Early Stages Of Galaxy Formation Research Articles

Four-hundred Very Metal-poor Stars Studied with LAMOST and Subaru. III. Dynamically Tagged Groups and Chemodynamical Properties

Very metal-poor (VMP) stars record the signatures of early accreted galaxies, making them essential tools for unraveling the early stages of Galaxy formation. Understanding the origin of VMP stars requires comprehensive studies of their chemical compositions and kinematics, which are currently lacking. Hence, we conduct a chemodynamical analysis of 352 VMP stars selected from one of the largest uniform high-resolution VMP star samples, jointly obtained from LAMOST and Subaru. We apply a friends-of-friends clustering algorithm to the master catalog of this high-resolution sample, which consists of 5778 VMP stars. It results in 131 dynamically tagged groups with 89 associated with known substructures in the Milky Way, including Gaia-Sausage-Enceladus (GSE), Thamnos, Helmi streams, Sequoia, Wukong, Pontus, and the very metal-poor disk (VMPD). Our findings are: (i) the VMPD shows lower Zn abundances than the rest, which indicates that it could be a relic of small stellar systems; (ii) Sequoia shows moderately high r-process abundances; (iii) Helmi streams show deficiencies in carbon and light neutron-capture elements; (iv) the fraction of carbon-enhanced metal-poor stars with no enhancement in heavy elements (CEMP-no stars) seems low in the VMPD and the Helmi streams; and (v) a subgroup in GSE exhibits a very high fraction of r-process enhanced stars, with four out of five showing [Eu/Fe]> +1.0. The abundance patterns of other elements in VMP substructures largely match the whole VMP sample. We also study large-scale correlations between abundance ratios and kinematics without classifying stars into substructures, but it does not yield significant correlations once the overall chemical evolution is considered for most elements.

Understanding the Early Stages of Galaxy Formation Using Very Metal-poor Stars from the Hamburg/ESO Survey

We explore the chemodynamical properties of a sample of very metal-poor (VMP) stars selected from the Hamburg/ESO survey, matched with Gaia EDR3, in the phase space identified by the three integrals of motion (L z , E, I 3). Disk and halo orbits are separated by using the criteria defined in Carollo & Chiba. We found 26 stars with [Fe/H] ≤ −2.5 possessing disk kinematics, of which 13 are extremely metal-poor. At these metallicities, the number of stars with disk kinematics is three times its retrograde counterpart. In the same range of metallicity we also identified 37 halo stars most tightly bound to the gravitational potential of the progenitor halo. The origins of these stars are investigated by comparing the observational results with simulated galaxies from the Aquarius Project and the IllustrisTNG simulations. We found two mechanisms of formation of VMP stars with disk kinematics: accretion from early satellites (which is dominant), and in situ formation. These stars are very old, with ages >12.5 Gyr (z > 5), and they are α-enriched. Accretion and in situ formation are also found for the retrograde counterparts, with accretion also being the dominant mode. Contributing accreted satellites have stellar masses in the range 106–109 M ☉ and are very gas-rich. The most bound halo stars are the oldest detected, with a median age of ∼13.3 Gyr (z ∼ 11), and are α-enriched. Our finding clearly shows that very old, VMP stars store important information on the first stages of assembly of our Galaxy and its halo.

Catastrophic dark matter particle capture

In this paper we describe a new idea which may be relevant to the formation of galaxies via the infall of baryonic matter (BM) and dark matter (DM) onto a preexisting overdensity. While BM can under certain circumstances be captured by thermal processes, DM particles fly through a static overdensity without being captured. We propose a simple model for DM capture: if during the passage through it, the mass of the overdensity increases, then slow DM particles are captured by it, further increasing its mass, while faster particles slow down, transferring part of their energy to the galaxy. We estimate the minimum initial velocity of a particle required for a passage without capture through the center of the galaxy and derive a nonlinear equation describing the rate of galaxy mass increase. An analysis carried out using the ideas of catastrophes theory shows that if the increase in the mass of baryonic matter exceeds a certain threshold value, this can lead to a very intensive capture of dark matter. We speculate that this process may be associated on the one hand with the accretion of matter during the early stages of galaxy formation and, on the other hand, also later with the merger of galaxies. For the studied process to take place, the density of intergalactic DM must exceed some threshold value. Then the rate of increase in the mass of DM can be much higher than the one of baryonic matter. The capture sharply decreases after the DM density drops below the threshold value, e.g., due to the expansion of the Universe.

