|
|
Thread Tools | Display Modes |
#111
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
In article , Oh No
writes: The cosmological parameters are pretty well tied down by supernova observations, and that should tie q=q(t) down too. Within the context of GR cosmology, knowing q_0 and Omega_0 (or lambda_0, or any two independent cosmological parameters except H_0) completely determines q(t). The only thing is, q is used in a series expansion, which we don't much use any more, and it True. At small redshifts, q is the most important term, so historically this was important. However, it is just Omega/2 - lambda, so is well defined in a much larger context. Practically no modern cosmological test "measures" q in the sense that the gradient of probability in cosmological parameter space is shallowest along lines of constant q, so this is another reason it is not used as often as it once was. However, it does tell one whether or not the universe is accelerating. makes not a lot of sense to use q(t) for any purpose I can think of. What is usually discussed is q0=q(t0), i.e. now. Right, though one might say something like "at a redshift of x, corresponding to a time y years ago, the universe was still decelerating". |
#112
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
Thus spake Phillip Helbig---remove CLOTHES to reply
LOTHESvax.de In article , Oh No writes: The cosmological parameters are pretty well tied down by supernova observations, and that should tie q=q(t) down too. Within the context of GR cosmology, knowing q_0 and Omega_0 (or lambda_0, or any two independent cosmological parameters except H_0) completely determines q(t). Right. I could have expressed myself more strongly. Sloppy of me. I only intended the looseness of expression to allow for error bounds. The only thing is, q is used in a series expansion, which we don't much use any more, and it True. At small redshifts, q is the most important term, so historically this was important. However, it is just Omega/2 - lambda, so is well defined in a much larger context. Practically no modern cosmological test "measures" q in the sense that the gradient of probability in cosmological parameter space is shallowest along lines of constant q, so this is another reason it is not used as often as it once was. However, it does tell one whether or not the universe is accelerating. makes not a lot of sense to use q(t) for any purpose I can think of. What is usually discussed is q0=q(t0), i.e. now. Right, though one might say something like "at a redshift of x, corresponding to a time y years ago, the universe was still decelerating". Yes. But that bears on something else which this thread has caused me to ponder. q is termed the acceleration parameter, or more properly the deceleration parameter. I think if someone said that I would think the were talking about acceleration not the acceleration parameter. I would take that to be adotdot. Mind you, I suppose its academic in this case. Neither a nor adot changes sign, so q and adotdot have the same sign. Regards -- Charles Francis substitute charles for NotI to email |
#113
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
Oh No wrote:
Thus spake "John (Liberty) Bell" Oh No wrote: Thus spake Joseph Lazio "J(B" =3D=3D John (Liberty) Bell = writes: JB John (Liberty) Bell wrote: I think I've posted previously a reference to a paper that claimed that Type II supernovae are difficult to detect beyond z ~ 0.5. The most distant Type Ia supernova is about 1.7, IIRC. Yes, that's right. SN1997ff is at z=3D1.755. In the Riess gold set the= re are only nine useful data points above z=3D1. Current surveys like the Supernova Legacy Survey aren't even looking above about z=3D1, because= of measurement problems and risk of statistical bias in the data. We have to wait for SNAP which should turn up hundreds, or even thousands of SN at red shifts up to 2. The reference (ApJ, 649, 563-569, 2006) provided by (http://groups.google.com/group/sci.a..._frm/thread/1a 777a781e67a3a2/#) indicates that there is no conclusive _observational_ evidence of q becoming positive at z0.5. That leads to a post by you, not by Rob. No. It leads to a thread started by me which contains various postings by Rob. If you provide a complete ref, with author and preferably arxiv, I will look at it. The ref. provided by Rob was ApJ, 649, 563-569, 2006. I found the paper by checking the contents of the most recent issues of the Journal. This paper is less than 2 months old. Authors: Charles Shapiro and Michael S. Turner . The ref. given by The Astrophysical Journal at the top of the paper is 649:563-569, 2006 =A9 2006Date: Oct. 1 2006 Although, actually The cosmological parameters are pretty well tied down by supernova observations, and that should tie q=3Dq(t) down too. The only thing is, q is used in a series expansion, which we don't much use any more, and it makes