Tsukuba Uchu Forum

122nd Uchu Forum

Molecular cloud collisions and machine learning

Shinji Fujita

Osaka Prefecture University


Massive stars are very important for the physical and chemical evolution of the interstellar medium and the galaxy as a whole. The key to their formation is the accretion of mass onto their parent molecular cloud cores. In recent years, the rapid compression caused by supersonic collisions between molecular clouds has been attracting much attention as a major mechanism for the growth of diffuse interstellar to molecular cloud cores and, finally, stars. Evidence of molecular cloud collisions has been reported in nearly 100 objects, ranging from relatively small (single OB-type stars) to large (mini-starbursts). In this talk, I will present observational evidence for examples. After introducing ongoing work on Spitzer bubble identification and molecular cloud near-far distance identification through machine learning, I will discuss the future prospects of construction machine learning systems for molecular cloud collision identification.