Skip to main content Skip to secondary navigation
Main content start

KIPAC Tea Talk: Characterizing Turbulence in Circumgalactic Gas / Calibrating Semi-Analytic Models to Beyond Cold Dark Matter

Mandy Chen (Caltech & Carnegie Observatories), Dimple Sarnaaik (Carnegie Science Observatories & USC)
Campus, PAB 102/103

Event Details:

Tuesday, October 27, 2026
10:40am - 11:30am PDT

Location

Campus, PAB 102/103

This event is open to:

Faculty/Staff
Members
Students

Chen Abstract: The circumgalactic medium is the turbulent interface between galaxies and the intergalactic medium. I will present empirical measurements of its turbulent motions, the insights they offer into AGN feedback and the baryon cycle, and our use of simulations and forward modeling to interpret these observations robustly.

Sarnaaik Abstract: Dark matter makes up ~80% of all matter in the Universe, yet its fundamental nature remains unknown. The hunt spans a vast landscape of indirect approaches, including stellar streams, strong gravitational lensing, dwarf satellite counts, and subhalo mass function measurements. These ultimately depend on the accurate prediction of subhalo populations. While Cold Dark Matter (CDM) succeeds on large scales, small-scale discrepancies motivate alternatives including Warm Dark Matter (WDM), Fuzzy Dark Matter (FDM), and Interacting Dark Matter (IDM). Distinguishing between them requires fast, reliable predictions at a statistical scale, and semi-analytic models (SAMs) like Galacticus bridge that gap, generating the merger histories of millions of halos at a fraction of the computational cost of N-body simulations.Galacticus is built on the Extended Press-Schechter (EPS) formalism, an analytic methodology for building the merger histories that determine how halos and subhalo populations form and evolve. EPS has been extensively tested for CDM, but its merger tree branching rates remain uncalibrated for alternative DM models until now.We present the first calibration of Galacticus to beyond-CDM physics using three leading simulation suites: MultiDark Planck (MDPL2), Symphony, and COZMIC I, targeting WDM particle masses of 3–10 keV. Using MCMC methods, we calibrate against the progenitor mass function — the mass distribution of early-Universe halos that will eventually merge into a halo of interest, and our key point of comparison between Galacticus and N-body simulations. We optimize the merger tree branching rate modifier across both top-hat and sharp-k window functions. Both yield consistent best-fit parameters within 1σ, confirming that the underlying density field governs halo statistics rather than the smoothing prescription — and our calibrated model reproduces N-body progenitor mass functions to ~0.01% of the host halo mass. Crucially, preliminary results show that WDM+CDM best-fit branching rates describe FDM and IDM with excellent agreement, pointing toward the possibility of a universal calibration spanning the vast landscape of DM models.

Related Topics

Explore More Events