[1101.2720] Observational Scan Induced Artificial CMB Anisotropy

Authors:  Hao Liu, Ti-Pei Li
Abstract:  To reliably detect the cosmic microwave background (CMB) anisotropy is of great importance in understanding the birth and evolution of the Universe. One of the difficulties in CMB experiments is the domination of the measured CMB anisotropy maps by the Doppler dipole moment from the motion of the antenna relative to the CMB. For each measured temperature the expected dipole component has to be calculated separately and then subtracted from the date. A small error in dipole direction, antenna pointing direction, sidelobe pickup contamination, and/or timing synchronism, can raise significant deviation in the dipole cleaned CMB temperature. After a full-sky observation scan, the accumulated deviations will be structured with a pattern closely correlated to the observation pattern with artificial anisotropies on large scales, including artificial quadrupole, octopole, etc in the final CMB map. Such scan-induced anisotropies on large scales can be predicted by the true dipole moment and observational scan scheme. Indeed, the expected scan-induced quadrupole pattern of the WMAP mission is perfectly in agreement with the published WMAP quadrupole. With the scan strategy of the Planck mission, we predict that scan-induced anisotropies will also produce an artificially aligned quadrupole. The scan-induced anisotropy is a common problem for all sweep missions and, like the foreground emissions, has to be removed from observed maps. Without removing the scan-induced effect, CMB maps from COBE, WMAP, and Planck as well, are unreliable for studying the CMB anisotropy.
[PDF]  [PS]  [BibTex]  [Bookmark]

Discussion related to specific recent arXiv papers
Post Reply
Molly Swanson
Posts: 9
Joined: June 22 2009
Affiliation: University College London

[1101.2720] Observational Scan Induced Artificial CMB Aniso

Post by Molly Swanson » January 21 2011

Here's another contribution from Liu and Li on scan-induced artifacts in CMB maps, continuing on the discourse from here: http://cosmocoffee.info/viewtopic.php?t=1537

Here they derive a general way to express many possible observational effects (timing offsets as in their previous work, errors in determining the sidelobe response of the detector, etc) as an effective error in the antenna line of sight direction which is constant in the spacecraft coordinate frame. They also come up with a way to clean errors of this form from maps, and demonstrate that they can recover the true input quadrupole from a simulated map even with a large artificial quadrupole on top of it.

What I find especially interesting here is that they actually make some predictions for Planck, so this will help resolve the issue of what's actually going on. Although they are careful to say that they can't fully calculate what's expected from Planck before the Planck data is released: "Note that the exact templates of scan-induced anisotropies can be calculated when the real Planck data release is available. ... Our calculations are based on the simulated Planck scanning scheme with assumed initial conditions of scan geometry. The real pattern of the scan-induced Planck quadrupole could hence be some kind of rotation or sign reversion to the ones shown here." Regardless, should be very interesting to revisit when the first Planck cosmological results appear!

Post Reply