| Author(s) and Co-Author(s) with Affiliation: Pratyush Kumar Das(University of Queensland), Sarah Sweet(University of Queensland), Simon Deeley(University of Queensland), Holger Baumgardt(University of Queensland), Scott Croom(University of Sydney), Luca Cortese(University of Western Australia), Joss Bland Hawthorn(University of Sydney) |
| Abstract: The classical Hubble sequence provides a qualitative, visual classification of galaxies. A complementary and physically motivated framework instead uses stellar specific angular momentum (j*) to characterize galaxy structure and assembly. Decades of work have established the tight relation (known as the FALL relation) linking j* and stellar mass (M*), with rotationally supported disks and dispersion-dominated spheroids occupying distinct loci.
I present Spinny, a modular pipeline that operationalizes this relation into a data-driven map of galaxy morphology and evolution. Spinny (i) recovers galaxy geometry from imaging, (ii) constructs smooth stellar mass surface-density models (including bulge+disk decompositions where required), and (iii) derives rotation profiles from spatially resolved spectroscopy to compute integrated j* with controlled uncertainties. The use of physically motivated density and velocity models stabilizes the measurements at large radii, enabling reliable extrapolation of j* beyond the observational footprint.
Applied to the SAMI Galaxy Survey, which spans environments from the field to rich clusters, Spinny yields a component-resolved Fall relation: stellar disks follow a high-j* sequence, while bulges populate a lower-j*, shallower branch. This framework provides a unified quantitative link between ordered rotation, star-formation state, and internal structure.
To further interpret these trends, I employ the symbolic regression Machine Learning model to identify compact, interpretable relations connecting j* with global kinematic, structural, and environmental parameters. This approach reveals low-dimensional functional forms that capture key drivers of angular momentum retention and morphology, offering physically transparent alternatives to purely empirical scaling relations.
Finally, I demonstrate how Spinny is readily transferable to other integral-field spectroscopic surveys such as Hector, MaNGA, and MAGPI, and describe extensions incorporating simulation-informed velocity priors to constrain j* in low-signal outskirts and, ultimately, in high-redshift systems. |
