The Effect of Lensing on the Large-Scale CMB Anisotropy

Abstract

We first compare the CMB lensing model of Seljak (1996) with the empirical model of Lieu & Mittaz (2005) to determine if the latter approach implies a larger effect on the CMB power-spectrum. We find that the empirical model gives significantly higher results for the magnification dispersion, assuming standard cosmological parameters. However, when the empirical foreground model is modelled via correlation functions and used in the Seljak formalism, the agreement is considerably improved. Thus we conclude that the main difference may be in the assumed foregrounds. We then discuss a foreground mass distribution which gives a lensing dispersion which is constant with angle. In Seljak's formalism, we show this can lead to a smoothing of the CMB power-spectrum which may be able to move the first acoustic peak to smaller l, if the mass clustering amplitude is high enough. Evidence for such a high amplitude comes from the QSO magnification results of Myers et al (2003, 2005) who suggest that foreground galaxy groups may be more massive than expected, implying that Omega_m ~ 1 and strong galaxy anti-bias. We then show that an inflationary model with neither CDM nor a cosmological constant and that predicts a primordial first peak at l=330 may fit the first acoustic peak of the WMAP data. This fit may be regarded as somewhat contrived since it also requires high redshift reionisation at the upper limit of what is allowed by the WMAP results. But given the finely-tuned nature of the standard LCDM model, the contrivance may be small in comparison and certainly the effect of lensing and other foregrounds may still have a considerable influence on the cosmological interpretation of the CMB.