Big Bang ruled out as origin of lithium-6

[TrickyDicky]

http://physicsworld.com/cws/article/news/2014/sep/02/big-bang-ruled-out-as-origin-of-lithium-6

The hystory of BB nucleosynthesis is quite busy, Gamow et al. started postulating that all elements were originated in the BB nucleosynthesis process in the late 40s early 50s, it soon became quite evident through the work of Hoyle and others that it couldn't be so and the list was reduced to the light elements, slowly but steadily more elements fell off, now with Lithium ruled out, which BB origin was a central prediction of BB nucleosynthesis, it seems it's down to twoydrogen and Helium. Actually what the nucleosynthesis would predict in principle is their proportion, approx. 1/4 in mass for Helium, which is in basic accordance with its observed abundance.
That the BB nucleosynthesis theory actually came up with this proportion(1/4) has been critiziced in the past by Hoyle(and others) who claimed that other theories without Big-bang, including his, also predicted it.

Now BB nucleosynthesis used to be one of the three pillars of the Big-bang theory(the others being cosmological redshift and the CMB), is the ruling out of Lithium as originated in the Big-bang a big blow or just a scratch to the theory?

 

[phyzguy]

You're selling BBN short. It very accurately predicts the abundances of H, D, He-3, and He-4. The question is why the measured abundance of Li-7 in old stars is 2-4X lower than BBN predicts. There are a number of possible explanations, the most likely being (IMHO) that some of the Li-7 is consumed by nuclear reactions in these stars. This interpretation is supported by this Nature paper, which measured the abundance of Li-7 in interstellar gas (not in stars) in the SMC, and found about the amount predicted by BBN. I suspect it will be some time before this is all sorted out, but I predict that the standard Lambda-CDM model of cosmology will emerge unscathed.

 

[TrickyDicky]

"First Direct Measurement of the H2(α,γ)Li6 Cross Section at Big Bang Energies and the Primordial Lithium Problem"
Phys. Rev. Lett. 113, 042501 – Published 21 July 2014
M. Anders et al. (LUNA Collaboration)

This is the paper that the article linked above refers to. This specific issue seems to have already been sorted out.

I'm not selling anything short, just presented some apparently not controversial scientific facts and wondered how they affect the theory. I don't think this gets solved simply by appeals to faith in the current concordance model, which has been changed several times thru the years, the last important one as recently as 1998, to accommodate new facts.

 

[phyzguy]

When I said you were "selling BBN short", I was referring to your statement that, "it seems it's down to twoydrogen and Helium", meaning ( I think) that BBN only successfully predicts the abundance of two elements. However, it really successfully predicts the abundances of four nuclei, H, D, He-3, and He-4, and only misses on one - Li-7. Given this, it seems more logical to suspect the Li-7 measurements than BBN itself. But time will tell.

 

[TrickyDicky]

When I said you were "selling BBN short", I was referring to your statement that, "it seems it's down to twoydrogen and Helium", meaning ( I think) that BBN only successfully predicts the abundance of two elements. However, it really successfully predicts the abundances of four nuclei, H, D, He-3, and He-4, and only misses on one - Li-7. Given this, it seems more logical to suspect the Li-7 measurements than BBN itself. But time will tell.

Ok, but I was obviously referring to the 2 elements generically(as it's done in the physicsworld article), D and He-3 are obviously isotopes of H and He respectively. But it doesn't make much difference if it's down to 2 or 4 isotopes, because the abundances are closely connected in BB nucleosynthesis in the sense that it is a process and therefore abundances of some influence the others. That's the reason why the observed discrepancy in abundance of Lithium-6 and -7 is a real problem for the whole BBN model.

 

[CKH]

My reading of the article is that the BB nucleosynthesis calculations agree with experiment (at least under the conditions of this experiment). So this is just a verification of the problem. My understanding is that in stellar measurements there a consistency, quite different from BBN that needs to be explained. That is, the lithium level doesn't seem to depend on precise stellar age or precise mass.

I've only read the abstract in the Nature article. I'm skeptical that some measurement of gas in the Magellanic cloud can be taken as primordial.

We measure only the stellar surface (outer atmosphere). Yet it seems that these abundances are taken as uniform throughout the star. That seems unlikely in that mass and atomic size ought to have some migration effects. Are these effects somehow taken into account?

Regarding "more logical to suspect Li-7 measurements than BBN itself". That's interesting because it's like saying that the theory trumps measurement. The theory is apparently considered so well proven that usually when there is disagreement with measurement, it is called "tension" and everyone scrambles to find a way to relieve it. Many, if not most, papers that find discrepancies focus on reducing this "tension" by searching for other effects to bridge the gap. On the other hand, measurements apparently in agreement with theory, don't get much focus. That is we don't usually look for other effects to explain a result in agreement with theory, even though such effects may well exist.

 

[Drakkith]

On the other hand, measurements apparently in agreement with theory, don't get much focus. That is we don't usually look for other effects to explain a result in agreement with theory, even though such effects may well exist.

