Impossible Early Galaxies Firmly Established

The New Paper In Nature

A new paper in Nature (and its abstract) are as follows (footnotes in the abstract omitted):

Galaxies with stellar masses as high as ~ 10^11 solar masses have been identified out to redshifts z ~ 6, approximately one billion years after the Big Bang. It has been difficult to find massive galaxies at even earlier times, as the Balmer break region, which is needed for accurate mass estimates, is redshifted to wavelengths beyond 2.5 μm. Here we make use of the 1-5 μm coverage of the JWST early release observations to search for intrinsically red galaxies in the first ≈ 750 million years of cosmic history. In the survey area, we find six candidate massive galaxies (stellar mass > 10^10 solar masses) at 7.4 ≤ z ≤ 9.1, 500–700 Myr after the Big Bang, including one galaxy with a possible stellar mass of ~10^11 solar masses. If verified with spectroscopy, the stellar mass density in massive galaxies would be much higher than anticipated from previous studies based on rest-frame ultraviolet-selected samples.

Labbé, I., van Dokkum, P., Nelson, E. et al. "A population of red candidate massive galaxies ~600 Myr after the Big Bang." Nature (February 22, 2023). https://doi.org/10.1038/s41586-023-05786-2 (Open access version available at https://arxiv.org/abs/2207.12446).

Why Is This Paper Important?

The Bottom Line

The paper is notable because it is the most extreme and definitive example of astronomy observations that give rise to the "impossible early galaxy problem" of the LamdaCDM model which is often described as the "Standard Model of Cosmology."

Simply put, the LambdaCDM model predicts that the formation of galaxies of this size should take place far later after the Big Bang than the time frames in which they have been observed by astronomers to exist.

Previous LambdaCDM Galaxy Formation Predictions Compared

One widely accepted prediction was made by astrophysicist Carlos Frenk at a scientific conference in October of 1998 based upon LambdaCDM simulations had been that there would be no galaxies before redshift z=7. This corresponds to a look back time of 13.01 billion years, which is about 770 million years after the Big Bang, which implies that galaxies start forming about 55% to 10% more slowly in the LambdaCDM model than the earliest galaxies must start to form given what has been observed to form so far by the JWST. And, new observations of earlier galaxies over the lifetime of the JWST's mission can only make that gap worse, not weaker, if even earlier galaxies are located.

Like most things in astronomy and cosmology, this LambdaCDM prediction for the time period in which galaxies start to form had a margin of error, but it wasn't a big one and had largely been confirmed by later calculations and simulations made in using the bare LambdaCDM model over the next quarter of a century. And, the LambdaCDM model was so admirable, in part, because it had so few complications and parameters to leave wiggle room in its predictions, and yet was still a good fit to the data for decades.

But now, the new paper in Nature is reporting six reasonably large galaxies at redshifts of z=7.4 ≤ z ≤ 9.1, which at the high end, is much earlier in time than Frenk's LambdaCDM model prediction of z=7. Frenck's 1998 prediction has been only slightly tweaked over the next quarter century and is a precise enough prediction to be significantly different, statistically, from the results announce in the new paper in Nature.

The Impossible Early Galaxies Problem

A published paper from 2016 articulated the "impossible early galaxies" problem as follows:

The current hierarchical merging paradigm and ΛCDM predict that the z∼ 4-8 universe should be a time in which the most massive galaxies are transitioning from their initial halo assembly to the later baryonic evolution seen in star-forming galaxies and quasars. However, no evidence of this transition has been found in many high-redshift galaxy surveys including CFHTLS, Cosmic Assembly Near-infrared Deep Extragalactic Survey (CANDELS), and Spitzer Large Area Survey with Hyper-Suprime-Cam (SPLASH), which were the first studies to probe the high-mass end at these redshifts. Indeed, if halo mass to stellar mass ratios estimated at lower-redshift continue to z∼ 6-8, CANDELS and SPLASH report several orders of magnitude more M∼ 10^12-13M⊙ halos than is possible to have been formed by those redshifts, implying that these massive galaxies formed impossibly early.

We consider various systematics in the stellar synthesis models used to estimate physical parameters and possible galaxy formation scenarios in an effort to reconcile observation with theory. Although known uncertainties can greatly reduce the disparity between recent observations and cold dark matter merger simulations, there remains considerable tension with current theory even if taking the most conservative view of the observations.

Steinhardt, Charles. L. ; Capak, Peter; Masters, Dan; Speagle, Josh S., "The Impossible Early Galaxies Problem" 824(1) The Astrophysical Journal, article id. 21, 9 pp. (June 2016). DOI: 10.3847/0004-637X/824/1/21 (open access copy available at arXiv:1506.01377).

Thus, there were strong observational hints that there might be an "impossible early galaxy problem", for example, from the Hubble space telescope's observations and other "telescopes" (using the term loosely to describe a variety of astronomy instruments) that can probe highly redshifted objects long before the paper above published yesterday in Nature based upon JWST observations was released.

But, since the James Webb Space Telescope is so much more powerful than any earlier telescope when it comes to probing high redshift objects, what were previously strong hints that there might be big galaxies very soon after the Big Bang turned almost immediately after the JWST came online into multiple clear and unequivocal examples of galaxies of given sizes long before LambdaCDM said that they should exist. The JWST has also seen more galaxies at higher redshifts (i.e. more recently after the Big Bang) than any other "telescopes" before it has (even though it has been on line only briefly). This paper discusses the oldest galaxies seen to date, just 500-700 million years after the Big Bang, that look far more like modern galaxies than the LambdaCDM model should have made possible at this early time period in the universe.

