The sky-line list provided at the NIRSPEC website is incomplete, comprised of OH lines only, missing prominent lines from other species. This incompleteness alone is not a problem, since a GP could catch the rest of the lines. The problem will arise when a known OH line is near an unknown line, and a better fit can arise by sliding the wavelengh solution towards the unknown line. We should consider making an improved line list, perhaps from fiatlux or some other spectrum generator. In pinciple these sky spectrum generators have pressure shifts that depend on the atmospheric properties, but in practice those shifts are small enough to ignore for our purposes.
The sky-line list provided at the NIRSPEC website is incomplete, comprised of OH lines only, missing prominent lines from other species. This incompleteness alone is not a problem, since a GP could catch the rest of the lines. The problem will arise when a known OH line is near an unknown line, and a better fit can arise by sliding the wavelengh solution towards the unknown line. We should consider making an improved line list, perhaps from fiatlux or some other spectrum generator. In pinciple these sky spectrum generators have pressure shifts that depend on the atmospheric properties, but in practice those shifts are small enough to ignore for our purposes.