Abstract
Cosmological N-body simulations of galaxies operate at the level of "star
particles" with a mass resolution on the scale of thousands of solar masses.
Turning these simulations into stellar mock catalogs requires üpsampling" the
star particles into individual stars following the same phase-space density. In
this paper, we demonstrate that normalizing flows provide a viable upsampling
method that greatly improves on conventionally-used kernel smoothing algorithms
such as EnBiD. We demonstrate our flow-based upsampling technique, dubbed
GalaxyFlow, on a neighborhood of the Solar location in two simulated galaxies:
Auriga 6 and h277. By eye, GalaxyFlow produces stellar distributions that are
smoother than EnBiD-based methods and more closely match the Gaia DR3 catalog.
For a quantitative comparison of generative model performance, we introduce a
novel multi-model classifier test. Using this classifier test, we show that
GalaxyFlow more accurately estimates the density of the underlying star
particles than previous methods.
Users
Please
log in to take part in the discussion (add own reviews or comments).