Application of Convolutional Neural Networks to Identify Stellar Feedback Bubbles in CO Emission

Duo Xu,Stella S. R. Offner,Robert Gutermuth,Colin Van Oort
DOI: https://doi.org/10.3847/1538-4357/ab6607
IF: 5.521
2020-02-12
The Astrophysical Journal
Abstract:We adopt the deep learning method called the Convolutional Approach to Shell Identification (<span class="small-caps">casi</span>) and extend it to 3D (<span class="small-caps">casi-3d</span>) to identify signatures of stellar feedback in molecular line spectra. We use magnetohydrodynamics simulations modeling the impact of stellar winds in a turbulent molecular cloud to generate synthetic <sup>13</sup>CO (<i>J</i> = 1 − 0) observations. We train two <span class="small-caps">casi-3d</span> models: ME1 predicts only the position of feedback, while MF predicts the fraction of the mass coming from feedback in each voxel. We adopt 75% of the synthetic observations as the training set and assess the accuracy of the two models with the remaining data. Both models identify bubbles in simulated data within 5% error. We use bubbles previously visually identified in Taurus in <sup>13</sup>CO to validate the models and show that both perform well on the highest confidence bubbles. Models ME1 and MF predict total feedback gas mass of 2894 <i>M</i> <sub>⊙</sub> and 302 <i>M</i> <sub>⊙</sub>, respectively. After correcting for missing energy due to the limited velocity range, model ME1 predicts feedback kinetic energies of 4.0 <b>×</b> 10<sup>46</sup> erg and 1.5 <b>×</b> 10<sup>47</sup> erg with and without subtracting the cloud velocity gradient. Model MF predicts feedback kinetic energies of 9.6 <b>×</b> 10<sup>45</sup> erg and 2.8 <b>×</b> 10<sup>46</sup> erg with and without subtracting the cloud velocity gradient. Model ME1 predicts bubble locations and properties consistent with previous visual identifications. However, model MF demonstrates that feedback properties computed using visual identifications significantly overestimate feedback impact, due to line-of-sight confusion and contamination from background and foreground gas.
What problem does this paper attempt to address?