Atmosphere Theme Report
Australia State of the Environment Report 2001 (Theme Report)
Lead Author: Dr Peter Manins, Environmental Consulting and Research Unit, CSIRO Atmospheric Research, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06746 9
Climate Variability and Change (continued)
Extreme temperatures [A Indicator 1.7]
As explained above, greater warming occurred during the second half of the 20th century, particularly for night temperatures. Increases in both annual mean minimums and in annual mean maximum temperatures have resulted in an increase in the area of Australia experiencing temperatures above the 90th percentile and a decrease in the area with temperatures below the 10th percentile (Figure 30). The frequency of extreme warm days and nights has increased while that of extreme cool days and nights has decreased over the second half of the 20th century (Plummer et al. 1999; Collins et al. 2000). Similar changes are observed when extremes spanning consecutive days are examined and, again, more significant trends are associated with increases in minimum temperatures.
Figure 30: Percentage area of Australia over which extreme annual mean temperatures occurred of greater than the 90th percentile (red) or less than the 10th percentile (teal) of temperature calculated over the 1961 to 1990 period.
Source: BoM
Overnight temperatures have increased more than daytime temperatures, resulting in a decrease in the diurnal temperature range (the difference between the daily maximum and minimum temperatures). For areas affected by frosts, Collins et al. (2000) found that the annual number of frost days declined by an average of 5.6 from 1957 to 1996 and the average length of the frost season shortened by around 40 days. The shortening of frost season has contributed to an increase in wheat yield in Australia between 1952 and 1990 (Nicholls 1997).
Implications
Greater increases in minimum temperatures have led to an increase in the frequency of extreme warm days and nights and a decrease in cool days and nights, particularly over the second half of the 20th century. For areas affected by frosts, the annual number of frost days and the length of the frost season have declined. Continuation of such an increase in hot days and a reduction in cool days under enhanced greenhouse conditions could lead to greater heat stress in humans, increases in heat-related diseases to human and livestock, changes to ecosystems, agriculture and building materials, increased frequency of bushfires and droughts, higher energy demand for air-conditioning and increased demand on water supply. An increase in temperature is also major concern for coral reefs around Australia (see the Coasts and Oceans Theme Report).
Temperature of the lower and upper atmosphere [A Indicator 1.9]
Global average surface temperature measured in standard meteorological screens at 1.5 m above the ground has increased since 1861. There are, however, major differences between temperature measured near the earth's surface and the temperature data obtained from satellites because satellites measure temperature over a considerable depth of the lower atmosphere. Satellite measurements of this layer (from the surface to 8 km above it) have shown smaller temperature rises than those measured at the surface (from 0 to 0.2C) from 1980 to 2000, the period where satellite measurements exist. Analysing temperature data taken from radiosondes (instruments attached to balloons) is another approach used to obtain a consensus pattern of change in global temperature, and these measurements tend to agree with those made by satellites. Jones (1994) computed temperature trends for 1979 to 1993 of 0.10C per decade based on rawindsonde (equipment used for gathering meteorological data) data for the layer between 850 and 300 hPa (between about 1.5 and 7 km above sea level), and 0.17C per decade based on combined land-surface air temperature.
The National Research Council of the United States of America (USA) considered this issue in detail (National Research Council 2000). The differences between the temperature measurements can be explained partly by changes in the techniques used to calculate temperatures from satellites, and partly by real differences between the temperature trends on the surface and those in the lower atmosphere. There is some evidence that lower atmosphere temperatures may have warmed less rapidly than the surface after 1979 because of volcanic eruptions and the cooling of the upper troposphere due to ozone depletion in the stratosphere. Recent works by the Hadley Centre of the British Meteorological Office also shows a link between tropospheric temperature variations and ENSO. It was concluded that these differences in temperature trends did not invalidate the assessment that surface temperatures have been rising.
Latitude-height profiles of zonal-mean temperature changes since the 1960s show significant cooling in the lower stratosphere, especially in the mid-latitudes and high latitudes of the Southern Hemisphere (Parker et al. 1997). Warming dominates the troposphere, and is greatest in the high latitudes of the Southern Hemisphere.
A warming trend in the troposphere and a cooling trend in the stratosphere are observed over Australia (Figure 31). The cooling trend (-0.042C per year) in the Australian stratosphere (about 17 km above the surface) is statistically significant, but the warming trend in the Australian troposphere (+0.011C per year) is not statistically significant. These trends are consistent with simulations by global climate models under enhanced greenhouse conditions, although simulated magnitudes are greater than observed magnitudes; however, many factors contribute to trends caused by global warming, and the association between simulated and observed trends is not definitive.
Figure 31: Upper air temperature over Australia from 1958 to 1999. Temperature values have been averaged for 850 to 500 hPa, 300 to 150 hPa and 100 to 50 hPa.
Source: BoM
The influence of the SOI is also evident in global tropospheric temperature, with reasonably strong negative relationships demonstrated between the SOI and Southern Hemisphere tropospheric temperature.
Implications
An observed increase in tropospheric temperature and a decrease in stratospheric temperature are consistent with simulations of several global climate models under increased greenhouse gases. However, uncertainties are associated with the observational record of upper-air temperature and satellite measurements and also model simulations. Continued monitoring of surface, tropospheric and stratospheric temperature and correction of satellite-measured values can give better estimates of global temperature change.