Chapter 4 Climate Impacts and Responses

Impacts of Climate Change on Australia > The drying trend, drought and fire

The significant rainfall deficiencies in the period since the publication of Climate Activities in Australia 2001 warrants separate consideration to the more general research on water resources. By November 2002, almost 62% of Australia had serious or severe 9-month rainfall deficiencies, making it the most widespread 9-month drought on record (Figure 4.3).

Figure 4.3 Rainfall anomalies for April 2002 to January 2003 compared with records back to 1900.

High temperatures were also a feature of this drought, with Australian-average maximum temperature the highest on record. On an annual basis, maximum temperature and rainfall are negatively correlated, i.e. droughts tend to have anomalously high temperatures. However, since about 1973, temperatures have tended to be higher for a given rainfall amount, i.e. droughts have become hotter, consistent with simulations of climate change due to increasing greenhouse gases. Australian researchers concluded that the drought of 2002 and the associated impacts on agriculture, water resources and fire were made more severe than past droughts due to the high temperatures and evaporation which may be partly due to human-induced global warming (Karoly et al 2003, Nicholls 2003).

Impacts of the 2002-2003 drought have been summarised by Australian Bureau of Agricultural and Resource Economics (ABARE), Bureau of Rural Sciences (BRS), Murray Darling Basin Commission (MDBC), Monash University and the Bureau of Meteorology. Key findings are:

• 62% of Australia with serious rainfall deficiency from March to December 2002 (most widespread 9-month deficiency on record);
• Australian average maximum temperature in 2002 hottest on record, especially in the Murray-Darling Basin where evaporation was also exceptionally high;
• Large dust storms over inland regions, one measuring 1500 by 400 km reaching Sydney and Brisbane on 23 October 2002, another affecting Griffith and Sydney on 13 November 2002, and another affecting Griffith again on 29 November 2002 reducing visibility to less than 300 metres;
• 30% reduction in 2002-03 agricultural output, equivalent to 1% of GDP;
• Net effect on 2002-03 GDP expected to be a loss of 1.6%;
• Loss of 70,000 jobs likely, mainly in wholesale, retailing and repairs, transport, business services, agricultural services, and food processing and beverages;
• Government drought relief package totalling $728 million over three years for farmers and businesses affected in rural regions;
• Depleted water storages in the Murray- Darling Basin, e.g. Burrendong Dam at 18% of capacity in December 2002, Copeton Dam 23%, Burrinjuck Dam 28%, Hume Dam 14%, Dartmouth Dam 54%, Lake Eildon 20% and expected to decrease further;
• No net flow past the Barrages at the Murray River mouth for more than 15 months; and
• Water restrictions in Melbourne (first time since 1982) and Canberra (first time since 1966), reduced water allocations in many rural areas, and water carting needed in 21 NSW towns (9 having exhausted dam supplies).

CSIRO analysed the output of nine climate models driven by projected increases in greenhouse gases over the 21st century. Results suggest a tendency for increased summer-autumn rainfall, especially in the north where human water use is low, but decreased rainfall in southern and eastern Australia, especially in winter and spring, with increased potential evaporation. This would exacerbate the existing moisture deficit across most of the country (Figure 4.4). For example, in western NSW, the average water balance is -1500 to -2000 mm (net moisture deficit) and the average change by 2030 is -40 to -160 mm (greater moisture deficit). These estimates could be a reflection of possible changes in ENSO, which affects the frequency of severe weather such as hailstorms. Over the past decade, El Niño conditions have been observed to increase and climate variability is expected to continue under enhanced greenhouse conditions, possibly with more intense rainfall in La Niña years and more intense drought in El Niño years.

Figure 4.4 Annual average Australian water balance (rainfall minus evaporation) for 1961- 90 (a) and projected changes for 2030 and 2070 relative to 1990 (b).

CSIRO researchers input the above estimated changes in rainfall and potential evaporation into a rainfall-runoff model for the Macquarie River, to find the decrease in average streamflow was up to 20%. Critical thresholds, defined as missed bird breeding cycles and economic failure on farms, were exceeded in 2030 by 20-30% in a droughtdominated climate and much less than 1% in a flood-dominated climate. However, by 2070 these risks were 70-80% and 10-20%, respectively. These reduced flows were increased by afforestation, seen as a method of addressing salinity and carbon sequestration. For example, BRS researchers found an increase of 10% in tree cover led to a decrease of 17% of river flow. Flow reductions have also been found by other researchers using different methods, and in different catchments.

The Indian Ocean Climate Initiative (IOCI) is a climate research program established by the Western Australian Government in collaboration with the BMRC and CSIRO to identify the causes of the serious rainfall decreases and the consequential impact on water resources experienced in southwest Western Australia since the 1970s (Figure 4.5). The strategic research program undertaken by IOCI over the last 5 years has improved the understanding of climate variability and trends for southwestern Australia. Research directed at developing inter-seasonal forecasting and better understanding climate variability on longer time-scales became progressively focussed on issues of decadal variability and longer term change because of the dominance of these issues in any analyses of the observed record.

Figure 4.5 Yearly inflow to Perth water supply (columns), showing 10-year moving average, 1911- 2001 average (287GL) and 1975-2001 average (167GL). From Climate variability and seasonal change in south west Western Australia, Perth, September 2002 (www.wrc.wa.gov.au/ioci).

Key findings include (IOCI 2002):

• Winter rainfall has decreased sharply and suddenly in the region since the mid 1970s;
• The decline was not gradual but more of a switching to an alternative rainfall regime;
• The rainfall decrease accompanied and was apparently associated with documented change in large scale atmospheric circulation at the time;
• The decrease in rainfall and associated circulation changes bear some resemblance to model projections for an enhanced greenhouse effect (EGE) but are not sufficiently similar to indicate, beyond doubt, that the EGE is responsible. Most likely both natural variability and the EGE have contributed to the rainfall decrease;
• The climatic shifts, which include warming, have resulted in an even sharper fall in regional streamflows;
• The changing climate will exhibit wetter and drier periods throughout the 21st century due to natural variability, overlaid on trends of continued warming and of probable decline in mean rainfall consequent on the EGE; and
• Decision-makers need to alter their decision base-lines to reflect observed and projected changes and also to include increased levels of uncertainty.

There is a strong link between interannual climate variability – particularly that associated with ENSO – and the intensity of the subsequent fire season. Bushfires were a major emergency in Australia over the 2002-03 summer. Fires raged in southeast Queensland in October 2002, in Western Australia in November 2002 124,000 ha burnt), in eastern NSW in early December 2002 ($100 million damage), in northeast Victoria in mid-December 180,000 ha burnt), in ACT in mid-January 2003 (4 deaths, 520 buildings destroyed, 250 million damage) and in the Great Dividing Range in January-February 2003 over 1 million ha burnt, $60 million firefighting costs).

In July 2003 the Cooperative Research Centre for Bushfire Research will commence operation, with seven years of funding. The research centre includes participants from Emergency Management Australia, CSIRO, the Bureau of Meteorology, State and Territory fire authorities and emergency service departments, and forestry organisations from NSW, Queensland, Tasmania, Western Australia, Victoria and New Zealand. Existing climate prediction and monitoring systems will be adapted and extended to provide advance warning of conditions likely to affect the level of fire danger.