How we can keep trusting weather forecasts
How we can keep trusting weather forecasts

How we can keep trusting weather forecasts

Information from precipitation radars can help us better predict showers and storms. But the quality of these radars can deteriorate. Iwan Holleman has developed a universally applicable method to monitor their quality. He recently showed that this method not only works for individual radars, but also for an entire network of radars. “To forecast the weather, you need good observations from your neighbours, not just your own.” 

You’re probably familiar with using an app on your phone to quickly and easily see what the weather will be like today. But a lot of data has to be exchanged on the back end to make that possible. Companies like Buienradar buy data from the Royal Netherlands Meteorological Institute (KNMI), which operates two radars to detect precipitation and wind over the Netherlands. Meteorologists use these radar data to create numerical models for weather forecasts and warnings about dangerous weather situations, such as heavy precipitation or whirlwinds.

High-quality precipitation radars are very important. “Everyone looks at these data”, Holleman says. “And it’s not just the Buienradar app that uses radar data: they’re also used by aviation, fire brigades, water management authorities, you name it. For example, consider the heavy rainfall in Limburg last summer. At times like that, we need to be confident that the radars are working properly and that we can use them to make reliable weather forecasts.”

Sunlight

Some time ago, Holleman developed a method that makes it easy to assess whether radar observations are still correct. “Then you can act fast and, for example, carry out repairs.” The method has since been used by KNMI and various other national meteorological institutes in Europe, the United States, Canada, Indonesia, Australia and South Africa.

The method uses the sunlight that is incidentally detected by the radar antenna several times a day. Because the position and strength of the sun are known data that are very precise, the sensitivity and alignment of the radar can be deduced from these solar signals. These quantities determine the accuracy of the precipitation and wind observations made by the radar.

Your neighbours’ radar data

In their most recent publication, Holleman and colleagues show that the method not only works for individual radars, but also for a continental-scale network of radars. “The ultimate test”, says Holleman. “Our monitoring method provides useful information, as we have now demonstrated for a network of more than 140 radars across the contiguous US. We chose the American network because the precipitation radars throughout the country are very similar. This is more difficult in Europe because radars in different countries have different hardware and settings.”

Precipitation data are increasingly shared between countries. This is something Holleman worked on in his previous job at KNMI. Now that he has demonstrated that the monitoring method also works well in a large network of radars, this offers new opportunities, including for the data exchange in Europe. “As the quality of radars improves, so does the usability of the data. To forecast the weather, you need observations from your neighbours, not just your own. Christoph Buys Ballot, 19th-century meteorologist and the first director of KNMI, already recognised that,” Holleman concludes.

Contact information

Theme
Innovation, Laws of nature