Hubble Just Found a Ghost Galaxy (99% Dark Matter)

David Li here, the first author of the paper. First of all, I would really love to thank Anton for covering this discovery, I’ve been a follower of the channel since I was in undergrad (almost 10 years ago), Never in a million years have I imagined my work will be featured in the channel. But most importantly, I never thought me and my team would have discovered a dark galaxy. Now back to the science bit. Firstly, the most important thing I would like to note is that CDG-2 is still a galaxy candidate. To put the hammer on the nail and confirm that this is indeed a galaxy, we need spectra and kinematic (motion) data for the globular clusters (GCs) to confirm they’re in a gravitationally bound system. However, simply based on the spatial distribution of the GCs, the chance that they are just randomly grouped together like in CDG-2 is astronomically low — we estimated that the probability of this happening is 1.5x10^-5. Basically near impossible. The actual thing that made us very sure that this is indeed a galaxy is the extremely faint glow around the 4 GCs that some people in the comment have noted. This faint, triangular shape of diffuse light is what we have been hunting for using Hubble, Euclid, and Subaru. Because this faint structure appeared in all the data from these telescopes, that’s how we’re sure this is a galaxy, and it may be the darkest one we’ve ever found so far. Combining the four GCs and the faint diffuse light, we can say CDG-2 is indeed a galaxy with 99% confidence. However, statistics and shapes don’t make a galaxy, we need actual physics, which is why we need the GC kinematic and spectra data to put that hammer on the nail — cue JWST. This is what we need to say that CDG-2 is a galaxy with 100% confidence. The second thing is the dark matter of CDG-2. The reason we estimated that this thing could be 99.94-99.98% dark matter is because of the GCs, since there’s a very well observed empirical relationship between a galaxy’s GC mass/number and the galaxy’s dark matter halo mass. This relationship is what we used to estimate the dark matter content of CDG-2. But still, to actually confirm the dark matter content, we once again need the kinematic and spectra of GCs in CDG-2. Lastly, for CDG-1, or what we call the twin of CDG-2, it was also made up of four GCs, but CDG-1 had absolutely no detectable faint light surrounding it like in CDG-2. That is why CDG-1 is still not confirmed because our current galaxy formation model simply would not support the existence of CDG-1. Put it in another way, if CDG-1 is indeed a galaxy, it will rewrite our understanding of how galaxies form. Our team have submitted multiple proposals to observe CDG-1, but they were all denied because the proposals were deemed too risky. But now with the discovery of CDG-2, we may have a better idea on what CDG-1 could be. If CDG-1 is confirmed to be a system of four self rotating GCs, that would literally change the game because CDG-1 will be the literal definition or poster child of Dark Galaxy.

Galaxy has not been unlocked as a playable character yet 😂 Thank you Anton for providing these videos. You are one of my favorites to see every day and I can't help but share them with anyone that I can. You're wonderful and you're appreciated!

It is so wild to me that just within my lifetime Dark Matter went from being an extremely fringe theory that some people even doubted could qualify as a theory, because it might be imposible to disprove, to something that was regarded to be probably a calculation error, to something that has a credible canditate of (almost) an entire galaxy being made of it.

I know I put that galaxy somewhere!

5:13 am i the only one that can see a faint triangle in the circled region

Thanks, Anton. It's heartening to see the cosmologists that are your fans, engaging directly in the comments.

8:42 "galaxy stripped of everything " i resemble that remark

Congratulations to David Li and the team — amazing result. The 1.50×10⁻⁵ probability plus the faint diffuse structure seen in Hubble, Euclid and Subaru data makes CDG-2 a very compelling galaxy candidate. From an RTRT angle this kind of system is interesting because the baryonic domain is Nyquist-limited. If parts of the underlying dynamics evolve faster than that sampling rate, they don’t show up as luminous structure and only appear through gravitational effects. In a system like CDG-2 where almost all baryons are gone, the globular clusters become slow tracer particles mapping the potential well. Really looking forward to the JWST GC kinematics, because once their velocities are measured we’ll finally know if the clusters are dynamically bound and what the true halo mass is.

Is in Dark galaxy Dark star with Dark planet orbiting it and Dark aliens living on it, observing universe and wondering if life can exist in Light galaxies?

Keep up the great work Anton, and how cool that one of the authors of the paper mentioned left a comment!