A change in the weather?

Extremes like a drought aren't reliable indicators of global warming, reports Michelle Gilchrist

"The official record of one of Australia's worst droughts is depressingly familiar. "The five years had been intermittently dry over most of the country ... very dry conditions set in across eastern Australia during the spring, and became entrenched over the following months," the Bureau of Meteorology reports. "The long drought and its severe climax had devastated stock numbers, and began focusing attention on planning for irrigation, especially in the three states through which the Murray River flows."

To rural Australians, these words are no surprise. But the report is an official history of the great drought, which lasted from 1896 to 1901 .... Mathematicians and meteorologists say extreme events give little useful information about long-term climate change ... "Statistically speaking, looking at extremes is not the right thing to do if you are concerned with long-term changes," says Kevin Judd, of the University of Western Australia. "Extremes don't tell you much. Nothing may have changed yet you might get some extreme [events] occurring." Using everyday weather - high and low temperatures, their mean or average over time - can provide more useful information in a much shorter period....

"We are really, in many ways, at the very early stage of understanding the variability of the weather system," says William Kininmonth, a former head of the Bureau of Meteorology's National Climate Centre. "We understand the overall annual weather cycle, and we know quite a lot about the seasons and some of their variability. But there are other subtle factors that we don't really know." Kininmonth says new technology compounds the confusion. "We have started to develop computer models and people believe such models can tell us the answer. But they are ... only simulations," he says.

Australia's former chief meteorologist John Zillman agrees that the study of longterm weather remains a nascent science. "The atmosphere is a tricky and complex creature." says Zillman. "The climate is hugely variable and every now and again we do get extremes that go way beyond the previous records we have in Australia." Queensland in 1994 recorded almost 8OOmm of rain in one month, with figures off the scale for Brisbane. "If we had, say, another 300 years of highly reliable rainfall records for Australia, say back until the 1500s, we would perhaps be able to say it falls within the range of natural variability," Zillman says. "But we don't have those records and we therefore cannot say things like that. The shortage of records limits us."

Those claiming weather patterns have changed beyond natural variation turn to work by Bureau of Meteorology researcher Neville Nicholls, who argues the 2002-03 drought is at least the worst since the 1950s. Last month Nicholls wrote that the previous drought was a triple whammy of high mean temperatures, high evaporation rates and very low rainfall figures. especially in the Murray/Darling Basin. In 2002, the mean daily temperature in that region was 0.7C higher than previous records; and more than 1C warmer than the droughts of 1982 and 1994. Importantly, though, the Nicholls paper does not say the 2002-03 drought is the worst on record; instead, it can only say that the weather appears to have changed during the past 50 years....

Average rainfall in the southwestern corner of Western Australia has dipped by 10 percent to 20 percent since the '70s, puzzling meteorologists and scientists. "With rainfall, more so than temperature, you get very large year to year variations and these shifts mean we need much longer records before you can unequivocally say we have had a significant shift", says Neil Plummer of the Bureau of Meteorology.

Kininmonth agrees the climate has changed but believes it is for natural, not man-made, reasons. "All of the evidence points to the fact that the globe has warmed slightly over the past 100 years. That's something almost everybody agrees on. Whether that's unusual or not is what becomes debatable."

Plummer says the rise in mean temperatures may be within normal variability but adds that it appears to be unusual... Climate-change sceptic Warwick Hughes questions that data, arguing that the bureau's long-term mean temperature trend is tweaked. "My belief is that a true Australian trend has less warming than they show and their time series should extend a few decades back to cover the warm spell in the late 19th century." he says. Hughes argues the mean temperature trends also disregard the recent effect of 'urban heat islands" - cities such as Sydney and Melbourne and even smaller places such as Canberra, which create their own heat and raise ground temperatures."

More here

It should be noted that Australia is a large country (roughly the same size as the United States) and is one of only four large Southern hemisphere land-masses. So if there is no warming going on in Australia, it would be most odd if there was warming going on in the Southern hemisphere generally. And in one of the other large Southern land-masses (Antarctica) significant cooling (as signified by recent thickening in the West Antarctic ice sheet) seems to be going on at the moment