university of groningen the role of neutral hydrogen in the life of … · 2016. 3. 8. · the role...

169
University of Groningen The role of neutral hydrogen in the life of galaxies and AGN Gereb, Katinka IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2014 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Gereb, K. (2014). The role of neutral hydrogen in the life of galaxies and AGN: A spectral stacking analysis. [S.n.]. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date: 17-02-2021

Upload: others

Post on 05-Oct-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

University of Groningen

The role of neutral hydrogen in the life of galaxies and AGNGereb, Katinka

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2014

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Gereb, K. (2014). The role of neutral hydrogen in the life of galaxies and AGN: A spectral stacking analysis.[S.n.].

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 17-02-2021

Page 2: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

The Role of Neutral Hydrogen in theLife of Galaxies and AGN

A Spectral Stacking Analysis

Proefschrift

ter verkrijging van de graad van doctor aan deRijksuniversiteit Groningen

op gezag van derector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

vrijdag 26 september 2014 om 14:30 uur

door

Katinka Geréb

geboren op 4 juli 1986Sangeorgiu de Padure, Romania

Page 3: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

PromotorProf. dr. R. Morganti

CopromotorProf. dr. T. Oosterloo

BeoordelingscommissieProf. dr. J.M. van der HulstProf. dr. J. van GorkomProf. dr. E. Sadler

ISBN 978-90-367-7268-6ISBN 978-90-367-7267-9 (electronic version)

Page 4: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

I am prepared to go anywhere, provided it be forward.- David Livingstone

Page 5: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Cover:Top left: Andromeda galaxy (M31). Credit: Galaxy Evolution Explorer, NASA/JPL-Caltech.Top right: Hercules A radio galaxy. Credit: NASA, ESA, S. Baum and C. O’Dea(RIT), R. Perley and W. Cotton (NRAO/AUI/NSF), and the Hubble Heritage Team(STScI/AURA).The background image displays the Westerbork Synthesis Radio Telescope. Credit: AS-TRON Nico Vermaas - Harm Jan Stiepel.

Page 6: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Contents

1 Introduction 11.1 A short history of H I observations . . . . . . . . . . . . . . . . . . . . . . 11.2 H I gas properties of galaxies . . . . . . . . . . . . . . . . . . . . . . . . . 31.3 Stacking of H I spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.4 Gas and the evolution of radio AGN . . . . . . . . . . . . . . . . . . . . . 61.5 Questions addressed in this thesis . . . . . . . . . . . . . . . . . . . . . . . 10

2 Gas and galaxy properties from a stacking experiment 172.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.2 The data: piggyback from the continuum observations . . . . . . . . . . . 192.3 Characteristics of the galaxies in the redshift range 0.06 < z < 0.09 . . . 202.4 HI stacking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

2.5.1 HI content and color separation . . . . . . . . . . . . . . . . . . . . 252.5.2 The connection between cold and ionized gas in red galaxies . . . . 262.5.3 HI and weak radio sources . . . . . . . . . . . . . . . . . . . . . . . 272.5.4 H I and 24 µm emission in LINERs . . . . . . . . . . . . . . . . . . 30

2.6 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . 302.7 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3 The global cold gas content up to z = 0.12 353.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363.2 Observations and sample selection . . . . . . . . . . . . . . . . . . . . . . 373.3 Data reduction and H I stacking . . . . . . . . . . . . . . . . . . . . . . . . 383.4 Characteristics of the galaxies in the selected sample . . . . . . . . . . . . 383.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

3.5.1 Stacking in color . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423.5.2 The H I properties of LINERs and optical AGN . . . . . . . . . . . 443.5.3 AGN and SF properties of the radio population . . . . . . . . . . . 463.5.4 The global SFR and SFE up to z = 0.12 . . . . . . . . . . . . . . . 47

3.6 Discussion and summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503.7 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Page 7: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

vi CONTENTS

4 HI absorption stacking of radio galaxies 574.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584.2 Sample selection and observations . . . . . . . . . . . . . . . . . . . . . . 59

4.2.1 Characterization of the AGN sample . . . . . . . . . . . . . . . . . 604.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

4.3.1 H I detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624.3.2 H I stacking: the H I distribution of AGN at low optical depth . . . 634.3.3 H I and radio morphology . . . . . . . . . . . . . . . . . . . . . . . 674.3.4 The nature of the H I absorbing systems . . . . . . . . . . . . . . . 69

4.4 Conclusions and implications for the future . . . . . . . . . . . . . . . . . 704.5 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

5 The HI absorption ‘Zoo’ 775.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 785.2 Description of the sample and observations . . . . . . . . . . . . . . . . . 795.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

5.3.1 Fitting complex H I absorption profiles with the busy function . . 835.3.2 Characterization of the profiles with BF parameters . . . . . . . . 87

5.4 The nature of H I absorption in flux-selected radio galaxies . . . . . . . . . 915.4.1 Are powerful AGN interacting with their ambient gaseous medium? 925.4.2 Fraction and time-scale of candidate H I outflows . . . . . . . . . . 945.4.3 Gas rich mergers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

5.5 The H I properties of compact and extended sources . . . . . . . . . . . . 975.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 995.7 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1005.8 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

5.8.1 Notes on the individual detections . . . . . . . . . . . . . . . . . . 1015.8.2 Summary table of non-detections . . . . . . . . . . . . . . . . . . . 106

6 HI, radio continuum, and optical properties of AGN 1176.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1176.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1176.3 Sample selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1196.4 Observations and data reduction . . . . . . . . . . . . . . . . . . . . . . . 1226.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

6.5.1 SDSS6: 4C 29.30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1266.5.2 SDSS8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1266.5.3 SDSS4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1276.5.4 Galaxies detected in H I emission . . . . . . . . . . . . . . . . . . . 127

6.6 Rejuvenated radio AGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1276.7 H I and optical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 1286.8 Results from the literature . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

6.8.1 H I and radio properties . . . . . . . . . . . . . . . . . . . . . . . . 1316.8.2 Optical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

6.9 Conclusions, and future perspectives . . . . . . . . . . . . . . . . . . . . . 1346.10 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

6.10.1 SDSS images and spectra . . . . . . . . . . . . . . . . . . . . . . . 135

Page 8: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

CONTENTS vii

Summary 1437.1 Chapter 2: The Lockman Hole project . . . . . . . . . . . . . . . . . . . . 1437.2 Chapter 3: From star forming to inactive galaxies . . . . . . . . . . . . . . 1447.3 Chapter 4: Probing the gas content of AGN with H I absorption stacking 1447.4 Chapter 5: The H I absorption ‘Zoo’ . . . . . . . . . . . . . . . . . . . . . 1457.5 Chapter 6: H I, radio continuum, and optical properties of radio AGN . . 1467.6 Future Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Nederlandse Samenvatting 1518.1 H I gas eigenschappen van sterrenstelsels . . . . . . . . . . . . . . . . . . . 1528.2 Stapelen van H I spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1538.3 Gas en de evolutie van radio AGN . . . . . . . . . . . . . . . . . . . . . . 1538.4 Deze thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Bibliography 151

Acknowledgements 159

Page 9: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis
Page 10: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Chapter 1Introduction

1.1 A short history of H I observationsWe happen to live in a galaxy called the Milky Way. The name arose due to the fortunatefact that human beings are endowed with rich imagination, which has got people thinkingabout the wonders of the sky for thousands of years. Galileo Galilei was the first one touse, and point an optical telescope on the Milky Way. It probably disappointed a lot ofpeople that instead of seeing highly resolved dairy products, he found himself staring atbillions of stars. The Milky Way appears to cross the sky as a bright line because we areliving inside this flat galaxy, about 8 kpc distance away from the Galactic center (see Fig.1.1). The central part of the galaxy is obscured by interstellar dust, hence our knowledgeof the detailed structure of the Milky Way has remained limited for a long time. Due toa great discovery of the 20th century, this has changed for good, and galaxies literallyappeared in a whole new light.

Each chemical element, including H I, emits and absorbs radiation at one specific fre-quency. It was predicted by Dutch astronomer Henk van de Hulst that atomic hydrogen(H I) would emit electromagnetic radiation at the frequency of 1420.405 MHz due to achange in the energy state of the hydrogen atom. This frequency falls in the radio regimeof the electromagnetic spectrum. Shortly after the spectral line of neutral hydrogen wasfirst detected in 1951 by Ewen and Purcell at Harvard University, the first maps of neu-tral hydrogen of our galaxy were created by van de Hulst in collaboration with Jan Oortand Lex Muller. For the first time, the H I maps revealed the spiral structure of the MilkyWay (see Fig. 1.1). Later, linked telescope arrays, radio interferometers have been built,allowing astronomers to study the H I gas in extragalactic sources at high resolution.

The main difference between optical and radio telescopes is that while in optical onecan measure the number of received photons, radio telescopes detect electromagneticwaves. The technology of radio interferometry was developed by British and Australianengineers and radio astronomers, including Ruby Payne-Scott, the first female radioastronomer. Taking advantage of the reflective surface of the sea, they converted asingle radar antenna into a sea-cliff interferometer near Sydney, Australia. In this simpleconfiguration, the radio detector was placed on top of a cliff, measuring the interferencepattern of radio waves reflected off the water surface.

Page 11: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2 chapter 1: Introduction

Figure 1.1: On the left: Artistic view of the Milky Way. On the right: early-typegalaxy, randomly selected from the sample discussed in this thesis

How do radio interferometers work?The u-v plane is the projection of the baselines i.e., spacings between antenna-pairs, inwavelengths (with respect to the central frequency). A two-element interferometer mea-sures the amplitude and phase of one spatial frequency (one point in the u-v plane). Theentire interferometer measures the complex visibilities at a series of different frequenciesthat can be Fourier-transformed to create the brightness distribution of the sky. Con-tinuum images of the brightness distribution are created by averaging the flux measuredin each frequency channel. Furthermore, by removing the continuum emission, one cancreate the H I spectral line cube of the observed volume of the sky. In the case of lineobservations, every frequency channel will provide one H I image that will form the datacube.

Due to the fact that we are living in an expanding Universe, the frequencies emitted bychemical elements from within distant galaxies appear redshifted toward lower frequen-cies compared to the rest frame. According to Hubble’s law, the redshift is approximatelyproportional to the distance of the galaxies. Hence, the spectroscopic redshift can serveas a rather accurate distance (or lookback time) measurement for the galaxies. To beable to detect H I in galaxies from the nearby to the distant Universe, radio telescopesneed to observe a large range of frequencies. This has become possible thanks to im-provements achieved in radio astronomy. Wide-band observations with telescopes likethe Westerbork Synthesis Radio Telescope (WSRT), the Karl G. Jansky Very Large Ar-ray (VLA), the Australia Telescope Compact Array (ATCA), and many more facilitiesthat are mentioned in this thesis, made it possible to observe large volumes of the sky inH I up to relatively high redshift.

H I observations has lead to important discoveries over the years. Hydrogen is themost abundant gas in the Universe, and it provides the primordial gas from which starseventually form. For example, in the disk of the Milky Way about 70% of the gas isin atomic form. We have learnt that galaxies can have billions of solar masses (M⊙) of

Page 12: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

1.2: H I gas properties of galaxies 3

H I gas, sometimes extending way beyond the stellar disk of galaxies. Star formation isrelated with the presence of gas, as stars are born in ‘stellar nurseries’ of dense molec-ular hydrogen clouds. Neutral hydrogen is also a great tracer of how interactions andmerging events disrupt or construct galaxies in the Universe. Furthermore, observationssuggest that H I could be partly responsible for the triggering of one of the most ener-getic phenomena in the Universe, by feeding supermassive black holes in the centres ofgalaxies.

The signal emitted by distant sources is weaker, and H I observations in the higherredshift Universe are limited by the sensitivity of current telescopes. The next generationof radio telescopes, e.g., Apertif (Oosterloo et al. 2010b), the Australian Square Kilome-tre Array Pathfinder (ASKAP, DeBoer et al. 2009), MeerKat (Booth et al. 2009) andlater on the Square Kilometre Array (SKA) are expected to bring H I studies to large,cosmologically significant distances. Building these facilities requires a great amount ofwork, financial support, and time, and it will take a few more years until they becomefully operational.

However, in recent years also statistical methods have become available to push thedetection limit of current radio telescopes, allowing one to to make full use of broad-bandreceivers and to extend H I observations to relatively high redshift. H I stacking has beenused effectively to study, globally, the average H I content of large galaxy samples.

With the goal of studying large samples of faraway galaxies statistically, the workpresented in this thesis uses spectral stacking analysis of H I. Using stacking techniques,we investigate the H I properties of hundreds of galaxies to get a better understandingof the role of H I in the evolution of galaxies over the past 1.5 Gyr. We also look at thegas properties of accreting supermassive black holes, i.e. active galactic nuclei (AGN),to investigate the interplay between AGN activity and the surrounding gas.

1.2 H I gas properties of galaxiesAlmost a century ago, by looking at the photographic images of 400 extra-galactic neb-ulae, Hubble (1926) found that galaxies can be separated into two main types based ontheir morphological appearance: spiral galaxies and ellipticals. Spirals, also known aslate-type galaxies, are rotating systems with bright stellar disks, while in galaxies whichhave spherical or elliptical shapes (early-type galaxies), the dynamics of stars is morechaotic. The two types of galaxies are presented in Fig. 1.1.

The morphological separation formed the basis of galaxy classification, and later thesetwo main galaxy types were found to follow systematic trends in many properties. Thecolors and luminosities of galaxies have long been know to correlate with galaxy mor-phology (de Vaucouleurs 1961; Chester & Roberts 1964). Late-type spirals are typicallyblue with lower surface brightness, early-type galaxies are predominantly red and moreluminous. These observational properties are easy to interpret in the context of starformation and stellar evolution history. Spiral galaxies are actively forming stars, hencetheir blue colors are given by young stellar populations. Early-type galaxies are typicallynon-star-forming, with older and redder stellar populations.

The interstellar medium of galaxies is not empty. Besides stars, galaxies can also con-tain gas and dust. Although the amount of the interstellar matter is usually only a smallfraction of the total galaxy mass, the content and physical conditions of the interstellarmedium are of great importance, as the formation of new stars will continuously affect

Page 13: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4 chapter 1: Introduction

the evolution and appearance of galaxies. Thanks to 21 cm observations of H I, earlystudies have shown that spiral galaxies almost always contain relatively large amountsof H I, while early-type galaxies contain very little, if any, gas (Roberts 1972). Henceearly-type galaxies seemed to be typically ‘red and dead’ systems.

Later, through systematic H I studies of larger samples of galaxies, we have learntthat early-type galaxies are way more interesting than simple red and dead galaxies.In fact this group shows very complex H I properties. In recent years, the SAURON(Morganti et al. 2006; Oosterloo et al. 2010a), and ATLAS3D (Serra et al. 2012) stud-ies have carried out detailed analysis of the H I properties of early-type galaxies in thenearby Virgo cluster. These studies have shown that many early-type galaxies containfast rotating stellar systems, and cold gas is present in a high fraction (about 40%) ofearly-type galaxies outside of the cluster environment. H I is in a settled disk/ring struc-ture in about half of the detected cases. However the frequent presence of unsettled gasand tidal tails suggests that interactions play a prime role in affecting the gas content.In spiral galaxies the H I mass is known to correlate well with the optical diameter andluminosity of the disk (Haynes & Giovanelli 1984; Toribio et al. 2011); however, in early-type galaxies no such relation is seen. In fact, H I in early-type galaxies covers a broadrange of H I masses and column densities, and along with the fact that the H I disk isoften kinematically misaligned with respect to the stellar disk, these are strong indica-tions that H I gas in early type galaxies is of external origin. In about 70% of galaxieswhich have H I gas in their central regions, also signs of on-going star formation are seen.This means that even though it is a modest effect, gas accretion and subsequent starformation do play role in the evolution of early type galaxies until the present days.

The mentioned studies have been carried out using direct observations in the nearbyUniverse. However, in order to understand the full picture of galaxy evolution, it is im-portant to extend H I observations to larger distances.

