Datasets:

---

Recag

/

Rp_CommonC_93

like 0

ERROR: type should be string, got "https://cosmocoffee.info/\nHubble Finds Evidence for Dark Energy in the Young Universe\nhttps://cosmocoffee.info/viewtopic.php?f=13&t=742\nPosted: November 17 2006\nby Geraint Lewis\nScientists using NASA's Hubble Space Telescope have discovered that dark energy is not a new constituent of space, but rather has been present for most of the universe's history. Dark energy is a mysterious repulsive force that causes the universe to expand at an increasing rate.\nInvestigators used Hubble to find that dark energy was already boosting the expansion rate of the universe as long as nine billion years ago. This picture of dark energy is consistent with Albert Einstein's prediction of nearly a century ago that a repulsive form of gravity emanates from empty space.\nData from Hubble provides supporting evidence that help astrophysicists to understand the nature of dark energy. This will allow scientists to begin ruling out some competing explanations that predict that the strength of dark energy changes over time.\nResearchers also have found that the class of ancient exploding stars, or supernovae, used to measure the expansion of space today look remarkably similar to those that exploded nine billion years ago and are just now being seen by Hubble. This important finding gives additional credibility to the use of these supernovae for tracking the cosmic expansion over most of the universe's lifetime.\n\"Although dark energy accounts for more than 70 percent of the energy of the universe, we know very little about it, so each clue is precious,\" said Adam Riess, of the Space Telescope Science Institute and Johns Hopkins University in Baltimore. Riess led one of the first studies to reveal the presence of dark energy in 1998 and is the leader of the current Hubble study. \"Our latest clue is that the stuff we call dark energy was relatively weak, but starting to make its presence felt nine billion years ago.\"\nTo study the behavior of dark energy of long ago, Hubble had to peer far across the universe and back into time to detect supernovae. Supernovae can be used to trace the universe's expansion. This is analogous to seeing fireflies on a summer night. Fireflies glow with about the same brightness, so you can judge how they are distributed in the backyard by their comparative faintness or brightness, depending on their distance from you. Only Hubble can measure these ancient supernovae because they are too distant, and therefore too faint, to be studied by the largest ground-based telescopes.\nEinstein first conceived of the notion of a repulsive force in space in his attempt to balance the universe against the inward pull of its own gravity, which he thought would ultimately cause the universe to implode.\nHis \"cosmological constant\" remained a curious hypothesis until 1998, when Riess and the members of the High-z Supernova Team and the Supernova Cosmology Project used ground-based telescopes and Hubble to detect the acceleration of the expansion of space from observations of distant supernovae. Astrophysicists came to the realization that Einstein may have been right after all: there really was a repulsive form of gravity in space that was soon after dubbed \"dark energy.\"\nOver the past eight years astrophysicists have been trying to uncover two of dark energy's most fundamental properties: its strength and its permanence. These new observations reveal that dark energy was present and obstructing the gravitational pull of the matter in the universe even before it began to win this cosmic \"tug of war.\"\nPrevious Hubble observations of the most distant supernovae known revealed that the early universe was dominated by matter whose gravity was slowing down the universe's expansion rate, like a ball rolling up a slight incline. The observations also confirmed that the expansion rate of the cosmos began speeding up about five to six billion years ago. That is when astronomers believe that dark energy's repulsive force overtook gravity's attractive grip.\nThe latest results are based on an analysis of the 24 most distant supernovae known, most found within the last two years.\nBy measuring the universe's relative size over time, astrophysicists have tracked the universe's growth spurts, much as a parent may witness the growth spurts of a child by tracking changes in height on a doorframe. Distant supernovae provide the doorframe markings read by Hubble. \"After we subtract the gravity from the known matter in the universe, we can see the dark energy pushing to get out,\" said Lou Strolger, astronomer and Hubble science team member at Western Kentucky University in Bowling Green, Ky. Further observations are presently underway with Hubble by Riess and his team which should continue to offer new clues to the nature of dark energy.