Signals from mobile phone masts have been used to measure rainfall patterns in Israel, scientists report.
Storms disrupt the signals between mobile phone masts
A team from the University of Tel Aviv analysed information routinely collected by mobile networks to make their estimates.
Writing in the journal Science, the researchers say their technique is more accurate than current methods used by meteorological services.
Similar work has been done by scientists in the UK.
Researchers funded by the Natural Environment Research Council (NERC) set up a mock mobile network near Bolton in the north of England to show how these signals can measure rainfall.
Further demonstrations were done in Germany and Italy as part of an EU project.
Researchers from all countries believe the technique has promise for issuing flood warnings.
Although both sets of work are similar, the team from Tel Aviv University say their work differs because they are using real data from an existing mobile network.
"I see our innovation as taking a network that is designed for one purpose and using it for another," Professor Hagit Messer-Yaron, of the University of Tel Aviv, told the BBC News website.
Both methods exploit the fact that the strength of electromagnetic signals is weakened by certain types of weather and particularly rain.
According to Professor Messer-Yaron, the effect was commonly seen in the days before cable television.
"Whenever there was a storm outside, you had a bad picture on your TV," she said. "This is well known."
To make the measurements the researchers analysed data collected automatically at mobile base stations around Tel Aviv, Haifa and Jerusalem.
The data is a by-product of mobile network operators' need to monitor signal strength.
If bad weather causes a signal to drop, an automatic system analysing the data boosts the signal to make sure that people can still use their mobile phones.
The amount of reduction in signal strength gave the researchers an indication of how much rain had fallen.
When they compared their estimates with measurements from traditional monitoring methods, such as radar and rain gauges, they discovered that the readings from all three closely matched.
But overall the new technique seems to give more precise measurements than radar and was able to monitor a greater area.
"It was win-win because we got both coverage and accuracy," said Professor Messer-Yaron.
Researchers in the UK have also shown that a similar method works using signals from global positioning satellites.
Scientists at the Institute of Engineering Surveying and Space Geodesy (IESSG) - part of the University of Nottingham - have demonstrated that the fluctuations in these signals are a good measure of atmospheric humidity.
However, the team from Tel Aviv University believes one of the advantages of its method is that it can measure rainfall at the surface and the technique therefore gives an accurate picture of the weather on the ground.
At the moment, meteorological services have to use rain gauges, which are expensive and therefore not widespread in some countries.
They are also not accurate at measuring low rates of rainfall and in places like the UK they are prone to freezing up.
But the information necessary for this novel approach is effectively free, continuous and comes from a dense network of masts that already span almost the entire globe.
Professor Messer-Yaron believes the technique could act as a cheap and valuable complement to existing systems for meteorological services.
The challenge is to get mobile network operators to routinely provide the data they collect for a national monitoring systems.
The next step for the team is to make use of the mobile phone users themselves - to start analysing their signal data to see if even greater accuracies can be achieved.
To do this, Professor Messer-Yaron must work out how to differentiate between changes in signal strength that are due to mobile phone users moving around and those that are due to weather conditions.
"This is a scientific challenge that is still ahead of us, but I believe it is possible," she said.
The UK project is a collaboration between the University of Essex, University of Salford, the Rutherford Appleton Laboratory (RAL), the Met Office, United Utilities and the Environment Agency, Crown Castle International and Your Communications