Speaker: Xiaolong Du, Carnegie Observatories & UCLA
Abstract: Roughly 85% of the matter in the Universe is in the form of invisible matter, i.e. dark matter. Under gravitational interactions, dark matter clusters and forms hierarchical structure. Over the last decades, simulations with higher and higher resolutions have greatly improved our understanding of the formation of dark matter halos and the properties of their substructure (subhalo). Simulations of the standard cold dark matter (CDM) show that CDM halos have cuspy density profiles and contain a large number of small subhalos, which are not fully consistent with the current observations of dwarf galaxies. Thus different dark matter models, such as warm dark matter, self-interacting dark matter, fuzzy dark matter, have been proposed, which predict very different phenomena on small scales (~kpc). Running numerical simulations under different dark matter assumptions requires a large amount of computing resource. On the hand, even the state-of-art simulations suffer from numerical artifacts, limiting their predicting power on small scales. In this talk, I will talk about some of our latest progresses in modeling the evolution of subhalos using the semi-analytic code Galacticus and how we can model and control the numerical artifacts. Using our semi-analytic models, we can get accurate predictions for the abundance of subhalos and the evolution of their density profiles due to tidal effects. Our semi-analytic models are ~10^4 times faster than direct simulations, making it possible to generate a large number of realizations of dark matter halo systems with correct substructure, which are useful for getting robust constraint on different dark matter models from observations such as strong gravitational lensing.