ALMA-IMF Core Working Group, April 22 2021

Connection:

https://zoom.us/j/96841488585?pwd=dURySjB2ZHQ1Z2lnL1FDclJWaTd4QT09
ID de réunion : 968 4148 8585
Code secret : Zw11YV

Google doc :
https://docs.google.com/document/d/1RPZPeDeEo1HrBgN_avnoAnH4OOLOHJ0RZfH2BBPuqrs/edit


Attendees: Sasha, fab,Yohan, Fabrice, Nathalie, Sylvain, Nicolás, Maxime, Thomas, Allison
Not attending: Adam (at CMZOOM), Gemma, Fred

For memory:
Minutes from our last telecon
Activities of the Core Working Group
Members and their main interests

Agenda:

Debrief on the ALMA-IMF workshop (22 to 26 of March)
Mélanie: G012 overviewResult: Gext2D gives higher fluxes than the getsf catalog. Gext2D gives a flatter CMF.
But: Free-free sources were not excluded(?)
Sylvain: A priori they (free-free sources) are removed in the GExt2D source list I sent to you. I remove all sources with Fl_1mm/Fl_3mm < 3.5. This is why sources on the filament seem undetected by GExt2D.
Here is the Fl_1mm vs Fl_3mm for GExt2D all sources for G012.80:


Future goal: look for outflows, especially in the HII region
Friedrich question: Why not using cm maps? Glostar has a too coarse angular resolution. Recombination line are indeed less sensitive.  (Adam’s note: VLA followup data with appropriate resolution, UV coverage are being reduced by UF undergraduate students now)
Pierre D : the difference of the CMF slope between the two extraction is not so important?

Sasha: The differences in CMF slopes (if significant) between the tools are very important for ALMA-IMF and the reasons must be clearly understood. Clearly, some properties of the codes are different. Yohan has the benchmark results for simalma50 and simalma240 for both codes and they must be analyzed in terms of the ratios A/A_true, I/I_true, F/F_true as a function of the S/N ratio I/sigma, as well as a function of the true source size A_true. Such scatter plots are very instructive and I use them a lot in my benchmarking paper. They will help us understand the reasons for the differences in the fluxes and in CMFs. Without such plots and analysis, the reasons cannot be established and this would be very bad, because no clear conclusion would be possible from the CMF work.

Bump in the CMF: denoising might help here, as it permits to extract cores deeper.
Sasha: the difference between the fits of getsf and getext2d are small (1.06 versus 1.17)? Did you try to keep only the common sources?  And for GExt2D, try to include sources even with bad flx_qu because they are detected.

Yohan: concerning the ratio 1/3mm, I recently updated the code to plot ratio as function of S/N to define the most reliable sources to remove. I can re-run G012 to produce this figure or send you the piece of code to do it yourself.

1mm/3mm: permits to separate the free-free contaminated and thermal sources quite efficiently ; perhaps not so efficient when 3mm is hardly detected. Sylvain made tests measuring 3mm fluxes down to 2\sigma -> this is a correct approach if there is a strong 1mm detection at this exact spot (Karl)

Open questions

1. Who wants to share submited ALMA project?

ALMA Proposal (PI: Nichol Cunningham): two mosaics at high resolution (100 AU) for the central regions of NGC2264 to see the fragmentation cascade (at different scales from the clump size down to the core size)
Fab: none as PI but I was involved in a couple of proposals toward outflows of Class II protostars (trying to know if MHD disk winds are a generic/universal process). One additional proposal toward a massive clump forming at the junction of two large finalements (testing conveyor belt scenario).

1. Status of the continuum data?

Last data reduction telecon suggested to re-run again continuum maps including the latest re-calibrations done by Luke. It is important to know if we will have a new data release (especially true for core extractions)
Adam: New maps are not expected to change any features of the  continuum data.  I recommend sticking with the current release.

1. getsf corrections for the sizes and fluxes of the unresolved sources with an overestimated background

if you overestimate the background, the associated component that getsf will substract from the source will be wrong. You will lose some faint parts (the “wings”) leading to a core size smaller than expected. Lot of tests done in benchmarks by Sasha and Yohan suggests that these corrections are improving measurements and mass estimation are closer to the true values.