Why is it important to push the redshift limit of H I observations?Semi-analytic modelling of galaxy formation predicts that the cosmic density of H Iremains constant as a function of redshift (Obreschkow & Rawlings 2009; Lagos et al.2011; Popping et al. 2013). This is an intriguing result given that the star formationrate (SFR) displays a dramatic evolution, decreasing since z ∼ 2 (Madau et al. 1996;Hopkins & Beacom 2006). In the local Universe one can use direct observations toconstrain the cosmic H I mass density (ΩHI) of neutral hydrogen (Zwaan et al. 2005;Martin et al. 2010). At higher redshift, because the sensitivity of H I emission observa-tions is limited, observational constraints on the cosmic H I mass density are providedby Damped Lyman-alpha (DLA) measurments from the absorption spectra of back-ground quasars (Rao et al. 2006; Prochaska & Wolfe 2009), or by statistical methodssuch as intensity mapping analysis (Chang et al. 2010), and H I stacking of emissionspectra (Lah et al. 2007; Delhaize et al. 2013). The overall product of these studies re-veals a modest evolution of the global H I content as function of redshift (see Fig. 8 ofDelhaize et al. 2013), however, the increase is not significant up to z ∼ 0.1.

1.3 Stacking of H I spectraH I stacking is the process of co-adding the spectra of individual galaxies, with the goalof recovering the average H I signal in the stacked sample. During the stacking process,

Page 14: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

1.3: Stacking of H I spectra 5

the spectra of distant galaxies are shifted to the rest frame, and the noise-weighted sumof the spectra is produced. This way, one can measure the average H I signal of thestacked galaxies. A great advantage of stacking is that by co-adding the spectra, onecan boost the signal-to noise ratio, as the noise in the stacked spectra decreases with thesquare root of the number of stacked galaxies. Thus, this method is particularly usefulto enhance the sensitivity of individually non-detected sources, not only in the nearbyUniverse, but also at large distances where the flux emitted by individual sources fallsbelow the sensitivity limit of current radio telescopes.

The first H I stacking study was carried out by Lah et al. (2007) with the goal ofmeasuring the average H I mass and cosmic density of neutral gas for galaxies at redshiftz = 0.24. In a later study, Delhaize et al. (2013) carried out a similar stacking analysisup to z = 0.1. Besides cosmic density measurements of the neutral gas, stacking hasbeen used for several other purposes in the last years. In two studies which were focusingon cluster galaxies at z = 0.2 and z = 0.37, Verheijen et al. (2007) and Lah et al. (2009)demonstrated that stacking techniques can be efficiently used to study the environmentaldependence of the H I content. Stacking was used by Fabello et al. (2011a,b) to estimatethe H I gas fractions in a volume limited sample of galaxies within the footprint of theALFALFA survey in the redshift range 0.025 < z < 0.05. They find that in massive galax-ies (with stellar masses greater than 1010 M⊙), the H I content most strongly correlateswith the color and stellar mass surface density of the galaxies. These works ascertainthat stacking is an efficient technique and it can be used in many ways to study, globally,the H I content of galaxies.

As the final noise depends on the number of stacked galaxies, one would need rela-tively large number of galaxies with available redshifts to achieve an adequate signal-to-noise ratio. Obtaining such a large number of spectroscopic redshifts requires a lot ofobserving time. To overcome these limitations, one possibility is stacking of wide-bandinterferometric data taken in sky areas covered by optical spectroscopic surveys, wheremany sources are available thanks to the large field-of-view of radio telescopes. In well-selected fields, optical surveys can provide the sufficient number of redshifts to reducethe noise significantly, i.e. the choice of radio data is set by the availability of the opticalinformation.

In addition, confusion with nearby objects is less of a problem for interferometric ob-servations. Spectroscopic surveys such as the Sloan Digital Sky Survey (SDSS, York et al.(2000)), not only provide the redshifts needed for stacking, but also plenty of informationconcerning the optical properties of galaxies, e.g., magnitude, color, emission line fluxmeasurements. Consequently, an effective way to make full use of stacking is to combinethe obtained H I and optical information with other multiwavelength data, e.g., infrared(IR), ultraviolet (UV). To get a more comprehensive view of the nature of the selectedgalaxies, the collected information can be used to define various properties of galaxygroups. One can measure the relative H I content of the sub-samples using H I stackinganalysis, making it possible to study the role of H I in the evolution of different types ofgalaxies.

Why are multiwavelength observations useful?One can collect multiwavelength information using the publicly available catalogs of skysurveys in the optical (SDSS), IR (Spitzer, WISE), UV (Galex), radio (NVSS, FIRST)

Page 15: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6 chapter 1: Introduction

wavelength regimes. Multiwavelength information allows us to study a number of phys-ical processes for the following reasons. The radiation emitted by young stars falls inthe optical and UV bands. Star-forming regions are often embedded in a dusty medium,where the emission coming from the young stars is re-emitted at IR wavelengths by thedust. Finally, supernova remnants can accelerate cosmic rays, producing synchrotron ra-dio emission in the magnetic field of galaxies. Thus, flux measurements in the mentionedwavelength regimes can be used as an estimator for the star formation activity in galax-ies (van der Kruit 1973; Condon 1992; Kennicutt 1998; Yun, Reddy, & Condon 2001).Furthermore, multiwavelength observations have been successfully used to study the ac-tive galactic nuclei of galaxies. For example, optical emission line diagnostics have beensuccessfully used to separate star forming galaxies from Low Ionization Nuclear Emis-sion Region galaxies (LINERs) and AGN (Kewley et al. 2001; Kauffmann et al. 2003).Mid-IR color-color diagrams can be used to identify AGN which are highly obscured bydust, and several studies have used this method when searching for AGN in the highredshift Universe (Lacy et al. 2004; Stern et al. 2005). These examples show that multi-wavelength analysis is most useful to get a deeper understanding of the processes whichdominate the electromagnetic radiation in galaxies.

1.4 Gas and the evolution of radio AGNAGN activity is associated with accreting supermassive black holes (BH) in the centralregion of galaxies. Such BHs with masses in the range 106 - 109.5 M⊙ are thought to existin every galaxy with a bulge component (Kormendy & Gebhardt 2001). The mass of theBHs was found to correlate well with the bulge velocity dispersion (Ferrarese & Merritt2000; Gebhardt et al. 2000), suggesting that the formation of black holes and galaxiesmay be closely linked. Observational evidence suggests that black holes can grow throughaccretion of matter and by mergers with other black holes, and it is thought that gas playsan important role in these processes. It is believed that AGN are responsible for affectingthe gas properties and evolution of galaxies by heating or driving the gas outside of thegalaxy. AGN feedback was implemented in cosmological simulations, as an importantelement for making the predicted number counts of massive present-day galaxies matchthe observed values (Croton et al. 2006; Booth et al. 2009; Debuhr et al. 2012). Feedbackcan act through radiation, accretion-driven winds, radio jets, and these effects are thoughtto be responsible for suppressing star formation in massive galaxies and for regulatingthe growth of the BH. The estimated lifetime of radio AGN is relatively short. Although,the radio activity can be rejuvenated and galaxies can experience the feedback effects oftheir radio AGN recurrently (Saikia & Jamrozy 2009). Thus, one important questionsregarding our understanding of active nuclei is whether AGN activity is usually episodicand if so, what is the cycle of the activity.

A commonly accepted model is that AGN activity is powered by material that istransported to the BH from the interstellar medium (Rees 1984). Due to conservation ofangular momentum, the infalling material will form a flattened structure around the ac-creting BH. AGN can efficiently convert gravitational energy into radiant energy throughan accretion disk, hence the classical AGN model (see Fig 1.2) has been built around thistheory (Barthel 1989; Antonucci 1993; Urry & Padovani 1995). The main energy out-put of an accretion disks falls in the optical, UV, and X-ray part of the electromagneticspectrum. Observational evidence supports that gaseous accretion disks are present in

Page 16: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

1.4: Gas and the evolution of radio AGN 7

Figure 1.2: A model of AGN

many AGN (van Langevelde et al. 2000; Peck & Taylor 2001; Struve et al. 2010).However, nuclear activity can reveal its presence in many ways. A particularly fasci-

nating type of AGN are radio galaxies, which group are known for launching relativisticjets of radio plasma to large distances in the intergalactic medium (see Fig 1.3). Becauseaccelerated particles in the magnetic fields of relativistic jets emit synchrotron emissionat radio wavelengths, radio telescopes can be used to study the radio phase of AGNactivity. In the classical scheme of unification theory it was suggested that orientationeffects play a prime role in affecting our perception and observations of the AGN. If theaccretion disk is oriented edge-on, we would detect the radio emission coming from thejets, while if the radio jets are pointed towards us, we can only detect the radiation ofthe accretion disk.

Even though orientation effects can play an important role, more recent studies arguethat intrinsic differences could exist between different types of AGN (Hardcastle et al.2007, 2009; Best & Heckman 2012). The main difference can be accounted trough themode of accretion, as AGN activity occurs in at least two modes. The quasar-mode orcold-mode accretion happens through a radiatively efficient, optically thick and geometri-cally thin accretion disk, radiating across a broad range of the electromagnetic spectrum.

Page 17: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

8 chapter 1: Introduction

Figure 1.3: On the left: A classical case of FR II galaxies: Cygnus A (courtesy of C.Carilli, NRAO/AUI). On the right: FR I radio source in 3C 31 (credit: NRAO/AUI1999)

However, it seems that not all AGN have an accretion disk in the classical sense of theword. It has been suggested that accretion can also happen through radiatively inefficientaccretion. In this second AGN activity mode, the energy radiated through accretion isinsignificant. However, the activity is revealed in form of highly energetic radio jets.

Based on the observed properties, radio galaxies can be separated into two main typesaccording to the Fanaroff-Riley classification. These are the FR I and FR II sources shownin Fig. 1.3 (Fanaroff & Riley 1974). Although, the difference between the two types isfar from being understood. FR II sources are powerful radio sources considered to beformed during gas-rich galaxy mergers. FR II-s have relativistic jets and edge-brightenedlobes. FR I-s are less powerful sources, with edge-darkened lobes. A favored idea for thefuelling of FR I AGNs is quasi-spherical accretion of gas from the galactic halo or IGM(inter galactic matter) (Best et al. 2006; Hardcastle et al. 2007). However, the possibilityof less violent ‘dry mergers’ – which occur amongst red, gas-poor early type galaxies – isalso one possibility (Colina & de Juan 1995; Emonts et al. 2010).

Because nuclear activity and the mode of accretion in galaxies is regulated by theavailability of the gas, it is crucial to get a better understanding of the physical andkinematical conditions of the gas in the circumnuclear region of AGN.

Probing the circumnuclear region of radio AGN via H I absorptionAn important method of probing the circumnuclear region of galaxies is via H I absorp-tion. If H I is located in front of a strong radio galaxy, it can be detected in absorptionagainst the radio continuum of the AGN. Unlike H I emission, H I absorption stronglydepends on the level of background continuum of the target source. Independently of themass, H I absorption can be detected if the gas is optically thick. Thus, one can efficientlydetect even small amounts of hydrogen in absorption at relatively high redshifts, whilethe sensitivity of present-day radio telescopes allows to observe H I emission only in thevery local Universe.

The natural width of the hyperfine H I line is extremely small, about 1 km s−1 for gasof 100 K kinematic temperature. The line can appear broader due to non-zero tempera-ture of the emitting regions or turbulence in the interstellar medium, however the mostcommonly observed broadening is due to Doppler shifts caused by bulk motions of theemitting gas relative to the observer, e.g. rotation, fast gas outflows. This means that H I

Page 18: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

1.4: Gas and the evolution of radio AGN 9

can be used as a powerful tracer of the gas kinematics. Over the years, H I absorption ob-servations contributed on a major scale for our understanding of the complex processesthat occur in the nuclear region of galaxies. Absorption was found to trace a varietyof structures in AGN: regularly rotating gas disks (Emonts et al. 2010; Gallimore et al.1999), infalling H I clouds associated with the feeding mechanisms of the central blackhole (van Gorkom et al. 1989; Morganti et al. 2009) and H I outflows tracing interactionsbetween the jets and the surrounding medium (Morganti et al. 1998, 2003, 2005, 2013).Thus, the complexity of the H I kinematics in AGN suggests that gas can play manydifferent roles in AGN.

A particular type of radio galaxies, compact steep spectrum (CSS) and gigahertz-peaked spectrum (GPS) sources are intrinsically small AGN (< 10 kpc), younger than< 104 yr (Owsianik & Conway 1998, Muria 1999). This type of radio source seems to beparticularly rich in H I, while a large fraction of extended, FR I sources are undetected(Gupta et al. 2006; Emonts et al. 2010). According to the currently accepted paradigm,AGN activity is triggered in the compact phase by the infall of gas, thus CSS and GPSsources are the progenitors of extended FR I radio galaxies. van Gorkom et al. (1989)estimated that infalling clouds carry sufficient amounts of H I to trigger the nuclear activ-ity in a sample of compact CSS and GPS sources. The estimated lifetime of radio AGNactivity is 107 − 108 years (Parma et al. 1999, 2007) and, according to van Gorkom et al.(1989), small amounts of H I with masses between 103 - 105 M⊙ can fuel the AGN overthis lifetime.

Other evolutionary theories suggest that not all compact sources evolve into extendedsources (Emonts et al. 2010). This could happen for two reasons. Inefficient fuelling ofthe central black hole could prevent the source from becoming extended. The otherpossibility is that large amounts of dense interstellar gas could frustrate or even confinethe growth of the central low-power AGN (van Breugel et al. 1984; Fanti et al. 1990;De Young 1993; Pihlström et al. 2003). However, it is not clear how severe the frustrationof radio sources is, as hydrodynamical simulations (Wagner et al. 2012) and observationalevidence (Morganti et al. 1998, 2005, 2013) show that interactions between the radio jetsand the surrounding medium often result in fast outflows of cold gas. These resultssupport that radio AGN feedback can have a major influence on the gas properties ofgalaxies.

One would expect that not just mechanical, but also radiative feedback has an ef-fect, and gas outflows are quickly ionized by jet-cloud interactions. However, interest-ingly, previous studies have found that outflows in radio galaxies are predominantly cold.Both molecular and neutral gas outflows show relatively large mass outflow rates of afew × 10 M⊙ yr−1. It was estimated that the corresponding time depletion time scaleis shorter than the typical lifetime of radio galaxies, suggesting that (cold) outflows areonly present for a limited time in the life of a galaxy.

Broad absorption features associated to outflows are typically faint, showing, on aver-age, low optical depth τ ∼ 0.005 (Morganti et al. 2005). Over the last years, the numberof H I outflow detections has been increasing thanks to sensitive, broad-band observa-tions, however the observed samples are still limited. It is clear that a full understandingof the properties of H I in radio AGN can only be achieved through the study of largestatistical samples of galaxies. The observation of a large number of sources has thefurther advantage to allow stacking experiments.

Page 19: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

10 chapter 1: Introduction

Because both mechanical and radiative feedback processes can have important effectson the cold gas content of AGN, it is of interest to understand whether AGN activity isrecurrent, and if so, how is this related to the H I properties. The structure and spectralindex distribution of the lobes of radio galaxies provide important information on thehistory of the source (Saikia & Jamrozy 2009). Spectral and dynamical ages of theselobes could be used to constrain time scales of episodic activity. Over the years severalfeatures in radio sources have been suggested to be the signature of past activity. Afterthe nucleus switches off, for lack of fuelling the lobe structure will fade away. However,for a limited time we can identify these structures through their fossil emission.

The main limitation of these studies is that at the moment only a handful of suchradio relics are known. However rejuvenated sources represent a key element for ourunderstanding of the AGN activity cycle, hence these objects are receiving more andmore attention and the samples are increasing. It is particularly interesting to notethat many of the restarted sources also show H I detection in their central regions(Saikia & Jamrozy 2009; Chandola et al. 2010; Shulevski et al. 2012). The frequent pres-ence of H I in restarted AGN has been interpreted as a possible link between the presenceof cold gas and AGN fuelling. In a way this is an intriguing discovery, as it is expectedthat AGN would get rid of the gas in the previous cycle of activity. The presence ofH I suggests that cold accretion is not entirely suppressed by the AGN and, in fact, coldgas could be one of the key elements responsible for the reactivation of the radio source.A possibility other than continuous accretion from the environment is the hypothesis ofpositive feedback. In this scenario, if the velocity of the gas is too low to escape thegravitational potential of the galaxy, the gas will cool radiatively and resettle in a disk.However, at the moment this possibility is rather unexplored.

