Search

We present spatially resolved two-dimensional maps and radial trends of the stellar populations and kinematics for a sample of six compact elliptical galaxies (cE) using spectroscopy from the Keck Cosmic Web Imager (KCWI). We recover their star formation histories, finding that all except one of our cEs are old and metal rich, with both age and metallicity decreasing toward their outer radii. We also use the integrated values within one effective radius to study different scaling relations. Comparing our cEs with others from the literature and from simulations we reveal the formation channel that these galaxies might have followed. All our cEs are fast rotators, with relatively high rotation values given their low ellipticites. In general, the properties of our cEs are very similar to those seen in the cores of more massive galaxies, and in particular, to massive compact galaxies. Five out of our six cEs are the result of stripping a more massive (compact or extended) galaxy, and only one cE is compatible with having been formed intrinsically as the low-mass compact object that we see today. These results further confirm that cEs are a mixed-bag of galaxies that can be formed following different formation channels, reporting for the first time an evolutionary link within the realm of compact galaxies (at all stellar masses). [ABSTRACT FROM AUTHOR]

The dominant physical formation mechanism(s) for ultradiffuse galaxies (UDGs) is still poorly understood. Here, we combine new, deep imaging from the Jeanne Rich Telescope with deep integral field spectroscopy from the Keck II telescope to investigate the formation of UDG1137+16 (dw1137+16). Our new analyses confirm both its environmental association with the low density UGC 6594 group, along with its large size of 3.3 kpc and status as a UDG. The new imaging reveals two distinct stellar components for UDG1137+16, indicating that a central stellar body is surrounded by an outer stellar envelope undergoing tidal interaction. Both the components have approximately similar stellar masses. From our integral field spectroscopy, we measure a stellar velocity dispersion within the half-light radius (15 ± 4 km s−1) and find that UDG1137+16 is similar to some other UDGs in that it is likely dark matter dominated. Incorporating literature measurements, we also examine the current state of UDG observational kinematics. Placing these data on the central stellar velocity dispersion–stellar mass relation, we suggest there is little evidence for UDG1137+16 being created through a strong tidal interaction. Finally, we investigate the constraining power current dynamical mass estimates (from stellar and globular cluster velocity dispersions) have on the total halo mass of UDGs. As most are measured within the half-light radius, they are unable to accurately constrain UDG total halo masses. [ABSTRACT FROM AUTHOR]

The ultra diffuse galaxy in the NGC 5846 group (NGC 5846_UDG1) was shown to have a large number of globular cluster (GC) candidates from deep imaging as part of the VEGAS survey. Recently, Müller et al. published a velocity dispersion, based on a dozen of its GCs. Within their quoted uncertainties, the resulting dynamical mass allowed for either a dark matter free or a dark-matter-dominated galaxy. Here, we present spectra from KCWI that reconfirms membership of the NGC 5846 group and reveals a stellar velocity dispersion for UDG1 of σGC = 17 ± 2 km s−1. Our dynamical mass, with a reduced uncertainty, indicates a very high contribution of dark matter within the effective radius. We also derive an enclosed mass from the locations and motions of the GCs using the tracer mass estimator, finding a similar mass inferred from our stellar velocity dispersion. We find no evidence that the galaxy is rotating and is thus likely pressure supported. The number of confirmed GCs, and the total number inferred for the system (∼45), suggests a total halo mass of ∼2 × 1011 M⊙. A cored mass profile is favoured when compared to our dynamical mass. Given its stellar mass of 1.1 × 108 M⊙, NGC 5846_UDG1 appears to be an ultra diffuse galaxy with a dwarf-like stellar mass and an overly massive halo. [ABSTRACT FROM AUTHOR]

