# Dark Energy in Light of the Cosmic Horizon

1. Nov 29, 2007

### wolram

arXiv:0711.4810
Title: Dark Energy in Light of the Cosmic Horizon
Authors: Fulvio Melia
Subjects: Astrophysics (astro-ph)
Based on dramatic observations of the CMB with WMAP and of Type Ia supernovae with the Hubble Space Telescope and ground-based facilities, it is now generally believed that the Universe's expansion is accelerating. Within the context of standard cosmology, the Universe must therefore contain a third dark' component of energy, beyond matter and radiation. However, the current data are still deemed insufficient to distinguish between an evolving dark energy component and the simplest model of a time-independent cosmological constant. In this paper, we examine the role played by our cosmic horizon R0 in our interrogation of the data, and reach the rather firm conclusion that the existence of a cosmological constant is untenable. The observations are telling us that R0=c t0, where t0 is the perceived current age of the Universe, yet a cosmological constant would drive R0 towards ct (where t is the cosmic time) only once, and that would have to occur right now. In contrast, scaling solutions simultaneously eliminate several conundrums in the standard model, including the coincidence' and `flatness' problems, and account very well for the fact that R0=c t0. We show here that for such dynamical dark energy models, either R0=ct for all time (thus eliminating the apparent coincidence altogether), or that what we believe to be the current age of the universe is actually the horizon time th=R0/c, which is always shorter than t0. Our best fit to the Type Ia supernova data indicates that t0 would then have to be ~16.9 billion years. Though surprising at first, an older universe such as this would actually eliminate several other long-standing problems in cosmology, including the (too) early appearance of supermassive black holes (at a redshift > 6) and the glaring deficit of dwarf halos in the local group.

2. Nov 29, 2007

### sysreset

In this scenario, no light was emitted for the first 4 billion years after t0. Any thoughts as to what was going on for 4 billion years?

3. Nov 29, 2007

### Wallace

I haven't fully digested this paper yet but it does seem to contain some glaring problems. The alternative cosmology requires that there be 70% energy density of dark energy at recombination. Studies of the effect of even 1% of w<-1/3 material at this time have found that to be inconsistent with the observed CMB. This was not even given the slightest mention.

It also claims to resolve the substructure problem (more small DM halos are seen in simulations that in observations) but doesn't show how this would be the case in any detail. The overall normalisation of the amplitude of fluctuations is also in very good agreement for LCDM between the CMB and galaxy surveys. The model proposed in this paper would not agree with this and again this issue isn't given any thought.

New proposals need to explain all the data, not just one small part, to be reasonable.

Methodologicaly the approach to science taken is also flawed. The author finds that it is completely inconceivable that a number measured in nature be close to unity, and this is the sole basis for the rejection of LCDM, yet makes this rather odd statement about his own model

The proposal is that if an observations gives a model with a value that is near a number that has a pre-conceived notion of being special (in this case unity) the model must be wrong. However, a model that in the opinion of its inventor is 'elegant' should be right, even if unsupported by data.

Such thinking was attractive to Aristotle, but science has moved on a little since then...

4. Nov 29, 2007

### patty144

Well, I've read through it once and still need to think about all the details as well, but I
don't see any glaring problems yet. First of all, I don't see any new models presented
here. He's just comparing existing models and showing that LCDM doesn't work. And it
doesn't work, not because something has a magical number of 1, but because it has a
magical number of one ONLY NOW, at this very moment, not 2 billion years ago, not
1/2 billion years ago, but now. That's weird by any account.

I think what's happening is that if w is < -1/3, then the universe has to be older than
13.7 billion years. Light produced before that time is gravitationally redshifted to infinity
at and beyond the horizon radius, at 13.5 billion years (if I got that right).

He says in the paper that the dwarf halo problem goes away since there would have been
more time for hierarchical merging to deplete the lower end of the mass distribution. I
don't know if 3 or 4 billion years is enough to do that, but it would help I guess.

I also took a look at the earlier paper (The Cosmic Horizon). I didn't understand much
of that one since it's more mathematical, but this one is based heavily on that one.
You should take a look.

5. Nov 29, 2007

### Wallace

I've had a brief look at the other paper you refer too, but haven't looked at it in detail.

Why is it weird? If the number that it was now and would only ever be once was 2.6 would you be amazed? Clearly not. The only reason this is at all remarkable is by having a pre-concieved notion that the number one is special. This is not good science, but merely numerology!

Exactly, there would be more time for the dwarfs to merge, but is 3-4 Gyrs enough? Who knows? You can't make the statement that this new model (where by new model I mean a dark energy that tracks the matter density perfectly for all time) removes the sub-structure problem without making any attempt to actually make even the most rudimentary analysis of structure growth in this new model.

