Re: My last mark-ups.

From: Chris Lidman (clidman@eso.org)
Date: Fri May 28 2004 - 09:36:08 PDT

  • Next message: Chris Lidman: "VLT spectroscopy paper"

    Hi Saul (with cc to Andy),

    On Thu, 2004-05-27 at 20:09, Saul Perlmutter wrote:
    > Hi Chris,
    > That's funny-- I looked at what I sent you and it appears marked up
    > when I look at it with Adobe Acrobat -- but maybe it's because I have a
    > rather recent version of Acrobat. I'm re-saving it a different way,
    > and resending it. Let me know if this works.
    >

    Yes, it works. Thanks.

    > The two things that I think are *most* important in mark-ups are the
    > things I pointed out about the abstract and the histogram figure.
    > Just to add to what I said on the note about the histogram figure -- I
    > believe that the supernovae above z ~ 1 that we call Type "Ia?" (with a
    > question mark) might be just as confirmed as Type Ia as the ones above z
    > ~ 1 that Riess et al.'s recent HST paper called Type Ia. In other
    > words, we might be being more stringint on these than he was (we should
    > probably check this by comparing the figures in his paper and ours), and
    > if we don't make this clear then we are likely to have problems when we
    > publish our set of z > 1 SNe and people think that the Riess et al SNe
    > were "better confirmed."
    >

    Our classifications are more stringent than Riess et al. Look at
    SN 2002 ki in Riess et al. (ApJ 607, 672) as an example. They classify
    this noisy spectrum as a SN~Ia with high confidence. This would be
    neither a Ia nor a Ia? in our classification scheme. Andy, would you
    agree with my assessment on the classification scheme used in the
    Riess et al. paper?

    They even contradict themselves. SN 2003aj has notes e and g, which
    stand for "Classified as SN Ia with high confidence from early-type, red
    host" and "Uncertain type" respectively.

    Overall, they present 17 spectra. On our scheme, 8 would be Ia, 2 would
    be Ia? and 7 would be unclassified. I have not done a detailed analysis
    so these numbers could move around by 1 or 2 units.

    However, some of their z > 1 SNe are clearly SN Ia from the spectra, but
    it is worth noting that all of these were done with ACS. They have
    several z > 1 candidates that were observed from the ground, but none
    of these would be classified as Ia or Ia? in our scheme. Compared to our
    competitors, we have done much better with the ground based spectroscopy
    of z > 1 SNe.

    In the VLT paper, we've done the classification on the spectra alone and
    I hope that I have been able to make this point clear. Once you add
    additional information, whether it is from the nature of the host, the
    colour of the SNe, or the light curve, then one is making inferences on
    the spectral type. In the Riess et al. paper about half of the
    supernovae are inferred to be Ia's.

    Do we want to add this additional information for the supernovae in this
    paper, especially for the ones that have been classified as "Ia?"? We
    could, but someone has to look at the photometric data for these SNe.
    I know that a preliminary analysis has been done on SN 2002kn and SN
    2002kr, but I do not know the status of SN 2002fe, SN2002fg and 2002gj.

    In regards to the abstract, I've expanded it a bit by adding one
    additional sentence. The abstract now reads

    We present VLT FORS1 and FORS2 spectra of 39 candidate
    high-redshift supernovae that were discovered as part of a
    cosmological study using Type Ia supernovae (SNe~Ia) over a wide range
    of redshifts. From the spectra alone, 15 candidates are securely
    identified as SN~Ia with redshifts ranging from $z=0.212$ to $z=1.181$
    and an additional 5 candidates are identified as possible SN~Ia with
    redshifts ranging from $z=0.44$ to $z=1.086$. Of the remaining 19
    candidates, 1 is a possible Type II supernova and 9 have broad
    supernova like spectral features and/or supernovae like light curves.
    We present the spectra of all candidates, including those that could
    not be spectrally classified as supernova.

    I do not think our overall success rate is unusually low, so I have
    not added this abstract. One has to be careful how one defines it.

    At z=0.5, it is 75% and this is good.

    At z~1, it is much lower, but for the ground based observations made in
    the Riess et al. paper it is zero.

    Regards, Chris.

    > Talk to you soon. (If you want to reach me I should be around
    > tomorrow.) Regards, --Saul
    >
    >
    >
    > Chris Lidman wrote:
    >
    > >Hi Saul,
    > > The marked up version you sent me appears to be unmarked. Could you
    > >send me the marked up version.
    > >
    > >Chris.
    > >
    > >On Tue, 2004-05-25 at 00:47, Saul Perlmutter wrote:
    > >
    > >
    > >>Hi Chris,
    > >> I think the paper's looking good, and should be submitted. Here
    > >>are few final things that would be great to fix before submitting if
    > >>possible.
    > >> Talk to you in a few hours! --Saul
    > >>
    > >>
    > >>
    > >>
    > >
    > >
    > >



    This archive was generated by hypermail 2.1.4 : Fri May 28 2004 - 09:36:31 PDT