We present new spectra obtained using Keck/KCWI and perform kinematics and stellar population analyses of the shell galaxy NGC 474, from both the galaxy centre and a region from the outer shell. We show that both regions have similarly extended star formation histories although with different stellar population properties. The central region of NGC 474 is dominated by intermediate-aged stars (8.3 ± 0.3 Gyr) with subsolar metallicity ([ Z /H] = −0.24 ± 0.07 dex) while the observed shell region, which hosts a substantial population of younger stars, has a mean luminosity-weighted age of 4.0 ± 0.5 Gyr with solar metallicities ([ Z /H] = −0.03 ± 0.09 dex). Our results are consistent with a scenario in which NGC 474 experienced a major to intermediate merger with a log |$(M_*/\rm M_\odot) \sim 10$| mass satellite galaxy at least |${\sim}2$|  Gyr ago which produced its shell system. This work shows that the direct spectroscopic study of low-surface brightness stellar features, such as shells, is now feasible and opens up a new window to understanding galaxy formation and evolution. [ABSTRACT FROM AUTHOR]

We present a catalogue of low surface brightness (LSB) galaxies in the Coma cluster obtained from deep Subaru/Suprime-Cam V - and R -band imaging data within a region of |$\mathord {\sim }4$| deg2. We increase the number of LSB galaxies presented in Yagi et al. (2016) by a factor of |$\mathord {\sim }3$| and report the discovery of 29 new ultradiffuse galaxies (UDGs). We compile the largest sample of UDGs with colours and structural parameters in the Coma cluster. While most UDGs lie along the red-sequence relation of the colour–magnitude diagram, |$\mathord {\sim }16$|  per cent are outside (bluer or redder) the red-sequence region of Coma cluster galaxies. Our analyses show that there is no special distinction in the basic photometric parameters between UDGs and other LSB galaxies. We investigate the clustercentric colour distribution and find a remarkable transition at a projected radius of |$\mathord {\sim }0.6$|  Mpc. Within this cluster core region and relative to the red-sequence of galaxies, LSB galaxies are on average redder than co-spatial higher surface brightness galaxies at the 2σ level, highlighting how vulnerable LSB galaxies are to the physical processes at play in the dense central region of the cluster. The position of the transition radius agrees with expectations from recent cosmological simulation of massive galaxy clusters within which ancient infalls are predicted to dominate the LSB galaxy population. [ABSTRACT FROM AUTHOR]

Ultra-diffuse galaxies (UDGs) reveal extreme properties. Here, we compile the largest study to date of 85 globular cluster (GC) systems around UDGs in the Coma cluster, using new deep ground-based imaging of the known UDGs and existing imaging from the Hubble Space Telescope of their GC systems. We find that the richness of GC systems in UDGs generally exceeds that found in normal dwarf galaxies of the same stellar mass. These GC-rich UDGs imply haloes more massive than expected from the standard stellar mass–halo mass relation. The presence of such overly massive haloes presents a significant challenge to the latest simulations of UDGs in cluster environments. In some exceptional cases, the mass in the GC system is a significant fraction of the stellar content of the host galaxy. We find that rich GC systems tend to be hosted in UDGs of lower luminosity, smaller size, and fainter surface brightness. Similar trends are seen for normal dwarf galaxies in the Coma cluster. A toy model is presented in which the GC-rich UDGs are assumed to be 'failed' galaxies within massive haloes that have largely old, metal-poor, alpha-element-enhanced stellar populations. On the other hand, GC-poor UDGs are more akin to normal, low surface brightness dwarfs that occupy less massive dark matter haloes. Additional data on the stellar populations of UDGs with GC systems will help to further refine and test this simplistic model. [ABSTRACT FROM AUTHOR]

