sign in sign up
<<
Home , Active galactic nucleus

GRAVITY Science DOI: 10.18727/0722-6691/5166

Spatially Resolving the Broad Emission Line Region

GRAVITY Collaboration Juan Pablo Gil 8 Anne-Lise Maire 23, 3 Stefan Gillessen1 Leander Mehrgan 8 Roberto Abuter 8 Frédéric Gonté 8 Antoine Mérand 8 Matteo Accardo 8 Paulo Gordo 6 Florentin Millour 37 Tobias Adler 3 Damien Gratadour 2 Paul Mollière 3 António Amorim 6 Alexandra Greenbaum 40 Thibaut Moulin 5 Narsireddy Anugu7,29,30 Rebekka Grellmann 4 André Müller 3 Gerardo Ávila 8 Ulrich Grözinger 3 Eric Müller 8, 3 Michi Bauböck1 Patricia Guajardo 8 Friedrich Müller 3 Myriam Benisty 5,12 Sylvain Guieu 5 Hagai Netzer 32 Jean-Philippe Berger 5 Maryam Habibi1 Udo Neumann 3 Joachim M. Bestenlehner 22, 3 Pierre Haguenauer 8 Mathias Nowak 2 Hervé Beust 5 Oliver Hans1 Sylvain Oberti 8 Nicolas Blind9 Xavier Haubois 8 Thomas Ott 1 Mickaël Bonnefoy 5 Marcus Haug 8 Laurent Pallanca 8 Henri Bonnet 8 Frank Haußmann1 Johana Panduro 3 Pierre Bourget 8 Thomas Henning 3 Luca Pasquini 8 Jérôme Bouvier 5 Stefan Hippler 3 Thibaut Paumard 2 Wolfgang Brandner 3 Sebastian F. Hönig 27 Isabelle Percheron 8 Roland Brast 8 Matthew Horrobin 4 Karine Perraut 5 Alexander Buron1 Armin Huber 3 Guy Perrin 2 Leonard Burtscher 14 Zoltan Hubert 5 Bradley M. Peterson 24,25, 26 Faustine Cantalloube 3 Norbert Hubin 8 Pierre-Olivier Petrucci 5 Alessio Caratti o Garatti 16, 3 Christian A. Hummel 8 Andreas Pflüger 1 Paola Caselli 1 Gerd Jakob 8 Oliver Pfuhl 8 Frédéric Cassaing 10 Annemieke Janssen 36 Than Phan Duc 8 Frédéric Chapron 2 Alejandra Jimenez Rosales1 Jaime E. Pineda1 Benjamin Charnay 2 Lieselotte Jochum 8 Philipp M. Plewa1 Élodie Choquet 37 Laurent Jocou 5 Dan Popovic 8 Yann Clénet 2 Jens Kammerer 8, 41 Jörg-Uwe Pott 3 Claude Collin 2 Martina Karl 20, 21 Almudena Prieto 39 Vincent Coudé du Foresto 2 Andreas Kaufer 8 Laurent Pueyo 11 Ric Davies1 Stefan Kellner1 Sebastian Rabien1 Casey Deen1 Sarah Kendrew 11, 3 Andrés Ramírez 8 Françoise Delplancke-Ströbele 8 Lothar Kern8 José Ricardo Ramos 3 Roderick Dembet 8 Pierre Kervella 2 Christian Rau1 Frédéric Derie 8 Mario Kiekebusch 8 Tom Ray 16 Willem-Jan de Wit 8 Makoto Kishimoto 31 Miguel Riquelme 8 Jason Dexter 1 Lucia Klarmann 3 Gustavo Rodríguez-Coira 2 Tim de Zeeuw 1,14 Ralf Klein 3 Ralf-Rainer Rohloff 3 Catherine Dougados 5 Rainer Köhler 3 Daniel Rouan 2 Guillaume Dubus 5 Yitping Kok1 Gérard Rousset 2 Gilles Duvert 5 Johann Kolb 8 Joel Sanchez-Bermudez 3, 17 Monica Ebert 3 Maria Koutoulaki 16,19,3,8 Marc Schartmann 1,33,34 Andreas Eckart 4,13 Martin Kulas 3 Silvia Scheithauer 3 Frank Eisenhauer 1 Lucas Labadie 4 Markus Schöller 8 Michael Esselborn 8 Sylvestre Lacour 2, 8 Nicolas Schuhler 8 Fabio Eupen 4 Anne-Marie Lagrange 5 Dominique Segura-Cox1 Pierre Fédou 2 Vincent Lapeyrère 2 Jinyi Shangguan1 Miguel C. Ferreira 6 Werner Laun 3 Thomas T. Shimizu1 Gert Finger 8 Bernard Lazareff 5 Jason Spyromilio 8 Natascha M. Förster Schreiber 1 Jean-Baptiste Le Bouquin 5 Amiel Sternberg1, 32 Feng Gao1 Pierre Léna 2 Matthias Raphael Stock 21 César Enrique García Dabó 8 Rainer Lenzen 3 Odele Straub 1, 2 Rebeca Garcia Lopez 16, 3 Samuel Lévêque 8 Christian Straubmeier 4 Paulo J. V. Garcia 7 Chien-Cheng Lin 3,18 Eckhard Sturm1 Éric Gendron 2 Magdalena Lippa1 Marcos Suárez Valles 8 Reinhard Genzel 1,15 Dieter Lutz1 Linda J. Tacconi1 Ortwin Gerhard1 Yves Magnard 5 Wing-Fai Thi1

