By 
In order to better forecast the weather, a lot of atmospheric data needs to be collected and crunched, and all this processing takes up a lot of time. This means that during a day, these numbers can only be processed a certain number of times, and by the time it is delivered to users, it might be out of date based on newer data.
This is where the use of machine learning comes in, and that’s something that Google is implementing to its Weather app in a new feature called “Nowcast”. Google actually unveiled the tech a few years ago, but it is now finally being used in its Weather app.
Image credit – 9to5Google
The data is being pulled from various sources, like Multi-Radar/Multi-Sensor System (MRMS) and High-Resolution Rapid Refresh (HRRR) data from the National Oceanic and Atmospheric Administration, for example. It will be freshed mutliple times per hour, meaning that it can deliver very close and semi-accurate weather forecasts for users.
The Nowcast will be displayed on top of the regular forecast where for example, it can tell users if it’s abot to rain soon, whether it will be light or heavy, and show users a timeline of the various states of rain. It’s actually pretty cool and if you’re someone who relies on weather information quite a bit, then this might come in handy. At the moment the feature is only available in the continental US and for devices like the Pixel Tablet and Pixel Fold, but hopefully it will expand to more regions soon.
Source: 9to5Google
Comments