Simulations of ram-pressure stripping in galaxy-cluster interactions

¹ Institut für Astro- und Teilchenphysik Universität Innsbruck, Technikerstrasse 25/8, 6020 Innsbruck, Austria
e-mail: dominik.steinhauser@uibk.ac.at
² Heidelberger Institut für Theoretische Studien, Schloss-Wolfsbrunnenweg 35, 69118 Heidelberg, Germany
³ Zentrum für Astronomie der Universität Heidelberg, Astronomisches Recheninstitut, Mönchhofstr. 12-14, 69120 Heidelberg, Germany

Received: 5 November 2015
Accepted: 7 March 2016

Abstract

Context. Observationally, the quenching of star-forming galaxies appears to depend both on their mass and environment. The exact cause of the environmental dependence is still poorly understood, yet semi-analytic models (SAMs) of galaxy formation need to parameterise it to reproduce observations of galaxy properties.

Aims. In this work, we use hydrodynamical simulations to investigate the quenching of disk galaxies through ram-pressure stripping (RPS) as they fall into galaxy clusters with the goal of characterising the importance of this effect for the reddening of disk galaxies. In particular, we compare our findings for the mass loss and evolution of the star formation rate in our simulations with prescriptions commonly employed in SAMs. We also analyse the gaseous wake of the galaxy, focusing on gas mixing and metal enrichment of the intracluster medium (ICM).

Methods. Our set-up employs a live model of a galaxy cluster that interacts with infalling disk galaxies on different orbits. We use the moving-mesh code AREPO, augmented with a special refinement strategy to yield high resolution around the galaxy on its way through the cluster in a computationally efficient way. Cooling, star formation, and stellar feedback are included according to a simple sub-resolution model. Stellar light maps and the evolution of galaxy colours are computed with the stellar synthesis code FSPS to draw conclusions about quenching timescales of our model galaxies.

Results. We find that the stripping models employed in current SAMs often differ substantially from our direct simulations. In most cases, the actual stripping radius of the simulated disk galaxies is larger than assumed in the SAMs, corresponding to an over prediction of the mass loss in SAMs. As long as the disk is not completely stripped in peaks of RPS during pericentre passage, some gas that remains bound to the galaxies is redistributed to the outer parts of disks as soon as the ram pressure becomes weaker again, an effect that is not captured in simplified treatements of RPS. Star formation in our model galaxies is quenched mainly because the hot gas halo is stripped, depriving the galaxy of its gas supply. The cold gas disk is only stripped completely in extreme cases, leading to full quenching and significant reddening on a very short timescale. Depending on the inclination angle, this can light up a galaxy for a few hundred Myrs until all of the gas is stripped or consumed and star formation drops to almost zero, suggesting a typical quenching timescale of ~200 Myr. On the other hand, galaxies experiencing only mild ram pressure actually show an enhanced star formation rate that is consistent with observations. Stripped gas in the wake is mixed efficiently with intracluster gas already a few tens of kpc behind the disk, and this gas is free of residual star formation.