20 The Messenger 178 – Quarter 4 | 2019 Konrad R. W. Tristram 8 15 Department of Physics, Le Conte Hall, 40 University of Michigan Department of Javier J. Valenzuela 8 University of California, Berkeley, USA , Ann Arbor, USA Roy van Boekel 3 16 Dublin Institute for Advanced Studies, 41 Research School of Astronomy & Ewine F. van Dishoeck14 Dublin, Ireland , Australian National Pierre Vermot 2 17 Instituto de Astronomía, Universidad ­University, Canberra, Australia Frédéric Vincent 2 Nacional Autónoma de México, Ciudad Sebastiano von Fellenberg1 de México, Mexico Idel Waisberg1 18 Institute for Astronomy, University of The angular resolution of the Very Large Jason J. Wang 28 Hawai’i, Honolulu, USA Telescope Interferometer (VLTI) and Imke Wank 4 19 School of Physics, University College the excellent sensitivity of GRAVITY Johannes Weber 1 Dublin, Ireland have led to the first detection of spa- Gerd Weigelt 13 20 Max Institute for Physics, tially resolved kinematics of high veloc- Felix Widmann1 Munich, Germany ity atomic gas near an accreting super- Ekkehard Wieprecht1 21 TUM Department of Physics, Technical massive , revealing rotation Michael Wiest 4 University of Munich, Garching, on sub- scales in the quasar Erich Wiezorrek1 Germany at a distance of 550 Mpc. The Markus Wittkowski 8 22 Department of Physics and Astronomy, observations can be explained as the Julien Woillez 8 University of Sheffield, UK result of circular orbits in a thick disc Burkhard Wolff 8 23  Institute, Liège, Belgium configuration around a 300 million solar Pengqian Yang 3, 35 24 Department of Astronomy, The Ohio mass black hole. Within an ongoing Senol Yazici 1, 4 State University, Columbus, USA Large Programme, this capability will be Denis Ziegler 2 25 Center for and AstroParticle used to study the kinematics of atomic Gérard Zins 8 Physics, The Ohio State University, gas and its relation to hot dust in a Columbus, USA sample of and Seyfert . 26 Space Telescope Science Institute, We will measure a new radius- 1 Max Planck Institute for Extraterrestrial Baltimore, USA relation from spatially resolved data and Physics, Garching, Germany 27 School of Physics & Astronomy, test the current methods used to meas- 2 LESIA, Observatoire de Paris, Université ­University of Southampton, UK ure black hole mass in large surveys. PSL, CNRS, Sorbonne Université, 28 Department of Astronomy, California ­Université de Paris, Meudon, France Institute of Technology, Pasadena, 3 Max-Planck-Institut für Astronomie, USA Introduction Heidelberg, Germany 29 Steward Observatory, Department 4 I Physikalisches Institut, Universität zu of Astronomy, University of Arizona, Emission lines of atomic gas velocity- Köln, Germany Tucson, USA broadened to widths of 3000– 5 Univ. Grenoble Alpes, CNRS, IPAG, 30 University of Exeter, School of Physics 10 000 km s–1 are a hallmark of quasars Grenoble, France and Astronomy, Exeter, UK and are thought to trace the gravitational 6 CENTRA and Universidade de Lisboa 31 Kyoto Sangyo University, Department potential of the central supermassive − Faculdade de Ciências, Lisboa, of Astrophysics and Atmospheric black hole. Despite decades of study Portugal Sciences, Japan their physical origin remains unclear. The 7 CENTRA and Universidade do Porto – 32 School of Physics and Astronomy, observed properties can be explained Faculdade de Engenharia, Porto, Tel Aviv University, Israel by emission from discrete, collapsed Portugal 33 Excellence Cluster Origins, Ludwig- clouds or high-density regions of a con- 8 ESO Maximilians-Universität München, tinuous medium. The gas may be part 9 Observatoire de Genève, Université de Garching, Germany of the inflow feeding the black hole or a Genève, Versoix, Switzerland 34 Universitäts-Sternwarte München, continuous equatorial outflow. Assuming 10 DOTA, ONERA, Université Paris- Munich, Germany a gravitational origin, line widths com- Saclay, Châtillon, France 35 Shanghai Institute of Optics and Fine bined with a measurement of the emis- 11 European Space Agency, Space Mechanics, Chinese Academy of sion region size provide an estimate of ­Telescope Science Institute, Baltimore, Sciences, China the black hole mass. USA 36 NOVA Optical Instrumentation 12 Unidad Mixta Internacional Franco- Group at ASTRON, Dwingeloo, the Extensive monitoring campaigns use Chilena de Astronomía (CNRS UMI Netherlands echoes in a technique called rever- 3386), Departamento de Astronomía, 37 Aix Marseille Univ, CNRS, CNES, LAM, beration mapping to measure the emis- Universidad de Chile, Las Condes, France sion size, with ongoing work expanding Santiago, Chile 38 Observatoire de la Côte d’Azur the sample size from tens (Kaspi et al., 13 Max Planck Institute for Radio Astron- Lagrange, Boulevard de l’Observatoire, 2000; Peterson et al., 2004) to hundreds omy, Bonn, Germany Nice, France (Du et al., 2016; Grier et al., 2017). The 14 Sterrewacht Leiden, Leiden University, 39 Instituto de Astrofísica de Canarias, key result of these studies is that the Leiden, the Netherlands La Laguna, Spain size of the emitting region increases with

