We all have a reasonable understanding of how weather forecasting works. Satellites and ground-based radar monitor weather conditions across the globe and those conditions are then fed into computer systems that use complex mathematical formulas to predict the weather days or weeks out.
But there’s a startup that says it can predict the weather 60% more accurately than the current models used by government agencies—and it can do so by using the wireless signal of your smartphone and other devices. ClimaCell, based in Boston, was founded in 2015 and says its technology uses the quality of the reception of wireless signals to our devices as a proxy for local weather conditions. Wireless signals can deteriorate based on factors such as precipitation and air quality. The strength of these signals can therefore be used to predict weather conditions. As MIT Technology Review explains:
The approach makes sense, in principle. Other forecasters use proxies, such as radar signals. But by using information from millions of everyday wireless devices, ClimaCell claims it has a far more fine-grained view of most of the globe than other forecasters get from the existing network of weather sensors, which range from ground-based devices to satellites. (ClimaCell taps into those, too.)
ClimaCell’s technique essentially turns every wireless device in the world into a potential weather detector, which gives it a lot more weather “sensors” than typical weather forecasting institutions have. For example, ClimaCell says there are a little less than 12,000 traditional weather sensing apparatuses worldwide. This includes 5 satellites, 940 Doppler radar stations, and 11,000 weather stations.
But thanks to ClimaCell’s technique, there are now over 567,000,000 million additional devices on the planet that can help detect the weather including 4 million wireless networks and 450 million Internet of Things devices, including smartphones, laptops, and tablets—basically anything that has a wireless signal.