MapBasedRb

From CMB-S4 wiki
Jump to: navigation, search

(David Alonso writing)

Summary

This is an updated version of the results shown in our previous post, where we looked at the r=0,\,f_{sky}=0.1 case in Victor's forecasts using our map-based component separation code on PySM simulations. Main changes with respect to our previous iteration:

  • Updated noise levels assuming Victor's levels are in polarization (instead of intensity).
  • Assumed a delensing factor A_L=0.18
  • Introduced correlated noise. For this we considered noise curves made up of a flat component with the quoted levels and a power-law component that starts dominating at some scale \ell_{\rm knee}. The power law index \alpha=-1.9 was determined from Victor's noise curves, and we explored \ell_{\rm knee}=0 (uncorrelated noise), \ell_{\rm knee}=50 and \ell_{\rm knee}=100.

The final numbers agree qualitatively with Victor's forecast:

  • \ell_{\rm knee}=0  \,\rightarrow\,\sigma(r)=0.62\times10^{-3}
  • \ell_{\rm knee}=50 \,\rightarrow\,\sigma(r)=0.94\times10^{-3}
  • \ell_{\rm knee}=100\,\rightarrow\,\sigma(r)=1.60\times10^{-3}

See our previous post for further details on the method.