The Messenger 178 – Quarter 4 | 2019 21 GRAVITY Science GRAVITY Collaboration, Spatially Resolving the Quasar Broad Emission Line Region

a Radial velocity (km s–1) b 30 –4,000 –2,000 0 2,000 4,000 0.8 PAJet = 222° 2.178 1.6 20 Photocentre Pa flux α m) 0.6 2.176 (µ 1.4 ) )

10 th (µas x

e (° 0.4 1.2 2.174 leng on flu ti 0 phas 1.0 2.172 wave

al 0.2 ti ed

en 0.8 –10 rv Δ Declina er 2.170

0.0 Normalised ff

Di 0.6 Obse Phase –20 –0.2 2.168 0.4 BLR model BLR model –0.4 0.2 –30 2.14 2.15 2.16 2.17 2.18 2.19 2.20 30 20 10 0 –10 –20 –30 Observed wavelength (µm) Δ Right ascension (µas)

c Yv d 150 3

PAJet = 222° Y 100 2 )

/ –1 s

) 50 1 km

ev e¦ 3 (µas

H 0 (1

 ion "KNTC 0 0 loci ty e¦

v ve , e %+ –50 –1 MFTK@Q Δ Declin at BKNTC

CDMRHSX Radial –100 –2

1@CH@K ` BKNTC CDMRHSX –150 –3 150 100 50 0 –50 –100 –150

1LHM 1!+1 Q Δ Right ascension (µas)

