Home > Research > Publications & Outputs > A Machine-learning Method for Identifying Multi...

Associated organisational unit

Electronic data

  • 1806.06859

    Rights statement: This is an author-created, un-copyedited version of an article accepted for publication/published in Astrophysical Journal. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at doi:10.3847/1538-4357/aacdaa

    Accepted author manuscript, 2.51 MB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies: Training and Testing Using AS2UDS and ALESS

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies: Training and Testing Using AS2UDS and ALESS. / An, Fang Xia; Stach, S. M.; Smail, Ian et al.
In: The Astrophysical Journal, Vol. 862, No. 2, 101, 01.08.2018.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

An, FX, Stach, SM, Smail, I, Swinbank, AM, Almaini, O, Simpson, C, Hartley, W, Maltby, DT, Ivison, RJ, Arumugam, V, Wardlow, JL, Cooke, EA, Gullberg, B, Thomson, AP, Chen, C-C, Simpson, JM, Geach, JE, Scott, D, Dunlop, JS, Farrah, D, van der Werf, P, Blain, AW, Conselice, C, Michałowski, M, Chapman, SC & Coppin, KEK 2018, 'A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies: Training and Testing Using AS2UDS and ALESS', The Astrophysical Journal, vol. 862, no. 2, 101. https://doi.org/10.3847/1538-4357/aacdaa

APA

An, F. X., Stach, S. M., Smail, I., Swinbank, A. M., Almaini, O., Simpson, C., Hartley, W., Maltby, D. T., Ivison, R. J., Arumugam, V., Wardlow, J. L., Cooke, E. A., Gullberg, B., Thomson, A. P., Chen, C.-C., Simpson, J. M., Geach, J. E., Scott, D., Dunlop, J. S., ... Coppin, K. E. K. (2018). A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies: Training and Testing Using AS2UDS and ALESS. The Astrophysical Journal, 862(2), Article 101. https://doi.org/10.3847/1538-4357/aacdaa

Vancouver

An FX, Stach SM, Smail I, Swinbank AM, Almaini O, Simpson C et al. A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies: Training and Testing Using AS2UDS and ALESS. The Astrophysical Journal. 2018 Aug 1;862(2):101. Epub 2018 Jul 27. doi: 10.3847/1538-4357/aacdaa

Author

An, Fang Xia ; Stach, S. M. ; Smail, Ian et al. / A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies : Training and Testing Using AS2UDS and ALESS. In: The Astrophysical Journal. 2018 ; Vol. 862, No. 2.

Bibtex

@article{474de65713054976ae0af981e9bd7b46,
title = "A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies: Training and Testing Using AS2UDS and ALESS",
abstract = "We describe the application of supervised machine-learning algorithms to identify the likely multiwavelength counterparts to submillimeter sources detected in panoramic, single-dish submillimeter surveys. As a training set, we employ a sample of 695 (S 870μm ≳ 1 mJy) submillimeter galaxies (SMGs) with precise identifications from the ALMA follow-up of the SCUBA-2 Cosmology Legacy Survey{\textquoteright}s UKIDSS-UDS field (AS2UDS). We show that radio emission, near-/mid-infrared colors, photometric redshift, and absolute H-band magnitude are effective predictors that can distinguish SMGs from submillimeter-faint field galaxies. Our combined radio + machine-learning method is able to successfully recover ˜85% of ALMA-identified SMGs that are detected in at least three bands from the ultraviolet to radio. We confirm the robustness of our method by dividing our training set into independent subsets and using these for training and testing, respectively, as well as applying our method to an independent sample of ˜100 ALMA-identified SMGs from the ALMA/LABOCA ECDF-South Survey (ALESS). To further test our methodology, we stack the 870 μm ALMA maps at the positions of those K-band galaxies that are classified as SMG counterparts by the machine learning but do not have a >4.3σ ALMA detection. The median peak flux density of these galaxies is S 870μm = (0.61 ± 0.03) mJy, demonstrating that our method can recover faint and/or diffuse SMGs even when they are below the detection threshold of our ALMA observations. In future, we will apply this method to samples drawn from panoramic single-dish submillimeter surveys that currently lack interferometric follow-up observations to address science questions that can only be tackled with large statistical samples of SMGs.",
keywords = "cosmology: observations, galaxies: evolution, galaxies: formation, galaxies: high-redshift, galaxies: starburst, submillimeter: galaxies",
author = "An, {Fang Xia} and Stach, {S. M.} and Ian Smail and Swinbank, {A. M.} and O. Almaini and C. Simpson and W. Hartley and Maltby, {D. T.} and Ivison, {R. J.} and V. Arumugam and Wardlow, {J. L.} and Cooke, {E. A.} and B. Gullberg and Thomson, {A. P.} and Chian-Chou Chen and Simpson, {J. M.} and Geach, {J. E.} and D. Scott and Dunlop, {J. S.} and D. Farrah and {van der Werf}, P. and Blain, {A. W.} and C. Conselice and M. Micha{\l}owski and Chapman, {S. C.} and Coppin, {K. E. K.}",
note = "This is an author-created, un-copyedited version of an article accepted for publication/published in Astrophysical Journal. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at doi:10.3847/1538-4357/aacdaa",
year = "2018",
month = aug,
day = "1",
doi = "10.3847/1538-4357/aacdaa",
language = "English",
volume = "862",
journal = "The Astrophysical Journal",
issn = "0004-637X",
publisher = "Institute of Physics Publishing",
number = "2",

}

RIS

TY - JOUR

T1 - A Machine-learning Method for Identifying Multiwavelength Counterparts of Submillimeter Galaxies

T2 - Training and Testing Using AS2UDS and ALESS

AU - An, Fang Xia

AU - Stach, S. M.

