To reiterate, Please see http://sci.esa.int/science-e/www/object/index.cfm?fobjectid=51551 which states:

"... the Planck data reveals the presence of subtle anomalies in the CMB pattern that might challenge the very foundations of cosmology. The most serious anomaly is a deficit in the signal at large angular scales on the sky, which is about ten per cent weaker than the standard model would like it to be. Other anomalous traits that had been hinted at in the past - a significant discrepancy of the CMB signal as observed in the two opposite hemispheres of the sky and an abnormally large 'cold spot' - are confirmed with high confidence. Planck's new image of the CMB suggests that some aspects of the standard model of cosmology may need a rethink, raising the possibility that the fabric of the cosmos, on the largest scales of the observable Universe, might be more complex than we think."

Bernard,

Planck is so far so good, half of the data have been analyzed. Just wait!

Yes indeed - but they have posted a library of papers and I shall have to find time to start reading them. No choice!

There's no doubt that it is successful and (fortunately) leaves open as many questions as it answers - in the first place why there are such substantial changes from our previously "accepted" values from WMAP 9-year. Obviously, the two sets of parameter estimates overlap (we would be upset if they didn't), but the differences appear to be systemic as if resulting from calibration uncertainties. That's why thorough reading is essential.

I'll now have to change all the parameter estimates in my book, but it will be nice to have new pictures.

I am a little puzzled by the statement that Planck reveals an almost perfect Universe.

If I understand it correctly our existence was made possible thanks to these small imperfections, alias fluctuations (?)

I think that was a marketing slogan for the press conference, but it is indeed "perfect" in the sense that is accords extremely well with our preconceptions based on the "Standard Model" that has emerged with the supernovae and WMAP data. There are no ugly surprises which might have indicated that we have got something wrong. It's also perfect in the sense that everything on the mission appears to have worked as it should. So nothing to be unhappy about.

The one report from the conference that puzzled (or excited - depends on who you are) was the detection of a dipole anisotropy component having a direction that lines up with the "axis of evil" and the plane of the ecliptic.

The most exciting thing that was presented from my perspective was their view of the very largest scale structure on degree scales.

But I need to read the papers before making remarks about that.

Bernard, before you update all the parameters in your book do have a look also on "The Atacama Cosmology Telescope: Cosmological parameters from three seasons of data", arXiv:1301.0824. For instance the Hubble constant in Table 2, when the standard model is extended with one more parameter for the number of neutrinos (N_eff) or for the sum of neutrino masses, yields the same value as Planck.

Bernard,

I am a co-founder of the Particle Data Group, I have done statistical analyses of data all my life. Therefore I am not the least worried or impressed by disagreements of the order of a few standard deviations. Thus 2.79 +- 0.56 is the same thing as 3 or 3.23, they are not "off" one another. Rather they confirm each other.

In the case of the Higgs boson, there are indeed two experiments, statistically in good agreement. Don't worry about your capacity as an "outsider"!

Bernard Jones,

As I understand, two different detectors ('experiments') within the LHC detected expected Higgs decay products, the CMS at 125 GeV and the ATLAS at around 126 GeV. See http://cms.web.cern.ch/news/observation-new-particle-mass-125-gev and http://www.atlas.ch/news/2012/latest-results-from-higgs-search.html.

I've only read the announcement document, http://www.esa.int/Our_Activities/Space_Science/Planck/Planck_reveals_an_almost_perfect_Universe, and glanced at the more detailed report http://sci.esa.int/science-e/www/object/index.cfm?fobjectid=51551

I think the confirmation of previously ignored conflicts with predictions of the Lambda-CDM model should be cause for concern.

The announcement document states:

"One of the most surprising findings is that the fluctuations in the CMB temperatures at large angular scales do not match those predicted by the standard model – their signals are not as strong as expected from the smaller scale structure revealed by Planck.

"Another is an asymmetry in the average temperatures on opposite hemispheres of the sky. This runs counter to the prediction made by the standard model that the Universe should be broadly similar in any direction we look.

"Furthermore, a cold spot extends over a patch of sky that is much larger than expected.

http://spaceinimages.esa.int/Images/2013/03/Planck_enhanced_anomalies

"The asymmetry and the cold spot had already been hinted at with Planck’s predecessor, NASA’s WMAP mission, but were largely ignored because of lingering doubts about their cosmic origin.

""The fact that Planck has made such a significant detection of these anomalies erases any doubts about their reality; it can no longer be said that they are artefacts of the measurements. They are real and we have to look for a credible explanation," says Paolo Natoli of the University of Ferrara, Italy."

That the current standard Lambda-Cold Dark Matter model of cosmology produces results that can be well fit to current, exceedingly complex interpretations of observations does not preclude the possibility that it includes fundamental misconceptions – as demonstrated for more than a millennium by the Ptolemaic model of the cosmos.

Matt, thanks for the reassurance!

My concern went a little further than that however - there are 6 fundamental parameters measured and thy are all off. No single difference is significant, but they are all off. Given that the covariance between these should be very small, should I not just multiply the (1 - probability ) of the individual deviations to get an estimate of my confidence level?

The signal to noise of Planck is an order of magnitude greater than WMAP and it has better resolution. It is also calibrated using Atacama and South Pole Telescope. All good stuff. As you remark (and thanks) - taking Atacama plus WMAP (a post-hoc choice) is consistent with Planck, even better news.

