I am a Ph.D. student at the Department of Astronomy and Theoretical Physics of Lund University with the goal of figuring out how galaxies form and evolve in our Universe. To do this I investigate how simluations of galaxies are effected by different physical processes as well as in different environments. I am also interested in comparing my results to data from large surveys of our Milky Way, in order to understand the detailes of how our own galaxy came to be.
In my current project I am testing how moving supernovae projenitors to low density gas, i.e., tracing trajectories of runaway OB stars, affects global properties of star forming galaxies. [Spoiler alert] The increased energy coupling strength for supernovae explosions in low density regions leads to the driving of a strong galactic winds in my simulations including massive runaway stars.
If you are interested in the work I do, don't hessitate to contact me.
Massive OB stars can be ejected out of their birth environment through dynamical scattering (Poveda et al. 1967) or by kicks produced as a result of the supernovae (SN) explosion of a binary companion (Blaauw 1961). I refer to these stars as runaways. Typically 30% of the OB population belong to this class of objects. Using estimates of their velocity distribution and models for their main sequence lifetime these stars will on average travel ~100 pc before exploding as SN type II.
Because runaway stars can travel significant distances in the galaxy before SN, they will inject energy into a gas phase which differ from the cold molecular phase where stars form. In cases where the medium adjensent to a SN is diffuse, e.g. in the interarm regions, the energy couples very efficiently to ISM leading to a significant increase in temperature. This heating process will cause an adiabatic expansion of the gas, driving a very strong galactic outflow. Current models of galaxy evolution completely ignore this effect.
To adress this issue and quantify the significanse of runaway stars on global evolution of galaxies I have implemented a model for these stars in simulations of entire galaxies. The model includes a new star-by-star treatment for massive stars and has been implemented in the hydro+N-body code RAMSES (Teyssier 2002). The right figure shows shows two simulations of a Milky Way-like disc, both including the same physics model but one taking runaway star into account and one ignoring this effect.
As hypothesised above, the runaway stars result in a strong galactic wind for the Milky Way model, however, since the runaway stars affect the low-density gas, much of the cold molecular phase is undisturbed. This implies that the star formation rate (SFR) is unchanged. However, the prolonged heating in low-density eventually reduces the total gas mass budget leading a reduction in SFR. Nonetheless, the runaway stars leads to a five-fold increase in mass loading factor. If you are interested in a detailed description of the effect of runaway stars check out for my paper (Andersson et al., 2020)
Left figure: The luminosity of the galaxy after 200 Myr of evolution both face-on (upper) and edge-on (lower). The model taking runaway stars into account is shown in the left panels, while the model ignoring runaway stars is shown in the right panels.
Right figure: Upper subplot shows the star formation rate while the lower subplot shows vertical gas outflows measured by taking the average flow in cells between 20 and 40 kpc above and below the disc plane. Both shows this as function of time in the simulation. Note that the first 80 Myr are the result of the simulation settling to the resolution of the simulation and should not be considered physical.
The globular cluster (GC) population of large star forming galaxies can be divided into two different populations: one which formed in-situ and one compriesed of GCs accreted from other galaxies. In this work I studied the GC population of the M31 galaxy, with a focus on the peculiar cluster MGC1. This object is one of the most isolated clusters known in the local group positioned at 200 kpc from the centre of M31. Understanding the origin of this cluster is challening since its remote location prohibit it from forming here. The hypothesis we tested is to have MGC1 form in a dwarf galaxy from which it was tidally stripped away during a close encounter with M31. The challenge is to leave to dwarf galaxy intact since to sign of tidal debris is found in the region.
In this work we set up a simple model for the potential of M31 and a moving dwarf galaxy in which GC where placed. The position of the dwarf galaxy was integrated in time to simulate an encounter and the resulting positions of tidally accreted globular clusters. We divied the clusters into 4 different classes: 1) Cluster which where tidally stripped from the dwarf and captured by M31 (Captured); Clusters retained by the dwarf galaxy (Retained); Clusters which where unbound from both M31 and the dwarf (Unbound); Cluster in the Captured population on orbits which at some point place the clusters a the orbital distance of MGC1 (MGC1-like). We tested a broad range of dwarf galaxy encounter to determine the likelyhood of producing MGC1-like clusters depending on initial orbital parameters of the dwarf galaxy.
What we find is that even for resonable orbital parameters of the dwarf galaxy there is a significant likelyhood of producing MGC1-like clusters (up to 20% for ideal cases.) Furthermore, using radial velocity measurements of the observed GCs and satellite galaxies we determine that MGC1-likely originates from the dwarf galaxy And XXV if it has been accreted from a dwarf which still exists in M31 today.
Furthermore, motivated be our simluations we device a method for determining the number distribution of in-situ versus accreted GCs with simple statistical arguments. Using this method we find that even close in roughly 50% of the M31 GC population is accreted, whilst this figure rises to more than 80% far out. If you are interested in the details of this study then check out my paper Andersson & Davies (2019).