ArticlePDF Available

Hbeta spectra of high-redshift QSOs: Eigenvector 1 at high luminosities

Authors:

Abstract

We present VLT-ISAAC spectra of unprecedented S/N and resolution for the H/3 region in a sample of n = 17 quasars in the range z = 0.8-2.5. The data represent our first attempt to test source occupation and line properties for the Eigenvector I parameter space that was defined using a.sample of sources with z < 0.8. We find no strong luminosity/redshift dependent effects with the possible exception of an increase in the minimum broad line profile width from 1000 to 3000 km s(-1) between M(B) approximate to -20 and -28.
The Interplay among Black Holes, Stars and ISM
in Galactic Nuclei
ProceedingsIAU S ympo siumN o.222, 2004
T. Storchi-Bergmann, L. C. Ho &H.R.Schmitt, eds.
c
2004 International Astronomical Union
DOI: 10.1017/S1743921304003199
Hβspectra ofhigh-redshift QSOs:
Eigenvector 1at high luminosities
G. M. Stirpe1,J.W.Sulentic
2,P.Marziani3,
R. Zamanov4,M.Calvani
3and V.Braito5
1INAF Osservatorio Astronomico diBologna, Via Ranzani 1, 40127 Bologna, Italy
2Dept. ofPhysics and Astronomy,Univ. of Alabama, Tuscaloosa, AL35487, USA
3INAF Osservatorio Astronomico diPadova, Vicolo dell’ Osservatorio 5, 35122 Padova, Italy
4Astrophysics ResearchInst., LiverpoolJohn Moores Univ., TwelveQuaysHouse, Egerton
Wharf, Birkenhead CH41 1LD, UK
5INAF Osservatorio Astronomico diBrera, Via Brera 28, 20121 Milano, Italy
Abstract. WepresentVLT–ISAACspectra of unprecedented S/N and resolution for the Hβ
region in asample ofn=17 quasars inthe range z=0.8–2.5. The data representour rst
attempt totest source occupation and line properties for the Eigenvector 1parameter space
that wasdefined using asample ofsources with z<0.8. Wefind no strong luminosity/redshift
dependenteffects with the possible exception of an increase in the minimum broad line profile
width from 1000 to 3000 km s1between MB≈−20 and 28.
Wehaveobtained high S/N spectra of the Hβregion of asample of 17 HE quasars
with intermediate redshift (0.85 z2.5) and high luminosity(26 MB30).
Forthe first time wehaveasample of NIR spectra with resolution and S/N compa-
rable tothose obtained for 200+ lower-redshift Type1AGN (Marziani etal. 2003
(M03)). Spectra (Fig. 1) were obtained between 2001 November and 2002 February using
ESO/VLT1+ISAACproviding aresolution of 300 km s1in all bands (Z, sZ, J,H).
The observations representour first attempt totest the robustness ofthe Eigenvector 1
(E1) parameter space previously defined using lower redshift sources (Boroson &Green
1992; Sulentic et al. 2000). Our goals include testing: 1) domain space occupation in the
optical E1 plane and 2) the apparentluminosityindependence of all E1 properties.
The VLTspectra enable ustomeasure both E1 optical parameters: FWHM(Hβ)and
the broad-line equivalentwidth ratio W(FeIIopt )/W(Hβ)=R
Fe ,where W(FeIIopt )mea-
sures the λ4570 blend. Broad Hβ,andthenarrow[OIII]λλ5007,4959, were isolated by
subtracting appropriately scaled FeIIemission from the spectra using atemplate based
on IZw1. FeII emission properties shownodifferences from our lower-redshift sample
(M03) and the FeII template works equally well at these high redshifts.
Wefind no significantdifferences in source occupation for the optical plane ofE1 as
defined previously (Fig. 2a). There maybeaslight tendency for higher redshift source
to displace towards the upper rightbut this requires verification with amuchlarger
sample. The results are consistentwith our previous suggestion that E1 occupation is
driven neither bysource luminositynor BH mass but rather bythe Eddington ratio
convolved with source orientation to the line ofsight.
Figure 2b showspossible evidence for asystematic increase in minimum observed
FWHM(Hβ)with source luminosity.Ageneral increase in FWHM(Hβ)with luminosity
is not certain because radio-loud sources are over-represented inthe higher luminosity
Based onobservations collected at the European Southern Observatory,Chile. Proposal
ref.: ESO 68.B–0364(A)
539
540 G. M. Stirpeet al.
Figure 1. Twoexamples ofthe 17 spectra ofHEQSOs obtained at the ESO/VLT
Figure 2. Left: the main E1 diagram, whichshowsthe relation between broad Hβline width
and relativeFeIIintensity.The HE QSOs are represented bythe lled circles with error bars.