I'm glad I am a theorist and grateful to experimentalists for what they do so skillfully! Thanks for putting me straight.

Now I'll get down to reading it all and perhaps understanding 50%.

To reiterate, Please see http://sci.esa.int/science-e/www/object/index.cfm?fobjectid=51551 which states:

"... the Planck data reveals the presence of subtle anomalies in the CMB pattern that might challenge the very foundations of cosmology. The most serious anomaly is a deficit in the signal at large angular scales on the sky, which is about ten per cent weaker than the standard model would like it to be. Other anomalous traits that had been hinted at in the past - a significant discrepancy of the CMB signal as observed in the two opposite hemispheres of the sky and an abnormally large 'cold spot' - are confirmed with high confidence. Planck's new image of the CMB suggests that some aspects of the standard model of cosmology may need a rethink, raising the possibility that the fabric of the cosmos, on the largest scales of the observable Universe, might be more complex than we think."

Bernard,

this not how to get an estimate of the confidence level because there are hidden assumptions, fitting one parameter and marginalizing over the others. Just read what they state. You cannot conclude much more than the people of the paper.

I do remember vaguely that the CMB signal could contain imprints of a collision with another Universe. Is it it somehow confirmed by the actual, more precise measurements ?

Adam, ask the people in the Planck team. I have no information (and I doubt such a claim)

Adam,

I also recall reading that - I also am highly skeptical. Please see http://en.wikipedia.org/wiki/Multiverse#Non-scientific_claims and

http://arxiv.org/abs/hep-th/0612142

Also see http://en.wikipedia.org/wiki/CMB_cold_spot#Possible_causes_other_than_primordial_temperature_fluctuation

Considering that the cold 'spot' in the sky seems to represent a low temperature 'tunnel' through spacetime (i.e., a temperature anomaly that is both directional and directionally persistent in time), perhaps it represents an origin point of expansion. While its often stated that there is no center of the universe because space appears to be expanding in all directions, the same can be said of an observer within a balloon filled with an expanding medium in which objects are suspended. All objects would appear to be receding from any observer, yet the as a whole, the spherical balloon is radially expanding away from a geometric center-point. Perhaps any observer might eventually see that the center-point is becoming increasingly disperse. Of course, the universe might not be spherical either, but I suspect there could be some large scale structure to an expanding universe while, at smaller scales, all observable objects appear to be receding omnidirectionally. This is of course wild speculation, but IMO a simpler explanation than interaction with other universes. There are potentially other, more mundane explanations that should be considered first...

Matt:: as always I appreciate your comments and in particular the nice statement "You cannot conclude much more than the people of the paper"". I am reading the papers, but I guess I have to study them too - takes more time.

James, I agree fully with you that what you propose is wild speculation.

Matts,

Obviously, this is not a proposal - merely an illustration of a potential alternative cold spot explanation to multiverses. Thanks for minding the gate, but where were you when the multiverse bandwagon piled through?

Meanwhile, I would presume the ESA Planck announcement documents were prepared under the auspices of the researchers, and that the quotations included in prior comments represent some consensus of their conclusions.

Someone in this discussion regretted that Planck had no polarization data. I just noted a comment in one of their papers that they do have polarization data, but since the analysis then is more complicated the results will be forthcoming later.

The Planck team showed me their polarization results in Fig. 11 in their paper XVI.

But they warned me that their polarization signal is weaker than the foreground noise, that they don't have a good way to calibrate it because Planck was not designed for that, so that they don't have a handle on systematic errors Once their full 4-years data have been analyzed they will know better.

With the Planck results, we have once again an announcement that all is well with the conventional model except for a few small “anomalies”, which of course should be balanced with the great success of Big Bang theory elsewhere. But the “anomalies” pointed out by the Planck team are scarcely the only ones that have been accumulating over the past decade and more. Again and again the fundamental, base line, predictions of the “concordance cosmology” have been contradicted by observations. Yet for most cosmologists no doubts can be entertained about the basic assumptions of the model that lead to these perfections.

The Big Bang theory leads to a universe that is isotropic and homogenous on the largest scales. Yet that prediction has been contradicted repeatedly by observations. In the Planck results (and in WMAP) there is the asymmetry of anisotropy power on two sides of the sky and there is the near-alignment of the octopole, quadrupole and dipole anisotropies. The polarization vectors of quasars show a major asymmetry whose axis is aligned with that of the CMB dipole. The direction of spin of spiral galaxies—directly observable from images—shows an asymmetry across the entire sky with an axis that does not line up with any of the other asymmetries. Yet another asymmetry is shown by measurements of the Hubble constant, based on supernovae brightness. In short, on the largest scales we can observe the universe is anisotropic in several different ways, in flat contradiction to the isotropic prediction that flows from the Big Bang hypothesis.

A second fundamental contradiction with observation is in the Big Bang Nucleosynthesis predictions. While not mentioned in the Planck results, BBN makes a strong prediction for the primordial abundance of lithium. But, as has been long known, this prediction is a factor of 3 above the abundances in old stars. What is much worse for the theory, the the older the star and the purer it is of admixture from later star-formation, the lower the lithium abundance. The oldest and purest stars yet detected have less than 1/100th of the amount of lithium predicted by BBN. In no other field of science could an error of two orders of magnitude be considered good agreement. Yet in “concordance cosmology” this is just a small anomaly.