There are three mains observational pillars of modern cosmology, distance redshift measurements (the most important of which are the SN1A), the CMB and large scale structure. The success of LCDM is that is agrees with all three of these, with the same parameters. Making 'first order' calculations of the predictions of a given cosmological model for these three observables is something any decent cosmologists is capable of doing. Any new model proposed must therefore address all three, not just one as in the case of this paper. The analysis doesn't have to be too intricate, but it must be shown to be consistent at least to first order with these observations for anyone to take a new idea seriously.

6. Nov 29, 2007

### patty144

hmm, when I look at figure 1 in the paper, I don't see numerology. I see R/ct as a
function of time and I see that it heads towards the value 1 only at the present time. I
don't know, it sure looks weird to me. Are we that special that in the entire history of
the universe we live just at the right time for this to be happening?

When I look at figure 3, LCDM looks deplorable. It doesn't fit the SN1 data at all. Yet figure
9 shows that a tracking solution fits it with an amazing chi^2. Again, I don't know, but the comparison looks pretty amazing to me.

Again, I don't see where the new model is here. He is not presenting any new model. He's comparing models that have been around for years. I still don't see anything wrong with
what's in the paper.

7. Nov 30, 2007

### marcus

J.G.Pereira and R. Aldrovandi have a model where dark energy tracks the matter density, as I understand it. this is one of the things that is worrying me about their model.
I was fearing that it was a fatal flaw.

Now this paper comes along and seems to make a virtue out of it! I will have a look.

8. Nov 30, 2007

### Wallace

You have to establish the specialness of 'what is happening just now' in order to determine if the fact that it only happens just now is weird. The co-incidence problem always looks a lot worse in a log-log plot as well, a very small difference in time either way (a billion or two years either side of now) suffers the same issue, so we aren't that special.

In any case the problem with testing model by co-incidence is that it comes down to an entirely subjective appraisal about what is 'special'. I prefer to let the data speak, rather than appeal to aesthetics.

Those plots are deplorable, but only because of the shiftyness the author has shown in putting them together! For the 'LCDM' plot, he has used first order canonical values of 0.3,0.7 rather than the actual best fit values for these parameters! If the LCDM plot was really that bad, we never would have settled on the model! I've played around with the SN data and cosmology enough to know that you only make LCDM look that bad by trying.

By contrast his scaling model has be fine-tuned to give the optimal fit. In any case, he still has only addressed one of the three main pieces of evidence for LCDM. It is easy to beat a model that satisfies all the data by one that only satisfies a small part of it.

Sure, the scaling type solutions have been around for years, but this idea is quite different. The regular approach is for the scaling to moderate the change over time between matter and Lambda domination epoch, making the transition longer and reducing the co-incidence problem. However this approach is much more drastic, taking in some sense a 'perfect' scaling solution that keeps the energy densities not just similar but identical at all times.

The reason why this model proposed makes no sense is that is completely alters the CMB signal and structure formation. The presence of all that negative pressure material in the early universe suppresses the growth of structure in a way that a few more billion years is not going to make up for. The linear growth factor as a function of time is an elementary calculation that should have been performed for this model. It has either been omitted because is would show that the model fails at the first hurdle for structure formation, or because the author isn't aware that cosmology extends beyond supernovae measurements.

Either way the paper is not a fair assessment of the two main models it compares, and should not make the firm statements that it does without doing these additional calculations.

9. Nov 30, 2007

### Chronos

Another 'problem' is the assumption Dark Matter is a single 'species' of particle. I'm very suspicious of this proposition. Introducing a variety of flavors of DM particles changes everything. We already know that neutrinos [a DM particle of sorts] come in a number of flavors. We also know they cannot comprise more than a tiny fraction of the total amount of DM in the universe.

10. Nov 30, 2007

### Garth

Fulvio Melia has also published this similar preprint already accepted in MNRAS The Cosmic Horizon
Garth

11. Nov 30, 2007

### marcus

This is just the sort of thing that was worrying me over the past few days about the Pereira and Aldrovandi paper. I wonder if there is some way out of the difficulty.

Patty and Wallace, excuse me for interrupting your conversation. I find these two papers by Fulvio Melia very interesting.