AU - Smail, Ian

AU - Swinbank, A. M.

AU - Almaini, O.

AU - Simpson, C.

AU - Hartley, W.

AU - Maltby, D. T.

AU - Ivison, R. J.

AU - Arumugam, V.

AU - Wardlow, J. L.

AU - Cooke, E. A.

AU - Gullberg, B.

AU - Thomson, A. P.

AU - Chen, Chian-Chou

AU - Simpson, J. M.

AU - Geach, J. E.

AU - Scott, D.

AU - Dunlop, J. S.

AU - Farrah, D.

AU - van der Werf, P.

AU - Blain, A. W.

AU - Conselice, C.

AU - Michałowski, M.

AU - Chapman, S. C.

AU - Coppin, K. E. K.

N1 - This is an author-created, un-copyedited version of an article accepted for publication/published in Astrophysical Journal. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at doi:10.3847/1538-4357/aacdaa

PY - 2018/8/1

Y1 - 2018/8/1

N2 - We describe the application of supervised machine-learning algorithms to identify the likely multiwavelength counterparts to submillimeter sources detected in panoramic, single-dish submillimeter surveys. As a training set, we employ a sample of 695 (S 870μm ≳ 1 mJy) submillimeter galaxies (SMGs) with precise identifications from the ALMA follow-up of the SCUBA-2 Cosmology Legacy Survey’s UKIDSS-UDS field (AS2UDS). We show that radio emission, near-/mid-infrared colors, photometric redshift, and absolute H-band magnitude are effective predictors that can distinguish SMGs from submillimeter-faint field galaxies. Our combined radio + machine-learning method is able to successfully recover ˜85% of ALMA-identified SMGs that are detected in at least three bands from the ultraviolet to radio. We confirm the robustness of our method by dividing our training set into independent subsets and using these for training and testing, respectively, as well as applying our method to an independent sample of ˜100 ALMA-identified SMGs from the ALMA/LABOCA ECDF-South Survey (ALESS). To further test our methodology, we stack the 870 μm ALMA maps at the positions of those K-band galaxies that are classified as SMG counterparts by the machine learning but do not have a >4.3σ ALMA detection. The median peak flux density of these galaxies is S 870μm = (0.61 ± 0.03) mJy, demonstrating that our method can recover faint and/or diffuse SMGs even when they are below the detection threshold of our ALMA observations. In future, we will apply this method to samples drawn from panoramic single-dish submillimeter surveys that currently lack interferometric follow-up observations to address science questions that can only be tackled with large statistical samples of SMGs.

AB - We describe the application of supervised machine-learning algorithms to identify the likely multiwavelength counterparts to submillimeter sources detected in panoramic, single-dish submillimeter surveys. As a training set, we employ a sample of 695 (S 870μm ≳ 1 mJy) submillimeter galaxies (SMGs) with precise identifications from the ALMA follow-up of the SCUBA-2 Cosmology Legacy Survey’s UKIDSS-UDS field (AS2UDS). We show that radio emission, near-/mid-infrared colors, photometric redshift, and absolute H-band magnitude are effective predictors that can distinguish SMGs from submillimeter-faint field galaxies. Our combined radio + machine-learning method is able to successfully recover ˜85% of ALMA-identified SMGs that are detected in at least three bands from the ultraviolet to radio. We confirm the robustness of our method by dividing our training set into independent subsets and using these for training and testing, respectively, as well as applying our method to an independent sample of ˜100 ALMA-identified SMGs from the ALMA/LABOCA ECDF-South Survey (ALESS). To further test our methodology, we stack the 870 μm ALMA maps at the positions of those K-band galaxies that are classified as SMG counterparts by the machine learning but do not have a >4.3σ ALMA detection. The median peak flux density of these galaxies is S 870μm = (0.61 ± 0.03) mJy, demonstrating that our method can recover faint and/or diffuse SMGs even when they are below the detection threshold of our ALMA observations. In future, we will apply this method to samples drawn from panoramic single-dish submillimeter surveys that currently lack interferometric follow-up observations to address science questions that can only be tackled with large statistical samples of SMGs.

KW - cosmology: observations

KW - galaxies: evolution

KW - galaxies: formation

KW - galaxies: high-redshift

KW - galaxies: starburst

KW - submillimeter: galaxies

U2 - 10.3847/1538-4357/aacdaa

DO - 10.3847/1538-4357/aacdaa

M3 - Journal article

VL - 862

JO - The Astrophysical Journal

JF - The Astrophysical Journal

SN - 0004-637X

IS - 2

M1 - 101

ER -