Figure 1. GRAVITY spatially resolves the broad emis- This is the result of net ordered rotation of the engines. The key components of AGN sion line kinematics of 3C 273. (a) Paa line profile line-emitting gas. By comparing a kinematic model are small on the sky, at micro- to milli-­ (black) and averaged differential phase (blue), show- of the emission region (c) to GRAVITY data, we ing non-zero phases and a change of sign across find that a thick disc configuration viewed at low arcsecond scales, requiring long baselines the broad emission line. (b) Photocentre positions inclination best explains the data (d). The model also at the VLTI and Keck Interferometer. AGN measured at each line channel, showing a clear sep- provides estimates of the mean emission radius are also relatively faint sources, so far only aration between red and blue which corresponds and central black hole mass. Adapted from GRAVITY detected in optical interferometry with to a velocity gradient at a position angle perpendicu- Collaboration (2018). lar to the large-scale radio jet of 3C 273 (black line). 8–10-metre-class telescopes and instru- mentation with excellent sensitivity. Con- tinuum measurements with the Keck luminosity, roughly as R ~ L1/2. That rela- campaigns to measure R via an estimate Interferometer (for example, Kishimoto et tionship can be understood as atomic based on L). Secondary methods so far al., 2011) and the Astronomical Multi- gas emission being produced under opti- provide all available active galactic nucleus BEam combineR (AMBER) on the VLTI mal photoionisation conditions (constant (AGN) black hole mass measurements (Weigelt et al., 2012) provide information received flux). This radius-luminosity rela- in large samples and out to high . about hot dust surrounding the nucleus. tion allows “secondary” methods for The broad line region (BLR) is even smaller ­estimating black hole masses using a Interferometry provides an independent (angular size < 0.1 milliarcseconds [mas]) ­single optical spectrum (replacing long method for spatially resolving AGN central and is impossible to resolve in standard

22 The Messenger 178 – Quarter 4 | 2019 Figure 2. AGN radius-luminosity relationships meas- GRAVITY BLR ured for hot dust and atomic gas. The hot dust 10 0 measurements include our new GRAVITY results GRAVITY hot dust (purple solid circles; see GRAVITY Collaboration, Hot dust RM/OI 2019a), as well as those from previous observations. Gas RM For atomic gas, we have detected velocity gradients and measured the emission region size for the ­quasar 3C 273 (GRAVITY Collaboration, 2018) with –1 another detection and upper limits in deep integra- c) 10 tions for two other sources. With an ongoing large programme, we aim to expand the sample to roughly 10 AGN spanning four orders of magnitude in ­luminosity. The results can be compared to the large scatter found in reverberation mapping samples Radius (p ­(different samples as smaller symbols) and to the 10 –2 R ~ L0.5 relations found for both dust and atomic gas.

marily set by rotation in the black hole gravitational potential, or by polar outflow 10 –3 driven by radiation pressure? And is the 10 43 10 44 10 45 10 46 10 47 velocity structure well ordered or randomised? Bolometric luminosity (erg s–1) imaging, even with the VLTI. Instead, we By adopting a kinematic model of the By modelling the line profile and differen- can study its kinematics by measuring Paa emission region as a collection of tial phase data, we will measure the the photocentre shift of the atomic gas orbiting gas clouds (following Pancoast et emission region size and construct a new relative to the hot dust, as a function of al., 2014 and Rakshit et al., 2015), we radius-luminosity relationship. Our results wavelength (or velocity) across the emis- measure physical properties of the gas can be compared with those obtained sion line. The photocentre shift results in distribution and black hole. The data are independently from reverberation tech- a small differential phase signal ~< 1 degree consistent with a thick disc (opening niques and used to constrain the physical + 9 (Rakshit et al., 2015) whose detection angle of 45–6 degrees) in Keplerian rota- origin of the atomic gas. We will also requires high sensitivity and deep integra- tion around a of study the connection of the atomic gas to 8 tions. This is now possible with GRAVITY. 1.5–4.1 × 10 M⊙. The inclination and that of the hot dust continuum which we position angles agree with those inferred obtain using the same data (for example, for the radio jet. The measured mean GRAVITY Collaboration, 2019a & b). The

