Probing Disruption Rates and Accretion Histories in High-Resolution CDM and WDM Simulations

Abstract

The true nature of dark matter is still not fully understood. The low mass scale structures in dark matter simulations are not fully aligned with observations yet, and new simulations with high enough resolution to distinguish these low mass scale structures are vital. In this work we analyzed a new set of high resolution dark-matter-only (DMO) simulations, comparing cold dark matter (CDM), 7.5 keV warm dark matter (WDM), and 3.0 keV WDM. The subhalo population of the host halo in CDM followed the expected power law of the subhalo mass function (SHMF), yielding α= 1.88 and confirming the accurate reproduction of CDM subhalo populations. We find a half-mode mass of (1.21 ± 0.09) × 107M⊙ and (2.63 ± 0.89) × 108M⊙ for the 7.5 keV and 3.0 keV model respectively. Tracking the tidal disruption of subhalo populations from z=2 onward showed that CDM had the highest initial subhalo count with 1710 followed by the 7.5 keV model with 582 and the 3.0 keV model with
182. The surviving fraction of these halos at z=0 was 20% for CDM, 24% for the 7.5 keV WDM and 9% for the 3.0 keV WDM. The accretion history of the main halo was analyzed and resulted in an dark matter (DM) ex-situ fraction of 66.3% for CDM, 67.2% for the 7.5 keV model and 57.4% for the 3.0 keV model, showing no significant relative difference between DM models compared to the stellar ex-situ fraction. The biggest limitation of the project is the low subhalo count in WDM and the resulting susceptibility to statistical effects, and the analysis was only done on one main halo, meaning its specific merger history potentially affected the results. Implementing stellar components into these simulations will give insights into observable signatures that future observations can use to constrain properties of dark
matter.

Popular Abstract

Around 85% of all the matter in the universe is invisible. It does not glow and does not absorb light. No telescope has ever directly detected it. However, we are very sure that it is there, because without it galaxies would never have been able to form. The ordinary visible
matter could not have clumped together fast enough on its own during the early universe. Understanding the nature of this so called dark matter is one of the central questions in modern physics.
The current assumption is that dark matter particles are cold, meaning that they barely moved during the early evolution of the universe, which made them very susceptible to gravity. This cold dark matter model is extremely successful at predicting how matter is distributed over a large scale. But it also predicts a lot more satellite galaxies around the Milky Way than we have actually observed. This tension has given rise to alternative particle models. One promising alternative is warm dark matter. Unlike how cold dark matter was not moving
during the early universe, warm dark matter has some thermal energy and can thus escape the gravitational pull of the smallest clumps and prevent them from ever forming. This has the effect that, in warm dark matter models, we expect to find a lot fewer satellites. Depending
on the particle mass, the scale at which structure formation is suppressed varies. This thesis looks at the differences between the cold dark matter model and two different warm
dark matter models, specifically comparing how galaxies like the Milky Way have formed. A new set of high-resolution dark matter simulations lets us look at these small clumps and satellites where we expect differences between the models. We track billions of particles and
compare where these particles came from, and study the growth history and internal structure of a simulated galaxy and how they change between the different models. The end goal is to determine what future telescopes may be able to detect, helping to understand the underlying
physics of dark matter better.
3

• author: Schuster, Florian
• supervisor: Santi Roca Fàbrega; Oscar Agertz
• organization: Department of Physics; Astrophysics
• course: FYSK04 20261
• year: 2026
• type: M2 - Bachelor Degree
• subject: Physics and Astronomy
• report number: 2026–EXA259
• other publication id: 2026–EXA259
• language: English
• id: 9246587
• date added to LUP: 2026-07-13 09:40:08
• date last changed: 2026-07-13 09:40:08