No Arabic abstract
In this work, we explore the diversity of ionised gas kinematics (rotational velocity $v_{phi}$ and velocity dispersion $sigma_{mathrm{g}}$) and gas-phase metallicity gradients at $0.1 leq z leq 2.5$ using a compiled data set of 74 galaxies resolved with ground-based integral field spectroscopy. We find that galaxies with the highest and the lowest $sigma_{mathrm{g}}$ have preferentially flat metallicity gradients, whereas those with intermediate values of $sigma_{mathrm{g}}$ show a large scatter in the metallicity gradients. Additionally, steep negative gradients appear almost only in rotation-dominated galaxies ($v_{phi}/sigma_{mathrm{g}} > 1$), whereas most dispersion-dominated galaxies show flat gradients. We use our recently developed analytic model of metallicity gradients to provide a physical explanation for the shape and scatter of these observed trends. In the case of high $sigma_{mathrm{g}}$, the inward radial advection of gas dominates over metal production and causes efficient metal mixing, thus giving rise to flat gradients. For low $sigma_{mathrm{g}}$, it is the cosmic accretion of metal-poor gas diluting the metallicity that gives rise to flat gradients. Finally, the reason for intermediate $sigma_{mathrm{g}}$ showing the steepest negative gradients is that both inward radial advection and cosmic accretion are weak as compared to metal production, which leads to the creation of steeper gradients. The larger scatter at intermediate $sigma_{mathrm{g}}$ may be due in part to preferential ejection of metals in galactic winds, which can decrease the strength of the production term. Our analysis shows how gas kinematics play a critical role in setting metallicity gradients in high-redshift galaxies.
Recent spatially resolved observations of galaxies at z=0.6-3 reveal that high-redshift galaxies show complex kinematics and a broad distribution of gas-phase metallicity gradients. To understand these results, we use a suite of high-resolution cosmological zoom-in simulations from the Feedback in Realistic Environments (FIRE) project, which include physically motivated models of the multi-phase ISM, star formation, and stellar feedback. Our simulations reproduce the observed diversity of kinematic properties and metallicity gradients, broadly consistent with observations at z=0-3. Strong negative metallicity gradients only appear in galaxies with a rotating disk, but not all rotationally supported galaxies have significant gradients. Strongly perturbed galaxies with little rotation always have flat gradients. The kinematic properties and metallicity gradient of a high-redshift galaxy can vary significantly on short time-scales, associated with starburst episodes. Feedback from a starburst can destroy the gas disk, drive strong outflows, and flatten a pre-existing negative metallicity gradient. The time variability of a single galaxy is statistically similar to the entire simulated sample, indicating that the observed metallicity gradients in high-redshift galaxies reflect the instantaneous state of the galaxy rather than the accretion and growth history on cosmological time-scales. We find weak dependence of metallicity gradient on stellar mass and specific star formation rate (sSFR). Low-mass galaxies and galaxies with high sSFR tend to have flat gradients, likely due to the fact that feedback is more efficient in these galaxies. We argue that it is important to resolve feedback on small scales in order to produce the diverse metallicity gradients observed.
We present a new model for the evolution of gas phase metallicity gradients in galaxies from first principles. We show that metallicity gradients depend on four ratios that collectively describe the metal equilibration timescale, production, transport, consumption, and loss. Our model finds that most galaxy metallicity gradients are in equilibrium at all redshifts. When normalized by metal diffusion, metallicity gradients are governed by the competition between radial advection, metal production, and accretion of metal-poor gas from the cosmic web. The model naturally explains the varying gradients measured in local spirals, local dwarfs, and high-redshift star-forming galaxies. We use the model to study the cosmic evolution of gradients across redshift, showing that the gradient in Milky Way-like galaxies has steepened over time, in good agreement with both observations and simulations. We also predict the evolution of metallicity gradients with redshift in galaxy samples constructed using both matched stellar masses and matched abundances. Our model shows that massive galaxies transition from the advection-dominated to the accretion-dominated regime from high to low redshifts, which mirrors the transition from gravity-driven to star formation feedback-driven turbulence. Lastly, we show that gradients in local ultraluminous infrared galaxies (major mergers) and inverted gradients seen both in the local and high-redshift galaxies may not be in equilibrium. In subsequent papers in this series, we show that the model also explains the observed relationship between galaxy mass and metallicity gradients, and between metallicity gradients and galaxy kinematics.
