Impact of dark matter sub-haloes on the outer gaseous discs of galaxies

Abstract

Late-type disc galaxies are observed to have extended gas discs that are significantly larger than their optical discs. The cold dark matter ( CDM) cosmological model predicts that luminous galaxies, such as the Milky-Way (MW), are hosts to a large number of low-mass dark matter sub-haloes (DMSHs), which are capable of interacting with the discs of galaxies. We aim to understand the kinematic impact of a DMSH collision with the extended gas disc of a galaxy. In order to study this, we use hydrodynamical simulations with a variety of parameters such as galaxy inclination, DMSH mass, and galaxy mass. To analyse the simulation results, we present two-dimensional (2D) kinematic and gas density maps of galaxies to investigate their unique features and structures. Along with this, we also implement a deep learning (DL) model to classify kinematic images of the simulated galaxies as either impacted (by the DMSH) or isolated. Through the use of these techniques, we have found that the impact of the DMSH causes features in both the 2D kinematic and gas density plots, most notably the development of a ∼1 kpc size H I hole in the gas-rich DMSH impact. Through the study of these maps, we identify certain features and the specific conditions required to induce them. The study of these features is important, as they can be observed with large upcoming H I surveys such as WALLABY and SKA.

Accretion of small satellites and gas inflows in a disc galaxy

“…Our results complement this work in the sense that even smaller satellites can still produce important effects in the morphology. In terms of triggered spiral arms, our results also agree qualitatively with those of the simulations of Chakrabarti et al (2011) and Shah et al (2019), who used satellites in the range of R ≈ 1 : 100 − 1 : 1000.…”

“…Other simulations have explored the effect of minor mergers with mass ratios in the range of ∼1:10 and their effects in the primary's morphology (e. g. Dobbs et al 2010;Purcell et al 2010;Purcell et al 2011;Qu et al 2011), while others those in the range 1 : 100 − 1 : 1000 (e. g. Chakrabarti et al 2011;Shah et al 2019). Our work shows and agrees with the latter ones in the sense that even satellite galaxies, smaller than roughly the size of the Small Magellanic Cloud, can produce lopsided density distributions in galaxies similar to the Milky Way.…”

“…The previous examples motivate a deeper analysis of minor mergers, but the study of these interactions has progressed at a somewhat slower rate. Several works have shown that they can lead to detectable and important morphological features in the evolution of spiral galaxies such as lopsidedness (e. g. Bournaud et al 2005), grand-design spiral arms (e. g. Dobbs et al 2010;Pettitt et al 2016;Pettitt & Wadsley 2018) as well as disturbed spiral arms (e. g. Starkenburg et al 2016), ring-like features in S0 galaxies (e. g. Mapelli et al 2015), as well as hole features in the HI distribution (e. g. Bekki & Chiba 2006;Kannan et al 2012;Shah et al 2019). Cox et al (2008) showed that the triggered starburst decreases when the satellite's mass is reduced.…”

“…Other orbits may be less effective in moving such gas quantities to the inner regions. Recent works have focused on reproducing features in spiral galaxies expected to be produced by minor interactions (e. g. Dobbs et al 2010;Chakrabarti & Blitz 2009;Chakrabarti et al 2011;Pettitt et al 2016;Hu & Sijacki 2018;Shah et al 2019).…”

Classifying the formation processes of S0 galaxies using Convolutional Neural Networks

“…In addition to replicating the abilities of human scientists, neural networks may also be used to extend those abilities into new realms by addressing scientific problems which would otherwise lie beyond reach. Recent work in this vein includes the use of CNNs to detect dark matter subhalos by their kinematic imprints in discs (Bekki et al 2019;Shah et al 2019), and using CNNs to constrain the orbits of cluster galaxies from the properties of stripped gas (Bekki 2019). In this paper, we contribute to this trend by training CNNs to classify S0 galaxies according to their formation processes.…”