Add ATR isotope data

[observingClouds]

Hi @leonie-villiger, Thanks for the PR! Great to see more datasets joining! 🥳 I committed a few changes to match the structure of the other airplane data. I think what is now left to do is:

• [x] add the additional flight days in ATR/isotope-analyzer-CRDS.yaml (@leonie-villiger could you do this?)
• [x] check if I used the correct instrument name (@leonie-villiger was it a CRDS on the ATR?) to match the syntax discussed in #153
• [ ] discuss how the metadata/flight segment data should be added I removed your initial indexing of the metadata because I realised that this metadata is following the flight segmentation syntax. I added the url to the metadata for now as an argument to the datasets metadata. However, I think we want to add this metadata to the flight-segmentation repository. Could the experts @Smpljack, @d70-t, @leifdenby comment on this and help us eventually on-board the yaml-files in https://observations.ipsl.fr/aeris/eurec4a-data/AIRCRAFT/ATR/vapour/?

[Smpljack]

Hi @leonie-villiger and @observingClouds , it has been some time for me to be involved in flight-segmentation issues, so let me know if I am following along well. I understand that Hauke's question basically is whether the ATR metadata files should be integrated to the flight-phase-separation repo, correct?

This would seem consistent from my point of view, the format mostly accommodates the HALO/P3 standards already (On first sight, I see that we call the segment kind rather "kinds" and allow several entries...). A small readme could be added to explain new types of segment kinds in the style of P3-segmentation-notes.md.

Then there are additional isotope meta data files (isotope-analyzer-CRDS.yaml). I understand that these should be kept independent of the flight segment files since they contain a lot of arguments that don't contain flight segment information. Maybe the description argument could be added to the remarks for each flight of the ATR metadata files.

How about opening a PR to the flight-phase-separation repo with the addition of the ATR metadata files to iterate on the details?

[observingClouds]

@Smpljack thanks for your response! This is already very helpful.

I understand that Hauke's question basically is whether the ATR metadata files should be integrated to the flight-phase-separation repo, correct?

Correct!

This would seem consistent from my point of view, the format mostly accommodates the HALO/P3 standards already (On first sight, I see that we call the segment kind rather "kinds" and allow several entries...). A small readme could be added to explain new types of segment kinds in the style of P3-segmentation-notes.md.

Ah, the kind <> kinds is a shame, but lets talk about this and the addition of new definitions over in the flight-segmentation repo.

Then there are additional isotope meta data files (isotope-analyzer-CRDS.yaml). I understand that these should be kept independent of the flight segment files since they contain a lot of arguments that don't contain flight segment information. Maybe the description argument could be added to the remarks for each flight of the ATR metadata files.

The (isotope-analyzer-CRDS.yaml) is currently only referencing the ATR metadata files. Because these ATR metadata files are (almost) flight segmentation files, they seem to be of much better use if they are handled like that, i.e. be accessible through the flight-segmentation "API"

[leonie-villiger]

@observingClouds:

I added the additional flights in ATR/isotope-analyzer-CRDS.yaml. Had to add the flight number to the identifier, because we sometimes had two flights on one day (ATR-0125 now reads ATR-0125-02).

Yes, it was a CRDS instrument on the ATR. All correct!

@observingClouds and @Smpljack There are two types of metadata:

1. ATR metadata files (isotopes) belonging to the isotope measurements, currently referenced in the ATR/isotope-analyzer-CRDS.yaml as note. This metadata is only relevant for the isotope data from the ATR.
2. ATR metadata files (flight segments). This is the official flight segmentation for all the ATR data (see description).

I assume it is rather number 2 which you want to include in the flight-phase-separation repository...

Also, when adding ATR metadata files (flight segments) it would make sense to add the ATR tracks (part of the SAFIRE-CORE data) somewhere. An overview of all ATR dataset is provided in Bony et al. (2022).

[observingClouds]

Alright, I think we can merge this for now and talk about the flight segmentation separately.

[observingClouds]

The integration of ATR flight segments is discussed at https://github.com/eurec4a/flight-phase-separation/issues/32