1.5 Questions addressed in this thesisWhat is the global H I content of different types of galaxies? Does the H Icontent evolve as function of redshift?One of our goals is to use stacking to measure the global H I content in galaxies up toredshift z = 0.12. We expand on previous studies mainly by using ‘multicolor’ infor-mation, and by separating different groups of galaxies based on optical emission lines,AGN diagnostics. Our analysis is also extended to higher redshift with respect to otherstacking experiments, e.g. studies using Arecibo Legacy Fast ALFA Survey (ALFALFA)observations (Fabello et al. 2011a,b). It is interesting to investigate whether the higherstar formation rate in the high redshift Universe is due to more efficient consumptionof similar amounts of gas, or it is due to larger amounts of available cold gas at earlierepochs of galaxy evolution. By tracking the redshift evolution of the global H I content,we are interested in studying the efficiency of star formation in galaxies. This allows us totest the availability of gas and star-forming conditions at different epochs of the Universe.

What is the connection of H I with star formation and AGN activity?An intriguing group of radio sources are the faint population of mJy and sub-mJy sources.At the sub-mJy level is difficult to disentangle whether star formation or radio AGN arethe main contributors for producing the radio continuum emission. As both phenom-ena are extremely relevant in galaxy evolution, one would need to separate star-formingand nuclear processes in galaxies. In this thesis we discuss several diagnostics to identify

Page 20: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

1.5: Questions addressed in this thesis 11

AGN, and we attempt to separate star-forming galaxies and AGN using multiwavelengthinformation. In particular, we focus on the connection of the H I gas with star formationand BH feeding processes.

Do feedback processes deplete cold gas reservoirs in galaxies?We investigate the effect of nuclear activity on the large scale gas content of galaxies. Itis of our interest to study whether AGN can be responsible for depleting cold gas reser-voirs in the not-so-distant Universe, or the gas is mainly consumed by star formationprocesses. To address this question, we target galaxies where feedback is more likely tobe caught in action, namely AGN, and galaxies in the green valley (with older stellarpopulations compared to actively star-forming galaxies).

What are the gas conditions in the central region of AGN, and how is thisrelated with the fuelling processes?In this thesis, for the first time, we explore the possibility to detect central H I in AGNusing H I absorption stacking. The detection rate and morphology of H I in early-typegalaxies with/without AGN component can provide information on the conditions underwhich the nucleus is activated. To address the question whether radio activity is trig-gered by accretion of cold (H I) gas, we study the gas properties of young vs. evolved (i.e.compact, extended) radio sources. The kinematics of the gas can tell us about gas accre-tion history (inflows) and feedback processes by interactions between the radio source andthe gas (outflows), thus we also study the gas kinematics in the two types of radio sources.

Are radio galaxies interacting with their ambient medium?One of our goals for using absorption stacking experiments is to test the presence of H Ioutflows in AGN at low optical depth detection limit. We would also like to know howthe gas depletion timescale compares to the life cycle of AGN. This can be important forconstraining feedback models, and for our understanding of the interplay between AGNactivity and H I gas throughout the lifetime of a radio AGN.

Restarted activity in AGN: does H I contribute to the fuelling processes?At the moment it is still not clear whether radio AGN activity occurs in every galaxy,and what is the actual duty cycle of the radio phase. It is even more unclear whether theradio activity would be reactivated in all cases, and what is the cause of the rejuvenation.Because the available samples are limited, it is difficult to fully understand the role ofH I in this process. We look for more cases of restarted radio sources and we study theH I properties of relic radio sources in this thesis.

Page 21: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

12 chapter 1: Introduction

Page 22: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Antonucci, R. 1993, ARA&A, 31, 473

Barthel, P. D. 1989, ApJ, 336, 606

Best, P.N.; Kaiser, C.N., Heckman, T.M. et al.; 2006; MNRAS; 368; 67

Best, P. N., & Heckman, T. M. 2012, MNRAS, 421, 1569

Booth, C. M., & Schaye, J. 2009, MNRAS, 398, 53

van Breugel, W., Miley, G., & Heckman, T. 1984, AJ, 89, 5

Chandola, Y., Saikia, D. J., & Gupta, N. 2010, MNRAS, 403, 269

Chang, T.-C., Pen, U.-L., Bandura, K., & Peterson, J. B. 2010, Nature, 466, 463

Chester, C., & Roberts, M. S. 1964, AJ, 69, 635

Colina, L.; de Juan, L.; 1995; ApJ; 448; 548

Condon, J. J. 1992, ARA&A, 30, 575

Croton, D. J., Springel, V., White, S. D. M., et al. 2006, MNRAS, 367, 864

DeBoer, D. R., Gough, R. G., Bunton, J. D., et al. 2009, IEEE Proceedings, 97, 1507

Debuhr, J., Quataert, E., & Ma, C.-P. 2012, MNRAS, 420, 2221

Delhaize J., Meyer M. J., Staveley-Smith L., Boyle B. J., 2013, MNRAS, 433, 1398

Emonts, B.H.C.; Morganti, R.; Struve, C.; 2010, MNRAS, 406, 987

Fabello S., Catinella B., Giovanelli R. et al., 2011, MNRAS, 411, 993

Fabello S., Kauffmann G., Catinella B. et al., 2011, MNRAS, 416, 1739

Fanaroff, B.L.; Riley, J.M.; 1974; MNRAS; 167; 31

Fanti, R., Fanti, C., Schilizzi, R. T., et al. 1990, A&A, 231, 333

Page 23: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

14 BIBLIOGRAPHY

Ferrarese, L.; Merritt, D.; 2000; ApJ; 539; 9

Gallimore, J. F.; Baum, S. A.; O’Dea, C. P.; 1999; ApJ; 524; 684

Gebhardt, K.; Bender, R.; Bower, G.; 2000; ApJ; 539; 13

van Gorkom, J. H.; Knapp, G. R.; Ekers, R. D. et al.; 1989; AJ; 97; 708

Gupta, N., Salter, C. J., Saikia, D. J., Ghosh, T.,&Jeyakumar, S. 2006, MNRAS, 373,972

Hardcastle, M.J.; Evans, D.A.; Croston, J.H.; 2007; MNRAS; 376; 1849

Hardcastle, M. J., Evans, D. A., & Croston, J. H. 2009, MNRAS, 396, 1929

Haynes, M. P., & Giovanelli, R. 1984, AJ, 89, 758

Hopkins, A. M., & Beacom, J. F. 2006, ApJ, 651, 142

Hubble, E. P. 1926, ApJ, 64, 321

Kauffmann, G., Heckman, T. M., Tremonti, C., et al. 2003, MNRAS, 346, 1055

Kennicutt, R. C., Jr. 1998, ARA&A, 36, 189

Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., & Trevena, J. 2001, ApJ,556, 121

Kormendy, J., & Gebhardt, K. 2001, 20th Texas Symposium on relativistic astrophysics,586, 363

van der Kruit, P. C. 1971, A&A, 15, 110

van der Kruit, P. C. 1973, A&A, 29, 263

Lacy, M., Storrie-Lombardi, L. J., Sajina, A., et al. 2004, ApJS, 154, 166

Lah, P., Chengalur, J. N., Briggs, F. H., et al. 2007, MNRAS, 376, 1357

Lah, P., Pracy, M. B., Chengalur, J. N., et al. 2009, MNRAS, 399, 1447

Lagos, C. D. P., Baugh, C. M., Lacey, C. G., et al. 2011, MNRAS, 418, 1649

van Langevelde, H. J., Pihlström, Y. M., Conway, J. E., Jaffe, W., & Schilizzi, R. T.2000, A&A, 354, L45

Lara, L., Márquez, I., Cotton, W. D., et al. 1999, A&A, 348, 699

Madau, P., Ferguson, H. C., Dickinson, M. E., et al. 1996, MNRAS, 283, 1388

Martin, A. M., Papastergis, E., Giovanelli, R., et al. 2010, ApJ, 723, 1359

Morganti, R., Oosterloo, T.,&Tsvetanov, Z. 1998, AJ, 115, 915

Morganti, R., Oosterloo, T. A., Emonts, B. H. C., van der Hulst, J. M.,&Tadhunter,C. N. 2003, ApJ, 593, L69

Page 24: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

BIBLIOGRAPHY 15

Morganti, R., Oosterloo, T. A., Tadhunter, C. N., van Moorsel, G.,&Emonts, B. 2005,A&A, 439, 521

Morganti, R., de Zeeuw, P. T., Oosterloo, T. A., et al. 2006, MNRAS, 371, 157

Morganti, R., Peck, A. B., Oosterloo, T. A., et al. 2009, A&A, 505, 559

Morganti, R., Fogasy, J., Paragi, Z., Oosterloo, T., & Orienti, M. 2013, Science, 341, 1082

Obreschkow, D., & Rawlings, S. 2009, ApJ, 696, L129

Oosterloo, T., Morganti, R., Crocker, A., et al. 2010, MNRAS, 409, 500

Oosterloo, T., Verheijen, M., & van Cappellen, W. 2010, ISKAF2010 Science Meeting

Parma, P., Murgia, M., Morganti, R., et al. 1999, A&A, 344, 7

Parma, P., Murgia, M., de Ruiter, H. R., et al. 2007, A&A, 470, 875

Peck, A. B., & Taylor, G. B. 2001, ApJ, 554, L147

Pihlström, Y. M., Conway, J. E.,&Vermeulen, R. C. 2003, A&A, 404, 871

Popping, G., Somerville, R. S., & Trager, S. C. 2013, arXiv:1308.6764

Prochaska, J. X., & Wolfe, A. M. 2009, ApJ, 696, 1543

Rao, S. M., Turnshek, D. A., & Nestor, D. B. 2006, ApJ, 636, 610

Rees, M. J. 1984, ARA&A, 22, 471

Roberts, M. S. 1972, External Galaxies and Quasi-Stellar Objects, 44, 12

Saikia, D. J., & Jamrozy, M. 2009, Bulletin of the Astronomical Society of India, 37, 63

Serra, P., Oosterloo, T., Morganti, R., et al. 2012, MNRAS, 422, 1835

Shulevski, A., Morganti, R., Oosterloo, T., & Struve, C. 2012, A&A, 545, A91

Stern, D., Eisenhardt, P., Gorjian, V., et al. 2005, ApJ, 631, 163

Struve, C., & Conway, J. E. 2010, A&A, 513, A10

Toribio, M. C., Solanes, J. M., Giovanelli, R., Haynes, M. P., & Martin, A. M. 2011,ApJ, 732, 93

Urry, C. M., & Padovani, P. 1995, PASP, 107, 803

de Vaucouleurs, G. 1961, ApJS, 5, 233

Verheijen M., van Gorkom J. H., Szomoru A., Dwarakanath K. S., Poggianti B. M.,Schiminovich D., 2007, ApJ, 668, L9

Wagner, A. Y., Bicknell, G. V., & Umemura, M. 2012, ApJ, 757, 136

Page 25: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

16 BIBLIOGRAPHY

York D. G., et al., 2000, AJ, 120, 1579

De Young, D. S. 1993, ApJ, 405, L13

Yun M.S., Reddy N.A., Condon J.J., 2001, ApJ, 554, 803

Zwaan M. A., Meyer M. J., Staveley-Smith L., Webster R. L., 2005, MNRAS, 359, L30

Page 26: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Chapter 2The Lockman Hole project: gas andgalaxy properties from a stackingexperiment

– K. Geréb, R. Morganti, T. Oosterloo, G. Guglielmino, I. Prandoni –Published in Astronomy & Astrophysics, 2013, 558, 54

AbstractWe perform an H I stacking analysis to study the relation between H I content and optical/radio/IRproperties of galaxies located in the Lockman Hole area. In the redshift range covered by theobservations (up to z = 0.09), we use the SDSS to separate galaxies with different optical char-acteristics, and we exploit the deep L-band continuum image (with noise 11 µJy beam−1) toidentify galaxies with radio continuum emission. Infrared properties are extracted from theSpitzer catalog.We detect H I in blue galaxies, but H I is also detected in the group of red galaxies - albeit withsmaller amounts than for the blue sample. We identify a group of optically inactive galaxieswith early-type morphology that does not reveal any H I and ionized gas. These inactive galaxieslikely represent the genuine red and dead galaxies depleted of all gas. Unlike inactive galaxies,H I is detected in red LINER-like objects.Galaxies with radio continuum counterparts mostly belong to the sub-mJy population, whoseobjects are thought to be a mixture of star-forming galaxies and low-power AGNs. After usingseveral AGN diagnostics, we conclude that the radio emission in the majority of our sub-mJyradio sources stems from star formation.LINERs appear to separate into two groups based on IR properties and H I content. LINERswith a 24 µm detection show relatively large amounts of H I and are also often detected in radiocontinuum as a result of ongoing star formation. The LINER galaxies which are not detectedat 24 µm are more like the optically inactive galaxies by being depleted of H I gas and havingno sign of star formation. Radio LINERs in the latter group are the best candidates for hosting

Page 27: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

18 chapter 2: Gas and galaxy properties from a stacking experiment

low-luminosity radio AGN.

2.1 IntroductionCold gas is known to play an important role in the formation and evolution of galaxies(Kereš et al. 2005; Crain et al. 2009; van de Voort et al. 2011). Therefore, our under-standing of the structure, properties and evolution of galaxies can not be complete with-out knowing about the various phases of the gas, including atomic hydrogen (H I), in andaround galaxies, and their dependence on other galaxy characteristics. The relation of H Iwith luminosity, morphological type, color, environment, etc., has been widely investi-gated for the z = 0 Universe, using large single-dish H I surveys, such as the H I Parkes AllSky Survey (HIPASS) (Meyer et al. 2004; Zwaan et al. 2005), the Arecibo Legacy FastALFA (ALFALFA) survey (Martin et al. 2005), and detailed imaging surveys like WHISP(van der Hulst et al. 2004), THINGS (Walter et al. 2008), SAURON (Morganti et al. 2006;Oosterloo et al. 2010a), and ATLAS3D (Serra et al. 2012).

For work at moderate redshifts (i.e. z ∼ 0.1), present-day radio telescopes are limitedby low sensitivity, and directly detecting samples of galaxies requires huge investments ofobserving time (Jaffe 1991). A possibility of going beyond these sensitivity limitationsand explore the gas content of galaxies at larger distances is given by stacking H I profiles.Although limited so far, some studies using this technique have already been carried out(Lah et al. 2007, 2009; Fabello et al. 2011a,b; Verheijen et al. 2007; Delhaize et al. 2013).

Here we expand on these studies by using stacking techniques, to investigate theproperties of the H I in galaxies located in the Lockman Hole (LH) area, one of thewell studied fields where multiwavelength data are available (Fotopoulou et al. 2012;Guglielmino et al. 2012), allowing different classes of objects to be identified and stud-ied. In our study, broad-band radio observations aimed at obtaining deep continuumimages are used to extract additional information about the H I. This is the first time acombined analysis of line and deep continuum data is attempted. A simultaneous lineand continuum setup is becoming standard for current radio telescopes, and it will beeven more common in future radio surveys. With this work we want to explore this pos-sibility. Furthermore, the relatively high spatial resolution (∼ 10′′) of the observationsused in this study reduces the risk of confusion from companion galaxies, often presentin previous, single-dish stacking experiments.

The availability of deep radio continuum data (reaching 11 µJy noise) allows us toinvestigate the relation between the gas (H I) and the radio continuum emission in sub-mJy radio sources. The nature of the faint radio population responsible for the excessin number counts below ∼ 1 mJy is controversial. Various studies (e.g. Simpson etal. 2006; Seymour et al. 2008; Smolcic et al. 2008, Mignano et al. 2008, Padovaniet al. 2009) have identified this population with star-forming (SF) galaxies (starbursts,spirals or irregulars) and low power radio-loud and/or radio-quiet AGN (e.g. faint FRI, Seyfert galaxies), and many different approaches have been taken to disentangle thesetwo phenomena (see Prandoni et al. 2009 for an overview). This study investigates thecharacteristics of sub-mJy sources and in particular their H I content at low redshift, andit provides a first step in preparing the work to be done at higher redshift.

Throughout this paper the standard cosmological model is used, with parameters Ωm

= 0.3, Λ = 0.7 and H0 = 70 km s−1 Mpc−1.

Page 28: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.2: The data: piggyback from the continuum observations 19

Figure 2.1: Redshift distribution of the SDSS galaxies in the Lockman Hole region,used for our H I stacking experiment

2.2 The data: piggyback from the continuum obser-vations

For our H I analysis we use data taken with the Westerbork Synthesis Radio Telescope(WSRT), originally aimed for the study of the radio continuum in the Lockman Holearea. The observations (Guglielmino et al. 2012, 2013 in prep.) were carried out at 1.4GHz and centered on the coordinates R.A. = 10:52:16.6, Dec = +58:01:15 (J2000). Anarea of about 6.6 square degree was covered by 16 pointings, each observed for 12 hour.Thanks to the deep observations, an rms noise of 11 µJy beam−1 was obtained in thecentral region of the final continuum mosaic, and about 6000 radio continuum sourceswere detected above a 5-σ flux density threshold. The observations were optimized forthe study of the radio continuum. However, as it is the case for many radio telescopesnowadays, the observations were carried out in spectral line mode, giving the possibilityof deriving the H I properties of the galaxies in the field to be. The setup uses 8×20 MHzbands (1300 - 1460 MHz) covering the redshift range 0 < z < 0.09 with 512 frequencychannels, corresponding to a velocity resolution of ∼75 km s−1. The bands do not overlap,some gaps between bands are present to avoid well-known RFI-dominated regions.