We study the globular clusters (GCs) in the spiral galaxy NGC 5907 well-known for its spectacular stellar stream – to better understand its origin. Using wide-field Subaru/Suprime-Cam gri images and deep Keck/DEIMOS multi-object spectroscopy, we identify and obtain the kinematics of several GCs superimposed on the stellar stream and the galaxy disc. We estimate the total number of GCs in NGC 5907 to be 154 ± 44, with a specific frequency of 0.73 ± 0.21. Our analysis also reveals a significant, new population of young star cluster candidates found mostly along the outskirts of the stellar disc. Using the properties of the stream GCs, we estimate that the disrupted galaxy has a stellar mass similar to the Sagittarius dwarf galaxy accreted by the Milky Way, i.e. |$\sim 10^8~\rm M_\odot$|⁠. [ABSTRACT FROM AUTHOR]

The NGC 1052 group, and in particular the discovery of two ultra-diffuse galaxies with very low internal velocity dispersions, has been the subject of much attention recently. Here we present radial velocities for a sample of 77 globular clusters associated with NGC 1052 obtained on the Keck telescope. Their mean velocity and velocity dispersion are consistent with that of the host galaxy. Using a simple tracer mass estimator, we infer the enclosed dynamical mass and dark matter fraction of NGC 1052. Extrapolating our measurements with a Navarro–Frenk–White (NFW) mass profile we infer a total halo mass of 6.2(±0.2) × 1012 M⊙. This mass is fully consistent with that expected from the stellar mass–halo mass relation, suggesting that NGC 1052 has a normal dark matter halo mass (i.e. it is not deficient in dark matter in contrast to two ultra-diffuse galaxies in the group). We present a phase-space diagram showing the galaxies that lie within the projected virial radius (390 kpc) of NGC 1052. Finally, we briefly discuss the two dark matter-deficient galaxies (NGC 1052-DF2 and NGC 1052-DF4) and consider whether modified Newtonian dynamics (MOND) can account for their low observed internal velocity dispersions. [ABSTRACT FROM AUTHOR]

Ultradiffuse galaxies (UDGs) are a population of extended galaxies but with relatively low luminosities. The origin of these objects remains unclear, largely due to the observational challenges of the low surface brightness Universe. We present here a detailed stellar population analysis of a relatively isolated UDG, DGSAT I, based on spectroscopic data from the Keck Cosmic Web Imager integral field unit. The star formation history of DGSAT I seems to be extended, with a mean luminosity-weighted age of ∼3 Gyr, in agreement with previous photometric studies. However, we find a very high [Mg/Fe] abundance ratio, which is extreme even in the context of the highly alpha-enhanced massive ellipticals and ultrafaint dwarfs. The [Mg/Fe] enhancement of DGSAT I appears to be 10 times higher than the most magnesium-enhanced stellar systems discovered to date, and suggests that the chemical enrichment of this object was dominated by core-collapse supernovae. Intriguingly, this breaks the canonical relation between [Mg/Fe] and star formation time-scale. With a measured velocity dispersion of 56 ± 10 km s−1, DGSAT I also shows a high-mass-to-light ratio, which indicates that it is highly dark matter dominated. The metal-poor conditions of DGSAT I may have enhanced the formation of massive stars, while at the same time, additional mechanisms are needed to prevent iron-rich yields from being recycled into stars. These results suggest that some UDGs could have experienced chemical enrichment episodes similar to the first building blocks of galaxies. [ABSTRACT FROM AUTHOR]

Using data from the WiFeS Atlas of Galactic Globular cluster Spectra, we study the behaviour of the calcium triplet (CaT), a popular metallicity indicator in extragalactic stellar population studies. A major caveat of these studies is that the potential sensitivity to other stellar population parameters such as age, calcium abundance, and the initial mass function has not yet been empirically evaluated. Here, we present measurements of the strength of the CaT feature for 113 globular clusters in the Milky Way and its satellite galaxies. We derive empirical calibrations between the CaT index and both the iron abundance ([Fe/H]) and calcium abundance ([Ca/H]), finding a tighter relationship for [Ca/H] than for [Fe/H]. For stellar populations, 3 Gyr and older the CaT can be used to reliably measure [Ca/H] at the 0.1 dex level but becomes less reliable for ages of ∼2 Gyr and younger. We find that the CaT is relatively insensitive to the horizontal branch morphology. The stellar mass function however affects the CaT strengths significantly only at low metallicities. Using our newly derived empirical calibration, we convert our measured CaT indices into [Ca/H] values for the globular clusters in our sample. [ABSTRACT FROM AUTHOR]

