Do we need CO simulations?

simonsobs / map_based_simulations

Map based simulations for the Simons Observatory

4 stars 1 forks source link

Do we need CO simulations? #23

Closed zonca closed 2 years ago

zonca commented 5 years ago

The CO component is available in so_pysm_models, see the documentation, it has been implemented by @giuspugl.

We need to check more realistic bandpasses and see if we really need to add CO to our simulations. @giuspugl would you volunteer to do this?

giuspugl commented 5 years ago

sure can do that .

zonca commented 5 years ago

@giuspugl would you have any update on this?

zonca commented 5 years ago

@giuspugl we have some realistic bandpasses, but @smsimon suggested to wait for updated release of this in September. Let's wait for that. Unless someone requests urgently CO maps, then we can do a preliminary release with what is available now.

msyriac commented 5 years ago

Sara Simon @smsimon has alerted me to existence of updated bandpass models here: https://github.com/smsimon/bandpasses/ . She tells me these will be incorporated in to sotodlib, but one could fetch them from there and use them right away.

zonca commented 5 years ago

@msyriac @giuspugl I don't see that repo, is it private?

msyriac commented 5 years ago

Yes, it's private. I'll ask Sara to make a team or add us to an existing team that gives us access.

smsimon commented 5 years ago

@zonca it is here now: https://github.com/simonsobs/instrument_model. We decided to make a general place where instrument modeling things can go and people/sotodlib can access them.

zonca commented 4 years ago

@giuspugl would you like to check which CO lines impact which channel using those bandpasses?

giuspugl commented 4 years ago

@smsimon for CO studies shall I consider the Bandpassess without any beam effect (i.e. files with just On-chip bandpass filters + Gain effects?

zonca commented 4 years ago

@smsimon @giuspugl any update on this?

giuspugl commented 4 years ago

Yes, working on that.

giuspugl commented 4 years ago

CO lines with realistic bandpasses

State of the art bandpasses to date can be found in the instrument model repo, as pointed out by @smsimon.

LAT

I considered bandpasses from the files MF_wOMT_wLPF_v1_LAT_beam.txt and UHF_wOMT_LAT_beam.txt

SAT

I considered bandpasses from the files MF_wOMT_wLPF_v1_SAT_beam.txt and UHF_wOMT_SAT_beam.txt

Summary

Bandpasses for MF are designed in such a way that the CO 1-0 rotational line is pretty much avoided (transmission at the line frequency, i.e. 115.3 GHz is 1e-4 and 1e-3 respectively for LAT and SAT channels. This also the case for CO3-2 line (at 345.8 GHz) whose transmission is at around 1e-3 for both LAT and SAT
Conversely, the CO2-1 line (at 230.5 GHz) is essentially transmitted at 80 % in the 220 GHz channel for both LAT and SAT.
Therefore, we should include CO2-1 emission in our map based simulations

zonca commented 2 years ago

in the plan for #39