retry of retry of comments on paper

From: Alex Kim (akim@panisse.lbl.gov)
Date: Mon Feb 24 2003 - 21:02:32 PST

  • Next message: Serena Nobili: "Comments on HST paper"

    attached mail follows:

    Rob,

    I lost my comments, oh well, here goes again.

    Need a systematic error section. A lot of if can reference P99.

    Systematic errors that should be addressed in detail:
    1. Error in intrinsic supernova color. Your ridgeline work is important
    meriting more mention or a paper itself. But at the least in this paper I
    would like to see
    - plot of stretch vs. intrinsic B-V
    - comparison of your B-V with RPK and our assumed colors in P99
    - how does the new colors change cosmological parameter determination?
    - you give the color dispersion, but what is the error in the mean?
    - the error in the mean color should be included as a correlated error in
    the cosmology analysis.

    You discuss improved determination of U-B colors but the K-corrections are
    seriously problematic as well. Lentz et al. show that variations in flux
    within U is stongly wavelength dependent. Make the case for the U-band
    K-corrections you use.

    2. Malmquist bias. Knowing that we were going progressively deeper in our
    searches, when I saw the omega_m - omega_lambda confidence region rise, I
    thought "Malmquist bias in the P99 sample". Do we have enough confidence
    in the P99 analysis that this is not the case?

    Smaller comments ensue
    Generally...
    I think capitalized Universe is appropriate for our universe.
    w is not the equation of state of dark energy. rho=wp is the equation of
    state. w is rho/p or the equation of state parameter.

    p1
    abstract
    "yield a measurement of $\Omega_M$ for a flat Universe"

    p2 col1 par 2
    "lends confidence"

    p2 col2 par2
    You say "just these 11 high-redshift supernovae provide limits" but don't
    forget about the nearby sample.

    p3 col1 par2
    the word "effect" twice in a sentance.
    "known colors of SNe Ia, can provide an upper"

    p3 col2 par1
    "from these data alone and combined with recent"

    p4 col1 par1
    May mention that SN redshifts are always consistent with galaxy redshifts.

    par2
    Put upper and lower bound on the inequalities.

    col2 par 1
    May mention that x_{0i} and x_{0i} are fit unlike P99 wide-field ground
    data.

    par2
    (Fruchter 2002)

    p6 col1 par 2
    In ApJ I notice that websites are generally given as footnotes.

    col2 par 2
    First sentance incomprehensible.
    Why is WIYN 3.5M whereas there rest xm?

    par 3
    "11 SNe"

    par 4
    "The exceptions are the seven"

    p7 col1 par 1
    "de-redshifted R and I bands"

    col2 par 1
    z=0.18
    I didn't look that hard but I didn't find SN97I and 97N's R-band
    magnitudes tabulated. How did you use these numbers in the analysis? An
    R to I K-correction?

    p8 col1 par 3
    Bessell I think, as opposed to Bessel functions.

    col2 par 3
    "BVRI spectral"

    Table 3-5
    I think readers would be interested in knowing the \chi^2/dof

    p14 col1 par2
    "values were then used"

    col2 par1
    Figure labeling screwed up.
    par2
    "this paper shows" (subject is data I think)

    p15 col2 par1
    R99 instead of H99.

    Figure 2
    It bothers me that H96 E(B-V) distribution looks clearly negative. Makes
    me doubt the ridgeline color determination.

    p17
    Table 7
    "Weighted Mean E(B-V) Values"
    "Fit 3" SNe (otherwise it looks like you only fit 3 SNe.

    col1 par1
    "one is omitted as an outlier."

    par2
    "then the mean"
    E(B-V) $< 3\sigma$

    p18 col1 par 2
    "which plot results from both the"

    col2 par 1
    "a dispersion"
    Should explicitly say that B and colors are not intrinsically correlated.

    par2
    Should explicitly write script M = M + 5log(H0) or whatever it is.

    p 19 col 1 par 2
    Could you delineate the cuts early in the paragraph? Its a little
    confusing mentioning them one by one as you go.

    p21 col1 par2
    "explicitly correct for host galaxy"
    Its not clear to me if you make cuts on "reddened" supernova and/or "red"
    supernova.

    p25 col1 par 1
    "higher redshift supernovae"

    par2
    Repetitive with the "inconsistent with a flat ..." Look at p21 col2 par
    1, almost the same wording.
    Will you give a best fit value for SNe alone?

    p25 col2 par2
    Punishment for taking a while to finish a paper means you have to include
    WMAP results.
    Omega_Lambda > 0

    p27 col1 par2
    fix lower error bar.
    Can you explain why inclusion of P99 supernovae keeps errors the same size
    but move the best-fit by so much? What are the chi^2/dof of the two
    fits?

    col2 par 1
    Omega_Lambda

    Should we acknowledge observatories?

    Alex

    This archive was generated by hypermail 2.1.4 : Mon Feb 24 2003 - 21:02:32 PST