From: Chris Lidman (clidman@eso.org)
Date: Wed Oct 13 2004 - 06:21:59 PDT
Hi Serena,
We've just had an executive meeting and we have discussed the
problematic case of SN 1999Q. Here are some suggestions.
i) SN99Q is plotted with a different symbol in Figures 13 and 14.
ii) SN99Q is plotted as a point and not a band in Fig. 14. This means
that it has to be plotted in the upper plot Figure 13 and listed in
Table 11, which it is currently not.
iii) We point out the peculiarities of this SNe, i.e. a
relatively large stretch, a faint restframe I-band magnitude, a
relatively red B-V colour, a relatively blue B-I colour
and a restframe B-V colour evolution that is suggestive of low stretch.
However, we should also note that the colours have been analysed by two
teams (we measured the restframe B-V colour and they measured the
restframe B-I colour) and that there is disagreement on the fit to
the date of maximum light.
iv) SN99Q is not used in the grey dust analysis. I.e. it should be
removed from Table 12.
Unfortunately, the results of the reanalysis of the SOFI data does
help us to understand this candidate any better. The restframe
B-I colour becomes redder and the restframe I-band magnitude
becomes brighter. To make further progress, we need the Keck data, and
I have asked for it.
As I am writing this e-mail, another thought has occurred to
me. Vitaly measured the restframe B and V band magnitudes, so we
should be able to compute a restframe B-I colour and a peak
B band magnitude. This leads to two questions.
a) How does this SN look in the B-band Hubble diagram. Is it also an
outlier?
b) If we use our restframe B-band magnitudes and their restframe I band
what is the B-I colour of SN99Q and how does it compare with what they
have published.
Cheers, Chris.
We all agree that SN 1999Q should be plotted differently in
figures 13 and 14 of your paper.
In the latest version of the paper, you state that you plan
to plot SN 1999ff differently as well. We understand that the
reasons for doing this is to highlight our own work on SN 2000fr.
But we should be cautious here. We do not want to cast SN 1999Q
and SN1999ff in the same light.
On Tue, 2004-10-12 at 11:04, Chris Lidman wrote:
> Hi Serena,
> The problem with the photometry is not related to the problem that
> had existed when the data were taken.
>
> Originally, I had suspected that they had used the time recorded in
> the EXPTIME keyword and not the DIT keyword. In 1999, EXPTIME was set
> to NDIT * DIT and not DIT as it should have been. This does not
> appear to be the case. Hence the reason why we find a different
> flux for the SN99Q is now unknown.
>
> If we keep 99Q in the paper then I agree that it should be plotted
> in a differently and that we should be forthcoming with our doubts.
>
> Regards, Chris.
>
>
> On Tue, 2004-10-12 at 08:18, Serena Nobili wrote:
>> Dear Chris and SCPexec members,
>>
>> given this result, I am changing the paper according to what I
>> believe was
>> decided in the previous executive meeting. That is, SN 1999Q will be
>> in
>> the paper, it will appear light gray in the plots, and a good
>> discussion
>> will point out all the doubts and problems we have about it, both in
>> the
>> optical lightcurve discrepancy (about the time of Bmax) and the
>> infrared
>> data reduction of the SOFI data.
>>
>> ** Chris, is there a reference or a web page explaining what was the
>> general problem everybody encountered for the SOFI data taken at the
>> time?
>> That would help justifying why we went back checking the IR data
>> reduction for this SN.
>>
>> I hope to be ready with a new version within a few days.
>> Cheers,
>>
>> Serena
>>
>>
>> On Mon, 11 Oct 2004, Chris Lidman wrote:
>>
>>> Dear Serena,
>>> I've looked at the SOFI 99Q data again. I downloaded the standard
>>> star
>>> observations and I can confirm that the data were taken during
>>> photometric conditions. With the newly computed ZPs (previously, I
>>> had
>>> taken the ZPs from the ESO web pages), I compute a magnitude of
>>> J=22.64
>>> +/- 0.15 for 99Q, which means that we still find that this object is
>>> 0.4
>>> magnitudes brighter than the magnitude reported in Riess et al.