That's simply not true. Theories are developed to explain observations and measurements. New observations and measurements are taken and old ones are refined, and if the theory disagrees with the observations then we have to figure out why. Many times it turns out that our measurements are either wrong, or our understanding of how the measurement applies to the theory is wrong. In addition, many times competing theories are explored even though the observations don't contradict the accepted mainstream theory. A good example is the various alternatives to special relativity: http://en.wikipedia.org/wiki/Special_relativity_(alternative_formulations)

I think the problem is that people rarely hear of alternative theories that have been explored or all the experiments that are performed to verify that current theories are correct.

Also, remember that science is a work in progress and if you aren't one of the ones actively working in a field you're extremely unlikely to understand all the work going on and how it affects the current working theory.

 

[CKH]

I agree with what you say above, but I don't see how it contradicts what you quoted.

I often see the phrase in papers "in good agreement with LCDM" referring to observations and their interpretation supporting LCDM. However when the conclusion is "in substantial tension with LCDM", that paper is going to draw at lot more attention. The observation and it's analysis are going to scrutinized. Researchers are going to pick it apart piece by piece. If the result of this scrutiny is that the paper is valid, then researchers are going to put considerable effort into explaining that discrepancy in the context of LCDM.

Even if they cannot explain such a discrepancy, they expect that an answer will come that is consistent with the theory. The decades old Lithium discrepancy is a case in point.

Right now, as far as I know, there is no alternative theory that agrees with cosmological observations as well as LCDM. However it is also true that LCDM has many problems on the scale of galaxies which could mean something is wrong despite the agreement at the level of the entire cosmos..

 

[e.bar.goum]

It's worth noting that those 6Li abundance measurements published by Asplund in 2006 disappear if you go to a 3D NLTE model of the solar atmosphere (see Lind, Asplund, Collet, Melendez 2012). This is hardly surprising if you know how measurements of 6Li in stellar atmospheres are made - they manifest as a slight asymmetry in the Li line shape.

I don't think anyone really thinks there's a "primordial 6Li problem" any more.

Now, the 7Li problem, on the other hand, is somewhat more intractable. My personal bias is that it will be solved by a combination of increased mixing in stars (which gets you about half-way to BBN abundances), with maybe a variation of nuclear reaction rates associated with 7Be destruction.

 

[Bob Enyart]

Hi TrickyDicky! Thinking about your question, how much "the ruling out of Lithium as originated in the Big-bang" affects the theory's track record regarding predictions, and phyzguy brought up hydrogen and helium, we should also keep in mind its prediction on antimatter.

An extensive search for the antimatter predicted by the BB ended with a paper in the Astrophysics Journal concluding that "a matter-antimatter symmetric universe is empirically excluded" with the journal Science reporting a physicist's assessment that, "The work is extremely compelling and gives me fresh pessimism" that the theory's prediction has failed, i.e., that there is not an entire universe's worth of antimatter out there. Fermi lab produced a short video to explain the problem:



TD, you also mention the BB's CMB prediction. That's an interesting one because it was more of a post-diction, made after observations had already indicated the data. In his lecture given on the very occasion of sharing the Nobel Prize for discovering the Cosmic Microwave Background Radiation, Robert Wilson, http://www.nobelprize.org/nobel_prizes/physics/laureates/1978/wilson-lecture.pdf that, "The first confirmation of the microwave cosmic background that we knew of, however, came from a totally different, indirect measurement. This measurement had, in fact, been made thirty years earlier by [Mount Wilson Observatory's] Adams and Dunhan... from the first rotationally excited state. McKellar using Adams' data... calculated [via absorption lines from cyanide detected in outer space] that the excitation temperature of CN was 2.3 K. This rotational transition occurs at 2.64mm wavelength, near the peak of a 3 K black body spectrum." [Robert Wilson, Nobel Lecture, http://www.nobelprize.org/nobel_prizes/physics/laureates/1978/wilson-lecture.pdf, and see also http://www.nobelprize.org/nobel_prizes/physics/laureates/1978/penzias-lecture.pdf]

So TD, if lithium-6 is a scratch, antimatter must be another scratch. :) Now I'm wondering about phyzguy's statement about hydrogen and helium.

 

[Torbjorn_L]

if lithium-6 is a scratch, antimatter must be another scratch. :) Now I'm wondering about phyzguy's statement about hydrogen and helium.

I'm a bit confused over what you are trying to say here.

- The Li-7 problem is an observation with stress re cosmology, but as the thread notes it is spurious, likely partial (only in stars, where nucleosynthsis is less well understood), and small (less than an oom).

- Matter/anti-matter asymmetry on the other hand is a big problem. Not too big, since the basic mechanism was discovered by Sakharov. But AFAIU the lack of predicted constraints puts it many oom from the matter content that is observed.

- Incidentally, the observed matter/anti-matter asymmetry is perfectly consistent with the CMB relict photons. Yes, its existence was a post-diction according to the timeline [ http://en.wikipedia.org/wiki/Cosmic_microwave_background#History ], but the characterization of its anisotropy has many features that are predicted (multipole moments vs total energy, dark energy, dark matter and matter; polarization).

