Aims. This study aims to trace the chronological evolution of galaxy spectra over cosmic time. Focusing on the VIPERS dataset, we seek to understand the diverse population of galaxies within narrow ...redshift bins, comparing our findings with the previously mapped diversity of SDSS galaxies. Methods. We used Fisher-EM, an unsupervised sub-space model-based classification algorithm to classify a dataset of 79 224 galaxy spectra from the VIMOS Public Extragalactic Redshift Survey (VIPERS). The dataset was divided into 26 samples by bins of redshift ranging from z ∼ 0.4 to z ∼ 1.2, which were classified independently. Classes of subsequent bins were linked through the k-nearest neighbours method to create a chronological tree of classes at different epochs. Results. Based on the optical spectra, three main chronological galaxy branches emerged: (i) red passive, (ii) blue star forming, and (iii) very blue, possibly associated with AGN activity. Each of the branches differentiates into sub-branches, discriminating finer properties such as D4000 break, colour, star-formation rate, and stellar masses, and/or disappear with cosmic time. Notably, these classes align remarkably well with the branches identified in a previous SDSS analyses, indicating a robust and consistent classification across datasets. The chronological ‘tree’ constructed from VIPERS data provides valuable insights into the temporal evolution of these spectral classes. Conclusions. The synergy between VIPERS and SDSS datasets enhances our understanding of the evolutionary pathways of galaxy spectra. The remarkable correspondence between independently derived branches in both datasets underscores the reliability of our unsupervised machine-learning approach. The three sub-trees show complex branching structures that highlight different physical and evolutionary behaviours. This study contributes to the broader comprehension of galaxy evolution by providing a chronologically organised framework for interpreting optical spectra within specific redshift ranges.
Context.
Defining templates of galaxy spectra is useful to quickly characterise new observations and organise databases from surveys. These templates are usually built from a pre-defined ...classification based on other criteria.
Aims.
We present an unsupervised classification of 702 248 spectra of galaxies and quasars with redshifts smaller than 0.25 that were retrieved from the Sloan Digital Sky Survey (SDSS) database, release 7.
Methods.
The spectra were first corrected for redshift, then wavelet-filtered to reduce the noise, and finally binned to obtain about 1437 wavelengths per spectrum. The unsupervised clustering algorithm Fisher-EM, relying on a discriminative latent mixture model, was applied on these corrected spectra. The full set and several subsets of 100 000 and 300 000 spectra were analysed.
Results.
The optimum number of classes given by a penalised likelihood criterion is 86 classes, of which the 37 most populated gather 99% of the sample. These classes are established from a subset of 302 214 spectra. Using several cross-validation techniques we find that this classification agrees with the results obtained on the other subsets with an average misclassification error of about 15%. The large number of very small classes tends to increase this error rate. In this paper, we do an initial quick comparison of our classes with literature templates.
Conclusions.
This is the first time that an automatic, objective and robust unsupervised classification is established on such a large number of galaxy spectra. The mean spectra of the classes can be used as templates for a large majority of galaxies in our Universe.
Aims.
Our study aims at providing deeper insight into the power and limitation of an unsupervised classification algorithm (called Fisher-EM) on spectra of galaxies. This algorithm uses a Gaussian ...mixture in a discriminative latent subspace. To this end, we investigate the capacity of this algorithm to segregate the physical parameters used to generate mock spectra and the influence of the noise on the classification.
Methods.
With the code CIGALE and different values for nine input parameters characterising the stellar population, we simulated a sample of 11 475 optical spectra of galaxies containing 496 monochromatic fluxes. The statistical model and the optimum number of clusters are given in Fisher-EM by the integrated completed likelihood (ICL) criterion. We repeated the analyses several times to assess the robustness of the results.
Results.
Two distinct classifications can be distinguished in the case of the noiseless spectra. The classification with more than 13 clusters disappears when noise is added, while the classification with 12 clusters is very robust against noise down to a signal-to-noise ratio (S/N) of 3. At
S
/
N
= 1, the optimum is 5 clusters, but the classification is still compatible with the previous classification. The distribution of the parameters used for the simulation shows an excellent discrimination between classes. A higher dispersion both in the spectra within each class and in the parameter distribution leads us to conclude that despite a much higher ICL, the classification with more than 13 clusters in the noiseless case is not physically relevant.
Conclusions.
This study yields two conclusions that are valid at least for the Fisher-EM algorithm. Firstly, the unsupervised classification of spectra of galaxies is both reliable and robust to noise. Secondly, such analyses are able to extract the useful physical information contained in the spectra and to build highly meaningful classifications. In an epoch of data-driven astrophysics, it is important to trust unsupervised machine-learning approaches that do not require training samples that are unavoidably biased.