The concordance cosmology requires the presence of large amounts of “dark matter”, non-baryonic matter different from any ever observed on earth. Dark matter models strongly predict that the satellites of galaxies should be randomly clustered around the central galaxy, guided by the gravitational field of the dark matter. Yet in the two galaxies where detailed studies have been made, our own Milky Way and our neighbor Andmoeda, M31, the satellite galaxies are arranged in a thin disk like a large-scale model of the solar system, in complete contradiction with the predictions of dark matter theories.

These are only three major contradictions of observation with concordance cosmology. It is by no means an exhaustive list. At what point do astronomers stop patching up the current Ptolemaic theory with dark-matter, dark-energy and inflation epicycles and start looking for a Keplerian theory which does not require one new fudge factor for each new observation? A non-expanding universe, without a Big Bang, but with the known laws of plasma physics taken into account, provides a far better starting point then does concordance cosmology. For decades, this model has predicted just what was discovered, an anisotropic, inhomogeneous universe with helium, deuterium and lithium produced in stars as today, supplying the energy we see as the cosmic background radiation.

@Simon

Simon: most interesting! Thanks for that perspective on the US show. Big Multinational Science tends to be that way.

My own worry about what I saw was that, without the polarisation data, this was never going to be more than a "much better WMAP". As you remark, they did not emphasise that this was in fact a significant step forward. Personally, given the situation, I think they should have put more weight on the "precision cosmology" aspect of this - the next decimal place! Perhaps they should also have made more of the impressive Sunyaev Zel'dovich results and the stuff on secondary anisotropies.

And I would have preferred to see "Planck:nailing the cosmos!" instead of this "Near perfect Universe" which did not light any fires. They should be proud of their achievements, technical and scientific.

But I'm glad it happened this month so I can update the draft of my CUP book to be all-Planck and put WMAP in the historical perspective it deserves.

Please reference a publication of the CMB power spectrum predicted from the "known laws of plasma physics." Of course there are no "laws of plasma physics," there are only Maxwell's equations. Just because there are phenomena that are unexplained, it does not follow that everything else that the Big Bang theory explains quantitatively should be ignored. Before COBE, the CMB showed deviations from a blackbody. Before the Key project, the universe seemed younger than some objects (globular clusters) in it. But better observations removed these anomalies, and the science moved forward.

But the contradictions with observation have been getting worse with more observation, not better. For example, as stars with less and less heavy elements have been discovered, the amount of lithium found keeps getting less, and thus further from BB predictions. The anisotropies get worse with more data, or are confirmed--they don’t go away. Science is about prediction, and the quantitative predictions of BB theorists—that is those made BEFORE data was available-- have been consistently wrong. Check out papers from the 1970’s calculating from BB theory the order of magnitude of the CMB anisotropies—the predictions were 100 times too big. When the predictions were refuted by observation, new free parameters were put into the theory. Fitting already-available data with equations that have many free parameters is not too difficult, so long as the number of free parameters keeps going up as the data improves. But that is not predictive science and it does not test the validity of a theory.

@Eric Lerner

You're just wrong about the the history of CMB anisotropy calculations. There was no first-principles theoretical basis for such calculations before inflation theory, and inflation was not invented--as you imply--to solve a problem of CMB anisotropy. New free parameters were not "put into the theory," but the discovery of dark matter and dark energy became part of the total energy and matter in the universe, just as neutrinos had to be put into the FRW equation when they were discovered experimentally--all decades after the FRW equation. The CMB itself was not even known to exist when the FRW equation was published, and yet that same equation is still used today to describe all the energy and matter content of the universe. By your logic, putting anything into the FRW equation other than baryons and starlight is dishonest.

The parameters for the CMB, baryons, dark matter, and dark energy are not free. Not only must they be consistent with many independent observations (rotation curves for galaxies, mass/light ratios for galaxies and clusters of galaxies, abundances of He, D, H, etc., large scale structure of galaxies), but they must be self-consistent. That's why I challenged you to cite published alternative theories in reputable journals that reproduce the CMB power spectrum. Until you can meet that challenge, I don't know the purpose of this discussion. You incorrectly argue, based on a few discrepancies, that "contradictions have been getting worse," as if cosmology is only based on several observations; it is based on much more. And my only reason for commenting is that the tone of your post is that of someone asserting a world-wide conspiracy among astrophysicists to support an untenable theory. Cosmologists know about the discrepancies that you highlight, but they also know that the Big Bang theory explains many observations in a consistent framework (FRW) that was established in the first half of the 20th century. Until they find overwhelming evidence to discard the Big Bang, they will keep it. But at present, nobody can make a case based on these discrepancies alone that the theory is completely invalidated.

@Daniel:

Excellently said!

Education is one area of science which needs far more resources than are currently given, with far more credit being given to teachers at all levels who do this. H-indexes depend on citations, not the number of enlightened persons.

While the dumbing-down of science by TV programs and magazines is undeniably of benefit in bringing science to a wider public, it does have deleterious side-effects in frequently propagating a more sensationalist view of science, aimed at promoting mystery even where there may be none. The result is often questions based on such experience, generally asked in earnest. Practising scientists must deal respectfully with this - as you have done, so clearly.

Of course this mystery of science is why practitioners of science do it :).

