Playing with FIRE: A Galactic Feedback-Halting Experiment Challenges Star Formation Rate Theories

TL;DR

This study uses a FIRE-2 simulation to demonstrate that stellar feedback critically regulates star formation rates and molecular cloud properties, revealing that feedback suppression leads to significant increases in star formation and cloud density.

Contribution

It introduces a feedback-halting experiment in galaxy simulations, showing feedback's direct role in moderating star formation efficiency and cloud evolution beyond density effects.

Findings

Feedback suppression causes a 15-20 fold increase in SFR.

Halting feedback leads to more bound and denser molecular clouds.

Existing turbulence regulation theories underpredict the SFR surge.

Abstract

Stellar feedback influences the star formation rate (SFR) and the interstellar medium of galaxies in ways that are difficult to quantify numerically, because feedback is an essential ingredient of realistic simulations. To overcome this, we conduct a feedback-halting experiment starting with a Milky Way-mass galaxy in the FIRE-2 simulation framework. Terminating feedback, and comparing to a simulation in which feedback is maintained, we monitor how the runs diverge. We find that without feedback, interstellar turbulent velocities decay. There is a marked increase of dense material, while the SFR increases by over an order of magnitude. Importantly, this SFR boost is a factor of $\sim$15-20 larger than is accounted for by the increased free fall rate caused by higher densities. This implies that feedback moderates the star formation efficiency per free-fall time more directly than simply through the density distribution. To probe changes at the scale of giant molecular clouds (GMCs), we identify GMCs using density and virial parameter thresholds, tracking clouds as the galaxy evolves. Halting feedback stimulates rapid changes, including a proliferation of new bound clouds, a decrease of turbulent support in loosely-bound clouds, an overall increase in cloud densities, and a surge of internal star formation. Computing the cloud-integrated SFR using several theories of turbulence regulation, we show that these theories underpredict the surge in SFR by at least a factor of three. We conclude that galactic star formation is essentially feedback-regulated on scales that include GMCs, and that stellar feedback affects GMCs in multiple ways.