SN9784. Greg has this host on the HST image which looks like an inclined disk galaxy, with the SN out of the disk fo the galaxy. Peter has objected that no SN has ever been seen 5kpc away from the disk of the galaxy. Greg responds that, hey, this is out of something like 50 known SNe. (Presumably this doesn't count ours, just "nearby" ones.) If the SN rate goes up at redshifts of 0.8, then there is less of a limit. As such, it may not be unreasonable to have seen such an event.
(The question is whether we should use this host, with the SN far away from the disk of a galaxy, as an arguement against extinction.)
Greg also argues that the rate of SN~Ia in the halo ought to be higher in the past. Locally, the halo is 1% of stars. The statement that none have been found is consistent with this. If the rate is enhanced in the halo, then it gets up into the regime where you don't dismiss it out of hand.Rob (that's me) points out that because we were trying to avoid AGN by selecting things without a host or far from the core, we may also have been slightly biased towards finding these rarer events. (This is probably something we shouldn't say in a paper, or if we do, we should do it with great care.)
Greg did a histogram in the coadded HST image, randomly laying down apertures of the same size as where he saw the host. In this distribution of fluxes in a 1" aperture, our object is well above most of the fluxes, sort of a 2-3% kind of event. This means the probability of randomly having something that bright within 1" of the SN is low. The detection is out there in someting like the 3*sigma regime.
For extinction arguements, Greg recommends that you do the color first. It's consistent with no extinction (or close), even though there is a lot of uncertainty.
We are also still disagreeing over exactly how to do the first (fix or not the R-band zero level?), do we use the sub or nosub version? (If this paper ever gets submitted, I'll eat my shirt.)
Gerson has a value of Omega for 38 SNe. He went through the Hamuy, and did the fits allowing the K correction to stretch. He got constants for color corrected and not color corrected (by which he means correcting for the extinction in the host galaxy; B&H was already included, and will henceforth be ignored). He says that the difference is small, of about a tenth of a magnitude.
OK, we're confused about what he did. He had a formula from Peter where E(B-V)=B_V+0.2*stretch-0.14 (recited from memory by gerson). Then somehow this E(B-V) gets turned into A_B; was this a multiplication by 4? Gerson doesn't remember right now.
Gerson goes to our data and looks at 38 SCP SNe, throwing out some really bad ones (in terms of really poor measurements). (Perhaps others should be argued to be thrown out.) If s<0.8 or s>1.4, he limited the stretch to those extreme values. This is the first time that B&H extinction is included. He has a hubble diagram, and made a histogram of Omega_M in the flat universe. It comes out peaking somewhere near 0.2, with more or less symmetric tails on either side (so much for the host eextinction tail). (Well, OK, the shoulder to the higher values of Omega_M is bigger than the other one.) (None of this is host extinction corrected.) All 7 of our first 7 SNe have Omega_M values greater than the peak.
Saul thinks that we are probably more affected by extinction; the assertion is that we have more dynamic range, and that the nearby guys are more affected by Malmquist than we are, meaning we are probably sampling a greater range of extinction.
Re: the first 7 being higher, Saul denys that Malmquist bias of the first seven could explain this, since the efficiency curves were investigated, and that we were well away from the edge and not suffereing from Malmquist bias.
Our recent SNe: we can make cuts for their redshift for how close they are to the limit. (I.e. at a given magintude SN, cut out those SN whose redshifts would put them towards low efficiency.) (This requires thought.) Malmquist bias is a potential explanation for the higher shoulder towards higher Omega_M.
In any event, there seems to be a real peak at Omega_M=0.2 for the flat universe.
Peter says that there seems to be a correlation between extinction and stretch, i.e. wider SNe tend to higher extinction. Rob says he thinks this means that the color-stretch relationship could be misunderstood. Greg points out that this could be Malmquist bias, because you'd preferentially find the wider (brighter) extinction ones. This could affect things if the nearby SNe are more effected by extinction than ours.
Peter argues that the only fair way to do it is to somehow correct for extinction independently for both sets. For instance, you could just do cuts on both; only keep the SNe which have colors close to the Branch normal set, with B-V=0.0+-0.2. (The actual distribution has far more on the red side than the blue side, probably because of extinction.) Peter says that $20 says you get rid of 3+-1 of the ones in the high shoulder and 2+-1 of those in the low shoulder. Gerson says that he will attempt this.
OK, we're puzzled again. Why is Gerson's distribution so different from the same distribution we did earlier? (Basically, it seems that Saul and Gerson produce a different histogram.) We hemed and hawed, and didn't really decide much. We still need to worry some more.
Peter said that Adam has shown him some data from CFHT of two 91T-like SNe. He says that when you apply extinction, differing as a function of time, the lightcurve starts looking like its unstretched. (I think that's what he said.) In short, R_V changes as a function of time! He says it varies from 3.1 at max to 3.4 at day 15. Consider, for example, E(B-V) of 0.8, so AB changes by 0.25 out of 2 magnitudes out of that range. Delta m15 goes from 1 to 1.25, which is the difference between a normal guy and an 19T-like guy. Lesson: extinction corrections have to be done as a function of time. Peter says that this should go into the new Kim templates.
This relationship could be different at different redshifts, as well, depending on which parts of the spectrum get shifted into our R and I filters.
Gerson says he will keep working on this. He would also like somebody else to check it.
I've been known to make mistakes.
-G. Goldhaber
Back to the SN9784 paper.
How to present this: one way is to go as straight as possible and say we have Omega=0.6 in a flat arguement, based on the assumption that we don't have exinction. We then give several arguements that there is no extinction: we don't see host galaxy, if we do, it's far, our colors aren't well known, but they are more or less consistent with no extinction. Make it clear, however, that we shouldn't base all of our judgements on this one SN, especially considering that it's a preliminary result. Rather, you need final results, and on more SNe at higher redshift.
That can all be handled with weasel words.
-G. Aldering
Discussion of the paper: Gerson agrees with Ariel that we shouldn't put the Omega result in the abstract. Saul worries that if we don't put it in the Abstract, it won't sound like a Nature paper. This convinces Gerson that we should leave it in the abstract. We should leave in the weasel words that make it clear that its preliminary.
Don is worried about talking about extinction in the abstract and its affects on Omega, since later we make strong arguements for there only being weak extinction. Next, we're talking about making the weasel words more weaselley.
Authors. Saul says he's narrowed it down to fifty or sixty of our closest friends. It has (in these meeting minutes, in no particular order whatsoever) the immediate people in the group, the two Richards, Ariel, Reynald, Isobel, Pelar, Nick Walton, Brad Schaefer, Chris Lidman, Andy Fruchter, Nina Panagia. The BTC team will get a strong acknowledgement.