Re: Phase II

From: Andrew Fruchter (fruchter@stsci.edu)
Date: Thu Apr 08 2004 - 09:12:28 PDT

  • Next message: Vitaliy Fadeyev: "Re: more info about 076 location"

    Adam told me he had a direct image in every visit.

    Your Phase II relies on a sole pointing for the offset between the
    grism and the direct image. We can probably align the grism spectra
    using something like tweakshifts. The only problem then may be a
    small error in the spectral zeropoint (of order a fraction of an ACS
    pixel -- perhaps 30A if we are not particularly lucky).

    Andy

    On Apr 7, 2004, at 3:26 PM, Rachel G. wrote:

    >
    > Hi Andy,
    >
    > I realized this would mean we'd have to think more, as
    > opposed to using the Pirzkal et al. software in the
    > straightforward way with the direct image and so forth. We
    > do have some of Adam's data from last year, and I'd planned
    > to start figuring out how to deal with it as soon as I'm
    > back in Berkeley. Vallery Stanishev in Stockholm will also
    > be working on this.
    >
    > Have you heard or do you have a sense for how large the
    > calibration errors that will creep in might get? Because, I
    > also thought Adam worked in this same mode. He could not
    > have done all his observations in one Visit. And, as you
    > know, his results look super.
    >
    > Ciao,
    >
    > Rachel
    >
    > On Wed, 7 Apr 2004, Andrew Fruchter wrote:
    >
    >> Hi Searchers,
    >>
    >> I have just approved visits F0-F5 for observation. However, you
    >> should
    >> know that you are using a very non-standard approach to grism
    >> observations. You are not taking a direct image in each visit (indeed
    >> the standard mode takes one associated with each dither position).
    >> This opens you up to the possibility of sub-pixel shifts between the
    >> direct image and the grism spectra.
    >>
    >> In principle one generally should not need a direct image. One should
    >> be able to use the zeroth order image (and this could be a more
    >> accurate way to align spectra in many cases). However, the
    >> calibration
    >> work necessary for this approach to be used has not been done, as far
    >> as I know.
    >>
    >> Cheers,
    >> Andy
    >>
    >

    This archive was generated by hypermail 2.1.4 : Thu Apr 08 2004 - 09:15:29 PDT