LCDM
plus everything else we know about star formation appears to get luminosities right (
http://arxiv.org/abs/astro-ph/0403293) but this is obviously a less precise job, so we're doing OK on (c) too.
I want to add one more comment to this. Why is a mainstream physicist (like me) allowed to say "you test LCDM theory with SNe1a, not with Tolman surface brightness", but I get to complain if Lerner says "You don't test PC with SNe1a, you test it by looking for close quasars pairs"? Isn't that a double standard? I don't think it is.
The reason SNe1a is a good test of LCDM is not "because it matches". SNe1a is a good test of LCDM because this test has a strong potential for
falsification. It is known a priori that the theory calculations are very precise. It is known a priori that the measurements can be done with low systematic error---SN1a are called "standard candles" for a reason. Therefore, you embark on this measurement knowing that the results will stand up to scrutiny
no matter what they tell you.
Tolman surface brightness? Observationally it's a mess (as you know ahead of time); it's a morass of filters, flux-limited area estimators, foreground stars, etc.. Theoretically it's worse; you need all of this star-formation history and merger history in the
model, and you need it all to fit together well enough to synthesize spectra. You know this ahead of time. You know, ahead of time, that if Calculation A disagrees with Observation B, someone will try again with a different filter set and star-formation model and get a different result. (That's how science works.)
Nobody is allowed to ignore a test
which has high falsification potential---a test, like SNe1a are for LCDM, that's precise and unambiguous both on the theory side and the observation side. That's not to say that the tests are the same for all theories.
All of the tests that
could have falsified LCDM---BBN, SNe1a, CMB, CMBpol, weak lensing, etc., I've been through this before---have
agreed with it. They have not forced it to add parameters.
What are the list of tests that
could have falsified PC? Let's look. We're looking for the list of phenomena for which PC can make a rock-solid and precise theoretical calculation---not an approximate one, not a photo of a terella and the words "sort of scale this up", not an elliptical screed about how that there's something weird about LCDM. SNe? Nope. CDM? Nope. Tolman? Sort of; Lerner presents a theory paper of extraordinarily low quality,
with missing data and entirely without error bars, on a test where we know the observations suck, so No. BBN? (My home institution, an extremely large research university, doesn't subscribe to the obscure journal where the supposed PC BBN theory is published.) BAO? LSS? Weak lensing? I'm trying here, Zeuzzz.
What precise cosmology calculation has strong falsification potential for PC?
None. Go down the list of
precisely observed bits of cosmology, and right down the list you'll see PC avoiding precise calculations. Go down the list of PC's precise predictive calculations, and you'll see ... oh wait. Go down the list of PC's
vaguely mathematical sort-of-calculations and you'll see a list of qualitative or otherwise imprecise observations.
Why the anticorrelation? What makes precision-observables so hard for PC to calculate?