Starburst Galaxies and AGN

H. Bushouse, L. Colina, R. Lucas (ST ScI), and K. Borne (Hughes STX) are obtaining a WFPC2 snapshot survey of the ultraluminous IRAS galaxy sample. The survey images are being used to study the cores and starburst regions of the galaxies, with particular emphasis on searching for young stellar clusters, as have already been found in HST observations of similar objects. Nearly all of the galaxies in the survey show evidence from groundbased images for a very recent tidal interaction, thus a search is being made for evidence of the interaction progenitors by looking for multiple cores near the center of each system. These images also serve as a baseline for future HST observations designed to catch the type II supernovae that are expected to be produced at a prodigious rate in these strongly starbursting systems.

D. Calzetti, in collaboration with M. Giavalisco (OCIW) and A. P. Koratkar (ST ScI), analyzed IUE spectra of a sample of local, low-metallicity starburst galaxies to investigate the correlation between Ly- emission, UV extinction, and the metal content. They find evidence that the transport of the Lyman- photons is primarily controlled by the geometry of the highly inhomogeneous galaxy ISM, rather than by the amount of dust contained in the galaxy. Primeval galaxies with characteristics similar to those of the present low-metallicity galaxies would have been unnoticed in typical narrow-band surveys tuned to detect high-redshift Lyman- emission. The spectral classification of high-redshift galaxies requires, for the zeroth-order comparison, UV and optical spectral energy distributions (SEDs) of present-day galaxies.

Calzetti, collaborating with A. L. Kinney, R. C. Bohlin, K. McQuade (ST ScI), T. Storchi-Bergmann, and H. Schmitt (UFRGS, Brazil), derived multiwavelength SEDs using aperture-matched IUE and optical spectra of local galaxies. For quiescent galaxies, the SEDs are separated according to the morphological classification of the galaxy; for starburst galaxies, according to color excess. The reddening-corrected SEDs of starburst galaxies have been used by Calzetti to investigate the star formation history of the starburst region. A large contribution from intermediate age population (age --500 Myr) is found, suggesting that the starburst activity has proceeded through various episodes over a long timescale. In the same galaxies, there is no indication for a high-mass-truncated or a low-mass-deficient stellar Initial Mass Function.

E. Colbert is continuing his thesis work with S. Baum (ST ScI) and S. Veilleux (U. MD) on large-scale (galactic) outflows in edge-on Seyfert galaxies. Optical and radio surveys have been completed for galaxies selected from a distance-limited sample of 22 objects. There is evidence for large-scale outflows in about 50% of the objects surveyed. Evidence includes extended H nebulae along the minor axis of the galaxy disk, luminous H nebulae located --10 kpc out of the disk, double-peaked H emission lines suggesting the presence of a wind blown bubble, and radiocontinuum emission extending one or more kpc from the nucleus. The three best candidates (NGC 2992, NGC 5506, and NGC 4438) show excellent evidence for an outflow in both the radio and optical regimes, and have extended soft X-ray nebulae which are roughly co-spatial. Follow-up work is being done to study in more detail the properties of the large-scale outflows in these three galaxies. The physics of these flows is fascinating. A plausible explanation of their origin is that they are larger scale extensions of sub-kpc nuclear outflows (e.g., jets) that are known to be present in Seyfert galaxies. Starburst driven superwinds also remain a plausible explanation in some cases.

Colbert, S. Baum and C. O'Dea (ST ScI) have obtained very deep radio continuum observations of the Seyfert/starburst hybrid galaxy Mrk 231 in order to study properties of the radio outflow. This galaxy was already known to have a galactic outflow but its origin (starburst- or AGN-driven) is not known. Data were taken with the VLA at 6 and 20 cm in all array configurations. Preliminary results show the southern extended radio feature to be highly polarized (30--40%), steep-spectrum emission which is likely to be produced by a large-scale jet from the nucleus.

Colbert has won an NRC Research Associateship Award to do postdoctoral research at NASA/GSFC on X-ray imaging and spectroscopy of Seyfert galaxies.

M. Goad and his collaborators have an ongoing project to monitor the variable broad absorption lines in NGC 3516 (PI---A. Koratkar). They have four epochs of HST/FOS data so far and are awaiting the last observation in December. Meanwhile, Goad continued analysis of the archival NGC 5548 data. They have determined robust constraints on the physical conditions within the broad line region gas for this object, using the technique of Simulated Annealing.

Similar constraints are to be placed on several other objects which have moderately high resolution spectroscopic data available in the HST archive. This work was started with the aid of a very able summer student (M. Wang), who spent 10 weeks here as part of the summer student programme.

