A deep-learning approach to the 3D reconstruction of dust density and temperature in star-forming regions

¹ Universität Heidelberg, Zentrum für Astronomie, Institut für Theoretische Astrophysik, Albert-Ueberle-Str. 2, 69120 Heidelberg, Germany
e-mail: v.ksoll@uni-heidelberg.de
² Universität Heidelberg, Interdisziplinäres Zentrum für Wissenschaftliches Rechnen, Im Neuenheimer Feld 205, 69120 Heidelberg, Germany
³ Department of Earth, Environment, and Physics, Worcester State University, Worcester, MA 01602, USA
⁴ Center for Astrophysics, Harvard & Smithsonian, 60 Garden Street, Cambridge, MA 02138, USA
⁵ Istituto di Astrofisica e Planetologia Spaziali (IAPS), INAF, Via Fosso del Cavaliere 100, 00133 Roma, Italy
⁶ Alma Mater Studiorum Università di Bologna, Dipartimento di Fisica e Astronomia (DIFA), Via Gobetti 93/2, 40129 Bologna, Italy
⁷ INAF – Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, 50125 Firenze, Italy
⁸ Université Paris-Cité, Université Paris-Saclay, CEA, CNRS, AIM, 91191 Gif-sur-Yvette, France
⁹ School of Physics and Astronomy, University of St Andrews, North Haugh, St Andrews, KY16 9SS, UK
¹⁰ Jodrell Bank Centre for Astrophysics, Department of Physics and Astronomy, University of Manchester, Oxford Road, Manchester M13 9PL, UK

Received: 18 August 2023
Accepted: 26 January 2024

Abstract

Aims. We introduce a new deep-learning approach for the reconstruction of 3D dust density and temperature distributions from multi-wavelength dust emission observations on the scale of individual star-forming cloud cores (<0.2 pc).

Methods. We constructed a training data set by processing cloud cores from the Cloud Factory simulations with the POLARIS radiative transfer code to produce synthetic dust emission observations at 23 wavelengths between 12 and 1300 µm. We simplified the task by reconstructing the cloud structure along individual lines of sight (LoSs) and trained a conditional invertible neural network (cINN) for this purpose. The cINN belongs to the group of normalising flow methods and it is able to predict full posterior distributions for the target dust properties. We tested different cINN setups, ranging from a scenario that includes all 23 wavelengths down to a more realistically limited case with observations at only seven wavelengths. We evaluated the predictive performance of these models on synthetic test data.

Results. We report an excellent reconstruction performance for the 23-wavelength cINN model, achieving median absolute relative errors of about 1.8% in log(n/m⁻³) and 1% in log(T_dust/K), respectively. We identify trends towards an overestimation at the low end of the density range and towards an underestimation at the high end of both the density and temperature values, which may be related to a bias in the training data. After limiting our coverage to a combination of only seven wavelengths, we still find a satisfactory performance with average absolute relative errors of about 2.8% and 1.7% in log(n/m⁻³) and log(T_dust/K).

Conclusions. This proof-of-concept study shows that the cINN-based approach for 3D reconstruction of dust density and temperature is very promising and it is even compatible with a more realistically constrained wavelength coverage.