Joint Cosmic Shear Measurements with the Keck and William Herschel Telescopes

TL;DR

This work presents joint cosmic shear measurements from Keck II/ESI and the William Herschel Telescope to map the matter distribution via weak gravitational lensing. By building two independent datasets with careful PSF corrections and comprehensive error covariances, the authors extract shear correlation functions and constrain ΛCDM parameters, emphasizing the importance of sample variance and cross-instrument systematics. The combined analysis yields σ8(Ωm/0.3)0.68 = 0.97 ± 0.13, supporting the older cluster-normalisation and offering a cross-check against other cosmic shear surveys. Overall, the study demonstrates the power of multi-telescope, multi-field cosmic shear to robustly measure the amplitude of mass fluctuations while highlighting tensions with recent cluster normalisations that warrant further investigation.

Abstract

The recent measurements of weak lensing by large-scale structure present significant new opportunities for studies of the matter distribution in the universe. Here, we present a new cosmic shear survey carried out with the Echelle Spectrograph and Imager on the Keck II telescope. This covers a total of 0.6 square degrees in 173 fields probing independent lines of sight, hence minimising the impact of sample variance. We also extend our measurements of cosmic shear with the William Herschel Telescope (Bacon, Refregier & Ellis 2000) to a survey area of 1 square degree. The joint measurements with two independent telescopes allow us to assess the impact of instrument-specific systematics, one of the major difficulties in cosmic shear measurements. For both surveys, we carefully account for effects such as smearing by the point spread function and shearing due to telescope optics. We find negligible residuals in both cases and recover mutually consistent cosmic shear signals, significant at the 5.1 sigma level. We present a simple method to compute the statistical error in the shear correlation function, including non-gaussian sample variance and the covariance between different angular bins. We measure shear correlation functions for all fields and use these to ascertain the amplitude of the matter power spectrum, finding sigma_8 (Omega_m/0.3)^0.68 = 0.97 \pm 0.13 with 0.14<Omega_m<0.65 in a Lambda-CDM model with Gamma=0.21. These 68% CL uncertainties include sample variance, statistical noise, redshift uncertainty, and the error in the shear measurement method. The results from our two independent surveys are both consistent with measurements of cosmic shear from other groups. We discuss how our results compare with current normalisation from cluster abundance.

Figures (17)

• Figure 1: The sky location of the cosmic shear fields. Galactic latitudes of $0^\circ$, $\pm 20^\circ$, $\pm 40^\circ$, $\pm 60^\circ$ are shown as contours; the Galactic centre and poles are shown as a cross.
• Figure 2: Histogram of seeing (FWHM) in all survey shear fields after stacking dithered exposures. Keck data are shown in white; WHT data are shown as shaded. Each $8' \times 16'$ WHT chip is counted separately. Combining survey area from the two telescopes, the median seeing is 0.73"; no data with seeing worse than 1" is used.
• Figure 3: Median reddening-corrected $R$-band magnitude of the galaxies in each field that are used in the final object catalogue. The median depth of the combined surveys is $R$=24.0.
• Figure 4: Example instrumental shear pattern for ESI. Each bar represents the magnitude and orientation of astrometric distortions averaged over 20 sets of three dithered exposures. The illuminated area of the CCD is rotated from the $x-y$ axes by $7^\circ$, accounting for the occupied bins seen here.
• Figure 5: Example stellar ellipticity pattern for ESI. (Left) Before correction; the pattern varies qualitatively between fields but mean ellipticity here is typical of the field-to-field average of 0.035. (Right) After correction; mean ellipticity reduced to $<$0.001 across the survey, with residual orientations essentially randomised.