Structure and kinematics of a massive galaxy at z ∼ 7

Context. Observations of the rest-frame UV emission of high-redshift galaxies suggest that the early stages of galaxy formation involve disturbed structures. Imaging the cold interstellar medium (ISM) can provide a unique view of the kinematics associated with the assembly of galaxies. Aims. In this paper, we analyze the spatial distribution and kinematics of the cold ionized gas of the normal star-forming galaxy COS-2987030247 at z = 6.8076, based on new high-resolution observations of the [C II] 158 μm line emission obtained with the Atacama Large Millimeter/submillimeter Array (ALMA). Methods. The analysis of these observations allowed us to: compare the spatial distribution and extension of the [C II] and rest-frame UV emission, model the [C II] line data-cube using the 3DBAROLO code, and measure the [C II] luminosity and star formation rate (SFR) surface densities in the galaxy subregions. Results. The system is found to be composed of a main central source, a fainter north extension, and candidate [C II] companions located 10-kpc away. We find similar rest-frame UV and [C II] spatial distributions, suggesting that the [C II] emission emerges from the star-forming regions. The agreement between the UV and [C II] surface brightness radial profiles rules out diffuse, extended [C II] emission (often called a [C II] halo) in the main galaxy component. The [C II] velocity map reveals a velocity gradient in the north-south direction, suggesting ordered motion, as commonly found in rotating-disk galaxies. However, higher resolution observations would be needed to rule out a compact merger scenario. Our model indicates an almost face-on galaxy (i ∼ 20°), with a average rotational velocity of 86 ± 16 km s−1 and a low average velocity dispersion, σ < 30 km s−1. This result implies a dispersion lower than the expected value from observations and semi-analytic models of high redshift galaxies. Furthermore, our measurements indicate that COS-2987030247 and its individual regions systematically lie within the local L[CII]-SFR relationship, yet slightly below the local Σ[CII]-ΣUV relation. Conclusions. We argue that COS-2987030247 is a candidate rotating disk experiencing a short period of stability which will possibly become perturbed at later times by accreting sources.

Mechanical feedback from stellar winds with an application to galaxy formation at high redshift

ABSTRACT We compute different sets of stellar evolutionary tracks in order to quantify the energy, mass, and metals yielded by massive main-sequence and post-main-sequence winds. Our aim is to investigate the impact of binary systems and of a metallicity-dependent distribution of initial rotational velocities on the feedback by stellar winds. We find significant changes compared to the commonly used non-rotating, single-star scenario. The largest differences are noticeable at low metallicity, where the mechanical-energy budget is substantially increased. So as to establish the maximal (i.e. obtained by neglecting dissipation in the near circumstellar environment) influence of winds on the early stages of galaxy formation, we use our new feedback estimates to simulate the formation and evolution of a sub-L* galaxy at redshift 3 (hosted by a dark-matter halo with a mass of 1.8 × 1011 M⊙) and compare the outcome with simulations in which only supernova (SN) feedback is considered. Accounting for the continuous energy injection by winds reduces the total stellar mass, the metal content, and the burstiness of the star-formation rate as well as of the outflowing gas mass. However, our numerical experiment suggests that the enhanced mechanical feedback from the winds of rotating and binary stars has a limited impact on the most relevant galactic properties compared to the non-rotating single-star scenario. Eventually, we look at the relative abundance between the metals entrained in winds and those ejected by SNe and find that it stays nearly constant within the simulated galaxy and its surrounding halo.

