On the caveats of tracing molecular gas with CO emission
The kinematics, masses and column density distributions of molecular clouds are fundamental parameters and indicators of the star formation process. Often these quantities are inferred from the observation of carbon monoxide (CO) isotope emission. A number of simplifications and assumptions, such as fixed CO abundance and isotope ratios, a single line of sight excitation temperature, local thermal equilibrium (LTE) excitation, virial equilibrium, etc., are involved in the inference methods.
By combining hydrodynamic simulations of turbulent self-gravitating molecular clouds, which include a detailed treatment of chemistry and thermal balance, with a non-LTE line radiative transfer model, these assumptions can be tested, one by one, and the induced systematic error can be estimated as a function of cloud properties (e.g. metallicity, interstellar radiation field strength, cosmic ray rate, overall virial state and cloud masses).
This talk tries to raise the attention and quantify the systematic errors in molecular cloud mass estimates, obtained by commonly used observational methods, especially focusing on the virial state of the CO bright molecular gas and the so-called XCO conversion factor. The traceability of molecular gas in regions irradiated by a high flux of cosmic ray particles is also addressed.
If time allows, I will discuss briefly the caveats of the semi-analytical calculation of dust temperatures in hydrodynamic simulations, compare the common expressions to observations and suggest improvements.