Thread by @astrocaits

JWST has now found 100s of new galaxies at z>8 ✨, <650 million years after the Big Bang. But nearly all of them are pretty faint.

In COSMOS-Web, we found some VERY bright 10<z<14 candidates we’re excited to share. New paper out today 🧵⬇️
arxiv.org/abs/2308.10932

Short version:
* 15 z>10 candidates with MUV ~<-21 across 0.28sq deg, 12 of them robust (lots of tests, etc.)
* derived masses are HIGH.  Some as high as 10^(9.5-10)Mstar at z~12.
* Those masses don’t break ΛCDM, but are also very challenging to build
...

...
* You can make them with short-lived, efficient star-forming bursts, maybe
* Need spectra
ApJ submitted, comments welcome. Now, the slightly longer story ⬇️

First an important aside: some of you may recall GN-z11 — it is the brightest, farthest galaxy found by Hubble. It now has this STUNNING JWST spectrum, confirming its redshift at z=10.6.

https://twitter.com/aaysaxena/status/1625771668260913153?s=20

GN-z11 is extraordinary and unusual. It’s much brighter than other galaxies at this z, it has Ly-alpha in emission (weird for z>10), might have an AGN, and be embedded in a large-scale overdense environment. In isolation, it’s unclear why it’s so bright.

https://twitter.com/sandrotacchella/status/1625786278519226369?s=20

Well, we set out to find more GN-z11’s in COSMOS-Web. At the beginning of the summer, we finally had a good chunk of it (0.28sq deg) to search through to find these needles in the haystack.

(Reminder about COSMOS-Web and our Survey Overview Paper here ⬇️)

https://twitter.com/astrocaits/status/1593668052230479873?s=20

We found 15 candidate galaxies between 10<z<14 with similar luminosities as GN-z11 (with MUV~<-21).
✔️12 robust candidates
❌3 likely low-z contaminants (but who knows until you get a spectrum)

COSMOS-Web doesn’t have all the JWST filters other fields do, so we relied on careful photometric measurements from ground-based data from UltraVISTA and HSC to build confidence in the sample. Shout out to our catalog team led by Marko Shuntov and @Louise_Paq for their heroics

@Louise_Paq Here are the SEDs of the four brightest sources. One at z~10, three at z~12.

Also, before you say 'obvious AGN' ...all candidates are spatially resolved, average size Reff ~ 500pc.

@Louise_Paq 3 galaxies stand out: COS-z12-1, COS-z12-2, and COS-z12-3. All are brighter and higher-z candidates than GN-z11 🤯.
Their stellar masses, 10^(9.5-10), are VERY high at z~12. Not so high to be in tension with ΛCDM, but … we shouldn’t feel comfortable about these masses either.

@Louise_Paq Measuring the implied stellar mass density at z~12 and comparing against the halo mass function, these galaxies need to have stellar baryon fractions (or efficiency to transform baryons to stars) as high as ε⋆=0.2-0.5 to be observable in the COSMOS-Web volume.

@Louise_Paq Those efficiencies/stellar baryon fractions are wild.  They’re integrated over the history of the galaxy, meaning that the star formation process needs to be *maximally* efficient (ε~1) if there’s any "down time" at all.

@Louise_Paq If you make reasonable assumptions about gas content too (Kennicutt-Schmidt), you quickly account for all possible baryons in just stars and molecular gas. Those are the open points on the last plot in this thread ⬆️ (stellar mass density vs. stellar mass)

@Louise_Paq How to ~10^10 Msun quickly? We conclude it is most likely happening through short-lived starburst episodes that must have a very high efficiency.  We compare to a number of theoretical models, which you can read more about in the paper. 🤗

@Louise_Paq Here’s a plot showing the modeled star-formation history of those three super-bright z~12 sources with redshift.  They probably grew fast (!), by ~10x since z~14 (70Myr prior).

@Louise_Paq All of this was fun to work on, but we don’t have answers until we get spectroscopy — hopefully something the community can get behind soon.  These are likely to be among the brightest sources JWST ever discovers at z>10.
(Looking at you👀 @NASAWebb /@almaobs!)

@Louise_Paq @NASAWebb @almaobs Many thanks to all co-authors, @cosmosastro, and special shout-outs to our imaging+catalog team.  Also thanks to my family for putting up with me writing this paper during maternity leave (it takes a village to raise 3 under 3 while also trying to do science 🙏🏼).