Smartphone batteries contain tiny temperature sensors, designed to keep the phone from overheating. While those sensors do measure the heat generated within the phone, their readings are also affected by the temperature of the phone’s external environment. With that in mind, British app developer OpenSignal has created a system that allows multiple users’ phones to provide real-time, location-specific weather reports.
The system is based around the company's existing OpenSignal app, which was designed to collect data sent voluntarily from users’ Android phones in order to create maps of Wi-Fi access points and phone coverage.
NEW ATLAS NEEDS YOUR SUPPORT
Upgrade to a Plus subscription today, and read the site without ads.
It's just US$19 a year.UPGRADE NOW
That app can also provide phone temperatures. When a sampling of London-based users’ phone temperatures was compared to local weather temperatures for the same time period, a correlation between the two sets of numbers became apparent – although the temperatures weren’t the same, they did go up and down by the same amount.
The OpenSignal team then went on to analyze readings and weather reports from users in Los Angeles, Paris, Mexico City, Moscow, Rome, San Paulo and Buenos Aires. The same correlation popped up. Once the consistent difference between the phone and weather temperatures was compensated for, it was found that the phone temperatures could be used to determine the weather temperatures within an accuracy range of 1.5ºC (2.7ºF).
Of course, some phones might be running particularly hot, due to the task they’re performing or the place that they’re being kept. By incorporating the data from thousands of users at once, however, the overall numbers still end up meshing with the ambient temperature. As more users get on board and the database increases in size, the accuracy should further improve.
The company has now launched the spin-off WeatherSignal app, for use in phones equipped with air temperature, humidity and pressure sensors. Ultimately, it is hoped that the app could be used in a crowd-sourced weather forecasting system, which would provide real-time weather information for areas as specific as a given city block. It could perhaps even be combined with the CitiSense app/device, that provides localized crowd-sourced air quality readings.
A paper on the research was published today in the journal Geophysical Research Letters.
Source: American Geophysical Union