Properties of the brightest globular cluster in M 81 based on multicolour observations

Context.Researching the properties of the brightest globular cluster (referred to as GC1) in M 81 can provide a fossil record of the earliest stages of galaxy formation and evolution. The Beijing–Arizona–Taiwan–Connecticut (BATC) Multicolour Sky Survey has carried out deep exposures of M 81.Aims.We derive the magnitudes in intermediate-band filters of the BATC system for GC1 and determine its age, mass, and structural parameters.Methods.GC1 was observed by BATC using 14 intermediate-band filters covering a wavelength range of 4000–10 000 Å. Based on photometric data in BATC and Two Micron All Sky Survey near-infraredJHKsfilters, we constructed an extensive spectral energy distribution of GC1, spanning the wavelength range from 4000 to 20 000 Å. By comparing multicolour photometry with theoretical single stellar population synthesis models, we derived the age and mass of GC1. In addition, we obtained ellipticities, position angles, and surface brightness profiles for GC1 based on the images of deep observations with the Advanced Camera for Surveys on theHubbleSpace Telescope. GC1 is better fitted by the Wilson model than by the King and Sérsic models in theF606Wfilter, and it is better fitted by the Sérsic model than by the King and Wilson models in theF814Wfilter. The ‘best-fit’ half-light radius of GC1 obtained here is 5.59 pc, which is larger than the majority of normal globular clusters (GCs) of the same luminosity.Results.The age and mass of GC1 estimated here are 13.0 ± 2.90 Gyr and 1.06 − 1.48 × 107 M⊙, respectively. TheRhversusMVdiagram shows that GC1 occupies the same area as extended star clusters. Therefore, we suggest that GC1 is more likely an accreted former nuclear star cluster than a classical GC similar to most of those in the Milky Way.

Formation of the First Stars and Black Holes

We review the current status of knowledge concerning the early phases of star formation during cosmic dawn. This includes the first generations of stars forming in the lowest mass dark matter halos in which cooling and condensation of gas with primordial composition is possible at very high redshift ($z > 20$), namely metal-free Population III stars, and the first generation of massive black holes forming at such early epochs, the so-called black hole seeds. The formation of black hole seeds as end states of the collapse of Population III stars, or via direct collapse scenarios, is discussed. In particular, special emphasis is given to the physics of supermassive stars as potential precursors of direct collapse black holes, in light of recent results of stellar evolution models, and of numerical simulations of the early stages of galaxy formation. Furthermore, we discuss the role of the cosmic radiation produced by the early generation of stars and black holes at high redshift in the process of reionization.

Surface brightness fluctuation spectra to constrain stellar population properties

ABSTRACT We present a new set of surface brightness fluctuation (SBF) spectra computed with the E-MILES stellar population synthesis models. The model SBF spectra cover the range λλ1680–50 000 at moderately high resolution, all based on extensive empirical stellar libraries. The models span the metallicity range $-2.3\le \mbox{$\mbox{[M/H]}$}\le +0.26$ for a suite of intial mass function types with varying slopes. These predictions can complement and aid fluctuation magnitude studies, permitting a first-order approximation by applying filter responses to the SBF spectra to obtain spectroscopic SBF magnitudes. We provide a recipe for obtaining the latter and discuss their uncertainties and limitations. We compare our spectroscopic SBF magnitudes to photometric data of a sample of early-type galaxies. We also show that the SBF spectra can be very useful for constraining relevant stellar population parameters. We find small (<5 per cent) mass fractions of extremely metal-poor components ($\mbox{$\mbox{[M/H]}$}\lt -1$) on the top of the dominant, old, and metal-rich stellar population. These results put stringent constraints on the early stages of galaxy formation in massive elliptical galaxies. This is remarkable given the high degree of degeneracy of the standard spectral analysis to such metal-poor stellar populations in the visible and in the near-infrared. The new SBF models show great potential for exploiting ongoing surveys, particularly those based on narrow-band filters.