In this second paper of the series we study, with new Keck/DEIMOS spectra, the stellar populations of seven spectroscopically confirmed ultradiffuse galaxies (UDGs) in the Coma cluster. We find intermediate to old ages (∼7 Gyr), low metallicities ([Z/H] ∼ −0.7 dex) and mostly supersolar abundance patterns ([Mg/Fe] ∼ 0.13 dex). These properties are similar to those of low-luminosity (dwarf) galaxies inhabiting the same area in the cluster and are mostly consistent with being the continuity of the stellar mass scaling relations of more massive galaxies. These UDGs’ star formation histories imply a relatively recent infall into the Coma cluster, consistent with the theoretical predictions for a dwarf-like origin. However, considering the scatter in the resulting properties and including other UDGs in Coma, together with the results from the velocity phase-space study of the Paper I in this series, a mixed-bag of origins is needed to explain the nature of all UDGs. Our results thus reinforce a scenario in which many UDGs are field dwarfs that become quenched through their later infall onto cluster environments, whereas some UDGs could be genuine primordial galaxies that failed to develop due to an early quenching phase. The unknown proportion of dwarf-like to primordial-like UDGs leaves the enigma of the nature of UDGs still open. [ABSTRACT FROM AUTHOR]

Recovering the origins of lenticular galaxies can shed light on the understanding of galaxy and compare them with the kinematics of planetary nebulae (PNe). The PNe and GC data come from the Planetary Nebulae Spectrograph and the SLUGGS Surveys. Through photometric spheroid-disc decomposition and PNe kinematics. we find the probability for a given GC to belong to either the spheroid or the disc of its host galaxy or be rejected from the model. We find that there is no correlation between the components that the GCs are likely to belong to and their colours. Particularly, for NGC 2768, we find that its red GCs display rotation preferentially at inner radii (Re < 1). In the case of the GC system of NGC 3115, we find a group of GCs with similar kinematics that are not likely to belong to either its spheroid or disc. For NGC 7457, we find that 70 per cent of its GCs are likely to belong to the disc. Overall, our results suggest that these galaxies assembled into S0s through different evolutionary paths. Mergers seem to have been very important for NGC 2768 and NGC 3115 while NGC 7457 is more likely to have experienced secular evolution. [ABSTRACT FROM AUTHOR]

We use Keck/DEIMOS spectroscopy to confirm the cluster membership of 16 ultradiffuse galaxies (UDGs) in the Coma cluster, bringing the total number of spectroscopically confirmed UDGs from the Yagi et al. (Y16) catalogue to 25. We also identify a new cluster background UDG, confirming that most (∼95 per cent) of the UDGs in the Y16 catalogue belong to the Coma cluster. In this pilot study of Coma UDGs in velocity phase space, we find evidence of a diverse origin for Coma cluster UDGs, similar to normal dwarf galaxies. Some UDGs in our sample are consistent with being late infalls into the cluster environment, while some may have been in the cluster for ≥8 Gyr. The late infallen UDGs have higher absolute relative line-of-sight velocities, bluer optical colours, and within the projected cluster core, are smaller in size, compared to the early infalls. The early infall UDGs, which may also have formed in situ, have been in the cluster environment for as long as the most luminous galaxies in the Coma cluster, and they may be failed galaxies that experienced star formation quenching at earlier epochs. [ABSTRACT FROM AUTHOR]

