No Arabic abstract
We present a novel Bayesian methodology to jointly model photometry and deep Hubble Space Telescope (HST) 2d grism spectroscopy of high-redshift galaxies. Our requiem2d code measures both unresolved and resolved stellar populations, ages, and star-formation histories (SFHs) for the ongoing REQIUEM (REsolving QUIEscent Magnified) Galaxies Survey, which targets strong gravitationally lensed quiescent galaxies at z~2. We test the accuracy of texttt{requiem2d} using a simulated sample of massive galaxies at z~2 from the Illustris cosmological simulation and find we recover the general trends in SFH and median stellar ages. We further present a pilot study for the REQUIEM Galaxies Survey: MRG-S0851, a quintuply-imaged, massive ($log M_* / M_odot = 11.02 pm 0.04$) red galaxy at $z=1.883pm 0.001$. With an estimated gravitational magnification of $mu = 5.7^{+0.4}_{-0.2}$, we sample the stellar populations on 0.6 kpc physical size bins. The global mass-weighted median age is constrained to be $1.8_{-0.2}^{+0.3}$ Gyr, and our spatially resolved analysis reveals that MRG-S0851 has a flat age gradient in the inner 3 kpc core after taking into account the subtle effects of dust and metallicity on age measurements, favoring an early formation scenario. The analysis for the full REQUIEM-2D sample will be presented in a forthcoming paper with a beta-release of the requiem2d code.
We take advantage of gravitational lensing amplification by Abell 1689 (z=0.187) to undertake the first space-based census of emission line galaxies (ELGs) in the field of a massive lensing cluster. Forty-three ELGs are identified to a flux of i_775=27.3 via slitless grism spectroscopy. One ELG (at z=0.7895) is very bright owing to lensing magnification by a factor of ~4.5. Several Balmer emission lines detected from ground-based follow-up spectroscopy signal the onset of a major starburst for this low-mass galaxy (M_* = 2 x 10^9 solar masses) with a high specific star formation rate (~20 /Gyr). From the blue emission lines we measure a gas-phase oxygen abundance consistent with solar (12+log(O/H)=8.8 +/- 0.2). We break the continuous line-emitting region of this giant arc into seven ~1kpc bins (intrinsic size) and measure a variety of metallicity dependent line ratios. A weak trend of increasing metal fraction is seen toward the dynamical center of the galaxy. Interestingly, the metal line ratios in a region offset from the center by ~1kpc have a placement on the blue HII region excitation diagram with f([OIII])/f(Hbeta) and f([NeIII])/f(Hbeta) that can be fit by an AGN. This asymmetrical AGN-like behavior is interpreted as a product of shocks in the direction of the galaxys extended tail, possibly instigated by a recent galaxy interaction.
A galaxys stellar mass-to-light ratio ($M_star/L$) is a useful tool for converting luminosity to stellar mass ($M_star$). However, the practical utility of $M_star/L$ inferred from stellar population synthesis (SPS) models is limited by mismatches between the real and assumed models for star formation history (SFH) and dust geometry, both of which vary within galaxies. Here, we measure spatial variations in $M_star/L$ and their dependence on color, star formation history, and dust across the disk of M31, using a map of $M^mathrm{CMD}_star$ derived from color-magnitude diagrams of resolved stars in the Panchromatic Hubble Andromeda Treasury (PHAT) survey. First, we find comparable scatter in $M_star/L$ for the optical and mid-IR, contrary to the common idea that $M_star/L$ is less variable in the IR. Second, we confirm that $M_star/L$ is correlated with color for both the optical and mid-IR and report color vs. $M_star/L$ relations (CMLRs) in M31 for filters used in the Sloan Digital Sky Survey (SDSS) and Widefield Infrared Survey Explorer (WISE). Third, we show that the CMLR residuals correlate with recent SFH, such that quiescent regions are offset to higher $M_star/L$ than star-forming regions at a fixed color. The mid-IR CMLR, however, is not linear due to the high scatter of $M_star/L$ in star-forming regions. Finally, we find a flatter optical CMLR than any SPS-based CMLRs in the literature. We show this is an effect of dust geometry, which is typically neglected but should be accounted for when using optical data to map $M_star/L$.
