1. Home
  2. / Science and Technology
  3. / Astronomers Found A Galaxy That Cannot Be Seen — It Was Detected By The Gravitational Trace Of Four Star Clusters Floating In The Void, And When Hubble, Euclid, And Subaru Pointed To The Location, They Confirmed That 99% Of What Is There Is Dark Matter, The Most Abundant Substance In The Universe That No Detector Has Ever Captured Directly
Reading time 7 min of reading Comments 0 comments

Astronomers Found A Galaxy That Cannot Be Seen — It Was Detected By The Gravitational Trace Of Four Star Clusters Floating In The Void, And When Hubble, Euclid, And Subaru Pointed To The Location, They Confirmed That 99% Of What Is There Is Dark Matter, The Most Abundant Substance In The Universe That No Detector Has Ever Captured Directly

Written by Débora Araújo
Published on 13/03/2026 at 15:03
Astrônomos encontraram uma galáxia que não pode ser vista — ela foi detectada pelo rastro gravitacional de quatro aglomerados de estrelas flutuando no vazio, e quando Hubble, Euclid e Subaru apontaram para o local, confirmaram que 99% do que está ali é matéria escura, a substância mais abundante do universo que nenhum detector jamais captou diretamente
Astrônomos encontraram uma galáxia que não pode ser vista — ela foi detectada pelo rastro gravitacional de quatro aglomerados de estrelas flutuando no vazio, e quando Hubble, Euclid e Subaru apontaram para o local, confirmaram que 99% do que está ali é matéria escura, a substância mais abundante do universo que nenhum detector jamais captou diretamente
  • Reação
Uma pessoa reagiu a isso.
Reagir ao artigo

Astronomers Detect Invisible Galaxy Formed Almost Entirely of Dark Matter After Observing the Gravitational Effect on Four Clusters of Stars.

Dayi Li is a statistician and astrophysicist. This uncommon combination was exactly what made possible the discovery he published in June 2025 in the Astrophysical Journal Letters. Li, a postdoctoral researcher at the University of Toronto, did not find CDG-2 by looking at it. He found it by looking at what shouldn’t be there.

The method was developed from a simple idea: normal galaxies accumulate globular clusters around them — spherical clumps of hundreds of thousands of old stars, gravitationally bound to each other, that orbit galaxies like compact satellites. The Milky Way has more than 150 of these clusters.

What Li realized is that if there are dark galaxies — objects with almost no stars, invisible by definition — they should still have globular clusters floating where their stars should be. It was like looking for a person in a crowd by the trail they leave while walking, not by their face.

The Method That Nobody Had Tried

The PIPER survey — Program for Imaging of the PERseus cluster — was designed to photograph the Perseus cluster with Hubble. Perseus is one of the largest known galaxy clusters, located about 300 million light-years from Earth, containing hundreds of galaxies interacting gravitationally within a colossal volume of space.

Within this chaos, Li and his team used a statistical model called cluster Poisson process — a technique adapted from spatial pattern analyses in epidemiology and ecology — to distinguish globular clusters belonging to known galaxies from those that are anomalously clustered, with no visible galaxy nearby.

The result, in March 2025, was a list of candidates. One caught attention: four globular clusters grouped in a space of just 1.2 kiloparsecs in diameter, with no visible galaxy around. The group was named CDG-2 — Candidate Dark Galaxy-2. The probability that this grouping was mere random coincidence was statistically negligible.

Three Telescopes, One Ghost

The next step was trying to see what was there. Li returned to the PIPER images and stacked two different exposures of the same spot in the sky, adding the weak signal that each one had captured separately. In the combined result, something appeared: a diffuse, extremely faint emission around the four clusters. A haze of light that barely distinguished itself from the background noise.

This was not enough to confirm anything. A single source can produce artifacts. Li then used a completely independent dataset: images from the Euclid satellite of the European Space Agency, obtained in the same field as the Perseus cluster as part of the telescope’s first scientific observations. Euclid was specifically designed to detect diffuse emission with low surface brightness — faint galaxies that conventional instruments cannot distinguish from the background sky.

The diffuse emission appeared again. In the same place. With the same shape. “The morphology of the diffuse emission in the Hubble and Euclid data is almost identical,” the authors wrote in the paper. Two telescopes, two detectors, two datasets collected at different times, showing the same ghost structure at the same point in space. Additional confirmation with the Subaru, a ground telescope in Hawaii, strengthened the signal. CDG-2 was real. And it was different from anything cataloged before.

What The Numbers Reveal

CDG-2 shines with light equivalent to about 6.2 million suns. This number seems large until it is put in context: the Milky Way shines with the light of approximately 200 to 400 billion suns. CDG-2 is between 30,000 and 65,000 times less luminous than our galaxy — yet it has dimensions comparable to those of a dwarf galaxy. What occupies that empty space where stars should be?

The four detected globular clusters account for at least 16% of all visible light from the galaxy. In some analytical scenarios, this percentage rises to 33%. This means that CDG-2 is the galaxy most dominated by globular clusters ever discovered — nearly all of its light comes from the only four compact objects that have survived.