TL;DR: Not much of a problem with cosmology as such. But the matter/anti-matter symmetry is a long-standing open question.

 

[Bob Enyart]

I'm a bit confused over what you are trying to say here.

Hello Torbjorn_L! These two, Anders and Johnston, are more qualified than I am to put the lithium problem in perspective.

After M Anders' Physics Review Letters paper concluded that "The much higher Li6/Li7 values reported for halo stars will likely require a nonstandard physics explanation," Physics World's Hamish Johnston wrote:

The BBN model predicts that lithium-6 should account for about two out of every 100,000 lithium nuclei in "metal-poor" stars, which are believed to be among the first stars to have formed and so should reflect the composition of the early universe. However, observations... suggest that the abundance of lithium-6 is more than a thousand times greater in such stars, accounting for about 5% of all the lithium present.
​

And so Johnston, who is the editor of Physics World, summed it up this way: "Big Bang nucleosynthesis (BBN) theory... fails miserably when it comes to the two stable lithium isotopes: lithium-6 and lithium-7."

So when TrickyDicky asked, "is the ruling out of Lithium as originated in the Big-bang a big blow or just a scratch to the theory?", I was thinking that when folks talk about predictions as evidence in support of the big bang, they should also just as readily acknowledge where the predictions were actually post-dictions, and where other predictions, like with antimatter and lithium, just haven't held up.

Then there's phyzguy's statement about the BB predictions of abundances of hydrogen and helium. As with all of science history of course, there's an interesting story there too.

 

[mfb]

I often see the phrase in papers "in good agreement with LCDM" referring to observations and their interpretation supporting LCDM. However when the conclusion is "in substantial tension with LCDM", that paper is going to draw at lot more attention. The observation and it's analysis are going to scrutinized. Researchers are going to pick it apart piece by piece. If the result of this scrutiny is that the paper is valid, then researchers are going to put considerable effort into explaining that discrepancy in the context of LCDM.

If you expect to measure 0 (as deviation from a theory from example) and 99 measurements are compatible with 0 while one gives a significant deviation, it should be obvious that this one gets the most attention. It is clear evidence that something is not understood. Sure, there could be smaller errors in the other experiments, but the chance that two significant problems happen at the same time (to give a null result again) is very small.

 

[e.bar.goum]

Hello Torbjorn_L! These two, Anders and Johnston, are more qualified than I am to put the lithium problem in perspective.

After M Anders' Physics Review Letters paper concluded that "The much higher Li6/Li7 values reported for halo stars will likely require a nonstandard physics explanation," Physics World's Hamish Johnston wrote:

The BBN model predicts that lithium-6 should account for about two out of every 100,000 lithium nuclei in "metal-poor" stars, which are believed to be among the first stars to have formed and so should reflect the composition of the early universe. However, observations... suggest that the abundance of lithium-6 is more than a thousand times greater in such stars, accounting for about 5% of all the lithium present.
​

And so Johnston, who is the editor of Physics World, summed it up this way: "Big Bang nucleosynthesis (BBN) theory... fails miserably when it comes to the two stable lithium isotopes: lithium-6 and lithium-7."

So when TrickyDicky asked, "is the ruling out of Lithium as originated in the Big-bang a big blow or just a scratch to the theory?", I was thinking that when folks talk about predictions as evidence in support of the big bang, they should also just as readily acknowledge where the predictions were actually post-dictions, and where other predictions, like with antimatter and lithium, just haven't held up.

Then there's phyzguy's statement about the BB predictions of abundances of hydrogen and helium. As with all of science history of course, there's an interesting story there too.

I need to emphasize again that no-one (apart from Anders and Johnston, apparently) really thinks there is a 6Li problem after re-analysis of the measurements in 2012 by Lind, Asplund, Collet and Melendez (Mem. S.A.It. Suppl. 22, 142). Martin Asplund the second author is the the guy who made the original 6Li measurements (APJ 644 2006), by the way. ETA: And he says that he was never as certain about the 6Li measurements as other people were, in the first place, FTR.

There are three approaches to solutions to the primordial lithium-7 problem:

1: We don't understand the mechanisms for 7Li depletion in metal poor halo stars. Perhaps there is more diffusion and turbulent mixing between the surface layers and burning layers than we previously thought. Scatter in the low metalicity end of abundances suggests there is some mechanism of depletion we don't fully understand. See for example, Pinsonneault et. al. 2002, Pizzone et. al. 2007. Lind 2009 is another good one. There are hundreds of papers on this.

2: We don't fully understand the nuclear reactions that go into BBN. While all significant reactions (we think) have been measured, large uncertainties exist for some reaction rates. Measurements at BBN energies are very challenging. Is there perhaps, some resonance that we're missing, or some mechanism that we've not investigated fully? See for example, Coc, A. 2011, Civitarese, 2013 & 2014, Broggini 2012, Chakraborty, 2011. Kirsebom 2011. Again, hundreds of papers on this possibility.

