| dc.description.abstract | The gravitational interaction around the event horizon of black holes presents theoretical challenges. With the advent of the Event Horizon Telescope (EHT), we are now
entering an era in physics where we can probe the structure of spacetime on horizon scales.
The EHT presents the first opportunity to directly image the supermassive black holes at
the center of the Milky Way and M 87. By imaging the central black hole, we can directly
learn about the nature of spacetime and plasma physics on horizon scales.
The black hole images produced by the EHT are dominated by a bright ring. The
ellipticity of the ring could potentially signal deviations from general relativity. However,
whether the EHT imaging techniques can robustly detect ellipticity has not been fully
explored. Chapters 2–5 analyze the EHT’s ability to measure ellipticity in four parts.
First, in Chapter 2, we develop a method to extract image features (e.g., ring ellipticity)
called variational image domain analysis. Second, in Chapter 3, we apply variational
image domain analysis to the M 87 image reconstruction pipeline and demonstrate that it
is unable to measure ellipticity. The core reason for this failure is that traditional radio
imaging techniques cannot quantify image uncertainty. To solve this issue, in Chapters 4
and 5 we use Themis, a Bayesian parameter estimation framework for the EHT, to robustly
measure the ellipticity of M 87. To apply Themis to the problem of Bayesian imaging, we
developed a new sampler interface in Chapter 4. In Chapter 5 we apply Themis to M 87
and construct the first Bayesian estimates of its ellipticity. Furthermore, we demonstrate
that the measured ellipticity is consistent with the expected ellipticity from an accretion
disk around a Kerr black hole.
In Chapter 6 we describe a novel method to measure spacetime around Sgr A∗ using
hot spots. While M 87 is static over an observation, Sgr A∗ is dynamic, changing on
minute timescales. Furthermore, Sgr A∗ flares 1–3 times a day in sub-mm, infrared, and
X-ray. The Gravity Collaboration recently demonstrated that hot spots near the innermost
stable circular orbit explain Sgr A∗ flares. Using Themis, we construct an efficient semi-
analytical model of hotspots and fit simulated Sgr A∗ data from the 2017 EHT observations.
We demonstrate that the EHT could potentially make a sub-percent spin measurement of
Sgr A∗ by tracking the evolution of these flares. Furthermore, by observing multiple flares,
we can tomographically map spacetime around Sgr A∗ , providing a test of general relativity
in the strong-field regime. | en |
