Physical Evidence for Dark Energy

TL;DR

The study cross-correlates SDSS Luminous Red Galaxies with WMAP CMB maps to detect the late Integrated Sachs-Wolfe (ISW) effect and constrain Sunyaev-Zel'dovich (SZ) contributions. Using four redshift slices of LRGs and multi-frequency CMB data, the authors compute the cross-correlation w_gT(θ) and rigorously assess uncertainties via jackknife and random-CMB simulations. They find a statistically significant, achromatic cross-correlation for the three highest-redshift samples, consistent with ISW in a flat ΛCDM universe, with SZ playing a sub-dominant role. A simple ISW+SZ model yields good fits for most high-redshift cross-correlations, providing independent evidence for dark energy and supporting the ΛCDM paradigm. The work combines precise data handling, robust error analysis, and theoretical modeling to connect large-scale structure growth with CMB anisotropies, reinforcing the physical reality of dark energy.

Abstract

We present measurements of the angular cross-correlation between luminous red galaxies from the Sloan Digital Sky Survey and the cosmic microwave background temperature maps from the Wilkinson Microwave Anisotropy Probe. We find a statistically significant achromatic positive correlation between these two data sets, which is consistent with the expected signal from the late Integrated Sachs-Wolfe (ISW) effect. We do not detect any anti-correlation on small angular scales as would be produced from a large Sunyaev-Zel'dovich (SZ) effect, although we do see evidence for some SZ effect for our highest redshift samples. Assuming a flat universe, our preliminary detection of the ISW effect provides independent physical evidence for the existence of dark energy.

Figures (3)

• Figure 1: The photometric redshift distributions for the four LRG subsamples, taking into account the covariance between photometric redshift and galaxy type. To select these galaxies, we imposed a color cut of $0.7(g - r) + 1.2((r - i) - 0.18) > 1.6$ and $(g - r) > 1$. We then define $d_\perp = (r - i) - (g - r)/8$, and step through $d_\perp$ in increments of 0.2 starting at $0.2 < d_\perp < 0.4$ and ending at $d_\perp > 0.8$ to determine our four redshift samples, from lowest to highest redshift respectively. There are 0.4 million LRGs in the $z \sim 0.35$ subsample, 0.8 million in the $z \sim 0.43$ subsample, 1 million in the $z \sim 0.49$ subsample, and 0.7 million in the $z \sim 0.55$ subsample.
• Figure 2: The uniform-weighted galaxy--WMAP cross-correlation functions for the four galaxy subsamples. The shaded boxes show the measurement and the $1\sigma$ jack--knife errors for the "clean" galaxy--WMAP cross-correlation function. The errors on the other functions are nearly identical. The Q map is given by the red line, V by the green line, W by the blue line and smoothed map by the cyan line.
• Figure 3: The comparison of our theoretical predictions to our measurement of the W-channel WMAP map cross-correlated with the $z \sim 0.49$ galaxy subsample. The fitted ISW cross-correlation function is given by the red line, the SZ by the green line and the sum of the two is given by the blue line. We find $\bar{b} = 5.47 \pm 1.82$ and $\overline{T_{\rm e} b_{\rm P}} = 0.15 \pm 0.61$, where the errors are unmarginalized.