The close environment of the central supennassive black hole of our Galaxy has been studied thoroughly for decades in order to shed light on the behavior of the central regions of galaxies in general ...and of active galaxies in particular. This study is made possible thanks to L-band spectroscopy. This spectral band, from 2.8 to 4.2 mu m, hosts important signatures of the circumstellar medium and interstellar dense and diffuse media among which deep absorption features are attributed to water-ices and hydrocarbons. We observed the GC in the L band of the ISAAC spectrograph located on the UT1/VLT ESO telescope. By mapping the central half parsec using 27 slit positions, we were able to build the first data cube of the region in this wavelength domain. Thanks to a calibrator spectrum of the foreground extinction in the L band derived in a previous paper, we corrected our data cube for the line-of-sight extinction and validated our calibrator spectrum.
Aims.The centre of the Milky Way is the nearest nucleus of a galaxy and offers a unique possibility to study the structure and dynamics of a dense stellar cluster around a super-massive black hole. ...Methods.We present high-resolution seeing limited and AO NIR imaging observations of the stellar cluster within about one parsec of Sgr A*, the massive black hole at the centre of the Milky Way. Stellar number counts and the diffuse background light density were extracted from these observations in order to examine the structure of the nuclear stellar cluster. A detailed map of the variation of interstellar extinction in the central ~0.5 pc of the Milky Way is presented and used to correct the stellar number counts and diffuse light density. Results.Our findings are as follows: (a) a broken-power law provides an excellent fit to the overall structure of the GC nuclear cluster. The power-law slope of the cusp is Γ = 0.19 ± 0.05, the break radius is Rbreak = 6.0” ± 1.0” or 0.22 ± 0.04 pc, and the cluster density decreases with a power-law index of Γ = 0.75 ± 0.1 outside of $R_{\rm break}$. (b) Using the best velocity dispersion measurements from the literature, we derive higher mass estimates for the central parsec than assumed until now. The inferred density of the cluster at the break radius is 2.8 ± 1.3$\times$106 $M_{\odot}$ pc-3. This high density agrees well with the small extent and flat slope of the cusp. Possibly, the mass of the stars makes up only about 50% of the total cluster mass. (c) Possible indications of mass segregation in the cusp are found (d) The cluster appears not entirely homogeneous. Several density clumps are detected that are concentrated at projected distances of R = 3” and R = 7” from Sgr A*. (e) There appears to exist an under-density of horizontal branch/red clump stars near R = 5”, or an over-density of stars of similar brightness at R = 3” and R = 7”. (f) The extinction map in combination with cometary-like features in an L'-band image may provide support for the assumption of an outflow from Sgr A*.
It is often assumed that the strong gravitational field of a super-massive black hole disrupts an adjacent molecular cloud preventing classical star formation in the deep potential well of the black ...hole. Yet, young stars have been observed across the entire nuclear star cluster of the Milky Way including the region close (<0.5 pc) to the central black hole, Sgr A*. Here, we focus particularly on small groups of young stars, such as IRS 13N located 0.1 pc away from Sgr A*, which is suggested to contain about five embedded massive young stellar objects (<1 Myr). We perform three-dimensional hydrodynamical simulations to follow the evolution of molecular clumps orbiting about a 4 × 106 M⊙ black hole, to constrain the formation and the physical conditions of such groups. The molecular clumps in our models are assumed to be isothermal containing 100 M⊙ in <0.2 pc radius. Such molecular clumps exist in the circumnuclear disc of the Galaxy. In our highly eccentrically orbiting clump, the strong orbital compression of the clump along the orbital radius vector and perpendicular to the orbital plane causes the gas densities to increase to values higher than the tidal density of Sgr A*, which are required for star formation. Additionally, we speculate that the infrared excess source G2/DSO approaching Sgr A* on a highly eccentric orbit could be associated with a dust-enshrouded star that may have been formed recently through the mechanism supported by our models.
Context. Processes driving mass assembly are expected to evolve on different timescales along cosmic time. A transition might happen around z similar to 1 as the cosmic star formation rate starts its ...decrease. Aims. We aim to identify the dynamical nature of galaxies in a representative sample to be able to infer and compare the mass assembly mechanisms across cosmic time. Methods. We present an analysis of the kinematics properties of 50 galaxies with redshifts 0.9 \textless z \textless 1.6 from the MASSIV sample observed with SINFONI/VLT with a mass range from 4.5 x 10(9) M-circle dot to 1.7 x 10(11) M-circle dot and a star formation rate from 6 M-circle dot yr(-1) to 300 M-circle dot yr(-1). This is the largest sample with 2D kinematics in this redshift range. We provide a classification based on kinematics as well as on close galaxy environment. Results. We find that a significant fraction of galaxies in our sample (29%) experience merging or have close companions that may be gravitationally linked. This places a lower limit on the fraction of interacting galaxies because ongoing mergers are probably also present but harder to identify. We find that at least 44% of the galaxies in our sample display ordered rotation, whereas at least 35% are non-rotating objects. All rotators except one are compatible with rotation-dominated (V-max/sigma \textgreater 1) systems. Non-rotating objects are mainly small objects (R-e \textless 4 kpc). They show an anti-correlation of their velocity dispersion and their effective radius. These low-mass objects (log M-star \textless 10.5) may be ongoing mergers in a transient state, galaxies with only one unresolved star-forming region, galaxies with an unstable gaseous phase or, less probably, spheroids. Combining our sample with other 3D-spectroscopy samples, we find that the local velocity dispersion of the ionized gas component decreases continuously from z similar to 3 to z = 0. The proportion of disks also seems to be increasing in star-forming galaxies when the redshift decreases. The number of interacting galaxies seems to be at a maximum at z similar to 1.2. Conclusions. These results draw a picture in which cold gas accretion may still be efficient at z similar to 1.2 but in which mergers may play a much more significant role at z similar to 1.2 than at higher redshift. From a dynamical point of view, the redshift range 1 \textless z \textless 2 therefore appears as a transition period in the galaxy mass assembly process(star star star star).
