Tue, August 17, 2021


You need to sign in to add a comment

Sign In

Not an OR Society Member

Find out the benefits of being a member

Become a member

Algorithm used to predict where lightning will strike

A machine-learning algorithm developed at the University of Wisconsin uses artificial intelligence to accurately predict the location of lightning up to an hour ahead of time.

Developed by John Cintineo, the programme works alongside the space-based lightning monitoring system GOES-R to make predictions of where lightning will strike next.  

The National Oceanic and Atmospheric Administration (NOAA) launched the GOES-R satellite into space with a range of sensors that could track severe storms like tornadoes. For example, lightning intensity can be used to indicate when tornadoes are about to form.

All the data gathered by GOES-R is then fed into the machine learning programme. It uses the data to understand what atmospheric patterns exist before lightning forms and also during lightning events.

It analyses current weather patterns and identifies what past examples are similar to this. The algorithm then uses the information to make predictions on where the next lightning strike will be.

This technology is still being developed and currently cannot give exact locations for strikes, only areas and the percentage chance of lightning occurring.

It is hoped that this technology can be used to deliver early warnings for lightning events that could save lives and homes in the future.

It is also not the first time that the GOES-R satellite has been used for groundbreaking weather research. A team at NASA recently been able to use lightning to help predict hurricane intensity.

NASA scientist Patrick Duran said, "we were able to prove that the lightning flashes in Hurricane Dorian were larger and more energetic when the storm was intensifying than when it was weakening… We also argue that changes in the location of lightning flashes could help to identify processes that affect a storm's intensity."

You can read more about this project here and here.