Meteorologists should add the words 'don't know' to their vocabulary

More broadly, it is interesting to hear what meteorologists in a more general context say about the accuracy of their forecasts.

The BOM says that "the chance of rainfall" forecasts continue to be reliable. "When we say it is going to be a dry day, it is dry 95 per cent of the time. Our four-day forecasts today are as good as a one-day forecast was in the 1980s".

A US Government site linked to NASA states that "A seven-day forecast can accurately predict the weather about 80 percent of the time and a five-day forecast can accurately predict the weather approximately 90 percent of the time. However, a 10-day or longer forecast is only right about half the time. Meteorologists use computer programs called weather models to make forecasts. Since we can't collect data from the future, models have to use estimates and assumptions to predict future weather. The atmosphere is changing all the time, so those estimates are less reliable the further you get into the future".

The ABC noted that "the Bureau of Meteorology provides seven-day forecasts, and even they would admit it gets a bit error-prone towards the end of the week-long outlook. The forecast for tomorrow and the day after is pretty much spot-on; up to day five it's not too bad, and by the time you reach day seven things get a little shaky". I would concur with this observation, which applies to other forecasters as well.

The Ten Day Precipitation Forecast for Australia has proved broadly indicative (though generally on the high side) for rainfall in the first five days of its forecast period. The forecast for the next five days is a lot less so and can change radically from one day to the next. I have come to broadly similar conclusion for Weatherzone's 28 day predictions, which dare to predict up to a month ahead.

The bottom line seems to be that weather forecasts for up to a week ahead are reasonably reliable, with reliability increasing as we get closer to the day being forecast. In stable weather conditions (especially those dominated by a slow moving high pressure system) accurate forecasts (usually for continued dry conditions) can sometimes be extended for a bit longer. Longer term forecasts, however, should be taken with a grain of salt. "Climate predictions" for coming decades should be treated with even greater scepticism.

As a general comment, there are few people more dangerous, if listened to, than those who claim to know when they really don't. I believe that meteorologists fall into this category in respect of their medium and long-term forecasts, or (more particularly) insofar as what the public interprets them as saying. The BOM would be more helpful if it said it could not accurately forecast more than a week or two in advance, instead of issuing long term and seasonal outlooks full of gobbledegook and veiled caveats.

Worthless seasonal outlooks have cost BOM a lot of credibility in the bush. Such loss of standing is not without precedent. The same happened with the Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES). The agricultural outlook conferences of its predecessor (the BAE) were a big event 30 years ago but (following rounds of dud predictions) few in the bush bother to listen to them anymore. BOM is facing the same outcome in respect of its seasonal outlooks.

In 2018-19 the net cost of BOM services to the taxpayer was $313.7 million (made up of expenses of $406.9 million less own source revenue of $93.2 million).