What we call large scales in the power spectrum should have actually nothing special from a cosmological point of view.... am i right?

HOwever from the observer point of view those large scales are very special in the sense that these are the largest scales (up to the dipole) one can probe observationally now.

The previous remark and the fact that the manifested preferred directions are close to the one defined by the ecliptic plane lead me to suspect that there is some kind of foreground that we dont understand and that we are not able to remove either in Wmap or Planck maps...may be new physics producing these foregrounds with properties that can hardly been distinguished from those of primordial fluctuations.

I'm aware that what makes this way of thinking difficult to follow much further is the fact that LCDM works impressively well on the small scales... so my question is: is it actually an extraordinary result to be able to fit the CMB power spectrum oscillations with six parameters?... I mean six is already a large number isn't it ?

Let me be more provocative : suppose i take the spectrum of a random AMI bipolar or Manchester signal with jitter to avoid the zeros (3 parameters) multiplied by a polynomial (3 additional parameters) ... would it be surprising to get a good fit with such phenomenological fit function?

@Frederic

The problem on the largest scales is, of course, the "statistics of one". And it's worse than that: there will only ever be the one sample of the Universe! This is what gets labelled as "cosmic variance" and the likelihood of these, possibly surprising, alignments and so on occurring is assessed in the light of that.

Has anyone computed or presented a "multiverse" model which might provide a framework within which we can assess this? Does that even make sense? Would anyone believe it if you did it?

It's not an area where I am at all knowledgeable so if anyone out there has answers please post!

@bernard jones

i 'm aware that there is a large cosmic variance on the largest scales.

But:

1- If there is something we dont understand in the foregrounds it's certainly on the largest scales that we have chance to see it as an anisotropic signature simply because the more anisotropic distribution of foregrounds is also that of neighbour structures (Milky way , neighbour galaxies, and may be our own solar system ) ... and as a matter of fact the anomalous preferred directions revealled up to now both in Planck and Wmap maps are surprisingly close to the ones related to the eclipic plane or the galactic plane.

2- Yes the probability that there is something really anomalous regarding the low power in the lowest multipoles is only at the percent level (they say)... however when they started to have a closer look at those anomalies looking at other observables then they discovered (both in Wmap and Planck) a dipole modulation over almost all the multipoles , and the probability of this is much lower as far as i remember!!

they also see incredibly low powers in some multipoles (l=6 (even negative!), l=22, l=26 !! in one hemisphere:

see at 47:17

http://spaceinvideos.esa.int/Videos/2013/03/Replay_of_Planck_media_briefing_-_Part_2

3- LCDM + inflation has all the characteristics of an epicyclic construction ... (add some new parameter each time you find something new, ...wait until you have 6 or a little bit more parameters, put under the carpet anything that's not completely understood and you will soon ... explain everything ;-) )

There is an intriguing possibility which seems to being overlooked. Maybe the low l multipoles are intrinsically extremely low and it's only because of this we can see some local large angular scale foreground which aligns with the ecliptic plane (large qualitites of very cold dust (

@Bob

Well you are in on the action and I guess that the team must have discussed such possibilities at great length - they (you) obviously know more about the foreground removal than anyone else!

For those of us with strong interest in this we now have to (or wish to) read and appreciate a significant fraction of those 30 or so papers, some exceeding 60 pages in length! So, in the final analysis, all we can do is believe the published conclusions. It's a great team. On top of that, lots of ground-based stuff.

My own interest is my upcoming book on "Precision Cosmology" (CUP) - I have to hand it in real soon now and I only wish the Planck data had arrived a few months earlier! (The book is at the post-grad post-doc level). As I read and try to understand those papers I shall be posting more questions in various places.

Currently I am on the E-mode and B-mode chapter, even though there is as yet relatively little Planck polarization data.

If there is no multiverse, is there no Big Bang?

@Cj

This "multiverse" idea is just that - an idea - it's not a theory and it could probably never be tested. I am not sure we even know how different universes might interact (tidal forces??). But I'm far from being an expert on that.

So if we get simple and suppose there is no multiverse, our Universe is all there is and it is the way it is because it was the way it was (Herman Bondi's favorite remark). It started, it inflated (probably), it stopped inflating but continued expanding and eventually there will be nothing for us to see (or will the protons decay before that happens?). Some people evidently have a problem with that "one Universe - this is it" concept, hence multiverses.

Wikipedia has a nice short bit on the string theory landscape.

Bernard, thank you for your time in entertaining my question. I don't feel there's a multiverse either. But then that would bring into question, From what did the Big Bang start? In other words, if we accept no multiverse, then we would also have to accept no Big Bang or start. Unless, although as unplausible, we admit a universe ex nihilo, narrowing realistically the most plausible of possibilities down to our well-established, well-known only universe with no Big Bang and a better explanation for its 'expansion' or the redshift and everything else that occurs. Don't you think? How can a 'Big Bang' or 'start' be otherwise plausibly accepted without something, even if a mystery or unknown, in the SAME 4-D universe preceding or precipitating that Big Bang?

@robert

yes it's possible that the low l multipoles are intrinsically low and even vanish for the quadrupole after removal of all kind of dipolar modulation effects.

