[fitswcs] comments and questions about ongoing WCS discussions

[Don Wells]

Perry etal,

I am in the process of preparing an itemized change order for approval
by this TG and then by the sci.astro.fits/fitsbits readership. This
change order will summarize our negotiations and will give Eric
Greisen, Mark Calabretta and Doug Mink the solid mandate that they
need to modify the WCS documents and software.  I currently expect to
post this document to fitswcs next week. My document references Bill
Pence's document on BINTABLE WCS conventions, and he tells me that he
will post a revised version of his document soon after I post mine.

Perry Greenfield writes:
 > ..As to whether the proposals satisfy the needs of existing and planned HST
 > instruments I cannot be definitive because I have lost track of what
 > proposals are on the table for handling instrumental and optical distortions.
..
 > 3) Combinations of lower order polynomials with existing, standard projections
 > (e.g., TAN with 3rd order) or radial terms.

Your item (3) is what we are now discussing. Appendix-A will be deleted.

 > .. [re] HST instruments..
 > In general, no distortion information is handled within any wcs-type 
 > system.  Distortion information is generally stored outside of the
 > science image and special tasks are run to either resample the data or
 > to provide coordinate mappings (e.g., sky coordinates corresponding to
 > pixel coordinates)..

A key objective of our discussions is to agree on a FITS-keyword
distortion interchange notation which is strong enough to represent
your instruments sufficiently well, and which will be implemented in
astronomical software systems everywhere. The same notation would also
represent ground-based instruments sufficiently well, and your support
for your own instruments in your ST-SDAS software would then imply
that you would handle distorted ground-based imagery properly, while
non-STScI software would handle your imagery properly.  I.e.,
observers could intercompare STScI-optical, ground-optical, radio,
.. and X-ray imagery using either STScI software or other software,
and get comparable results. Any such imagery could be regridded
accurately onto any projection with any of the software systems.  Note
that the imminent availability of Monet's 5e8 stars on the Hipparchos
system means that we will soon be able to put almost any ground-based
optical/infrared imagery onto the same coordinate system that
HST/AXAF/VLA/MERLIN/etc are using.

 > .. The distortion model
 > for the FOC is not well fit by polynomials (even of order 10) and is instead
 > generated by a spline model. Offhand, I am not familiar with the MAMA
 > distortion characteristics for STIS but I am sure that it is much better
 > behaved than the FOC. ..

The FOC probably exhibits S-distortion. We could add S-distortion
terms to the [TBD] set, but I think we should do so only if there is a
demonstrated need using current instruments. My impression is that FOC
data has always been heavily proecessed, including almost automatic
re-gridding. The pixel scale of FOC satisfies the Sampling Theorem
(unlike WF/PC), so regridding need not lose any information.

A key question is whether MAMA detectors have significant S-distortion
(their barrel/pincushion will be handled by the r^2 and r^4
terms). Many of the image tube detectors used in the 70s had
S-distortion; I once had software which could fit such terms.

 > .. For FOC, I doubt anything short of a spline
 > model or finely sampled pixel regularization image would suffice..
 > .. unsure.. splines.. necessary for the STIS MAMA detectors..

Figure 7.4 of the STIS Instrument Handbook makes it clear that this is
a microchannel plate [MCp] detector with an anode array close to the
back of the plate. This means that barrel/pincushion- and
S-distortions are probably minimal. The window and the curved front
surface of the MCP plate will probably produce some radial distortion
terms. Summary: my guess is that the STIS MAMA detectors in imaging
mode will be fitted nicely by the [TBD] terms.

 > Surprisingly, bilinear interpolation for a coarse pixel regularization
 > image may not be sufficient for some applications. We tried that for FOC
 > and were surprised to see that the effect on the appearance of images 
 > geometrically corrected using bilinear interpolation on a coarse grid lead
 > to quite apparent discontinuities in apparent brightness when flux conserving
 > geometric corrections were used. Basically the Jacobian is discontinuous
 > on the grid and this was very apparent when applying the distortion correction
 > to flat fields.

Fascinating. Thanks for this report!

 > If splines were part of the WCS, I am confident they would obviate any need
 > for using pixel regularization images for our existing or anticipated
 > instruments (well, with one exception: It is possible that CCD's may have
 > pixel offsets arising form manufacture that could be modeled and that would
 > require a fully sampled pixel regularization image)

If some form of Appendix-A ultimately reappears for optional
differential corrections to the TAN+terms projection, perhaps it
should be an image of 2-D spline terms rather than a map for bilinear
interpolation.

 > ..the G&C draft consistently refers to keywords when describing naming
 > conventions for WCS parameters in such tables which leads me to believe
 > that it is describing header keywords in which case different rows must all 
 > share the same WCS parameters. If so, it will not be usable for some data
 > (STIS for example). 

See Appendix B of G&C96, especially Table 8 on page 37, and see the
postings by Bill Pence and Mark Calabretta in recent days which update
Appendix B. 

-Don

-- 
  Donald C. Wells         Associate Scientist         dwells at nrao.edu
		    http://www.cv.nrao.edu/~dwells
  National Radio Astronomy Observatory                +1-804-296-0277
  520 Edgemont Road,   Charlottesville, Virginia       22903-2475 USA