1207.6322 (Issha Kayo et al.)

Information content of weak lensing bispectrum: including the non-Gaussian error covariance matrix    [PDF]

Issha Kayo, Masahiro Takada, Bhuvnesh Jain

We address a long-standing problem, how can we extract information in the non-Gaussian regime of weak lensing surveys, by accurate modeling of all relevant covariances between the power spectra and bispectra. We use 1000 ray-tracing simulation realizations for a Lambda-CDM model and an analytical halo model. We develop a formalism to describe the covariance matrices of power spectra and bispectra of all triangle configurations, which extend to 6-point correlation functions. We include a new contribution arising from coupling of the lensing Fourier modes with large-scale mass fluctuations on scales comparable with the survey region via halo bias theory, which we call the halo sample variance (HSV) effect. We show that the model predictions are in excellent agreement with the simulation results for the power spectrum and bispectrum covariances. The HSV effect gives a dominant contribution to the covariances at multipoles l > 10^3, which arise from massive halos with masses of about 10^14 solar masses and at relatively low redshifts z<0.4. Since such halos are easy to identify from a multi-color imaging survey, the effect can be estimated from the data. The bispectra add information to the power spectra, and increase the cumulative signal-to-noise up to a maximum multipole of a few 10^3 by up to 50%, which is equivalent to a factor of about 2 in survey area. However, the total information content of the power spectrum and bispectrum is still significantly smaller than that for the corresponding Gaussian field, mostly due to the HSV effect. Thus bispectrum measurements are useful for cosmology, but using information from upcoming surveys requires that non-Gaussian covariances are carefully estimated.

View original: http://arxiv.org/abs/1207.6322