Http://arxiv.org/abs/1203.5720

@Bernard. Thanks, although I'm more on the systematics data analysis side. A big part of Planck is the component separation thanks to the many frequency channels, but something unexpected with a spectrum close to CMB and doesn't correlate to the warmer dust might get through, I guess. That's a question for the component separtation team using the full data set!

@Cj

Stephen Hawking, in his book Brief History of Time, addresses the issue of "what is the Big Bang and where did it come from?". This was a radical departure from earlier thinking where people had simply dismissed the question as being "non-scientific". It is non-scientific in the sense of being unfalsifiable, but theorists don;t mind about that since the question is whether we can construct a theory which includes a description, however bizarre, for this event and perhaps its precursor. Steve came up with just such a theory: in essence that there was no such thing as time prior to this event. Read the book - many many people did.

Our terms of reference are somewhat different now. We think in terms of our being higher dimensional spaces - "Brane universes". Unprovable? Ridiculous? - that's a matter of opinion. Just think - we are contemplating verifying and studying the period when we think the Universe was "inflating"! We propose to do this using the gravitational waves that would be associated with the inflation process, and observing them via the B-mode of the polarization of the Cosmic Microwave background (which Planck is not sensitive enough to see).

We already have an interesting clue from Planck ( bit technical but incredibly important): the Planck constraint on the Gaussianity of the spectrum of primordial density fluctuations are quite strong, even in this first data release. This is somewhat unexpected on the basis of many discussions of the mechanism through which these fluctuations are thought to be generated. This is exciting stuff because it is an additional hint about the nature of the big bang itself.

The other hint, again mighty technical, is the slope of the power spectrum of those density fluctuations that led to the origin of stars and galaxies. Elementary thinking inclines us to the belief that this slope is 1.00... It turns out that this slope is a little less than that, as expected on the basis of one of the simplest inflation models (Alex Starobinsky). But that's a technical issue for a different discussion.

WMAP, Planck and all those fantastic ground based CMBR experiments are laying the ground-work for a future in which we can discuss the earliest moments of our Universe - imagine!

As for multiverses and big bang, I don't see the need for a connection. Steinhardt's ekpyrotic universe (witth lots of problems) is cyclic . Someone has proposed that the Universe was just coasting along forever without any time evolution until big bang. At that time it was governed by quantum mechanics, that does not predict the occurrence of single events, nor their outcome. Multiverses seem like fairy tales.

Bernard, Thanks once again for your gentle exposition and for the Hawking reference, it may be in agreement in a sense with my concept of time as nonexistent before the Big Bang and in that sense important to consider.

As to the branes concept, I find that partially correct, but they are not a complete description of the universe in and of themselves. However, that will be a smart approach to measure the gravitational waves against the cosmic microwave background, that ratio would be roughly close to the Planck length, might it not (this before reading your Planck-Gauss paragraph)? If the Planck length cannot be surpassed, i.e., turns out to be that ratio, then there really isn't any big bang theory or 'start' as I previously proposed, or as you phrase, "an additional hint about the nature of the big bang itself," no?

I'm glad the slope is found to be less than 1 (a good starter), considering a logarithmic nature of the universe. Yes, all awesome.

A news report describes some controversial interpretations of the Plank data in conjunction with CERN assessments of the initial state of the Higgs field. The new report casts doubts on inflationary models, but it's meeting with opposition. See http://www.nature.com/news/higgs-data-could-spell-trouble-for-leading-big-bang-theory-1.12804 & references.

"In particular, the researchers note that as the Planck team narrowed down the list of possible inflaton fields, the models that best fit the data — known as ‘plateau models’ because their potential-energy profiles level off at relatively low energies — are far less likely to occur naturally than the models that Planck ruled out.

"But the news for these plateau models gets dramatically worse, the researchers say, when the results are analysed in conjunction with the latest results about the Higgs field coming from CERN's Large Hadron Collider (LHC) near Geneva. Particle physicists working at the LHC have calculated that the Higgs field is likely to have started out in a high-energy, ‘metastable’ state rather than in a stable, low-energy configuration."

Thank you James for the information. Knowing Steinhardt I'm one of the people who does not get too worried. He is mainly and provocatively defending his own cyclic (ekpyrotic) model - he may be right, of course.

For both time periods before and after Planck there is the same central conceptual question about the physical sense of the Cosmic Microwave Background Radiation.

This is the problem of energy-momentum conservation in the expanding space of the Friedmann’s cosmological model (see e.g. Harrison 1981 p.276; Peebles 1993 p.139).

“The conclusion, whether we like it or not, is obvious: energy in the universe is not conserved”. For the case of the photon gas it means that during space expansion the huge initial energy (temperature) of the hot photons disappear nowhere so now we see 3K degree radiation as the proof of the violation of energy conservation in the Big Bang model.

Yurij Baryshev,

Shouldn't it be considered that the physical expansion of spacetime also require some conversion of energy? In this case, perhaps the 'lost' thermal energy of photon redshift is dissipated within the expanded spacetime, contributing to vacuum energy?

Yurij, James:

I don't get it!

For most of the expansion of the Universe the ratio of photons to baryons remains constant, so cosmic photons are neither created nor destroyed. Their energy behaves precisely as you would expect from general relativity, and so nothing is gained and nothing is lost.

Maybe you are asserting a disbelief inEinstein's general relativity? Yet it is an extraordinarily good approximation (my sat-nav works beautifully)! Even Fred Hoyle's continuous creation theory managed to conserve energy.