To oversimplify the narrative somewhat, LambdaCDM assumes that in the early universe clumps of dark matter start to accumulate from random differences in matter density, which make it possible for ordinary matter to become concentrated enough to form stars, which in turn end up in clumps of proto-galaxies. These clumps undergoing a "hierarchal" process of repeated mergers and the collide into each other, until they eventually form modern sized galaxies.

The LambdaCDM model, once its six parameters are fitted to astronomy observations, provides data that makes it possible to make reasonable estimates of the rate at which dark matter becomes clumpy, the rate at which stars form, and the rate at which mergers occur, from which it is possible to make reasonable estimates of when galaxies of a certain size ought to appear.

But evidence provided by the JWST, exemplified by this paper with the earliest examples of early galaxies, makes clear that either something about the calculations used to make those predictions, or the LambdaCDM model itself, are wrong in this respect.

The jury is out regarding why galaxies formed much earlier than expected in the LambdaCDM model, even though something is clearly amiss.

What About The LambdaCDM Model Does This Paper Show Is Broken?

One of the main pieces of experimental evidence that previously inspired confidence in the accuracy of the LambdaCDM model was its ability to describe essentially all cosmological scale observations at a large scale structure level with just six astronomy observation fixed parameters, each of which has been measured with some precision.

The LambdaCDM model, once its six parameters are fitted, is, for example, an extraordinarily good match to the observed pattern of the cosmic background radiation (CMB) which came into being during the "recombination era" about 0.38 million years after the Big Bang. (See generally here for the conventional chronology of the universe after the Big Bang in cosmology.) Star and galaxy formation can't happen in any appreciable amount until after the recombination era when ordinary matter and photons (i.e. "radiation") decouple from each other because protons and electrons have been bound together into neutral atoms and the temperature of the universe has cooled sufficiently to allow for greater clumping of matter. So, we think that LambdaCDM gets the "starting line" of the star and galaxy formation process right even after this new paper.

But, the new paper's observations imply that the universe went from giving birth to its first stars after recombination ended, to producing galaxies similar in scale to many galaxies we see today at very low redshifts (about 13,780 million years after recombination), in just 500 to 700 million years. This strongly suggests that some other part of the LambdaCDM model, that crops up later in the history of galaxy formation in the universe and the larger span of the chronology of cosmology, is broken.

Are The Observations In This Paper Credible And Likely To Hold Up?

This result is highly credible because it confirms pre-JWST strong hints that these early galaxies existed. As noted above "high-redshift galaxy surveys including CFHTLS, Cosmic Assembly Near-infrared Deep Extragalactic Survey (CANDELS), and Spitzer Large Area Survey with Hyper-Suprime-Cam (SPLASH)" were already pointing to this result seven years ago, although much more equivocally than the new data from the JWST does.

The new paper's results are also credible observations because many other JWST observations also identify early galaxies, even though they have not yet been published, have been released to the public, and in some cases have been disclosed in preprints of articles that have been accepted for publication in leading astronomy journals (e.g. here and here and here). So, independent confirmations of the existence of early galaxies established in the Nature article using JWST observation of other candidate galaxies at very high redshifts over the course of the next year or so, are a near certainty.

Another reason that this result is credible is that the impossible early galaxy problem with the LambdaCDM model isn't the only recent crack in that model's predictions.

A full review of discrepancies between astronomy observations and the LambdaCDM model is beyond the scope of this thread. But it suffices to say that there are half a dozen or a dozen or so distinct and independent discrepancies between LambdaCDM model predictions and astronomy observations that are currently being actively investigated in areas that range from galaxy and galaxy cluster observations to different kinds of cosmology scale observations.

Some of those discrepancies may turn out to be experimental methodology issues or require slight tweaks to how the LambdaCDM predictions are calculated that are no big deal in the greater scheme of things. Others discrepancies, however, may turn out to be as serious a challenge to the LambdaCDM model as the impossible early galaxy problem that the new paper in Nature has highlighted.

This is a very different situation than the one that exists for the Standard Model of Particle Physics, where every time a discrepancy or tension between the Standard Model's predictions and experiment has cropped up over the last half century (except with respect to neutrino mass), it has subsequently promptly been ruled out with more experiments and better analysis, and there are only a few weak discrepancies between the Standard Model and experiment currently in play.

So, the scientific community is already primed right now to be more receptive to challenges to the "Standard Model of Cosmology" than it is to tensions between experimental results and the Standard Model of Particle Physics.

Alternatives To The LambdaCDM Model

If the LambdaCDM model is "broken" what alternatives exist to it?

There are multiple possible ways that the LambdaCDM model could be tweaked. One of those many possible alternatives is to look at a gravity based explanation of dark matter phenomena.

Models that attempt to explain dark matter phenomena with a modification to the laws of gravity or how they operate, rather than with dark matter particles, generically predict earlier galaxy formation than the LambdaCDM model does, consistent with the new JWST observations.

For example, galaxies were predicted to form in the time frame now observed by Bob Sanders in October of 1997 in the MOND (modified Newtonian dynamics) paradigm which is the most widely discussed gravitation based approach to explaining dark matter phenomena, even though it is itself a mere phenomenological toy model theory. Sanders predicted that: “Objects of galaxy mass are the first virialized objects to form (by z=10) and larger structure develops rapidly.” This is a little less than 500 million years after the Big Bang.

A JWST observation of a galaxy that may be as earlier as z=9.1 after only an initial quick search shortly after it has become operational, is consistent with that prediction and is at odds with the LambdaCDM prediction.

(This post substantially recaps a Physics Forum post here).