There isnoevidence that their distribution isdifferentfrom that ofthe open circles (sources
from M03), despite the differentluminositydistribution. Right: the broad Hβline width vs.
luminosity.Triangles representHEQSOs, circles are AGN from M03. Symbols are filled for
radio-quiet sources, emptyfor radio-loud. The curveis the estimated lowerlimit to the FWHM
as afunction of luminosity.
part of our sample and they are known to showsystematically larger FWHM(Hβ). The
increase in minimum FWHM with Lmightbe expected if we:
assume virialized motions inthe Broad Line Region (BLR)
use the RBLR Lαrelation found byKaspi etal. (2000), with α=0.7
assume that low-redshift NLS1 havean Eddington ratio close to 1
By combining these relations, weobtain the lowerboundary to the broad-line widths
showninFig. 2b.
Using L(5100˚
A) (rest wavelength) and the FWHM(Hβ)asdescribed byVestergaard
(2002), wecan estimate the mass ofthe central blackhole, assuming that the BLR-
size vs. luminosityrelation found byKaspi etal. (2000) can be extrapolated tohigh
luminosities and redshifts. Weused the formula derived byMcLure &Jarvis (2002). The
mass estimates yield large values, close to or reaching 1010 M,for the most luminous
sources. Estimates ofthe Eddington ratios indicate that most QSOs are radiating near
0.3L
Edd.
References
Boroson, T.A., &Green, R.F.1992, ApJS, 80, 109
Kaspi, S., etal. 2000, ApJ, 533, 631
Marziani, P., et al. 2003, ApJS, 145, 99 (M03)
McLure, R.J., &Jarvis, M.J.2002, MNRAS, 337, 109
Sulentic, J. W., Zwitter, T., Marziani, P., &Dultzin-Hacyan, D. 2000, ApJ, 536, L5
Vestergaard, M. 2002, ApJ, 571, 733
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
We identify a correlation space involving optical and UV emission line parameters as well as the soft X-ray spectral index that provides optimal discrimination between all principal classes of active galactic nuclei. Most of the sources in our three high quality data samples show a strong intercorrelation with narrow line Seyfert 1 (NLSy1) galaxies and steep spectrum radio galaxies occupying opposite extrema in the space. NLSy1 sources show a clear continuity with broader line sources indicating that they are not a disjoint class of AGN as is sometimes suggested. We interpret the principal intercorrelation in the parameter space as being driven by AGN luminosity to black hole mass ratio (L/M prop. to the Eddington ratio). Source orientation no doubt also plays an important role but it is not yet clear whether FWHM H beta or CIV 1549 line shift is the better indicator. We tentatively identify two RQ populations: an almost pure RQ population A with FWHM<=4000 and population B which occupies the same parameter domain as the flat spectrum RL sources. A possible interpretation sees Pop. A/NLSy1 as lower mass/high accretion rate sources and Pop. B/RL sources the opposite.
Article
The present study obtains spectra covering the 4300-5700 wavelength region for all 87 QSOs in the BQS catalog having redshifts less than 0.5. A novel empirical technique which allows the measurement and subtraction of the many Fe II lines in this region is applied to these spectra. Measurements of the strengths of H-beta, forbidden O III 5007, and He 4686, and a 4D parameterization of the H-beta profile are combined with optical, radio, and X-ray continuum information from the literature to elucidate how these properties are related. It is concluded that the dominant source of variation in the observed properties of low-redshift QSOs is a physical parameter which balances Fe II excitation against the illumination of the narrow line region. It is argued that this property is not driven by external orientation, i.e., the viewing angle. A comparison of subsamples defined by their radio properties suggest that radio-quiet and radio-loud objects cannot be parallel sequences because there is a large deficit of radio-quiet analogs to the steep-spectrum QSOs.
