Executive summary
Extremes are the infrequent events at the high and low end of the range of values of a particular climate or weather variable. A small change in the average of a climate variable such as temperature can cause a large change in the frequency of extreme temperatures such as frosts. Extreme weather and climate events can cause severe impacts on society, the economy, and the environment. Several climate and weather extremes have cause severe impacts in Australia in recent years. For example, Eastern Australia experienced record temperatures during the period 1‑22 February 2004 which led to "the most significant medical emergency in the south-east corner [of Queensland] on record" (Canberra Times, 24 February). Tropical Cyclone Larry left 100 square kilometres of World Heritage listed rainforest in north Queensland as "coleslaw and sticks" in March 2006, according to the Queensland Parks and Wildlife Service (The Age, 11 April 2006, p 13). But are these examples of a human influence on weather and climate extremes? A wide range of extreme weather events is possible even with an unchanging climate, so it is difficult to attribute an individual event to a changed climate. As well, until recently the quality and quantity of data to study changes in extremes have been insufficient to allow credible examination of whether extremes are changing.
Over the past decade there have been increased international efforts to improve the quality and availability of data suitable for determining whether or not extremes have changed. Analyses of these data indicate that more than 70% of the global land area sampled exhibited a statistically significant decrease in the annual occurrence of cold nights and a significant increase in the annual occurrence of warm nights. Precipitation extremes showed a widespread and significant increase, but the changes are much less spatially coherent compared with temperature change. Other extremes such as tornadoes are much harder to monitor and it is thus much harder to determine whether or not there has been a change in their frequency and/or intensity. Changes in Australian extremes are generally similar to the changes that have been observed globally. Some of these changes, at least in the case of extreme temperatures, now appear to be at least partly attributable to human influences on the climate.
The changes in Australian extremes likely to accompany anticipated future increases in atmospheric concentrations of greenhouse gases include (from various sources—see text for details):
- Increase in frequency of days over 35°C by 2020
- Decrease in frequency of days below 0°C by 2020
- Increases in intensity of heavy daily rainfall events, although there appears likely to be considerable spatial variation in this
- Decrease over north-east Australia of the number of tropical cyclones, accompanied by an increase in intensity
- Decreased hail frequency in some places
- Increase in large hail (2cm diameter) and reduction in average recurrent interval for hail exceeding 6cm diameter in Sydney
- More droughts over most of Australia by 2030
- Increased frequency of extreme fire danger days (except Tasmania).
There is, however, considerable uncertainty in these projections, arising from the limited number of climate simulations from which they are derived, as well as model deficiencies.
Further work is needed to refine our understanding of extremes and their possible changes, including:
- A reanalysis of tropical cyclone data, to facilitate comparisons of the frequency and intensity of current-day cyclones with those in the past.
- Analysis of historical changes in drought frequency, intensity and duration, using multiple drought indices, e.g. rainfall deficiency, standardised precipitation index, soil moisture deficit and Palmer Drought Severity Index. Climate change projections are required for the same indices, including estimation of drought return periods relevant for assessment of Exceptional Circumstances.
- A regional reanalysis, including homogenisation of upper air data, to facilitate studies linking specific extremes with synoptic patterns.
- Improved down scaling of small-scale synoptic events such as tornadoes, hailstorms and thunderstorms that are difficult to monitor using conventional meteorological networks and approaches, to facilitate an increased focus on studies of these small-scale events.
- Improved climate models, with higher resolution and improved parameterisation of small-scale processes that lead to extremes. This will require involvement of the user community in the design of ACCESS (the Australian Community Climate Earth System Simulator).
- Development of high quality historical data sets for wind speed and hail, to facilitate documentation of any trends in these extremes.
- Improved historical data sets of rainfall and temperature should include the effects of urban heating, rather than removing such effects.
- An increased emphasis is required on sub-daily precipitation extremes, and the analysis of historical changes in extremes would be facilitated by increased palaeo-climatic emphasis on extreme events.
- Joint analyses of multiple extremes (e.g. strong winds and heavy rainfall) that might exacerbate the impacts either extreme would have on its own.
- Studies to determine how much of the recent trends in extreme temperatures are attributable to human actions, and how this varies seasonally and spatially.
- Integrated assessments are needed to determine how communities could or should react to changes in extremes.
- A comprehensive assessment of projected changes in extreme daily temperature, rainfall, wind, fire danger, tropical cyclones, hail, tornadoes and storm surges. To ensure internal consistency, this would require a suite of simulations from selected climate models that perform well in the Australian region.