No Arabic abstract
We study the star formation rates (SFRs) of galaxies as a function of local galaxy density at 0.6<z<0.9. We used a low-dispersion prism in IMACS on the 6.5-m Baade (Magellan I) telescope to obtain spectra and measured redshifts to a precision of sigma_z/(1+z)=1% for galaxies with z<23.3 AB mag. We utilized a stellar mass-limited sample of 977 galaxies above M>1.8x10^{10} Msun to conduct our main analysis. With three different SFR indicators, (1) Spitzer MIPS 24-micron imaging, (2) SED fitting, and (3) [OII]3727 emission, we find the median specific SFR (SSFR) and SFR to decline from the low-density field to the cores of groups and a rich cluster. For the SED and [OII] based SFRs, the decline in SSFR is roughly an order of magnitude while for the MIPS based SFRs, the decline is a factor of ~4. We find approximately the same magnitude of decline in SSFR even after removing the sample of galaxies near the cluster. Galaxies in groups and a cluster at these redshifts therefore have lower star formation (SF) activity than galaxies in the field, as is the case at z~0. We investigated whether the decline in SFR with increasing density is caused by a change in the proportion of quiescent and star forming galaxies (SFGs) or by a decline in the SFRs of SFGs. Using the rest-frame U-V and V-J colors to distinguish quiescent galaxies from SFGs we find the fraction of quiescent galaxies increases from ~32% to 79% from low to high density. In addition, we find the SSFRs of SFGs, selected based on U-V and V-J colors, to decline with increasing density by factors of ~5-6 for the SED and [OII] based SFRs. The MIPS based SSFRs for SFGs decline with a shallower slope. The order of magnitude decline in the SSFR-density relation at 0.6<z<0.9 is therefore driven by both a combination of declining SFRs of SFGs as well as a changing mix of SFGs and quiescent galaxies [ABRIDGED].
We study the relations between gas-phase metallicity ($Z$), local stellar mass surface density ($Sigma_*$), and the local star formation surface density ($Sigma_{rm SFR}$) in a sample of 1120 star-forming galaxies from the MaNGA survey. At fixed $Sigma_{*}$ the local metallicity increases as decreasing of $Sigma_{rm SFR}$ or vice versa for metallicity calibrators of N2 and O3N2. Alternatively, at fixed $Sigma_{rm SFR}$ metallicity increases as increasing of $Sigma_{*}$, but at high mass region, the trend is flatter. However, the dependence of metallicity on $Sigma_{rm SFR}$ is nearly disappeared for N2O2 and N2S2 calibrators. We investigate the local metallicity against $Sigma_{rm SFR}$ with different metallicity calibrators, and find negative/positive correlations depending on the choice of the calibrator. We demonstrate that the O32 ratio (or ionization parameter) is probably dependent on star formation rate at fixed local stellar mass surface density. Additional, the shape of $Sigma_*$ -- $Z$ -- $Sigma_{rm SFR}$ (FMR) depends on metallicity calibrator and stellar mass range. Since the large discrepancy between the empirical fitting-based (N2, O3N2) to electronic temperature metallicity and the photoionization model-dependent (N2O2, N2S2) metallicity calibrations, we conclude that the selection of metallicity calibration affects the existence of FMR on $Sigma_{rm SFR}$.
The connection between galaxy star formation rate (SFR) and dark matter (DM) is of paramount importance for the extraction of cosmological information from next generation spectroscopic surveys that will target emission line star forming galaxies. Using publicly available mock galaxy catalogs obtained from various semi-analytic models (SAMs) we explore the SFR-DM connection in relation to the speed-from-light method (Feix et al. 2016) for inferring the growth rate, $f$, from luminosity/SFR shifts. Emphasis is given to the dependence of the SFR distribution on the environment density on scales of 10s-100s Mpc. We show that the application of the speed-from-light method to an Euclid-like survey is not biased by environmental effects. In all models, the precision on the measured $beta=f/b$ parameter is $sigma_beta < 0.17$ at $z=1$. This translates into errors of $sigma_f sim 0.22$ and $sigma_{(fsigma_8)}sim 0.1$, without invoking assumptions on the mass power spectrum. These errors are in the same ballpark as recent analyses of the redshift space distortions in galaxy clustering. In agreement with previous studies, the bias factor, $b$ is roughly a scale-independent, constant function of the SFR for star forming galaxies. Its value at $z=1$ ranges from $1.2$ to $1.5$ depending on the SAM recipe. Although in all SAMs denser environments host galaxies with higher stellar masses, the dependence of the SFR on the environment is more involved. In most models the SFR probability distribution is skewed to larger values in denser regions. One model exhibits an inverted trend where high SFR is suppressed in dense environment.
