PHAT: PHoto-z Accuracy Testing

By H. Hildebrandt, Stéphane Arnouts, P. Capak, L. A. Moustakas, C. Wolf, Filipe B. Abdalla, R. Assef, M. Banerji, Narciso Benitez, G. Brammer and 15 more. 2010

208 citations recorded by the source corpus.

Context. Photometric redshifts (photo-z's) have become an essential tool in extragalactic astronomy. Many current and upcoming observing programmes require great accuracy of photo-z's to reach their scientific goals. Aims. Here we introduce PHAT, the PHoto-z Accuracy Testing programme, an international initiative to test and compare different methods of photo-z estimation. Methods. Two different test environments are set up, one (PHAT0) based on simulations to test the basic functionality of the different photo-z codes, and another one (PHAT1) based on data from the GOODS survey including 18-band photometry and similar to 2000 spectroscopic redshifts. Results. The accuracy of the different methods is expressed and ranked by the global photo-z bias, scatter, and outlier rates. While most methods agree very well on PHAT0 there are differences in the handling of the Lyman-alpha forest for higher redshifts. Furthermore, different methods produce photo-z scatters that can differ by up to a factor of two even in this idealised case. A larger spread in accuracy is found for PHAT1. Few methods benefit from the addition of mid-IR photometry. The accuracy of the other methods is unaffected or suffers when IRAC data are included. Remaining biases and systematic effects can be explained by shortcomings in the different template sets (especially in the mid-IR) and the use of priors on the one hand and an insufficient training set on the other hand. Some strategies to overcome these problems are identified by comparing the methods in detail. Scatters of 4-8% in Delta z/(1 + z) were obtained, consistent with other studies. However, somewhat larger outlier rates (\textgreater 7.5% with Delta z/(1 + z) \textgreater 0.15; \textgreater 4.5% after cleaning) are found for all codes that can only partly be explained by AGN or issues in the photometry or the spec-z catalogue. Some outliers were probably missed in comparisons of photo-z's to other, less complete spectroscopic surveys in the past. There is a general trend that empirical codes produce smaller biases than template-based codes. Conclusions. The systematic, quantitative comparison of different photo-z codes presented here is a snapshot of the current state-of-the-art of photo-z estimation and sets a standard for the assessment of photo-z accuracy in the future. The rather large outlier rates reported here for PHAT1 on real data should be investigated further since they are most probably also present (and possibly hidden) in many other studies. The test data sets are publicly available and can be used to compare new, upcoming methods to established ones and help in guiding future photo-z method development.

Subfields: Astronomy and Astrophysics, Instrumentation

View Paper: https://www.semanticscholar.org/paper/ed3164e3f3f14bd3cb4093fae4151786ebea08f4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Astronomy and Astrophysics

In Instrumentation

Citations in the corpus

The Canadian Cluster Comparison Project: detailed study of systematics and updated weak lensing masses (2015, methodology): “EAZY is well-tested and performs well amongst the best of publicly-available photometric redshift codes (e.g. Hildebrandt et al. 2010).”
Dark Energy Survey Year 1 Results: redshift distributions of the weak-lensing source galaxies (2017, background): “…from broad-band fluxes (e.g. Arnouts et al. 1999; Benı́tez 2000; Bender et al. 2001; Collister & Lahav 2004; Feldmann et al. 2006; Ilbert et al. 2006; Hildebrandt et al. 2010; Carrasco Kind & Brunner 2013; Sánchez et al. 2014; Rau et al. 2015; Hoyle 2016; Sadeh, Abdalla & Lahav 2016; De Vicente,…”
CFHTLenS: the Canada–France–Hawaii Telescope Lensing Survey – imaging data and catalogue products (2012, background): “…and deepwide survey observations have been paralleled by significant community efforts in algorithm development in order to overcome the challenges of this non-trivial observation (Heymans et al. 2006a; Massey et al. 2007a; Bridle et al. 2010; Hildebrandt et al. 2010; Kitching et al. 2012a,b).”
A CRITICAL ASSESSMENT OF PHOTOMETRIC REDSHIFT METHODS: A CANDELS INVESTIGATION (2013, methodology): “This includes Hogg et al. (1998), Abdalla et al. (2008), and Hildebrandt et al. (2008, 2010).”
UVUDF: ULTRAVIOLET THROUGH NEAR-INFRARED CATALOG AND PHOTOMETRIC REDSHIFTS OF GALAXIES IN THE HUBBLE ULTRA DEEP FIELD (2015, methodology): “The Bayesian Photometric Redshift (BPZ) algorithm (Beńıtez 2000; Beńıtez et al. 2004; Coe et al. 2006) is the primary redshift software used due to its robust performance (Hildebrandt et al. 2010), familiarity of the code to the authors, and best performance as detailed in Section 4.2.”
A HIGHLY CONSISTENT FRAMEWORK FOR THE EVOLUTION OF THE STAR-FORMING “MAIN SEQUENCE” FROM z ∼ 0–6 (2014, methodology): “See Appendix G for more discussion on some of the uncertainties present our methodology.”
The LABOCA survey of the Extended Chandra Deep Field South: A photometric redshift survey of submillimetre galaxies (2010, methodology): “We also note that Hildebrandt et al. (2010) recently performed extensive testing of the accuracy and reliability of 19 different photometric redshift codes, including several neural networks and others with training modes, on galaxies in the GOODS with a median spectroscopic redshift z < 1.”
Photometric redshift analysis in the Dark Energy Survey Science Verification data (2014, methodology): “Such technique has been succesfully applied by LeP- hare in this work, as was also the case in Hildebrandt et al. (2010).”
CLASH: THREE STRONGLY LENSED IMAGES OF A CANDIDATE z ≈ 11 GALAXY (2012, methodology): “These two methods, BPZ (§4.1) and LePHARE (§4.2) were the top two performers out of 17 methods tested in Hildebrandt et al. (2010).”
CFHTLenS: Improving the quality of photometric redshifts with precision photometry (2011, methodology): “It should be noted that the spec-z catalogues used here are incomplete for i>∼ 22 and might paint too positive a picture of the photoz accuracy (see Hildebrandt et al. 2008, 2010, for extensive discussions of these effects).”
KiDS-450: cosmological parameter constraints from tomographic weak gravitational lensing (2016, background): “The reason for this approach is that the output of BPZ (and essentially every photo-z code; see e.g. Hildebrandt et al. 2010) is biased at a level that cannot be tolerated by contemporary and especially future cosmic shear measurements (for a discussion, see Newman et al. 2015).”
CFHTLenS: the relation between galaxy dark matter haloes and baryons from weak gravitational lensing (2013, methodology): “…the dispersion in log stellar mass and the lower panels show the bias of the CFHTLenS stellar mass estimates relative to the WIRDS stellar mass estimates. mates as the alternative Bayesian LEPHARE2 (Arnouts et al. 1999; Ilbert et al. 2006) software (see Hildebrandt et al. 2010, for a comparison).”