Raging fires have caused death and destruction in Australia. Credit: Punchstock

Australian firefighters are desperately trying to extinguish the worst bushfires that the country has seen in decades, causing more than 170 deaths so far in the state of Victoria.

While strong winds continue to fan the flames, thousands of homeless families in search of shelter and food are flooding Red Cross relief centres set up around the state. Nature asks if savage blazes such as these will get more frequent in a warming world, and looks at the environmental fallout from the disaster.

Why is this fire so severe?

The vegetation in the region, dominated by bushes and tall eucalyptus trees, burns extremely well when dry. Combined with record-breaking temperatures, unusually strong winds and alleged arson attacks, the fires have been particularly quick to spread through suburban areas that back onto bushland.

As the populations of cities such as Melbourne grow, an increasing number of homes are being built in areas that have a greater risk of wildfire — a problem that also exists in the greater Los Angeles and San Diego areas of California.

Lightning usually triggers wildfires, but there were no reported thunderstorms in the region until yesterday. This lends weight to the claim that at least some of the fires may have been started by people, perhaps deliberately — acts described as "mass murder" by Australian Prime Minister Kevin Rudd.

Could climate change be responsible for the wildfire, or the weather that has sustained it?

No single weather event can be attributed to climate change with any degree of certainty. Warm and dry summers are a typical feature of Australia's subtropical climate, as are frequent wildfires. Moderate fires are actually ecologically beneficial, and can provide nutrients for fresh plant growth, for example. Most trees and tall vegetation can survive normal bushfires, and smaller, more frequent fires reduce the risk of a catastrophic blaze.

However, climate models do suggest that Australian summers will get warmer and drier as the century proceeds, and there is little doubt that this will have an effect on fire risk.

"Heatwaves and fires are virtually certain to increase in intensity and frequency," the Intergovernmental Panel on Climate Change (IPCC) concluded in its most recent Fourth Assessment Report in 2007.

By 2080, average temperatures in central Australia could increase by up to 8°C, according to the IPCC's most extreme regional prediction. Within 400 kilometres of the coast, the continent could still warm by up to 5.4°C, and the amount of rainfall could decrease by up to 80%.

In southeast Australia, the frequency of very high and extreme fire danger days is likely to rise by 4–25% by 2020 and by 15–70% by 2050, according to the IPCC's report.

Other regions exposed to high wildfire risk, including southern Europe, South Africa and the southwestern United States, face similar changes. Increased fire frequency could lead to vegetation changes that would reduce plant growth, decreasing the amount of carbon that plants remove from the atmosphere — further exacerbating greenhouse warming by carbon dioxide.

How will we mitigate against such fires in the future?

A forest-management method called prescribed burning is already being tested in Australia, California and Portugal. The idea is to deliberately set the vegetation on fire from time to time, without letting fires get too large. This means that the accumulating ground layer of flammable dry wood never gets thick enough to provide a major source of fuel for an enormous fire.

Ecologists are still assessing the best time of year for setting different types of vegetation on fire, to find the best strategy that causes the least damage to plants. Probably the best ways to adapt to increased wildfire risk include sensible urban planning, installation of early-warning systems and preparedness among the population.

Do wildfires like this add a significant load of CO 2 to the atmosphere?

Disastrous though they are, the Australian bushfires are not large enough to leave a signal on global atmospheric CO2 levels. The only known fire events in recent decades that apparently did release substantial amounts of CO2 were the Indonesian peat fires in 1997–98, and the vast Siberian forest fires of 2003.