Application of Convolutional Neural Networks to Identify Protostellar Outflows in CO Emission

Abstract

We adopt the deep learning method CASI-3D (Convolutional Approach to Structure Identification-3D) to identify protostellar outflows in molecular line spectra. We conduct magneto-hydrodynamics simulations that model forming stars that launch protostellar outflows and use these to generate synthetic observations. We apply the 3D radiation transfer code RADMC-3D to model 12CO (J=1-0) line emission from the simulated clouds. We train two CASI-3D models: ME1 is trained to predict only the position of outflows, while MF is trained to predict the fraction of the mass coming from outflows in each voxel. The two models successfully identify all 60 previously visually identified outflows in Perseus. Additionally, CASI-3D finds 20 new high-confidence outflows. All of these have coherent high-velocity structures, and 17 of them have nearby young stellar objects, while the remaining three are outside the Spitzer survey coverage. The mass, momentum and energy of individual outflows in Perseus predicted by model MF is comparable to the previous estimations. This similarity is due to a cancelation in errors: previous calculations missed outflow material with velocities comparable to the cloud velocity, however, they compensate for this by over-estimating the amount of mass at higher velocities that has contamination from non-outflow gas. We show outflows likely driven by older sources have more high-velocity gas compared to those driven by younger sources.

arXiv commentRevised version submitted to ApJ, comments welcome

References in corpus (39)

• Theory of Star Formation
• A Spitzer Survey of Young Stellar Clusters within One Kiloparsec of the Sun: Cluster Core Extraction and Basic Structural Analysis
• The Energy Dissipation Rate of Supersonic, Magnetohydrodynamic Turbulence in Molecular Clouds
• A Catalog of Extended Green Objects (EGOs) in the GLIMPSE Survey: A new sample of massive young stellar object outflow candidates
• Spitzer Observations of NGC 1333: A Study of Structure and Evolution in a Nearby Embedded Cluster
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• Infrared Extinction Toward Nearby Star-Forming Regions
• The Effects of Radiative Transfer on Low-Mass Star Formation
• The COMPLETE Survey of Star-Forming Regions: Phase I Data
• Embedding Lagrangian Sink Particles in Eulerian Grids
• Modeling jet and outflow feedback during star cluster formation
• The COMPLETE Survey of Outflows in Perseus
• Radiation-Hydrodynamic Simulations of Massive Star Formation with Protostellar Outflows
• Evolution of Protostellar Outflow around Low-mass Protostar
• Molecular Outflows Driven by Low-Mass Protostars. I. Correcting for Underestimates When Measuring Outflow Masses and Dynamical Properties
• Star Cluster Formation in Turbulent, Magnetized Dense Clumps with Radiative and Outflow Feedback
• A submillimetre survey of the kinematics of the Perseus molecular cloud - II. Molecular outflows
• A Bubbling Nearby Molecular Cloud: COMPLETE Shells in Perseus
• Protostellar outflow-driven turbulence
• Impact of Protostellar Outflows on Turbulence and Star Formation Efficiency in Magnetized Dense Cores
• Feedback Effects on Low-Mass Star Formation
• Investigations of Protostellar Outflow Launching and Gas Entrainment: Hydrodynamic Simulations and Molecular Emission
• Dense Cores in Perseus: The Influence of Stellar Content and Cluster Environment
• Cluster Formation Triggered by Filament Collisions in Serpens South
• The formation of the W43 complex: constraining its atomic-to-molecular transition and searching for colliding clouds
• A Stable, Accurate Methodology for High Mach Number, Strong Magnetic Field MHD Turbulence with Adaptive Mesh Refinement: Resolution and Refinement Studies
• The Milky Way Project: Leveraging Citizen Science and Machine Learning to Detect Interstellar Bubbles
• Star-Gas Surface Density Correlations in Twelve Nearby Molecular Clouds I: Data Collection and Star-Sampled Analysis
• Outflows and Bubbles in Taurus: Star-formation Feedback Sufficient to Maintain Turbulence
• The Milky Way Project Second Data Release: Bubbles and Bow Shocks
• Star cluster formation with stellar feedback and large-scale inflow
• Protostar Mass Due to Infall and Dispersal
• Molecular Outflows Identified in the FCRAO CO Survey of the Taurus Molecular Cloud
• The COMPLETE Nature of the Warm Dust Ring in Perseus
• CASI: A Convolutional Neural Network Approach for Shell Identification
• Classifying structures in the ISM with Support Vector Machines: the G16.05-0.57 supernova remnant
• Application of Convolutional Neural Networks to Identify Stellar Feedback Bubbles in CO Emission
• Searching for Molecular Outflows with Support Vector Machines: Dark Cloud Complex in Cygnus
• Assessing the Performance of a Machine Learning Algorithm in Identifying Bubbles in Dust Emission

Cited by in corpus (10)

• The Single-Cloud Star Formation Relation
• Turbulence, Coherence and Collapse: Three Phases for Core Evolution
• A Morphological Classification of 18190 Molecular Clouds Identified in ${}^{12}$CO Data from the MWISP Survey
• Denoising Diffusion Probabilistic Models to Predict the Density of Molecular Clouds
• PGNets: Planet mass prediction using convolutional neural networks for radio continuum observations of protoplanetary disks
• A Census of Outflow to Magnetic Field Orientations in Nearby Molecular Clouds
• A Census of Protostellar Outflows in Nearby Molecular Clouds
• Application of Convolutional Neural Networks to Predict Magnetic Fields Directions in Turbulent Clouds
• Effects of stellar feedback on cores in STARFORGE
• Modeling turbulent and self-gravitating fluids with Fourier neural operators