• On-orbit dispersion solution must be made with celestial sources

• Pre-flight dispersion solutions were derived using emission line lamps with large number of lines.
• “Full” modeling of the dispersion requires polynomials of order >30.
• “Most” of the structure is taken care of by 4:th order polynomials

• Bright (point) sources (but not too bright; TTAG better)
• As many lines as possible, evenly spread over the “band-pass”
• Lines originating in a single location (cloud)
• Simple intrinsic line structure

• Yields very different results from ISM lines

• H₂ and HI show discrepancies in v_r
• H₂ and many neutral species show discrepancies in v_r
• Do high-J H₂ lines agree in v_r with low-J lines?

• With information about the velocity-ionization structure
• With enough H₂, to get strong lines, but not too much to saturate many levels
• “Long wavelength” lines of coexisting low ionization species:
  • Use H₂, Fe II & P II
• Ar I, N I, Si II all seem to have different vr from H₂
• Preferably TTAG data (to allow spectral shift compensation)
• e.g. sk180

• D v_rel<5km s^-1 over much of the band-pass.
• Locally (much) worse
• Significant edge effects
  • Caveat Emptor!
• We are working on improvements
  • It is possible that we could complement the on-orbit solution with the higher order terms from the ground based solution.
  • Manpower constrained

Return to the FUSE Data Workshop page.