[daip] Spectral Index maps

[Eric Greisen]

Mercedes Esteves Filho writes:
 > Hi. Ive data from the VLBA and would like to construct
 > spectral index maps. Now Ive tapered and convolved the
 > data to the same beam. Image size and increment are the
 > same for both images. So added and divided images using COMB
 > look ok, but the spectral index maps dont look ok.
 > 
 > I guess that this will have to do with my choices
 > for aparm and bparm...But could you help me
 > out with this?
 > 

I am not sure what to suggest.  There is a task SPIXR if you have more
than 2 images for the spectral index determination.  In COMB, for
APARM you want 1,0,c1,c2 where c1 and c2 are cutoff levels below which
the spectral inde3x in blanked.  These should be at least 1 sigma and
perhaps 2 or 3 sigma.  Use IMEAN to determine sigma.

BPARM offers alternatives for limiting the computation - put the
correct sigmas from IMEAN (rms from fitting histogram) into BPARM of 1
and 2, bparm(3)=0, bparm(4)=1, bparm(5)=x where x is the maximum
uncertainty in spix that you can tolerate.

Eric Greisen