The director of the broker, Francisco Förster, talked to the international media about the challenges involved in being chosen by the Rubin Observatory as the only project in the southern hemisphere that will have access to its data.
“The future of astronomy is in AI”, headlined the prestigious US magazine Forbes after covering the international event “Unveiling the dynamic universe: cosmic streams in the era of Rubin“, or simply Cosmic Streams, which brought together in Puerto Varas scientists currently working at the main observatories of the planet and who, with the start-up in 2025 of the Vera Rubin Observatory at Cerro Pachón in Coquimbo, will have to monitor a gigantic amount of astronomical data per night.
The meeting took place in December in southern Chile, was organized as part of the 10 years of operation of the Instituto Milenio de Astrofísica (MAS), in collaboration with the Center for Mathematical Modeling (CMM) of the University of Chile, and was covered by this North American magazine. The report states that “the biggest buzz in ground-based astronomy these days is the soon-to-be-completed Rubin Observatory and its upcoming Large Synoptic Survey of the wide-field sky. From atop a lonely mountain in the Atacama Desert of northern Chile, the observatory’s 8.4-meter optical telescope will scan the southern sky approximately every three to four nights.”
He adds that “for more than a decade, their observations will generate an unprecedented amount of raw data, much of it related to so-called transient astronomical phenomena. Such events are usually active for short periods of days or weeks and can involve highly energetic and destructive astrophysical phenomena, such as supernovae or gamma-ray bursts. In fact, the LSST (Legacy Survey of Space and Time) survey is expected to generate so much data that it will require a level of scientific data management that will use software and technology bordering on artificial intelligence.”
Francisco Förster, MAS – CMM researcher and lead organizer of the conference spoke to Forbes at the time, noting that “the telescope’s repeated scans of its 9.6 square degree field of view (roughly the size of 40 full moons) will use a 3.2-megapixel camera to create a nightly plethora of some 10 million astronomical alerts (…) Within 60 seconds of hitting the telescope’s primary mirror, the photons from these events will be transferred through a high-speed optical relay to huge amounts of cloud storage. ..) Within 60 seconds of hitting the telescope’s primary mirror, photons from these events will be transferred through a high-speed optical relay to huge amounts of cloud storage. From there, this raw data will be processed and sent to astronomers around the world by so-called alert brokers. An alert broker is an intermediary between the survey telescope, its observational science data and the follow-up telescopes.”
One of these intermediaries is the ALeRCE broker, which was created under the auspices of MAS and CMM, with the Data Observatory Foundation and the University of Concepción as partners. This Chilean project was selected by the Vera C. Rubin to process the LSST data, through an interdisciplinary group of astronomers, engineers, experts in computer science and statistics. All of them have been preparing for this challenge for years by training the artificial intelligence -which the report deepens- with the analysis of data from the Zwicky Transient Facility (ZTF) telescope. This work has enabled ALeRCE to process some 300 million alerts in real time and to discover at least 20,000 supernova candidates.
“And while the telescope will not use thinking artificial intelligence in the classical sense of machine thinking, it is clear that the future of astronomy is through A.I. The amount of data that future telescopes will produce will demand A.I. capability that allows astronomers to analyze raw data with speeds and accuracies that would hitherto be considered science fiction. If we are going to apply machine learning, it must be super fast. You can’t wait more than one second per object to classify it,” Förster states in the note.
In closing, Forbes asks, “Will human eyes continue to be needed to interpret astronomical data?” This was answered by Alexander Gagliano, a postdoctoral researcher at MIT’s Institute for Artificial Intelligence and Fundamental Interactions, who said that “humans have an innate talent for generalizing to completely new situations: you can walk into a room with a lamp you’ve never seen before and figure out how to turn it on, says Gagliano. Most algorithms can’t do this kind of fuzzy reasoning.”
By Alonso Farías Ponce, CMM journalist and Makarena Estrella, MAS journalist.