LBNL-2016: Foregrounds for Inflation and Lensing

From CMB-S4 wiki
Jump to: navigation, search

The purpose of this session is to discuss the astrophysical foregrounds that could bias or contaminate measurements for inflation and lensing.

Those agreeing to present should not prepare full-blown talks, but rather a slide or two with some ideas that will foment discussion.


Foregrounds - Inflation:

  • Foreground residuals: how do we model those at this moment?
    • Josquin
  • Astrophysics: What do we know today and what remains uncertain about the diffuse galactic foregrounds?
  • Foreground modeling:
    • Alex, Ben and Flavien

[ for discussion: ]

  • Forecast validation: What is the role of map-level simulations and data challenges in validating forecasting efforts, which presumably will use various approaches to component separation?
    • AGREED: all sigma(r) forecasting needs to explicitly include FG separation, AND needs to be validated with map-space sims using diffuse FG sky models.
      • Both spectral and pixel-space component separation can be Fisherized, but BOTH can also be validated with map-space sims and biases from residuals quantified.

  • Band Selection: What is the plan to study how FG complexity feeds into band selection, and may indicate different optimizations for different kinds of S4 survey.
    • CHARGE FOR TOMORROW's sky modeling session: We need an ensemble of FG sky models (multiple realizations, several stages of complexity)
      • this ensemble needs to be agreed as a standard test set for various sigma(r) forecast efforts
      • need first set by ~May
      • we can expect future versions, updated with more sophisticated physics and better (S3) data

Kam: What format? Maps every 10 GHz? Or tools for which bandpasses can be specified?

Charles: realistic BPs must be included.

Lyman: use Planck as worked example of what level of FG residual did they stop using for parameter estimation.

Foregrounds - Lensing:

9:45 Alexander van Engelen - how foregrounds bias lensing measurements (5+5)

9:55 Blake Sherwin - quantifying and mitigating foreground bias (5+5)

Plenary Summary