We have developed a novel Markov Chain Mote Carlo (MCMC) chemical painting technique to explore possible radial and vertical metallicity gradients for the thick disc progenitor. In our analysis we match an N-body simulation to the data from the Apache Point Observatory Galactic Evolution Experiment (APOGEE) survey. We assume that the thick disc has a constant scale-height and has completed its formation at an early epoch, after which time radial mixing of its stars has taken place. Under these assumptions, we find that the initial radial metallicity gradient of the thick disc progenitor should not be negative, but either flat or even positive, to explain the current negative vertical metallicity gradient of the thick disc. Our study suggests that the thick disc was built-up in an inside-out and upside-down fashion, and older, smaller and thicker populations are more metal poor. In this case, star forming discs at different epochs of the thick disc formation are allowed to have different radial metallicity gradients, including a negative one, which helps to explain a variety of slopes observed in high redshift disc galaxies. This scenario helps to explain the positive slope of the metallicity-rotation velocity relation observed for the Galactic thick disc. On the other hand, radial mixing flattens the slope of an existing gradient.
We investigate how HII region temperature structure assumptions affect direct-method spatially-resolved metallicity observations using multispecies auroral lines in a galaxy from the SAMI Galaxy Survey. SAMI609396B, at redshift $z=0.018$, is a low-mass galaxy in a minor merger with intense star formation, analogous to conditions at high redshifts. We use three methods to derive direct metallicities and compare with strong-line diagnostics. The spatial metallicity trends show significant differences among the three direct methods. Our first method is based on the commonly used electron temperature $T_e$([OIII]) from the [OIII]$lambda$4363 auroral line and a traditional $T_e$([OII]) -- $T_e$([OIII]) calibration. The second method applies a recent empirical correction to the O$^+$ abundance from the [OIII]/[OII] strong-line ratio. The third method infers the $T_e$([OII]) from the [SII]$lambdalambda$4069,76 auroral lines. The first method favours a positive metallicity gradient along SAMI609396B, whereas the second and third methods yield flattened gradients. Strong-line diagnostics produce mostly flat gradients, albeit with unquantified contamination from shocked regions. We conclude that overlooked assumptions about the internal temperature structure of HII regions in the direct method can lead to large discrepancies in metallicity gradient studies. Our detailed analysis of SAMI609396B underlines that high-accuracy metallicity gradient measurements require a wide array of emission lines and improved spatial resolutions in order to properly constrain excitation sources, physical conditions, and temperature structures of the emitting gas. Integral-field spectroscopic studies with future facilities such as JWST/NIRSpec and ground-based ELTs will be crucial in minimising systematic effects on measured gradients in distant galaxies.
We report on the gas-phase metallicity gradients of a sample of 264 star-forming galaxies at 0.6 < z < 2.6, measured through deep near-infrared Hubble Space Telescope slitless spectroscopy. The observations include 12-orbit depth Hubble/WFC3 G102 grism spectra taken as a part of the CANDELS Lya Emission at Reionization (CLEAR) survey, and archival WFC3 G102+G141 grism spectra overlapping the CLEAR footprint. The majority of galaxies (84%) in this sample are consistent with a zero or slightly positive metallicity gradient across the full mass range probed (8.5 < log M_*/M_sun < 10.5). We measure the intrinsic population scatter of the metallicity gradients, and show that it increases with decreasing stellar mass---consistent with previous reports in the literature, but confirmed here with a much larger sample. To understand the physical mechanisms governing this scatter, we search for correlations between the observed gradient and various stellar population properties at fixed mass. However, we find no evidence for a correlation with the galaxy properties we consider---including star-formation rates, sizes, star-formation rate surface densities, and star-formation rates per gravitational potential energy. We use the observed weakness of these correlations to provide material constraints for predicted intrinsic correlations from theoretical models.