Relatively precise redshift measurements and sky positions are indispensable for stack-ing, thus we use the Sloan Digital Sky Survey (SDSS, York et al. 2000) catalog to selectour spectroscopic galaxy sample.

The ∼60 km s−1 typical error in SDSS redshifts provides a suitable match with theLockman Hole spectral resolution, making this dataset appropriate for stacking. Upto z = 0.09, in total 120 SDSS galaxies can be used for stacking, 50 of them beingassociated with a radio source in the catalog of Guglielmino et al. (2012, 2013 in prep.).

Page 29: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

20 chapter 2: Gas and galaxy properties from a stacking experiment

Figure 2.2: BPT diagram of 26 blue (SF+LINER) and 31 red (SF+LINER) LockmanHole galaxies with SDSS spectra available in the highest redshift bin. Arrows mark thelimits for galaxies for which only three optical lines are measured above S/N > 2 (in theother galaxies all four lines have S/N > 2). The vertical and horizontal lines representthe conventional separation for LINERs (i.e. [[O III]]/Hβ < 3 and [[N II]]/Hα > 0.6) andSeyfert galaxies (i.e. [[O III]]/Hβ > 3 and [[N II]]/Hα > 0.6). In this work we adopt thedemarcation by Kauffmann et al. (2003) (dashed line) to separate star-forming galaxiesfrom LINERs. The dotted line indicates the more stringent demarcation to identifyAGN, proposed by Kewley et al. (2001). The radio subsample is marked by circles (31sources). This figure excludes the inactive galaxies, see Sec 2.5.2 for more detail.

The distribution in redshift of the final sample is shown in Fig. 2.1.In order to avoid biases and selection effects due to the relatively large redshift and

luminosity distribution (−23 < Mr < −18) of the objects, we limit our H I analysis tothe group of galaxies within 0.06 < z < 0.09 (73 galaxies). This is also motivated by thelarger number of objects (critical for stacking), and by the interest in terms of bridgingfuture higher redshift studies. In the remainder of the paper we refer to the 0.06 < z <0.09 redshift range as the highest redshift bin.

2.3 Characteristics of the galaxies in the redshift range0.06 < z < 0.09

In addition to the spectroscopic redshifts and positions, other galaxy parameters can bederived from the SDSS database. We use the SDSS DR8 Structured Query Language

Page 30: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.3: Characteristics of the galaxies in the redshift range 0.06 < z < 0.0921

(SQL) tool1 to extract optical parameters for the Lockman Hole stacked sample. Theg and r band magnitudes are extracted together with the [[N II]], Hα, [[O III]], Hβ linefluxes to constrain the galaxy colors and properties of the ionized gas. Blue and redgalaxy samples are divided according to Blanton et al. (2001), Lockman Hole galaxieswith optical colors g−r < 0.7 are classified as blue (26 sources), while g−r > 0.7 galaxiesare classified as red (47 sources). Infrared (IR) information is extracted from the SWIRELockman Region 24 µm Spring ’05 Spitzer Catalog2.

We separate galaxies with different optical line properties using the Baldwin, Phillips& Terlevich (BPT) line ratio diagnostic diagram (Baldwin, Philips & Terlevich 1981).According to the BPT diagnostics shown in Fig. 3.2, our sample mainly consists ofstar-forming [H II] galaxies and transition/LINER (Low Ionization Nuclear Emission Re-gion) objects (Ho, Filippenko, & Sargent 1993; Heckman 1980). Star-forming galaxiesare mostly associated with blue galaxies, while transition/LINER objects have red col-ors. Commonly, LINERs are defined to have [[O III]]/Hβ < 3 and [[N II]]/Hα > 0.6(indicated by the vertical and horizontal solid lines in Fig. 3.2). Various results fromthe literature indicate that LINER-type spectra, ionized by a harder continuum, can beobtained from different processes able to ionize the gas. Among these, the most oftenproposed mechanism is ionization by AGN, but more recent studies (Sarzi et al. 2010,and ref. therein), reveal that interaction of the warm ionized gas with the hot ISMand/or ionization by old pAGB stars is more likely. Sarzi et al. (2010) found that therole of AGN photoionization is confined to the central 2-3 arcsec, but only in a handfulof galaxies.

Given the fact that this class may include a variety of ionizing sources not neces-sarily dominated by an AGN, we generically define as LINERs all galaxies above theKauffmann et al. (2003) demarcation in Fig. 3.2 (dashed line).

Besides star-forming galaxies and LINERs, our red sample involves a third group ofobjects. We call these galaxies optically inactive, defined to have less than three opticallines ([[N II]], Hα, [[O III]], Hβ) detected with a S/N> 2. For this reason, inactive galaxiesdo not appear in the BPT diagram.

In addition to this, and particularly relevant in our analysis, in our SDSS-selectedsample a group of objects is identified with a radio continuum counterpart (31 sources inthe redshift range 0.06 < z < 0.09). They are indicated with a circle in the BPT diagramof Fig. 3.2. The 1.4 GHz radio flux distribution of these sources is shown in Fig. 4.4(black hatched region). These objects mainly belong to the intriguing composite class ofsub-mJy sources. At the distance of our galaxies (z < 0.09) these fluxes (S < 2 mJy)correspond to radio powers < 1022.5 W Hz−1. By looking at the BPT diagram in Fig.3.2, we can already notice that the radio sources are spread among star-forming galaxiesand LINERs. This further supports the composite nature of this faint radio population.

Before proceeding, we investigate whether biases and selection effects are present inthe SDSS-selected sample. We start by showing in Fig. 4.4 the distribution of radiofluxes in our selected sample compared to those of the entire sample of radio sources inthe Lockman Hole area. The two distributions are not completely identical: comparedto the full sample, our sample contains only a small fraction of the faintest sources (S< 0.1 mJy). Due to the radio continuum observations being deep, most of the missingsources in our SDSS-selected sample are probably higher redshift galaxies. This effect1 http://skyserver.sdss3.org/dr8/en/tools/search/sql.asp2 http://irsa.ipac.caltech.edu/cgi-bin/Gator/nph-dd?catalog=lockman_24_cat_s05

Page 31: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

22 chapter 2: Gas and galaxy properties from a stacking experiment

Figure 2.3: Flux distribution of the highest redshift radio sample used in this study(black hatched region, i.e. 31 radio sources with available SDSS redshift) compared tothe distribution of the entire radio sample in the Lockman Hole (grey area, Guglielminoet al. 2012). For the Guglielmino et al. (2012) sample, sources with larger fluxes are alsofound, however here we only plot sources with peak flux < 2 mJy, in order to match thedistribution of sources in this study. The latter distribution is normalized by the totalnumber of galaxies (below 2 mJy) detected in the catalog of Guglielmino et al. 2012.The radio power corresponding to the fluxes is log(P1.4GHz) < 22.5 W Hz−1.

can also be due to the optical magnitude limit (r = 17.77) of the SDSS sample, togetherwith the fact that the radio flux and optical magnitude tend to be correlated (Condon1980; Mignano et al. 2008).

We also explore possible biases in optical luminosity and/or color as a function ofredshift. The magnitude limit of the SDSS sample introduces a bias against low opticalluminosity objects, and this bias is increasing with redshifts. However, the optical mag-nitude biases are not present in the highest redshift bin of our sample (0.06 < z < 0.09),where both blue and red galaxies (either radio-detected or not) show the same opticalluminosity distribution in Fig. 2.4. Red and blue galaxies with radio counterparts alsoshow similar (even if not identical) radio power distributions (see Fig. 2.5).

Following the above described criteria, we define various samples of galaxies. Bluegalaxies are classified as SF (23) and LINERs (3), while red galaxies can be classifiedinto three groups: SF (7), LINERs (24) and inactive galaxies (16). We remark that theblue LINER and red SF samples are too small for stacking, therefore in this study we donot carry out any further analysis on these galaxies.

In Sec. 2.5.3 we use the Spitzer 3.6 µm, 4.5 µm, 5.8 µm and 8.0 µm fluxes to constructthe IR color-color plot of the LH galaxies. In Sec. 2.5.4 we also make use of the 24 µminformation. The sample of blue SF IR galaxies is complete to the 95 percent level (both

Page 32: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.4: HI stacking 23

Figure 2.4: Magnitude of the blue (26) and red (47) galaxies in the highest redshiftbin. Circles indicate radio detected objects.

in the four IR bands and at 24 µm). The sample of red LINERS is complete at 3.6 µm,4.5 µm, 5.8 µm and 8.0 µm, however the completeness drops to 60 percent at 24 µm.Almost 90 percent of the red inactive galaxies are detected in the four Spitzer bands, butnone of the inactive galaxies has a 24 µm detection.

2.4 HI stackingWe used the radio-continuum calibrated uv-data (from Guglielmino et al. 2013 in prep.)to create H I data cubes of the 16 pointings. The cubes were produced after subtract-ing the continuum in the uv-plane. The subtraction was achieved by using the cleancomponents from the final continuum image.

The stacking script, written in Python, performs a number of operations startingfrom the spectrum extraction in the data cubes at the location of the SDSS galaxies, asecond continuum subtraction, de-redshifting, weighting of the spectra with the primarybeam and noise, and making the final stack.

We extract the spectra over an extended region around the optical position. One hasto select this region carefully, for the reason that a small box may not include all of theH I flux, while a larger region reaching beyond the H I extent would only increase thenoise and the risk of including companion galaxies. After extensive tests, we derive aboxsize of 30 kpc and integrate the spectrum over a region corresponding to this radiusfor each galaxy. With larger boxsizes, the mass-luminosity ratios in the higher redshiftbins do not change significantly, although the errors start to increase.

After the extraction of the spectrum, the stacking script performs a second continuumsubtraction. This is needed, because after the first subtraction in the uv-plane, continuum

Page 33: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

24 chapter 2: Gas and galaxy properties from a stacking experiment

Figure 2.5: Distribution of radio power for blue (16) and red (15) galaxies with radiocounterparts in the highest redshift bin.

residuals remained in all cubes. The second subtraction is done directly in the extractedspectra by fitting a second order polynomial to the line-free channels, i.e. excluding avelocity range of 500 km s−1 around the predicted position of the redshifted H I frequency.

To optimally weigh the stacked spectra, one has to apply corrections for the noisein every channel. This correction is needed, because the noise varies with frequency,becoming an important effect in our relatively large frequency range (∼ 40 MHz, 0.06 <z < 0.09). For the Lockman Hole data, we create the cubes separately for every pointing,and handle them individually in our stacking procedure. We use the information on thenoise along with the WSRT primary beam (Popping & Braun 2008) to correspondinglyweigh the stacked spectra (Eq. 2.1). The weighted sum of the source spectra, assuminga small range in redshift is:

S(ν) =

∑ pi(ν)σ2

i(ν) Si(ν)∑

(p2i (ν)/σ2

i (ν))(2.1)

where S(ν) is the stacked spectrum, pi(ν), σi(ν) and Si(ν) are the primary beam correc-tion, the channel noise as a function of frequency and the extracted spectrum of sourcei. Flagged data and frequency gaps are given zero-weights.In the stacking process, the noise of the co-added spectra is expected to decrease with1/

√N , where N is the number of stacked galaxies (Fabello et al. 2011a). By stacking all

120 Lockman Hole objects with spectroscopic redshift we reach a noise level of ∼20 µJyper channel, consistent with a factor of ten noise improvement compared to the initialrms (∼ 0.18 mJy/beam).

From the stacked profiles we derive the H I mass. Considering a small redshift range,H I masses can be derived from the formula:

Page 34: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.5: Results 25

Section 2.5.1MHI (109 M⊙) MHI/Lr (M⊙/L⊙)

Blue (SF+LINER) (26) 6.12 ± 0.40 0.38 ± 0.02Red (SF+LINER+Inactive) (47) 1.80 ± 0.20 0.08 ± 0.01

Section 2.5.2MHI (109 M⊙) MHI/Lr (M⊙/L⊙)

Red LINER (24) 2.40 ± 0.45 0.09 ± 0.02Red Inactive (16) < 0.95 ± 0.32 < 0.06 ± 0.02

Section 2.5.4MHI (109 M⊙) MHI/Lr (M⊙/L⊙)

LINER ’IR SF ’ (14) 3.62 ± 0.94 0.14 ± 0.03LINER ’IR inactive’ (10) < 1.20 ± 0.40 < 0.04 ± 0.015

Table 2.1: H I content and 3-σ upper limits derived for various groups in the highestredshift bin. The number of stacked sources is indicated in the brackets in Col. 1.Related results are discussed in the denoted sections.

MHI

M⊙= 235600

(1 + z)

(Sν

Jy

)(dL

Mpc

)2(∆V

km/s

)(2.2)

where z is the mean redshift of the stacked sample, Sν is the average flux integrated overthe ∆V velocity width in the emitter’s frame of the H I profile and dL is the averageluminosity distance. H I masses are further used for mass-luminosity evaluation, wherethe luminosities are calculated from SDSS r band magnitudes.

2.5 ResultsIn Table 2.1 we list the values of H I mass and MHI/Lr for all the groups we examine.The related results are discussed in the following sections.

2.5.1 HI content and color separationBlue galaxies (26 objects) are almost all star-forming (23 SF) as can be seen in Fig. 3.2.As presented in Fig. 2.6 and Table 2.1, the blue sample is detected with the highest H Icontent.

For red galaxies (47 in total) we find lower H I masses and mass-luminosity ratios(see Fig. 2.6 and Table 2.1). We compare the numbers detected at z = 0.09 for the redsample, with the H I content derived by Serra et al. (2012) for early-type galaxies within42 Mpc distance (corresponding to redshift z = 0.009). These nearby early-type galaxiesare found to span a large range of H I mass-luminosity, making it difficult to do a directcomparison with values from a stacking experiment as the one presented here. Because

Page 35: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

26 chapter 2: Gas and galaxy properties from a stacking experiment

1.405 1.410 1.415 1.420 1.425 1.430 1.435Frequency (GHz)

0

0.2

0.4

0.6

0.8

1

Flux density (mJy)

0.06 < z < 0.09

Blue (SF+LINER)Red (SF+LINER+Inactive)

Figure 2.6: H I spectra obtained for blue and red galaxies.

galaxies in our sample tend to be optically brighter compared to ATLAS3D, we apply aluminosity cut (4.5 × 109 L⊙) for ATLAS3D galaxies in order to match the LH sample.

The average mass-luminosity M/Lr = 0.003 (M⊙/L⊙) of ATLAS3D galaxies is lowercompared to the red LH sample (Table 2.1). The difference in H I mass-luminosity ratioslikely suggests that, compared to ATLAS3D, our color-selected LH sample may stillcontain a variety of morphological types, including more H I-rich late-type galaxies. Thepossibility will be further investigated in Sec. 2.5.4.

The above presented red sample includes all the red galaxies (according to the classi-fication presented in Sec. 3.4). In the next section we investigate in more detail the H Iproperties of the two main red groups, namely LINERs and inactive galaxies.

2.5.2 The connection between cold and ionized gas in red galax-ies

As can be seen in Table 2.1, the H I mass of red LINERs is higher compared to the entiresample of red galaxies, but the H I mass-luminosity ratios are similar.

An intriguing result of this stacking exercise is that while red LINER galaxies (24objects) are clearly detected in the stacked profile, the group of inactive galaxies doesnot appear to show an H I gas detection (see Fig. 2.7 and Table 2.1). Stacking of thisgroup (16 objects) results in H I non-detection with a 3-σ upper limit presented in Table2.1. This result is limited to a small group of objects and needs to be confirmed by largersamples, allowing the detection limit derived for the H I mass to be lowered.