AGNs at the cosmic dawn: predictions for future surveys from a ΛCDM cosmological model

ABSTRACT Telescopes to be launched over the next decade and a half, such as JWST, EUCLID, ATHENA, and Lynx, promise to revolutionize the study of the high-redshift Universe and greatly advance our understanding of the early stages of galaxy formation. We use a model that follows the evolution of the masses and spins of supermassive black holes (SMBHs) within a semi-analytic model of galaxy formation to make predictions for the active galactic nucleus luminosity function at $z$ ≥ 7 in the broadband filters of JWST and EUCLID at near-infrared wavelengths, and ATHENA and Lynx at X-ray energies. The predictions of our model are relatively insensitive to the choice of seed black hole mass, except at the lowest luminosities (Lbol < 1043 erg s−1) and the highest redshifts ($z$ > 10). We predict that surveys with these different telescopes will select somewhat different samples of SMBHs, with EUCLID unveiling the most massive, highest accretion rate SMBHs, Lynx the least massive, lowest accretion rate SMBHs, and JWST and ATHENA covering objects inbetween. At $z$ = 7, we predict that typical detectable SMBHs will have masses, MBH ∼ 105–8 M⊙, and Eddington normalized mass accretion rates, $\dot{M}/\dot{M}_{\mathrm{Edd}}\sim 0.6{-}2$. The SMBHs will be hosted by galaxies of stellar mass M⋆ ∼ 108–10 M⊙, and dark matter haloes of mass Mhalo ∼ 1011–12 M⊙. We predict that the detectable SMBHs at $z$ = 10 will have slightly smaller black holes, accreting at slightly higher Eddington normalized mass accretion rates, in slightly lower mass host galaxies compared to those at $z$ = 7, and reside in haloes of mass Mhalo ∼ 1010–11 M⊙.

X-Ray Spectrum of RBS 315: Absorption or Intrinsic Curvature

Abstract X-ray absorption of high-redshift quasars is enigmatic, because it remains unclear where in the universe the absorbing gas is. If absorption occurs near the high-z host, it could help us understand early stages of galaxy formation. If it is in the intergalactic medium (IGM), it provides a unique way to probe this elusive baryon component. We report on observations of one of the brightest X-ray sources at a high redshift, RBS 315 (z = 2.69). Despite several previous analyses, no definite conclusion as to the source of the curvature in its spectrum, whether absorption or intrinsic, could be reached. We present observations by XMM-Newton’s European Photon Imaging Camera (EPIC) and Reflection Grating Spectrometers (RGS) as well as NuSTAR and Swift’s X-Ray Telescope (XRT). The XMM-Newton spectra of this source are of unprecedented quality. A purely statistical analysis of the charge-coupled device (CCD) spectra yields no clear results—the spectrum is as likely to be photoelectrically absorbed as it is to be curved at the source, and no constraint on the position of the absorber can be obtained. Assuming absorption governs the spectral curvature, the lack of absorption lines in the grating spectra indicates the absorber is not well localized in redshift space, and could be dispersed over the cosmological scales of the IGM. Intrinsic curvature, however, can not be unambiguously ruled out.

Witnessing globular cluster formation at z ∼ 3–10 with JWST and ELT

AbstractThe opportunities offered by JWST and the ELT for the detection and study of forming/just formed globular clusters at high redshifts are illustrated, also alluding at the unique insight we may get on the very early stages of galaxy formation.

The role of AGB stars in Galactic and cosmic chemical enrichment

AbstractThe role of asymptotic giant branch (AGB) stars in chemical enrichment is significant for producing 12,13C, 14N, F, 25,26Mg, 17O and slow neutron-capture process (s-process) elements. The contribution from super-AGB stars is negligible in classical, one-zone chemical evolution models, but the mass ranges can be constrained through the contribution from electron-capture supernovae and possibly hybrid C+O+Ne white dwarfs, if they explode as Type Iax supernovae. In addition to the recent s-process yields of AGB stars, we include various sites for rapid neutron-capture processes (r-processes) in our chemodynamical simulations of a Milky Way type galaxy. We find that neither electron-capture supernovae or neutrino-driven winds are able to adequately produce heavy neutron-capture elements such as Eu in quantities to match observations. Both neutron-star mergers (NSMs) and magneto-rotational supernovae (MRSNe) are able to produce these elements in sufficient quantities. Using the distribution in [Eu/(Fe, α)] – [Fe/H], we predict that NSMs alone are unable to explain the observed Eu abundances, but may be able to together with MRSNe. In order to discuss the role of long-lifetime sources such as NSMs and AGB stars at the early stages of galaxy formation, it is necessary to use a model that can treat inhomogeneous chemical enrichment, such as in our chemodynamical simulations. In our cosmological, chemodynamical simulations, we succeed in reproducing the observed N/O-O/H relations both for global properties of galaxies and for local inter-stellar medium within galaxies, without rotation of stars. We also predict the evolution of CNO abundances of disk galaxies, from which it will be possible to constrain the star formation histories.

