Attendance: Clem, Jacques, Charles, Graca, Brandon, Susan Clark. Mathieu,
Notes
Soumen's latest presentation on NILC results. Results look good for 00-03 (Gaussian and PySM's), 96 marginal (MHD), 99 is quite bad (Vansyngel). AI's for Soumen:
Run on some more realizations.
Try varying the mask area.
We can also take the delta on reextracted r CMB only versus fully combined on a realization-by-realization basis (to take out the sample variance on r). Soumen will also try varying the mask.
Graca update:
Soumen NILC maps are B-mode only, Graca needs IQU, has emailed Soumen who will provide.
Also has Belen's SMICA maps (which are IQU), had permissions problem but fixed as of last night.
Will run r estimation on both component separated map and also tensor only input. Taking delta cancels sample variance on tensors making the residual foreground bias clearer.
Clem will take an action to post dust decorrelation plots for each model.
Discussion of which models to put in report. Agreed that they should all go in unless they can be shown to be inconsistent with current data (or totally crazy from an astrophysics point of view).
Clem reports that files containing Jacques et al model are nearly ready to go - will upload to NERSC this afternoon.
Report:
Brandon has provided a paragraph on MHD model.
Jacques has started writing but didn't get access to git repository yet.
Soumen will send short description of NILC to Jacques.
Mathieu will send a paragraph about situation with COMMANDER to Jacques.
Clem will put in text from CDT report on PySM models etc