Lightning Spotting Sats Used to Study Fireballs

Scientists have repurposed data from a lightning spotting instrument aboard the GOES 16 and 17 weather satellites to find bolides using machine learning.

IMAGE: Each year, spotting scopes and ground telescopes witness only about a dozen or so bolides with enough detail to extract data from them. In a new study, researchers used machine learning to increase the detection rate of bolides with data from lightning-spotting satellites. CREDIT: Jijian Fan, CC BY-NC-ND 2.0

Whether looking for stars around the black hole at the center of our galaxy or looking for fireballs in the night sky, scientists appreciate the struggle of observing the small, the distant, and the rare. And when it comes to those fireballs, we have mostly relied on a few all-sky video camera systems, scattered dashcams, and eye-witness accounts. So astronomers are always trying to find a way to collect more data and, of course, do so at a low cost.

Now, a team of scientists led by Jeffrey Smith from the SETI Institute has not only repurposed data from two weather satellites, but they have also programmed a machine-learning algorithm to analyze the data. First off, the data they are using was lightning mapping data collected by the GOES 16 and 17 satellites. Second, the machine learning analysis increased the detection rate of fireballs from 0.2% to 80%.

And that’s amazing news for planetary defense researchers.

Here’s how it works. The Geostationary Lightning Mapper instruments aboard the two satellites detect a bright flash. The…