[METIS] [other instruments] missing functionality in Scopesim

[oczoske]

Action item from the weekly A*/METIS videocon, 13 September 2023, following a request from the METIS project office for a development schedule for the "fully finished" simulator.
These are some issues that ought to be implemented before Scopesim can be considered a fully functional simulator for METIS (and other instruments). I include some very vague ideas for implementation, which may or may not work.

• Sequences of exposures: Observing blocks for METIS typically involve sequences of exposures (dithered LM images, chop-nodded N images, LMS dithered and rotated exposures,...). These sequences need to take into account either field rotation or pupil (PSF) rotation from one exposure to the next. In principle this can always be done using scripts that modify sources and parameters, effectively creating each exposure with an independent Scopesim simulation; however, this requires a lot of effort from the user.

  • implement field rotation and field shift through Scopesim parameters, maybe effective when creating the FieldOfView object from the Source
  • implement PSF rotation, probably as part of the PSF effects
  • implement a wrapper (replacement or extension of OpticalTrain.observe()?) that runs a sequence of observations with the necessary parameter changes from one exposure to the next.

• Differential atmospheric refraction: Some work has been done, but I don't think this is fully functional yet. The effect should take a DAR model as input, which can be the full atmospheric effect or the residuals from METIS' fixed ADCs.

• LMS (nominal) is nearly finished, except for an unsolved issue connected to the FieldOfView (TestObserveSpectroscopyMode::test_runs in test_basic_instrument.py fails). Should be tested for flux throughput.

• LMS (extended) is missing. Can only be implemented once spectral mapping model is available from UK/ATC. The complexity of the implementation depends on the format of that model.