Cosmology from cosmic shear power spectra with Subaru Hyper Suprime-Cam first-year data

TL;DR

This study analyzes cosmic shear in the Subaru HSC first-year data to constrain cosmology via tomographic power spectra measured with a pseudo-C_ℓ method. The authors carefully model systematics, including intrinsic alignments, photo-z uncertainties, PSF leakage, and baryonic effects, and validate their covariance with analytic halo-model predictions against realistic mock catalogs. They report a robust S8 constraint of 0.800^{+0.029}_{−0.028} (α=0.45) in ΛCDM and find general consistency with Planck within uncertainties, while exploring extensions to wCDM and joint analyses with external datasets. The work demonstrates the power of deep, high-density imaging (n_g,eff ~ 17 arcmin^−2) over wide areas to tighten cosmological constraints and sets the stage for the improved precision anticipated from the full HSC survey. The framework combines precise nuisance marginalization with tomographic leverage to mitigate degeneracies and provides a blueprint for future Stage III/IV lensing analyses.

Abstract

We measure cosmic weak lensing shear power spectra with the Subaru Hyper Suprime-Cam (HSC) survey first-year shear catalog covering 137deg$^2$ of the sky. Thanks to the high effective galaxy number density of $\sim$17 arcmin$^{-2}$ even after conservative cuts such as magnitude cut of $i<24.5$ and photometric redshift cut of $0.3\leq z \leq 1.5$, we obtain a high significance measurement of the cosmic shear power spectra in 4 tomographic redshift bins, achieving a total signal-to-noise ratio of 16 in the multipole range $300 \leq \ell \leq 1900$. We carefully account for various uncertainties in our analysis including the intrinsic alignment of galaxies, scatters and biases in photometric redshifts, residual uncertainties in the shear measurement, and modeling of the matter power spectrum. The accuracy of our power spectrum measurement method as well as our analytic model of the covariance matrix are tested against realistic mock shear catalogs. For a flat $Λ$ cold dark matter ($Λ$CDM) model, we find $S_8\equiv σ_8(Ω_{\rm m}/0.3)^α=0.800^{+0.029}_{-0.028}$ for $α=0.45$ ($S_8=0.780^{+0.030}_{-0.033}$ for $α=0.5$) from our HSC tomographic cosmic shear analysis alone. In comparison with Planck cosmic microwave background constraints, our results prefer slightly lower values of $S_8$, although metrics such as the Bayesian evidence ratio test do not show significant evidence for discordance between these results. We study the effect of possible additional systematic errors that are unaccounted in our fiducial cosmic shear analysis, and find that they can shift the best-fit values of $S_8$ by up to $\sim 0.6σ$ in both directions. The full HSC survey data will contain several times more area, and will lead to significantly improved cosmological constraints.

Figures (21)

• Figure 1: Tomographic cosmic shear power spectra of EE ( red filled circles), BB ( blue open triangles), and EB ( yellow crosses) modes. The galaxy samples are divided into four tomographic redshift bins using the Ephor AB photo-$z$ code. The redshift ranges of the four tomographic bins are set to [$0.3$, $0.6$], [$0.6$, $0.9$], [$0.9$, $1.2$], and [$1.2$,$1.5$], for binning number 1 to 4 (see also Table \ref{['tab:info']}). The right-bottom panel shows the non-tomographic cosmic shear power spectrum. The multipole ranges of $\ell<300$ and $\ell>1900$ ( shaded regions) are excluded in the cosmological analysis. The combined total detection significance of the tomographic EE-auto spectra is $16\sigma$ in the range of $300<\ell<1900$ ( unshaded regions), whereas both BB and EB-mode spectra are consistent with zero.
• Figure 2: The auto power spectrum of the PSF leakage $\tilde{\alpha}^2C_\ell^{pp}$ ( open triangles), that of the residual PSF $\tilde{\beta}^2C_\ell^{qq}$ ( filled squares), and their cross-spectrum $\tilde{\alpha}\tilde{\beta} |C_\ell^{pq}|$ ( crosses) are compared with the non-tomographic cosmic shear power spectrum $C^{\rm gg}_\ell$ ( filled circles) (see Section \ref{['subsec:resPSF']} for details). The shaded region indicates the range of multipoles that is excluded in our fiducial cosmological analysis. It is found that both of the PSF systematics are subdominant in the fiducial multipole range.
• Figure 3: Comparison of COSMOS-reweighted redshift distribution for the four tomographic bins with corresponding redshift distributions obtained by stacking the photo-$z$ PDFs using different photo-$z$ codes; Ephor AB (photo-$z$'s used to define the tomographic bins), MLZ, Mizuki, FRANKEN-Z, NNPZ, and DEmP. Shaded regions show our definition of 4 tomographic bins for best photo-$z$'s, i.e., $0.3<z<0.6$, $0.6<z<0.9$, $0.9<z<1.2$, and $1.2<z<1.5$. The shaded region around each curve in the top panel shows the statistical error of the COSMOS-reweighted redshift distributions estimated by the bootstrap resampling technique. In our cosmology analysis, we take account of these differences of redshift distributions $\bar{P}(z)$ using $z$-shift parameter $\Delta z$ [equation (\ref{['eq:deltaz']})]. We also discuss the impact of different $\bar{P}(z)$ on cosmological parameters in Section \ref{['subsec:robustness']}.
• Figure 4: Comparison of the measured tomographic shear power spectra with our theoretical model with best-fit values for the fiducial $\Lambda$CDM model. Best-fit IA power spectra of $C_{\rm GG}$ ( dotted), $-C_{\rm GI}$ ( short dashed), and $C_{\rm II}$ ( long dashed) as well as power spectra arising from PSF leakage and PSF model error [equation (\ref{['eq:cl_psfsys']})] ( dash-dotted) are also plotted. The redshift range of $z_{\rm best}$ in each tomographic bin is =$[0.3,0.6]$, $[0.6,0.9]$, $[0.9,1.2]$, and $[1.2,1.5]$ from 1 to 4. The right-bottom panel shows the measured non-tomographic cosmic shear power spectrum and the model spectra with the best-fit values from the tomographic analysis. The $C_{\rm II}$ term is so small that it is absent from all panels except for 11.
• Figure 5: Marginalized posterior contours in the $\Omega_{\rm m}$-$\sigma_8$ plane ( left) and in the $\Omega_{\rm m}$-$S_8(\alpha=0.45)$ plane ( right), where $S_8 (\alpha)\equiv \sigma_8 (\Omega_{\rm m}/0.3)^\alpha$, in the fiducial $\Lambda$CDM model. Both 68% and 95% credible levels are shown. For comparison, we plot cosmic shear results from KiDS-450 with correlation function (CF) estimators Hildebrandt17 and with quadratic estimators (QE) Kohlinger17 and DES Y1 Troxel18b with the same set of cosmological parameters and priors as adopted in this paper, as well as WMAP9Hinshaw13 ( yellow) and Planck 2015 CMB constraints without CMB lensing Planck16 ( purple).