A. Koratkar's research has made use of UV variability, and UV spectropolarimetry to study AGN.

The nature of the powerful central engine in Active Galactic Nuclei (AGN) is related to the structure of the circumnuclear gas in these objects. Clues to the nature of activity can be found by studying the characteristic broad emission lines observed in AGN spectra, and comparing statistical properties of AGN at various wavelengths and luminosities.

One of the observational facts that favors the thermal disk models in AGN is the UV emission component of the quasar continuum. In addition to the above signature, there are other observational signatures of thermal accretion disks, such as discontinuity of the continuum at the Lyman edge both in polarized light and in total intensity. Koratkar and collaborators have observed with HST many AGN which are accretion disk candidates. The FOS data show polarized flux in emission contrary to the model predictions! (Koratkar, Antonnuci, Bushouse & Kinney 1995; Koratkar et al. 1997, in prep). They are presently developing models to explain the observed polarized flux in emission.

Along with strong UV emission lines, some AGN display intrinsic UV absorption lines. The origin and physical properties of the absorbing gas are not well understood. They have recently completed a study of the Seyfert 1 galaxy, NGC 3516, (Koratkar et al. 1996 in press) in which they have established that NGC 3516 has the most variable absorption lines. Dissipation of the variable absorption line gas implies a dynamic absorber, possibly the ionized surface layer of a molecular torus.

M. Livio, in collaboration with Pringle (Cambridge), examined models for the formation of double-peaked emission lines in AGN and proposed critical observations that can distinguish between the different models.

Livio, collaborating with Baum, O'Dea, de Koff, Sparks, Hayes, Golombek (all ST ScI), discovered obscuration rings in Hercules A, and examined their implications for energy transport in radio galaxies.

In collaboration with Borne, Bushouse, and Colina, R. Lucas is investigating the nature of IRAS Ultraluminous Galaxies using the HST WFPC2 in a SNAPshot program to obtain I-band images of several IRAS ULG samples, including the remainder of galaxies not yet imaged by HST from the samples of Sanders, and also of Melnick and Mirabel. Other subsamples in this program include those from Kim et al. and Clements, as well as the QDOT and NGW surveys of Lawrence et al. At the time of this writing, it is still early in the observing Cycle, but new, previously unresolved, features have already been discovered in a number of the galaxies, and the majority seem to be part of apparently interacting pairs or groups, which is consistent with the view that these highly luminous systems are triggered and/or fueled by interactions and mergers. The SNAP allocation is for a sample of up to 160 galaxies. To date, images of about 45 galaxies have been obtained, and a paper describing the results seen in these initial observations is in preparation.

Lucas continued work on HST WFPC2 images of the Cartwheel Ring Galaxy. Some new results are described in Struck, Appleton, Borne, and Lucas (accepted for publication in The Astronomical Journal) in which the dust lanes and cometary structures in the inner disk of the Cartwheel are described and analyzed. The previously undetected cometary structures may be explained in several ways, all of which raise still more questions. They may be supersonic orbit crossings of the inner ring by massive clouds, or cloud complexes radially mixed in the disk as a result of the collision; they may be formed from dense clouds raining down on the disk from a bridge which connects the Cartwheel to the intruder galaxy; or they may be disk clouds traversing a tenuous high-velocity accretion stream. More work is needed to better understand this very complex and fascinating galaxy, and the history of the interactions it shares with its companions. Lucas also presented results of these observations at the workshop on Interacting Galaxies in Pairs, Groups, and Clusters in Sant'Agata, Italy, in September 1995, and some initial results will be described in the published proceedings from this conference. Initial results are also described in Borne et al. in the proceedings of the Paris conference on Science With HST: II, which took place in December 1995. More papers are currently in preparation on these initial HST observations of the Cartwheel Galaxy and two of its three companions. (NOTE: It has recently been found by Higdon 1995 using the VLA that the third, more distant companion seems to be connected to the Cartwheel by a plume or bridge of H I gas, so we hope to follow up this result also if possible.)

J. E. Pesce, Urry, Scarpa, Falomo (Obs. Padua), and Treves (SISSA, Trieste) are working on a series of projects to study the host galaxies of BL Lac objects. The first set of HST data has been analyzed and more work is in progress. In particular, they continue to receive HST WFPC2 images from their Cycle 6 Snapshot project. See Urry, below, for a description of the results obtained.

