Make synthetic interferometer

[eldond]

1. Define three points along interferometer line: source, reflection point, and end point. The source and end points are often the same, but not always.
2. Integrate the electron density along the interferometer path.
3. Determine the path length in plasma (usually just using the distance between intersections of the laser with the separatrix)
4. average density = integrated density / path length
   • Don't forget that the path length is for both line segments (or the same line segment twice if start and end points are the same)

Work consists of

• [x] Define interferometer instrument function. For this, perhaps a json format OMAS dd instance on disk should be checked in. The setup data for most diagnostics should be minimal.
• [x] Do the interferometer calculation and store results in the appropriate place in the data dictionary.
• [x] Provide a plot recipe for overlaying interferometer instrument function vs R-Z; it should be compatible with 2D SOLPS mesh and quantity plots.
• [x] Provide a plot recipe for plotting interferometer signals vs time

SynthDiag list on SPARC sharepoint

Requires

• https://github.com/ProjectTorreyPines/SD4SOLPS.jl/issues/1
• https://github.com/ProjectTorreyPines/SD4SOLPS.jl/issues/2

[anchal-physics]

Questions:

a) Is the path length in 4. same as the defined "path length in plasma" in 3.? b) If yes, does that mean that we average over a shorter path length even though interferometer integrates over the whole distance (far SOL, SOL, core, and back) because most contribution is assumed to come from the core?

[eldond]

a) yes b) if there were significant density in the SOL, assuming the path length is core-only would produce an error. The truth is the line integral. Converting it into an average is convenient to look at because the units become compatible with other density measurements. It's usually fine. If you happen to think of a clever way of checking for significant error in the path length assumption (maybe comparisons of chords in different locations?), we could add that.

[eldond]

Actually, my earlier comment is based on experience with interferometers that don't pass through the divertor. I wonder what's being done with newer designs that do go through the divertor. The WEST tokamak has such an interferometer chord, for example.

[anchal-physics]

Finished with https://github.com/ProjectTorreyPines/SynthDiag.jl/pull/9 and https://github.com/ProjectTorreyPines/GGDUtils.jl/pull/30