On March 19, 2025, the European Space Agency’s (ESA) Euclid mission released its first major batch of survey data, providing an early look into the vast cosmic structures it is set to map.
This dataset includes a preview of Euclid’s deep fields, revealing hundreds of thousands of galaxies in different shapes and sizes, offering a glimpse into the large-scale cosmic web that defines the structure of the Universe.
A first look at Euclid’s deep fields
This data release consists of three large mosaics covering a substantial portion of the sky. These images contain galaxy clusters, active galactic nuclei, and transient astronomical phenomena, offering a wealth of information for scientists.
Among the most notable features is a classification survey that has cataloged over 380,000 galaxies and identified 500 gravitational lens candidates, leveraging a combination of artificial intelligence and citizen science contributions.
ESA’s Director of Science, Professor Carole Mundell, emphasized the mission’s significance:
“Euclid shows itself once again to be the ultimate discovery machine. It is surveying galaxies on the grandest scale, enabling us to explore our cosmic history and the invisible forces shaping our Universe.”
The Euclid mission, launched in July 2023, began routine science observations on February 14, 2024. Previous image releases in November 2023 and May 2024 showcased the telescope’s capabilities, followed by the unveiling of the first section of Euclid’s grand map of the Universe in October 2024.
By March 2025, Euclid had surveyed approximately 2,000 square degrees of the sky—equivalent to about 14% of its planned 14,000-square-degree coverage. The three deep fields featured in this release span 63.1 square degrees, providing an initial preview of Euclid’s ability to trace the cosmic web.
Advertisement
Mapping the Cosmic Web
Euclid’s observations of deep fields have already captured 26 million galaxies, some as far as 10.5 billion light-years away. The data also reveal a population of bright quasars at even greater distances. Over the next six years, Euclid will repeatedly scan these regions, drastically increasing the number of detected galaxies and enhancing the resolution of its observations.
According to ESA’s Euclid project scientist, Valeria Pettorino:
“It’s impressive how one observation of the deep field areas has already given us a wealth of data that can be used for a variety of purposes in astronomy: from galaxy shapes, to strong lenses, clusters, and star formation, among others.”
Euclid’s instruments play a crucial role in this endeavor. The visible imaging instrument (VIS) enables precise measurements of galaxy shapes, while the near-infrared spectrometer and photometer (NISP) determines distances and masses of galaxies
Together, these instruments provide important insights into how galaxies are distributed across the Universe and how they fit within the cosmic web—a structure shaped by the interaction of ordinary matter and dark matter.
Euclid Consortium scientist Clotilde Laigle, based at the Institut d’Astrophysique de Paris, highlighted the mission’s long-term goals:
“The full potential of Euclid to learn more about dark matter and dark energy from the large-scale structure of the cosmic web will be reached only when it has completed its entire survey. Yet the volume of this first data release already offers us a unique first glance at the large-scale organisation of galaxies, which we can use to learn more about galaxy formation over time.”
Advertisement
AI for galaxy classification
Over the course of its six-year mission, Euclid is expected to image more than 1.5 billion galaxies and transmit around 100 gigabytes of data daily. To efficiently process this vast dataset, researchers are employing artificial intelligence alongside human classification efforts.
“We’re at a pivotal moment in terms of how we tackle large-scale surveys in astronomy. AI is a fundamental and necessary part of our process in order to fully exploit Euclid’s vast dataset,” explained Mike Walmsley, a Euclid Consortium scientist at the University of Toronto.
The AI-based “Zoobot” algorithm, trained with input from nearly 10,000 human volunteers via the Galaxy Zoo platform, has produced the first detailed classification of over 380,000 galaxies. These classifications include features such as spiral arms, central bars, and tidal tails indicative of merging galaxies.
This initial dataset represents just 0.4% of the total number of galaxies that Euclid is expected to catalog at similar resolution. As the mission progresses, this growing catalog will enhance scientists’ understanding of galaxy formation and evolution.
Advertisement
A gravitational lensing discovery machine
One of Euclid’s key objectives is to study gravitational lensing, a phenomenon where light from distant galaxies is bent by the gravitational influence of foreground matter, including dark matter. Strong lensing, which creates highly visible distortions such as Einstein rings and arcs, is a particularly valuable tool for mapping the distribution of dark matter.
The new data release includes a catalog of 500 strong gravitational lens candidates, most of which were previously unknown. These discoveries were identified through a combination of AI analysis, citizen science contributions, and expert validation.
Future releases are expected to increase the number of known strong gravitational lenses dramatically. By the mission’s conclusion, Euclid is projected to catalog approximately 100,000 strong lenses—about 100 times more than currently known. In addition, Euclid will detect more subtle “weak” lensing effects across billions of galaxies, creating a three-dimensional map of dark matter distribution across cosmic history.
“This data release highlights the incredible potential we have by combining the strengths of Euclid, AI, citizen science, and experts into a single discovery engine that will be essential in tackling the vast volume of data returned by Euclid,” said Pierre Ferruit, ESA’s Euclid mission manager.
The telescope’s rapid progress is just at its beginning: this first dataset serves as a preview of what’s to come, with the first full cosmology data release scheduled for October 2026.
Advertisement