Pesce, with Urry, Scarpa, Falomo, Paske (Pomona, summer student), and Kim (Towson State, summer student) is completing a project using the digitized sky survey images to study the extended, Mpc-scale environments of low redshift () BL Lac objects and radio galaxies (FR type I and II). This is the first study using a complete sample of FR I and II radio galaxies, and preliminary results indicate no great difference between the environments of FRI/II radio galaxies and BL Lac objects (papers are in preparation).

The HST WFPC2 images of about 50 BL Lac fields (both GO targets and Snapshot targets) are being analyzed by Pesce to determine the environmental properties of these objects. In addition to number counts of galaxies around the BL Lac targets, the high resolution of the HST allows morphological classification of these galaxies as well. A paper is in preparation.

Pesce, Pian, and Urry (along with a substantial fraction of the astronomical community!) have finished a series of papers describing the large multiwavelength campaign to observe PKS 2155--304 in May 1994 (see also below). The ground-based data show a correlation between the optical and ultraviolet; the optical flux and polarized flux seem to be well correlated with a flare seen in the ultraviolet and harder energies. Pesce continues to coordinate ground-based observations for various multiwavelength campaigns, in particular observations with GRO, collaborating with Hartman (GSFC).

E. Pian is investigating the continuum variability of blazars with the aim of clarifying the emission processes in these luminous sources and the mechanisms which convert matter into radiated energy. Multiwavelength monitorings of selected sources have led to a significant improvement in the interpretation of the emission and spectral variations. Particularly, the observing campaign of the BL Lac class prototype PKS 2155--304 in May 1994 from radio to hard X-ray frequencies detected a prominent flare of wavelength-dependent profile amplitude. This characteristic, together with the observation of a temporal lag, point to the evolution of a synchrotron flare within a shock or disturbance propagating down an inhomogeneous jet. A recent EGRET campaign on the gamma-ray bright blazar 3C 279, coordinated with observations with space- and ground-based facilities at lower frequencies, enabled mapping the development of an outburst at high energies, which seems consistent with an inverse Compton flare produced by scattering of relativistic electrons off high ionization emission lines (likely Lyman-) photons.

Using ground spectrophotometric observations, R. Scarpa is studying the optical spectrum of a large sample of 73 blazars, investigating differences and similarities between radio-loud highly polarized quasars (HPQ) and BL Lacertae objects (BLL). This study is carried out as part of a larger collaboration with R. Falomo (Padova Obs.). The analysis of line luminosities shows that a continuum exists from BL Lacs to HPQ, indicating that, in spite of the classification criteria, they constitute a single class of object. This result puts strong constraints on the theoretical models for blazars.

In collaboration with C. M. Urry, A. Treves (SISSA, Trieste), R. Falomo, and J. E. Pesce, Scarpa is carrying out a detailed study of BL Lac host galaxies. This research is based on a large amount of HST WFPC2 images. Results will permit better determination of the average morphological properties of the hosts, to be compared with those of the proposed parent population.

C. Simpson and collaborators finally obtained data for HST GO-5411, emission-line imaging of 8 Seyfert 2 galaxies using the Linear Ramp Filters, after a long delay. Papers based on the two most dramatic datasets are in press, and a paper is in preparation on the remaining objects.

The origin of the near-infrared [Fe II] emission in Seyfert galaxies has been a matter of debate for some time, and a thorough analysis of literature data and our own data (obtained with CGS4 on UKIRT) came out strongly in favour of photoionization; a paper is in press for MNRAS.

In addition to projects which were finally completed and the results published, a number of new projects are underway. A combination of infrared data (from ESO and UKIRT) and X-ray spectroscopy (from the ASCA satellite) has revealed an extremely large gas-to-dust ratio (an order of magnitude greater than the canonical Galactic value) along the line-of-sight to the Seyfert nucleus of NGC 3281. Thermal infrared imaging of radio galaxies at (from UKIRT) has indicated the existence of non-stellar contamination in the near-infrared Hubble diagram. A similar project (from the ANU 2.3m at Siding Spring) is turning up insights into the nature of the obscuring torus in low-redshift radio galaxies. Broad-band imaging of Seyfert galaxies (again from UKIRT) revealed the presence of a kpc-scale dust disk in Mrk 348, and follow-up Fabry-Perot imaging in a line of molecular hydrogen is scheduled for early next year.

H. Stockman continues to collaborate with Schmidt and B. Jannuzi (NOAO) on imaging polarimetry and spectropolarimetry of high redshift objects. The data for most of these objects appears consistent with obscured/lensed AGNs, rather than extended star forming regions. A good example is the radio galaxy, 3C 256, which displays extended polarized radiation (P(0.19 m rest = 16.4%) which is consistent with much of the light being scattered nuclear radiation.