\nhttp://hubblesite.org/newscenter/archiv ... 6/52/full/\nDoes anyone know the redshifts of these SNs?\nby Matthew Francis\nThere's a bit of info over at Cosmic Variance\nby Tommy Anderberg\nWhat I get out of the little info posted is that combining CMB data with O(10) new SNe (not clear whether 21 or only the 13 spectroscopically confirmed as Ia) at z > 1 shows no evidence of evolving equation of state, assuming that dark energy was not actually the dominating term at z > 1.8 (which probably means that's the highest z they have). By the way, an early hint: astro-ph/0601319, which singles out 1.5 < z < 3.0 as particularly interesting.\nBottom line: a plain cosmological constant still does the job as far as data fitting is concerned. Or, if you will, observation still lends no support to quintessence etc.\nAnyone wondering exactly how this news (?) warranted the press release posted above might do well to read Steinn Sigurdsson's Anatomy of a Press Release.\nOther interesting posts by Sigurdsson which may help shed light on the goings-on are:\nNASA: double down on Science\nBeyond Einstein Review - Important Info\nOrderly Withdrawal: NSF Astronomy Senior Review is Out\nAdvice to my friends\n>Bottom line: a plain cosmological constant still does the job as far as data fitting is concerned. Or, if you will, observation still lends no support to quintessence etc.\nWell - the error is large:\nw(z > 1) = -0.8+0.6-1.0, and w(z > 1) < 0 at 98% confidence.\nSo, it doesn't really rule out quntessence dark energy - just that there is an accelerating component at z>1\nAs for press releases - given the finicky way the press deal with science, the fact that any astronomy appears at all should be taken as a positive.\nby Eric Linder\nWe should not expect \"interesting\" improvements on cosmological parameter constraints from handfuls of even high redshift supernovae. An upper limit on w(z>1)<-0.1 (or -0.2 at lesser confidence) comes almost wholly from the distance to the CMB last scattering surface, not from current SN, assuming a smooth evolution of w(z), and was given two years ago. What is interesting, if much less sexy from a press release point of view, is what the data (when released) will show about systematics. An interesting data set and analysis, from the HST \"Decelerating and Dustfree\" program, coming out soon, has roughly the same number of new SN Ia at z>1, but observed in cluster elliptical galaxies, where the systematics uncertainties due to dust dimming are much less. This will provide an interesting crosscheck on the cosmology analysis, as opposed to simple statistical growth.\nHere's the answer - in >80 rolling pages!\nastro-ph/0611572 [abs, ps, pdf, other] :\nTitle: New Hubble Space Telescope Discoveries of Type Ia Supernovae at z > 1: Narrowing Constraints on the Early Behavior of Dark Energy\nAuthors: Adam G. Riess (JHU, STScI), Louis-Gregory Strolger (UWK), Stefano Casertano (STScI), Henry C. Ferguson (STScI), Bahram Mobasher (STScI), Ben Gold (JHU), Peter J. Challis (CfA), Alexei V. Filippenko (UCB), Saurabh Jha (UCB), Weidong Li (UCB), John Tonry (IfA), Ryan Foley (UCB), Robert P. Kirshner (CfA), Mark Dickinson (NOAO), Emily MacDonald (NOAO), Daniel Eisenstein (UofA), Mario Livio (STScI), Josh Younger (CfA), Chun Xu (STScI), Tomas Dahlen (STScI), Daniel Stern (JPL)\nComments: 82 pages, 17 figures, 6 tables. Data also available at: this http URL Accepted, Astrophysical Journal vol. 656 for Feb 10, 2007\nWe have discovered 21 new Type Ia supernovae (SNe Ia) with the Hubble Space Telescope (HST) and have used them to trace the history of cosmic expansion over the last 10 billion years. These objects, which include 13 spectroscopically confirmed SNe Ia at z > 1, were discovered during 14 epochs of reimaging of the GOODS fields North and South over two years with the Advanced Camera for Surveys on HST. Together with a recalibration of our previous HST-discovered SNe Ia, the full sample of 23 SNe Ia at z > 1 provides the highest-redshift sample known. Combined with previous SN Ia datasets, we measured H(z) at discrete, uncorrelated epochs, reducing the uncertainty of H(z>1) from 50% to under 20%, strengthening the evidence for a cosmic jerk--the transition from deceleration in the past to acceleration in the present. The unique leverage of the HST high-redshift SNe Ia provides the first meaningful constraint on the dark energy equation-of-state parameter at z >1.