3: There is some new physics solution. These include Axion dark matter (Ereken 2012), SUSY (so many papers. Look at Pospelov 2010 or Fields 2011 for an overview) , late-time neutron injection (through any mechanism the authors can think of e.g. Vasquez 2012), variation of fundamental constants (Coc 2007), cosmic rays during BBN (Kang 2011)... it goes on.

But, it is important to note that these new physics mechanisms are not replacing BBN, but modifying it, and modifying it in a small way, otherwise you lose the nice agreement with other BBN elements. And options 1 and 2 are not yet ruled out, either, allowing for unmodified BBN to be true.

It is a bit of a stretch to say that because the 7Li abundances are 3-4 times larger in BBN than in metal poor halo stars, that they provide evidence against the Big Bang. In fact, I'd say it is flat out wrong.

 

[e.bar.goum]

I encourage anyone with an interest in the primordial lithium problem to read the excellent 2011 review article by Brian Fields, which happily, is also available on the arxiv.

http://www.annualreviews.org/doi/abs/10.1146/annurev-nucl-102010-130445
http://arxiv.org/abs/1203.3551

 

[Torbjorn_L]

So when TrickyDicky asked, "is the ruling out of Lithium as originated in the Big-bang a big blow or just a scratch to the theory?", I was thinking that when folks talk about predictions as evidence in support of the big bang, they should also just as readily acknowledge where the predictions were actually post-dictions, and where other predictions, like with antimatter and lithium, just haven't held up.

Ah, thanks Bob! Now I understand what your comment is intending. Well, I don't really know anything about primordial nucleosynthesis or the observation problems, but what I read in this thread is encouraging: 6Li is no longer an obvious problem (if it ever where), if I understand e.bar.goum's comments with its many good references which I haven't had time to read. (While your reference neatly supports the BBN prediction.)

The other stuff you mention seems plain - "flat out" in the parlance of this thread - wrong. (Wikipedia clearly states the post-diction timeline; there is no finalized antimatter prediction - they are still measuring cross-sections in the standard model for particles (and see e.bar.goum's comment on this) - except that the observed asymmetry not necessarily need new physics for its prediction).

 

[Bob Enyart]

If you expect to measure 0 (as deviation from a theory from example) and 99 measurements are compatible with 0 while one gives a significant deviation, it should be obvious that this one gets the most attention.

Hello mfb! Yes, that's true. What's also true, I'm sure everyone here would agree, is the importance of having an objective assessment of how often the bb predictions of primordial abundances have actually matched actual observations, and how often, like with the CMB, we are seeing post-dictions and after-the-fact adjustments. An open letter was published in New Scientist, signed by dozens (and since by hundreds) of researchers working at prestigious institutions, including the Max-Planck Institute for Astrophysics, Sheffield University, George Mason University, Jet Propulsion Laboratory at CalTech, Cambridge University, Lawrenceville Plasma Physics, Penn State, Cal State Fullerton, University of Virginia, European Southern Observatory, and scores of others, affirming the opinion published in the journal Nature, that:

It is commonly supposed that the so-called primordial abundances of D (Deuterium, i.e., heavy hydrogen, N+P), 3He (Helium N+2P), and 4He (2N+2P) and 7Li (Lithium 3P+4N) provide strong evidence for Big Bang cosmology. But a particular value for the baryon-to-photon ratio needs to be assumed ad hoc to obtain the required [predicted] abundances.
- 1990 Nature http://www.nature.com/nature/journal/v346/n6287/pdf/346807a0.pdf​

The many astrophysicists and similarly qualified scientists at scores of leading institutions claimed in that New Scientist open letter that:

the big bang theory can boast of no quantitative predictions that have subsequently been validated by observation. The successes claimed by the theory's supporters consist of its ability to retrospectively fit observations with a steadily increasing array of adjustable parameters...​

Of course, they may be wrong mfb. But they are many. And unlike much of the difficulties that cosmology faces, these scientists are making claims that are verifiable, or falsifiable, via something as relatively simplistic as history. The more I look into actual bb predictions, including that there should be no "mature" galaxies among those that that are most distant, but per Science, observation has shown otherwise; and that gravity should have enabled less galaxy clustering far away, and more clustering nearby, yet Hubble keeps finding clusters that defy that patter, and not to delve into the unexpected anisotropy of the Axis of Evil and the apparent quantized redshift from hundreds of thousands of galaxies, much of the cosmos doesn't seem to fit neatly into the predictions of the theory, which might be why even a group like the National Academy of Sciences would publish as recently as in 2003 an alternative cosmological model for a bounded universe.

 

[Drakkith]

I'm not sure why adjusting a theory to match observations is looked down on, especially in the case of cosmology, which requires knowledge of macroscopic physics, microscopic physics, the conditions of the very early universe, and the ability to compute massive amounts of data representing both short term and long term changes. It seems obvious to me that the chances of coming up with a near-perfect theory that requires no adjusting is effectively zero.