We present and publicly release (https://www.gclasshst.com) the first spatially resolved H$alpha$ maps of star-forming cluster galaxies at $zsim1$, made possible with the Wide Field Camera 3 (WFC3) G141 grism on the Hubble Space Telescope (HST). Using a similar but updated method to 3D-HST in the field environment, we stack the H$alpha$ maps in bins of stellar mass, measure the half-light radius of the H$alpha$ distribution and compare it to the stellar continuum. The ratio of the H$alpha$ to stellar continuum half-light radius, $R[mathrm{H}alpha/mathrm{C}]=frac{R_{mathrm{eff, H}alpha}}{R_{mathrm{eff, Cont}}}$, is smaller in the clusters by $(6pm9)%$, but statistically consistent within $1sigma$ uncertainties. A negligible difference in $R[mathrm{H}alpha/mathrm{C}]$ with environment is surprising, given the higher quenched fractions in the clusters relative to the field. We postulate that the combination of high quenched fractions and no change in $R[mathrm{H}alpha/mathrm{C}]$ with environment can be reconciled if environmental quenching proceeds rapidly. We investigate this hypothesis by performing similar analysis on the spectroscopically-confirmed recently quenched cluster galaxies. 87% have H$alpha$ detections, with star formation rates $8pm1$ times lower than star-forming cluster galaxies of similar stellar mass. Importantly, these galaxies have a $R[mathrm{H}alpha/mathrm{C}]$ that is $(81pm8)%$ smaller than coeval star-forming field galaxies at fixed stellar mass. This suggests the environmental quenching process occurred outside-in. We conclude that disk truncation due to ram-pressure stripping is occurring in cluster galaxies at $zsim1$, but more rapidly and/or efficiently than in $zlesssim0.5$ clusters, such that the effects on $R[mathrm{H}alpha/mathrm{C}]$ become observable just after the cluster galaxy has recently quenched.
We present an analysis of STIS/HST optical spectra of a sample of ten Seyfert galaxies aimed at studying the structure and physical properties of the coronal-line region (CLR). The high-spatial resolution provided by STIS allowed us to resolve the CLR and obtain key information about the kinematics of the coronal-line gas, measure directly its spatial scale, and study the mechanisms that drive the high-ionisation lines. We find CLRs extending from just a few parsecs (~10 pc) up to 230 pc in radius, consistent with the bulk of the coronal lines (CLs) originating between the BLR and NLR, and extending into the NLR in the case of [FeVII] and [NeV] lines. The CL profiles strongly vary with the distance to the nucleus. We observed line splitting in the core of some of the galaxies. Line peak shifts, both red- and blue-shifts, typically reached 500 km/s, and even higher velocities (1000 km/s) in some of the galaxies. In general, CLs follow the same pattern of rotation curves as low-ionisation lines like [OIII]. From a direct comparison between the radio and the CL emission we find that neither the strength nor the kinematics of the CLs scale in any obvious and strong way with the radio jets. Moreover, the similarity of the flux distributions and kinematics of the CLs and low-ionisation lines, the low temperatures derived for the gas, and the success of photoionisation models to reproduce, within a factor of few, the observed line ratios, point towards photoionisation as the main driving mechanism of CLs.
To investigate star formation and assembly processes of massive galaxies, we present here a spatially-resolved stellar populations analysis of a sample of 45 elliptical galaxies (Es) selected from the CALIFA survey. We find rather flat age and [Mg/Fe] radial gradients, weakly dependent on the effective velocity dispersion of the galaxy within half-light radius. However, our analysis shows that metallicity gradients become steeper with increasing galaxy velocity dispersion. In addition, we have homogeneously compared the stellar populations gradients of our sample of Es to a sample of nearby relic galaxies, i.e., local remnants of the high-z population of red nuggets. This comparison indicates that, first, the cores of present-day massive galaxies were likely formed in gas-rich, rapid star formation events at high redshift (z>2). This led to radial metallicity variations steeper than observed in the local Universe, and positive [Mg/Fe] gradients. Second, our analysis also suggests that a later sequence of minor dry mergers, populating the outskirts of early-type galaxies (ETGs), flattened the pristine [Mg/Fe] and metallicity gradients. Finally, we find a tight age-[Mg/Fe] relation, supporting that the duration of the star formation is the main driver of the [Mg/Fe] enhancement in massive ETGs. However, the star formation time-scale alone is not able to fully explain our [Mg/Fe] measurements. Interestingly, our results match the expected effect that a variable stellar initial mass function would have on the [Mg/Fe] ratio.