Events

Modeling emission from super-Eddington supermassive black holes

Tidal disruption events (TDEs) light up hidden supermassive black holes and encode information about the
properties of both these black holes and their environments. They also regularly produce super-Eddington
mass return rates, providing a local test of extreme black hole growth. The light curves and spectra of TDEs
show signatures of these super-Eddington feeding rates, including evidence of large outflows, very high
electron densities, and strongly ionizing but obscured emission. Many of these signatures are also seen in
other candidates for super-eddington accretion, including luminous fast blue optical transients, ambiguous
nuclear transients, and little red dots. However, properties of the system such as the origin, mass/energy
budget and black hole mass are often better understood in TDEs. I’ll discuss how we can use theoretical
models to connect the observed light curves and spectra to the actual source emission as well as to the
properties of the gas surrounding the black hole in these transients. More broadly I’ll discuss how we can
combine theoretical models with observations to investigate the black hole mass function, probe emission
from forming and evolving accretion disks, and study the properties of stars on size scales we cannot
observe outside our own galactic nuclei.

Monday, October 19, 2026 11:00 AM

Inperson
G10

Virtual
Zoom

The Approximate Cosmological Principle and Permanent Underdetermination in Cosmology

In this talk, I will discuss the implications of some recent work in observational cosmology and mathematical general relativity (GR) for the epistemology of spacetime. My primary aim is to articulate a non-trivial and physically relevant underdetermination thesis concerning the global, topological structure of models of GR. I will argue that my thesis is philosophically interesting for at least two reasons: (i) it affects “sensible realists,” viz., those who counsel realism only about theoretical claims that are amenable to empirical investigation (Stanford 2006); and (ii) it is “permanent” (in the sense of Pitts 2011). In order to formulate my thesis and to defend my claims, I begin by surveying cosmological research devoted to empirically determining the spatial topology of the universe (e.g., Akrami et al. 2024). This research typically assumes the Cosmological Principle (CP); however, CP is a very strong constraint on our theorizing, and its justification is fraught with difficulties. Indeed, strictly speaking, it is known not to hold, even in the observable universe. As such, it has been suggested that a more accurate reflection of our observational situation would be to assume an “approximate” CP. Drawing on recent mathematical work, I discuss one possible way of making this notion precise (Ringström 2013). Interestingly, when we pass to Ringström’s approximate CP, it appears that one cannot draw any conclusions about the topology of space. I carefully analyze why this occurs, and demonstrate that the resulting underdetermination is permanent. Time permitting, I compare my thesis to related results and sketch how my discussion can serve as a powerful starting point for anti-realist positions in the epistemology of spacetime.

Monday, October 19, 2026 9:30 AM

Inperson
Speaker will be in person – BHI Conference Room

Virtual
Zoom

Monday, October 26, 2026 11:00 AM

Inperson
G10

Virtual
Zoom

Intuition and Algorithms in Observational Astronomy

In this talk, I will use the history of gravitational lensing in observational astronomy to
explore the role of human intuition in an age of algorithmic inference. Between the late
1970s, when the first gravitational lens was observed, and the 2010s, astronomers
discovered only a few hundred strong lenses. To find these sources, scientists relied on
visual intuition. With a vast amount of sky survey data, fast visual analysis (just a few
seconds per image!) was a key aspect in the data-processing pipeline. In a glance, the
trained astronomer could spot a lens while combing through a survey. This is changing.
The launch of Euclid in 2023 marked the onset of a project that forecasted it would find
over 100,000 lenses in its 6-year mission. The oncoming deluge will overwhelm even
the quick glance, and recent work has focused on developing and deploying algorithms
that can automate the search. Yet, this has not spelled the end for the intuitive eye.
Moving from the 1970s to the present, I will trace how astronomers cultivated a specific
form of visual intuition, the well-trained glance, and how this skill has become even
more important within the infrastructure of automated data processing. Recognizing the
persistence of intuition, I contend, raises new questions about how we integrate
algorithms into scientific practice.

Monday, October 26, 2026 9:30 AM

Inperson
Speaker will be in person – BHI Conference Room

Virtual
Zoom

Monday, November 2, 2026 11:00 AM

Inperson
G10

Virtual
Zoom

Monday, November 2, 2026 9:30 AM

Inperson
Speaker will be in person – BHI Conference Room

Virtual
Zoom

Monday, November 9, 2026 11:00 AM

Inperson
G10

Virtual
Zoom

Monday, November 16, 2026 9:30 AM

Inperson
Speaker will be in person – BHI Conference Room

Virtual
Zoom

Monday, November 23, 2026 – Monday, November 30, 2026

Monday, December 7, 2026 11:00 AM

Inperson
G10

Virtual
Zoom

Monday, December 14, 2026 11:00 AM

Inperson
G10

Virtual
Zoom

Monday, December 21, 2026 – Monday, January 18, 2027