\nThe result remains consistent with a cosmological constant (w(z)=-1), and rules out rapidly evolving dark energy (dw/dz >>1). The defining property of dark energy, its negative pressure, appears to be present at z>1, in the epoch preceding acceleration, with ~98% confidence in our primary fit. Moreover, the z>1 sample-averaged spectral energy distribution is consistent with that of the typical SN Ia over the last 10 Gyr, indicating that any spectral evolution of the properties of SNe Ia with redshift is still below our detection threshold.\nby Pier Stefano Corasaniti\nThe constraints on dark energy do not seem to have changed and the claim that rapidly evolving dark energy is ruled out seems to be more the fruit of bias than of the new (or previous data). Just a quick look at the 2d likelihoods at page 66 in the w_0-w_a plane show that the ellipses are spread over a large range of values. For those who have passed Calculus I they should notice that w_a, the first order parameter of the linear expansion in the scale factor of w(a) (=w_0+w_a(1-a)) is >1 and even under strong priors the ellipses shrink but still allowing w_a~1 at 1\\sigma. Which basically implies that at 1\\sigma higher order corrections must be included in the analysis and higher order simply means rapid variation. Although if you include such terms you end up having also cancellations between the different orders, but this is a problem of using a Taylor expansion when the function you are trying to fit is not slowly varying. In the end the information that you get is equal to the one in the present likelihood. Namely dark energy can be Lambda or can also change in time.\nOn a more provocative subject I have noticed that the HST data still provide w_a shifted in the region of positve values, which is consequence of the fact that the high-z SN are slightly brighter than in LCDM. This is also the reason why the LCDM in the likelihoods at page 66 remains at the edge ot the 1 and 2\\sigma contours (exactly where it was in Riess et al. 04). If I remember there is plot in the Spergel et al paper (astro-ph/0611572 at page 31 fig 8) which shows the two binned datasets separately (the old data for HST) and there you can appreciate the small difference. Perhaps it is exactly because of this that the SNLS likelihood in the w_0-w_a plane does not exceed the w_a~1 region (see for instance the nice analysis by Huterer and Peiris astro-ph/0610427). Of course this discrepancy is 1\\sigma statistics and one should take it as a grain of salt, but I am wondering whether anyone has thought about possible experimental systematics which may affect in different ways one of the two experiments.\nPier-Stefano\nPier-Stefano raises some good points about priors and systematics. As I said in my previous post, ruling out positive w(z>1) comes essentially entirely from the distance to CMB last scattering, not SN. You can see this in the likelihood curves of Fig. 13 and 14 of R06, but it's been well known for years. As a most naive model, if w(z<1)=-1 and w(z>1)=w_1, then d_lss limits w_1<-0.2. You get similar results for a w_a model or most any monotonic w(z). I do want to correct the impression that my w(a)=w_0+w_a(1-a) is some sort of Taylor expansion. Unlike in Chevalier & Polarski 2001 (who properly restricted it to not being used cosmologically, z<<1), this is a fit function not an expansion, and in particular w_a=-dw_{true}/da at z=1, not at z=0 as in a Taylor expansion. In several articles it has been shown robust against bias from rapid time variation; even when \\dot w=H the bias is only 0.3\\sigma. Of course it does not handle non-monotonic w(z).\nRegarding positive w_a arising from slightly brighter high-z SN and systematics, many people have pointed out that there may well be a systematic involving instead dimmer local (0.03<z<0.1) SN. In particular, an offset between the local sample (observed very differently to date from z>0.1 SN) and rolling or blind search SN at z>0.1 indeed mimics a rapid time variation in w(z). This is one reason why the recent carefully designed local SN searches are so exciting.\nRe: Hubble Finds Evidence for Dark Energy in the Young Unive\nby Ben Gold\nPier Stefano Corasaniti wrote: The constraints on dark energy do not seem to have changed ...\nI don't really disagree with your statements about the [tex]w_0[/tex] -- [tex]w_a[/tex] parameterization, but if you look at the later figures where [tex]w(z)[/tex] is contrained in bins, you'll find a comparison of constraints with and without the high-[tex]z[/tex] supernovae (figure 15 on page 72).\nEven with the \"weak\" prior the high redshift SNe make a difference in the two higher redshift bins, and in fact shift [tex]w[/tex] down in the middle bin (0.45 < [tex]z[/tex] < 0.935) so that it's more consistent with -1 (and less likely that [tex]w>0[/tex]).