not a lot of sense to use q(t) for any purpose I can think of. What is usually discussed is q0=3Dq(t0), i.e. now. If you are aware of different information, please clarify, with appropriate references. No, that is where I think we do have a problem. It's early days, and a= ll the data is too preliminary to be sure, but galaxy evolution models do not tie in well with the age of the universe, when one looks at galaxi= es at z=3D6 and above. There are even large galaxies at z=3D10, though the image we get of them is a bit inconclusive in terms of saying what kind of stars they contain. A reference for this too, would also be appreciated # I already gave you references in response to the above mentioned post. I would refer you to two review papers in Natu Glazebrook K. et. al., 2004, Nature, 430, 181-184. http://www.pha.jhu.edu/~kgb/MiscPub/...iii-nature.pdf Cimatti. et. al., 2004, Old Galaxies in the Young Universe, Nature, 430, 184-188. astro-ph/0407131 I had already read the first paper and the abstract of the second. Neither support your assertion of large galaxies at z=3D10. The first paper covers z =3D~ 0.5 to 2. The second paper describes 4 galaxies at 1.6z1.9 I'll let you search arxiv yourself for galaxies around z=3D10. I don't think there are many found so far. I am only aware of 1, which was subsequently discredited. That is why I asked you for your references. John Bell |
#114
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
John (Liberty) Bell wrote:
Oh No wrote: Thus spake "John (Liberty) Bell" Oh No wrote: Thus spake Joseph Lazio "J(B" =3D=3D John (Liberty) Bell = writes: JB John (Liberty) Bell wrote: I think I've posted previously a reference to a paper that claimed that Type II supernovae are difficult to detect beyond z ~ 0.5. The most distant Type Ia supernova is about 1.7, IIRC. Yes, that's right. SN1997ff is at z=3D1.755. In the Riess gold set the= re are only nine useful data points above z=3D1. Current surveys like the Supernova Legacy Survey aren't even looking above about z=3D1, because= of measurement problems and risk of statistical bias in the data. We have to wait for SNAP which should turn up hundreds, or even thousands of SN at red shifts up to 2. The reference (ApJ, 649, 563-569, 2006) provided by (http://groups.google.com/group/sci.a..._frm/thread/1a 777a781e67a3a2/#) indicates that there is no conclusive _observational_ evidence of q becoming positive at z0.5. That leads to a post by you, not by Rob. No. It leads to a thread started by me which contains various postings by Rob. If you provide a complete ref, with author and preferably arxiv, I will look at it. The ref. provided by Rob was ApJ, 649, 563-569, 2006. I found the paper by checking the contents of the most recent issues of the Journal. This paper is less than 2 months old. Authors: Charles Shapiro and Michael S. Turner . The ref. given by The Astrophysical Journal at the top of the paper is 649:563-569, 2006 =A9 2006Date: Oct. 1 2006 Although, actually The cosmological parameters are pretty well tied down by supernova observations, and that should tie q=3Dq(t) down too. The only thing is, q is used in a series expansion, which we don't much use any more, and it makes not a lot of sense to use q(t) for any purpose I can think of. What is usually discussed is q0=3Dq(t0), i.e. now. If you are aware of different information, please clarify, with appropriate references. No, that is where I think we do have a problem. It's early days, and a= ll the data is too preliminary to be sure, but galaxy evolution models do not tie in well with the age of the universe, when one looks at galaxi= es at z=3D6 and above. There are even large galaxies at z=3D10, though the image we get of them is a bit inconclusive in terms of saying what kind of stars they contain. A reference for this too, would also be appreciated # I already gave you references in response to the above mentioned post. I would refer you to two review papers in Natu Glazebrook K. et. al., 2004, Nature, 430, 181-184. http://www.pha.jhu.edu/~kgb/MiscPub/...iii-nature.pdf Cimatti. et. al., 2004, Old Galaxies in the Young Universe, Nature, 430, 184-188. astro-ph/0407131 I had already read the first paper and the abstract of the second. Neither support your assertion of large galaxies at z=3D10. The first paper covers z =3D~ 0.5 to 2. The second paper describes 4 galaxies at 1.6z1.9 I'll let you search arxiv yourself for galaxies around z=3D10. I don't think there are many found so far. I am only aware of 1, which was subsequently discredited. That is why I asked you for your references. Technical addendum. In the above posting of Wed, Nov 22 2006 4:27 pm, the Astrophysical Journal copyright symbol was corrupted to =A9. This is understandable since it is an 8 bit character code, not 7 bit ascii. However, each instance of =, including those quoted, was also corrupted, this time to =3D. The reason for this is less obvious. (I found in a later posting that this second corruption can be supressed by typing = over every displayed =, before posting the response.) John Bell [Mod. note: MIME encoding (http://en.wikipedia.org/wiki/Mime) has this effect. It can be avoided by posting in plain ASCII only, which is why I continue to encourage posters to be careful to do so -- mjh] |
