More than 15,000 people from all over the world try to locate exoplanets with similar conditions to the earth using ‘machine learning’ and the cloud.
The chances of discovering new planets multiply. The detection of stars is not only in the hands of scientists thanks to the Microsoft cloud and the algorithm developed by the team of researchers from the Institute of Astrophysics of the Canary Islands (IAC). More than 15,000 people from around the world have begun looking for exoplanets (planets outside the solar system) during the Global Azure Bootcamp event, an event about data storage and the search for celestial bodies. The headquarters of Microsoft of Madrid has hosted about 30 conferences and workshops. In one of the talks, those responsible for the project have taught attendees to run the program through which they will analyze 85% of our galaxy.
The day of Global Azure Bootcamp has had a mostly male audience dedicated to R + D + i or computing. However, professionals from other sectors have come to the event because anyone (not only astrophysicists) has the possibility of searching for stars through distributed computing (the use of thousands of computers spread around the world, which process data at the same time). that are stored in the Microsoft cloud). Each computer processes a certain amount of data, which when added together, produces a great advance in a short time.
“Exoplanet research is one of the fastest growing areas in the area of modern astrophysics,” says Enric Pallé, researcher at the IAC. The Kepler satellite (launched into space in 2009) orbited around the sun for almost 10 years. I was looking for extrasolar stars, especially those similar in size to the earth. Its mission concluded last year with a balance of 1,500 new planets detected.
However, Kepler was out of date. In July of last year, NASA launched TESS (Transiting Exoplanet Survey Satellite) into space. An observation mechanism for exoplanets that will be in operation for 18 months. TESS has four lenses that capture images with a fairly high resolution. The satellite divides the space into groups and analyzes them in detail. “It examines thirteen sectors for each hemisphere, and photographs each sector every 30 minutes for a month,” explains Alberto Marcos, head of universities at Microsoft. This new system, more innovative and powerful, could be able to discover a multitude of planets thanks to its field of vision 400 times larger than Kepler.
TESS detects the reflex movement that the planets leave the mother star and the darkening that they cause when they are in movement. “The brightness of light is measured from the images. If the planet and the star are aligned with the earth, there is an almost imperceptible descent in the image, “says Sebastián Hidalgo, Doctor in Astrophysics and researcher at the IAC.
In the last 25 years, more than 4,000 new worlds have been discovered around stars by detecting the reflex movement that these planets cause
From Saturday, all those who have executed the program (about 15,000 people) will analyze millions of data using machine learning algorithms. “They will not only look for candidates to be new planets, also stars that are disintegrating or binary stars (one star spinning around another),” adds Hidalgo.
In the last 25 years, more than 4,000 new worlds have been discovered around other stars by detecting the reflex movement they cause in their stars or during transit (periodic attenuation of starlight if the planet crosses the star disk during its orbit).
In addition, scientists have discovered that on average there is more than one planet per star and that small stars (the most abundant in our galaxy) tend to have several planets of a rocky nature such as the earth. The great challenge of TESS is the location of similar planets. “First with the same atmospheric characteristics and later, the search for life signs or biological markers,” concludes Diego Hidalgo, an astrophysicist at the IAC.