–low polarity of the SAM index for March–May, June–August, September–November, and December–February. CI for rainfall differences is 0.5 mm day−1 with negative difference dashed. Differences that are deemed to be significantly different from zero based on a resampled Monte Carlo test are shaded. The number of days in the high and low index polarities of the SAM is listed in the upper right of each panel. The vector wind scale is the same in all panels, but the maximum plotted vector is different as indicated.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 4.
Fig. 4.

Rate of the daily occurrence of easterly flow at 850 hPa for the (top) low, (middle) high, and (bottom) high-minus-low polarity of the daily SAM index in the December–February season 1979–2005. CI is 4%.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 5.
Fig. 5.

Std dev of low-pass-filtered (7-day running mean) daily rainfall (mm day−1) for MAM, JJA, SON, and DJF for the period 1979–2005. Data-void regions are unshaded.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 6.
Fig. 6.

Ratio of the frequency of occurrence of exceeding the highest weekly quintile rainfall accumulation in the high index polarity to the low index polarity of the SAM. Solid contours are for ratios 1.5:1, 2:1, 2.5:1, 3:1, and 3.5:1. Dashed contours are ratios 1:1.5, 1:2, 1:2.5, 1:3, and 1:3.5. Shading indicates regions where the ratio is significantly different than 1 based on a resampled Monte Carlo test. The number of days in the high and low index polarities of the SAM is indicated in the upper right of each panel.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 7.
Fig. 7.

Composite daily maximum temperature differences (°C) between the high and low index polarities of the SAM. Differences are plotted solid (open) where they are deemed to be significantly different from 0 at the 95% (90%) level based on a t test. Positive (negative) differences are indicated by circles (triangles).

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 8.
Fig. 8.

Same as in Fig. 7, but for daily minimum temperature.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 9.
Fig. 9.

Composite daily rainfall and 850-hPa wind differences between high and low index polarities of the SAM index during DJF in non-ENSO years. Plotting convention is same as in Fig. 3.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 10.
Fig. 10.

(top) Observed rainfall trend for the DJF season 1979–2005 based. (bottom) Rainfall trend that can be attributed to the SAM. CI is 0.2 mm day−1 (27 yr)−1.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Fig. 11.
Fig. 11.

Maximum temperature trend [°C (25 yr)−1] in DJF for (top) 1960–2005 and (middle) 1979–2005. (bottom) Maximum temperature trend that can be attributed to the SAM in the 1979–2005 period.

Citation: Journal of Climate 20, 11; 10.1175/JCLI4134.1

Table 1.

Representative stations (from west to east) that exhibit significant changes in maximum temperature (TMax) associated with swings in the SAM for the summer season (DJF). Columns are station name and location, mean summer maximum T and its daily std dev, difference in maximum T for high polarity and low polarity composites of the SAM, climatological threshold for highest decile of maximum T, and rates of occurrence of highest decile in the high and low polarity composites of the SAM (days exceeding threshold in each phase and total number of days in each phase indicated in parentheses). Bold indicates differences significant at the 95% level.

Table 1.
Table 2.

Representative stations (from west to east) that exhibit significant changes in minimum temperature associated with swings in the SAM for the winter season (JJA). Columns are station name and location, mean winter minimum T and its daily std dev, minimum T difference for the high polarity and low polarity composites of the SAM, climatological threshold for lowest decile of minimum T, and rates of occurrence of lowest decile in the high and low polarity composites of the SAM (days exceeding threshold in each phase and total number of days in each phase indicated in parentheses). Bold indicates differences significant at the 95% level.

Table 2.
Table 3.

Correlation of SAM index with Niño-3.4 SST index (average SST for 5°S–5°N, 120°–170°W) for the period 1979–2005 (bold indicates significant correlation at the 95% level).

Table 3.
Table 4.

Trend in SAM index for the period 1979–2005. Units are daily std devs.

Table 4.
Save