Chandra and ALMA observations of the nuclear activity in two strongly lensed star-forming galaxies

Aim. According to coevolutionary scenarios, nuclear activity and star formation play relevant roles in the early stages of galaxy formation. We aim at identifying them in high-redshift galaxies by exploiting high-resolution and high-sensitivity X-ray and millimeter-wavelength data to confirm the presence or absence of star formation and nuclear activity and describe their relative roles in shaping the spectral energy distributions and in contributing to the energy budgets of the galaxies. Methods. We present the data, model, and analysis in the X-ray and millimeter (mm) bands for two strongly lensed galaxies, SDP.9 (HATLAS J090740.0-004200) and SDP.11 (HATLAS J091043.1-000322), which we selected in the Herschel-ATLAS catalogs for their excess emission in the mid-IR regime at redshift ≳1.5. This emission suggests nuclear activity in the early stages of galaxy formation. We observed both of them with Chandra ACIS-S in the X-ray regime and analyzed the high-resolution mm data that are available in the ALMA Science Archive for SDP.9. By combining the information available in mm, optical, and X-ray bands, we reconstructed the source morphology. Results. Both targets were detected in the X-ray, which strongly indicates highly obscured nuclear activity. ALMA observations for SDP.9 for the continuum and CO(6-5) spectral line with high resolution (0.02 arcsec corresponding to ~65 pc at the distance of the galaxy) allowed us to estimate the lensed galaxy redshift to a better accuracy than pre-ALMA estimates (1.5753 ± 0.0003) and to model the emission of the optical, millimetric, and X-ray band for this galaxy. We demonstrate that the X-ray emission is generated in the nuclear environment, which strongly supports that this object has nuclear activity. On the basis of the X-ray data, we attempt an estimate of the black hole properties in these galaxies. Conclusions. By taking advantage of the lensing magnification, we identify weak nuclear activity associated with high-z galaxies with high star formation rates. This is useful to extend the investigation of the relationship between star formation and nuclear activity to two intrinsically less luminous high-z star-forming galaxies than was possible so far. Given our results for only two objects, they alone cannot constrain the evolutionary models, but provide us with interesting hints and set an observational path toward addressing the role of star formation and nuclear activity in forming galaxies.

Age dating of old field stars: Challenges from the uncertain efficiency of atomic diffusion

AbstractAge‐dating the Milky Way stellar populations plays a major role in studies of galaxy formation and evolution. Ages of old stars that represent the fossil record of the early stages of galaxy formation are particular important. Current spectroscopic surveys provide effective temperature, surface gravity and photospheric composition of large samples of Milky Way stars, and these data can be used to determine stellar ages by employing stellar evolution models. Here it is shown how current uncertainties on the efficiency of atomic diffusion in the interiors of low‐mass metal poor stars can affect the derived ages at the level of several 10 %. Asteroseismic constraints on the stellar masses can reduce these errors and in the limit of high accuracy can indirectly constrain the efficiency of diffusion in the interiors of these stars.

WHERE ARE MOST OF THE GLOBULAR CLUSTERS IN TODAY’S UNIVERSE?

ABSTRACT The total number of globular clusters (GCs) in a galaxy rises continuously with the galaxy luminosity L, while the relative number of galaxies decreases with L following the Schechter function. The product of these two very nonlinear functions gives the relative number of GCs contained by all galaxies at a given L. It is shown that GCs, in this universal sense, are most commonly found in galaxies within a narrow range around L ⋆. In addition, blue (metal-poor) GCs outnumber the red (metal-richer) ones globally by 4 to 1 when all galaxies are added, pointing to the conclusion that the earliest stages of galaxy formation were especially favorable to forming massive, dense star clusters.

