The physics of multiphase gas flows: fragmentation of a radiatively cooling gas cloud in a hot wind

TL;DR

This paper uses high-resolution simulations to study how radiative cooling causes large cold gas clouds in galactic winds to fragment, revealing important physics for modeling the circumgalactic medium and related phenomena.

Contribution

It demonstrates that radiative cooling induces fragmentation of cold clouds in hot winds, with detailed analysis of the process in 2D and 3D simulations, highlighting the need for high resolution.

Findings

Cooling causes clouds larger than the cooling length to fragment.

Fragmentation affects the density power spectra and cloud dissolution.

State-of-the-art cosmological simulations lack sufficient resolution to capture this physics.

Abstract

Galactic winds exhibit a multiphase structure that consists of hot-diffuse and cold-dense phases. Here we present high-resolution idealised simulations of the interaction of a hot supersonic wind with a cold cloud with the moving-mesh code arepo in setups with and without radiative cooling. We demonstrate that cooling causes clouds with sizes larger than the cooling length to fragment in two- and three-dimensional simulations (2D and 3D). We confirm earlier 2D simulations by McCourt et al. 2018 and highlight differences of the shattering processes of 3D clouds that are exposed to a hot wind. The fragmentation process is quantified with a friends-of-friends analysis of shattered cloudlets and density power spectra. Those show that radiative cooling causes the power spectral index to gradually increase when the initial cloud radius is larger than the cooling length and with increasing time until the cloud is fully dissolved in the hot wind. A resolution of around 1 pc is required to reveal the effect of cooling-induced fragmentation of a 100 pc outflowing cloud. Thus, state-of-the-art cosmological zoom simulations of the circumgalactic medium (CGM) fall short by orders of magnitudes from resolving this fragmentation process. This physics is, however, necessary to reliably model observed column densities and covering fractions of Lyman-$\alpha$ haloes, high-velocity clouds, and broad-line regions of active galactic nuclei.