We apply the Jeans Anisotropic Multi-Gaussian Expansion dynamical modelling method to SAGES Legacy Unifying Globulars and GalaxieS (SLUGGS) survey data of early-type galaxies in the stellar mass range 1010 < M*/Mʘ < 1011.6 that cover a large radial range of 0.1-4.0 effective radii.We combine SLUGGS and ATLAS3D data sets to model the total-mass profiles of a sample of 21 fast-rotator galaxies, utilizing a hyperparameter method to combine the two independent data sets. The total-mass density profile slope values derived for these galaxies are consistent with those measured in the inner regions of galaxies by other studies. Furthermore, the total-mass density slopes (γ tot) appear to be universal over this broad stellar mass range, with an average value of γ tot= -2.24± 0.05 , i.e. slightly steeper than isothermal. We compare our results to model galaxies from the Magneticum and EAGLE cosmological hydrodynamic simulations, in order to probe the mechanisms that are responsible for varying total-mass density profile slopes. The simulated-galaxy slopes are shallower than the observed values by ***0.3--0.5, indicating that the physical processes shaping the mass distributions of galaxies in cosmological simulations are still incomplete. For galaxies with M* > 1010.7Mʘ in the Magneticum simulations, we identify a significant anticorrelation between total-mass density profile slopes and the fraction of stellar mass formed ex situ (i.e. accreted), whereas this anticorrelation is weaker for lower stellar masses, implying that the measured total-mass density slopes for low-mass galaxies are less likely to be determined by merger activity. [ABSTRACT FROM AUTHOR]

We use globular cluster kinematics data, primarily from the SAGES Legacy Unifying Globulars and GalaxieS (SLUGGS) survey, to measure the dark matter fraction (fDM) and the average dark matter density (〈ρDM〉) within the inner 5 effective radii (Re) for 32 nearby early-type galaxies (ETGs) with stellar mass log (M∗/M⊚) ranging from 10.1 to 11.8. We compare our results with a simple galaxy model based on scaling relations as well as with cosmological hydrodynamical simulations where the dark matter profile has been modified through various physical processes. We find a high fDM (≥0.6) within 5 Re in most of our sample, which we interpret as a signature of a late mass assembly history that is largely devoid of gas-rich major mergers. However, around log (M∗/M⊚) ∼ 11, there is a wide range of fDM which may be challenging to explain with any single cosmological model. We find tentative evidence that lenticulars (S0s), unlike ellipticals, have mass distributions that are similar to spiral galaxies, with decreasing fDM within 5 Re as galaxy luminosity increases. However, we do not find any difference between the 〈ρDM〉 of S0s and ellipticals in our sample, despite the differences in their stellar populations.We have also used 〈ρDM〉 to infer the epoch of halo assembly (z ∼ 2– 4). By comparing the age of their central stars with the inferred epoch of halo formation, we are able to gain more insight into their mass assembly histories. Our results suggest a fundamental difference in the dominant late-phase mass assembly channel between lenticulars and elliptical galaxies. [ABSTRACT FROM AUTHOR]

We present the WiFeS Atlas of Galactic Globular cluster Spectra, a library of integrated spectra of Milky Way and Local Group globular clusters. We used the WiFeS integral field spectrograph on the Australian National University 2.3 m telescope to observe the central regions of 64 MilkyWay globular clusters and 22 globular clusters hosted by the MilkyWay's low-mass satellite galaxies. The spectra have wider wavelength coverage (3300-9050 Å) and higher spectral resolution (R = 6800) than existing spectral libraries of Milky Way globular clusters. By including Large and SmallMagellanic Cloud star clusters, we extend the coverage of parameter space of existing libraries towards young and intermediate ages. While testing stellar population synthesis models and analysis techniques is the main aim of this library, the observations may also further our understanding of the stellar populations of Local Group globular clusters and make possible the direct comparison of extragalactic globular cluster integrated light observations with well-understood globular clusters in the Milky Way. The integrated spectra are publicly available via the project website. [ABSTRACT FROM AUTHOR]

