South-eastern Australia including the Murray Basin has experienced a range of climate extremes in recent times, including the worst drought of the instrumental record; the Millennium drought (1997–2009), which was broken by Australia’s wettest two-year period on record (2010–11). While the Millennium drought was largely restricted to southern Australia and including a large decline during the cool season in particular autumn, the frequent heavy rain events in the spring and summer of 2010/11, and again in 2011/12, affected most of the Australian continent. However, across south-eastern Australia, there was a continuation of below average rainfall during the cool season (April to October).

The observed long-term reductions occurring in cool season rainfall and stream flow across the region appears related to changes in the mean meridional circulation. Evidence indicates that there is an expansion of the tropics, with the Hadley circulation expanding at the rate of approximately 50 km per decade, pushing mid-latitude storm tracks further south and leading to an expanded subtropical dry zone and reduced rainfall across southern Australia. These changes are at least partly attributable to global warming and can be reproduced by global climate models only when human influences (in the form of greenhouse gases, aerosols and stratospheric ozone depletion) are included albeit with a much reduced magnitude (typically a third to a fifth of the observed changes depending on which aspect is considered)

Future climate projections using the latest CMIP5 database confirm the importance of the Mean Meridional Circulation change to drive the mean rainfall changes and are consistent across many other related variables: means sea level pressure increasing, humidity decreasing, wind decreasing, and solar radiation increasing. Although future rainfall projections have similarities with the observed trends during the last 30 years, the magnitude of the observed decline is equivalent to the worst case scenario by the end of the 21^st century. Several factors are likely to explain this; ranging from natural variability playing a large part in the observed trends to plausible climate models lack of sensitivity. A very simple factor may be that climate model projections at the scale of the Murray Basin lack accuracy due to the coarse resolution of the climate models and hence may not be directly comparable to the observed trends which are based on high density observation networks. Results from downscaling approaches are used to assess this possible mismatch.