Site Map
Contacts
Follow us on Facebook Follow us on Twitter YouTube channel
Centro de Astrofísica da Universidade do Porto
22-26 August 2011, Porto, Portugal

 

Precision Cosmology with Weak Gravitational Lensing
Hearin, Andrew (University of Pittsburgh)

Forthcoming projects such as DES, LSST, Euclid, and WFIRST aim to measure weak lensing shear correlations with unprecedented precision. These data sets have the potential to illuminate the nature of the cosmic acceleration by constraining the dark energy equation of state at the percent level and providing precision tests of the consistency of general relativity. We address several of the leading challenges to this program. Because the weak lensing signal is most sensitive to scales from 0.2-100 Mpc/h, percent-level uncertainty in predicting the matter power on nonlinear scales is a threatening source of systematic error. Another major source of uncertainty for these surveys derives from their reliance on photometrically determined redshifts. Cosmological parameter estimation depends quite sensitively on the redshift distribution of sources: current estimates are that both the bias and scatter of the photo-z distribution must be controlled to 0.003 or better in order to avoid degrading dark energy constraints by ~50% or more. Meanwhile, uncalibrated "catastrophic" redshift outliers must be controlled at a level of 10^-3 or less or their associated systematic errors will dominate the error budget. We present a new analysis of these stringent calibration requirements that addresses the interplay between photo-z uncertainty and errors in the calibration of the matter power spectrum. We find that including galaxy clustering statistics in a joint analysis with cosmic shear not only strengthens the survey's constraining power but can also have a profound impact on the calibration demands. For contemporary levels of uncertainty in the photo-z distribution of galaxies, a joint analysis can improve upon the cosmic shear constraints on w_0 by ~50%. We also study the potential to exploit the complementarity of these statistics to detect and calibrate the most damaging sources of systematic error. Finally, we compare these calibration requirements to the contemporary state-of-the-art in predictions of the lensing power spectrum and photometric redshift estimation and suggest strategies to utilize forthcoming data optimally.