Intermediate-mass black holes in globular clusters: observations and simulations - Update

AbstractThe study of intermediate-mass black holes (IMBHs) is a young and promising field of research. If IMBH exist, they could explain the rapid growth of supermassive black holes by acting as seeds in the early stage of galaxy formation. Formed by runaway collisions of massive stars in young and dense stellar clusters, intermediate-mass black holes could still be present in the centers of globular clusters, today. We measured the inner kinematic profiles with integral-field spectroscopy for 10 Galactic globular cluster and determined masses or upper limits of central black holes. In combination with literature data we further studied the positions of our results on known black-hole scaling relations (such asM•− σ) and found a similar but flatter correlation for IMBHs. Applying cluster evolution codes, the change in the slope could be explained with the stellar mass loss occurring in clusters in a tidal field over its life time. Furthermore, we present results from several numerical simulations on the topic of IMBHs and integral field units (IFUs). N-body simulations were used to simulate IFU data cubes. For the specific case of NGC 6388 we simulated two different IFU techniques and found that velocity dispersion measurements from individual velocities are strongly biased towards lower values due to blends of neighbouring stars and background light. In addition, we use the Astrophysical Multipurpose Software Environment (AMUSE) to combine gravitational physics, stellar evolution and hydrodynamics to simulate the accretion of stellar winds onto a black hole. We find that the S-stars need to provide very strong winds in order to explain the accretion rate in the galactic center.

MODELING THE FORMATION OF GLOBULAR CLUSTER SYSTEMS IN THE VIRGO CLUSTER

The mass distribution and chemical composition of globular cluster (GC) systems preserve fossil record of the early stages of galaxy formation. The observed distribution of GC colors within massive early-type galaxies in the ACS Virgo Cluster Survey (ACSVCS) reveals a multi-modal shape, which likely corresponds to a multi-modal metallicity distribution. We present a simple model for the formation and disruption of GCs that aims to match the ACSVCS data. This model tests the hypothesis that GCs are formed during major mergers of gas-rich galaxies and inherit the metallicity of their hosts. To trace merger events, we use halo merger trees extracted from a large cosmological N-body simulation. We select 20 halos in the mass range of 2 × 1012 to 7 × 1013 M☉ and match them to 19 Virgo galaxies with K-band luminosity between 3 × 1010 and 3 × 1011 L☉. To set the [Fe/H] abundances, we use an empirical galaxy mass–metallicity relation. We find that a minimal merger ratio of 1:3 best matches the observed cluster metallicity distribution. A characteristic bimodal shape appears because metal-rich GCs are produced by late mergers between massive halos, while metal-poor GCs are produced by collective merger activities of less massive hosts at early times. The model outcome is robust to alternative prescriptions for cluster formation rate throughout cosmic time, but a gradual evolution of the mass–metallicity relation with redshift appears to be necessary to match the observed cluster metallicities. We also affirm the age–metallicity relation, predicted by an earlier model, in which metal-rich clusters are systematically several billion younger than their metal-poor counterparts.

Intermediate-mass black holes in globular clusters: observations and simulations

AbstractThe study of intermediate-mass black holes (IMBHs) is a young and promising field of research. If IMBHs exist, they could explain the rapid growth of supermassive black holes by acting as seeds in the early stage of galaxy formation. Formed by runaway collisions of massive stars in young and dense stellar clusters, intermediate-mass black holes could still be present in the centers of globular clusters, today. Our group investigated the presence of intermediate-mass black holes for a sample of 10 Galactic globular clusters. We measured the inner kinematic profiles with integral-field spectroscopy and determined masses or upper limits of central black holes in each cluster. In combination with literature data we further studied the positions of our results on known black-hole scaling relations (such asM•− σ) and found a similar but flatter correlation for IMBHs. Applying cluster evolution codes, the change in the slope could be explained with the stellar mass loss occurring in clusters in a tidal field over its life time. Furthermore, we present results from several numerical simulations on the topic of IMBHs and integral field units (IFUs). We ranN-body simulations of globular clusters containing IMBHs in a tidal field and studied their effects on mass-loss rates and remnant fractions and showed that an IMBH in the center prevents core collapse and ejects massive objects more rapidly. These simulations were further used to simulate IFU data cubes. For the specific case of NGC 6388 we simulated two different IFU techniques and found that velocity dispersion measurements from individual velocities are strongly biased towards lower values due to blends of neighboring stars and background light. In addition, we use the Astrophysical Multipurpose Software Environment (AMUSE) to combine gravitational physics, stellar evolution and hydrodynamics to simulate the accretion of stellar winds onto a black hole.