The lack of optical emission lines along with the H I non-detection and early-typemorphology suggest that inactive galaxies must represent the true red and dead galaxysystems already depleted of cold and ionized gas. Thus, this finding - apart from tellingus about the actual mechanism responsible for ionizing the gas in LINERs - suggests

Page 36: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.5: Results 27

1.405 1.410 1.415 1.420 1.425 1.430 1.435Frequency (GHz)

0

0.2

0.4

0.6Fl

ux d

ensi

ty (m

Jy)

0.06 < z < 0.09

Red LINERRed Inactive

Figure 2.7: Stacked spectra of red LINERs (red line) and red inactive galaxies (blackline)

the key importance of gas (of which H I can represent one of the tracers) content tomake it possible for a galaxy to be classified as LINER. It appears that a galaxy needsto contain gas (cold and/or warm phase) in order to become a LINER, regardless whatthe mechanism causing the ionized emission is. In addition, the H I properties found forLINERs and inactive galaxies confirm the connection between the presence of neutraland ionized gas, already noted in the study of the SAURON sample (Morganti et al. 2006).In the detailed SAURON study, a strong link is observed between these two phases of thegas, both in terms of detection rate as well as kinematics. All SAURON and LH galaxieswhere H I is detected also contain ionized gas, whereas no H I is found in galaxies withoutionized gas.

2.5.3 HI and weak radio sourcesAs shown in a number of studies, nuclear activity is associated with massive black holestypically hosted by early-type galaxies with massive bulges. In our sample, the mostlikely group where nuclear activity can be found is the group of red galaxies, as alreadymentioned in the introduction and discussed in a number of papers (e.g. Kauffmannet al. 2003, and references in Sec. 3.4). An active nucleus can reveal its presencein different ways, not only connected with optical emission line properties, and AGNactivity can also be indicated by radio emission. However, radio emission can originatefrom star-formation processes even in early-type galaxies, as it was found for the SAURONsample (Oosterloo et al. 2010a) Thus, we now turn to investigating, for the group of redgalaxies, the nature of the radio emission detected in some of these objects. Our interest,in particular, is to understand what fraction of the radio emission may be due to nuclearactivity. Despite the depth of the radio data available, none of the optically inactive

Page 37: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

28 chapter 2: Gas and galaxy properties from a stacking experiment

RadioNr (24µm), Stacked SFRobserved

1.4GHz [M⊙ yr−1] SFR24µm1.4GHz [M⊙ yr−1]

Red LINERs (8), 10 1.87 ± 0.09 1.74 ± 0.01Blue SF (15), 15 3.86 ± 0.15 3.58 ± 0.02

Non-RadioNr (24µm), Stacked SFRobserved

1.4GHz [M⊙ yr−1] SFR24µm1.4GHz [M⊙ yr−1]

Red LINERs (6), 14 - 0.40 ± 0.005Blue SF (7), 8 - 0.95 ± 0.01

Table 2.2: Average SFR for red LINER and blue SF galaxies, derived from differentSFR indicators in the highest redshift bin (0.06 < z < 0.09). In Col. 2, numbers inthe brackets indicate the number of objects for which the 24 µm flux is detected aboveS/N > 2. However, in order to match the total sample contributing to the H I content,the SFRs are averaged over the total number of stacked objects (undetected galaxies aregiven zero SFR). Errors are calculated from the flux uncertainty of the 24 micron andradio measurements.

Figure 2.8: Infrared color-color plot in the highest redshift bin for galaxies with Spitzerdetection (22 blue SF, 24 red LINER and 14 inactive galaxies). Circles are proportionalto the radio power distribution. The dashed line indicates the region for powerful AGN.

galaxies is found to have a radio counterpart. Thus, we are looking only at the group ofred LINERs.

Using the formula by Yun, Reddy, & Condon (2001), we derive the radio SFR from

Page 38: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.5: Results 29

1.405 1.410 1.415 1.420 1.425 1.430 1.435Frequency (GHz)

0

0.2

0.4

0.6Fl

ux d

ensity

(m

Jy)

0.06 < z < 0.09

LINER "IR SF"LINER "IR inactive"

Figure 2.9: H I spectra obtained for LINERs found in the IR SF (red line) and IRinactive (black line) region.

the observed 1.4 GHz luminosity, and compare it to the radio SFR inferred from the 24µm flux density (assuming the well known radio/IR correlation holding for SF galaxies,Wu et al. 2005). The 24 µm continuum is thought to be a good tracer of the warm dustcomponent associated with current star formation in galaxies (Wu et al. 2005). Underthe assumption that all radio emission is due to star-formation processes, the SFR derivedfrom the two indicators (24 µm and 1.4 GHz) should be similar, while the presence of aradio AGN would result in a radio excess and a significantly higher value of the observed1.4 GHz SFR. In Table 2.2 we present the SFRs averaged over the total number ofstacked objects. The SFRs reveal no significant outliers, suggesting that no significantAGN counterpart is present in our sample. In the same table, for a comparison the valuesfor blue SF galaxies are also indicated.

It is also interesting to compare the star-formation properties of the radio and non-radio samples. In Table 2, the SFR of radio-detected sources (both LINERs and SFgalaxies) is always higher compared to non-radio galaxies.

The results above indicate that the radio emission in these sub-mJy radio sourcesat relatively low redshift (up to z = 0.09) is arising mostly from SF. As in early-typegalaxies in the nearby SAURON sample (Oosterloo et al. 2010a), the radio detection of LHgalaxies is mostly connected to the level of star formation i.e. more enhanced SFR ingalaxies with radio counterparts. In the SAURON sample, galaxies with H I in the centralregions were found to be more likely detected in radio continuum. This was shown tobe due to a higher probability for star formation to occur in galaxies with H I gas, andnot to H I-related AGN fuelling. This could be the case also in the LH sample and weexplore this more in the next session.

To further confirm the absence of any powerful AGN, we have created the infraredcolor-color plot based on Spitzer colors, first proposed by Lacy et al. (2004) and then

Page 39: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

30 chapter 2: Gas and galaxy properties from a stacking experiment

revisited by Stern et al. (2005). The resulting plot is shown in Fig. 2.8, illustrating thatthe red LINERs studied here - and more in particular the radio detected - appear toavoid the region of powerful AGN, indicated with a dashed line. Although the color-color plot confirms that no powerful AGN are present, several studies based on X-rayselected samples suggest that low (X-ray) luminosity or obscured AGN are missed bythe IR color-color selection wedge, and tend to appear in the region of inactive galaxies(Cardamone et al. 2008; Brusa et al. 2010; Eckart et al. 2010). A similar result was ob-tained by Prandoni, Morganti, & Mignano (2009), who found that radio-selected AGN,typically hosted by elliptical galaxies, can be located both in the AGN region (definedby the dashed line in Fig. 2.8) and in the region of inactive galaxies. Thus, based on thisplot, we can not yet completely exclude the presence of weak AGN in, at least, a subsetof the red galaxies in the LH sample.

2.5.4 H I and 24 µm emission in LINERs:are there two groups in terms of SF?

We investigate now in some more detail the apparent split of the red galaxies (red LIN-ERs and inactive) around the line corresponding to log(S8.0/S4.5) = −0.2 in Fig. 2.8.This rather arbitrary separation does actually correspond to the 24 µm detection of thesources, i.e above the line all but two (one blue SF and one red LINER) sources aredetected in 24 µm, whereas below the line all but one (one red LINER) galaxies have noassociated 24 µm dust emission.

Interesting is the connection between the detection/non-detection of the 24 µm starformation tracer and the H I detections. The stacked profiles and H I content of the twogroups of red (LINER) galaxies are presented in Fig. 2.9 and Table 2.1, the two regionsbeing referred to as IR SF region and IR inactive region. LINERs in the IR SF region aredetected with relatively high H I content and often have associated radio counterparts.However, the LINER group in the IR inactive region does not show any H I detectionand is largely populated by non-radio LINERs.

These properties indicate a strong correlation between the presence of 24 µm emissionand H I content, pointing towards SF-driven 1.4 GHz emission in the IR SF region, asseen in Sec. 2.5.3. This is consistent with what was found for early-type galaxies in thenearby Universe (see the study of the ATLAS3D sample by Serra et al. 2012) where theneutral hydrogen seems to provide material for star formation, and galaxies containingcentral H I exhibit signatures of ongoing star formation five times more frequently thangalaxies without central H I. For the group of radio LINERs in the IR inactive region,the lack of H I and 24 µm dust emission suggests that radio emission is less likely arisingfrom SF, but could be, indeed, related to weak AGN activity.

These LH results suggest that, using IR information, LINERs can be disentangledinto two groups. One group is actively star-forming, while other LINERs show moreresemblance to optically inactive galaxies and are the best candidates for hosting low-luminosity AGN.

2.6 Summary and ConclusionsThis study has investigated the H I content - using stacking techniques - of the galaxiesin the LH area, observed with the WSRT telescope and combined with SDSS spectra.

Page 40: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

2.7: Acknowledgements 31

We have focused our analysis on the redshift bin 0.06 < z < 0.09, i.e. a redshift rangerelatively clean from selection biases. The main results of this study can be summarizedin:– Both red and blue galaxies are detected in the H I stacked profile. Blue galaxies aremore H I rich, but also red galaxies show interesting amounts of H I.– Inactive (in terms of optical emission lines) galaxies are not detected in H I. This groupof early-type galaxies appears to be genuinely depleted of cold and ionized gas.– The H I properties found for red LINERs and inactive galaxies confirms the strong con-nection between the presence of neutral and ionized gas, already noted in the study ofthe SAURON sample (Morganti et al. 2006). All galaxies where H I is detected also containionized gas, whereas no H I is found around galaxies without ionized gas.– For the majority of radio LINERs, the radio emission appears to be connected withenhanced star formation.– LINERs can be separated into two groups based on 24 µm emission properties. LINERsdetected at 24 µm show relatively large amounts of H I and are often detected in radioas a result of ongoing SF. The lack of H I for the group of 24 µm undetected LINERspoints towards resemblance with optically inactive galaxies. Radio LINERs in the lattergroup are the best candidates for hosting low luminosity AGN.

This study has been limited by the number of objects and by the available SDSS red-shifts. The next step will be to increase by at least an order of magnitude the availableobjects. To do this, an observational campaign is in progress to extend the area coveredby observing many fields. With a larger number of sources, the noise level in the stackedspectra can be significantly reduced, making it possible to study various samples, and todetect H I down to lower limits compared to this study. Furthermore, we expect that in alarger sample the number of potential AGN (optical and/or radio) will increase, makingit possible to further investigate the connection between the central activity and gas.

2.7 AcknowledgementsWe thank the reviewer for the useful and detailed comments that helped us to improvethe manuscript.

Funding for the SDSS and SDSS-II has been provided by the Alfred P. Sloan Foun-dation, the Participating Institutions, the National Science Foundation, the U.S. De-partment of Energy, the National Aeronautics and Space Administration, the JapaneseMonbukagakusho, the Max Planck Society, and the Higher Education Funding Councilfor England. The SDSS Web Site is http://www.sdss.org/.

The SDSS is managed by the Astrophysical Research Consortium for the Participat-ing Institutions. The Participating Institutions are the American Museum of NaturalHistory, Astrophysical Institute Potsdam, University of Basel, University of Cambridge,Case Western Reserve University, University of Chicago, Drexel University, Fermilab,the Institute for Advanced Study, the Japan Participation Group, Johns Hopkins Uni-versity, the Joint Institute for Nuclear Astrophysics, the Kavli Institute for Particle As-trophysics and Cosmology, the Korean Scientist Group, the Chinese Academy of Sciences(LAMOST), Los Alamos National Laboratory, the Max-Planck-Institute for Astronomy(MPIA), the Max-Planck-Institute for Astrophysics (MPA), New Mexico State Univer-sity, Ohio State University, University of Pittsburgh, University of Portsmouth, Princeton

Page 41: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

32 chapter 2: Gas and galaxy properties from a stacking experiment

University, the United States Naval Observatory, and the University of Washington.This work is based [in part] on observations made with the Spitzer Space Telescope,

which is operated by the Jet Propulsion Laboratory, California Institute of Technologyunder a contract with NASA.

Page 42: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Baldwin J. A., Phillips M. M., Terlevich R., 1981, PASP, 93, 5

Blanton, M. R., Dalcanton, J., Eisenstein, D., et al. 2001, AJ, 121, 2358

Brusa, M., Civano, F., Comastri, A., et al. 2010, ApJ, 716, 348

Cardamone, C. N., Urry, C. M., Damen, M., et al. 2008, ApJ, 680, 130

Condon J. J., 1980, ApJ, 242, 894

Crain, R. A., Theuns, T., Dalla Vecchia, C., et al. 2009, MNRAS, 399, 1773

Delhaize J., Meyer M. J., Staveley-Smith L., Boyle B. J., 2013, MNRAS, 433, 1398

Eckart M. E., McGreer I. D., Stern D., Harrison F. A., Helfand D. J., 2010, ApJ, 708,584

Fabello S., Catinella B., Giovanelli R. et al., 2011, MNRAS, 411, 993

Fabello S., Kauffmann G., Catinella B. et al., 2011, MNRAS, 416, 1739

Fotopoulou, S., Salvato, M., Hasinger, G., et al. 2012, ApJS, 198, 1

Guglielmino, G., Prandoni, I., Morganti, R., & Heald, G. 2012, in ”ResolvingThe Sky - Radio Interferometry: Past, Present and Future”. Published online athttp://pos.sissa.it/cgi-bin/reader/conf.cgi?confid=163

Heckman T. M., 1980, A&A, 87, 152

Ho L. C., Filippenko A. V., Sargent W. L. W., 1993, ApJ, 417, 63

van der Hulst J. M., van Albada T. S., Sancisi R., 2001, ASPC, 240, 451

Jaffé Y. L., Poggianti B. M., Verheijen M. A. W., Deshev B. Z., van Gorkom J. H., 2012,ApJ, 756, L28

Kauffmann, G., Heckman, T. M., Tremonti, C., et al. 2003, MNRAS, 346, 1055

Kereš D., Katz N., Weinberg D. H., Davé R., 2005, MNRAS, 363, 2

Page 43: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

34 BIBLIOGRAPHY

Kewley L. J., Heisler C. A., Dopita M. A., Lumsden S., 2001, ApJS, 132, 37

Lacy, M., Storrie-Lombardi, L. J., Sajina, A., et al. 2004, ApJS, 154, 166

Lah, P., Chengalur, J. N., Briggs, F. H., et al. 2007, MNRAS, 376, 1357

Lah, P., Pracy, M. B., Chengalur, J. N., et al. 2009, MNRAS, 399, 1447

Loveday, J., Norberg, P., Baldry, I. K., et al. 2012, MNRAS, 420, 1239

Martin A. M., Papastergis E., Giovanelli R. et al., 2010, ApJ, 723, 1359

Meyer, M. J., Zwaan, M. A., Webster, R. L., et al. 2004, MNRAS, 350, 1195

Mignano A., Prandoni I., Gregorini L. et al., 2008, A&A, 477, 459

Morganti, R., de Zeeuw, P. T., Oosterloo, T. A., et al. 2006, MNRAS, 371, 157

Oosterloo, T., Morganti, R., Crocker, A., et al. 2010, MNRAS, 409, 500

Padovani, P., Mainieri, V., Tozzi, P., et al. 2009, ApJ, 694, 235

Popping A., Braun R., 2008, A&A, 479, 903

Prandoni I., Morganti R., Mignano A., 2009, arXiv, arXiv:0909.3761

Sarzi, M., Shields, J. C., Schawinski, K., et al. 2010, MNRAS, 402, 2187

Serra, P., Oosterloo, T., Morganti, R., et al. 2012, MNRAS, 422, 1835

Seymour, N., Ogle, P., De Breuck, C., et al. 2008, ApJ, 681, L1

Simpson, C., Martínez-Sansigre, A., Rawlings, S., et al. 2006, MNRAS, 372, 741

Stern, D., Eisenhardt, P., Gorjian, V., et al. 2005, ApJ, 631, 163

Verheijen M., van Gorkom J. H., Szomoru A., Dwarakanath K. S., Poggianti B. M.,Schiminovich D., 2007, ApJ, 668, L9

van de Voort F., Schaye J., Booth C. M., Dalla Vecchia C., 2011, MNRAS, 415, 2782

Smolčić, V., Schinnerer, E., Scodeggio, M., et al. 2008, ApJS, 177, 14

Walter F., Brinks E., de Blok W. J. G., Bigiel F., Kennicutt R. C., Jr., Thornley M. D.,Leroy A., 2008, AJ, 136, 2563

Wu H., Cao C., Hao C.-N., Liu F.-S., Wang J.-L., Xia X.-Y., Deng Z.-G., Young C.K.-S.,2005, ApJ, 632, L79

York, D. G., Adelman, J., Anderson, J. E., Jr., et al. 2000, AJ, 120, 1579

Yun M.S., Reddy N.A., Condon J.J., 2001, ApJ, 554, 803

Zwaan M. A., Meyer M. J., Staveley-Smith L., Webster R. L., 2005, MNRAS, 359, L30

Page 44: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Chapter 3From star forming to inactive galaxies:the global cold gas content up toz = 0.12

– K. Geréb, R. Morganti, T. Oosterloo, L. Hoppmann, L. Staveley-Smith –Submitted to Astronomy & Astrophysics

AbstractWe investigate the global neutral hydrogen (H I) content and star formation properties of ∼1600galaxies up to z = 0.12 using stacking techniques. The observations were carried out withthe Westerbork Synthesis Radio Telescope (WSRT) in the area of the SDSS South GalacticCap (SSGC), where we selected a galaxy sample from the SDSS spectroscopic catalog. Multi-wavelength information is provided by SDSS, NVSS, GALEX, and WISE. We use the collectedinformation to study H I trends with color, star-forming, and AGN (Active Galactic Nuclei)properties.