The remainder of the mass — between 99.94% and 99.98% of the total, according to established relations between cluster mass and dark matter halo mass — is dark matter. “This is the first galaxy detected solely by its population of globular clusters,” Li declared. The phrase, technical in form, is radical in content: an entire galaxy identified not by what can be seen, but by the gravitational trail left on objects that survived where nearly everything else has vanished.

What Is Dark Matter — And Why Is It Still a Mystery

Dark matter does not emit light. It does not reflect light. It does not absorb light. It does not interact with photons in any known way. The only way to detect it is by the effect it has on ordinary matter through gravity.

The existence of dark matter was first proposed in the 1930s by Swiss astronomer Fritz Zwicky, upon observing that the galaxies in the Coma cluster were moving too fast to be held together merely by the gravity of visible matter. In the 1970s and 1980s, American astronomer Vera Rubin confirmed the phenomenon for individual galaxies: the stars at the edges of spiral galaxies orbit too fast, which only makes sense if there is a huge amount of invisible mass around them.

Since then, dark matter has been a pillar of the standard cosmological model. It makes up approximately 27% of the total content of the universe — five times more than all the ordinary matter that forms stars, planets, gases, and everything else that can be seen. But no terrestrial detector has ever captured a dark matter particle directly. Experiments in deep mines, particle accelerators, gamma-ray satellites — all without conclusive results after decades of search.

CDG-2 does not solve this mystery. But it offers something that physicists rarely have: a nearly pure laboratory of dark matter, without the complicating interference of gases, stars, and star formation processes that, in normal galaxies, scramble the gravitational signal of dark matter with that of everything else.

Why CDG-2 Reached This Extreme State

Most galaxies, even dwarf and low surface brightness ones, retain enough gas to form stars over billions of years. CDG-2 appears to have lost almost all of that.

The most likely explanation lies in the environment it exists in. The Perseus cluster is one of the most violent gravitational environments in the local universe — hundreds of galaxies moving at speeds of thousands of kilometers per second, interacting, deforming, exchanging material. In this scenario, smaller and less massive galaxies can have their gas and stars stripped away by tidal interactions with larger neighbors, in a process that astronomers call gravitational stripping.

What remains, after billions of years of erosion, are the most resilient objects: the globular clusters, dense and gravitationally cohesive enough to survive where diffuse stars cannot. And, holding all of this together, the dark matter halo — which, being distributed in massive volumes and not interacting through collisions like ordinary matter, resists the erosion process much better.

CDG-2 may be a galaxy that has been hollowed out from within, leaving only the invisible skeleton that keeps it cohesive and the few compact nodules that survived the process.

The Even Darker Twin

The confirmation of CDG-2 reignited an older question. In 2022, Li and his team had already identified, using the same method, a previous candidate: CDG-1, also in the Perseus cluster, also an anomalous grouping of globular clusters without a visible galaxy. However, unlike CDG-2, CDG-1 never produced a signal of diffuse emission in any dataset — neither in Hubble nor in Euclid.

This could mean two opposite things. Either CDG-1 is a false positive, a fortuitous grouping of clusters that does not belong to any galaxy. Or it is even more extreme than CDG-2 — a galaxy so devoid of stars that not even the residual diffuse emission can be detected with current instruments. A ghost within a ghost.

The authors of the article published in June 2025 left the question open: the confirmed existence of CDG-2 makes CDG-1 a more serious candidate than it was before — and also more disturbing.

What Comes Next

The definitive confirmation of CDG-2 as a galaxy — and not a fortuitous grouping — still depends on spectroscopy. If James Webb can separate the light from the diffuse emission into its individual wavelengths, it will be possible to measure the galaxy’s recession velocity and confirm that it is indeed in the Perseus cluster, 300 million light-years away, and not in some distant background plane.

Meanwhile, the implications of what has already been found are sufficiently unsettling. Dark galaxies — objects dominated by dark matter with almost no stars — were predicted by computational simulations of the universe decades ago. But finding them is difficult precisely for the obvious reason: they do not shine.

What Li demonstrated is that it is possible to find them indirectly, by the trail they leave on objects that withstand the emptying process. And, if this method works, the universe may be filled with invisible galaxies that have never been cataloged — entire structures of dark matter with only a few clusters of old stars as witnesses that something larger, and invisible, exists there.

Inscreva-se
Notificar de
guest
0 Comentários
Mais recente
Mais antigos Mais votado
Feedbacks
Visualizar todos comentários
Source
Débora Araújo

Débora Araújo é redatora no Click Petróleo e Gás, com mais de dois anos de experiência em produção de conteúdo e mais de mil matérias publicadas sobre tecnologia, mercado de trabalho, geopolítica, indústria, construção, curiosidades e outros temas. Seu foco é produzir conteúdos acessíveis, bem apurados e de interesse coletivo. Sugestões de pauta, correções ou mensagens podem ser enviadas para contato.deboraaraujo.news@gmail.com

Share in apps
0
Adoraríamos sua opnião sobre esse assunto, comente!x