Puzzled.

Bernard,

Sorry, but I don't understand how GR relates to cosmological redshift - can you explain briefly?

As I understand, the conservation of energy applies not just to numbers of particles - in effect photons are created when emitted and destroyed when detected. It is the total energy of an isolated system that must be conserved in accordance with the law.

I have some doubts if the energy conservation principle can be applied to the Universe as a whole, which certainly cannot be treated as a closed, isolated system.

Stretching space implies redshift so even if the number of photons stays constant, their energy decreases. Not even to mention dark energy evolution which starts to dominate, causing expansion acceleration.

Bernard,

Conservation of the number of particles in any local commoving volume is the only case of the energy conservation in expanding universe. But any kinetic energy of particles and also energy of photons in any local commoving volume (LCV) do not conserve and disappear nowhere from the LCV (not that happens locally and just as the sum of the effect also related to the whole universe). This is the well-known fact and it presented in all text-books on cosmology, e.g. lost of momentum of free particles inversely proportional to the radius of the universe.

So there is no doubt in the violation of energy-momentum within any LOCAL commoving volume in the frame of Hot Big Bang cosmology based on the general relativity. For clear derivation of corresponding formulae you may see

http://arxiv.org/abs/0810.0153.

James,

In the Friedmann cosmological model, which is a particular solution of the general relativity equations, the cosmological redshift is due to stretching of any individual photon due to expansion of the space. And during this process the energy of each photon disappear nowhere from our universe, while the number of photons conserved.

Adam,

As I note above the violation of energy conservation in the HBB model happens for any LOCAL commoving volume and the 3K microwave background radiation is the result of energy disappearing nowhere from any LCV containing initially hot photon gas.

This is a simple mathematical fact of general relativity, and the question is - will we accept it as the phenomenon of a dramatically new physics or we have here the limits of GR applicability?

@Yurij:

Yuij, I do make a point of reading some of your papers (and had read the one you referred me to, thank you). I certainly read yours more than those of your like-minded colleagues since your arguments are generally so well-put. While I strongly disagree with some of the points of view you advocate, I think it is important to be able to address those issues, particularly when teaching students.

Having said that, I should add that, I do agree that at this time there are things we cannot really know about the Universe and that may forever be mere speculation. But speculation is fun and is part of the essence of science. Perhaps I wish that people who criticize the current paradigm would offer some serious alternative that is founded on as strong a base as Einstein's theory, as has been done rather well by the Bekenstein - Milgrom - Sanders group who write on alternatives to dark matter.

But look at the inflation story: a couple of decades ago we would not have imagined that we could confront that with any kind of experimental data. The coming years will certainly impose constraints on these ideas, and we may be able to conclude that, within the framework of our current paradigm, some ideas on inflation are certainly more attractive than others. To me that is exciting and is part of the excitement that the CMBR experiments bring to the table.

Who knows? - in a couple of decades when I am long gone there may be experiments to touch some of those higher dimensions which some people speculate about.

Bottom line?

Constructive criticism is essential for the well-being of our science.

Dear Bernard,

thanks for your good answer and correct view the modern state of cosmology.

As for me I do not like when people hide real problems under the carpet.

It is interesting that only recently in the scientific literature we see discussion such conceptual problems of the standard cosmological model as the nature of cosmological redshift (which is not the Doppler effect), receding velocities of galaxies more than velocity of light, violation of energy conservation in expanding space, exact Newtonian character of the Friedmann equations.

Yes, I agree with you that to propose alternative explanation of existing cosmological observational facts one needs to know well the standard cosmological model and its foundation on modern physics. As Feynman wrote if one wishes to suggest a new physical theory he should take into account all existing physics and get it as limiting case of a new theory.

The true basis of HBB cosmological model is the Einstein’s general relativity, which has a good experimental verification in the weak gravity conditions. Now there are a lot of modifications of the geometrical gravity theories which is developed by the modern theoretical physics (e.g. Clifton et al., Phys.Rep., 513, 1, 2012), which also modify cosmology.

What is more there is an alternative approach to the theory of gravitational interaction as a relativistic quantum tensor field theory with natural definition of force, energy, quanta of the gravity field. The foundation of the Field Gravity was discussed partly by Poincare, Thirring, Feynman, and the first results on this way were obtained, see our recent Springer book:

http://link.springer.com/book/10.1007/978-94-007-2379-5/page/1

Here there is such possibility as cosmological gravitational redshift and evolution without expanding space.

Dear Yurij,

It would be interesting to argue with you about the modifications of GR that you think are most promising. You refer to the 312-page review paper of Clifton et al,and to your own 332-page book with Pekka Teerikorpi (thanks for the references!) , both of which cover such vast amount of material, that one cannot easily sift out some ideas that could usefully be discussed in this limited forum.

Let me pose a very specific question. The Universal Rotation Curve of spiral galaxies, discovered by P. Salucci and collaborators, shows clearly that ordinary luminous matter and dark matter interact near the center of the galaxies as if there were another force. As a consequence the cusped DM cores of the LambdaCDM model appear flattened, i.e. cored. F(R) models don't work. Perhaps you could rule out a large number of models and thereby simplify the search?

