BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Hierarchical Bayesian inference: constraining population distribut
 ion of dark matter halo shapes via stellar streams - David Chemaly (IoA)
DTSTART:20260216T160000Z
DTEND:20260216T170000Z
UID:TALK244180@talks.cam.ac.uk
CONTACT:65128
DESCRIPTION:Stellar streams\, the debris of tidally disrupted satellites\,
  trace their host’s gravitational potential and thus probe dark matter h
 alo structure. While six dimensional phase-space data of Galactic streams 
 enable precise dark matter halo modelling in the Milky Way\, streams aroun
 d external galaxies are typically available only as low surface brightness
  features without kinematics (i.e. two-dimensional photometric data)\, pro
 viding only weak constraints when considered individually. We present a hi
 erarchical Bayesian framework that infers the population distribution of h
 alo flattening using only projected stream tracks. Streams are forward-mod
 elled in StreaMAX\, a new JAX-accelerated particle-spray package that achi
 eves orders of magnitude faster stream generation when compared to traditi
 onal methods. For each stream we fit an axisymmetric dark matter halo mode
 l and obtain a posterior on the flattening. These posteriors are then comb
 ined through hierarchical reweighting to constrain the\npopulation distrib
 ution. Using mock data\, we show that individual fits recover the correct 
 flattening with modest precision and exhibit projection-induced multi-moda
 lities. Nevertheless\, aggregating these fits yields accurate and confiden
 t constraints on the underlying population distribution of dark matter hal
 o morphologies\, clearly distinguishing between oblate\, spherical\, and p
 rolate\npopulations. The total computational cost scales linearly with sam
 ple size. Our results demonstrate that ensembles of purely photometric str
 eams carry sufficient information to constrain dark matter halo shapes in 
 external galaxies at the population level. With the forthcoming samples fr
 om Euclid and Rubin/LSST\, this approach offers a practical path to popula
 tion-level inferences of halo morphology without any kinematic measurement
 s.
LOCATION:Martin Ryle Seminar Room\, KICC
END:VEVENT
END:VCALENDAR