The Köppen-Geiger climate classification scheme to classify world climate is based on the concept that native vegetation forms the best expression of climate. The modified Köppen-Geiger climate classification of Australia adopted by the Bureau of Meteorology (Stern et al., 2000) characterises most of the Cooper subregion as hot, persistently dry desert. A small area in the north-east is classified as hot, persistently dry grassland. The distribution of mean annual rainfall is shown in Figure 10 . Most of the Cooper subregion receives between 200 and 300 mm/year of rain, it is drier in the west with less than 200 mm/year and wetter in the north-east with 300 to 400 mm/year rainfall. Evapotranspiration is highest in the north-west and decreases towards the south and east. The range is from 2300 to 2400 mm/year at Cameron Corner to 2800 to 2900 mm/year at Canterbury. The distribution of mean annual point potential evapotranspiration is shown in Figure 11 . The contrast between Figure 10 and Figure 11 demonstrates the excess of potential evapotranspiration over precipitation across the region. Figure 12 (a) – which shows mean monthly rainfall and potential evapotranspiration – demonstrates that this applies to all months of the year.
Figure 10 Cooper subregion mean annual rainfall
Data: Bureau of Meteorology (Dataset 2)
Figure 11 Cooper subregion mean annual potential evapotranspiration
Data: Bureau of Meteorology (Dataset 3)
(a) (b)
Data: Bioregional Assessment Programme (Dataset 8)
Figure 12 (b) shows that the region experiences hot summers with mean maximum temperatures over 35 °C and minima over 20 °C. Winter maxima are warm (greater than 20 °C) and minima cool (less than 10 °C) and the relatively large diurnal range throughout the year reflects the continental setting. Precipitation is generally low and shows a distinct summer maximum. Mean January precipitation for Canterbury in the east is 50 mm compared to 28.5 mm for Innamincka in the west. The lowest rainfall occurs in the late winter to early spring period (August to September) and is lower in the west than the east, with Canterbury averaging 8.5 mm and Innamincka 6.9 mm for August. Importantly, rainfall is highly variable from year to year across the whole subregion with Canterbury recording the lowest on record to highest on record figures for January of zero to 263.3 mm and for August of zero to 69.9 mm. The corresponding figures for Innamincka are zero to 536.0 mm and zero to 56.7 mm, respectively (Bureau of Meteorology, 2014a, 2014b).
CSIRO and Bureau of Meteorology (2012) is based on historical records and shows a 0.9 °C increase in mean annual temperature between 1910 and 2011. The records show little consistent change between 1910 and 1950 followed by a progressive increase since 1950, with each decade since 1950 being warmer than the previous decade. This clear warming trend is complicated by a background of natural variability due to factors such as the El Niño-Southern Oscillation (ENSO) and regional differences. Overnight minimum temperatures have warmed more than daytime maxima. The records show that the annual average daily mean temperature of the Cooper subregion warmed by 0.8 to 1.2 °C between 1960 and 2011. The Cooper subregion does not lie in a region with a strong rainfall trend but is likely to have had a slight decrease over recent decades. The CSIRO and Bureau of Meteorology (2012) survey concluded that both natural and human influences have affected climate over the past 100 years and that anthropogenic greenhouse gases have caused most of the surface warming, temperature extremes, ocean warming and sea-level rise, which is consistent with global patterns. The reduction in rainfall in the south-west of Australia since the 1970s is also most likely to be due to increased anthropogenic greenhouse gases, though other less marked rainfall trends in other regions are more difficult to attribute.
CSIRO and Bureau of Meteorology (2007) provides projections of future changes in the Australian climate, under different emissions scenarios, and the application of those projections for impact studies and risk assessments. The study used 23 climate models used by climate research groups around the world with different Intergovernmental Panel on Climate Change (IPCC) greenhouse gas emission scenarios converted to radiative forcing of the climate system. Projections for annual and seasonal mean changes to temperature, rainfall, humidity, solar radiation, wind speed, potential evaporation and sea-surface temperature are provided in probabilistic form but only qualitative assessments were possible for other climate variables. Projections for 2030 showed varying patterns of regional change but minimal impact of different emission scenarios because such shorter-term future projections strongly reflect the impact of greenhouse gases already emitted. Projections for 2050 and 2070 show more variability according to emissions scenario. The median value (50th percentile) is used as the best estimate with the uncertainty range derived from the 10th and 90th percentile values.
According to CSIRO and Bureau of Meteorology (2007), the best estimate of Australia-wide warming is about 1.0 °C (uncertainty: 0.6 to 1.5 °C) by 2030; about 1.2 °C (uncertainty: 0.8 to 1.8 °C) under a low-emissions scenario by 2050; about 2.2 °C (uncertainty: 1.5 to 2.8 °C) under a high-emissions scenario by 2050; about 1.8 °C (uncertainty: 1.0 to 2.5 °C) under a low-emissions scenario by 2070; and about 3.4 °C (uncertainty: 2.2 to 5.0 °C) under a high-emissions scenario by 2070. Warming is higher inland than coastal and northern, which is higher than southern, and it is lower in winter than other seasons. In the Cooper subregion the projections of warming are generally higher than the continental average values. According to CSIRO and Bureau of Meteorology (2007) the 2030 projection is for a mean annual temperature increase in the range of 1.1 to 1.2 °C across the subregion. The 10th and 90th percentile figures are 0.6 to 0.8 °C and 1.4 to 1.8 °C, respectively. Under the 2050 low-emissions and high-emissions scenarios, projected increases are 1 to 1.5 °C and 2 to 2.5 °C, respectively; the corresponding values for 2070 are 1.5 to 2.0 °C and 3 to 4 °C. Under the low-emissions scenario there is a greater than 90% probability of the subregion exceeding a 1 °C warming by 2070 but less than 10% probability of exceeding 3 °C. Under the high-emissions scenario, there is a greater than 90% probability of exceeding 2 °C and a 20 to 40% probability of exceeding 4 °C.
CSIRO and Bureau of Meteorology (2007) provides the best current estimate of changes to Australia-wide precipitation. In the Cooper subregion the best estimate (50th percentile) of projected mean annual rainfall is for a slight decline in the range of 1 to 2.5% with a relatively large uncertainty range from a decrease of 10 to 15% (10th percentile) to an increase of 5 to 10% (90th percentile). However, the minimal annual change contains considerable seasonal variability with summer and autumn precipitation showing minimal increase or decrease, and winter and spring showing a stronger decrease from 2.5 to 10%. By 2050 the mean annual precipitation projection ranges from a decrease of 2 to 5% under the low-emissions scenario to 5 to 10% under the high-emissions scenario, with estimates of 2 to 10% and 5 to 20%, respectively, for 2070. By 2070 there is a 30 to 40% probability of a mean annual rainfall decline of at least 10% under the low-emissions scenario and a probability of 50 to 70% under the high-emissions scenario.
Solar radiation, relative humidity and potential evapotranspiration are projected to change little with a very slight increase for the former and a decrease for the latter two in 2030 with slightly greater changes for 2050 and 2070, especially under high-emission scenarios. Drought occurrence is projected to increase with increased effectiveness due to higher temperatures. There is a projected possibility of an increase in more-intense tropical cyclones but a decrease in the total number of cyclones. ENSO impact projections suggest that El Niño events will become drier and La Niña events will become wetter.
There was an update to the projections released in January 2015 (Watterson et al., 2015) after this assessment was prepared. The projected changes in the 2015 projections are broadly consistent with those included in this assessment.