#115
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
Thus spake "John (Liberty) Bell"
Oh No wrote: Thus spake "John (Liberty) Bell" Oh No wrote: Thus spake Joseph Lazio JB John (Liberty) Bell wrote: I think I've posted previously a reference to a paper that claimed that Type II supernovae are difficult to detect beyond z ~ 0.5. The most distant Type Ia supernova is about 1.7, IIRC. Yes, that's right. SN1997ff is at z=3D1.755. In the Riess gold set the= re are only nine useful data points above z=3D1. Current surveys like the Supernova Legacy Survey aren't even looking above about z=3D1, because= of measurement problems and risk of statistical bias in the data. We have to wait for SNAP which should turn up hundreds, or even thousands of SN at red shifts up to 2. The reference (ApJ, 649, 563-569, 2006) provided by (http://groups.google.com/group/sci.a..._frm/thread/1a 777a781e67a3a2/#) indicates that there is no conclusive _observational_ evidence of q becoming positive at z0.5. The ref. provided by Rob was ApJ, 649, 563-569, 2006. I found the paper by checking the contents of the most recent issues of the Journal. This paper is less than 2 months old. Authors: Charles Shapiro and Michael S. Turner . The ref. given by The Astrophysical Journal at the top of the paper is 649:563-569, 2006 =A9 2006Date: Oct. 1 2006 This paper takes the non-standard step of dispensing with the Friedmann equation. Not something I would be inclined to do, since the Friedmann equation is itself based on well established physics. The most I would consider reasonable is to allow the cosmological constant Lambda to vary in time. This is done in a somewhat arcane group of theories often called "quintescence". Of course if one does take a step like this then one is no longer working within the context of a mainstream GR cosmology, and obviously that opens up the possibility that q is not determined at all times by the current data. However, there is an immediate problem that, in this case, the amount of data is so little that it is very difficult to come to any physics conclusions at all. Mostly these theories are being tested by looking for supernova at z1, because that is where we are best able to collect data with which to test them. Other projects are underway, but I believe they are mostly in early design and build stage. As I understand, at the moment there is no evidence for varying Lambda. If you are talking about such theories outside of standard cosmology, you should make your context plain. Although the paper says that there is only week *observational* evidence of past deceleration, such deceleration is a feature of a standard GR cosmology, and it concludes that "none of the kinematic models studied here have revealed robust features about cosmic acceleration that differ from LambdaCDM or wCDM. Moreover, none of the kinematic models fit the data significantly better". I'll let you search arxiv yourself for galaxies around z=3D10. I don't think there are many found so far. I am only aware of 1, which was subsequently discredited. That is why I asked you for your references. I had said that the data was too thin to use. It does not altogether surprise me if the observation has been discredited. I gather the moderator gave references earlier in the thread. Regards -- Charles Francis substitute charles for NotI to email |
#116
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
Oh No wrote:
Thus spake "John (Liberty) Bell" Oh No wrote: Thus spake "John (Liberty) Bell" Oh No wrote: Thus spake Joseph Lazio JB John (Liberty) Bell wrote: I think I've posted previously a reference to a paper that claimed that Type II supernovae are difficult to detect beyond z ~ 0.5. The most distant Type Ia supernova is about 1.7, IIRC. Yes, that's right. SN1997ff is at z=3D1.755. In the Riess gold set the= re are only nine useful data points above z=3D1. Current surveys like the Supernova Legacy Survey aren't even looking above about z=3D1, because= of measurement problems and risk of statistical bias in the data. We have to wait for SNAP which should turn up hundreds, or even thousands of SN at red shifts up to 2. The reference (ApJ, 649, 563-569, 2006) provided by (http://groups.google.com/group/sci.a..._frm/thread/1a 777a781e67a3a2/#) indicates that there is no conclusive _observational_ evidence of q becoming positive at z0.5. The ref. provided by Rob was ApJ, 649, 563-569, 2006. I found the paper by checking the contents of the most recent issues of the Journal. This paper is less