Thaisa Storchi-Bergmann

Dust extinction of the stellar continua in starburst galaxies: The ultraviolet and optical extinction law

We analyze the International Ultraviolet Explorer (IUE) UV and the optical spectra of 39 starburst and blue compact galaxies in order to study the average properties of dust extinction in extended regions of galaxies. The optical spectra have been obtained using an aperture which matches that of IUE, so comparable regions within each galaxy are sampled. The data from the 39 galaxies are compared with five models for the geometrical distribution of dust, adopting as extinction laws both the Milky Way and the Large Magellanic Cloud laws. The commonly used uniform dust screen is included among the models. We find that none of the five models is in satisfactory agreement with the data. In order to understand the discrepancy between the data and the models, we have derived an extinction law directly from the data in the UV and optical wavelength range. The resulting curve is characterized by an overall slope which is more gray than the Milky Way extinction laws slope, and by the absence of the 2175 A dust feature. Remarkably, the difference in optical depth between the Balmer emission lines H(sub alpha) and H(sub beta) is about a factor of 2 larger than the difference in the optical depth between the continuum underlying the two Balmer lines. We interpret this discrepancy as a consequence of the fact that the hot ionizing stars are associated with dustier regions than the cold stellar population is. The absence of the 2175 A dust feature can be due either to the effects of the scattering and clumpiness of the dust or to a chemical composition different from that of the Milky Way dust grains. Disentangling the two interpretations is not easy because of the complexity of the spatial distribution of the emitting regions. The extinction law of the UV and optical spectral continua of extended regions can be applied to the spectra of medium- and high-redshift galaxies, where extended regions of a galaxy are, by necessity, sampled.

Template Ultraviolet to Near-Infrared Spectra of Star-forming Galaxies and Their Application to K-Corrections

Template UV-optical spectra of quiescent and starburst galaxies are presented and used to derive Kcorrections as a function of morphological type and redshift. IUE observations and archival data are used for the UV template spectra. The optical spectra are from ground-based observations obtained in apertures that match closely the 200 arcsec 2 IUE aperture. The templates of quiescent galaxies are built according to morphological type, elliptical, bulge, SO, Sa, Sb, and So, and the templates of starburst galaxies according to color excess. The unprecedented characteristics of these templates is that UV and optical spectra have been obtained in matched apertures to produce consistent spectral information from 1200 to 10,000 A. Despite the relatively small IUE aperture, the galaxy stellar populations are well represented in the elliptical, SO, Sa, and Sc, and in the starburst templates. The spectra are available digitally. The UV-optical templates can be applied to the classification of high-redshift galaxies and to the identification of the host galaxies of quasars. The templates predict that observed magnitudes from traditional ground-based photometric surveys can be uniquely interpreted. For example, U, B, and I magnitudes uniquely determine both the redshift and the morphological type of a galaxy. The template spectra are also used to calculate K-corrections for galaxies as a function of morphological type and redshift, up to z = 2. These improved K-corrections are not sufficient to explain the excess counts in faint blue galaxies. A subset of our galaxy templates are linked with published data from the radio to the X-ray for galaxies and quasars. A comparison between the quiescent galaxies and the quasars suggests that, in the optical band, the host galaxy is a factor of 10-100 fainter in flux than the quasar. Subject headings: galaxies: distances and redshifts --galaxies: photometry -galaxies: starburst -galaxies: stellar content -quasars: general

Ultraviolet to near-infrared spectral distributions of star-forming galaxies : metallicity and age effects