A. Suchkov and Berman, collaborating with T. Heckman, have continued hydrodynamical simulations of galactic winds and modeling of the wind induced X-ray emission of starburst galaxies. They concluded that, in the case of M82, the galactic wind must be a strongly collimated mass-loading wind whose mass flux is 3--6 times higher than the mass deposition rate due to supernova ejecta, with almost all mass loading taking place in the very center of the galaxy (Suchkov et al. 1996). The X-ray emission was found to be due exclusively to the wind material: any substantial contribution from the wind-shocked halo gas appears to be incompatible with the observed data. Preliminary results from a model applicable to Arp 220 have shown that a putative superwind responsible for extended X-ray emission in this galaxy should also be a collimated wind with central mass-loading (Suchkov et al., in preparation). The modeling suggests that, unlike the case of M82, the X-ray emission in this galaxy has a two-component structure, with an inner component (within --4 kpc) being due to the wind material and an outer component (prominent beyond kpc) associated with the shocked material of the wind and the wind-evaporated gas of an extended cloudy component of the galactic ISM. The intermediate region appears to be void of emitting material at all.

C. M. Urry and Pesce continue their HST WFPC2 imaging survey of BL Lac objects, investigating the properties of their host galaxies and environments. The first results, now in press (Falomo et al. 1997), describe the detection of two host galaxies at and and give interesting limits on the host luminosity for a BL Lac at . Additional data for several dozen BL Lacs is now being analyzed, and will be used to understand the nature of BL Lac objects, their connection to radio galaxies, their evolution from to z=1, and the role of interactions in triggering nuclear activity.

Urry has continued multiwavelength monitoring of blazars, with Maraschi (U. Milan), Pesce, Pian, Wehrle (IPAC), and collaborators, carrying out radio-through-gamma-ray campaigns on the superluminal quasar 3C 279 and the UV-bright BL Lac object PKS 2155--304. For 3C 279, coordinated variations at gamma-ray through optical/UV wavelengths imply associated changes in both Compton-scattering electrons and seed UV photons, as may happen if the broad-line region is photoionized significantly by the jet radiation. A re-analysis of archival IUE spectra (Koratkar et al. 1996) suggests that this may well be the case.

For PKS 2155--304, two epochs of monitoring have demonstrated a close connection between UV and X-ray emission, but dramatic differences between the two epochs argue for multiple flare mechanisms. In the papers describing the second-epoch X-ray, UV, and optical data recently been submitted for publication, a variety of possible flare mechanisms are discussed, including acceleration and decay events intrinsic to a relativistic jet, microlensing events in intervening galaxy cores, and changes in Doppler beaming.

The relativistic jet paradigm for blazars implies that radio galaxies are fundamentally the same objects. Urry has written a review paper for PASP with P. Padovani describing the unification of blazars with radio galaxies and quasars (BL Lac objects with low-luminosity radio galaxies and quasars with high-luminosity radio galaxies). The review describes the kinds of anisotropic emission that lead to aspect-dependent AGN classification and details the aspect-independent properties that can be used to test these unified schemes. Urry and Padovani then discuss the statistical basis for this kind of unification, with quantitative analysis of both prevailing schemes. They investigate in some detail the nature of BL Lac objects, particularly the distinctions between X-ray-selected and radio-selected BL Lacs, and the connection to blazar-like quasars. Finally, they discuss the problems and complications for unified schemes and outline the ten most important questions for future research in this area.

Urry continues to study the X-ray properties of AGN with ROSAT and ASCA. She and Sambruna (NASA/GSFC) have completed a ROSAT survey of radio-selected BL Lac objects which suggests the onset of a Compton-scattered X-gamma-ray component typically occurs above 1--2 keV. Sambruna and she have completed a related paper demonstrating that the multiwavelength properties of BL Lac objects and flat-spectrum radio quasars show systematic trends with luminosity and line emission properties. This study rules out a simple orientation effect as the sole distinction between radio-selected and X-ray-selected BL Lac objects. With Sambruna, the ASCA spectrum of the high-redshift blazar 0528+134 has been used to constrain the activity of the gamma-ray component, and ASCA, ROSAT, and BBXRT spectra of the BL Lac object 1426+428 have been used to constrain the presence of hot, X-ray-absorbing gas.

M. Voit has begun to evaluate the potential of large infrared interferometers for studying active galaxies at high resolution. Instruments like the VLTI and Keck interferometers will be able to resolve the 10-micron thermal emission from AGN and will be excellent for studying the interface between the accretion-dominated nucleus and the ambient ISM of the host galaxy (Voit 1996b, 1996c).