In case either of you is curious here is the link to the paper I was reading earlier, which seems to bear some relation to these.

http://arxiv.org/abs/0711.2274
de Sitter Relativity: a New Road to Quantum Gravity
R. Aldrovandi, J. G. Pereira
17 pages
(Submitted on 14 Nov 2007)

"The Poincaré group generalizes the Galilei group for high-velocity kinematics. The de Sitter group is here assumed to go one step further, generalizing Poincaré as the group governing high-energy kinematics. Algebraically, this is done by supplementing spacetime translations with proper conformal transformations. This change in special relativity implies concomitant changes in general relativity -- yielding a de Sitter general relativity. The source current turns out to include now, in addition to energy-momentum, the proper conformal current, which appears as the origin of the cosmological constant. In consequence, it is no longer a free parameter, and can be determined in terms of other quantities. When applied to the propagation of ultra-high energy photons, de Sitter general relativity gives a good estimate of the time delay observed in extragalactic gamma-ray flares. It can, for this reason, be considered a new approach to quantum gravity."

They calculate a value for the effective Lambda which agrees roughly with observation. Provide a mechanism explaining how the dark energy effect arises (don't put it in by hand). They also calculate a delay for 10 TeV gammaray which agrees roughly with what was reported by MAGIC from observations of Makarian 501 AGN flares. Both of these things would appear to be "too good to be true". Their gammaray dispersion relation exposes their model to immediate risk of falsification as soon as more AGN flare data appears.

their de Sitter General Relativity model is something they have been developing for quite a few years and are now beginning to derive numerical predictions from. I find it very interesting that it has some apparent overlap with what Fulvio Melia is saying.

Last edited: Nov 30, 2007
12. Nov 30, 2007

### marcus

Yes Garth, that is the paper which Patty was talking about earlier (post #4) called The Cosmic Horizon.
The two papers should be read together. It is encouraging that both were submitted to MNRAS and the first one has already been accepted for publication. Since the second is a continuation, extending rather than repeating the first, I expect it too will be promptly accepted.

Last edited: Nov 30, 2007
13. Nov 30, 2007

### Wallace

I'm not so sure. I still haven't looked at the first paper in detail, but it seems to be better than the second in that it outlines as far as I can see a previously unknown property of the FRW solution to GR, or at least illuminates the significance of a previously little known result. But as I say, I can't assess it too much as I haven't read it properly yet.

The second paper though falls short for the reasons I've already stated.

14. Nov 30, 2007

### Garth

Thank you marcus, sorry Patty144 - I didn't notice that comment of yours and BTW a very warm welcome to these Forums!

Wallace - although the LCDM model fits the data very well and the OP link paper could have only made the LCDM model "look that bad by trying", nevertheless the question arises of whether other models, such as suggested in these papers, might also eventually fit the data as well.

I note the number of free parameters that are necessary in the mainstream model and are constrained to make that model fit the data and my question is whether any proposed model is actually describing the real physics or just emulating it. A laboratory detection of the ‘Inflaton’ and DM particle with the required properties would help resolve this problem.

Garth

Last edited: Nov 30, 2007
15. Nov 30, 2007

### marcus

One of the main reasons you stated is structure formation and the paper I mentioned by P and A hints at an answer to that one.

To roughly paraphrase, you pointed out that if DE is proportional to matter density thru all history, then early universe has much more DE than we are used to supposing. So clusters that are trying to form would be dissipated scattered by expansion and never get a chance, or some would take much longer to form because an uphill fight against all that expansion.

This was worrying me for several days as I was trying to read the Pereira Aldrovandi paper---so I think I know what you mean. They also have this roughly constant proportionality. but for them, DE DOES NOT GRAVITATE, because it is an affect that comes out of deSitter GR. So they require a lot more DARK MATTER in their universe, so as to achieve critical density for spatial flatness.

this gives me the faint hope that (whether or not P and A are on the right track) something could work out that makes Fulvio Melia's idea OK----something that addresses your objection about structure formation.

16. Nov 30, 2007

### patty144

Heard an interesting journal-club style presentation on this today. Seems to be getting
a lot of attention. The reason his LCDM fit does poorly is that he's using basic LCDM, as
he states. The SN people get better fits because they need to introduce past
deceleration (beyond redshift 0.5) before acceleration. As such, his results agree with
them. They get a good fit only if they introduce new parameters into the model, which
is NOT basic LCDM.

After reading this paper through again carefully and listening to the talk, it's clear that he
is not a proponent of any one model. His main emphasis is simply to show that a
consideration of the Cosmic Horizon is necessary in order to interpret the data properly.
We rederived his equation (5), and as far as we can tell, it's correct and
cannot be ignored. LCDM doesn't seem to fit with this.

What's also interesting, and hadn't noticed/heard before, is that scaling solutions have
the potential of eliminating both the coindicence and flatness problems, as he shows
with his equations (10) - (13). That might be new.

