Overview

Both the ESA/NASA Euclid and NASA WFIRST missions will use galaxy weak lensing over large sky areas to measure the growth of structure with cosmic time. Structure growth is sensitive to the nature of dark energy and thus is a powerful cosmological probe. But in order to fully exploit the technique, the redshift distributions of galaxies in the weak lensing samples must be known with exquisite accuracy - i.e., they must be highly unbiased. While any particular galaxy's photo-z can be uncertain, the distribution of galaxy redshifts must be well-known.

The aim of C3R2 is to develop a systematic, empirical mapping of the relation of galaxy colors to redshift in order to enable redshift calibration at the level required by weak-lensing cosmology with Euclid and WFIRST. This is not easy because of the faintness of the galaxies in these surveys and the difficultly in obtaining secure redshifts for some populations.

The problem is that spectroscopic samples are not fully representative of the parameter space of galaxy properties (in particular colors). We developed a method to measure the high-dimensional color distribution of galaxies in a given survey in Masters et al 2015, by mapping the high-dimensional distribution of galaxies in color space using an unsupervised machine learning technique known as the self-organizing map (SOM):

Illustration of the self-organized map (SOM) used for the C3R2 DR2 source selection and analysis. Each cell of the SOM represents a particular spectral energy distribution (SED) that shows up with regularity in the deep field data; particular SEDs are illustrated on the sides. The axes should be thought of as indices to parts of the high-dimensional galaxy color space. The color shows the empirical relation of galaxy redshift to position in color space, while the white regions are "terra incognita", where no high-confidence redshifts exist. These are the primary targets of C3R2.

Our goal is to fill out the color space, thus providing a constrained color-redshift relation for galaxies to the Euclid depth.