[ldd-spectral] Support multi-order derived spectral data products

[mkelley]

Enhancement

I don't have the ability to assign a label, but it should be "enhancement."

The issue identified

A data provider is preparing a collection based on data products derived from cross-dispersed echelle spectral observations. The derived data products are 2D arrays, one dimension is spectral, the other is spectral order. This appears to be describable with an Array_2D_Spectrum class. However, the group facets under Primary_Result_Summary do not seem directly applicable:

• 2D (Spectroscopy) - seems to be better suited for spectral-spatial data: "each pixel value directly represents the spectral measurement at that point." The concept of "at that point" is not clear, but to me it implies a spatial relationship.
• Spectral Image (Spectroscopy) - although multiple orders may be present in data described by this facet, it is specifically for focal plane images.

A proposed solution

• Adding a new group facet, something to the effect of "Multi-order spectrum."
• Clarifying that 2D (Spectroscopy) is for spatial-spectral data.

These may or may not have consequences on the Spectral DD.

Alternatives considered

The alternative seems to be to not describe these data with Group_Facet1, which likely has an impact on discoverability.

PDS4 IM Version

1.24

Need-by Date

First half of 2025 to facilitate archiving of these data.