Measurement of the pairwise kinematic Sunyaev-Zeldovich effect with Planck and BOSS data

We present a new measurement of the kinetic Sunyaev-Zeldovich effect (kSZ) using Planck cosmic microwave background (CMB) and Baryon Oscillation Spectroscopic Survey (BOSS) data. Using the “LowZ North/South” galaxy catalogue from BOSS DR12, and the group catalogue from BOSS DR13, we evaluate the mean pairwise kSZ temperature associated with BOSS galaxies. We construct a “Central Galaxies Catalogue” (CGC) which consists of isolated galaxies from the original BOSS data set, and apply the aperture photometry (AP) filter to suppress the primary CMB contribution. By constructing a halo model to fit the pairwise kSZ function, we constrain the mean optical depth to be $\overline{\tau }=(0.53\pm 0.32)\times {10}^{-4}(1.65\sigma )$ for LowZ North CGC, $\overline{\tau }=(0.30\pm 0.57)\times {10}^{-4}(0.53\sigma )$ for LowZ South CGC, and $\overline{\tau }=(0.43\pm 0.28)\times {10}^{-4}(1.53\sigma )$ for DR13 Group. In addition, we vary the radius of the AP filter and find that the AP size of 7 arcmin gives the maximum detection for $\overline{\tau }$. We also investigate the dependence of the signal with halo mass and find $\overline{\tau }=(0.32\pm 0.36)\times {10}^{-4}(0.8\sigma )$ and $\overline{\tau }=(0.67\pm 0.46)\times {10}^{-4}(1.4\sigma )$ for DR13 Group with halo mass restricted to, respectively, less and greater than its median halo mass, $10^{12}{h}^{-1}{\mathrm{M}}_{\odot }$. For the LowZ North CGC sample restricted to $M_{\mathrm{h}}\gtrsim {10}^{14}{h}^{-1}{\mathrm{M}}_{\odot }$ there is no detection of the kSZ signal because these high mass halos are associated with the high-redshift galaxies of the LowZ North catalogue, which have limited contribution to the pairwise kSZ signals.

Figure 1

The redshift histogram for each catalogue. The total number of galaxies in each catalogue is shown in the legend. The DR7 CGC catalogue that was used in Planck Collaboration et al. [8] is also shown here as the black curve.

Figure 2

The jackknife subarea division for the ‘LowZ North’ and ‘LowZ South’ survey area shown in equatorial coordinates. The grey area shows the masked range of the Planck 2D-ILC CMB map.

Figure 3

The covariance matrix (the upper panels) and correlation matrix (the lower panels) for LowZ North CGC. The left panels show the covariance matrix estimated with the jackknife method, and the right panels show the covariance matrix estimated with CMB mock samples.

Figure 4

The errors estimated with the jackknife (red) and CMB mock (black) samples for ‘LowZ North CGC’. The green and blue step lines show the errors estimated with jackknife samples of ‘LowZ North CGC’ and ‘DR13 Group’, respectively.

Figure 5

Results of the null test with shuffled (green) and mock (red) CMB maps.

Figure 6

The measured $p_{\text{kSZ}}$ with different AP sizes. From top to the bottom, the AP radius is 3, 5, 7, 9 and 11 arcmin. The error bars shown are the square-root of the diagonal terms of the jackknife covariance matrix.

Figure 7

The best-fit $\overline{\tau }({\theta }_{\mathrm{AP}})$ and its $1\sigma$ errors with different AP size. The solid lines show the best-fit theoretical prediction of the mean optical depth as a function of ${\theta }_{\mathrm{AP}}$.

Figure 8

The joint probability of $z_{\mathrm{eff}}$ and $\log M_{\mathrm{h}}$ constrained with the measured $\overline{\tau }$ of LowZ North CGC (red) and DR13 Group (blue). The filled inner and empty outer contours represent the 68% and 95% C.L., respectively. The 68% C.L. contour for DR13 Group goes beyond the parameter ranges of the plot. The squares with dashed cross lines indicate the best-fit values of $[z_{\mathrm{eff}}=0.15,\log M_{\mathrm{h}}=13.24]$ for LowZ North CGC and $[z_{\mathrm{eff}}=0.05,\log M_{\mathrm{h}}=12.58]$ for DR13 Group. The circles with dotted cross lines indicate the median redshift and halo mass of the respective catalogues.

Figure 9

The ${\theta }_{200}$ distribution of the galaxy catalogues. The black dotted vertical line indicates ${\theta }_{200}=7\mathrm{arcmin}$.

Figure 10

The measured kSZ pairwise momentum with different halo mass bins. The catalogues are split into higher and lower halo mass subcatalogues relative to the median value. The upper panel shows the results of the DR13 Group catalogue with median halo mass of $10^{12.24}{h}^{-1}{\mathrm{M}}_{\odot }$. The lower panel shows the results of LowZ North CGC with median halo mass of $10^{14.14}{h}^{-1}{\mathrm{M}}_{\odot }$.

Figure 11

The redshift distributions of the higher/lower halo mass subcatalogues of LowZ North CGC.