Spectral distributions from the UV to the near-IR of a sample of 44 star-forming galaxies are used to calculate the metallicity (O/H), star-formation rate (SFR) and age of the starbursts. The oxygen abundance covers the range 8.3 less than O/H less than 9.4 and nitrogen (N) is found to be mostly a product of secondary nucleosynthesis for O/H greater than 8.4. Due to its secondary origin, N/O ratios up to approximately equals 4 times the solar value can be obtained for metal-rich starbursts. The SFR ranges 0.01 to 100 solar mass/year. The lower metallicity galaxies seem to be experiencing an instantaneous burst of star formation, with ages ranging from under 5 x 10(exp 6) to 10(exp 7) yr. The highest metallicity galaxies are most probably experiencing a continuous burst. Correlations between the calculated quantities and several spectral features are investigated. We found a highly significant correlation between the equivalent width W(C IV lambda 1550)-a stellar (absorption) feature- and the oxygen abundance of the emitting gas (O/H). Thus we show for the first time that the stellar metallicity is well correlated with the gas metallicity in star-bursting galaxies. The equivalent width W(Si IV lambda 1400) and the emission line ratio (N II) lambda lambda 6548.84/H(sub alpha) also correlate well with O/H, and all three features can be used as metallicity indicators for star-forming galaxies. The continuum color between lambda 1400 and lambda 3500 (C(14 - 35)) is shown to correlate with O/H, although it is better correlated with E(B - V). It was not possible to disentangle the metallicity from the reddening effect in C(14- 35). We estimate that the reddening affecting the UV continuum is about half the one derived from the Balmer decrement of the emitting gas. The SFR correlates well with the galaxy luminosity and there is no dependence of the continuum color on the SFR. The higher metallicities are only found in the more luminous galaxies, while low metallicities are found over the whole luminosity interval (-16 less than M(sub B) less than -23, H(sub 0) = 50 km/sec/Mpc) covered by the sample.

On the nature of Seyfert galaxies with high [O III] λ5007 blueshifts

We have studied the properties of Seyfert galaxies with high [O III] ?5007 blueshifts (blue outliers), originally identified because of their strong deviation from the -->MBH ? ? relation of normal, narrow-line Seyfert 1 (NLS1) and broad-line Seyfert 1 (BLS1) galaxies. These blue outliers turn out to be important test beds for models of the narrow-line region (NLR), for mechanisms of driving large-scale outflows, for links between NLS1 galaxies and radio galaxies, and for orientation-dependent NLS1 models. We report the detection of a strong correlation of line blueshift with ionization potential in each galaxy, including the measurement of coronal lines with radial velocities up to 500-1000 km s -->?1, and we confirm a strong correlation between [O III] blueshift and line width. All [O III] blue outliers have narrow widths of their broad Balmer lines and high Eddington ratios. While the presence of nonshifted low-ionization lines signifies the presence of a classical outer quiescent NLR in blue outliers, we also report the absence of any second, nonblueshifted [O III] component from a classical inner NLR. These results place tight constraints on NLR models. We favor a scenario in which the NLR clouds are entrained in a decelerating wind, which explains the strong stratification and the absence of a zero-blueshift inner NLR of blue outliers. The origin of the wind remains speculative at this time (collimated radio plasma, thermal winds, or radiatively accelerated clouds). It is perhaps linked to the high Eddington ratios of blue outliers. Similar, less powerful winds could be present in all Seyfert galaxies, but would generally only affect the coronal line region (CLR), or level off even before reaching the CLR. Similarities between blue outliers in NLS1 galaxies and (compact) radio sources are briefly discussed.

Double-peaked broad line emission from the LINER nucleus of NGC 1097

We report the recent appearance of a very broad component in the H-alpha and H-beta emission lines of the weakly active nucleus of the Sersic-Pastoriza galaxy NGC 1097. The FWZI of the broad component is about 21,000 km/s, and its profile is double-peaked; the presence of a blue, featureless continuum in the nucleus is also suggested. The broad component was first observed in H-alpha in November 2, 1991, and confirmed 11 months later. The H-alpha profile and flux did not change in this time interval. Comparison with previously published spectral data indicates that the broad lines have only recently appeared. Together with the relatively high X-ray luminosity and the compact nuclear radio source, our results characterize the presence of a Seyfert 1 nucleus in a galaxy which had previously shown only LINER characteristics. Obscuring material along our line of sight to the nucleus appears to have recently cleared, permitting a direct view of the active nucleus. We discuss two possible structures for the broad line region, biconical outflow and an accretion disk, that could give rise to the observed profile.

