Luminosity Evolution in Type 1a Supernovae?

Figure 1 of Kang et al.

During this afternoon’s very exciting Meeting of the Faculty of Science and Engineering at Maynooth University I suddenly remembered a paper I became aware of over Christmas but then forgot about. There’s an article here describing the paper that makes some pretty strong claims, which was what alerted me to it. The actual paper, by Kang et al., which has apparently been refereed and accepted for publication by the Astrophysical Journal, can be found on the arXiv here. The abstract reads:

The most direct and strongest evidence for the presence of dark energy is provided by the measurement of galaxy distances using type Ia supernovae (SNe Ia). This result is based on the assumption that the corrected brightness of SN Ia through the empirical standardization would not evolve with look-back time. Recent studies have shown, however, that the standardized brightness of SN Ia is correlated with host morphology, host mass, and local star formation rate, suggesting a possible correlation with stellar population property. In order to understand the origin of these correlations, we have continued our spectroscopic observations to cover most of the reported nearby early-type host galaxies. From high-quality (signal-to-noise ratio ~175) spectra, we obtained the most direct and reliable estimates of population age and metallicity for these host galaxies. We find a significant correlation between SN luminosity (after the standardization) and stellar population age at a 99.5% confidence level. As such, this is the most direct and stringent test ever made for the luminosity evolution of SN Ia. Based on this result, we further show that the previously reported correlations with host morphology, host mass, and local star formation rate are most likely originated from the difference in population age. This indicates that the light-curve fitters used by the SNe Ia community are not quite capable of correcting for the population age effect, which would inevitably cause a serious systematic bias with look-back time. Notably, taken at face values, a significant fraction of the Hubble residual used in the discovery of the dark energy appears to be affected by the luminosity evolution. We argue, therefore, that this systematic bias must be considered in detail in SN cosmology before proceeding to the details of the dark energy.

Of course evidence for significant luminosity evolution of Type Ia supernovae would throw a big spanner in the works involved in using these objects to probe cosmology (specifically dark energy), but having skimmed the paper I’m a bit skeptical about the results, largely because they seem to use only a very small number of supernovae to reach their conclusions and I’m not convinced about selection effects. I have an open mind, though, so I’d be very interested to hear through the comments box the views of any experts in this field.

7 Responses to “Luminosity Evolution in Type 1a Supernovae?”

  1. Phillip Helbig Says:

    The hype around papers of this sort, intentionally or not on the part of the authors, tends to be along the lines of “the universe is not accelerating”, “there is no evidence for dark energy”, etc. The concordance model is called the concordance model because many lines of evidence leave one with the same cosmological parameters. We know what they are even without the supernova data. We know what they are based just on the CMB. So if conclusions from the supernova data are wrong, so must all the others be, which I find rather unlikely. On the other hand, other papers have claimed other problems with the supernova data, so perhaps two wrongs make a right in that various effects (say, luminosity evolution and wrong subtraction of the dipole, or wrong addition of redshifts, or whatever) more or less cancel, leaving one with the right result for the wrong reasons.

    • Is it true that dark energy is firmly established if we completely threw out the supernovae data? (And also without adopting strong assumptions like flatness)

      I thought the various cosmological probes carved out orthogonal slices through parameter space so that we are left with a highly constrained set of parameters. If we erase the supernova contours will we still have a posterior on Omega_Lambda that is concentrated above zero? Can you point us to any papers/plots?

      • Obviously, the more tests involved, the tighter the constraints, and, indeed, in the old days, only a combination of constraints led to a small region in the allowed parameter space, more effectively than one might have thought, because of the (approximate )orthogonality you mention. But individual constraints have improved. Yes, combining them still results in an improvement, but individual constraints now allow small regions of parameters space, ruling out a non-positive cosmological constant.

