ClimaCell Bags $15M to Predict Weather from Wireless Network Signals | Xconomy
Local meteorologists might avoid being the butt of so many jokes, if the latest crop of weather technology startups have their way.

A number of young companies are trying to improve the methods for tracking and predicting the weather. Some are tackling the problem from space, like Spire Global, whose network of small satellites gather data for short-term weather forecasting models, among other uses. Spire announced earlier this month it raised $70 million in a Series C funding round. Other ventures capture weather data from Earth’s surface, like Madison, WI-based Understory, which installs solar-powered sensor devices on top of buildings. The company, which has raised at least $9.7 million from investors, has set up networks of sensors in five metropolitan areas thus far.

Now comes ClimaCell, a Boston-based weather tech startup that today announced a $15 million Series A funding round led by Canaan Partners (more on the investors below). What makes ClimaCell interesting is the company isn’t deploying any of its own sensors or other hardware. Instead, it uses software to measure the ways in which weather impacts wireless communication networks.

The basic principle is that weather and atmospheric conditions affect wireless signals traveling through the air between cellular towers and mobile phones and other devices. ClimaCell CEO and co-founder Shimon Elkabetz says his company has devised a way to glean weather insights from those signal effects.

“We reverse engineer, let’s call it, what happens to the signal,” Elkabetz says in an interview. The company says it uses several methods to access the signal data, including through partnerships with the wireless networks’ owners. Elkabetz declined to share more details.

ClimaCell combines that network information with data from traditional sensors, such as satellites and radar, to provide minute-by-minute weather observations, as well as make local weather forecasts for up to six hours into the future and at street-level precision, the company claims. Currently, most weather forecasts for the next half hour, say, are based on calculations performed 90 minutes ago, Elkabetz says.

Read the full article originally posted here.