Here we utilize recent measures of galaxy total dynamical mass and X-ray gas luminosities (LX, Gas) for a sample of 29 massive early-type galaxies from the SLUGGS survey to probe LX, Gas-mass scaling relations. In particular, we investigate scalings with stellar mass, dynamical mass within 5 effective radii (Re) and total virial mass. We also compare these relations with predictions from Λ cold dark matter simulations. We find a strong linear relationship between LX, Gas and galaxy dynamical mass within 5Re, which is consistent with the recent cosmological simulations of Choi et al, that incorporate mechanical heating from AGN. We conclude that the gas surrounding massive early-type galaxies was shock-heated as it fell into collapsing dark matter haloes so that LX, Gas is primarily driven by the depth of a galaxy's potential well. Heating by an AGN plays an important secondary role in determining LX, Gas. [ABSTRACT FROM AUTHOR]

The dark matter (DM) haloes around spiral galaxies appear to conspire with their baryonic content: empirically, significant amounts of DM are inferred only below a universal characteristic acceleration scale. Moreover, the discrepancy between the baryonic and dynamical mass, which is usually interpreted as the presence of DM, follows a very tight mass discrepancy acceleration (MDA) relation. Its universality, and its tightness in spiral galaxies, poses a challenge for the DM interpretation and was used to argue in favour of MOdified Newtonian Dynamics (MOND). Here, we test whether or not this applies to early-type galaxies.We use the dynamical models of fast-rotator early-type galaxies by Cappellari et al. based on ATLAS3D and SAGES Legacy Unifying Globulars and GalaxieS (SLUGGS) data, which was the first homogenous study of this kind, reaching ~4 Re, where DM begins to dominate the total mass budget. We find the early-type galaxies to follow an MDA relation similar to spiral galaxies, but systematically offset. Also, while the slopes of the mass density profiles inferred from galaxy dynamics show consistency with those expected from their stellar content assuming MOND, some profiles of individual galaxies show discrepancies. [ABSTRACT FROM AUTHOR]

We study mass distributions within and beyond 5 effective radii (Re) in 23 early-type galaxies from the SAGES Legacy Unifying Globulars and Galaxies Survey, using their globular cluster (GC) kinematic data. The data are obtained with Keck/DEep Imaging Multi-Object Spectrograph, and consist of line-of-sight velocities for ~3500 GCs, measured with a high precision of ~15?km?s-1 per GC and extending out to ~13?Re. We obtain the mass distribution in each galaxy using the tracer mass estimator of Watkins et al. and account for kinematic substructures, rotation of the GC systems and galaxy flattening in our mass estimates. The observed scatter between our mass estimates and results from the literature is less than 0.2 dex. The dark matter fraction within 5 Re (fDM) increases from ~0.6 to ~0.8 for low- and high-mass galaxies, respectively, with some intermediate-mass galaxies (M* ~1011 M⊙) having low fDM ~ 0.3, which appears at odds with predictions from simple galaxy models. We show that these results are independent of the adopted orbital anisotropy, stellar mass-to-light (M/L) ratio, and the assumed slope of the gravitational potential. However, the low fDM in the ~1011 M⊙ galaxies agrees with the cosmological simulations of Wu et al. where the pristine dark matter distribution has been modified by baryons during the galaxy assembly process. We find hints that these M* ~1011M⊙ galaxies with low fDM have very diffuse dark matter haloes, implying that they assembled late. Beyond 5 Re, the M/L gradients are steeper in the more massive galaxies and shallower in both low and intermediate mass galaxies. [ABSTRACT FROM AUTHOR]