We examine whether the super star-forming clumps (R~1-3 kpc; M~10^8-10^9.5 Msun) now known to be a key component of star-forming galaxies at z~2 could be the formation sites of the locally observed old globular cluster population. We find that the stellar populations of these super star-forming clumps are excellent matches to those of local metal-rich globular clusters. Moreover, this globular cluster population is known to be associated with the bulges / thick disks of galaxies, and we show that its spatial distribution and kinematics are consistent with the current understanding of the assembly of bulges and thick disks from super star-forming clumps at high redshift. Finally, with the assumption that star formation in these clumps proceeds as a scaled-up version of local star formation in molecular clouds, this formation scenario reproduces the observed numbers and mass spectra of metal-rich globular clusters. The resulting link between the turbulent and clumpy disks observed in high-redshift galaxies and a local globular cluster population provides a plausible co-evolutionary scenario for several of the major components of a galaxy: the bulge, the thick disk, and one of the globular cluster populations.
We present near-infrared spectroscopic observations of star-forming galaxies at z~1.4 with FMOS on the Subaru Telescope. We observed K-band selected galaxies in the SXDS/UDS fields with K<23.9 mag, 1.2<z_ph<1.6, M*>10^{9.5} Msun, and expected F(Halpha)>10^{-16} erg s^{-1} cm^{-2}. 71 objects in the sample have significant detections of Halpha. For these objects, excluding possible AGNs identified from the BPT diagram, gas-phase metallicities are obtained from [NII]/Halpha line ratio. The sample is split into three stellar mass bins, and the spectra are stacked in each stellar mass bin. The mass-metallicity relation obtained at z~1.4 is located between those at z~0.8 and z~2.2. We constrain an intrinsic scatter to be ~0.1 dex or larger in the mass-metallicity relation at z~1.4; the scatter may be larger at higher redshifts. We found trends that the deviation from the mass-metallicity relation depends on the SFR and the half light radius: Galaxies with higher SFR and larger half light radii show lower metallicities at a given stellar mass. One possible scenario for the trends is the infall of pristine gas accreted from IGM or through merger events. Our data points show larger scatter than the fundamental metallicity relation (FMR) at z~0.1 and the average metallicities slightly deviate from the FMR. The compilation of the mass-metallicity relations at z~3 to z~0.1 shows that they evolve smoothly from z~3 to z~0 without changing the shape so much except for the massive part at z~0.
We investigate the relationships between stellar mass, gas-phase oxygen abundance (metallicity), star formation rate, and dust content of star-forming galaxies at z$sim$1.6 using Subaru/FMOS spectroscopy in the COSMOS field. The mass-metallicity relation at $zsim1.6$ is steeper than the relation observed in the local Universe. The steeper MZ relation at $zsim1.6$ is mainly due to evolution in the stellar mass where the MZ relation begins to turnover and flatten. This turnover mass is 1.2 dex larger at $zsim1.6$. The most massive galaxies at $zsim1.6$ ($sim 10^{11}M_odot$) are enriched to the level observed in massive galaxies in the local Universe. The mass-metallicity relation we measure at $zsim1.6$ supports the suggestion of an empirical upper metallicity limit that does not significantly evolve with redshift. We find an anti-correlation between metallicity and star formation rate for galaxies at a fixed stellar mass at $zsim1.6$ which is similar to trends observed in the local Universe. We do not find a relation between stellar mass, metallicity and star formation rate that is independent of redshift; our data suggest that there is redshift evolution in this relation. We examine the relation between stellar mass, metallicity and dust extinction. We find that at a fixed stellar mass dustier galaxies tend to be more metal rich. From examination of the stellar masses, metallicities, SFRs and dust extinctions we conclude that stellar mass is most closely related to dust extinction.