>>>
>>> Even though Riess et al. report that they measured the flux of
>>> standard stars during three photometric nights, it is not explicitly
>>> stated that these nights correspond to data that were taken with the
>>> Keck telescope. The SOFI observations of 99Q were taken on two
>>> different nighs. I've checked the first one and it was photometric.
>>>
>>> I do not think we can make much additional progress on this SN
>>> until we get the Keck data.
>>>
>>> Regards, Chris.
>>>
>>>
>>>
>>>
>>> On Mon, 2004-10-04 at 13:13, Serena Nobili wrote:
>>>> Dear all,
>>>>
>>>> since Chris is back, I would like to share with you some thoughts I
>>>> have
>>>> on the issue of the first epoch of SN 1999Q. I hope to have some
>>>> feedback
>>>> from you.
>>>>
>>>> 1) Assuming the mistake Riess et al. made is on the first epoch
>>>> only, (see
>>>> #2 in this email) I played around with the light curve, making the
>>>> first
>>>> point 0.4 mag brighter (quoted by Chris), and fitted the light curve
>>>> again. I attach the plots showing the best fitted nearby template
>>>> in the
>>>> two cases:
>>>>
>>>> mysn99Q.bestfit.ps obtained using the published photometry
>>>> Imax= 23.86 +/- 0.08
>>>> chi2= 2.11629
>>>> template = SN1999ac
>>>>
>>>>
>>>> mysn99Q.plus0.4.ps obtained adding 0.4 mag to the first epoch
>>>> Imax= 23.61 +/- 0.08
>>>> chi2= 7.05273
>>>> template = SN1989B
>>>>
>>>> Although I find the second case a bit odder, since the first point
>>>> is
>>>> far too bright, it is still reasonable.
>>>>
>>>> Note that in both cases the two templates for under-luminous
>>>> (1991bg and
>>>> 1997cn) give a significantly poorer fit.
>>>>
>>>>
>>>> 2) I find difficult to believe that all the 5 points are wrong by
>>>> 0.4 mag.
>>>> According to the paper, they used 3 photometric night to calibrate
>>>> all 5
>>>> nights. This, if true, would exclude the possibility that they use
>>>> the
>>>> first epoch (taken with SOFI-NTT) to calibrate the other 4 nights
>>>> (taken at
>>>> NIRC-Keck), and get them all wrong. Having said that, of course, if
>>>> this
>>>> statement in their paper is true, is likely to be the only true
>>>> statement
>>>> in the whole paper. This is certainly making things more
>>>> complicated for
>>>> us to explain.
>>>>
>>>> 3) Assuming all epochs to be 0.4 mag brighter, the Imax will be 0.4
>>>> mag
>>>> closer to the concordance model in the Hubble diagram.
>>>> At the same time, B-I for this SN would be about 0.4 mag too blue
>>>> in the
>>>> B-I color evolution. I am not saying this is necessarily a problem,
>>>> I am
>>>> only pointing out all of this.
>>>>
>>>> I hope this email will help the discussion and we will be able to
>>>> find a
>>>> way out as soon as possible.
>>>> Cheers,
>>>>
>>>> Serena
>>>>
>>>> --
>>>> www.physto.se/~serena
>>>> Tel +33 1 44 27 72 52
>>>>
>>>> Give free food at:
>>>> http://www.porloschicos.com/
>>>> http://www.thehungersite.com/
>>>>
>>>
-- European Southern Observatory Alonso de Córdova 3107, Vitacura Casilla 19001, Santiago 19 CHILEPh. +56 2 463 3106 FAX +56 2 463 3101
This archive was generated by hypermail 2.1.4 : Wed Oct 13 2004 - 09:50:50 PDT