 

[Bob Enyart]

Drakkith, hi! It seems obvious to me that you are EXACTLY CORRECT:

I'm not sure why adjusting a theory to match observations is looked down on, especially in the case of cosmology, which requires knowledge of macroscopic physics, microscopic physics, the conditions of the very early universe, and the ability to compute massive amounts of data representing both short term and long term changes. It seems obvious to me that the chances of coming up with a near-perfect theory that requires no adjusting is effectively zero.

Adjusting a theory to match observations is properly done all the time. And equally obvious, we should not be revisionists and claim that a theory predicted something when it did not, because then our revisionism itself becomes misinterpreted as actual evidence for the validity of the theory.

The long-predicted shadow of the big bang is a handy example. A 2006 paper in the Astrophysical Journal reported on a "vital test of the present cosmological paradigm" that "taken at face value, one may even hold the opinion that there is in fact no strong evidence" for the CMB's shadow from behind 31 nearby galaxy clusters. About this, Hunstville's Univ of Alabama professor of physics Richard Lieu said in Science Daily, that, "These shadows are a well-known thing that has been predicted for years. If you see a shadow… it means the radiation comes from behind the cluster. If you don’t see a shadow, then you have something of a problem", and so, "Either... the Big Bang is blown away or ... there is something else going on."

Of course, something else may be going on. Regardless though, regarding evidence, we should let the chips fall where they may. Once the wheel stops, so to speak, nobody gets to move their chips. So based on the current state of our knowledge,
- if major predictions appear to be simply wrong, like with a 50% antimatter universe, or
- if they weren't predictions but post-dictions, like with the CMB, or
- if they require secondary adjustments after-the-fact, like with initial abundances, or
- if they require significant secondary assumptions, like with lithium, or
- if they even require the supposition of multiple hypothetical entities,
then rather than overselling one's claims, a more frank assessment might not present the theory's predictive powers as reason to squelch challenges.

 

[Drakkith]

Adjusting a theory to match observations is properly done all the time. And equally obvious, we should not be revisionists and claim that a theory predicted something when it did not, because then our revisionism itself becomes misinterpreted as actual evidence for the validity of the theory.

I think there may be some confusion with the word "prediction" here. If I input a set of conditions into my model and get a result, then that model "predicts" the results. Similarly, the standard model of cosmology (based on the big bang theory) predicts certain things like the CMB, nucleosynthesis, etc. Changing the parameters of the model alters the prediction.

From wiki: http://en.wikipedia.org/wiki/Prediction#Prediction_in_science

In science, a prediction is a rigorous, often quantitative, statement, forecasting what will happen under specific conditions; for example, if an apple falls from a tree it will be attracted towards the center of the Earth by gravity with a specified and constant acceleration.

So a "prediction" does not necessarily mean that we predicted something prior to discovery, though that may happen.

 

[e.bar.goum]

It is commonly supposed that the so-called primordial abundances of D (Deuterium, i.e., heavy hydrogen, N+P), 3He (Helium N+2P), and 4He (2N+2P) and 7Li (Lithium 3P+4N) provide strong evidence for Big Bang cosmology. But a particular value for the baryon-to-photon ratio needs to be assumed ad hoc to obtain the required [predicted] abundances.
- 1990 Nature http://www.nature.com/nature/journal/v346/n6287/pdf/346807a0.pdf​

Yeah, in 1990. These days, in the era of precision cosmology (since about 2003), we know very well what the baryon-to-photon ratio is, independent of BBN abundances. Thank you, WMAP and Planck!

So now, the rather than BBN abundances being used as a probe for the baryon to photon ration, the baryon to photon ratio is used as a probe for BBN abundances.

 

[Bob Enyart]

Hi Drakkith! Again I agree with what you've posted:

I think there may be some confusion with the word "prediction" here. If I input a set of conditions into my model and get a result, then that model "predicts" the results. Similarly, the standard model of cosmology (based on the big bang theory) predicts certain things like the CMB, nucleosynthesis, etc. Changing the parameters of the model alters the prediction.

Meanwhile though, we shouldn't blur the difference between true predictions, and post-dictions. I wrote software for the Army's AH64 Apache Helicopter simulators and so I know a bit about creating models. It's comparatively easy to create a climate model that "predicts" the previous decade's average annual temperature for the entire globe. It's much harder to build a model that predicts the next decade's average annual temperatures. That's one reason why scientists place more value on predictions than they do on post-dictions.

 

[Drakkith]

So based on the current state of our knowledge, if major predictions appear to be simply wrong, like with a 50% antimatter universe, or if they weren't predictions but post-dictions, like with the CMB, or if they require secondary adjustments after-the-fact, like with initial abundances, or if they require significant secondary assumptions, then rather than overselling the predictive power of a theory, a more frank assessment might not present the theory's predictive powers as reason to squelch challenges.

Forgot to add this to my previous post:

It's important to remember that just because a specific part of a model doesn't match predictions it doesn't mean that the entire model is inaccurate. The imbalance of matter and anti-matter is a perfect example. We don't know why there was an imbalance, but we do know that if we place into our model the fact that there is an imbalance, we get predictions which are in excellent agreement with our overall observations. When part of a model doesn't work, but the rest of it does, it's usually a good sign that our understanding of how that specific part of the model works is incomplete, not the whole thing.

