Hottest year start keeps climate change in spotlight
The Bureau of Meteorology said last month was the hottest start to a year in just over a century of records.
Nationwide, the January average maximum temperature anomaly was 2.28 degrees, "a substantial increase" on the previous record of 2.17 degrees set in 1932.
Thanks to the massive heatwave that dominated the first half of January, all states and territories posted above-average temperatures, the bureau said.
This week's floods added to the extremes. Queensland Premier Campbell Newman warned damage to the state's economy was $2.4 billion and rising, eclipsing the $2.388 billion bill from the huge flooding of 2011. Insurers don't think it will be that bad for them.
Add record low rainfall for much of southern Australia, a flurry of bushfires and it looks like climate change is kicking in - or does it?
Professor John McAneney, the director of Risk Frontiers, an independent research group funded mostly by the insurance industry, said that based on a database of natural hazard events in Australia, including some dating from 1803, "there has been no increase in the frequency of natural hazard events since 1950".
But what of the spiralling insurance claims in the wake of hailstorms, floods, cyclones (think Yasi at $1.4 billion) and bushfires ($4 billion for Victoria's Black Saturday)?
"When this data set, and many other data sets in different jurisdictions across the world and for many different perils, is corrected for the increases in numbers of buildings at risk and their value, no long-term trend remains," Professor McAneney said. "It is indisputable that the rising toll of natural disasters is due to more people and assets at risk."
Professor McAneney said US hurricane modelling to identify a signal climate change is contributing to storm strength and suggested that it could be a while before the data is definitive. Averaging 18 different climate models, "it's going to take 260 years", he said. "This whole thing about climate change being responsible for an increase in extreme weather, or natural disasters, is just a fiction really," he said.
Cue howls of protest from climatologists and cries of "gotcha" from climate change doubters? Not quite.
Some climate change signals are clearer than others and there is no reason to ignore the direction most indicators are clearly pointed, said Andrew Ash, director of the Climate Adaptation flagship at the CSIRO.
"It doesn't mean all extremes are changing," Dr Ash said.
Take temperature, for instance. During 2001-11, the frequency of record high temperatures in Australia was 2.8 times (for maximum temperatures) and 5.2 times for minimums than the rate of record low temperatures.
Sea temperatures are also rising with Australian regional waters about 0.6-0.7 degrees warmer than in 1900, said Neil Plummer, assistant director of the weather bureau's climate information services.
Add a warmer atmosphere - with temperatures about 1 degree higher than pre-industrial levels and rising - and it means more moisture can be held and then dumped in the form of more severe rain deluges.
A peer-reviewed report for the American Meteorological Society's Journal of Climate by researchers including Seth Westra, a hydrologist at the University of Adelaide, bears that out. The report found statistically significant increasing trends globally of annual maximum daily precipitation using a data set of 8326 high-quality observing sites with more than 30 years of records.
The median intensity of extreme precipitation increases "in proportion with changes in global mean temperature at a rate of between 5.9 per cent and 7.7 per cent per degree, depending on the method of analysis", the report found.
The big wet, when it comes, is getting wetter.
But what of Australia? The weather bureau said it depended on where you look.
The annual number of days with more than 30 millimetres of rain from 1950-2012 has decreased in the southern and eastern parts of the country but increased in the north. That doesn't spell fewer heavy rainfall events as much as the trend for southern regions to miss out on decent soakings during winter.
As for the frequency of disasters, such as cyclones - aren't we getting fewer of those? The answer is complex because there aren't many instances in the record to count.
"Because you're dealing with a very small number of very extreme events that cause most of the damage, the size of the signal you would need to have before it was statistically significant as detectable is quite big," said Blair Trewin, a senior climatologist at the bureau. "The absence of evidence is not the evidence of absence."
Frequently Asked Questions about this Article…
January 2013 was the hottest start to a year in just over a century, with the Bureau of Meteorology reporting a nationwide January average maximum temperature anomaly of 2.28°C (beating the previous record set in 1932). A massive heatwave affected all states and territories, followed by floods that Queensland estimated had caused about $2.4 billion in damage. For everyday investors, these events highlight how extreme weather can affect the economy, infrastructure and insurance exposures—sectors often sensitive to climate-related losses.
The answer is not straightforward. Professor John McAneney (Risk Frontiers) says that after correcting for more buildings and assets at risk, there has been no long-term increase in the frequency of natural hazard events in Australia since 1950. However, climate scientists caution that some signals (like rising temperatures and heavier rainfall extremes) are clearer than others, and the lack of a detectable trend in some hazards does not prove climate change isn’t affecting them.
There is evidence that some extremes are intensifying. A peer‑reviewed study cited in the article found global increases in annual maximum daily precipitation at high‑quality observation sites, and reported that the median intensity of extreme precipitation rises by roughly 5.9–7.7% per degree of global warming. The Bureau of Meteorology also notes a warmer atmosphere (around 1°C above pre‑industrial levels) can hold more moisture, which can lead to heavier deluges.
The article notes insurers said the recent Queensland floods 'don't think it will be that bad for them.' Experts point out that much of the rise in insurance claims over time reflects more people and assets being located in harm’s way rather than a clear increase in event frequency. Historic big events cited include Cyclone Yasi (~$1.4 billion) and Victoria's Black Saturday bushfires (~$4 billion) to illustrate scale.
From 1950–2012 the Bureau reports the annual number of days with more than 30 mm of rain has decreased in southern and eastern Australia but increased in the north. That means southern regions may be missing winter soakings, even as some indicators point to more intense extreme precipitation when it does occur.
Uncertainty remains for some hazards. Professor McAneney referenced U.S. hurricane modelling and said that averaging 18 climate models suggested it could take around 260 years before a clear statistical signal tying climate change to storm strength is detectable. Experts also warn that because very extreme damaging events are relatively rare, detecting trends statistically can be difficult.
Several indicators show a warming trend: between 2001–2011 the frequency of record high temperatures in Australia was 2.8 times greater for maximums and 5.2 times greater for minimums than the frequency of record lows. Australian regional sea temperatures are also about 0.6–0.7°C warmer than in 1900, according to the weather bureau.
The article highlights that evidence is mixed by hazard and region: some climate signals (temperatures, sea warming, heavier extreme rainfall) are clear, while trends in event frequency for certain disasters are harder to detect and influenced by greater exposure of people and assets. Investors should be aware of regional variability and the potential for economic and insurance impacts, but also recognise the scientific nuance—absence of clear statistical trends for some hazards is not proof there is no climate effect.

