A ROBUST QUANTIFICATION OF GALAXY CLUSTER MORPHOLOGY USING ASYMMETRY AND CENTRAL CONCENTRATION

• Published in: The Astrophysical journal Vol. 779; no. 2; pp. 1 - 15
• Main Authors: Nurgaliev, D, McDonald, M, Benson, B A, Miller, E D, Stubbs, C W, VIKHLININ, A
• Format: Journal Article
• Language: English
• Published: United States 20.12.2013
• Subjects: ABUNDANCE; ALGORITHMS; ASTROPHYSICS, COSMOLOGY AND ASTRONOMY; ASYMMETRY; BRIGHTNESS; CLASSIFICATION; Clusters; CONCENTRATION RATIO; Counting; COVERINGS; GALAXY CLUSTERS; Morphology; PHOTONS; RED SHIFT; SIGNAL-TO-NOISE RATIO; STABILITY; Statistics; Substructures; SURFACES; X RADIATION; X-RAY GALAXIES
• ISSN: 0004-637X; 1538-4357
• DOI: 10.1088/0004-637X/779/2/112

We present a novel quantitative scheme of cluster classification based on the morphological properties that are manifested in X-ray images. We use a conventional radial surface brightness concentration parameter (c sub(SB)) as defined previously by others and a new asymmetry parameter, which we define in this paper. Our asymmetry parameter, which we refer to as photon asymmetry (A sub(phot)), was developed as a robust substructure statistic for cluster observations with only a few thousand counts. To demonstrate that photon asymmetry exhibits better stability than currently popular power ratios and centroid shifts, we artificially degrade the X-ray image quality by (1) adding extra background counts, (2) eliminating a fraction of the counts, (3) increasing the width of the smoothing kernel, and (4) simulating cluster observations at higher redshift. The asymmetry statistic presented here has a smaller statistical uncertainty than competing substructure parameters, allowing for low levels of substructure to be measured with confidence. A sub(phot) is less sensitive to the total number of counts than competing substructure statistics, making it an ideal candidate for quantifying substructure in samples of distant clusters covering a wide range of observational signal-to-noise ratios. Additionally, we show that the asymmetry-concentration classification separates relaxed, cool-core clusters from morphologically disturbed mergers, in agreement with by-eye classifications. Our algorithms, freely available as Python scripts (https://github.com/ndaniyar/aphot), are completely automatic and can be used to rapidly classify galaxy cluster morphology for large numbers of clusters without human intervention.