Agenda/Notes from Systematics WG telecon July 26 2017

on the call: Maurizio, Paolo, Brendan, Colin, Ranajoy, Shaul

1. Review CORE systematics paper (Paolo)
https://arxiv.org/pdf/1707.04224.pdf
  * Used the LBL TOAST tools (Kisner, Keskitalo, Borrill et al): flexible tool for generating timestreams
  * simulated timelines fed into optimal mapmaking code (MADAM)
  * Section 4: Started with checking whether clean reconstruction of polarization can be done with a space mission scan strategy and importantly, no modulator (i.e. rotating half waveplate); Fig 3: white-noise-only simulations showing correlation matricies. 1e-2 reciprocal condition number (for IQU mixing matrix) set to be a limit: this tells you how hard it will be to reconstruct polarization: systematics make this reconstruction harder.
  * noise knee frequency looked at: 10-20mHz is nearly white noise for ell<10.  Even 50 mHz, when averaged over many detectors, is not that bad (Fig 10).
  * Section 5: Detector Crosstalk.  Generates bias in power spectrum, see Fig 11
  * Section 6: Systematics: bandpass mismatch leakage from dust: projecting out at the mapmaking level (see Fig 13 for simulation of the effect and its correction for a 4-detector).  Edges of a tophat varied.  No consideration of synchrotron or CO or other discrete lines.
  * Section 7: Asymmetric beams: corrected in two ways, real-space deconvolution and power-spectrum-level correction
    * Real space deconvolution works very well, assuming effectively infinite S/N
    * Harmonic space correction: QuickPol (also needs good S/N)
  * Section 8: Calibration pipeline
    * the CORE scan strategy, see Fig 19 for example, gives a much better dipole signal for calibration than the Planck scan strategy
    * Galaxy-related systematics, assuming no correction, sets a "worst case" 

2. Review Systematics List
https://zzz.physics.umn.edu/ipsig/preliminary_list_of_systematic_effects_to_consider
 * Everyone look at this list offline, we'll go through this in detail next week.