Hintze Centre for Astrophysical Surveys

Abstract:

The next generation of cosmology experiments will be required to use photometric redshifts rather than spectroscopic redshifts. Obtaining accurate and well-characterized photometric redshift distributions is therefore critical for Euclid, the Large Synoptic Survey Telescope and the Square Kilometre Array. However, determining accurate variance predictions alongside single point estimates is crucial, as they can be used to optimize the sample of galaxies for the specific experiment (e.g. weak lensing, baryon acoustic oscillations, supernovae), trading off between completeness and reliability in the galaxy sample. The various sources of uncertainty in measurements of the photometry and redshifts put a lower bound on the accuracy that any model can hope to achieve. The intrinsic uncertainty associated with estimates is often non-uniform and input-dependent, commonly known in statistics as heteroscedastic noise. However, existing approaches are susceptible to outliers and do not take into account variance induced by non-uniform data density and in most cases require manual tuning of many parameters. In this paper, we present a Bayesian machine learning approach that jointly optimizes the model with respect to both the predictive mean and variance we refer to as Gaussian processes for photometric redshifts (GPz). The predictive variance of the model takes into account both the variance due to data density and photometric noise. Using the SDSS DR12 data, we show that our approach substantially outperforms other machine learning methods for photo-z estimation and their associated variance, such as tpz and annz2. We provide a matlab and python implementations that are available to download at https://github.com/OxfordML/GPz.

Abstract:

We present the stellar mass (M*) and Wide-Field Infrared Survey Explorer (WISE) absolute Band 1 magnitude (MW1) Tully-Fisher relations (TFRs) of subsets of galaxies from the CO Legacy Database for the Galex Arecibo SDSS Survey (COLD GASS). We examine the benefits and drawbacks of several commonly used fitting functions in the context of measuring CO(1-0) line widths (and thus rotation velocities), favouring the Gaussian Double Peak function. We find the MW1 and M* TFR, for a carefully selected sub-sample, to be MW1 = (-7.1 ± 0.6) [log(W50/sin i / km s^-1) - 23.83 ± 0.09 and log (M*/M⊙) = (3.3 ± 0.3) [log(W50/sin i / km s^-1) -2.58] + 10:51 ± 0.04, respectively, where W50 is the width of a galaxy's CO(1-0) integrated profile at 50% of its maximum and the inclination i is derived from the galaxy axial ratio measured on the SDSS r-band image. We find no evidence for any significant offset between the TFRs of COLD GASS galaxies and those of comparison samples of similar redshifts and morphologies. The slope of the COLD GASS M* TFR agrees with the relation of Pizagno et al. (2005). However, we measure a comparitively shallower slope for the COLD GASS MW1 TFR as compared to the relation of Tully and Pierce (2000). We attribute this to the fact that the COLD GASS sample comprises galaxies of various (late-type) morphologies. Nevertheless, our work provides a robust reference point with which to compare future CO TFR studies.