[daip] Equal-area projections in AIPS?
Ben Weiner
bjw at ociw.edu
Thu Apr 27 15:19:30 EDT 2000
Hello,
I am trying to produce a 2-D plot of an image
in Galactic longitude and latitude with a
reasonable (close to equal-area) projection.
I have a FITS file where the coordinates are
longitude (l) and latitude (b), I read it into
AIPS with IMLOD, and I make the contour with
GREYS. The image covers the entire sky with
0.5 degree pixels.
I have two difficulties: (1) the FITS file is not generated
with AIPS and so I need to twiddle the header to make
AIPS understand the transformation from (x,y) to (l,b)
and that the coordinates are longitude and latitude;
(2) I want to plot a large area, about 30x30 degrees,
and the deviations from a rectilinear (x,y) <-> (l,b)
are large.
I think I have problem 1 fixed; if I set the FITS header
fields 'CTYPE1' and 'CTYPE2' to 'GLON' and 'GLAT', and
give the transformation between pixel values and (l,b)
with 'CRVAL1', 'CRPIX1', and 'CRDELT1', then GREYS
labels the coordinate axes correctly. However,
it produces a rectilinear plot, i.e. lines of longitude
and latitude are all perpendicular, and the distortions
are large over a 30x30 degree area.
If I tweak the header fields to include a projection,
e.g. 'GLON-TAN' or -SIN or -ARC, then GREYS reports
a lot of error messages of the form:
waffle> GREYS1: TICINC: POSITION ROUTINES FAIL ON THIS IMAGE
waffle> GREYS1: NEWPOS: ANGLE IS TOO LARGE. L,M= -0.334E+01 0.823E+00
I think these projections are inappropriate for the data
anyway. An attempt to regrid the data into Aitoff using
REGRD also fails with similar errors.
Do you know how to do this in AIPS, or failing that, of
a plotting package that can do it (either standalone or
Fortran subroutine?) I think IDL will do it, but want to
put off the overhead of learning IDL.
Thanks very much,
Ben Weiner
____________________________________________________
Benjamin Weiner Carnegie Observatories
bjw at ociw.edu 813 Santa Barbara St
phone: 626-304-0274 fax: 626-795-8136_________________Pasadena, CA 91101
More information about the Daip
mailing list