Startup predicts weather using cell phone signals

ClimaCell claims its forecasts are 60 percent more accurate than most others

Boston-based startup ClimaCell is using cell phone signals and street cameras to create weather forecasts. The company says they’re more accurate too: as much as 60 percent better than existing providers such as the National Oceanic and Atmospheric Administration.

ClimaCell analyses the radar signals used by other forecasters and complements them with signals from millions of wireless devices, which it says results in a much more accurate weather prediction.

Luke Peffers, who runs one of the company’s research teams, previously worked for the US government. His job was to determine whether nuclear test bans had been violated using forensic analysis of the atmosphere.

The firm is working on a new mathematical model that can be adapted to focus on a specific region, the type of weather and the frequency of updates a subscriber wants. It is interested in working with national forecasters as well as providing weather forecasts to private businesses.