 

[Torbjorn_L]

Meanwhile though, we shouldn't blur the difference between true predictions, and post-dictions. I wrote software for the Army's AH64 Apache Helicopter simulators and so I know a bit about creating models. It's comparatively easy to create a climate model that "predicts" the previous decade's average annual temperature for the entire globe. It's much harder to build a model that predicts the next decade's average annual temperatures. That's one reason why scientists place more value on predictions than they do on post-dictions.

This is a distinction without a difference. If you use hypothesis testing, the constraints goes into the testing and everything is a prediction (hypothesis) test.

What you are alluding to is instead that prediction on processes is depending on the amount of and quality of data. It is easier to predict a short period with known constraints with an adjustable parameter model, since you have less data (smaller degree of freedom) and the process doesn't deviate as much due to noise or change.

I gather the "prediction/postdiction testing" sense used here derives from the same type of science denialism as the creationist attempt to distinguish between "experimental sciences" and "historical sciences" when they try to make out biology as not working. But I may easily be mistaken.

 

[Bob Enyart]

Torbjorn_L, I think that Drakkith has stated the matter so clearly that for this thread I'm content to let his be the last word:

The imbalance of matter and anti-matter is a perfect example. We don't know why there was an imbalance, but we do know that if we place into our model the fact that there is an imbalance, we get predictions which are in excellent agreement with our overall observations.

 

[D H]

So TD, if lithium-6 is a scratch, antimatter must be another scratch. :) Now I'm wondering about phyzguy's statement about hydrogen and helium.

Bob Enyart, I'm going to be blunt with you. You kindly gave us your name. I can google that name, and the results of that search aren't pretty. Your prior history with regard to your anti-scientific views strongly suggests you are trolling.

In the off chance that you are here to learn, here's a link to a partial list of unsolved problems in physics: http://en.wikipedia.org/wiki/List_of_unsolved_problems_in_physics. This is a big list, and it is not complete. Note that the issues you've sited as falsifying physics are mentioned in that list. These unsolved problems do not falsify physics.

The apparent current shortcomings of science is not how science is judged. You would not have been able to write your posts if science was fundamentally wrong. Science is judged on the basis of its successes rather than its current shortcomings. There's a tension here: As science progresses, scientists tend to find evermore (but smaller and smaller) gaps they cannot currently explain. The problem with looking at those gaps as disproof of science is that scientists have an uncanny ability of filling in those gaps.

 

[CKH]

Yeah, in 1990. These days, in the era of precision cosmology (since about 2003), we know very well what the baryon-to-photon ratio is, independent of BBN abundances. Thank you, WMAP and Planck!

So now, the rather than BBN abundances being used as a probe for the baryon to photon ration, the baryon to photon ratio is used as a probe for BBN abundances.

This is a very important point in evaluating the success of the theory. That is, what conclusions can be "cross-verified" by truly independent measurements?

Present day cosmologist proclaim an "era of precision cosmology" and claim that overall the BBT is essentially proven. The theory is only in need of tuning as problems arise. But as a layman it is difficult to evaluate these claims. I'd love to see a detailed explanation of the evidence that supports these views (for example the above evidence). Usually all I see are statements about the "pillars of BBT", redshift, nucleosynthesis and the CMB. However, I know that some six adjustable parameters are involved in some BBT fits to observations. Through adjustments, we can get the H and He isotopes to match the predictions of nucleosynthesis, but not Li.

It would be useful to have some sort of chart of the web of evidence that includes the assumptions, the successes and failed predictions as well, like the charts detectives draw in movies.

There are problems that are claimed to be solved by "new" physics such as DM, DE and (to some extent) inflation. It is even suggested that the Li problem may be a sign of new physics. DM in particular has a number of problems explaining the local universe. It is claimed that these will be explained by baryonic processes or some variation on the properties of DM. If you claim new physics to support a model, it is hard to assign a legitimacy to the model.

As already pointed out, before the deep studies of galaxies in the last decade, no one expected to find "mature" galaxies as early as we have. Has that problem somehow disappeared? Also, it is beginning to look like SMBH's developed early (because they are found at high z) excluding sufficient time for them to evolve through accretion of matter in galaxies. We have however no such prediction from theory.

Is there any evidence of evolution of metallicity in BH's or in galaxies over time? On the other hand, strong correlations across time have been found between other parameters such as mass and metallicity.

If you look only at the successes of nucleosynthesis and the CMB, maybe there is strong cross evidence of the BBT at early times, but when you look at galaxies it is not so clear that this theory of the early universe is helpful. Several predictions have been wrong.

