Weather forecasting is notoriously difficult (and ungrateful), but scientists are using the latest computer technology to make the process more accurate. Using a wealth of data, they hope to be able to predict possible weather conditions with greater certainty.
If their efforts are successful, this could allow meteorologists to anticipate important events in advance, not only in the near future, but even for decades.
This has the potential to save lives in the event of extreme weather conditions such as hurricanes, droughts and floods.
Experts from the Australian Bureau of Meteorology are using computer simulations to analyze a wealth of information about the planet's climate.
They rely on a new weather forecasting method called a "forecasting ensemble" that produces a number of possible results.
This allows you to simulate different conditions, which gives the most likely result.
By putting as much detail into their system as possible, including a range of different predicted climate change scenarios, the team hopes to be able to predict weather for decades into the future.
And in a world expected to soon suffer from more extreme weather caused by climate change, such a system could save millions of lives by offering early warning. Currently, detailed weather forecasts can only extend up to a month. But their accuracy decreases after about a week.
Promotional video:
To test the accuracy of the system's forecasts, the Australian team compared model predictions based on actual weather events over the past 100 years; with the results that actually happened.
According to the data, the model is drawing the right conclusions, which could mean a robust new method for long-term weather forecasting.
The modern method of weather forecasting has not changed significantly since the 1920s
In 1917, the British mathematician Lewis Fry Richardson, working with a group of forecasters on the Western Front during World War I, decided to improve the methods of weather forecasting at the time.
He developed a numerical process that used data collected from around the globe to predict the upcoming weather, a method he published in 1922.
Although the data collected today is more complex and varied, the basic principle that Richardson created still holds true.
And on condition: experts use computer simulations to analyze four petabytes of information. That's as much as one could store on the hard drives of eight million laptops. Analyzing this data requires processing power equivalent to 20,000 laptops.
Serg kite