Regulation of Star Formation Rates in Multiphase Galactic Disks: Numerical Tests of the Thermal/Dynamical Equilibrium Model

TL;DR

This study uses numerical simulations to test a self-regulated star formation model in galactic disks, demonstrating how feedback mechanisms maintain equilibrium states and influence star formation rates.

Contribution

It provides a detailed numerical validation of a new theory linking star formation rates to vertical dynamical and thermal equilibrium in multiphase galactic disks.

Findings

Star formation correlates with Sigma*(rho_sd)^1/2, not just Sigma.

Turbulent velocity dispersions are ~7 km/s, matching observations.

Star formation is highly efficient in energy and momentum production.

Abstract

We use vertically-resolved numerical hydrodynamic simulations to study star formation and the interstellar medium (ISM) in galactic disks. We focus on outer disk regions where diffuse HI dominates, with gas surface densities Sigma_SFR=3-20 Msun/kpc^2/yr and star-plus-dark matter volume densities rho_sd=0.003-0.5 Msun/pc^3. Star formation occurs in very dense, cold, self-gravitating clouds. Turbulence, driven by momentum feedback from supernova events, destroys bound clouds and puffs up the disk vertically. Time-dependent radiative heating (FUV) offsets gas cooling. We use our simulations to test a new theory for self-regulated star formation. Consistent with this theory, the disks evolve to a state of vertical dynamical equilibrium and thermal equilibrium with both warm and cold phases. The range of star formation surface densities and midplane thermal pressures is Sigma_SFR ~ 0.0001 - 0.01 Msun/kpc^2/yr and P_th/k_B ~ 100 -10000 cm^-3 K. In agreement with observations, turbulent velocity dispersions are ~7 km/s and the ratio of the total (effective) to thermal pressure is P_tot/P_th~4-5, across this whole range. We show that Sigma_SFR is not well correlated with Sigma alone, but rather with Sigma*(rho_sd)^1/2, because the vertical gravity from stars and dark matter dominates in outer disks. We also find that Sigma_SFR has a strong, nearly linear correlation with P_tot, which itself is within ~13% of the dynamical-equilibrium estimate P_tot,DE. The quantitative relationships we find between Sigma_SFR and the turbulent and thermal pressures show that star formation is highly efficient for energy and momentum production, in contrast to the low efficiency of mass consumption. Star formation rates adjust until the ISM's energy and momentum losses are replenished by feedback within a dynamical time.