CAMB: late-time ISW contribution to TE and EE spectra?

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Graeme Addison
Posts: 34
Joined: July 17 2014
Affiliation: Johns Hopkins University

CAMB: late-time ISW contribution to TE and EE spectra?

Post by Graeme Addison » July 17 2014

Hi,

I edited equations.f90 to 'turn off' the late-time ISW effect by setting ISW = 0 if z < 30. I see a suppression of power in TT for l < 30 or so, as expected, but the TE power is tens of per cent higher at the largest scales, while there is no change at all to the EE power (the EE column in the dat file is identical to the default equations.f90 case).

Is this working correctly? Naively I would expect a (probably very small) ISW EE contribution from evolution of potentials at reionization. And if the EE change really is negligible then I can't understand why there's any effect on TE...

I've been using the default params.ini parameters, flags etc. and looking at the lensedCls.dat files.

Thanks for any help.

Jason Dossett
Posts: 97
Joined: March 19 2010
Affiliation: The University of Texas at Dallas
Contact:

CAMB: late-time ISW contribution to TE and EE spectra?

Post by Jason Dossett » July 17 2014

That ISW term should not contribute to the EE spectrum as the EE spectrum is computed from sources(2) only. The ISW term in equations.f90 only contributes to sources(1) which is used in the computation of the TE and TT power spectrum.

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CAMB: late-time ISW contribution to TE and EE spectra?

Post by Antony Lewis » July 17 2014

There is a small effect of re-scattering of late-ISW-induced quadrupole at reionization, but it is absolutely tiny because reionization is very matter dominated when it starts, and the visibility drops quite rapidly to lower redshift when ISW starts to be more important. To see the effect you'd have to modify the Boltzmann hierarchy for the sources that contribute to E, not set just the temperature ISW = 0 when calculating the sources for the line of sight integrals.

Graeme Addison
Posts: 34
Joined: July 17 2014
Affiliation: Johns Hopkins University

CAMB: late-time ISW contribution to TE and EE spectra?

Post by Graeme Addison » July 17 2014

Thanks for the replies.

I'm afraid I still don't understand what is happening to TE. If I just set the temperature ISW = 0 then why would TE be affected?

Daniel Francisco Boriero
Posts: 16
Joined: March 09 2009
Affiliation: Bielefeld

CAMB: late-time ISW contribution to TE and EE spectra?

Post by Daniel Francisco Boriero » November 12 2014

Dear All,

I have a similar question, but for the TT itself. I set ISW=0 when z > 30 to extract the eISW, by subtracting the spectra without ISW from the complete spectra. However, the aparent contribution is much higher than what I expected.

Does the ISW parameter in CAMB contain other contribution than this effect? And is this procedure to extract the eISW possible?


Thanks,
Daniel

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CAMB: late-time ISW contribution to TE and EE spectra?

Post by Antony Lewis » November 13 2014

Changes in TE would be from the correlation of the E from reionization with the ISW part of T, i.e. large modes that contribute both to the quadrupole at reionization and the perturbations causing the ISW. cf. the lensing-E correlation explained in 1101.2234.

Early ISW is very correlated with the non-ISW signal (they are from the same potetial wells), which gives an effect substantially larger than the eISW-eISW auto spectrum.

Daniel Francisco Boriero
Posts: 16
Joined: March 09 2009
Affiliation: Bielefeld

CAMB: late-time ISW contribution to TE and EE spectra?

Post by Daniel Francisco Boriero » November 17 2014

Indeed, the subtracted spectrum with eISW cross-correlations is bigger (4x at l=100) than the eISW self-correlation. I got the self-correlation simply by keeping only the eISW contribution to the source and obtained the spectrum that is usually presented as the eISW.

Thanks for the explanation.

Post Reply