atmospheric model

[thomasgas]

the atmospheric model that we have now is just the one for paranal (called atmprof26 in corsika&simtelarray).
We should add also the model for La Palma. Is it a problem if I take the atmospheric model for La Palma and I add it to this repository with the same format used for paranal?

[kosack]

no that should be ok for now. In the future, there will be different service data for each site, but until that system is in place, we can just put a few here. Of course ctapipe-extra was just supposed to be defaults (not all calibration and configuration data possible), and the users can specify their own files by adding directories to $CTAPIPE_SVC_PATH, but perhaps that could lead to confusion?

[maxnoe]

The new simtel monte carlos include the atmospheric profile used for the simulation

[thomasgas]

what does "new" means? like prod4?
the header will just have the name of the atmospheric model or also the model itself?
thanks!

[maxnoe]

There is a new eventio type (1216) that contains the full atmprof data. See cta-observatory/pyeventio#188

I think this is a matter of the version of the iact/atmo extension. 1.60 I think.

[maxnoe]

Yes: https://www.mpi-hd.mpg.de/hfm/~bernlohr/iact-atmo/ChangeLog

[kosack]

we still need an atmospheric model when analyzing real data, so perhaps we

1. switch to reading the atmosphere model from MCs in SimTelEventSource
2. improve ctapipe-dump-instrument to also write out the model in a nice table format (FITS or ECSV, using astropy.table)
3. include any useful models as tables here, so they can be loaded later for computing h_max, etc.

Soon, ctapipe-extra will be deprecated and replaced with just a file store (already partially working), an we can manage multiple sub-configurations (e.g. prod3/*, prod4/*), in which there are subarray descriptions, atmosphere profiles, etc.