Matts Roos,

As I understand, the 'Universal Rotation Curve' indicates that a standard rotation curve can be fitted to many galaxies when they are configured in simulation models with a special purpose distribution of undetected dark matter. IMO, the better question is: what property of any candidate exotic dark matter particle could maintain any cuspy halo distribution, both at its center and its periphery - solely through its allowed gravitational interaction?

I would suggest that it's quite a stretch to infer the existence of some new force in order to explain the behavior of an inferred exotic form of matter necessary to fit very specific configurations of mass distributions to a generalized rotation curve model.

In other words, what property of undetectable dark matter causes it to so 'universally' be present in such large, very specific, varying proportions and distributions necessary to fit so many galaxies to a generalized rotation curve model?

Dear Matts and James,

the well-known “cusp problem” is the fact that standard LCDM model of galaxy formation predicts the fast increase of mass density close to the center of a halo as rho ~ 1 / r (Navarro-Frenk-White profile), while observations show that galaxies have cores with flat mass density profiles. In many papers this contradiction of the LCDM was used for developing different models which have self-interacting or new properties of the dark matter.

However in recent papers by Italian – France cosmologists team it was demonstrated that the “cusp problem” is an artifact of N-body simulations with non-sufficient accuracy at small scales of the simulations (see e.g. F. Sylos Labini, MNRAS, 429, 679 – arXiv:1211.1278 and arXiv: 1303.4531).

So observed flat density profiles in the galaxy cores do not need a new properties of dark matter.

Well Yurij,

you addressed the second half of my question: Sylos Labini indeed has a mechanism to flatten cuspy profiles, but only for objects which start out as a spatially uniform and isolated gravitating cloud. The end result can then be an elliptical galaxy with a flat, cored DM profile. In spirals the cored Burkert profile fits better than the cuspy ones, but these profiles are just empirical formuli, so Sylos Labinis demonstration is better if it can be applied to spirals.

But the first half of my question needs to be spelled out better. In spirals you can infer the DM halo profile from the difference between the observed rotation velocities of stars (ellipticals don't rotate) and the observed luminosity profile. No simulation is needed within the radius of visible stars.

Salucci then found that the DM profile has a shape depending on the total luminosity of the galaxy. He has no explanation for this except some hand-waving remarks, that it could depend on the formation history of the galaxy. The dependence is well visible in the URC picture, see e.g. Mon.Not.Roy.Astron.Soc.378:41-47,2007.

So my question is: given (what I call) the Salucci effect, can you rule out a host of DM models?

Dear Yurij and Matts,

Yurij - thanks for the interesting links. You're correct in your discussion of the core-cusp problem in LCDM cosmology - it is a conflict between the cuspy-core predictions of n-body simulations of initial galaxy formation that focuses on the collapse of a CDM halo, and the observed generally flat inner radii velocities of galaxy rotation curves.

However, the rotational velocities of objects at low radii in spiral galaxies is much lower than predicted by gravitational evaluations without dark matter. As I understand, _any_ dark matter within the central region of of a spiral galaxy would only increase the discrepancy between the predicted and observed rotational velocities at small radii. The requirement for dark matter halos to compensate for the discrepancies between the expected and observed rotation curves of spiral galaxies is that it tremendously increase peripheral galactic mass, in order to increase the expected rotational velocities of peripheral disk objects to match expectations.

Matts - see Yegorova, I. A., Babic, A., Salucci, P., Spekkens, K., Pizzella, A., (2011), "Rotation curves of luminous spiral galaxies,"

http://dx.doi.org/10.1002/asna.201111582

http://arxiv.org/abs/1110.1925

Its abstract concludes:

"... We find that in the outermost parts of the stellar disks of these massive objects, the rotation curves agree with the Universal Rotation Curve (Salucci et al. 2007), however a few rotation curves of the sample show a divergence."

I haven't read the paper - I presume from the qualified summary that for much of the galactic radii the rotation curves do not agree with the Universal Rotation Curve - but perhaps I'm wrong...

James,

In the Salucci paper that you refer to they state: on p.6: "...also in high luminosity galaxies, a disagreement between naive LCDM predictions and the actual data does clearly emerge". (I have read all the Salucci papers, not just one abstract.)

My point is not whether data agree or not with the URC, because the URC is empirical and therefore defined by the data. Let me instead direct you to Fig.5 of http://arxiv.org/abs/0707.4370, a figure often reproduced There you see that the DM curves (that is, the difference between the observed rotation curve and the observed luminosity) depend on the total luminosity of the galaxy. Why?

As Salucci points out in the Conclusions of 0707.4370:"The distribution of luminous and dark matter in galaxies shows amazing properties and a remarkable systematics..." Still in 2013 he writes (http://arxiv.org/abs/1302.2268): "In spirals there is fundamental evidence that dark and luminous matter are well linked together." This is an evidence not addressed by anybody.

Matts,

The issue I was noting was their qualification that for nearly all the 900+ galaxies studied, the Universal Rotation Curve agreed with the rotation curves _in_the_outermost_parts_of _the_stellar_disks.

Why do only the outermost rotation curves agree? As I mentioned, any additional mass at the center of galaxies would increase the expected rotational velocities at small radii, based on standard gravitational evaluation methods, which were already greater than those observed.

That the summary mentions that a few rotation curves diverge (from agreeing with the outermost parts of the stellar disks) is not at all the 'central' issue!

I have a question. If the CMB spectrum is produced on an assumption that the data that it uses is described by a Gaussian field, then how is it possible to measure the non-Gaussianity of the field? Or have I misunderstood something big?