Context. In the near- to mid-infrared wavelength domain, bright continuum sources in the central parsec of the Galactic center (GC) are subject to foreground absorption. These sources therefore ...represent ideal probes of the intervening material that is responsible for the absorption along the line of sight. Aims. Our aim is to shed light on the location and physics of the absorbing clouds. We try to find out which of the gaseous absorbing materials is intimately associated with the GC and which one is associated with clouds at a much larger distance. Methods. We used the capabilities of CRIRES spectrograph located at ESO Very Large Telescope in Chile to obtain absorption spectra of individual lines at a high spectral resolution of R = 50 000, that is, 5 km s−1. We observed the 12CO R(0), P(1), P(2), P(3), P(4), P(5), P(6), P(7) and P(9) transition lines, applied standard data reduction, and compared the results with literature data. Results. We present the results of CRIRES observations of 13 infrared sources located in the central parsec of the Galaxy. The data provide direct evidence for a complex structure of the interstellar medium along the line of sight and in the close environment of the central sources. In particular we find four cold foreground clouds at radial velocities vLSR of the order of −145, −85, −60, and −40 ± 15 km s−1 that show absorption in the lower transition lines from R(0) to P(2) and in all the observed spectra. We also find in all sources an absorption in velocity range of 50–60 km s−1, possibly associated with the so-called 50 km s−1 cloud and suggesting an extension of this cloud in front of the GC. Finally, we detect individual absorption lines that are probably associated with material much closer to the center and with the sources themselves, suggesting the presence of cold gas in the local region.
Context. There are a number of faint compact infrared excess sources in the central stellar cluster of the Milky Way. Their nature and origin is unclear. In addition to several isolated objects of ...this kind there is a small but dense cluster of comoving sources (IRS13N) located ~3′′ west of SgrA* just 0.5′′ north of the bright IRS13E cluster of Wolf-Rayet and O-type stars. Based on the analysis of their color and brightness, there are two main possibilities: (1) they may be dust-embedded stars older than a few Myr; or (2) very young, dusty stars with ages younger than 1 Myr. Aims. We present a first Ks-band identification and proper motions of the IRS13N members, the high-velocity dusty S-cluster object (DSO, also referred to as G2), and other infrared excess sources in the central field. Goal is to constrain the nature of these source. Methods. The L′- (3.8 μm) Ks- (2.2 μm) and H-band (1.65 μm) observations were carried out using the NACO adaptive optics system at the ESO VLT. Proper motions were obtained by linear fitting of the stellar positions extracted by StarFinder as a function of time, weighted by positional uncertainties, and by Gaussian fitting from high-pass filtered and deconvolved images. We also present results of near-infrared (NIR) H- and Ks-band ESO-SINFONI integral field spectroscopy of the Galactic center cluster ISR13N. Results. We show that within the uncertainties, the positions and proper motions of the IRS13N sources in Ks- and L′-band are identical. The HK−sL′ colors then indicate that the bright L′-band IRS13N sources are indeed dust-enshrouded stars rather than core-less dust clouds. The proper motions also show that the IRS13N sources are not strongly gravitationally bound to each other. Combined with their NIR colors, this implies that they have been formed recently. For the DSO we obtain proper motions and a Ks-L′-color. Conclusions. Most of the compact L′-band excess emission sources have a compact H- or Ks-band counterpart and therefore are likely stars with dust shells or disks. Our new results and orbital analysis from our previous work favor the hypothesis that the infrared excess IRS13N members and other dusty sources close to SgrA* are young dusty stars and that star formation at the Galactic center (GC) is a continuously ongoing process. For the DSO the color information indicates that it may be a dust cloud or a dust-embedded star.
The estimate of radial abundance gradients in high-redshift galaxies allows to constrain their star formation history and their interplay with the surrounding intergalactic medium. The authors ...present VLT/SINFONI integral-field spectroscopy of a first sample of 50 galaxies at z similar to 1.2 in the MASSIV survey. Using the N2 ratio between the NII 6584 and H alpha rest-frame optical emission lines as a proxy for oxygen abundance in the interstellar medium, they measured the metallicity of the sample galaxies. Seven galaxies display a positive gradient at a high confidence level. Four out of these are interacting, and one is a chain galaxy. They suggest that interactions might be responsible for shallowing and even inverting the abundance gradient. They also identify two interesting correlations in their sample: galaxies with higher gas velocity dispersion have shallower/positive gradients; and metal-poor galaxies tend to show a positive gradient, whereas metal-rich ones tend to show a negative one.