Identification of molecular line emission using Convolutional Neural Networks

Abstract

Complex organic molecules (COMs) are observed to be abundant in various astrophysical environments, in particular toward star forming regions they are observed both toward protostellar envelopes as well as shocked regions. Emission spectrum especially of heavier COMs may consists of up to hundreds of lines, where line blending hinders the analysis. However, identifying the molecular composition of the gas leading to the observed millimeter spectra is the first step toward a quantitative analysis. We develop a new method based on supervised machine learning to recognize spectroscopic features of the rotational spectrum of molecules in the 3mm atmospheric transmission band for a list of species including COMs with the aim to obtain a detection probability. We used local thermodynamic equilibrium (LTE) modeling to build a large set of synthetic spectra of 20 molecular species including COMs with a range of physical conditions typical for star forming regions. We successfully designed and trained a Convolutional Neural Network (CNN) that provides detection probabilities of individual species in the spectra. We demonstrate that the produced CNN-model has a robust performance to detect spectroscopic signatures from these species in synthetic spectra. We evaluate its ability to detect molecules according to the noise level, frequency coverage, and line-richness, and also test its performance for incomplete frequency coverage with high detection probabilities for the tested parameter space, and no false predictions. Ultimately, we apply the CNN-model to obtain predictions on observational data from the literature toward line-rich hot-core like sources, where detection probabilities remain reasonable with no false detection. We prove the use of CNNs facilitating the analysis of complex millimeter spectra both on synthetic spectra as well as first tests on observational data.

References in corpus (22)

• CASA, the Common Astronomy Software Applications for Radio Astronomy
• Formation of methyl formate and other organic species in the warm-up phase of hot molecular cores
• The ALMA Protostellar Interferometric Line Survey (PILS): First results from an unbiased submillimeter wavelength line survey of the Class 0 protostellar binary IRAS 16293-2422 with ALMA
• Evolutionary study of complex organic molecules in high-mass star-forming regions
• The Detection of Hot Cores and Complex Organic Molecules in the Large Magellanic Cloud
• History of the solar-type protostar IRAS16293-2422 as told by the cyanopolyynes
• A simple prescription for simulating and characterizing gravitational arcs
• Modeling and Analysis Generic Interface for eXternal numerical codes (MAGIX)
• Organic molecules, ions, and rare isotopologues in the remnant of the stellar-merger candidate, CK Vulpeculae (Nova 1670)
• Machine Learning of Interstellar Chemical Inventories
• Predicting binding energies of astrochemically relevant molecules via machine learning
• Small-scale physical and chemical structure of diffuse and translucent molecular clouds along the line of sight to Sgr B2
• Complex organic molecules detected in twelve high-mass star-forming regions with Atacama Large Millimeter/submillimeter Array (ALMA)
• The Detection of Hot Molecular Cores in the Small Magellanic Cloud
• Active galactic nuclei synapses: X-ray versus optical classifications using artificial neural networks
• Implementation of Rare Isotopologues into Machine Learning of the Chemical Inventory of the Solar-Type Protostellar Source IRAS 16293-2422
• Reexploring Molecular Complexity with ALMA: Insights into chemical differentiation from the molecular composition of hot cores in Sgr B2(N2)
• The physical and chemical structure of Sagittarius B2, VIII. Full molecular line survey of hot cores
• First detection of CF+ towards a high-mass protostar
• CH${}_3$OH as a User-Friendly Density Probe: Calibration and Beyond
• Complex Organic Molecules towards the central molecular zone of NGC 253
• Unmasking the physical information inherent to interstellar spectral line profiles with Machine Learning. I. Application of LTE to HCN and HNC transitions