than 2 months old. Authors: Charles Shapiro and Michael S. Turner . The ref. given by The Astrophysical Journal at the top of the paper is 649:563-569, 2006 =A9 2006Date: Oct. 1 2006 This paper takes the non-standard step of dispensing with the Friedmann equation. This is not strictly true. The paper takes the "non standard" step of examining the astronomical data objectively, for what it is. The authors DO examine the data in the context of the best fit EFE model. However, they also examine the data in the context of radically different models too, which puts the evidence into perspective. In particular, I draw your attention to Figure 1, and my discussion with John Bell under the subtitle "Better News for Big bang Theory". You will note from the text of the paper, that the displayed step function gives as good a match to the data, as does EFE (also displayed in that figure). You will also note from my discussion with John, that when I tanslated his predicted cosmic acceleration into the more commonly used acceleration parameter q, this gave rise to a predicted asymptotically decreasing q function as z got progressively lower. Thirdly, please note, from figure 1, that a best fit asymptotic function does appear to represent something mid way between the best fit step function, and the best fit EFE function at low z. Not something I would be inclined to do, That is clear from your own approach to the unification of QM and GR. since the Friedmann equation is itself based on well established physics. By the same token, 90 years ago, you would not have been inclined to examine the orbital anomaly of Mercury in the context of EFE, since Newtonian physics was then the well established physics, whereas EFE was not. This is not a personal attack, since the vast majority of physicists at that time, were of the same opinion. There was one notable exception, however, who mounted an expedition to test Einstein's predicted doubling of the Newtonian bending of starlight grazing the Sun, during a solar eclipse. The rest is history (and that is what got EFE recognised). The most I would consider reasonable is to allow the cosmological constant Lambda to vary in time. And how is that reasonable? As Phillip Helbig pointed out, a constant is, by definition, constant. There is no way that such a varying lambda can be predicted from Einstein's foundations of the theory. This is done in a somewhat arcane group of theories often called "quintescence". Of course if one does take a step like this then one is no longer working within the context of a mainstream GR cosmology, and obviously that opens up the possibility that q is not determined at all times by the current data. However, there is an immediate problem that, in this case, the amount of data is so little that it is very difficult to come to any physics conclusions at all. Mostly these theories are being tested by looking for supernova at z1, because that is where we are best able to collect data with which to test them. Other projects are underway, but I believe they are mostly in early design and build stage. As I understand, at the moment there is no evidence for varying Lambda. If you are talking about such theories outside of standard cosmology, you should make your context plain. No, neither I nor John have been looking at such 'arcane' (one might even say revisionist) approaches to cosmology and relativity theory. IMO they lack elegance, and are more reminiscent, in their historical context, to adding further epicycles to the Ptolemaic model of the Universe, before Copernicus, Galileo, Brahe, Kepler, and Newton came along. John has already explained to you, first at sci.physics.research (over 6 months ago), and then again here, that his job is to examine the consequences of a newer (as yet unpublished) solution of relativistic field theory, in the context of compatibility with both modern astronomical evidence and known (proven) physics. You still seem unwilling to accept that. Although the paper says that there is only week *observational* evidence of past deceleration, such deceleration is a feature of a standard GR cosmology, Yes. I only computed the situation at low z, since the situation at high z is far less clear cut. It is, however, the situation at low z that we can examine most fruitfully at present, because this is the area where the divergence between these models is greatest. You can confirm that, simply by looking at Figure 1 of the paper. and it concludes that "none of the kinematic models studied here have revealed robust features about cosmic acceleration that differ from LambdaCDM or wCDM. Moreover, none of the kinematic models fit the data significantly better". Agreed, but they did not examine the predicted asymptotic function. They did say, however, that the step function was as good a fit as EFE (which you should find astonishing, given your preconceptions). And our predicted asymptotic function does seem to lie between those two extremes over the vast majority of the range where statistically meaningful astronomical data is available. Such a new statistical examination looks to me like it could well make an excellent starting point for a further groundbreaking astronomical paper. I sincerely hope that some bright young (academic) astrophysicist has enough sense to try it out, and then publish his (or her) findings. Chalky. |
