Adding higher res delensing "band"

From CMB-S4 wiki
Jump to: navigation, search

Mar 28 2017, Clem Pryke

The standard 01.00 maps. assume only low res (small aperture) data and include realizations where delensing is faked by turning down A_L.

Eventually we need to switch to doing real delensing. For small/large aperture hybrid experiment this would be done by extracting the lensing potential from the high res data.

In this posting I therefore play around with adding an additional 155GHz band with higher resolution. I left the white noise level the same and simply jacked the 1/f knee up to 200 (500 for T), and set the ellmin to 100. I arbitrarily set the beam size to 4 arcmin.

Due to the smaller field of view large aperture experiments can have a much smaller edge region to the map so I generate a relative hits map which is flat to r=22 with a 3deg taper:

20170328 fig1.gif

This leads to the noisy edge being much narrower:

20170328 fig2.gif

And the combined map is LCDM dominated nearly all over:

20170328 fig3.gif

Taking the spectra we see as expected: the beam rolloff is at higher ell, and the ell cutoff & ell knee are higher. (I actually took these spectra using the softer edge mask to reduce what looked like E/B mixing.)

20170328 fig4.gif