The CMB angular power spectrum does indeed assume the field is Gaussian, that is to say the spherical harmonics are random and uncorrelated. Non-Gaussianity will introduce correlations between harmonics in ways which depends on the type of non-Gaussinity. it is weak then the effect on the power spectrum is tiny and very diluted. This doesn't stop the field from having non-Gaussian features, just that the power spectrum doesn't capture all the information of the field. For example point sources are non-Gaussian signals, which are easily found in the field but just add a small increase in the overall amplitude of the angular spectrum.

The assumption of Gaussianity, if made, is just that - an assumption. But we can test for non-Gaussianity by assuming a model in which there is a small deviation from Gaussianiity, the magnitude of which which we parameterise by a parameter called f_NL (ie: f with a subscript 'NL' which stands for 'nonlinear'). With that we can, via statistical analysis make an estimate with error estimates.

It turns out that the estimated value of f_NL from the Planck Satellite data is not significant and is consistent with being zero:

http://arxiv.org/abs/1303.5084

Of course, that does not exclude a very small value of f_NL - it just asserts consistency with the Gaussian assumption, at the level of sensitivity of the Planck data as currently presented.

The paper looks at several methods for analyzing the degree of nonlinearity - it's highly technical but well written. The main results are summarised in their tables 5,6,7 and 9. The last table, 9, looks at the consequences of using different map-cleaning methods, just to be sure that there is nothing amiss in the data cleaning.

Of course "consistent with zero" does not mean it is zero. The major source of any putative non-Gaussianity would be things happening during inflation - a very trendy topic these past weeks. So there is an entire zoo of papers on what different inflationary theories produce in terms of gravitational waves (that we have seen) and non-Gaussianity which we have not seen).

Exciting times!

@Bob

Sorry - we submitted responses at the same time.

I did once wonder about the situation of a distribution having Gaussian amplitudes but with a non-random distribution of phases due to some weird primordial process. I'm not sure whether the bi- and tri-spectra would pick that up.

Do you know the answer?

@Bernard

No problem. Your answer is more detailed.

Indeed I'm not sure the bi & tri would capture them, but the Minkowski functionals the Santander group like so much might. Not quite my field.

"If the CMB spectrum is produced on an assumption that the data that it uses is described by a Gaussian field, then how is it possible to measure the non-Gaussianity of the field?"

I know nothing about it, but in terms of the polarization of CMB light attributed to primordial gravitational waves, isn't the presumption of CMBR Gaussian compliance critical to the identification non-Gaussian features as having been introduced by external influences - in this case primordial gravitational waves?

If they had had any polarization data before they would have taken the contribution of both of the waves- density and gravitational -into account in CMBR measurements. I think both waves have different effect on the polarization data. I am not sure how though. But noone has confirmed the indirect detection of gravitational waves yet, right?

Heleri,

I'm pretty sure that no one has yet independently repeated the identification of B-mode signal in the CMBR.

For a skeptical evaluation, see the 3-part blog posting, beginning with the pre-announcement posting http://blankonthemap.blogspot.com/2014/03/b-modes-rumours-and-inflation.html.

The second article was posted on the Mar. 24, 2014 announcement day at http://blankonthemap.blogspot.com/2014/03/bicep2-reasons-to-be-sceptical-part-2.html. It addresses the question "How certain can we be that the observed B-mode signal is cosmological?" Most importantly, it simply points out that more complete dust data captured by Planck is now being analyzed (dust may also contribute to identified B-mode CMBR polarization):

"... Now one real test of these assumptions will come from Planck, because Planck will soon have the best map of dust in our Galaxy and therefore the best limits on the possible contamination. This is one of the reasons to look forward to Planck's own polarization results, due in about October or November."

It also suggests that phenomena other than primordial gravitational waves may have contributed to the B-modes CMB polarization, concluding:

"The chances are much higher – I'd be tempted to say perhaps even as much as better than even money – that foregrounds contribute a part of the observed signal, and that therefore the actual value of the tensor-to-scalar ratio will come down from r=0.2, perhaps to as low as r=0.1, when Planck checks this result using their better dust mapping."

As of this writing, the third article has not yet been written. It will address whether the identified BICEP2 gravitational wave signal is actually associated with inflation...

James,

there are already 64 papers in arXiv on BICEP2, you can forget about blogs

Heleri,

Sorry I don't think I answered the question you asked which perhaps implied that the B-mode signal was non-Gaussian. It is Gaussian but not covered by the purely temperature angular power spectrum (TT) as they are part of the polarized fields which are E and B. These E and B distributions can also be described by spherical harmonics which are uncorrelated and random just like the temperature distribution and so are Gaussian.

The EE signal was seen by DASI several years ago, but is caused by the local temparature isotropy scattering differently into different linear polarizations when the Universe became transparent. Just to complicate matters this causes correlations between T and E due to this scattering process off electrons, which has also seen. This is still Gaussian

It's the handedness or parity of the gravitational waves which causes the special spiral B-mode pattern in polarization everyone is looking for.

Pulsar observers will tell they saw indirect evident of gravitational waves, but very indirect in my point of view. Seeing the run down rate of a fast binary pulsar orbit (PSR 1913+16) consistent with gravitation radiation emission rate via General Relativity.