About the Climate Extremes Analyses

Extreme Climate Indices used

Extreme climate events such as heat waves, cold snaps, floods and dry spells have significant impacts on society. To examine whether such extremes have changed over time a variety of extreme climate indices can be defined, such as the number of days per year which exceed, or fail to exceed, fixed thresholds. However, since people tend to adapt to their local climate, a threshold considered extreme in one part of Australia could be considered quite normal in another. To overcome this problem, thresholds based on percentile values can also be defined.

The 30 extreme temperature and rainfall indices used here are based on those defined by the WMO Expert Team on Climate Change Detection Monitoring and Indices. Some of the extreme temperature indices have been modified for Australian purposes.

Extreme Temperature Indices	Definition
Very hot days	Annual count of days with maximum temperature > 40°C
Hot days	Annual count of days with maximum temperature > 35°C
Very hot nights	Annual count of nights with minimum temperature > 25°C
Hot nights	Annual count of nights with minimum temperature > 20°C
Cold days	Annual count of days with maximum temperature < 15°C
Very cold days	Annual count of days with maximum temperature < 10°C
Cold nights	Annual count of nights with minimum temperature < 5°C
Frost nights	Annual count of nights with minimum temperature < 0°C
Warm days	Percentage of days with maximum temperature > 90th percentile
Warm nights	Percentage of nights with minimum temperature > 90th percentile
Cool days	Percentage of days with maximum T < 10th percentile
Cool nights	Percentage of nights with minimum T < 10th percentile
Highest maximum temperature	Annual maximum value of daily maximum temperature
Highest minimum temperature	Annual maximum value of daily minimum temperature
Lowest maximum temperature	Annual minimum value of daily maximum temperature
Lowest minimum temperature	Annual minimum value of daily minimum temperature
Warm spell duration	Annual count of days with at least 4 consecutive days when daily maximum temperature > 90th percentile
Cold spell duration	Annual count of nights with at least 4 consecutive nights when daily minimum temperature < 10th percentile
Growing season length	Annual (1st July to 30th June) count between first span of 6 or more days with daily mean temperature > 15°C and first span of 6 or more days with daily mean temperature < 15°C

Extreme Precipitation Indices	Definition
Wet days	Annual count of days with daily precipitation ≥ 1 mm
Heavy precipitation days	Annual count of days with daily precipitation ≥ 10 mm
Very heavy precipitation days	Annual count of days with daily precipitation ≥ 30 mm
Maximum 1-day precipitation	Annual maximum 1-day precipitation total
Maximum 5-day precipitation	Annual maximum consecutive 5-day precipitation total
Very wet day precipitation	Annual total precipitation when daily precipitation > 95th percentile
Extremely wet day precipitation	Annual total precipitation when daily precipitation > 99th percentile
Annual total wet day precipitation	Annual total precipitation on wet days (daily precipitation ≥ 1 mm)
Simple daily intensity	Annual total precipitation divided by the number of wet days (daily precipitation ≥ 1 mm)
Consecutive dry days	Maximum number of consecutive days with daily precipitation < 1 mm
Consecutive wet days	Maximum number of consecutive days with daily precipitation ≥ 1 mm

Percentile calculations

Percentile-based threshold levels are calculated for 5-day windows across the annual cycle using the standard 1961 to 1990 normal period ie. daily rainfall and temperatures values are compared relative to varying thresholds throughout the year. So it is possible to have warm spells during winter, cold spells during summer, or extreme precipitation recorded at any time of the year.

Trend maps available

Australian extreme climate trend maps are available for periods starting from the beginning of each decade from 1900 to 1970 for rainfall, and from 1950 to 1970 for temperature. Only a limited amount of digitised daily temperature data is available prior to 1950. Analysis periods starting after 1970 are considered too short to calculate meaningful trend values. Post-1950 trend maps for the extreme temperature indices show fewer calculated trends than the post-1960 and post-1970 periods because digitised data is not available until 1957 at many locations.

Interpreting the trend maps

Trends in extreme indices are presented using "bubble" plots, rather than analysed fields, due to the sparse network of locations for which some indices are relevant. Trend values are only calculated for a particular location if its series of annual extreme values are non-zero for at least 50% of years for temperature indices, and at least 25% for rainfall indices. Consequently trend values are not calculated or plotted for some indices in some regions eg. frost nights in northern Australia. Trends in extreme values also tend to be "noisier" than for the mean climate making them more difficult to analyse onto a continuous field.

The trend maps are a useful way to compare how climate extremes have changed in different regions of Australia over time. They are also interesting to compare with changes in the mean climate over the same period. However, they need to be interpreted with caution. Trend values have been determined from a linear (straight line) fit to the data, but the change may not have been gradual. For example, a calculated trend could be due to a relatively rapid "step" change, with the remainder of the series being fairly flat (see some of the timeseries graphs). Also, trend values are highly dependent on the start and end dates of the analysis. Consequently trend maps starting in different decades can look remarkably different. Users are advised to keep in mind the period over which trend values have been calculated and interpret them alongside the timeseries of averaged values.

In addition, the trend values calculated here using past observations should not be used to imply future rates of change. Due to the complex interactions between the natural and human drivers of climate change and variability, the climate of any location is always changing. Future rates of change will depend on how these drivers interact in future, which will not necessarily be the same as in the past.

Timeseries available

Spatially-averaged timeseries start in 1900 for the extreme rainfall indices but not until 1957 for the extreme temperature indices because digitised daily temperature series are not available at many locations until that year.

Interpreting the timeseries

Timeseries of extreme climate indices are calculated using a simple average of all locations for which the index is considered relevant ie. all locations at which the index series is non-zero for more than 50% of years for temperature indices, and more than 25% of years for rainfall indices. Consequently these averages only represent those regions of the country where the index has meaning. To determine which locations have been used to calculate the spatial average timeseries for a particular index examine the corresponding trend maps.

Data used

The extreme climate analyses are calculated using "high-quality" daily temperature and rainfall datasets developed specifically for monitoring long-term trends and variability in the Australian climate (see maps below). The development of these datasets involved identifying and removing problem records using statistical techniques, visual checks and station history information (metadata). The high-quality daily temperature and rainfall datasets are available at http://www.bom.gov.au/climate/change/datasets/datasets.shtml.

Please note that any use of these data should be acknowledged to the Bureau of Meteorology. Apart from the purposes of study, research, criticism and review, no part of these data may be reproduced, or redistributed for any commercial purposes, or distributed to a third party for such purpose, without written permission from the Director of Meteorology.

Locations of records included in Australia's high-quality daily temperature dataset

Locations of records included in Australia's high-quality daily rainfall dataset

Further information

Alexander, L., Hope, P., Collins, D., Trewin, B., Lynch, A. and Nicholls, N. 2006. Trends in Australia's climate means and extremes: a global context. Australian Meteorological Magazine, (accepted).

Collins, D., Della-Marta, P., Plummer, N. and Trewin, B. 2000. Trends in annual frequencies of extreme temperature events in Australia. Australian Meteorological Magazine, 49, 277-292.

Haylock, M. and Nicholls, N. 2000. Trends in extreme rainfall indices for an updated high quality data set for Australia. International Journal of Climatology, 20, 1533-1541.

Lavery, B., Kariko, A. and Nicholls, N. 1992. A historical rainfall data set for Australia. Australian Meteorological Magazine, 40, 33-39.

Trewin, B.C. 2001. Extreme temperature events in Australia. PhD Thesis, School of Earth Sciences, University of Melbourne, Australia.