A case study in 3C 273 emission radius of RBLR = 0.12 ± 0.03 pc angular size of both the hot dust and the (at an angular diameter distance of atomic gas scales with optical flux, which We observed 3C 273 with GRAVITY 548 Mpc) is a factor of about two smaller makes interferometry well suited for using the four Unit Telescopes (UTs) over than reported in earlier RM studies (Kaspi ­studying luminous quasars like 3C 273 as eight nights between July 2017 and et al., 2000; Peterson et al., 2004) well as nearby Seyfert galaxies. A future May 2018, with a total on-source integra- although it is consistent with a recent one upgrade to the sensitivity of GRAVITY tion time of 8 hours. By combining the (Zhang et al., 2019). This first result sup- could further obtain kinematics, broad data from all epochs, we measure the ports the fundamental assumptions used emission line region size, and black hole interferometric phase with a precision of in reverberation mapping and the sec- mass estimates for large samples out to ~ 0.1–0.2 degrees per baseline. An aver- ondary methods used to measure black a redshift z ~ 2. age of three of the six baselines shows hole mass. For more details, see GRAVITY the detection of an S-shaped phase sig- Collaboration (2018). nal, corresponding to a spatially resolved Acknowledgements velocity gradient across the otherwise This research was supported by Paris Observatory, featureless broad Paa emission line (Fig- Outlook Grenoble Observatory, by CNRS/INSU, by the Pro- ure 1a). From the phase data, we fit for a gramme National Cosmologie et Galaxies (PNCG) model-independent photocentre position With an approved large programme of CNRS/INSU with INP and IN2P3, co-funded by at wavelength channels where the line we are carrying out observations of CEA and CNES, by the Programme National GRAM of CNRS/INSU with INP and IN2P3, co-funded by emission is strong. We find a clear sepa- ~ 10 sources over the next two years, CNES, by the Programme National Hautes Energies ration between blue and red channels (a spanning four orders of magnitude in (PNHE) of CNRS/INSU with INP and IN2P3, co-funded velocity gradient, Figure 1b), with an ori- AGN luminosity. The data will provide by CEA and CNES, and by the Programme National entation perpendicular to the large-scale information on the dominant kinematics de Physique Stellaire (PNPS) of CNRS/INSU, co- funded by CEA and CNES. It has also received fund- radio jet. This demonstrates net rotation and the degree of ordered motion in ing from the following programmes: European of the line emission region. The photo- atomic gas in the broad emission line Union’s Horizon 2020 research and innovation pro- centre positions are measured with a typ- region, helping us to address the follow- gramme (OPTICON Grant Agreement 730890), from ical precision of 5 µas per channel. ing questions: are the line widths pri­ the European Research Council (ERC) under the

The Messenger 178 – Quarter 4 | 2019 23 GRAVITY Science

European Union’s Horizon 2020 research and inno- FIS/00099/2013, SFRH/BSAB/142940/2018 [P. G.] GRAVITY Collaboration 2018, Nature, 563, 657 vation programme (Grant Agreement No. 743029), and PD/BD/113481/2015; M. F. in the framework of GRAVITY Collaboration 2019a, submitted to A&A, from the Irish Research Council (IRC Grant: the Doctoral Programme IDPASC Portugal), by NSF arXiv:1910.00593 GOIPG/2016/769) and SFI Grant 13/ERC/12907, grant AST 1909711, by the Heising-Simons Founda- GRAVITY Collaboration 2019b, submitted to A&A from the Humboldt Foundation Fellowship and the tion 51 Pegasi b postdoctoral fellowship, from the Bentz, M. C. et al. 2013, ApJ, 767, 149 ESO Fellowship programes, from the European Direction Scientifique Générale of Onera and by a Du, P. et al. 2018, ApJ, 856, 6 Research Council under the European Union’s Hori- Grant from Science Foundation Ireland under Grant Grier, C. J. et al. 2017, ApJ, 851, 21 zon 2020 research and innovation programme (Grant number 18/SIRG/5597. Pancoast, A. et al. 2008, MNRAS, 445, 3073 Agreement Nos. 2016–ADG–74302 [EASY], 2015- Kishimoto, M. et al. 2011, A&A, 527, 121 StG-677117 [SFH], 694513, and 742095 [SPIDI]), and Weigelt, G. et al. 2012, A&A Letters, 451, 9 was supported in part by the German Federal Minis- References Zhang, Z.-X. et al. 2019, ApJ, 876, 49 try of Education and Research (BMBF) under the grants Verbundforschung #05A08PK1, #05A11PK2, Peterson, B. M. et al. 2004, ApJ, 613, 682 #05A14PKA and #05A17PKA, by Fundação para a Kaspi, S. et al. 2000, ApJ, 533, 631 Ciência e a Tecnologia, Portugal (Grants UID/ Rakshit, S. et al. 2015, MNRAS, 447, 2420