Article
(abridged) We have spectrophotometrically monitored a well-defined sample of 28 Palomar-Green quasars in order to obtain measurements of their BLRs and to investigate the relationships between quasar luminosity, central black hole mass, and broad emission line region (BLR) size in active galactic nuclei (AGN). Spectrophotometry was obtained every 1-4 months for 7.5 years, yielding 20-70 observing epochs per object. Both the continuum and emission line fluxes of all of the quasars were observed to change during the duration of the observing program. Seventeen of the 28 objects were observed with adequate sampling to search for correlated variations between the Balmer emission lines and the continuum flux. For each of these 17 objects, a significant correlation was observed, with the Balmer line variations lagging those of the continuum by \~100 days (rest frame). Our work increases the available luminosity range for studying the size-mass-luminosity relations in AGN by two orders of magnitude and doubles the number of objects suitable for such studies. Combining our results with comparable published data available for Seyfert 1 galaxies, we find the BLR size scales with the rest-frame 5100 A luminosity as L^{0.70+-0.03}. This determination of the scaling of the size of the BLR as a function of luminosity is significantly different from those previously published, and suggests that the effective ionization parameter in AGN may be a decreasing function of luminosity. We are also able to constrain, subject to our assumption that gravity dominates the motions of the BLR gas, the scaling relationship between the mass of the central black holes and the luminosity in AGN: M \propto L^{0.5+-0.1}. This is inconsistent with all AGN having optical luminosity that is a constant fraction of the Eddington luminosity. Comment: 21 pages, 8 figures included, LaTeX emulateapj.sty, accepted for publication in the Astrophysical Journal. Resubmitting to Correct few confusing sentences in subsection 5.2
Article
A new technique is presented for determining the black-hole masses of high-redshift quasars from optical spectroscopy. The new method utilizes the full-width half maximum (FWHM) of the low-ionization MgII emission line and the correlation between broad-line region (BLR) radius and continuum luminosity at 3000 Angstroms. Using archival UV spectra it is found that the correlation between BLR radius and 3000 Angstrom luminosity is tighter than the established correlation with 5100 Angstrom luminosity. Furthermore, it is found that the correlation between BLR radius and 3000 Ang luminosity is consistent with a relation of the form $R_{BLR} \propto lambda L_{lambda}^{0.5}$, as expected for a constant ionization parameter. Using a sample of objects with broad-line radii determined from reverberation mapping it is shown that the FWHM of MgII and H beta are consistent with following an exact one-to-one relation, as expected if both H beta and MgII are emitted at the same radius from the central ionizing source. The resulting virial black-hole mass estimator based on rest-frame UV observables is shown to reproduce black-hole mass measurements based on reverberation mapping to within a factor of 2.5 (1 sigma). Finally, the new UV black-hole mass estimator is shown to produce identical results to the established optical (H beta) estimator when applied to 128 intermediate-redshift (0.3<z<0.9) quasars drawn from the Large Bright Quasar Survey and the radio-selected Molonglo quasar sample. We therefore conclude that the new UV virial black-hole mass estimator can be reliably used to estimate the black-hole masses of quasars from z~0.25 through to the peak epoch of quasar activity at z~2.5 via optical spectroscopy alone.
Article
An empirical relationship, of particular interest for studies of high redshift active galactic nuclei (AGNs) and quasars, between the masses of their central black-holes and rest-frame ultraviolet (UV) parameters measured in single-epoch AGN spectra is presented. This relationship is calibrated to recently measured reverberation masses of low-redshift AGNs and quasars. An empirical relationship between single-epoch rest-frame optical spectrophotometric measurements and the central masses is also presented. The UV relationship allows reasonable estimates of the central masses to be made of high-redshift AGNs and quasars for which these masses cannot be directly or easily measured by the techniques applicable to the lower luminosity, nearby AGNs. The central mass obtained by this method can be estimated to within a factor of ~3 for most objects. This is reasonable given the intrinsic uncertainty of a factor less than 2 in the primary methods used to measure the central masses of nearby inactive and active galaxies, namely resolved gas and stellar kinematics in the underlying host galaxy and reverberation-mapping techniques. The UV relationship holds good potential for being a powerful tool to study black-hole demographics at high redshift as well as to statistically study the fundamental properties of AGNs. The broad line region size - luminosity relationship is key to the calibrations presented here. The fact that its intrinsic scatter is also the main source of uncertainty in the calibrations stresses the need for better observational constraints to be placed on this relationship. The empirically calibrated relationships presented here will be applied to quasar samples in forthcoming work. Comment: 14 pages, plus 7 tables and 7 figures. To appear in ApJ, June 1, 2002
  • P Marziani
Marziani, P., et al. 2003, ApJS, 145, 99 (M03) McLure, R. J., & Jarvis, M. J. 2002, MNRAS, 337, 109
  • S Kaspi
Kaspi, S., et al. 2000, ApJ, 533, 631
  • T A Boroson
  • R F Green
Boroson, T. A., & Green, R. F. 1992, ApJS, 80, 109
  • P Marziani
Marziani, P., et al. 2003, ApJS, 145, 99 (M03)
  • R J Mclure
  • M J Jarvis
McLure, R. J., & Jarvis, M. J. 2002, MNRAS, 337, 109