Elliptical accretion disks in active galactic nuclei

We present a calculation of the profiles of emission lines originating in a relativistic, eccentric disk, and show examples of the resulting model profiles. Our calculations are motivated by the fact that in about one-quarter of the double-peaked emission lines observed in radio-loud active galactic nuclei (and in the mildly active nucleus of NGC 1097), the red peak is stronger than the blue peak, which is contrary to the prediction of relativistic, circular disk models. Using the eccentric disk model we fit some of the observed profiles that cannot be fitted with a circular disk model. We propose two possible scenarios for the formation of an eccentric disk in an active galactic nucleus: (a) tidal perturbation of the disk around a supermassive black hole by a smaller binary companion, and (b) formation of an elliptical disk from the debris resulting from the tidal disruption of a star by the central black hole. In the former case we show that the eccentricity can be long-lived because of the presence of the binary companion. In the latter case, although the inner parts of the disk may circularize quickly, we estimate that the outer parts will maintain their eccentricity for times much longer than the local viscous time. We suggest that it may be possible to detect profile variability on much shorter timescales than those ranging from a decade to several centuries by comparing the evolution of the line profile with detailed model predictions. We argue that line-profile variability may also be the most promising discriminant among competing models for the origin of asymmetric, double-peaked emission lines.

A Strong Correlation between Circumnuclear Dust and Black Hole Accretion in Early-Type Galaxies

We present a detailed investigation of the incidence of circumnuclear dust structure in a large, well-matched sample of early-type galaxies with and without active galactic nuclei (AGNs). All 34 early-type AGN hosts in our sample have circumnuclear dust, while dust is only observed in 26% (9) of a pair-matched sample of 34 early-type, inactive galaxies. This result demonstrates a strong correlation between the presence of circumnuclear dust and accretion onto the central, supermassive black hole in elliptical and lenticular galaxies. This correlation is not present at later Hubble types, where a sample of 31 active and 31 inactive galaxies all contain circumnuclear dust. These archival, Hubble Space Telescope observations reveal a wide range of mostly chaotic dust morphologies. Current estimates suggest the dust settling or destruction time is on order of 108 yr, and therefore the presence of dust in ~50% of early-type galaxies requires frequent replenishment and similarly frequent fueling of their central supermassive black holes. The observed dust could be internally produced (via stellar winds) or externally accreted, although there are observational challenges for both of these scenarios. Our analysis also reveals that approximately one-third of the early-type galaxies without circumnuclear dust have nuclear stellar disks. These nuclear stellar disks may provide a preferred kinematic axis to externally accreted material, and this material may in turn form new stars in these disks. The observed incidence of nuclear stellar disks and circumnuclear dust suggests that episodic replenishment of nuclear stellar disks occurs and is approximately concurrent with the fueling of the central AGN.

Feeding versus feedback in NGC 4151 probed with Gemini NIFS – II. Kinematics

We have used the Gemini Near-infrared Integral Field Spectrograph (NIFS) to map the gas kinematics of the inner ∼200 x 500 pc 2 of the Seyfert galaxy NGC 4151 in the Z, J, H and K bands at a resolving power of ≥5000 and spatial resolution of ∼8 pc. The ionized gas emission is most extended along the known ionization bi-cone at position angle PA = 60°-240°, but is observed also along its equatorial plane. This indicates that the active galactic nucleus (AGN) ionizes gas beyond the borders of the bi-cone, within a sphere with ≈1 arcsec radius around the nucleus. The ionized gas has three kinematic components: (1) one observed at the systemic velocity and interpreted as originating in the galactic disc; (2) one outflowing along the bi-cone, with line-of-sight velocities between -600 and 600 km s ―1 and strongest emission at ±(100―300) km s ―1 ; and (3) another component due to the interaction of the radio jet with ambient gas. The radio jet (at PA = 75°-255°) is not aligned with the narrow-line region (NLR) and produces flux enhancements mostly observed at the systemic velocity, suggesting that the jet is launched close to the plane of the galaxy (approximately plane of the sky). The mass outflow rate, estimated to be ≈ 1 M ⊙ yr ―1 along each cone, exceeds the inferred black hole accretion rate by a factor of ∼100. This can be understood if the NLR is formed mostly by entrained gas from the circumnuclear interstellar medium by an outflow probably originating in the accretion disc. This flow represents feedback from the AGN, estimated to release a kinetic power of Ė ≈ 2.4 x 10 41 erg s ―1 , which is only ∼0.3 per cent of the bolometric luminosity of the AGN. There is no evidence in our data for the gradual acceleration followed by gradual deceleration proposed by previous modelling of the [O III] emitting gas. Our data allow the possibility that the NLR clouds are accelerated close to the nucleus (within 0.1 arcsec, which corresponds to ≈7 pc at the galaxy) after which the flow moves at essentially constant velocity (≈600 km s ―1 ), being consistent with near-infrared emission arising predominantly from the interaction of the outflow with gas in the galactic disc. The molecular gas exhibits distinct kinematics relative to the ionized gas. Its emission arises in extended regions approximately perpendicular to the axis of the bi-cone and along the axis of the galaxys stellar bar, avoiding the innermost ionized regions. It does not show an outflowing component, being observed only at velocities very close to systemic, and is thus consistent with an origin in the galactic plane. This hot molecular gas may only be the tracer of a larger reservoir of colder gas which represents the AGN feeding.

THE STELLAR POPULATIONS OF LOW-LUMINOSITY ACTIVE GALACTIC NUCLEI. I. GROUND-BASED OBSERVATIONS

We present a spectroscopic study of the stellar populations of low-luminosity active galactic nuclei (LLAGNs). Our main goal is to determine whether the stars that live in the innermost (100 pc scale) regions of these galaxies are in some way related to the emission-line properties, which would imply a link between the stellar population and the ionization mechanism. High signal-to-noise ratio, ground-based long-slit spectra in the 3500-5500 A interval were collected for 60 galaxies: 51 LINERs and LINER/H II transition objects, two starburst galaxies, and seven nonactive galaxies. In this paper, the first of a series, we (1) describe the sample; (2) present the nuclear spectra; (3) characterize the stellar populations of LLAGNs by means of an empirical comparison with normal galaxies; (4) measure a set of spectral indices, including several absorption-line equivalent widths and colors indicative of stellar populations; and (5) correlate the stellar indices with emission-line ratios that may distinguish between possible excitation sources for the gas. Our main findings are as follows: (1) Few LLAGNs have a detectable young (107 yr) starburst component, indicating that very massive stars do not contribute significantly to the optical continuum. In particular, no features due to Wolf-Rayet stars were convincingly detected. (2) High-order Balmer absorption lines of H I (HOBLs), on the other hand, are detected in ~40% of LLAGNs. These features, which are strongest in 108-109 yr intermediate-age stellar populations, are accompanied by diluted metal absorption lines and bluer colors than other objects in the sample. (3) These intermediate-age populations are very common (~50%) in LLAGNs with relatively weak [O I] emission ([O ]/Hα ≤ 0.25) but rare (~10%) in LLAGNs with stronger [O I]. This is intriguing since LLAGNs with weak [O I] have been previously hypothesized to be transition objects in which both an AGN and young stars contribute to the emission-line excitation. Massive stars, if present, are completely outshone by intermediate-age and old stars in the optical. This happens in at least a couple of objects where independent UV spectroscopy detects young starbursts not seen in the optical. (4) Objects with predominantly old stars span the whole range of [O I]/Hα values, but (5) sources with significant young and/or intermediate-age populations are nearly all (~90%) weak-[O I] emitters. These new findings suggest a link between the stellar populations and the gas ionization mechanism. The strong-[O I] objects are most likely true LLAGNs, with stellar processes being insignificant. However, the weak-[O I] objects may comprise two populations, one where the ionization is dominated by stellar processes and another where it is governed by either an AGN or a more even mixture of stellar and AGN processes. Possible stellar sources for the ionization include weak starbursts, supernova remnants, and evolved poststarburst populations. These scenarios are examined and constrained by means of complementary observations and detailed modeling of the stellar populations in forthcoming communications.

Streaming Motions toward the Supermassive Black Hole in NGC 1097

We have used GMOS-IFU and high-resolution HST-ACS observations to map, in unprecedented detail, the gas velocity field and structure within the 0.7 kpc circumnuclear ring of the SBb LINER/Seyfert 1 galaxy NGC 1097. We find clear evidence of radial streaming motions associated with spiral structures leading to the unresolved (<3.5 pc) nucleus, which we interpret as part of the fueling chain by which gas is transported to the nuclear starburst and supermassive black hole.