There is evidence that dark matter constitutes a majority of the Universe's matter content. Yet, we are ignorant about its nature. Understanding dark matter requires new physics, possibly in the form of a new species of fundamental particles. So far, the evidence supporting the existence of dark matter is purely gravitational, ranging from mass measurements on galactic scales, to cosmological probes such as the cosmic microwave background radiation. For many proposed models of particle dark matter, the strongest constraints to its properties do not come from particle collider or direct detection experiments on Earth, but from the vast laboratory of space. This thesis focuses on such extra-terrestrial probes, and discusses three different indirect signatures of dark matter.

(1) A first part of this thesis is about the process of dark matter capture by the Sun, whereby dark matter annihilating in the Sun's core could give rise to an observable flux of high-energy neutrinos. In this work, I was the first to thoroughly test the common assumption that captured dark matter particles thermalise to the Sun's core temperature in negligible time. I found that the thermalisation process is short with respect to current age of the Sun, for most cases of interest. (2) A second part concerns a radio signal associated with the epoch when the first stars were born. A measurement of this signal indicated an unexpectedly low hydrogen gas temperature, which was speculated to be explained by cooling via dark matter interactions. In my work, I proposed an alternative and qualitatively different cooling mechanism via spin-dependent dark matter interactions. While bounds coming from stellar cooling excluded significant cooling for the simple model I considered, perhaps the same cooling mechanism is allowed in an alternative dark matter model. (3) Thirdly, a significant part of this thesis is about the mass distribution of the Galactic disk, which can be measured by analysing the dynamics of stars under the assumption of equilibrium. Although most of the matter in the Galactic disk is made up of stars and hydrogen gas, exact measurements can still constrain the amount of dark matter. Potentially, dark matter could form a dark disk that is co-planar with the stellar disk, arising either from the Galactic accretion of in-falling satellites or by a strongly self-interacting dark matter subcomponent. Together with my collaborators, I made significant progress in terms of the statistical modelling of stellar dynamics. I measured the matter density of the solar neighbourhood using Galactic disk stars and data from the Gaia mission. I found a surplus matter density close to the Galactic mid-plane, with respect to the observed baryonic and extrapolated dark matter halo densities. This result could be due to a dark disk structure, a misunderstood density of baryons, or due to systematics related to the data or equilibrium assumption. I also developed an alternative method for weighing the Galactic disk using stellar streams. This method does not rely on the same equilibrium assumption for stars in the Galactic disk, and will be used to provide a complementary mass measurement in future work.

The different indirect probes of dark matter discussed in this thesis span a great range of spatial scales − from stellar interactions relevant to our own solar system, to the matter distribution of the Milky Way, and even cosmological signals from the dawn of the first stars. Through the macroscopic phenomenology of dark matter, the microscopic particle nature of dark matter can be constrained. Doing so is a window into new physics and a deeper understanding of the Universe we live in.