#117
|
|||
|
|||
Galactic Evolution (was: Still lower noise radio astronomy )
Thus spake Chalky
Oh No wrote: The ref. provided by Rob was ApJ, 649, 563-569, 2006. I found the paper by checking the contents of the most recent issues of the Journal. This paper is less than 2 months old. Authors: Charles Shapiro and Michael S. Turner . The ref. given by The Astrophysical Journal at the top of the paper is 649:563-569, 2006 =A9 2006Date: Oct. 1 2006 This paper takes the non-standard step of dispensing with the Friedmann equation. This is not strictly true. No, that is strictly true and is clearly stated in the paper. The paper takes the "non standard" step of examining the astronomical data objectively, for what it is. If one were really to do that the data would be just a bunch of numbers, and nothing could be said of it at all. There is no such thing as science without assumptions, for example a basic assumption that reality exists independent of our perceptions, or the assumption that it obeys laws which are always and everywhere the same as encapsulated in the general principle of relativity. It is trivially obvious that if you drop that assumption you will be able to say very little about any data from distant astronomical objects, so to start making a song and dance about it, and to suggest that the analysis given under general relativity is in some way unscientific because of it, is not only thoroughly offensive, but plain ignorant because actually there is a good deal of scientific work, both theoretical and observational, which does look into well defined ways in which the assumption can be varied. The authors DO examine the data in the context of the best fit EFE model. However, they also examine the data in the context of radically different models too, which puts the evidence into perspective. A better perspective would be to understand the structure of General Relativity based on differential geometry, to understand the EFE, and the derivation of Friedmann's equation and to know just what assumptions are involved, and in what manner it is reasonable to vary them. Not something I would be inclined to do, That is clear from your own approach to the unification of QM and GR. Indeed I have varied a much more subtle and less robust assumption in GR, while retaining the EFE and Friedmann's equation, and I have found a statistically better fit to the data than standard for a model with no cosmological constant at all, let alone a variable one. since the Friedmann equation is itself based on well established physics. By the same token, 90 years ago, you would not have been inclined to examine the orbital anomaly of Mercury in the context of EFE, since Newtonian physics was then the well established physics, whereas EFE was not. I really think you need to understand the physics involved before you make that kind of judgement. Newton's law of gravity was always an unexplained phenomenon, as was his *assumption* of an absolute background space. These things were well understood by the great mathematical physicists, Leibniz, Gauss, Riemann, and it was on their painstaking and inspirational work and thought over two hundred years that Einstein based both the special and general theory of relativity. This is not a personal attack, since the vast majority of physicists In matters of the theoretical advance of science, such as those due to Newton and Einstein, the vast majority of physicists have very little to say. They come along afterwards and check the detail through experiment. But to understand the insights involved in generating the theory, it would be wiser to look only at what was said by great mathematicians. The most I would consider reasonable is to allow the cosmological constant Lambda to vary in time. And how is that reasonable? As Phillip Helbig pointed out, a constant is, by definition, constant. That is a matter of semantics. There is no way that such a varying lambda can be predicted from Einstein's foundations of the theory. There is also no way that constant lambda can be predicted from Einstein's foundations. There is not even any way that any lambda at all can be predicted from the foundations of the theory. Nor, for that matter, is there any other theory, in particle physics for example, that can predict lambda or a value for it. This is done in a somewhat arcane group of theories often called "quintescence". Of course if one does take a step like this then