APALL extras data

Frank Valdes wrote on Jul 25, 1997

Q:  Hi.  I'm working with a spectroscopic file.  It's an array of 50
columns and 512 lines.  I run it through APALL fitting a background to
it, and using a reference image rather than tracing (reference image is
a photometric standard star).  Anyway, after I extract the data, if I
observe it with IMPLOT or DISPLAY, it's my extracted spectrum, a 1 by
1024 array.  However, in SPLOT there are two bands.  In band one [*,1,1]
is my extracted spectrum.  We are wondering what is in band two [*,1,2].
It is obviously somwhat related to the extracted spectum, but
we can not figure out how.  Also, someone has mention that it is
possible for APALL to create an error array as it extracts each
spectrum.  I have not been able to find info on either of these, and
any light you could shed would be greatly appreciated.  Thank you . . .


A:  The parameter "extras" in APALL controls whether your output 1D spectra
have "associated" spectra in the higher dimensions of the image.  So
you must have this set to yes.  The associated data you get depends
on what options you use in APALL.  If you do background subtractions
you get the background spectrum subtracted.  If you use "variance"
weighting or cleanning you get the simple unweighted/uncleanned spectrum
in addition.  Finally if you use the variance weighting you get a
sigma array which gives the estimated sigma of the extracted pixels.
In recent versions you will see a description of what is what in
the header under the keywords BANDIDn.  The basic order (with some
missing if not selected) is

	1: always the primary spectrum
	2: the simple unweighted/uncleanned spectrum
	3: the background spectrum
	4: the sigma spectrum
	
This is described somewhat under the topics "extras" (relatively new
topic), "apextract.package", and "apall".

---

Last post on Jul 25, 1997