RIX LCc : Difference between x_LSS_RSRF_PINH_RAW and x_DISTORTION_RAW

[astronomyk]

https://jira.eso.org/browse/MET-2053

What is the difference between x_LSS_RSRF_PINH_RAW (used by metis_x_lss_trace) and x_DISTORTION_RAW (used by metis_x_img_distortion). Can one calibration type fulfil the purposes of both recipes, or do we really need to take both?

@wkausch I assigned you, as you were the first to come to my mind. But feel free to pass this on to someone else

@wkausch @ivh @hugobuddel @sesquideus - why (or why not) is this a valid question? Maybe my understanding of the distortion maps is wrong, but the RSRF is a wavelength and field dependent map that is used for extracting the traces, while the IMG distortion map is used to correct field distortion over the full imaging FOV. My gut feeling says that because they have two seperate purposes, they are not the same data type. Is this a wrong assumption?

Thanks for your opinions!

[wkausch]

Can't really answer that because I don't know what the x_DISTORTION_RAW is....

The x_LSS_RSRF_PINH_RAW are observations where the slit is replaced by the pinhole mask, we therefore get traces at the position of the pinholes with the flatfield lamp.

[hugobuddel]

It seems x_LSS_RSRF_PINH_RAW and x_DISTORTION_RAW are indeed similar. @wkausch names the data product based on how it is made, and I name it based on what it is for.

@wkausch I must admit that I don't fully understand how the slits and the grisms work together to produce the traces. You say that the slit is replaced with the pinhole mask and that still gives you the traces?

The METIS_spec_lm_cal_rsrfpinh does not exist yet (this is known), but I looked up METIS_spec_lm_cal_rsrf, and it seems to also have a pinhole mask:

A grid is inserted in the optical path of the instrument. The grid is illuminated with the black body source, simulating a set of point sources at various positions of the detector.
https://metis.strw.leidenuniv.nl/wiki/doku.php?id=operations:metis_spec_lm_cal_rsrf

@wkausch what would be the difference between METIS_spec_lm_cal_rsrf and METIS_spec_lm_cal_rsrfpinh?

That aside, back to the RIX. We could potentially harmonize the names. I could use x_IMG_FLAT_PINH_RAW and IFU_FLAT_PINH_RAW. I should have added the IMG part in the first place.

Because the actual question is: "do we really need to take both?", yes, we do need to take all 5, because they are (SPEC, IMG) * (LM, N) + IFU. This would have been clear if I at least named them x_IMG_DISTORTION_RAW.

So I propose to reply:

x_LSS_RSRF_PINH_RAW and x_DISTORTION_RAW are indeed similar, however, they correspond to different observation modes:

• LM_LSS_RSRF_PINH_RAW: LM LSS
• N_LSS_RSRF_PINH_RAW: N LSS
• LM_DISTORTION_RAW: LM IMG
• N_DISTORTION_RAW: N IMG
• IFU_DISTORTION_RAW: IFU

It is therefore essential to observe all 5; they cannot be combined. We will however rename the LM_DISTORTION_RAW to LM_IMG_DISTORTION_RAW and rename N_DISTORTION_RAW to N_IMG_DISTORTION_RAW to clarify the difference.

What do you think @wkausch ?

[wkausch]

I agree we need both. We cannot do anything with an imaged pinhole mask, we need the traces from this pinholes. Concerning the naming: Yes, we should sync the naming convention, but can do that after FDR.

@your question: The slit "cuts out" a portion of the sky, usually a long rectangular format. This image is projected onto a grism in a way that the "long" side of the slit is in parallel to the etched lines in the grism. This is a fixed alignment creating a spectrum in direction the "short" side, resulting in an image which has a spatial direction along the "long" side and wavelength along the "short" side. If we replace the slit by pinholes, we do exactly the same, but only at the places for the pinholes. Therefore we see traces as the part between the pinholes is covered and no light is reaching the grism. YOu can imagine the pinhole mask as interrupted slit.

[hugobuddel]

Answered. I'll rename the IMG ones, because those should have IMG in the name. Keeping this open because I still have to do that renaming.