        The CMB results are usually presented differently, since lambda and Omega are “derived” rather than “basic” parameters, so one has to read between the lines. For example, Planck ( finds, using only CMB data without including any other constraints, that the sum of lambda and Omega is about 1.05. Formally, a flat universe is ruled out at 1 sigma but, depending on which Planck data one uses, only marginally at 2 sigma and not at 3 sigma. (Combining with other tests, however, “pulls parameters back into consistency with a spatially flat universe to well within 2 sigma”.) See section 7.3 of the paper linked to above. Table 1 in that paper (section 3) quotes Omega_m as 0.3158 with an uncertainty of about 0.007, assuming flatness, which implies that lambda is 0.6842. But, as noted above, the sum of the two is about 1.05. While it would be interesting to see constraints in the lambda–Omega plane from just the Planck data, it is clear from these numbers that, even using just the Planck data and pushing the uncertainties in the “preferred” direction, a non-positive cosmological constant is ruled out at several sigma.

      • “strong assumptions like flatness”

        Saying that flatness is a strong assumption can be confusing depending on how it is interpreted. It is definitely not the same as “assuming the Einstein-de Sitter universe”, as some did 30 years ago. The latter was a result of theoretical prejudice and the desire for closed solutions. A flat universe (which incidentally does allow closed solutions for some quantities, but that hardly plays a role today), however, is an observation, not an assumption. The best such observation is that of the CMB. As mentioned above, just the CMB tells us that the sum of Omega and lambda (or Omega_m and Omega_lambda for those who like subscripts) is about 1.05 (with a 1-sigma uncertainty of about 0.023).

        Why are CMB results now often presented assuming absolute flatness? Several reasons. First, there are enough parameters as it is, so why allow for an extra one even if we know that it will deviate from the strict assumption only slightly, even based on just the CMB data, and, when other data are used, the universe is compatible with perfect flatness? (Although not at all in the same category, one usually assumes that the gravitational constant doesn’t vary with time, even though some variation would be allowed by observations.) I think that that assumption is mostly harmless, unless one is investigating constraints on flatness as one’s primary goal. Another reason is that many people believe in inflation (for which there is some evidence, including the tilt in the power spectrum of CMB fluctuations), which generically implies a very flat universe, so flat that it is not worth the trouble to consider a deviation when analysing current observations, as the uncertainties in the observations are much larger than any possible deviation.

        All the same, I think that an actual detection of curvature, especially positive curvature, would be very interesting.

  2. Peter and others: still waiting for comments on

    • Richard Feynman once said to a journalist “Listen, buddy, if I could explain it in two minutes, it wouldn’t be worth a Nobel Prize.” Similarly, a blog-comment box is probably not the right place for subtle discussions. If there were some obvious goof, yes, but not for a detailed analysis. For what it’s worth, there is a lot of discussion, including some by yours truly and one of the authors of the paper, at Sabine Hossenfelder’s blog.

      One must also distinguish the hype surrounding such papers from the actual conclusions of the paper. Some pundit compared it to a smoky toaster leading to the headline “city on fire”, though on the other hand that doesn’t mean that we shouldn’t worry about smoky toasters.

      If anyone claims that the supernova data don’t support current acceleration of the universe, or even (a stronger claim) a positive cosmological constant, then the question is why the concordance model has the parameters it does, even ignoring the supernova data (see above). Something else is whether the data are consistent with no acceleration but also consistent with the concordance model. In the former case, one needs to explain the conspiracy of other tests which consistently led to the wrong result.

  3. It has been obvious for decades that an evolution of the zeropoint of SN brightness vs decline rate could mimic the effect of an accelerating Universe – see astro-ph/0201196 for example. But this excuse for the SNe data is no longer interesting. Back in 2003 the CMB data allowed any model along a geometric degeneracy line in the Omega_M,Omega_Lambda plane from the flat, accelerating LambdaCDM model with H~70 all the way to a super Sandage model with no dark energy, Omega_tot ~ 1.3, and H~30. But the current CMB data alone cut off the degeneracy and require enough dark energy to make the Universe accelerate. The combination of the BAO data with CMB data favors a flat accelerating model. As Phillip Helbig said, flat LambdaCDM is called the concordance model because it fits all the data, so no one dataset is essential.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: