We perform an objective classification of 170,000 galaxy spectra in the Sloan Digital Sky Survey (SDSS) using the Karhunen-Loève (KL) transform. With about one-sixth of the total set of galaxy spectra that will be obtained by the survey, we are able to carry out the most extensive analysis of its kind to date. The formalism proposed by Connolly and Szalay is adopted to correct for gappy regions in the spectra and to derive eigenspectra and eigencoefficients. From this analysis, we show that this gap-correction formalism leads to a converging set of eigenspectra and KL-repaired spectra. Furthermore, KL eigenspectra of galaxies are found to be convergent not only as a function of iteration, but also as a function of the number of randomly selected galaxy spectra used in the analysis. From these data a set of 10 eigenspectra of galaxy spectra are constructed, with rest-wavelength coverage 3450-8350 8. The eigencoefficients describing these galaxies naturally place the spectra into several classes defined by the plane formed by the first three eigencoefficients of each spectrum. Spectral types corresponding to different Hubble types and galaxies with extreme emission lines are identified for the 170,000 spectra and are shown to be complementary to existing spectral classifications. From a nonparametric classification technique, we find that the population of galaxies can be divided into three classes that correspond to early late-type through intermediate late-type galaxies. This finding is believed to be related to the color separation of SDSS galaxies discussed in earlier works. Bias in the spectral classifications due to the aperture spectroscopy in the SDSS is small and within the signal-to-noise limit for a majority of galaxies, except for the reddest nearby galaxies and large galaxies (>30 kpc) with prominent emissions. The mean spectra and eigenspectra derived from this work can be downloaded from the SDSS Web site.

Using an eigenspectrum decomposition technique, we separate the host galaxy from the broad line active galactic nucleus (AGN) in a set of 4666 spectra from the Sloan Digital Sky Survey (SDSS), from redshifts near zero up to about 0.75. The decomposition technique uses separate sets of galaxy and quasar eigenspectra to efficiently and reliably separate the AGN and host spectroscopic components. The technique accurately reproduces the host galaxy spectrum, its contributing fraction, and its classification. We show how the accuracy of the decomposition depends upon S/N, host galaxy fraction, and the galaxy class. Based on the eigencoefficients, the sample of SDSS broad-line AGN host galaxies spans a wide range of spectral types, but the distribution differs significantly from inactive galaxies. In particular, post-starburst activity appears to be much more common among AGN host galaxies. The luminosities of the hosts are much higher than expected for normal early-type galaxies, and their colors become increasingly bluer than early-type galaxies with increasing host luminosity. Most of the AGNs with detected hosts are emitting at between 1% and 10% of their estimated Eddington luminosities, but the sensitivity of the technique usually does not extend to the Eddington limit. There are mild correlations among the AGN and host galaxy eigencoefficients, possibly indicating a link between recent star formation and the onset of AGN activity. The catalog of spectral reconstruction parameters is available as an electronic table.Comment: 18 pages; accepted for publication in A

Void galaxies, residing within the deepest underdensities of the Cosmic Web, present an ideal population for the study of galaxy formation and evolution in an environment undisturbed by the complex processes modifying galaxies in clusters and groups, as well as provide an observational test for theories of cosmological structure formation. We have completed a pilot survey for the H i imaging aspects of a new Void Galaxy Survey (VGS), imaging 15 void galaxies in H i in local (d < 100 Mpc) voids. H i masses range from 3.5 × 10 8 to 3.8 × 10 9 M , with one nondetection with an upper limit of 2.1 × 10 8 M . Our galaxies were selected using a structural and geometric technique to produce a sample that is purely environmentally selected and uniformly represents the void galaxy population. In addition, we use a powerful new backend of the Westerbork Synthesis Radio Telescope that allows us to probe a large volume around each targeted galaxy, simultaneously providing an environmentally constrained sample of fore-and background control sample of galaxies while still resolving individual galaxy kinematics and detecting faint companions in H i. This small sample makes up a surprisingly interesting collection of perturbed and interacting galaxies, all with small stellar disks. Four galaxies have significantly perturbed H i disks, five have previously unidentified companions at distances ranging from 50 to 200 kpc, two are in interacting systems, and one was found to have a polar H i disk. Our initial findings suggest void galaxies are a gas-rich, dynamic population which present evidence of ongoing gas accretion, major and minor interactions, and filamentary alignment despite the surrounding underdense environment.

Extinction in galaxies affects their observed properties. In scenarios describing the distribution of dust and stars in individual disk galaxies, the amplitude of the extinction can be modulated by the inclination of the galaxies. In this work we investigate the inclination dependency in composite spectra of star-forming disk galaxies from the Sloan Digital Sky Survey Data Release 5. In a volumelimited sample within a redshift range 0.065 − 0.075 and a r-band Petrosian absolute magnitude range -19.5 to -22 mag which exhibits a flat distribution of inclination, the inclined relative to face-on extinction in the stellar continuum is found empirically to increase with inclination in the g, r and i bands. Within the central 0.5 intrinsic half-light radius of the galaxies, the g-band relative extinction in the stellar continuum for the highly-inclined objects (axis ratio b/a = 0.1) is 1.2 mag, agreeing with previous studies. The extinction curve of the disk galaxies is given in the restframe wavelengths 3700 −8000Å, identified with major optical emission and absorption lines in diagnostics. The Balmer decrement, Hα/Hβ, remains constant with inclination, suggesting a different kind of dust configuration and/or reddening mechanism in the H II region from that in the stellar continuum. One factor is shown to be the presence of spatially non-uniform interstellar extinction, presumably caused by clumped dust in the vicinity of the H II region.

We present a novel technique to overcome the limitations of the applicability of principal component analysis to typical real‐life data sets, especially astronomical spectra. Our new approach addresses the issues of outliers, missing information, large number of dimensions and the vast amount of data by combining elements of robust statistics and recursive algorithms that provide improved eigensystem estimates step by step. We develop a generic mechanism for deriving reliable eigenspectra without manual data censoring, while utilizing all the information contained in the observations. We demonstrate the power of the methodology on the attractive collection of the Visible Imaging Multi‐Object Spectrograph (VIMOS) Very Large Telescope (VLT) Deep Survey spectra that manifest most of the challenges today, and highlight the improvements over previous workarounds, as well as the scalability of our approach to collections with sizes of the Sloan Digital Sky Survey and beyond.