Catalog of Unidentified Cometary Emission Lines
Review - M.S.P. Kelley, 11 May 2015

1. The heliocentric distance is a relevant parameter that is missing from the files.

2. The data set needs some documentation.  Some recommended items to include:

  1. Please define Reference code (tell the user that this should be ignored), Information bulletin, etc.

  2. Please describe the method for the literature search and the criteria used to determine if a line is real and unidentified.  Include target selection and that molecular bands are also included (e.g., Wyckoff et al. 1999).  This is important for understanding the data completeness and level of fidelity.  For example, the PDS archived atlas of Cochran et al. 2002 lists 4055 unidentified lines for comet 122P/de Vico between 3830 and 10192 Å, but this data set lists 242 lines between 3862 and 4187.  Without documentation a user cannot determine the origin of the discrepancy.  The references searched span a wide range of time.  Documentation would also help the user understand if the unidentified lines are simply those that were unidentified at time, or if any currently understood lines have been removed from the lists.

  3. Why is SNR N/A for most?  Signal-to-noise ratio should always be applicable in this context.  If it is unknown, state that instead, and describe in the documentation why some are unknown.

  4. Similar to SNR: Why are some resolving powers == N/A?  Resolving power should always be applicable to these spectra.  If it is unknown, state that instead, and describe in the documentation why they are unknown.  Can no estimate be made?  R~1000 and ~100,000 are very different and indicate the difference between an unidentified line and an unidentified band.

PDS4 questions:

3. Some of the line lists are in the UV and near-IR.  Should the collection also include these wavelength ranges in the Primary_Result_Summary?  If so, then the individual labels should include their own wavelength_range to facilitate searching.

4. Are PDS4 units case sensitive?  angstrom vs. Angstrom

5. Are XML entities OK?  & found in the labels (within literature references).