Kinematic Downsizing at z~2

TL;DR

This study investigates the internal kinematics of star-forming galaxies at z~2, revealing that massive galaxies have formed primitive, rotation-supported disks, while low-mass galaxies are still assembling their disks, indicating a mass-dependent disk formation process.

Contribution

It provides new observational evidence of kinematic downsizing at z~2, showing how disk assembly varies with galaxy mass during peak star formation epoch.

Findings

Massive galaxies have rotation-dominated kinematics and follow the Tully-Fisher relation.

Low-mass galaxies show dispersion-supported kinematics and fall below the Tully-Fisher relation.

Kinematic downsizing suggests stronger or more active disk disruption in low-mass systems at z~2.

Abstract

We present results from a survey of the internal kinematics of 49 star-forming galaxies at z$\,\sim\,$2 in the CANDELS fields with the Keck/MOSFIRE spectrograph (SIGMA, Survey in the near-Infrared of Galaxies with Multiple position Angles). Kinematics (rotation velocity $V_{rot}$ and integrated gas velocity dispersion $\sigma_g$) are measured from nebular emission lines which trace the hot ionized gas surrounding star-forming regions. We find that by z$\,\sim\,$2, massive star-forming galaxies ($\log\,M_*/M_{\odot}\gtrsim10.2$) have assembled primitive disks: their kinematics are dominated by rotation, they are consistent with a marginally stable disk model, and they form a Tully-Fisher relation. These massive galaxies have values of $V_{rot}/\sigma_g$ which are factors of 2-5 lower than local well-ordered galaxies at similar masses. Such results are consistent with findings by other studies. We find that low mass galaxies ($\log\,M_*/M_{\odot}\lesssim10.2$) at this epoch are still in the early stages of disk assembly: their kinematics are often supported by gas velocity dispersion and they fall from the Tully-Fisher relation to significantly low values of $V_{rot}$. This "kinematic downsizing" implies that the process(es) responsible for disrupting disks at z$\,\sim\,$2 have a stronger effect and/or are more active in low mass systems. In conclusion, we find that the period of rapid stellar mass growth at z$\,\sim\,$2 is coincident with the nascent assembly of low mass disks and the assembly and settling of high mass disks.