No Arabic abstract
Why are the nuclei of some galaxies more active than others? If most galaxies harbor a central massive black hole, the main difference is probably in how well it is fueled by its surroundings. We investigate the hypothesis that such a difference can be seen in the detailed circumnuclear morphologies of galaxies using several quantitatively defined features, including bars, isophotal twists, boxy and disky isophotes, and strong non-axisymmetric features in unsharp masked images. These diagnostics are applied to 250 high-resolution images of galaxy centers obtained in the near-infrared with NICMOS on HST. To guard against the influence of possible biases and selection effects, we have carefully matched samples of Seyfert 1, Seyfert 2, LINER, starburst and normal galaxies in their basic properties, taking particular care to ensure that each was observed with a similar average scale (10-15 parsecs per pixel). Several morphological differences among our five different spectroscopic classifications emerge from the analysis. The HII/starburst galaxies show the strongest deviations from smooth elliptical isophotes, while the normal galaxies and LINERS have the least disturbed morphology. The Seyfert 2 galaxies have significantly more twisted isophotes than any other category, and the early-type Seyfert 2s are significantly more disturbed than the early-type Seyfert 1s. The morphological differences between Seyfert 1s and 2s suggest that more is at work than simply the viewing angle of the central engine. They may correspond to different evolutionary stages.
HST is used to study the power sources and the interaction-induced tidal disturbances within the most luminous galaxies in the local universe -- the Ultra-Luminous IR Galaxies (ULIRGs) -- through the use of I-band images with WFPC2 and H-band images with NICMOS. Such images are probing for the first time the fine-scale structures in the strong collision-disturbed morphologies of these rare and exotic galaxies.
We present results of Hubble Space Telescope NICMOS H-band imaging of 73 of most luminous (i.e., log[L_IR/L_0]>11.4) Infrared Galaxies (LIRGs) in the Great Observatories All-sky LIRG Survey (GOALS). This dataset combines multi-wavelength imaging and spectroscopic data from space (Spitzer, HST, GALEX, and Chandra) and ground-based telescopes. In this paper we use the high-resolution near-infrared data to recover nuclear structure that is obscured by dust at optical wavelengths and measure the evolution in this structure along the merger sequence. A large fraction of all galaxies in our sample possess double nuclei (~63%) or show evidence for triple nuclei (~6%). Half of these double nuclei are not visible in the HST B-band images due to dust obscuration. The majority of interacting LIRGs have remaining merger timescales of 0.3 to 1.3 Gyrs, based on the projected nuclear separations and the mass ratio of nuclei. We find that the bulge luminosity surface density increases significantly along the merger sequence (primarily due to a decrease of the bulge radius), while the bulge luminosity shows a small increase towards late merger stages. No significant increase of the bulge Sersic index is found. LIRGs that show no interaction features have on average a significantly larger bulge luminosity, suggesting that non merging LIRGs have larger bulge masses than merging LIRGs. This may be related to the flux limited nature of the sample and the fact that mergers can significantly boost the IR luminosity of otherwise low luminosity galaxies. We find that the projected nuclear separation is significantly smaller for ULIRGs (median value of 1.2 kpc) than for LIRGs (mean value of 6.7 kpc), suggesting that the LIRG phase appears earlier in mergers than the ULIRG phase.
We present results from our analysis of F160W NICMOS Parallel images. These data cover $sim$~9~sq. arcminutes and reach 3$sigma$ depths of H$=$ 24.3 $-$ 25.5 in a $0.6$ diameter aperture with integration times of 2,000 to 13,000 seconds. We derive the first deep H band galaxy counts. The slope of the counts for H$<$ 20 is 0.31, consistent with various K-band measurements from the Keck telescopes. The measured number counts vs. magnitude relation is reasonably well fitted with no-evolution models with a low $Omega$ value. The half-light radii of the galaxies declines steeply with apparent magnitude and reaches the NIC3 resoltion limit at H$=$23.5. Deep ground-based VRI imaging of one NICMOS field has revealed an extremely red galaxy with R$-$H $=$ 6 and H of 18.8. Our analyses of the grism data show that we can reach 3$sigma$ flux limits of of $1times 10^{-16}$ to $2times10^{-17}$ ergs/sec/cm$^2$ for integration times of 2,000 to 21,000~seconds. We have detected a total of 33 emission line galaxies. The comoving number density is $rm sim 2times 10^{-4} Mpc^{-3}$. The detected emission lines are probably H$_alpha$~6563AA. Thus, the derived star formation rates, without extinction correction, are $10 - 163 Modot$ per year for galaxies at redshifts between 0.7 and 1.9.
We discuss the critical importance of black hole mass indicators based on scaling relations in active galaxies. We highlight outstanding uncertainties in these methods and potential paths to substantial progress in the next decade.
We have obtained near-infrared (1.6 micron) images of 11 powerful 3CR radio galaxies at redshifts 0.8 < z < 1.8 using NICMOS on board HST. The high angular resolution permits a detailed study of galaxy morphology in these systems at rest-frame optical wavelengths, where starlight dominates over the extended, aligned UV continuum. The NICMOS morphologies are mostly symmetric and are consistent with dynamically relaxed, elliptical host galaxies dominated by a red, mature stellar population. The aligned structures are sometimes faintly visible, and nuclear point sources may be present in a few cases which manifest the ``unveiled AGN that is obscured from view at optical wavelengths. Our observations are consistent with the hypothesis that the host galaxies of z ~ 1-2 radio galaxies are similar to modern-day gE galaxies. Their sizes are typical of gE galaxies but smaller than present-day cD and brightest cluster galaxies, and their surface brightnesses are higher, as expected given simple luminosity evolution.