Abstract
Stellar feedback, such as stellar winds and outflows, plays a significant role in both physical and chemical evolution of molecular clouds. This energy and momentum leave identifiable signatures (bubbles and outflows) that affect the dynamics and structure of the cloud. Most feedback feature searches are performed "by-eye", which are usually time-consuming, subjective and difficult to calibrate. Automatic classifications based on machine learning make it possible to perform systematic, quantifiable and repeatable searches for stellar feedback features. I will introduce a new deep learning method CASI-3D (Convolutional Approach to Structure Identification-3D) to identify stellar feedback signatures in molecular line spectra. The CASI models are able to identify all previously identified feedback features in four nearby molecular clouds (Ophiuchus, Taurus, Perseus and Orion), and identify new feedback structures as well. Meanwhile, the CASI models indicate that the mass, momentum and energy from stellar wind-driven bubbles are overestimated by a large factor in previous studies. I will show the linear correlation between the outflow mass and the total number of YSOs for all four clouds. I will present the broken power law in the spatial power spectrum of the outflow regions. The break point might indicate the typical outflow mass and energy injection scale. I also compare the energy associated with outflows to the rate of turbulent dissipation and find that feedback is sufficient to maintain turbulent dissipation at the current epoch for all four clouds.