Using NUV - r colors, galaxies are divided into blue cloud, green valley and red sequencegalaxies. We detect H I in green valley objects with lower amounts of H I than blue galaxies,while stacking only produces a 3-σ upper limit for red galaxies with MHI < 5 × 108 M⊙ andMHI/Lr < 0.02 (M⊙/L⊙) (averaged over four redshift bins up to z = 0.12). We find that theH I content is more dependent on NUV - r and infrared color, and less on ionization properties,in the sense that regardless of the presence of an optical AGN (based on optical ionization linediagnostics), green galaxies always show H I, whereas red galaxies only produce an upper limit.This suggests that feedback from optical AGN is not the (main) reason for depleting large-scalegas reservoirs, or the effect of this type of feedback is not instantaneous.

Galaxies with NVSS radio counterparts are divided into IR late-type and IR early-typegalaxies based on the WISE color-color plot. We find that the radio emission in IR late-typegalaxies stems from enhanced star formation, and this group is detected in H I. However, IRearly-type galaxies lack any sign of H I gas and star formation activity, suggesting that radioAGN are likely to be the source of radio emission in this group.

The H I mass-luminosity ratio and H I-based star formation efficiency do not change signifi-

Page 45: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

36 chapter 3: The global cold gas content up to z = 0.12

cantly as function of redshift up to z = 0.12, corresponding to ∼ 1.5 Gyr in look-back time. Ourstacking study will be extended to higher redshift with the next generation of radio telescopes.Future, large surveys will provide enough data to test the global H I content at earlier epochs ofthe Universe at lower, currently rather unexplored H I detection limit (MHI < 107 M⊙).

3.1 IntroductionThe amount and conditions of cold H I gas in galaxies are, in a direct or indirect way, re-lated to star formation (SF) processes and to black hole fuelling, therefore our knowledgeof the H I properties is crucial to understand the intricate process of galaxy formation andevolution. Our knowledge of the gas content in various types of galaxies in the nearbyUniverse has increased substantially thanks to large single-dish H I surveys such as theH I Parkes All Sky Survey (HIPASS) (Meyer et al. 2004; Zwaan et al. 2005), the AreciboLegacy Fast ALFA (ALFALFA) survey (Martin et al. 2005; Grossi et al. 2009) and de-tailed imaging surveys like WHISP (van der Hulst et al. 2004), THINGS (Walter et al.2008), SAURON (Morganti et al. 2006; Oosterloo et al. 2010a) and ATLAS3D (Serra et al.2012).

In the nearby Universe, in late-type, star forming galaxies the H I mass is well cor-related with the total luminosity, the diameter of the stellar disk, and the maximumrotation speed (Toribio et al. 2011). Hence, scaling relations can be used to predict theamount of gas in these, typically blue galaxies. Red early-type galaxies, however, are anintriguing population when it comes to gas content and star formation properties. Thisgroup displays a large range of neutral hydrogen (H I) content from being very H I rich tobeing completely devoid of gas. Furthermore it is thought that feedback processes playan important role in affecting the gas reservoirs and consequently the star formationprocesses, particularly in massive, bulgy galaxies.

At higher redshift, H I studies are limited by sensitivity and bandwidth. However spec-tral stacking is an efficient technique to measure the average global H I content of galaxies.In combination with multiwavelength data, stacking also provides a powerful tool to studythe cold gas properties in different galaxy groups (Lah et al. 2007, 2009; Fabello et al.2011a,b; Verheijen et al. 2007; Delhaize et al. 2013; Geréb et al. 2013). Among thesestudies, Fabello et al. (2011a) noted that in massive galaxies the cold gas fraction moststrongly correlates with NUV - r color and stellar surface mass density, or in otherwords with the star formation history of galaxies. More recently, Geréb et al. (2013)reported the detection of H I gas not only in normal SF galaxies, but also in LINERs(Low Ionization Nuclear Emission Region), a group often associated with AGN activity(Kauffmann et al. 2003; Best & Heckman 2012). These studies show, in good agreementwith SAURON and ATLAS3D, that albeit in lower amounts, H I and star formation arepresent not just in typical SF galaxies (generally blue, late type spirals), but also ingalaxies with older stellar populations, or in AGN.

It is thought that quenching of star formation happens before the red sequence phase,in the green valley. The green valley is considered to be a transition population be-tween blue and red galaxies, displaying residual star-formation signatures (Yi et al. 2005).Therefore, it is also interesting to explore the presence and amount of H I in this pop-ulation, and to test possible differences in the gas and star formation properties withrespect to gas-rich blue and gas-poor red samples.

Page 46: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.2: Observations and sample selection 37

In Geréb et al. (2013), we used stacking techniques to test the H I properties of galax-ies located in the area of the Lockman Hole (LH). In the LH study our selection of dif-ferent groups of objects was limited by the small sample size. In this paper we presentthe global H I and SF properties of a larger sample, ∼ 1600 galaxies. We confirm severaltrends derived by Geréb et al. (2013) and expand on these results. This work is madepossible by the increase in the number of galaxies by an order of magnitude, allowing usto lower the detection limit by a factor of 4 using stacking techniques. In fact, stackingis a suitable and also efficient way to test H I content in such, relatively large samples.In addition to (Geréb et al. 2013), we study in more detail galaxies where quenchingand feedback by AGN (radio/optical) are thought to be affecting the gas reservoirs, e.g.green valley objects and LINERs.

Large samples of galaxies (following our selection) also contain an increased numberof potential AGN (optical and radio), making it possible to investigate the connection ofcold gas with nuclear activity. According to the 1.4 GHz radio luminosity function, starforming galaxies dominate the luminosity distribution below radio power < 1023 W Hz−1.Hence, in the low radio power regime we expect to detect a mix of AGN/SF galaxies. Inthis paper we explore the SF and AGN properties of the radio population extracted fromNVSS, and we discuss the possibility of separating these phenomena using IR colors.

It is clear that a better measure of the global H I content seems to be crucial for ourunderstanding of the cosmic evolution of H I. By tracking the redshift evolution of theglobal H I content and efficiency of star formation in galaxies, one might be able to testthe availability of gas and the conditions under which star formation occurs at differentepochs of the Universe. Recent results show that in the nearby Universe, the H I-basedstar formation efficiency (SFE = SFR/MHI) – or the equivalent inverse, the time scale ofcold gas consumption (t = MHI/SFR) – is independent of other galaxy properties, suchas stellar mass, stellar surface density, color, concentration (Schiminovich et al. 2010;Bigiel et al. 2011). This result was interpreted as a signature that external processesand/or feedback by SF/AGN, which processes regulate the H I gas fraction in galaxies,can be responsible for regulating star formation as well. In this paper we probe theglobal H I content and SFE properties up to z = 0.12. In a second paper, using the samedataset, the ΩHI will be investigated in the same redshift range.

Throughout this paper the standard cosmological model is used, with parameters Ωm

= 0.3, Λ = 0.7 and H0 = 70 km s−1 Mpc−1.

3.2 Observations and sample selectionThe H I observations were carried out with the Westerbork Synthesis Radio Telescope(WSRT) at 1.4 GHz, in the period May 2011 - October 2012, in the area of the SDSSSouth Galactic Cap (SSGC). Between the coordinates 21h < RA < 2h, 10 < DEC <16 (J2000), 35 WSRT pointings were observed and used for H I spectral stacking.

The redshift range 0 < z < 0.12 is covered by 8 × 20 MHz bands with 128 frequencychannels in each band (1280 - 1420 MHz, the bands overlap 3 MHz). The correspondingvelocity resolution is ∼38 km s−1. The integration time is 12 hours for most of theobservations. The synthesized beam is typically ∼70 × 9 arcsec, the elongation of thebeam being (mainly) the result of the low declination observations with the East-WestWSRT array.

Page 47: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

38 chapter 3: The global cold gas content up to z = 0.12

In each observed pointing, we use the Sloan Digital Sky Survey (SDSS, York et al.2000) to pre-select our spectroscopic galaxy sample. The pre-selected sample is cross-correlated with the Galaxy Evolution Explorer (GALEX, Martin et al. 2005), providing1595 galaxies that can be used for stacking. We also use the Wide-Field Infrared SurveyExplorer (WISE, Wright et al. 2010) to obtain infrared (IR) data for our galaxies. TheWISE sample is complete to the 99 percent level. In addition, a few (50, within a searchradius of 15 arcsec) galaxies are identified with radio counterparts in the NRAO VLA SkySurvey (NVSS, Condon et al. 1998) survey. The collected multiwavelength data allowsus to combine various galaxy parameters and investigate the H I properties of differentgalaxy groups.

3.3 Data reduction and H I stackingThe data were reduced using the MIRIAD package (Sault, Teuben, & Wright 1995). Baddata were flagged from the datasets, with extra care for the most prominent RFI in thelowest-frequency band (1280 - 1300 MHz).

The standard way to subtract the continuum is by fitting a low-order polynomialto the line-free channels. Our datasets cover a broad, 140 MHz bandwidth composedby 8 spectral windows. We fit each spectral band separately; however, we find thatpolynomial fitting is unsuccessful when strong H I lines are present close to the edgeof the bands. These H I lines create a dip in the stacked spectra because of the badcontinuum subtraction.

To avoid this effect, first we perform a continuum subtraction in the uv-plane, usingthe clean components of each field. The clean components were created as a resultof deconvolution of the continuum images with the dirty beam. This subtraction stepremoves most of the continuum from the H I data cubes. However, we perform a secondcontinuum subtraction by fitting a second-order polynomial to the spectra to subtractlow-level residual continuum emission coming from very bright sources. Following thesesteps, we eliminate the dip from the stacked spectra.

Stacking is done similarly as described in Geréb et al. (2013), centred on the redshiftof the galaxy to be stacked. Galaxies are stacked in four redshift ranges, between 0.02< z < 0.12 each redshift range covers ∆z = 0.02 (except the highest redshift range, 0.08< z < 0.12). The average rms noise of the cubes is ∼ 0.2 mJy/beam, which number isexpected to decrease with the square root of the number of co-added sources.

3.4 Characteristics of the galaxies in the selected sam-ple

Our main goal is to study/compare the H I properties of various groups of galaxies usinglarge samples with available multiwavelength information. In order to do this, we defineseveral sub-samples, using the collected multiwavelength data.

The color distribution of galaxies is known to consist of two main peaks, i.e. the bluecloud and the red sequence (Strateva et al. 2001; Baldry et al. 2004). At intermediatecolours between blue and red galaxies an excess population is present in the distributionat fixed absolute magnitudes (Wyder et al. 2007). Intermediate, green colors can be due

Page 48: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.4: Characteristics of the galaxies in the selected sample 39

Figure 3.1: 1. Color-magnitude diagram (top panel) 2. UV-optical color-color plot(bottom panel); The galaxies are color coded according to the color-magnitude selectionby Wyder et al. (2007), i.e. blue cloud, green valley and the red sequence. Radio sourcesare marked by yellow squares.

to a number of different phenomena, e.g. low level (residual) star formation activity(Yi et al. 2005), dusty galaxies, older stellar populations (Sarzi et al. 2010).

Previous stacking studies, including our LH analysis, were carried out on galaxiesseparated into blue/red samples, which colors were defined based on optical g − r, orultraviolet-optical NUV - r selections (Fabello et al. 2011a; Geréb et al. 2013). To expand

Page 49: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

40 chapter 3: The global cold gas content up to z = 0.12

Figure 3.2: 1. BPT diagram (top panel). Galaxies are color coded according tothe color-magnitude selection by Wyder et al. (2007), i.e. blue cloud, green valley andthe red sequence. The dashed line (Kauffmann et al. 2003) is separating SF galaxies(below the dashed line) from LINERs (between dashed and dotted line). Optical AGNare located above the dotted line (Kewley et al. 2001). Inactive galaxies do not appearin the diagram. 2. WISE IR color-color plot (bottom panel). The sources are color-coded according to the BPT selection. The vertical solid line is separating IR early-type([4.6µm] - [12µm] < 2) and IR late-type galaxies ([4.6µm] - [12µm] > 2). Radio sourcesare marked by yellow squares.

Page 50: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.4: Characteristics of the galaxies in the selected sample 41

on previous studies, in this paper we consider the green valley as a separate group.Following Wyder et al. (2007), in Fig 3.2 our objects are divided into blue cloud, greenvalley and red sequence objects based on NUV - r colors. To derive the color distributionof the galaxies, Wyder et al. (2007) utilizes the fit by Yi et al. (2005) to the NUV - rcolors in function of the Mr absolute magnitude: NUV - r = f(Mr) = 1.73 - 0.17Mr. Thered sequence is defined as the galaxies with NUV - r > f(Mr) - 0.5, blue galaxies haveNUV - r < f(Mr) - 2, whereas green galaxies are the excess population between blueand red galaxies with f(Mr) - 2 < NUV - r < f(Mr) - 0.5 colors. The optical and UVapparent magnitudes are extracted from GALEX and SDSS, and K-corrected followingChilingarian et al. (2010); Chilingarian & Zolotukhin (2012). The NUV - r colors arecorrected for Galactic extinction following Wyder et al. (2007).

From Fig. 3.1 (bottom panel) it is clear that in case of the Geréb et al. (2013)selection, our current green valley objects are part of the g − r > 0.7 (red) sample, andin the selection of Fabello et al. (2011a), green valley objects belong to the NUV - r <4.5 (blue) sample.

We also use the Baldwin, Phillips & Terlevich (BPT) line ratio diagnostic diagram(Baldwin, Philips & Terlevich 1981) to separate galaxies with different ionization proper-ties (Fig. 3.2, top panel). The selection is done by using line fluxes from SDSS, similarlyto Geréb et al. (2013), but with one major difference. Galaxies, which were defined asLINERs (Low Ionization Nuclear Emission Region galaxies) in Geréb et al. (2013), arenow separated into LINER and optical AGN samples, using the more stringent demar-cation by Kewley et al. (2001) for selecting AGN. Following the BPT classification, oursample includes star-forming (SF) galaxies, LINERs and optical AGN. Furthermore, apart (280 sources) of our sample is defined as optically inactive, with non-detected, orwith maximum two detected lines (among [NII], Hα, [OIII], Hβ). In the relatively smallsample of Geréb et al. (2013), optically inactive galaxies were not detected in the ra-dio continuum. However here, thanks to the increased number of objects, we also finda few optically inactive (non-star-forming) galaxies with associated radio counterparts.Such non-star-forming radio sources are good candidates for hosting low-luminosity ra-dio AGN. Star forming galaxies in the BPT diagram in Fig. 3.2 (top panel) are mainlyblue and green, while LINERs and optical AGN are green and red. Inactive galaxies aretypically red, as in Paper1.

Radio sources are among the optically most luminous objects in each group in thecolor-magnitude diagram in Fig. 3.1 (top panel). From the 1.4 GHz radio luminosityfunction it is expected that radio AGN become dominant over star formation at radiopower higher than P > 1023 W Hz−1 (Mauch and Sadler 2007). From the radio powerdistribution in Fig 3.3 we expect to have such powerful AGN, however the distributionshows that also low-power radio sources are present in the sample. In the low radiopower regime it becomes more complicated to disentangle the contribution of SF andAGN to the radio continuum emission. This effect is well illustrated in the BPT diagramin Fig. 3.2 (top panel), where radio emission is detected both in SF galaxies and inLINERs/AGN.