Also remember, an important point I heard this morning, is that just becauseyou
have a scaling solution doesn't mean it has to go all the way back to the beginning.
There may be some effect that drops off when z reaches 5, 10 or beyond. We just
don't know. I think the main point is that certainly for z<3-5, scaling solutions work
_much_ better than LCDM. Take a look at Fig 9, in particular. A chi^2 of 1.001 is
pretty darned impressive, and there is no need to introduce decelration followed by
acceleration.

This is fun!!

17. Nov 30, 2007

### jal

Is this talk available on line?

18. Nov 30, 2007

### Wallace

'basic' LCDM inevitably contains decceleration before z~1 and acceleration afterwards. This is not artificially introduced just to fit the SN data! The Riess et al papers used LCDM only and fit the data without any additional hacks. The reason Melia's LCDM plot doesn't look good is that he is using $$\Omega_m = 0.3$$,$$\Omega_{\Lambda} = 0.7$$ which are not the best fit LCDM parameters! Have a read of the Riess et al papers to see the better fits.

I made that point previously in this thread, that scaling solutions proposed in the past have not been the 'perfect' scaling as proposed in this paper, but as you say looks more like LCDM in the early universe.

The overriding point though is that we can speculate about models for as long as we want but when simple calculations to relate them to data can be performed, but are ignored, then the speculation is pointless.

19. Nov 30, 2007

### patty144

This is what is actually said in the paper: "Given the broad range of alternative
theories of dark energy that are still considered to be viable, it is beyond the scope
of this paper to exhaustively study all dynamical scenarios. Instead, we shall focus
on a class of solutions with particular importance to cosmology---those in which
the energy density of the scalar field mimics the background fluid energy density."

I don't think he's proposing any model. I may be wrong, but my interpretation is
different. I think the main point he's making is that one cannot ignore the role
of the Cosmic Horizon, as when he says: "In this paper, we examine
the role played by our cosmic horizon R0 in our interrogation of
the data..."

And then he claims that the behavior of LCDM shown in figure 1, which is far worse
than previously supposed for the coincidence problem, makes it very unlikely that
dark energy is due to Lambda.

To me, this odd behavior of LCDM, and the fact that a simple scaling solution fits
the SN data as well as shown in Figure 9, suggests that LCDM is in trouble. Then
there's the question of why Lambda should be so much smaller than predicted for
the vacuum energy in quantum mechanics. I looked up the Klypin paper, and according
to them, LCDM predicts 10 times as many small halos as are seen.

There's no doubt that with enough free parameters one can make LCDM fit the SN
data. But scaling solutions fit the SN data too. What I get out of the paper is that
LCDM doesn't explain R0=ct0 well at all, but scaling solutions do. It seems to me that
the balance is in favor of the latter. At the very least, one should keep an open mind.
I don't see why there's so much vested in LCDM that there's fear to consider something
else, especially if the data doesn't agree with it.

20. Dec 1, 2007

### Wallace

We only have one universe. If the one we got to look at has a coincidence, then it has a coincidence. If that is what the data says then so be it.

Right, so the Klypin paper carefully calculates the structure expected for LCDM, and a well known issue is found. The paper under question spends one sentence simply asserting that the scaling solution will simply fix this problem. No justification, no calculation of even an order of magnitude or first order estimate. It is easy to beat a model that has been rigorously interrogated with one in which you simply make up it's predictions to suit the data.

You are missing the point, the issue is that figure 1 intentionally uses the wrong values of the parameters of LCDM, not that is uses more or less 'free parameters'. It is clearly designed to make LCDM look bad. With the values of LCDM paramters that actually are fitted to the data, it is a much better fit.

Again, the paper compares a LCDM with intentionally ill-fitting parameters, with a scaling solution that has an extra free parameter compared to LCDM, and with the parameters properly fitted. Hardly a fair comparison!

Only when you consider one small part of available cosmological data, and when you fiddle with the numbers for maximum effect!

There really needs to be a rule, lets call is Wallace's Law for sake of argument, that as soon as you appeal to 'having an open mind' then you must have run out of intelligent things to say and have therefore lost the argument. I have a very open mind when it comes to cosmology, as do most people who get this banal like thrown at them. The point is that I'll consider any idea on it's merits. If an alternative idea has huge gaping holes compared to the current best bet, then of course I will express concern over those holes, and demand that they be addressed.

It's not about having an open mind, but an active one that question everything. I'll put my money on the model that best answers those questions.

My own research revolves almost entirely around non-LCDM models. The point is that you have to do the hard yards and actually consider the full implications of a model to do it properly, not make a smash and grab paper that takes a skewed view of a small sub-set of the data and then makes grandiose claims.

Again, your emotive appeals to 'a fear' of considering alternatives is ill-considered. We know there isn't perfect agreement between all data and LCDM, but if we try and pretend that we've fixed things by ignoring the majority of the data we aren't getting anywhere.