A change in the weather no-one predicted: forecasters are getting it right

Weather forecasts are no longer in vane. Syncop8ted/Flickr

Whenever there’s severe weather anywhere in Australia attention is focused on forecasts issued by the Bureau of Meteorology. People want to know how good the forecasts were, and if they could be improved.

The early days of forecasting

Of course this recent fascination with the forecast is just part of our eternal interest in the weather.

It is just over 250 years since Samuel Johnson wrote in The Idler, “It is commonly observed, that when two Englishmen meet, their first talk is of the weather; they are in haste to tell each other, what each must already know, that it is hot or cold, bright or cloudy, windy or calm.”

A later aphorism often (but probably wrongly) attributed to Mark Twain says, “Everybody talks about the weather, but nobody does anything about it.”

But scientists have always wanted not just to talk about the weather, but to do something about it as well.

Scientific efforts to forecast the weather date from the mid-19th century. On 30 June 1854, there was a debate in the British House of Commons about establishing what would become the Meteorological Office.

Mr M J Ball, the Member for Carlow, said, “observations made…upon land as well as at sea would be collected, as, if that were done, he anticipated that in a few years, notwithstanding the variable climate of this country, we might know in this metropolis the condition of the weather 24 hours beforehand.”

The response from the House was “Laughter”.

Despite the amusement of the politicians, the Meteorological Office was established under the guidance of Robert FitzRoy, Darwin’s Captain, and he started producing weather forecasts soon after.

But he had his critics. Referring to his forecasts, an editorial in The Times on 18 June 1864 said, “Whatever may be the progress of the sciences, never will observers who are trustworthy and careful of their reputations venture to foretell the state of the weather.”

Coming out of the meteorological closet

The Australian Government established our Bureau of Meteorology in 1908, just over a hundred years ago. So, what progress has been made in weather prediction?

Is it still a field that scientists “careful of their reputations” should avoid? Is even the idea of forecasting the weather still regarded as amusing?

That was certainly still the case when I joined the Bureau of Meteorology in 1970 – we meteorologists were careful not to mention our job at parties, because of the merciless ribbing that would surely follow.

I sense that this doesn’t happen nowadays, and for good reasons – weather forecasts are now generally accurate and useful.

Saving lives and homes with forecasting

A very severe windstorm did a great deal of damage in Melbourne on April 2 2008. I have a good reason for remembering this storm: a 6 metre branch from a neighbour’s tree went through the roof of my home that day.

I wasn’t the only person affected by the storm. The Victorian SES received 3300 requests for assistance (mainly for trees through roofs) and 200,000 homes were without power for various periods.

The Bureau of Meteorology did an excellent job of forecasting this storm. Five days prior to the storm they had forecast “windy” for that day.

The day before the storm they issued a severe weather warning. And on the morning of April 2 the Bureau issued a forecast of wind gusts exceeding 90 km/hr.

Of course these forecasts didn’t stop the branch going through my roof – but they should have allowed action to minimise some storm damages (by ensuring I didn’t park my car under a tree, for example).

The Bureau’s forecasts in the days leading up to Black Saturday, February 7 2009, were also very accurate. On the Wednesday preceding the bushfires, the Bureau forecast a Melbourne maximum temperature of 43˚C for Saturday.

The highest temperature previously recorded in Melbourne during February was 43.2˚C, so this forecast of extremely hot conditions was very unusual.

The forecast of extreme weather led to the Premier of Victoria using apocalyptic language to warn the public in fire-threatened areas to have their fire plans ready.

The improving accuracy of forecasts can be demonstrated by a simple statistic: forecasts of Melbourne’s maximum temperature four days in advance are now more accurate than a one-day-ahead forecast was in the early 1970s, when I joined the Bureau.

Why have forecasts improved?

There are two main reasons for the recent improvement in forecasting.

First, modern high-speed computers have allowed the development and operational use of skilful weather forecast models, based on well-established and understood physics of the atmosphere.

Second, satellites have dramatically increased the amount of data to input into these computer weather forecast models. We can expect to see even more accurate forecasts in the future.

But despite the accuracy of modern weather forecasts, not everyone enjoys the weather forecast. Zadie Smith’s novel White Teeth has one character who is especially anti-forecaster.

“And then there was Hortense’s horror of weather reports. Whoever it was, however benign, honey-voiced and inoffensively dressed, she cursed them bitterly for the five minutes they stood there, and then, out of what appeared to be sheer perversity, proceeded to take the opposite of whatever advice had been proffered (light jacket and no umbrella for rain, full cagoule and rain hat for sun).”

Although weather forecasters are doing a great job with predicting heat waves and tree-uprooting, roof-destroying storms, perhaps we cannot expect everyone to take notice, quite yet.