search

rwesson / NEAT

Empirical analysis of ionised nebulae, with uncertainty propagation
GNU General Public License v3.0
9 stars 5 forks source link

Line blending #15

Closed rwesson closed 8 years ago

rwesson commented 12 years ago

Means of accounting for line blending should be incorporated. It might be hellishly difficult. At the moment we allow a small number of common CEL blends to be accounted for, but we do not have any way to deal with weak ORL blends, of which there can be many.

rwesson commented 9 years ago

Recent progress on this - the code reads in line lists with blends, and flags them such that the flux of the blend is dereddened, but no blended lines are used in diagnostic or abundance calculations.

970eb458af7cb5056736d60efb83d2d1b74499b4 3b21c59e3031d33e479f48ebc62148babf9fef93

Further improvements would be:

  1. checking for blends where abundance can still be calculated, eg OII 4275-4277 3d-4f transitions.
  2. deblending with relative intensities predicted from other lines. that might be too much hassle in the end to do well and maintain speed.
rwesson commented 8 years ago

The code should also check things like OII 7320, if one line is in the line list and the other is not, then most likely it's an unresolved blend of the two lines, and NEAT should be given the average wavelength to properly treat the blend.

rwesson commented 8 years ago

The CEL blends that NEAT recognises are now checked for when the line list is read in, with the wavelength being rewritten as necessary. Further work on deblending using predicted relative intensities etc looks likely to be unjustifiably time consuming so I'll close this.

6a4f6ace4dbe80162dc4f7ba877c1e35e4dd1b90