Show simple item record

dc.contributor.advisorSzalay, Alexander S.
dc.creatorYang, Lin
dc.date.accessioned2018-05-22T03:39:24Z
dc.date.available2018-05-22T03:39:24Z
dc.date.created2017-12
dc.date.issued2017-08-11
dc.date.submittedDecember 2017
dc.identifier.urihttp://jhir.library.jhu.edu/handle/1774.2/58624
dc.description.abstractIn the successful concordance model of cosmology, dark matter is crucial for structures to form as we observe it in the universe. Despite the overwhelming observational evidence for its existence, it is not yet directly detected, and its nature is largely unknown. Physicists propose various dark matter candidates, with masses ranging over dozens of orders of magnitude. However, both indirect and direct detection experiments for dark matter have reported no convincing results. Dark matter research is therefore critically relying on computer simulations. Using supercomputer numerical simulations, we can test the correctness of the current cosmological model, as well as obtain guidance for future detection experiments. In this dissertation, we study dark matter from several perspectives using cosmological simulations: its possible radiation, its warmth, and other related issues. A commonly accepted candidate for dark matter is the weakly interacting massive particle (WIMP). WIMPs interact with normal matter only through the weak force (as well as gravity). It is thus extremely challenging to detect these particles directly. However, depending on the type of dark matter, they can %self-annihilate annihilate with other dark matter particles, or decay into high-energy photons (i.e., $\gamma$-ray). We studied the spatial distribution of possible emission components from dark matter annihilation or decay in a large simulation of a galaxy like the Milky Way. The predicted emission components can be used as templates for observations such as those from the {\it Fermi}/LAT $\gamma$-ray instrument, to constrain for the physical properties of dark matter. Structure formation theory suggests that dark matter is ``cold'', i.e., moving non-relativistically during structure formation. However, cold dark matter predicts many more dark-matter satellites, or subhaloes, around galaxies such as the Milky Way than observed. One well-established mechanism to bring the theory in line with observations is that many of these satellites are not visible because they are too small for baryons to form stars in them. Another way is to attenuate the small-scale structure directly, positing ``warm'' dark matter. Using simulation, we propose a method of testing this possibility in a complementary environment, by measuring the density profile of cosmic voids. Our results suggest that there are sufficient differences between warm and cold dark matter to test using future observations. Furthermore, our data analyzing methods are based on sophisticated data stream algorithms and newly developed Graphic Process Unit (GPU) hardware. These tools lead to other studies of dark matter as well. For example, we studied the spin alignment of dark matter halos with its environment. We show that the spin alignments are highly related to the hierarchical levels of the cosmic web, in which the halo is located. We also studied the responses in different density variables to ``ringing'' the initial density field at different spatial frequencies (i.e. putting spikes in the power spectrum at a particular scale). The conventional wisdom is that power generally migrates from large to small comoving scales from the initial to final conditions. But in this work, we found that this conventional wisdom is only true for a density variable emphasizing dense regions, such as the usual overdensity field. In the log-density field, however, power stays about at the same scale but broadens. In the reciprocal-density field, emphasizing low-density regions, power moves to larger scales. This is an example of voids as ``cosmic magnifying glasses.'' The GPU density-estimation technique was crucial for this study, allowing the density to be estimated accurately even when modestly sampled with particles. Our results provide guidance for designing future statistic analytics for dark matter and the large-scale structure of the Universe in general.
dc.format.mimetypeapplication/pdf
dc.language.isoen_US
dc.publisherJohns Hopkins University
dc.subjectDark Matter
dc.subjectCosmological Simulation
dc.subjectCDM
dc.titleINVESTIGATIONS OF DARK MATTER USING COSMOLOGICAL SIMULATIONS
dc.typeThesis
thesis.degree.disciplinePhysics
thesis.degree.grantorJohns Hopkins University
thesis.degree.grantorKrieger School of Arts and Sciences
thesis.degree.levelDoctoral
thesis.degree.namePh.D.
dc.date.updated2018-05-22T03:39:25Z
dc.type.materialtext
thesis.degree.departmentPhysics and Astronomy
dc.contributor.committeeMemberWyse, Rosemary
dc.contributor.committeeMemberBraverman, Vladimir
dc.contributor.committeeMemberBudavári, Tamás
dc.contributor.committeeMemberBroholm, Collin L.
dc.publisher.countryUSA


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record