All observations of distant galaxies are interpreted through the GR expansion model which strongly affects the measurements of distance, luminosity and time. Since observations are interpreted via this established theory, it is hard to know what to think about conclusions drawn from observations. For example, observations of distant galaxies at around z=2 (interpreted in this manner) tell us that galaxies were about 1/6th the size that they are now but had the same mass. How can that be? At high z, extreme star formation rates are measured such as 3,000/y. Can we explain that? Wouldn't such galaxies be so loaded with SN dust as to be nearly undetectable in the UV?

Apparently the cosmological theory demands mergers to account for large galaxies, but this looks shaky. If large galaxies form through mergers as claimed, how can you explain that 70% are nice disk galaxies instead of blobs? How do you explain that some large disk galaxies don't even have a bulge?

It is my impression that while BBT has some amazing successes in interpreting the CMB, I'm not so sure that it tells us much about the rest of the universe. It leaves us with many puzzles. At the same time it remains THE accepted theory and efforts appear focused on explaining away "tensions" with the theory in the belief that the theory is ultimately the correct explanation of the evolution of the universe.

I think it's fair to say that interpretation of the universe has been driven by cosmological theory. That is, we try to explain the local universe not directly from observations and known physics, but instead with a systems of beliefs derived from the study of the CMB and abundance of elements.

The above may be a distorted picture of the actual situation. It is just my lay impression from reading published papers.

 

[Bob Enyart]

D H, that's a worthwhile list, and yes, unanswered questions do *nothing* to undermine the validity of the scientific endeavor, of course. And I agree, the list isn't complete. I'll paste below a major issue that deserves to be on the list...

...here's a link to a partial list of unsolved problems in physics: http://en.wikipedia.org/wiki/List_of_unsolved_problems_in_physics. This is a big list, and it is not complete.

The National Academy of Sciences issued a report titled, Eleven Science Questions for the New Century which includes Question #10, “How were the heavy elements from iron to uranium made?” That should be on the list also.

Stellar nucleosynthesis can produce carbon, oxygen, nitrogen, etc., but the heaviest element that this fusion can construct is iron (common knowledge, as also described in Science News). The reason that the NAS has posed their Question 10 is that the long-accepted answer is now increasingly rejected: that zinc, silver, gold, etc., were said to be formed by supernovas. For "simulations show that these explosions have an insufficient quantity of neutrons" (SN, p. 16-17) and probably the nail in the coffin for that previous explanation is, as described in Nature, that when we actually look at a supernova explosion, we see "no spectroscopic evidence that r-process [heavy] elements have truly been produced" (http://www.nature.com/nature/journal/v500/n7464/full/500535a.html; see also Physics Reports, 2007; Astronomy & Astrophysics, 2011; and Physical Review Letters, 2013). So I'd add the origin of the heavy elements to Wikipedia's list of unanswered physics questions. (Although if I add it, other WP editors who dislike me, or my past, will almost certainly remove it; perhaps you D H or someone else here at PF could add it.)

Today astrophysicists are working to show that neutron star mergers, instead of supernovae, may be the source of our heavy elements, and they're trying to explain why we don't see what was initially expected by the standard model of chemical evolution, that is, the homogenous distribution of isotopes in the Sun, the Moon, Earth, and even in our crust and mantel (where, for example, as physicist Lawrence Krauss agreed when I asked him recently, that 90% of Earth's radioactive elements, like uranium, appear in the *continental* crust [and preferentially, of all things, near granite]).

These unsolved problems do not falsify physics. The apparent current shortcomings of science is not how science is judged. You would not have been able to write your posts if science was fundamentally wrong. Science is judged on the basis of its successes rather than its current shortcomings.

I agree with you completely D H. Perhaps you think that I would disagree with you because you are equating your philosophical beliefs with the words "physics" and "science". Physics and science are too cool and right on. People with enormously different philosophical beliefs all benefit from such progress. And what is worth noting is that when big changes occur in origins science, not a single one of our inventions stop working (cell phones, MRIs, etc.). Thus our technological society and amazingly sophisticated applied sciences are not dependent upon current speculation on abiogenesis, the nebular hypothesis, our cosmology, etc.

 

[Bob Enyart]

Guys, something I think we've glossed over in this whole thread is that the paper we've been talking about deals with actually a second lithium issue. The more traditional cosmological lithium problem, I think, is not this current isotope ratio problem, but simply a lithium problem.

An update on this first problem appeared this year in a paper in the Monthly Notices of the Royal Astronomical Society. Regardless of isotopes, the amount of observed lithium where theory attributes it to the big bang itself is apparently inconsistent with big bang nucelosynthesis (BBN). Earlier, a secondary assumption was that the inconsistency was possibly a "local problem", perhaps only manifesting itself in our own or similar galaxies. So the authors asked:

...is the Li problem a local problem, limited to our Galaxy, or is it independent of the environment? The analysis of the RGB stars in M54 confirms the findings in ω Centauri (Monaco et al. 2010), considered as the remnant of an accreted dwarf galaxy: the Li problem seems to be an universal problem, regardless of the parent galaxy.​

Thus Mucciarelli, et al., conclude:

Our result shows that this discrepancy is a universal problem concerning both the Milky Way and extra-galactic systems. Either modifications of BBN calculations, or a combination of atomic diffusion plus a suitably tuned additional mixing during the main sequence, need to be invoked to solve the discrepancy. MNRAS, 2014​

So it appears that BB predictions have led to two different lithium issues.

