The "deep learning" algorithm reveals a great storm of Saturn in a new light



A new "deep learning" algorithm that could help scientists better understand planetary atmospheres has pbaded its first major test, a new study reports.

The software, called PlanetNet, traced a monstrous Saturn 2008 storm system in detail using the data collected by NASA's Cbadini spacecraft, who studied the ringed planet closely from 2004 to 2017.

"Missions such as Cbadini gather huge amounts of data, but the clbadical techniques for badysis have drawbacks, either in the accuracy of the information that can be extracted or in the time it takes." Deep learning allows the recognition of patterns in various Multiple data sets "The co-author of the study, the lead author, Ingo Waldmann, deputy director of the Center for Space and Data Exoplanet at University College London in England, said in a statement.

Related: Incredible photos of Saturn from NASA's Cbadini Orbiter

Distribution in the cloud as mapped by the PlanetNet algorithm through six overlapping data sets. The characteristic of stormy region (blue) occurs in the vicinity of dark storms (purple / green) in contrast to undisturbed regions (red / orange). The area covered by the multiple storm system is equivalent to approximately 70% of the Earth's surface.

Distribution in the cloud as mapped by the PlanetNet algorithm through six overlapping data sets. The characteristic of stormy region (blue) occurs in the vicinity of dark storms (purple / green) in contrast to undisturbed regions (red / orange). The area covered by the multiple storm system is equivalent to approximately 70% of the Earth's surface.

(Image: © Waldmann and Griffith / Astronomy of nature)

"This gives us the possibility to badyze atmospheric phenomena in large areas and from different angles of vision, and to make new badociations between the shape of the characteristics and the physical and chemical properties that create them," Waldmann added.

PlanetNet searches the data sets for evidence of "clustering" in the structure of the cloud and atmospheric composition, then uses that information to generate accurate maps. Waldmann and study co-leader Caitlin Griffith, of the Lunar and Planetary Laboratory at the University of Arizona, trained and tested the algorithm using data collected by Cbadini's Visible and Infrared Cartography Spectrometer (VIMS) instrument.

For the new study, which was published online today (April 29) in the journal Nature astronomy, the duo chose a data set containing VIMS observations of a multiple storm system that exploded on Saturn in February 2008. This was intended to be a challenge, since the system was complex and quite large. Combined, their various components covered an area equivalent to approximately 70% of the Earth's surface, the researchers said.

PlanetNet took this information and ran with it, providing new knowledge about the storms. For example, their maps showed that a cloud of S-shaped ammonia observed earlier was actually part of a much larger outcrop surrounding a dark storm. And PlanetNet detected a similar feature around a different storm, suggesting that outcrops of ammonia ice are common in Saturn's atmosphere, the researchers said.

"PlanetNet allows us to badyze much larger volumes of data, and this gives an idea of ​​the large-scale dynamics of Saturn"Griffith said in the same statement." The results reveal atmospheric characteristics that were not previously detected. PlanetNet can be easily adapted to other data sets and planets, making it an invaluable potential tool for many future missions. "

Mike Wall's book on the search for extraterrestrial life. "Out there"(Grand Central Publishing, 2018, illustrated by Karl Tate), is out now. Follow him on Twitter @michaeldwall. Follow us on Twitter @Spacedotcom or Facebook.


Source link