IR colors were found to be efficient in disentangling SF and AGN activity in galaxies(Geréb et al. 2013). With this goal in mind, we extract 3.4 µm, 4.6 µm and 12 µmmagnitudes from WISE to constrain the IR color-color plot, presented in Fig 3.2 (bottompanel). The separation at the vertical line (at [4.6µm] - [12µm] = 2) in the IR color-colorplot (Fig. 3.2, bottom panel) is often used in the literature to disentangle IR early- and

Page 51: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

42 chapter 3: The global cold gas content up to z = 0.12

20 21 22 23 24 25 26 log(PIntegrated) [W/Hz]

0

2

4

6

8

10

12

14 Count

NVSS sources

Figure 3.3: Radio power histogram of NVSS sources in our sample.

late-type galaxies (Wright et al. 2010; Sadler et al. 2013). The sample in this figure iscolor-coded according to the BPT diagram. IR late-type galaxies include star-forminggalaxies and LINERs, whereas the IR early-type region is dominated by optical AGNand non-star-forming (inactive) galaxies. The IR early-type sample is associated with redgalaxies at NUV - r > 5 in the color-magnitude diagram in Fig. 3.1 (top panel) and inthe optical-UV color-color diagram (Fig. 3.1, bottom panel). As expected, radio sourcesin the IR late-type region are blue and green. Thus, star forming galaxies and potential(radio) AGN seem to be well separated by IR colors, and a more detailed analysis of thetwo radio groups is presented in Sec. 3.5.3.

3.5 Results

3.5.1 Stacking in color

Before we look at the H I content of blue cloud/green valley/red sequence objects, firstwe compare our current measurements with the results from the LH study. In Fig 3.4(top panel) we use g − r optical colors to evaluate the H I mass and mass-luminosityratio of blue/red galaxies in our current, larger sample. In the LH field, in the redshiftrange 0.06 < z < 0.09 we measured MHI = 6.12 ± 0.4 × 109 M⊙ and MHI/Lr = 0.38 ±0.02 (M⊙/L⊙) in the blue (g − r < 0.7) population, whereas red (g − r > 0.7) galaxiescontain lower amounts of gas, with MHI = 1.8 ± 0.2 × 109 M⊙ and MHI/Lr = 0.08± 0.01 (M⊙/L⊙). In the redshift range of the LH studies (0.06 < z < 0.09), the newmeasurements in Fig. 3.4 (top panel) are consistent with the results of the LH.

The H I mass and mass-luminosity ratio for blue cloud, green valley and red sequenceobjects are also presented in Fig 3.4 (bottom panel). We detect H I in blue and green

Page 52: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.5: Results 43

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

2

4

6

8

10

12

MHI (109 M

⊙)

g - r < 0.7

g - r > 0.7

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0.0

0.2

0.4

0.6

0.8

MHI / L r (M⊙ / L⊙)

g - r < 0.7

g - r > 0.7

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

2

4

6

8

10

12

MHI (109 M

⊙)

NUV - r < (f - 2)

(f - 2) < NUV - r < (f - 0.5)

NUV - r > (f - 0.5), upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0.0

0.2

0.4

0.6

0.8

1.0

MHI / L r (M⊙ / L⊙)

NUV - r < (f - 2)

(f - 2) < NUV - r < (f - 0.5)

NUV - r > (f - 0.5), upper limit

Figure 3.4: H I mass (left) and mass-luminosity ratio (right) of galaxies divided indifferent colors. (Top panel): Blue and red galaxies selected from g − r colors as inGeréb et al. (2013). (Bottom panel): our current color selection of blue cloud, greenvalley and red sequence objects.

objects, however unlike in the LH studies, red galaxies do not show an H I detection1.The non-detection of red galaxies is the consequence of the different color selection, andin fact, our non-detection is in good agreement with previous results from the literature.Fabello et al. (2011a) reported H I non-detection in red galaxies with NUV - r > 4.5,which color limit is similar to our red sequence definition (see color-color plot in Fig. 3.1,bottom panel). Using stacking techniques here we expand on the results of Fabello et al.(2011a), and we confirm the H I non-detection found in red galaxies at low 3-σ limit ofMHI < 5 × 108 M⊙ and MHI/Lr < 0.02 (M⊙/L⊙) (with values averaged over the fourredshift bins).

As expected, the global H I content is decreasing from the blue population towardsred galaxies in Fig. 3.4. Green valley objects are a transition population from H I pointof view, showing lower amounts of H I than the blue population, however unlike redgalaxies, green valley objects are not completely devoid of gas.

Finally we note that similarly to what previous studies have found (Freudling et al.2011; Delhaize et al. 2013), the global H I mass-luminosity ratio does not change signif-1 We note that in the first redshift bin we find a tentative H I detection in red galaxies at the 3-σ

level, with MHI < 3.8 × 108 M⊙ and MHI/Lr < 0.03 (M⊙/L⊙). Higher redshift bins are notdetected.

Page 53: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

44 chapter 3: The global cold gas content up to z = 0.12

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

2

4

6

8

10

12

MHI (109 M

⊙)

SFLINEROptical AGNInactive, upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0.0

0.2

0.4

0.6

0.8

MHI / L r (M⊙ / L⊙)

SFLINEROptical AGNInactive, upper limit

Figure 3.5: H I mass (left) and H I mass-luminosity ratio (right) measured in differentgroups from the BPT diagram.

icantly up to z = 0.12. We expand on previous studies by separating the sample intodifferent groups in Fig. 3.4. We show that even when galaxies are divided into sev-eral groups, the global H I content (mass-luminosity ratio) shows only little variations asfunction of redshift.

3.5.2 The H I properties of LINERs and optical AGNThe H I stacking results of SF, LINER, optical AGN and inactive galaxies are presentedin Fig. 3.5. Besides SF galaxies, H I is detected in LINERs and optical AGN, whichresult is in good agreement with the Geréb et al. (2013) study. Thanks to the highernumber of objects in the SSGC, we can lower the H I detection limit and confirm the H Inon-detection of optically inactive galaxies down to the 3-σ level of MHI < 6 × 108 M⊙and MHI/Lr < 0.03 (M⊙/L⊙) (averaged over the four redshift bins).

Within a certain population, the H I mass-luminosity ratio does not change signifi-cantly with redshift up to z = 0.12. However, the H I gas fraction in Fig. 3.5 seems todecrease from the SF population towards optical AGN, and inactive galaxies are devoidof gas.

In the BPT diagram in Fig. 3.2 (top panel), LINERs and optical AGN are mostlygreen or red. We want to test whether the lower H I content in LINERs/AGN is theresult of galaxies being redder (with older stellar populations, dustier), or it is relatedto ionization/AGN feedback properties. To do this, in the following analysis we select agroup of potential AGN based on the BPT diagram, and we use IR and NUV - r colorsto test the color dependence of H I in the selected groups.

First we create a combined sample of LINERs and optical AGN, considering all galax-ies above the dashed line in the BPT diagram in Fig 3.2 (top panel). In Paper1 we showthat IR colors are efficient in separating H I-rich and H I-poor LINERs. The formergroup is associated with star formation activity, however H I-poor LINERs are non-star-forming. In Fig 3.2 (bottom panel), 63% of LINERs/AGN are located in the IR late-typeregion in the WISE color-color plot in Fig. 3.2 (bottom panel). The late-type region isdominated by star forming galaxies, and this suggests that a large fraction of LINERs inthis region could be also associated with (low-level) star-formation. Based on our results

Page 54: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.5: Results 45

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

2

4

6

8

10

12

MHI (1

09 M

⊙)

LINER+AGN, IR late-typeLINER+AGN, IR early-type, upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0.0

0.2

0.4

0.6

0.8

MHI / L r

(M⊙

/ L⊙)

LINER+AGN, IR late-typeLINER+AGN, IR early-type, upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

2

4

6

8

10

12

MHI (109 M

⊙)

LINER+AGN, (f - 2) < NUV - r < (f - 0.5)

LINER+AGN, NUV - r > (f - 0.5), upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0.0

0.2

0.4

0.6

0.8

MHI / L r (M⊙ / L⊙)

LINER+AGN, (f - 2) < NUV - r < (f - 0.5)

LINER+AGN, NUV - r > (f - 0.5), upper limit

Figure 3.6: (Top panel): H I mass (left) and H I mass-luminosity ratio (right) of LINERsand optical AGN with green and red colors. (Bottom panel): The same H I measure-ments for LINERs and optical AGN separated into WISE late-type and early-type regionsamples. Upper limits are marked by dashed lines in all plots.

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

2

4

6

8

10

12

MHI (109 M

⊙)

IR late-type (radio)IR early-type (radio), upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0.0

0.2

0.4

0.6

0.8

MHI / L r (M⊙ / L⊙)

IR late-type (radio)IR early-type (radio), upper limit

Figure 3.7: H I mass (left) and mass-luminosity ratio (right) of radio sources in the IRlate- and early-type regions.

Page 55: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

46 chapter 3: The global cold gas content up to z = 0.12

from Paper1, we expect H I detection in these IR late-type LINERs.In Fig. 3.6 (top panel) the stacked profiles of LINERs and AGN are separated into

IR late- and early-type galaxies. As expected, we detect H I in the IR late-type sample.However, galaxies in the IR early-type region do not show a significant detection2 at the3-σ limit of MHI < 1.05 × 109 M⊙ and MHI/Lr < 0.04 (M⊙/L⊙) (averaged over the fourredshift bins). With this test we confirm the result of the LH study, that the presenceof H I is well correlated with IR colors, even in the sample of LINERs and optical AGN.

After testing the IR color dependence, we also stack LINERs/AGN separated intogreen and red samples based on NUV - r colors. In Fig. 3.6 (bottom panel), H I is con-centrated in green objects among LINERs/AGN, whereas red galaxies are not detected3

in H I at the 3-σ limit of MHI < 9.7 × 108 M⊙ and MHI/Lr < 0.04 (M⊙/L⊙) (averagedover the four redshift bins). This suggests that even those galaxies in which the presenceof an AGN is expected to be more likely (LINER/AGN) do show H I detections, how-ever this depends on their color. This is a strong indication that the H I content is wellcorrelated with the NUV - r color and the SF history of the galaxies, while the effect ofAGN feedback on the gas content is less significant.

These results show that both NUV - r and IR colors are efficient in separating H I-richand H I-poor galaxies. To understand the relation of the NUV - r vs. IR color selection,we study in more detail the color distribution of galaxies in the IR late- and early-typeregion. We find that red galaxies are the bulk of the IR early-type sample, while IRlate-type galaxies are mainly green and blue. Hence, the correlation between NUV andIR colors explains that we detect similar amounts of H I in the two different selections.

3.5.3 AGN and SF properties of the radio populationAs we have shown in Sec 3.5.2 and in Geréb et al. (2013), we can use IR colors toseparate star forming galaxies from non-star-forming samples (potential AGN). Now wedo the same analysis with radio-selected NVSS objects (see Sec 3.4 for information onthe separation of radio sources).

The stacking results of the two radio groups, IR early- and late-type galaxies separatedat [4.6µm] - [12µm] = 2, are presented in Fig. 3.7. We detect H I in IR late-type galaxies,however in the IR early-type region H I reveals a non-detection with a mass and mass-luminosity upper limit of MHI < 1.95 × 109 M⊙ and MHI/Lr < 0.03 (M⊙/L⊙). TheH I non-detection of IR early-type galaxies is consistent with the fact that in our colorselection scenario these are red objects. IR late-type galaxies are green and blue. Thelow mass-luminosity ratio of H I detections in IR late-type galaxies with respect to e.g.LINERs/optical AGN in Fig. 3.6, is partly the result of the high optical luminosity ofradio-detected objects, which are among the brightest sources in the color-magnitudediagram in Fig. 3.1 (top panel).

The radio power distribution of IR late-type and IR early-type galaxies is presentedin Fig. 3.8. The two distributions have different shapes, with a wide, D = 0.504 max-imum distance between the two cumulative distribution functions. According to theKolmogorov-Smirnov test, the probability that the two distributions are different is 99%,2 We note that in the first redshift bin we tentatively detect H I in IR early-type LINERs/AGN at

the 5-σ level, with MHI < 9 × 108 M⊙ and MHI/Lr < 0.05 (M⊙/L⊙). Higher redshift bins arenot detected.

3 In the first redshift bin we tentatively detect H I in red AGN/LINERs at the 3-σ level with MHI

< 9 × 108 M⊙ and MHI/Lr < 0.05 (M⊙/L⊙). Higher redshift bins are not detected.

Page 56: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.5: Results 47

21.5 22.0 22.5 23.0 23.5 24.0 24.5log(P1.4 GHz) [W/Hz]

0.0

0.2

0.4

0.6

0.8

1.0

Cumulative fraction

IR late-type (radio)IR early-type (radio)D

Figure 3.8: Cumulative fraction of the radio power distribution in IR late-type and IRearly-type galaxies. We measure D = 0.504 for 35 IR late-type, and 15 IR early-typegalaxies.

implying that statistically IR late-type and IR early-type galaxies have a different radiopower distribution. The mean radio power of IR late-type galaxies is log(P) = 22.5 WHz−1, whereas as expected, IR early-type galaxies are typically more powerful, with amean log(P) = 23 W Hz−1. Along with the lack of star formation and H I gas, the highradio power supports that radio emission in the IR early-type region is due to radio AGN.

3.5.4 The global SFR and SFE up to z = 0.12In Geréb et al. (2013) we found that the presence of H I provides favourable conditions forstar formation not just in blue, but also in optically red (g − r > 0.7), LINER galaxies.In this paper, in Sec 3.5.1 and Sec 3.5.2 we show that the H I content of galaxies iswell correlated with their star formation history (color), therefore we want to test theefficiency of star formation in different types of galaxies, bearing in mind their H I content.The star formation rate (SFR) in our sample is derived from the NUV flux, followingSchiminovich et al. (2010). The SFR formula accounts for dust attenuation by combiningUV-optical colors (NUV - r) and the Dn(4000) index of galaxies. The latter index is anindicator of the presence of young stellar populations.

In each redshift bin, the H I-based star formation efficiency is defined as the averageSFR over stacked H I mass, i.e. ΣSFR/ΣMH I.

The plots in Fig. 3.9 show that up to z = 0.12, the SFE shows little variation withredshift. However the SFR properties do change for different galaxy groups, i.e. fromthe blue cloud towards the red sequence, or from SF galaxies towards optical AGN andinactive galaxies.

In Fig. 3.9, green valley objects show lower SFR than blue galaxies, however these

Page 57: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

48 chapter 3: The global cold gas content up to z = 0.12

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

1

2

3

4

5

SFR

[M⊙ yr−

1]

NUV - r < (f - 2)

(f - 2) < NUV - r < (f - 0.5)

NUV - r > (f - 0.5)

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

−11.0

−10.5

−10.0

−9.5

−9.0

−8.5

−8.0

log (

SFR

/MHI)

[yr−

1]

NUV - r < (f - 2)

(f - 2) < NUV - r < (f - 0.5)

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

1

2

3

4

5

SFR

[M⊙ yr−

1]

SFLINEROptical AGNInactive, upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

−11.0

−10.5

−10.0

−9.5

−9.0

−8.5

−8.0lo

g (

SFR

/MHI)

[yr−

1]

SFLINEROptical AGN

Figure 3.9: SFR (left) and SFE (right) of all groups of galaxies which have been studiedin Sec. 3.5.1, 3.5.2, 3.5.3. The error on the SFR is estimated from the 3-σ NUV fluxerrors. In the SFE figures, the blue dashed line indicates the region of high efficiencystar formation, whereas the low efficiency region is marked by a red dashed line.

two groups reveal similar, efficient star formation, with little variations around SFE =10−9.5 yr−1, corresponding to a gas consumption time scale of t ∼ 3×109 yr. However,red galaxies lack any sign of H I gas and star formation activity.

In Sec 3.5.2 we found that the H I content is better correlated with IR and NUV - rcolor rather than with ionisation/AGN properties, and from Fig. 3.9 it seems that thesame applies to the star formation properties as well. Similarly to the H I gas properties,star formation is concentrated in green/blue LINERs and optical AGN, while in red AGN(including LINERs) the level of star formation is negligible. Furthermore, the NUV -r and the IR WISE selections yield very similar results. The SFR is ∼1 M⊙ yr−1 ingreen and IR late-type LINERs in Fig. 3.9. The efficiency of SF is relatively low in bothgroups, with SFE . 10−9.6 yr−1.

Finally, the SFE of radio sources in the IR late-type region (Fig. 3.9) is the most en-hanced amongst all groups investigated in this paper, corresponding to gas consumptiontime scales of t = 109 yr.