\nIn fact, to address your later comments, I think the addition of the new SNe brings the Riess et al. dataset slightly more in line with the SNLS dataset, but I'd have to check the data to be sure.\nIt's been pointed out to me that the URL from the arXiv web entry to the data page doesn't actually work (a period snuck in at the end). The correct link is\nhttp://braeburn.pha.jhu.edu/~ariess/R06/\nfor the main page (which is still pretty bare at this point) and\nhttp://braeburn.pha.jhu.edu/~ariess/R06/sn_sample\nfor the magnitude/redshift data.\nby Syksy Rasanen\nBen Gold wrote:\nWhat you say is not inconsistent with Pier Stefano's comment. As I understand, the group of HST supernovae in figure 15 includes also old SNIa, not just the recently discovered ones.\nI don't know how much of a difference the new high-z supernovae make. But in the past the constraints on the evolution of the equation of state from SNIae have typically been overstated. Doing a model selection comparison instead of just evaluating confidence level contours shows how little one can say about the expansion history from the SNIa data alone.\nIn astro-ph/0512586, Shapiro and Turner conclude that one can say with confidence only that there has been acceleration, and that the deceleration parameter [tex]q[/tex] has decreased in time. However, with an approach that takes into account the penalties of introducing additional parameters Elgaroy and Multamaki show in astro-ph/0603053 that even the second statement is not supported by the statistics.\nIt's instructive to note that (for the Gold dataset) the \\chi^2 of a model with constant [tex]q[/tex] is 183, whereas the \\chi^2 of the [tex]\\Lambda[/tex]CDM model is 177. The number of parameters is the same, so this corresponds to odds of about 1:20. This may be considered suggestive, but hardly decisive.\nSo, strong conclusions about the evolution of [tex]w[/tex] have typically been driven by the assumed parametrisation; see for example astro-ph/0407364 by Bassett, Coranisiti and Kunz, or the debate in astro-ph/0311364, astro-ph/0403687, astro-ph/0404468 and astro-ph/0406672. (The case when data other than SNIa are also considered is discussed in astro-ph/0406608 by Coranisiti et al.)\nSyksy Rasanen wrote: What you say is not inconsistent with Pier Stefano's comment. As I understand, the group of HST supernovae in figure 15 includes also old SNIa, not just the recently discovered ones.\nTrue, that plot used 14 old HST SNIa out of a total of 32 (not all are at [tex]z \\ge 1[/tex]). But the new SNe do affect the likelihoods, in fact here's a similar figure but the comparison is that it's \"all\" versus \"no new HST\".\nTo fully interpret this you really need to also look at the window functions as they shift around slightly (which I don't have right at hand at the moment), which means it's not exactly an apples to apples comparison (which has something to do with the shift of the middle bin). But it should give a rough idea of how much the new SNe do.\nThe sensitivity to parameterization is one motivation behind doing things in bins in addition to [tex]w_0[/tex] -- [tex]w_a[/tex], though that itself is a parameterization. And the whole idea of the \"strong\" prior actually was to try to combine with other datasets (e.g. distance to last scattering) without goofing around with [tex]w(z)[/tex] in redshift ranges where no observations exist.\nAnyway, I definitely agree that trying to put general constraints on [tex]w(z)[/tex] without having some kind of model is a difficult problem! We tried a few different ways; I'm sure there are others.\nby Dragan Huterer\nBen Gold wrote: Even with the \"weak\" prior the high redshift SNe make a difference in the two higher redshift bins, and in fact shift [tex]w[/tex] down in the middle bin (0.45 < [tex]z[/tex] < 0.935) so that it's more consistent with -1 (and less likely that [tex]w>0[/tex]).\nThat's an interesting point, and I too think it's right. Riess07 find that the second bin of w is greater than -1 at somewhat higher than 1-sigma with new data, but looks like 2 sigma or more without new data. Cooray and I, using the same approach in astro-ph/0404062 (and also using the \"old\" Riess04 data), found that the second bin is pretty much on -1, but that the first bin is less than -1. So, both papers show some evidence for w'>0 roughly speaking (I am not too worried about the overall effective offset in w, since we are using the other, complementary data somewhat differently, and changes in those would change w everywhere).\nSo it is interesting that in Riess07 the new data helps reduce w' (which, as someone mentioned, is apparently not seen in SNLS). Not that the former was significant, but this might be saying something about the systematics, or lowz-highz offsets."