one is no longer working within the context of a mainstream GR cosmology, and obviously that opens up the possibility that q is not determined at all times by the current data. However, there is an immediate problem that, in this case, the amount of data is so little that it is very difficult to come to any physics conclusions at all. Mostly these theories are being tested by looking for supernova at z1, because that is where we are best able to collect data with which to test them. Other projects are underway, but I believe they are mostly in early design and build stage. As I understand, at the moment there is no evidence for varying Lambda. If you are talking about such theories outside of standard cosmology, you should make your context plain. No, neither I nor John have been looking at such 'arcane' (one might even say revisionist) approaches to cosmology and relativity theory. IMO they lack elegance, and are more reminiscent, in their historical context, to adding further epicycles to the Ptolemaic model of the Universe, before Copernicus, Galileo, Brahe, Kepler, and Newton came along. I am inclined to agree with that. Not just Lambda, but CDM, Inflation, the unresolved problems between quantum theory and general relativity, all point to the need for a radical shift in our accepted paradigms. John has already explained to you, first at sci.physics.research (over 6 months ago), and then again here, that his job is to examine the consequences of a newer (as yet unpublished) solution of relativistic field theory, in the context of compatibility with both modern astronomical evidence and known (proven) physics. You still seem unwilling to accept that. It is true that I am unwilling to accept that anyone who shows as little understanding of mathematics in general, or of general relativity in particular, can be equipped to do such a job. Also, as a great deal of work has gone into finding solutions of relativistic field theory, by mathematicians who are equipped for the job, I am sceptical that there really is such a thing as an unpublished solution of the sort you appear to be suggesting (not withstanding recent, genuine, published solutions such as those of Bekenstein and Moffat). If this is a solution, it should be publishable in its own right, even without data analysis. Otherwise anyone is likely to suspect that it is based on an errors, or makes assumptions outside of relativistic field theory which may well not make sense. Mathematically constructing a new solution of relativistic field theory is extremely difficult. It took Bekenstein years, and Moffat had the benefit of following his work. Even early solutions, Schwarzschild, Friedmann Cosmologies, were carried out by mathematicians of exceptional ability. But I have been explaining to you and John very basic stuff; the meaning of the scale factor, the meaning of curvature. The prospect that you could really be talking of a new unpublished solution appears to be practically zero. and it concludes that "none of the kinematic models studied here have revealed robust features about cosmic acceleration that differ from LambdaCDM or wCDM. Moreover, none of the kinematic models fit the data significantly better". Agreed, but they did not examine the predicted asymptotic function. They did say, however, that the step function was as good a fit as EFE (which you should find astonishing, given your preconceptions). Not at all. Given the quality of the data, I would expect all sorts of things to fit it once you drop Friedmann's equation. That is precisely why I find it unimpressive - the very opposite of astonishing. It is very easy to drop scientific assumptions and get arbitrary functions to fit data. Leibniz even furnished a proof that there are always an infinite number of ways to do that. What is much harder and far more subtle is to do what Einstein did, and replace scientific assumptions with, better, more rigorous, and more scientific, assumptions, and then to demonstrate a better fit with data. The ability to do that is what made him a genius. Don't expect me to believe that anyone can do it who can't even understand what Einstein said. Regards -- Charles Francis substitute charles for NotI to email |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[sci.astro] General (Astronomy Frequently Asked Questions) (2/9) | [email protected] | Astronomy Misc | 0 | May 3rd 06 12:33 PM |
[sci.astro,sci.astro.seti] Welcome! - read this first | [email protected] | Astronomy Misc | 9 | February 2nd 06 02:37 AM |
SCIENTIFIC PANEL TO REVIEW ED CONRAD'S BRAIN.... | Ed Conrad | Astronomy Misc | 0 | December 25th 05 03:24 PM |
CORRUPT THEORY 'WINS' ANOTHER ROUND -- Evolution vs. Intelligent Design | Ed Conrad | Astronomy Misc | 2 | December 21st 05 09:55 PM |
[sci.astro] General (Astronomy Frequently Asked Questions) (2/9) | [email protected] | Astronomy Misc | 0 | October 6th 05 02:35 AM |