Young stars, like young people, tend to soften flares. But star torches can burn anything around them, including the atmosphere of nearby planets that are beginning to form.
By finding out how often young stars erupt, scientists can better understand where to look for habitable planets. Previously, searching for these flares had to look through thousands of measurements of fluctuations in star brightness, known as light curves.
Now an international team of scientists, including Dr. Ben Montet of UNSW Sydney, used machine learning to make searches faster and more effective
Scientists taught a neural network – a type of artificial intelligence – to detect the tell-tale light patterns of a star torch.
“With the help of the neural network, we were able to find more than 23,000 torches in thousands of young stars,” said Dr. Montet, Scientia lecturer at UNSW Science and co-author of the study.
“Finding star flares – which can be deadly to the evolving atmosphere of nearby planets – can help us figure out where to look for habitable planets.”
The results, published over the weekend in the Astronomisches Journal and the Journal of Open Source Software offer a new benchmark for the use of AI in astronomy, as well as a better understanding of the evolution of young stars and their planets.
“When we say young, we only mean one million to 800 million years old,” said Ms. Adina Feinstein, a graduate student at the University of Chicago and first author of the paper.
“At this point in time, planets are still forming near a star. This is a particularly fragile time, and a flare from a star can easily vaporize water or atmosphere that has been collected. ”
Throw a neural network
NASA’s TESS telescope on board a satellite that has been orbiting the earth since 2018 was specially developed for the search for exoplanets. Flares from distant stars appear in TESS’s images, but traditional algorithms have a hard time figuring out their shape from the background noise of star activity.
However, neural networks are particularly good at looking for patterns – like Google’s AI which selects cats from Internet images – and astronomers are increasingly looking for patterns to classify astronomical data.
Mrs. Feinstein and Dr. Montet worked with a team of scientists from NASA, the Flatiron Institute, the Fermi National Accelerator Laboratory, the Massachusetts Institute of Technology, and the University of Texas at Austin to put together a set of identified flares and non-flares for training on the neural network.
“The neural network did very well at finding small flares,” said Dr. Montet, who was the main researcher on the study.
“These are really difficult to find with other methods.”
Once the researchers were happy with the neural network’s performance, they applied it to the entire dataset: more than 3,200 stars.
They found that stars like our sun have few flares and that these flares fall off after about 50 million years.
“This is good for promoting the planetary atmosphere – a calmer stellar environment means the atmospheres have a better chance of survival,” said Ms. Feinstein.
In contrast, cooler stars known as red dwarfs flickered out much more frequently.
“Red dwarfs are home to small rocky planets. If these planets are bombed at a young age, it could prove detrimental to the maintenance of an atmosphere, ”she said.
In search of habitable planets
The results will help scientists understand how likely it is for habitable planets to survive around different types of stars and how atmospheres are formed. This can help them find the most likely places to look for habitable planets elsewhere in the universe.
The scientists also studied the relationship between star flares and star spots as we see them on the surface of our own sun.
“The most blotchy spot our sun will ever get is maybe 0.3% of the surface,” said Dr. Mounts.
“For some of these stars that we see, the surface is basically all points. This reinforces the idea that points and flares are connected as magnetic events. ”
Next, the scientists want to adjust the neural network to look for planets lurking around young stars.
“Currently we only know a dozen or so younger than 50 million years, but they are so valuable for learning how planetary atmospheres develop,” said Ms. Feinstein.
Dr. Montet will also expand this neural network framework at UNSW.
“We will use the same methods to find young planets in the same dataset,” he said.
“Hopefully this will lead to a ‘rise of the machines’ where we can use machine learning algorithms to find a number of exciting new planets using the same methods.”
Recently discovered planets that are not as safe from star flares as initially thought
Adina D. Feinstein et al. Flare statistics for young stars from a convolutional analysis of neural networks of TESS data, The Astronomical Journal (2020). DOI: 10.3847 / 1538-3881 / abac0a
Provided by the University of New South Wales
Quote: Astronomers engage AI to search for “deadly” baby star eruptions (2020, October 26th), released on October 26th, 2020 from https://phys.org/news/2020-10-astronomers-ai-lethal -baby-star.html were retrieved
This document is subject to copyright. Except for fair trade for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.
These were the details of the news Astronomers engage AI in search of “deadly” baby star eruptions for this day. We hope that we have succeeded by giving you the full details and information. To follow all our news, you can subscribe to the alerts system or to one of our different systems to provide you with all that is new.
It is also worth noting that the original news has been published and is available at de24.news and the editorial team at AlKhaleej Today has confirmed it and it has been modified, and it may have been completely transferred or quoted from it and you can read and follow this news from its main source.