If you subtract the model image from the grism image, you see that the model is
frequently more spatially extended than the actual spectra, i.e. there is a
narrow core in the residual image.
Is this the result of parameters to the aXe IRAF tasks or is it coming from
somewhere deep within aXe?
I would think that it would get the profiles right if it's using the direct
image segmentation map, though perhaps there's a smoothing term somewhere when
it is translated to the grism exposures.
Original issue reported on code.google.com by gbrammer on 27 Sep 2010 at 3:05
Original issue reported on code.google.com by
gbrammer
on 27 Sep 2010 at 3:05