 

[Drakkith]

This thread is in serious danger of being locked for wandering off-topic. Let me remind everyone that PF supports the mainstream view of big bang theory and the standard cosmological model. PF exists to teach people about current, mainstream science, and arguments that any theory is wrong are usually not allowed. Claims that specific parts of the BBT or the standard cosmological model are inaccurate need to have a good deal of reputable sources, should be presented in as clear and concise a manner as possible, make logical sense, and the posting member needs to have sufficient knowledge of the details of the BBT, the standard cosmological model, and preferably GR and/or other relevant theories as well, otherwise there is a very good chance that the thread will go nowhere and will get locked.

Unfounded claims, off-topic posts, posts without reputable sources, or other problematic posts WILL be deleted and/or the thread locked and may result in an infraction for the posting member.

Whatever anyone's personal beliefs about any parts of science are, remember that this isn't you and your buddies arguing at starbucks, but a private forum with a clear goal and mission.

 

[PeterDonis]

the long-accepted answer is now increasingly rejected: that zinc, silver, gold, etc., were said to be formed by supernovas.

It's worth noting that this "long-accepted answer" was always a best guess based on the limited evidence then available. The fact that more evidence is causing scientists to re-evaluate the guess is not a "red flag" that the BBT is wrong; if it turns out that neutron star mergers, rather than supernovas, are what formed the elements heavier than iron, that doesn't invalidate other parts of the BBT; it just becomes a better-tested part of the same overall theory.

 

[e.bar.goum]

Guys, something I think we've glossed over in this whole thread is that the paper we've been talking about deals with actually a second lithium issue. The more traditional cosmological lithium problem, I think, is not this current isotope ratio problem, but simply a lithium problem.

An update on this first problem appeared this year in a paper in the Monthly Notices of the Royal Astronomical Society. Regardless of isotopes, the amount of observed lithium where theory attributes it to the big bang itself is apparently inconsistent with big bang nucelosynthesis (BBN). Earlier, a secondary assumption was that the inconsistency was possibly a "local problem", perhaps only manifesting itself in our own or similar galaxies. So the authors asked:

...is the Li problem a local problem, limited to our Galaxy, or is it independent of the environment? The analysis of the RGB stars in M54 confirms the findings in ω Centauri (Monaco et al. 2010), considered as the remnant of an accreted dwarf galaxy: the Li problem seems to be an universal problem, regardless of the parent galaxy.​

Thus Mucciarelli, et al., conclude:

Our result shows that this discrepancy is a universal problem concerning both the Milky Way and extra-galactic systems. Either modifications of BBN calculations, or a combination of atomic diffusion plus a suitably tuned additional mixing during the main sequence, need to be invoked to solve the discrepancy. MNRAS, 2014​

So it appears that BB predictions have led to two different lithium issues.

There is only one lithium problem.

I went into incredible detail above, providing a heap of references to approaches to solving the 7Li problem. You will note that only one option involves "new physics", and none of them are anything like getting rid of BBN. Glossed over, indeed! I heavily suggest you read Brian Fields 2011 review paper, that I linked above.

https://dl.dropboxusercontent.com/u/34677838/lithiumproblem.bib , for everything I've got on the 7Li problem, beyond what I referred to above. I don't pretend it's complete, and it is skewed towards nuclear solutions, but it may help to enlighten.

 

[e.bar.goum]

It's worth noting that this "long-accepted answer" was always a best guess based on the limited evidence then available. The fact that more evidence is causing scientists to re-evaluate the guess is not a "red flag" that the BBT is wrong; if it turns out that neutron star mergers, rather than supernovas, are what formed the elements heavier than iron, that doesn't invalidate other parts of the BBT; it just becomes a better-tested part of the same overall theory.

Definitely. And it's really not a settled question yet - we're not sure that the rate of neutron star mergers is enough. But it's important to note that supernovas aren't really ruled out either. Supernova models are hard! We're only just being able to actually get them to explode reliably!

Open questions aren't bad, or signs that a theory is wrong, they're opportunities to help us refine our models.

ETA: And I'm afraid I've got little idea as to what supernova explosions have to do with the validity of BBN, except for the general theme of nucleosynthesis. Help?

 

[Torbjorn_L]

To you who knew so much about the current Li situation, thanks for a valuable thread! I learned a lot, obviously.

Now this:

I think that Drakkith has stated the matter so clearly that for this thread I'm content to let his be the last word:

QFT? Or is it an implication of that something I said wasn't correct? I can't tell.

But FWIW, and it is quite clearly straying from the topic of BBN, that constraint is going into the testing, which was successful. Some day it may itself be predicted (explained), but we aren't there yet.

As noted already:

Open questions aren't bad, or signs that a theory is wrong, they're opportunities to help us refine our models.

Even more, open questions is a sign of a productive area of inquiry. Science for the win, literary.