Cloud Properties and Correlations with Star Formation in Numerical Simulations of the Three-Phase ISM

TL;DR

This study compares gravity-based and density-based methods to identify clouds in simulations of the three-phase ISM, analyzing their properties, boundedness, and correlation with star formation rates over time.

Contribution

It introduces a detailed comparison of cloud identification methods and their relation to star formation, revealing limitations of traditional virial parameters and the importance of density thresholds.

Findings

Bound clouds have masses ~10^3-10^4 M_solar and densities ~100 cm^-3.

High-mass clouds can be unbound despite low virial parameters.

Star formation efficiency per free-fall time is about 0.4, increasing with density.

Abstract

We apply gravity-based and density-based methods to identify clouds in numerical simulations of the star-forming, three-phase interstellar medium (ISM), and compare their properties and their global correlation with the star formation rate over time. The gravity-based method identifies bound objects, which have masses M ~ 10^3 - 10^4 M_solar at densities n_H ~ 100 cm^-3, and traditional virial parameters alpha_v ~ 0.5 - 5. For clouds defined by a density threshold n_H,min , the average virial parameter decreases, and the fraction of material that is genuinely bound increases, at higher n_H,min. Surprisingly, these clouds can be unbound even when alpha_v < 2, and high mass clouds (10^4 - 10^6 M_solar) are generally unbound. This suggests that the traditional alpha_v is at best an approximate measure of boundedness in the ISM. All clouds have internal turbulent motions increasing with size as sigma ~ 1 km/s(R/ pc)^1/2, similar to observed relations. Bound structures comprise a small fraction of the total simulation mass, with star formation efficiency per free-fall time epsilon_ff ~ 0.4. For n_H,min = 10 - 100 cm^-3, epsilon_ff ~ 0.03 - 0.3, increasing with density. Temporal correlation analysis between SFR(t) and aggregate mass M(n_H,min;t) at varying n_H,min shows that time delays to star formation are t_delay ~ t_ff(n_H,min). Correlation between SFR(t) and M(nH,min;t) systematically tightens at higher n_H,min. Considering moderate-density gas, selecting against high virial parameter clouds improves correlation with SFR, consistent with previous work. Even at high n_H,min, the temporal dispersion in (SFR-epsilon_ff M/t_ff )/<SFR> is ~ 50%, due to the large-amplitude variations and inherent stochasticity of the system.