The SAMI galaxy survey: predicting kinematic morphology with logistic regression

TL;DR

This study uses logistic regression on SAMI galaxy survey data to analyze the factors influencing galaxy kinematic morphology, revealing that environment alone does not predict slow rotator fractions beyond other galaxy properties.

Contribution

It introduces a logistic regression model that predicts galaxy kinematic type based on multiple parameters, clarifying the role of environment in morphology.

Findings

Local environment does not add predictive power beyond other galaxy properties.

Galaxy size, mass, SFR, and ellipticity are key predictors of kinematic morphology.

The model can be applied to other surveys to estimate slow rotator fractions.

Abstract

We use the SAMI galaxy survey to study the the kinematic morphology-density relation: the observation that the fraction of slow rotator galaxies increases towards dense environments. We build a logistic regression model to quantitatively study the dependence of kinematic morphology (whether a galaxy is a fast rotator or slow rotator) on a wide range of parameters, without resorting to binning the data. Our model uses a combination of stellar mass, star-formation rate (SFR), $r$-band half-light radius and a binary variable based on whether the galaxy's observed ellipticity ($\epsilon$) is less than 0.4. We show that, at fixed mass, size, SFR and $\epsilon$, a galaxy's local environmental surface density ($\log_{10}(\Sigma_5/\mathrm{Mpc}^{-2})$) gives no further information about whether a galaxy is a slow rotator, i.e. the observed kinematic-morphology density relation can be entirely explained by the well-known correlations between environment and other quantities. We show how our model can be applied to different galaxy surveys to predict the fraction of slow rotators which would be observed and discuss its implications for the formation pathways of slow rotators.