A Stellar Reddening based Dust Map for Cosmological Applications

Galactic dust is a troublesome foreground for cosmology. Its absorption reddens background galaxies and quasars, and its emission at far infrared and microwave wavelengths interferes with measurements of the cosmic microwave background. Dust maps based on emission or absorption allow us to partially correct for these effects, but risk introducing systematic biases in cosmological inferences. For example, existing dust maps have been found to have spatial correlations with distributions of extragalactic objects. Emission-based maps are contaminated by the cosmic infrared background, while maps inferred from stellar reddenings suffer from imperfect removal of galaxies and quasars from stellar catalogs.

We present a two-dimensional high-latitude dust extinction map based on stellar reddenings, that has a well-defined point spread function. We use a stricter selection of catalog objects to filter out galaxies and quasars. We examine the performance of our selection on the basis of two figures of merit — the removal of extragalactic objects on a subset of objects with spectroscopic labels and the spatial correlation of the resultant Galactic extinction map with extragalactic structure, using the clustering-redshift based method (Chiang et al 2018).