'It's only eerie if you don't accept the validity of the science': Garnaut's fire prediction a decade on

"This effect increases over time, but should be directly observable by 2020," Prof Garnaut wrote in 2008.

He also spelled out the health risks, including the impacts of severe weather events such as bushfires, the impacts of temperature extremes, including heatwaves, and increases in air pollution, for example, from bushfire smoke.

In 2020, with Canberra and NSW clouded in choking bushfire smoke and the South Coast hit by bushfires of shocking ferocity and spread, Prof Garnaut's words might seem eerily prescient, but not to him.

"It's only eerie if you don't accept the validity of the science," he responds to the suggestion. "It's not eerie that apples fall downwards towards the earth if you have absorbed Newton's laws of motions."

Prof Garnaut said the 2008 review had been based on "very good science".

The science in 2008 predicted that the "accumulation of badness" would have reached a point by 2020 when the effects were noticeable in everyday life, but there was nothing critical about 2020. The same science predicted it would continue to get worse, he said.

"Things are never so bad that they wont get worse until the world has achieved zero net emissions," he said.

Wherever he had looked for evidence, the picture was unfolding in ways predicted by the 2008 analysis.

"The science generally is playing out as my distillation said it would," Prof Garnaut said. "And that is true whether you are talking about temperature increases and heatwaves, or bushfires, or reduced water flows into the rivers and streams of southern Australia, or sea level rise."

Dr Garnaut also predicted an increase in heatwaves. Canberra had five days over 35 degrees in 2008. That was predicted to rise to eight by 2030, and 21 days by 2070.

In 2019, Canberra sweltered on 29 days over 35 degrees in just two months - January and December.

Prof Garnaut's report commissioned fire-danger work from scientists at the Bureau of Meteorology and the CSIRO.

That work, led by Dr Chris Lucas, predicted an increase in very high and extreme fire danger days

It looked at 26 sites across NSW, southeast Queensland and South Australia, and modelled temperature increases of 0.4 degrees, 1 degree and 2.9 degrees.

Dr Lucas predicted that the number of very high fire danger days would increase by 10 to 30 per cent in 2020 for the high-global warming scenario, depending on location. The number of extreme fire danger days would increase by 5 to 25 per cent in the low-warming scenario by 2020, and between 15 and 65 per cent with higher global warming.

Of the 26 sites, only 12 had recorded catastrophic fire weather days in the previous 35 years. Under a one degree global warming scenario, 20 of the locations would have catastrophic fire days.

"Fire seasons will start earlier and end slightly later, while being generally more intense throughout their length. This effect is most pronounced by 2050, although it should be apparent by 2020," the researchers concluded.

In 2008, Canberra had an average of 16.8 very high fire danger days. By 2020, that was predicted to rise to 18 or 19 in a low-warming scenario and 21 to 23 in the high-temperature scenario.

Nowra had 8.8 very high fire danger days. That was predicted to stay roughly similar by 2020 in low-temperature scenario, but rise to nine to 10 days in the high-temperature scenario.

Asked how he felt watching the bushfire crisis unfold in the wake of his warnings in 2008, Prof Garnaut, professorial fellow in economics at the University of Melbourne, said: "My feeling is one of personal disappointment that I was asked by all of the governments of Australia - federal, state and territory - to prepare a report on climate change and the appropriate policy response to it. And I failed in persuading the Australian community that it was in our interests to move decisively and strongly as part of a global effort."