We present an investigation of the globular cluster (GC) systems of NGC 3607 and NGC 3608 as part of the ongoing SLUGGS (SAGES Legacy Unifying Globulars and GalaxieS) survey. We use wide-field imaging data from the Subaru telescope in the g, r and i filters to analyse the radial density, colour and azimuthal distributions of both GC systems. With the complementary kinematic data obtained from the Keck II telescope, we measure the radial velocities of a total of 81 GCs. Our results show that the GC systems of NGC 3607 and NGC 3608 have a detectable spatial extent of ~15 and 13 galaxy effective radii, respectively. Both GC systems show a clear bimodal colour distribution. We detect a significant radial colour gradient for the GC subpopulations in both galaxies. NGC 3607 exhibits an overabundance of red GCs on the galaxy minor axis and NGC 3608 shows a misalignment in the GC subpopulation position angles with respect to the galaxy stellar component. With the aid of literature data, we discuss several relationships between the properties of GC systems and their host galaxies. A one-to-one relation between the ellipticities of red GCs and the galaxy stellar light emphasizes the evolutionary similarities between them. In our sample of four slowly rotating galaxies with kinematically decoupled cores, we observe a higher ellipticity for the blue GC subpopulation than their red counterparts. Also, we notice the flattening of negative colour gradients for the blue GC subpopulations with increasing galaxy stellar mass. Finally, we discuss the formation scenarios associated with the blue GC subpopulation. [ABSTRACT FROM AUTHOR]

Globular clusters (GCs) can be considered discrete, long-lived, dynamical tracers that retain crucial information about the assembly history of their parent galaxy. In this paper, we present a new catalogue of GC velocities and colours for the lenticular galaxy NGC 1023, we study their kinematics and spatial distribution, in comparison with the underlying stellar kinematics and surface brightness profile, and we test a new method for studying GC properties. Specifically, we decompose the galaxy light into its spheroid (assumed to represent the bulge+halo components) and disc components and use it to assign to each GC a probability of belonging to one of the two components. Then we model the galaxy kinematics, assuming a disc and spheroidal component, using planetary nebulae and integrated stellar light. We use this kinematic model and the probability previously obtained from the photometry to recalculate for each GC its likelihood of being associated with the disc, the spheroid, or neither. We find that the reddest GCs are likely to be associated with the disc, as found for faint fuzzies in this same galaxy, suggesting that the disc of this S0 galaxy originated at z ∍ 2. The majority of blue GCs are found likely to be associated with the spheroidal (hot) component. The method also allows us to identify objects that are unlikely to be in equilibrium with the system. In NGC 1023 some of the rejected GCs form a substructure in phase space that is connected with NGC 1023 companion galaxy. [ABSTRACT FROM AUTHOR]

NGC 4473 is a so-called double sigma (2σ) galaxy, i.e. a galaxy with rare, double peaks in its 2D stellar velocity dispersion. Here, we present the globular cluster (GC) kinematics in NGC 4473 out to ~10Re (effective radii) using data from combined Hubble Space Telescope/Advanced Camera for Surveys and Subaru/Suprime-Cam imaging and Keck/Deep Imaging Multi-Object Spectrograph. We find that the 2σ nature of NGC 4473 persists up to 3Re, though it becomes misaligned to the photometric major axis. We also observe a significant offset between the stellar and GC rotation amplitudes. This offset can be understood as a co-addition of counter-rotating stars producing little net stellar rotation. We identify a sharp radial transition in the GC kinematics at ~4Re suggesting a well defined kinematically distinct halo. In the inner region (<4Re), the blue GCs rotate along the photometric major axis, but in an opposite direction to the galaxy stars and red GCs. In the outer region (>4Re), the red GCs rotate in an opposite direction compared to the inner region red GCs, along the photometric major axis, while the blue GCs rotate along an axis intermediate between the major and minor photometric axes. We also find a kinematically distinct population of very red GCs in the inner region with elevated rotation amplitude and velocity dispersion. The multiple kinematic components in NGC 4473 highlight the complex formation and evolutionary history of this 2σ galaxy, as well as a distinct transition between the inner and outer components. [ABSTRACT FROM AUTHOR]