August 13, 2013
Using Smartphone Batteries To Better Predict The Weather
Michael Harper for redOrbit.com – Your Universe Online
The weather in London is often unpredictable and is able to jump from bright and sunny to dark and cloudy at a moment’s notice. This provides a challenge for smartphone weather apps, which are often pulling data from a central source. A group of app developers are looking to make localized weather predictions more accurate by using a thermometer already inside many smartphones that is currently used to determine the temperature of a smartphone’s battery. Doing so may help accurately determine the weather conditions in a given area by pulling data from many sources rather than one.
“Just sort of for fun we started looking to see if there was a correlation with anything else,” explained James Robinson, the cofounder of OpenSignal, an app development company based in London. “We got some London weather data for comparison and found the two sets of temperatures were offset, but they had the same sort of shape.”
“It was amazing how easily the correlation sort of popped out. We didn’t do any handpicking of data—it sort of just emerged.”
The data described the subtle differences in temperature between areas in a certain location such as London. While some areas may have bright and sunny skies, others were much colder with clouds overhead. After working with several weather experts, Robinson and team were able to develop a system that took normal smartphone operations into consideration to better determine the ambient air temperature.
Smartphones are often equipped with thermometers that can determine if the battery is overheating in order to protect the rest of the device. For instance, a smartphone might run hot when performing a graphics intense operation, such as playing a 3D game or watching a movie. On the other hand, smartphones that are sitting dormant in an air-conditioned room will be running at much cooler temperatures.
The challenge, then, was to create an app that is able to take these environments into consideration and determine the precise temperature and conditions of each smartphone, compile this into a larger pool of data, and use it to more precisely predict the conditions in a given area.
“There’s the wider promise when logging all this information that there will be something really interesting you can understand,” explained Robinson. His OpenWeather app is currently in beta testing. Robinson also plans to offer the data collected to academic researchers to conduct their own tests.
“The most obvious application is climate and weather tracking.”
Robinson believes that this data can go a long way in terms of building a better system for understanding incoming weather conditions, but also understands that weather stations are still critical in measuring actual data rather than anecdotal information.
“The challenge is whether we can take this technique and use it in places where we don’t already have reliable weather information to retune the model,” he said.
“That’s something we’re still working on.”