DOI: 10.18727/0722-6691/5167

An Image of the Dust Sublimation Region in the Nucleus of NGC 1068

GRAVITY Collaboration (see page 20) Since the first seminal paper addressing Under superb conditions, with seeing its physical properties (Krolik & ­Begelman, ~ 0.5 arcseconds and a coherence time 1988), and following numerous observa- of up to 13 ms, it was possible to The superb resolution of the Very Large tions at many different wavelengths, ­fringe-track on the nucleus of NGC 1068 Telescope Interferometer (VLTI) and the “torus” concept has evolved and despite its large size and moderate the unrivalled sensitivity of GRAVITY been modified considerably. At the same brightness. The data obtained were of have allowed us to reconstruct the first time, increases in computational power excellent quality, with typically < 1% visi- detailed image of the dust sublimation have facilitated detailed modelling of bility and closure-phase accuracy. The region in an active . In the nearby clumpy torus structures. Such models wealth of information provided by the six archetypal Seyfert 2 galaxy NGC 1068, are consistent with the near- to mid- VLTI baselines has enabled us to recon- the 2 µm continuum emission traces infrared spectral energy distribution as struct a K-band image based on the a highly inclined thin ring-like structure well as dust reverberation measurements. obtained closure phases and visibilities with a radius of 0.24 pc. The observed Observations of almost two dozen with 3-milliarcsecond (mas) resolution. morphology challenges the picture of a ­galaxies using the MID-infrared Interfero- geometrically and optically thick torus. metric instrument (MIDI) on the VLTI have We used the publicly available Multi-­ resolved the 1–3 pc scales where warm aperture image Reconstruction Algorithm dust is responsible for the mid-infrared (MiRA; Thiébaut, 2008) to generate the Introduction continuum (Burtscher et al., 2013 and ref- image shown in Figure 1, which contains erences therein). However, measuring a total flux of 155 mJy. The structures NGC 1068 is one of the best studied the size of the small (< 1 pc) region con- present are robust, having been repro- nearby active galactic nuclei (AGN), in taining hot dust that emits at near-infra- duced consistently over a wide variety of which onto a central super- red wavelengths has been possible in parameter settings, and with a signal massive black hole contributes a signifi- very few galaxies. Also, until GRAVITY level much higher than that expected for cant fraction of the galaxy’s total luminos- observed NGC 1068, there were no data spurious sources. Full details are in ity. The observation of broad polarised showing spatial structure in this dust sub- GRAVITY Collaboration (2019). emission lines by Antonucci & Miller limation region. (1985) in the nucleus of this Seyfert gal- axy was central to the development of A new view of NGC 1068 the unified model that explains the differ- Observations and ences between Seyfert 1 and Seyfert 2 Image Reconstruction The image in Figure 1 is dominated objects as being due to the presence by knots of continuum arranged in a ring of a nuclear equatorial structure that both Data on NGC 1068 were obtained in around a central hole, with the south- obscures and scatters the central emis- November and December 2018 using western side about a factor of two brighter sion depending on the line of sight. GRAVITY and the four 8-metre UTs. than the north-eastern side. Fitting an

24 The Messenger 178 – Quarter 4 | 2019