Unveiling Hidden Lyman Alpha Emitters in the DESI DR1 Data

Abstract We present an automatic method based on machine learning convolutional neural network (CNN) architecture to detect Ly α emitters (LAEs) hidden in the Data Release 1 spectroscopic data set of the Dark Energy Spectroscopic Instrument (DESI). Those LAEs mostly have incorrect redshift estimations because the current DESI pipeline is not designed to detect and measure the redshifts of galaxies at z > 2. To uncover those sources, we first visually inspect 4134 DESI spectra and construct a sample, consisting of both LAEs and non-LAEs, for training and testing the CNN-based model to (1) detect LAEs in DESI spectra and (2) determine their Ly α redshifts. The final model yields 95.2% purity and 95.9% completeness for detecting LAEs. We apply this model to approximately 2 × 10 6 spectra of sources targeted as emission-line galaxies and detect 19,685 LAEs from z ∼ 2 to 3.5 within 12 minutes with a single GPU, illustrating the high efficiency of this model for identifying LAEs. The detected LAEs are mostly at the bright end of the luminosity function with Ly α luminosity L Ly α ≳ 10 43 erg s −1 . The high signal-to-noise composite spectrum of the detected LAEs further shows various spectral features, including P-Cygni profiles of metal lines and Mg ii emission lines, possible indicators of Lyman continuum escape fraction, revealing the rich astrophysical information in this LAE sample. Finally, this sample can be used to train and validate the pipelines for redshift determination of LAEs for the preparation of the DESI-II survey.

Authors

Institutions

Publication Details

Journal
The Astrophysical Journal
Published
2026-10-28
DOI
https://doi.org/10.3847/1538-4357/aeaa44
Primary Topic
Galaxies: Formation, Evolution, Phenomena
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Unveiling Hidden Lyman Alpha Emitters in the DESI DR1 Data

Will J. Percival, Ting-Wen Lan, Stéphanie Juneau, Michael S. Schubnell et al.
2 links
The Astrophysical Journal
Galaxies: Formation, Evolution, Phenomena
article

Unveiling Hidden Lyman Alpha Emitters in the DESI DR1 Data

Will J. Percival, Ting-Wen Lan, Stéphanie Juneau, Michael S. Schubnell, S. Nadathur, J. Moustakas, Ignasi Pérez-Ràfols, Martin Landriau, Marc Manera, Satya Gontcho A Gontcho, R. Miquel, G. Gutierrez, J. E. Forero-Romero, Steven P. Ahlen, Aaron Meisner, Claire Poppett, J Xavier Prochaska, P. Doel, Aguilar Jl., D. Sprayberry, R. Joyce, H. Zou, J. Silber, E. Gaztañaga, Biprateep Dey, G. Tarlé, D Bianchi, G. Rossi, F. Prada, D. Brooks, Jui-Kuan Chan, E. J. Sanchez, J. Jimenez, Shun Saito, D. Schlegel, A. de la Macorra, A. Kremin, A. Font-Ribera, D. Kirkby, A. Cuceu, C. Hahn, B. A. Weaver
article en

Abstract

Abstract We present an automatic method based on machine learning convolutional neural network (CNN) architecture to detect Ly α emitters (LAEs) hidden in the Data Release 1 spectroscopic data set of the Dark Energy Spectroscopic Instrument (DESI). Those LAEs mostly have incorrect redshift estimations because the current DESI pipeline is not designed to detect and measure the redshifts of galaxies at z > 2. To uncover those sources, we first visually inspect 4134 DESI spectra and construct a sample, consisting of both LAEs and non-LAEs, for training and testing the CNN-based model to (1) detect LAEs in DESI spectra and (2) determine their Ly α redshifts. The final model yields 95.2% purity and 95.9% completeness for detecting LAEs. We apply this model to approximately 2 × 10 6 spectra of sources targeted as emission-line galaxies and detect 19,685 LAEs from z ∼ 2 to 3.5 within 12 minutes with a single GPU, illustrating the high efficiency of this model for identifying LAEs. The detected LAEs are mostly at the bright end of the luminosity function with Ly α luminosity L Ly α ≳ 10 43 erg s −1 . The high signal-to-noise composite spectrum of the detected LAEs further shows various spectral features, including P-Cygni profiles of metal lines and Mg ii emission lines, possible indicators of Lyman continuum escape fraction, revealing the rich astrophysical information in this LAE sample. Finally, this sample can be used to train and validate the pipelines for redshift determination of LAEs for the preparation of the DESI-II survey.

The Astrophysical JournalVol. 1011(1)
University of Siena (IT), Boston University (US), Institució Catalana de Recerca i Estudis Avançats (ES), Universitat Autònoma de Barcelona (ES), Institute of Astronomy and Astrophysics, Academia Sinica (TW), Fermi National Accelerator Laboratory (US), Consejo Superior de Investigaciones Científicas (ES), Lawrence Berkeley National Laboratory (US), University of Waterloo (CA), Universidad de Los Andes (CO), National Taiwan University (TW), University of Pittsburgh (US), Perimeter Institute (CA), University of California, Santa Cruz (US), University of Toronto (CA), Institute for High Energy Physics (ES), University of Milan (IT), Chinese Academy of Sciences (CN), Missouri University of Science and Technology (US), University of Michigan (US), Brera Astronomical Observatory (IT), University of California System (US), Sejong University (KR), Institut d'Estudis Espacials de Catalunya (ES), NSF NOIRLab (US), Instituto de Astrofísica de Andalucía (ES), Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (ES), University of California Observatories (US), University College London (GB), University of Virginia (US), University of Portsmouth (GB), The University of Tokyo (JP), The University of Texas at Austin (US), National Institute for Astrophysics (IT), Universidad Nacional Autónoma de México (MX), Universitat Politècnica de Catalunya (ES)
National Science Foundation, U.S. Department of Energy, Gordon and Betty Moore Foundation, Ministerio de Ciencia, Innovación y Universidades, Academia Sinica, Ministry of Education, India, National Taiwan University, Commissariat à l'Énergie Atomique et aux Énergies Alternatives, National Science and Technology Council, Office of Science, Science and Technology Facilities Council, Agencia Estatal de Investigación, Division of Astronomical Sciences, High Energy Physics
Openalex Percentile: Top 65%
Galaxies: Formation, Evolution, Phenomena
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon