Cosmological Constraints with Clustering-Based Redshifts

We demonstrate that observations lacking reliable redshift information, such as photometric and radio continuum surveys, can produce robust measurements of cosmological parameters when empowered by clustering-based redshift estimation. This method infers the redshift distribution based on the spatial clustering of sources, using cross-correlation with a reference dataset with known redshifts. Applying this method to the existing SDSS photometric galaxies, and projecting to future radio continuum surveys, we show that sources can be efficiently divided into several redshift bins, increasing their ability to constrain cosmological parameters. We forecast constraints on the dark-energy equation-of-state and on local non-gaussianity parameters. We explore several pertinent issues, including the tradeoff between including more sources versus minimizing the overlap between bins, the shot-noise limitations on binning, and the predicted performance of the method at high redshifts. Remarkably, we find that, once this technique is implemented, constraints on dynamical dark energy from the SDSS imaging catalog can be competitive with, or better than, those from the spectroscopic BOSS survey and even future planned experiments. Further, constraints on primordial non-Gaussianity from future large-sky radio-continuum surveys can outperform those from the Planck CMB experiment, and rival those from future spectroscopic galaxy surveys. The application of this method thus holds tremendous promise for cosmology.