The Stellar Abundances for Galactic Archaeology (SAGA) Database – III. Analysis of enrichment histories for elements and two modes of star formation during the early evolution of the Milky Way

We study the enrichment histories for nine elements, C, four alpha-elements of Mg, Si, Ca, and Ti, Sc, and three iron-peak elements of Co, Ni, and Zn, by using a large number of stellar data, collected by the Stellar Abundances for Galactic Archaeology (SAGA) database. We find statistically significant changes, or breaks, of the mean abundance ratios to iron at three metallicities of [Fe/H]-1.8, -2.2, and -3.3. Across the first one, the mean abundance ratios decrease with the metallicity by similar extents for all the elements with the sufficient data. Across the latter two, downward trends with the metallicity are also detected but for limited elements, C, Co, Zn, and possibly Sc, and for two of Co and Zn, respectively. The breaks define four stellar populations with the different abundance patters which are dominant in each metallicity range divided by the breaks, Pop IIa, IIb, IIc, and IId in order of increasing metallicity. We also explore their spatial distributions with the spectroscopic distances to demonstrate that Pops IIa and IIb spread over the Galactic halo while Pops IIc and IId are observed near the Galactic plane. In particular, Pop IIc stars emerge around [Fe/H] -2.6 and coexist with Pop IIb stars, segregated by the spatial distributions. Our results reveal two distinct modes of star formation during the early stages of Galaxy formation, which are associated with the variations of IMF and the spatial distribution of remnant low-mass stars. For the two lower-metallicity populations, the enhancements of Zn and Co indicate a high-mass and top-heavy IMF together with the statistics on the carbon-enhanced stars. We discuss the relevance to the kinematically resolved structures of the Galactic halo and the possible sites of these populations within the framework of hierarchical structure formation scenario.

FIRST SPECTROSCOPIC EVIDENCE FOR HIGH IONIZATION STATE AND LOW OXYGEN ABUNDANCE IN Lyα EMITTERS,

We present results from Keck/NIRSPEC and Magellan/MMIRS follow-up spectroscopy of Lya emitters (LAEs) at z=2.2 identified in our Subaru narrowband survey. We successfully detect Ha emission from seven LAEs, and perform a detailed analysis of six LAEs free from AGN activity, two out of which, CDFS-3865 and COSMOS-30679, have [OII] and [OIII] line detections. They are the first [OII]-detected LAEs at high-z, and their [OIII]/[OII] ratios and R23-indices provide the first simultaneous determinations of ionization parameter and oxygen abundance for LAEs. CDFS-3865 has a very high ionization parameter (q_{ion}=2.5^{+1.7}_{-0.8}x10^8 cm s^{-1}) and a low oxygen abundance (12+log(O/H)=7.84^{+0.24}_{-0.25}) in contrast with moderate values of other high-z galaxies such as Lyman-break galaxies (LBGs). COSMOS-30679 also possesses a relatively high ionization parameter (q_{ion}=8^{+10}_{-4}x10^7 cm s^{-1}) and a low oxygen abundance (12+log(O/H)=8.18^{+0.28}_{-0.28}). Both LAEs appear to fall below the mass-metallicity relation of z~2 LBGs. Similarly, a low metallicity of 12+log(O/H)<8.4 is independently indicated for typical LAEs from a composite spectrum and the [NII]/Ha index. Such high ionization parameters and low oxygen abundances can be found in local star-forming galaxies, but this extreme local population occupies only ~0.06% of the SDSS spectroscopic galaxy sample with a number density ~100 times smaller than that of LAEs. With their high ionization parameters and low oxygen abundances, LAEs would represent an early stage of galaxy formation dominated by massive stars in compact star-forming regions. High-q_{ion} galaxies like LAEs would produce ionizing photons efficiently with a high escape fraction achieved by density-bounded HII regions, which would significantly contribute to cosmic reionization at z>6.