ASTROPHYSICS
The Smallest Galaxies Providing Big Insights into Dark Matter
Principal Investigator:
Dr. Thales Gutcke
Affiliation:
University of Hawaii, Hilo, United States
Local Project ID:
pn73we
HPC Platform used:
SuperMUC-NG at LRZ
Date published:
Introduction
This project targets various central questions in modern astrophysics, including "What is the nature of dark matter?", "How do galaxies form and evolve?" and "Do we understand the extremes of the universe?". Dwarf galaxies provide a natural laboratory for confronting these questions as we will explain below. The formation of dwarf galaxies tracks an extreme situation in various ways. Dwarf galaxies are assumed to be the very first type of galaxy to form in the earliest Universe. This means they form from gas clouds with the simplest chemical composition and begin forming stars at the first moment when this becomes physically possible. This makes them the smallest entities in the Universe that are capable of forming stars. lt also means that studying the smallest dwarf galaxies can tell us about the conditions and physical processes at work in the very beginning of the Universe. In the structural build-up of galaxies, it is assumed that galaxies merge and form hierarchically. This means that small galaxies form first and then merge to create larger galaxies. Thus, dwarf galaxies are the building blocks of all other galaxies. They are also the most abundant type of galaxy in the Universe. All of these characteristics of dwarf galaxies make them highly sensitive to the nature of dark matter, too. The aim of this project is to create a highly accurate cosmological, hydrodynamical simulation model that can produce extremely realistic representations of dwarf galaxies right into the centers, where dark matter models can be tested.
Results and Methods
The first result of this project is a suite of six highly realistic dwarf galaxy simulations. The computational requirements are significant due to the precision gas physics calculations and the fact that every star is created and tracked individually. Of special note are the individual supernova explosions of dying stars that cause atomic bomb-like explosions, heating and moving gas clouds around at high speeds (see Fig. 1). These features make the newly developed code [2] state-ofthe-art, since it employs various numerical and astrophysical modeling techniques that have never before been used in cosmological simulations. The precision we aim for requires that the model is able to resolve individual supernova blast waves. Preliminary work has shown that this necessitates a mass resolution of 4 solar masses, meaning that a single dwarf galaxy is resolved with between 0.5-2 billion resolution elements. Thus, we are constrained to run all simulations at this resolution or better if we want to retain this vital feature of the model. We have carried out six ultra-high resolution cosmological zoom-in simulations of dwarf galaxies in that have masses of a few billion solar masses. These simulations constitute the highest resolution dwarf galaxies run fully cosmologically across the age of the Universe to date. The galaxies were selected to form in isolation and distanced from larger neighbors, the simulations probe universal aspects of cosmological evolution and of the effects of an external radiation field, without being influenced by effects from the environment or infall into larger halos. In the recently published study, we strove to validate the model. We studied the present-day properties of the simulated galaxies and compare the global stellar properties with measurements from Local Group dwarf galaxies. We show that the stellar mass, stellar size, stellar chemical composition, kinematics and morphology are well matched by our simulations [3]. These results provide the basis on which we can proceed to study alternative dark matter in these galaxies.
Figure 1: Gas clouds around a dwarf galaxy forming in the early Universe. Due to supemova explosions in the central region, gas is being accelerated and pushed outwards from the center at high speeds of hundreds of km per second.
Ongoing Research/Outlook
There exists a discrepancy between the inferred dark matter density profiles for observed dwarf galaxies and simulated ones (see Fig. 2). Observations seem to indicate a flatter inner profile that standard simulations produce. This is an ongoing area of active research, and various solutions have been suggested. To retain the current standard cosmological model, internal energetic processes such a supernova and other such processes have been proposed to play an important role in the flattening of the profile. As well, the formation channels for nuclear stellar clusters in the centers of galaxies are still under discussion, with the infall of globular clusters or the in-situ star formation being the main options discussed. In our ongoing work, we study the formation of dark matter density profiles in our simulated dwarf galaxies. Our findings so far show no signs of the formation of a core in the sample of simulated galaxies. However, we see the impact of the addition of gas and supernova in the dark matter profiles in these simulations. For one of the halos, it is possible to see an increase in the dark matter density in the inner regions. We argue that this is explained by a significant star formation event that might also be linked to the subsequent formation of a nuclear stellar cluster in the central region. While the flatter profiles we see in standard cosmology may be still reconcilable with better understanding the gas physics, self-interacting dark matter (SIDM) is a favored family of dark matter models that may provide an alternative explanation for inconsistencies. However, since gas physics is equally active in an SIDM universe, it remains crucial to test these models with the same set of detailed baryonic processes. Selfinteracting dark matter (SIDM) postulates new gauge bosons, which opens up a hidden sector of dark matter models. The boson facilitates annihilation and selfscattering between dark matter particles, altering the distribution. Our model is the ideal testbed for this study, since the small-scale gas physics is highly resolved. We have already performed the dark matter-only simulations and there is a clear flattening of the dark matter profile in the SIDM simulation (see Fig. 2, red region). We expect to present the results including gas physics that predict the dark matter profile shape for galaxies 4 orders of magnitude smaller than recent studies [4].
Figure 2: The centers of dwarf galaxies are a sensitive probe of different dark matter models. Our preliminary results show a lower central density for an alternative model: self-interacting dark matter.
References and Links
[1] https://thalesada.github.io/lyra
[2] Gutcke T. A., Pakmor R., Naab T., Springei V., 2021, MN RAS, 501, 5597.
[3] Gutcke T. A., Pakmor R., Naab T., Springei V., 2022, MN RAS, 513, 1372.
[4] Correa C. A., Schaller M., Ploeckinger S., Anau Montel N., Weniger C., Ando S., 2022, MN RAS, 517,3045.