Re: poor agreement with Riess 99

From: Alex Conley (aconley@panisse.lbl.gov)
Date: Wed May 07 2003 - 20:08:31 PDT

  • Next message: Alex Conley: "K-corrections for low-z SNe"

    I can tell you from experience that sn1995bd is a 'funny' supernova
    when it comes to lightcurve fitting. If you look at my SN by SN
    comments when I refit all of the SNe as a stability test, this is
    what I had to say about 95bd:

    sn1995bd -- a piece of junk. The fits move around a bit when the initial
       conditions change, but mostly this just misses the peak badly, en
       when a floating offset is used. I'm glad this one gets excluded from the
       fits.

    and 1996bo

    sn1996bo -- solid, both fixed and floating. However, the floating fit is
      worse. It seems to miss the peak. Again, using floating offsets has
      ruined a perfectly good fit

    Here I felt that floating offsets were a mistake.

    As for the other Riess SNe:

    sn1994M -- solid
    sn1994S -- pretty solid, but the V peak seems slightly off. Changing to
      floating offsets doesn't really improve this. In any case, changing
      the initial conditions changes the error bars a bit, but not the central
      values.
    sn1995ac -- solid
    sn1996bl -- solid

    The full reference for these fits is
    http://panisse.lbl.gov/collab/archive/hstpaper/0098.html
    and a bit more description of what I was doing
    http://panisse.lbl.gov/collab/archive/hstpaper/0093.html

    Attached to the first email is a tgz containing all of the output
    postscript for the fits.

    Alex

    On Wed, 7 May 2003, Greg Aldering wrote:

    > HERE IS THE REVISED VERSION:
    >
    > Hi Rob,
    >
    > I started to investigate "what'sup" with SN1995bd and SN1996bo and
    > found that your B-band magnitudes differ, sometimes by a lot, from the
    > published Bmax tables from Table 3 of Riess 1999 (the 22 SNe paper).
    > Further investigation showed generally poor agreement. Can you
    > comment?
    >
    > no extinction cor
    > Bmax mB mBcor Bmx-mB
    > R99 SCP03 SCP03
    > ---------------------------------------------------------------------------
    > SN1995bd 17.27 15.32 17.39 HUGE! (weird that mB_cor is much better)
    > SN1996bo 16.15 15.85 15.92 +0.30
    >
    > also, here are the other Riess 1999 SNe for comparison
    >
    > SN1994M 16.35 16.25 16.13 +0.10
    > SN1994S 14.79 14.78 14.93 +0.01
    > SN1995ac 17.19 17.05 17.37 +0.14
    > SN1996bl 17.08 16.67 17.14 +0.41 way off!
    >
    > These seem like large differences (some extremely large) given the
    > small uncertainty you quote from fitting the same data (few tenths of a
    > magnitude). Some of this is probably due to the crudeness of the
    > delta-m15 fits quoted in Riess 1999. I seem to recall that in P99 we
    > may have had a similar issue with the H96 data, but I haven't checked
    > that yet.
    >
    > These offsets aren't enough to explain the anomalous extinction for
    > 95bd and 96bo, unless the extinction estimate is off. So, I looked at
    > your E(B-V) values and compared them to the value in Phillips 1999
    >
    > your ph99 tail ph99 peak
    > E(B-V) E(B-V) E(B-V)
    > ------------------------------------------------------
    > SN1995bd 0.339 0.242 NOT GIVEN - dm15 < 0.9
    > SN1996bo 0.390 0.358 0.385
    >
    >
    > So here, the agreement for E(B-V) at max for 96bo is good. 95bd may be
    > another question, but it isn't as big an outlier as 96bo in the
    > extinction-corrected fits.
    >
    > On to more data sifting.
    >
    > - Greg
    >



    This archive was generated by hypermail 2.1.4 : Wed May 07 2003 - 20:08:31 PDT