Page 58: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.5: Results 49

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

1

2

3

4

5

SFR

[M⊙ yr−

1]

LINER+AGN, IR late-typeLINER+AGN, IR early-type, upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

−11.0

−10.5

−10.0

−9.5

−9.0

−8.5

−8.0

log (

SFR

/MHI)

[yr−

1]

LINER+AGN, IR late-type

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

1

2

3

4

5

SFR

[M⊙

yr−

1]

LINER+AGN, (f - 2) < NUV - r < (f - 0.5)

LINER+AGN, NUV - r > (f - 0.5), upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

−11.0

−10.5

−10.0

−9.5

−9.0

−8.5

−8.0

log (

SFR

/MHI)

[yr−

1]

LINER+AGN, (f - 2) < NUV - r < (f - 0.5)

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

0

1

2

3

4

5

SFR

[M⊙

yr−

1]

IR late-type (radio)IR early-type (radio), upper limit

0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10z

−11.0

−10.5

−10.0

−9.5

−9.0

−8.5

−8.0

log (

SFR

/MHI)

[yr−

1]

IR late-type (radio)

Figure 3.9: - continued.

Page 59: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

50 chapter 3: The global cold gas content up to z = 0.12

3.6 Discussion and summaryOur stacking results show that galaxies in the green valley are detected with loweramounts of H I than blue galaxies, but unlike red galaxies, they are not completely de-pleted of cold (H I) gas. This result indicates that green valley objects are an intermediatepopulation also from H I point of view.

In our previous paper (Geréb et al. 2013) we used g −r colors to disentangle blue andred objects (vertical line at g − r = 0.7), and obtained an H I detection in red galaxies.From Fig. 3.1 (bottom panel) it is clear that our current green valley objects are part ofthe g − r > 0.7 (red) sample. However, in other color selection scenario (Fabello et al.2011a), green valley objects belong to the NUV - r < 4.5 (blue) sample. This compari-son illustrates that different color selections result in different H I populations, and thecontribution of the green valley seems to play a crucial role.

We do not detect H I in red galaxies at the limit of MHI < 5 × 108 M⊙ and MHI/Lr< 0.02 (M⊙/L⊙) (averaged over four redshift bins). Even though this is a relativelylow detection limit, lower H I masses have been detected before by direct observationsof the SAURON and ATLAS3D samples. Stacking is a promising technique to lower thedetection limit and explore the < 107 M⊙ H I mass regime of galaxies using large samplesof galaxies. This will be made possible by future H I surveys with the next generation ofradio telescopes, e.g. Apertif (Oosterloo et al. 2010b), the Australian Square KilometreArray Pathfinder (ASKAP, DeBoer et al. 2009), and MeerKat (Booth et al. 2009).

There are indications that by the time galaxies reach the green valley, they developbulges and bigger central black holes (Schiminovich et al. 2007). These galaxies are morelikely to host AGN, which phenomena is thought to be able to deplete cold gas reservoirsby feedback processes. Therefore, intriguing is the detection of H I in AGN, where it isexpected that feedback processes deplete the cold gas reservoirs. It seems that, if they aregreen/blue, even galaxies with higher ionization properties (LINERs and optical AGN)do contain cold gas. This suggests that optical AGN are not the (main) reason fordepleting gas reservoirs, or that AGN-driven gas depletion is not an instantaneous effectin galaxies. In agreement with previous studies, our results show that the presence of H Iis better correlated with IR and NUV - r color rather than with ionization properties.

We do not detect any H I gas in radio sources located in the IR early-type region([4.6 µm] - [12 µm] < 2 in the WISE color-color plot) down to the 3-σ detection limit ofMHI < 2 × 109 M⊙ and MHI/Lr < 0.02 (M⊙/L⊙) (averaged over the 4 redshift bins).The lack of H I gas along with the high average radio power of log(P) = 23 W Hz−1

suggest that the radio emission in this population can not originate from star formation.Therefore, radio AGN are likely to be responsible for the radio continuum emission inIR early-type galaxies.

Our estimates for the SFR and SFE agree well with the results of previous studies(Schiminovich et al. 2010). Our results suggest that SF goes hand in hand with the H Iproperties, and in galaxies where cold gas (H I) is present, conditions are favourable for(residual) SF to be seen. Furthermore, it seems to be true for most of the sample thatin galaxies where there is more gas, also the SFR is higher (blue cloud, SF galaxies).However, exceptions can be found, e.g. in radio sources in the IR late-type region in Fig3.7, where small amounts of gas are associated with very efficient star formation. The lackof H I and the high level of residual star formation suggest that these galaxies recentlywent through an intense star-formation period, and this led to a significant depletion

Page 60: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

3.7: Acknowledgements 51

of H I in these galaxies. The radio emission is likely the result of this enhanced starformation activity from the recent past in these sources. We note that three SF galaxiesare located in the QSO/Seyfert region of the color-color plot ([3.4 µm] - [4.6 µm] > 0.5),and for these galaxies the presence of an AGN counterpart can not be excluded.

In Figs. 3.4, 3.5, 3.6 and in Fig 3.7, the H I mass-luminosity ratios do not changesignificantly in function of redshift, suggesting that the global H I content remains rela-tively constant up to z = 0.12. Furthermore, the global SFE displays a similar behaviour,remaining relatively constant in the covered redshift range. In fact, throughout our pa-per we see a good correlation between the presence of H I and star formation properties.As Schiminovich et al. (2010) argue, this can be interpreted as an indications that theH I content and SF are regulated by the same process, e.g. feedback effects, galaxyenvironment.

3.7 AcknowledgementsThe WSRT is operated by the ASTRON (Netherlands Foundation for Research in As-tronomy) with support from the Netherlands Foundation for Scientific Research (NWO).This research made use of the “K-corrections calculator” service available at:http://kcor.sai.msu.ru/

Page 61: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

52 chapter 3: The global cold gas content up to z = 0.12

Page 62: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Baldry, I. K., Glazebrook, K., Brinkmann, J., et al. 2004, ApJ, 600, 681

Baldwin J. A., Phillips M. M., Terlevich R., 1981, PASP, 93, 5

Best, P. N., & Heckman, T. M. 2012, MNRAS, 421, 1569

Bigiel, F., Leroy, A. K., Walter, F., et al. 2011, ApJ, 730, L13

DeBoer, D. R., Gough, R. G., Bunton, J. D., et al. 2009, IEEE Proceedings, 97, 1507

Booth, R. S., de Blok, W. J. G., Jonas, J. L., & Fanaroff, B. 2009, arXiv:0910.2935

Chang, T.-C., Pen, U.-L., Bandura, K., & Peterson, J. B. 2010, Nature, 466, 463

Chilingarian, I. V., Melchior, A.-L., & Zolotukhin, I. Y. 2010, MNRAS, 405, 1409

Chilingarian, I. V., & Zolotukhin, I. Y. 2012, MNRAS, 419, 1727

Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ, 115, 1693

Delhaize J., Meyer M. J., Staveley-Smith L., Boyle B. J., 2013, MNRAS, 433, 1398

Fabello S., Catinella B., Giovanelli R. et al., 2011, MNRAS, 411, 993

Fabello S., Kauffmann G., Catinella B. et al., 2011, MNRAS, 416, 1739

Geréb, K., Morganti, R., Oosterloo, T. A., Guglielmino, G., & Prandoni, I. 2013, A&A,558, A54

Grossi, M., di Serego Alighieri, S., Giovanardi, C., et al. 2009, A&A, 498, 407

Hopkins, A. M., & Beacom, J. F. 2006, ApJ, 651, 142

van der Hulst J. M., van Albada T. S., Sancisi R., 2001, ASPC, 240, 451

Kauffmann G., et al., 2003, MNRAS, 346, 1055

Kewley L. J., Heisler C. A., Dopita M. A., Lumsden S., 2001, ApJS, 132, 37

Lagos, C. D. P., Baugh, C. M., Lacey, C. G., et al. 2011, MNRAS, 418, 1649

Page 63: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

54 BIBLIOGRAPHY

Lah, P., Chengalur, J. N., Briggs, F. H., et al. 2007, MNRAS, 376, 1357

Lah, P., Pracy, M. B., Chengalur, J. N., et al. 2009, MNRAS, 399, 1447

Madau, P., Ferguson, H. C., Dickinson, M. E., et al. 1996, MNRAS, 283, 1388

Martig, M., Bournaud, F., Teyssier, R., & Dekel, A. 2009, ApJ, 707, 250

Martin D. C., et al., 2005, ApJ, 619, L1

Meyer M. J., et al., 2004, MNRAS, 350, 1195

Martin, A. M., Papastergis, E., Giovanelli, R., et al. 2010, ApJ, 723, 1359

Morganti R., et al., 2006, MNRAS, 371, 157

Obreschkow, D., & Rawlings, S. 2009, ApJ, 696, L129

Oosterloo T., et al., 2010, MNRAS, 409, 500

Oosterloo, T., Verheijen, M., & van Cappellen, W. 2010, ISKAF2010 Science Meeting

Popping, G., Somerville, R. S., & Trager, S. C. 2013, arXiv:1308.6764

Prochaska, J. X., & Wolfe, A. M. 2009, ApJ, 696, 1543

Rao, S. M., Turnshek, D. A., & Nestor, D. B. 2006, ApJ, 636, 610

Sadler E. M., Ekers R. D., Mahony E., Mauch T., Murphy T., 2013, arXiv,arXiv:1304.0268

Sarzi M., et al., 2010, MNRAS, 402, 2187

Sault R. J., Teuben P. J., Wright M. C. H., 1995, ASPC, 77, 433

Schiminovich, D., Wyder, T. K., Martin, D. C., et al. 2007, ApJS, 173, 315

Schiminovich D., et al., 2010, MNRAS, 408, 919

Serra P., et al., 2012, MNRAS, 422, 1835

Strateva, I., Ivezić, Ž., Knapp, G. R., et al. 2001, AJ, 122, 1861

Toribio, M. C., Solanes, J. M., Giovanelli, R., Haynes, M. P., & Martin, A. M. 2011,ApJ, 732, 93

Verheijen M., van Gorkom J. H., Szomoru A., Dwarakanath K. S., Poggianti B. M.,Schiminovich D., 2007, ApJ, 668, L9

Walter F., Brinks E., de Blok W. J. G., Bigiel F., Kennicutt R. C., Jr., Thornley M. D.,Leroy A., 2008, AJ, 136, 2563

Wright E. L., et al., 2010, AJ, 140, 1868

Wyder T. K., et al., 2007, ApJS, 173, 293

Page 64: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

BIBLIOGRAPHY 55

Zwaan M. A., Meyer M. J., Staveley-Smith L., Webster R. L., 2005, MNRAS, 359, L30

Yi, S. K., Yoon, S.-J., Kaviraj, S., et al. 2005, ApJ, 619, L111

York D. G., et al., 2000, AJ, 120, 1579

Freudling, W., Staveley-Smith, L., Catinella, B., et al. 2011, ApJ, 727, 40

Page 65: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

56 BIBLIOGRAPHY

Page 66: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Chapter 4Probing the gas content of radiogalaxies through HI absorptionstacking

– K. Geréb, R. Morganti, T. Oosterloo –Accepted to Astronomy & Astrophysics, 2014

AbstractUsing the Westerbork Synthesis Radio Telescope, we carried out shallow H I absorption obser-vations of a flux-selected (S1.4 GHz > 50 mJy) sample of 93 radio Active Galactic Nuclei (AGN)which have available SDSS (Sloan Digital Sky Survey) redshifts between 0.02 < z < 0.23. Ourmain goal is to study the presence of gas in radio sources down to S1.4 GHz flux densities notsystematically explored before using, for the first time, stacking of absorption spectra of extra-galactic H I. Despite the shallow observations, we obtained a direct detection rate of ∼ 29%,comparable with deeper studies of radio galaxies. Furthermore, detections are found at ev-ery S1.4 GHz flux level, showing that H I absorption detections are not biased toward brightersources. The stacked profiles of detections and non-detections reveal a clear dichotomy in thepresence of H I, with the 27 detections showing an average peak τ = 0.02 corresponding to N(H I)∼ (7.4±0.2)×1018 (Tspin/cf) cm−2, while the 66 non-detections remain undetected upon stack-ing with a peak optical depth upper limit τ < 0.002 corresponding to N(H I) < (2.26±0.06)×1017

(Tspin/cf) cm−2 (using a FWHM of 62 km s−1, derived from the mean width of the detections).Separating the sample into compact and extended radio sources, increases the detection rate,optical depth and FWHM for the compact sample. Among these two groups of objects, thedichotomy for the stacked profiles of detections and non-detections still holds. We argue thatorientation effects connected to a disk-like distribution of the H I can be partly responsible forthe dichotomy that we see in our sample. However, orientation effects alone can not explain allthe observational results, and some of our galaxies must be genuinely depleted of cold gas. Afraction of the compact sources in the sample are confirmed, by previous studies, as likely young

Page 67: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

58 chapter 4: HI absorption stacking of radio galaxies

radio sources (Compact Steep Spectrum and Gigahertz Peaked Spectrum sources). These showan even higher detection rate of 55%. Along with their high integrated optical depth and widerprofile, this reinforces the idea that young radio AGN are embedded in a medium rich in atomicgas. Part of our motivation is to probe the presence of faint H I outflows at low optical depthusing stacking. However, the stacked profiles do not reveal any significant blueshifted wing. Weare currently collecting more data to investigate the presence of outflows. The results presentedin this paper are particularly relevant for future surveys in two ways. The lack of bias towardbright sources is particularly encouraging for the search of H I in sources with even lower radiofluxes planned by such surveys. The results also represent a reference point for search of H Iabsorption at higher redshifts.

4.1 IntroductionThe nuclear activity in galaxies is regulated by the availability of gas in the central regionand by the conditions that make it possible for the gas to cool and feed the central blackhole. Our knowledge of the gas content in nearby early-type galaxies (the typical hostsof radio-loud AGN) has increased substantially in recent years. Projects like WSRT-SAURON (Morganti et al. 2006; Oosterloo et al. 2010a) and ATLAS3D (Serra et al. 2012;Young et al. 2011; Davis et al. 2013) have characterized the presence and properties ofH I and molecular gas in nearby early-type galaxies. For example, detailed studies of theH I content of early-type galaxies in the nearby Universe has shown that H I is detectedin emission in about 40 percent of early-type galaxies in the field (Serra et al. 2012).However, because these samples are limited to very nearby galaxies, only a minority ofthe objects host an AGN, and in particular a radio AGN.

In radio-loud AGN the presence and kinematics of the gas can be explored via H Iabsorption, and this has been done for a long time (e.g. Roberts 1970; De Young et al.1973; Dickey 1982; Shostak et al. 1983). Such observations of H I absorption are typi-cally done using strong radio sources as a bright background against which the gas canbe traced. A number of such studies have recently provided a better understanding of theH I absorption properties of radio AGN (van Gorkom et al. 1989; Morganti et al. 2001;Vermeulen et al. 2003; Gupta et al. 2006; Curran & Whiting 2010; Emonts et al. 2010).These samples show a detection rate of H I between a few percent up to 40%. Absorptionwas found to trace a variety of structures showing that H I can be present in regularlyrotating disks (Serra et al. 2012; Emonts et al. 2010; Gallimore et al. 1999; Allison et al.2014), in infalling clouds that have been associated with the feeding mechanisms of thecentral black hole (van Gorkom et al. 1989; Morganti et al. 2009) and in outflows trac-ing the interactions between the jets and the surrounding medium (Morganti et al. 1998,2005, 2013). Thus, the complexity of the H I kinematics in AGN suggests that gas canplay many different roles. One interesting finding of these studies is that there appearsto be a trend between the detection rate and the type of radio source, and in particularits evolutionary stage. Compact Steep Spectrum (CSS) and Gigahertz Peaked Spectrum(GPS) sources have been proposed to represent young ( <∼ 104 yr) radio AGN, basedon spectral ageing analysis and lobe expansion speed measurements (Fanti et al. 1995;Readhead et al. 1996; Owsianik & Conway 1998). A high detection rate of H I absorp-tion in these sources has been reported by a number of studies (Pihlström et al. 2003;Gupta et al. 2006; Emonts et al. 2010; Chandola et al. 2010). This has been interpreted

Page 68: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.2: Sample selection and observations 59

as evidence for a relation between the recent triggering of the AGN activity and thepresence of H I gas.

Finally, the presence of a strong interaction between the radio jets and the interstellarmedium has been proposed for many of these young sources. Although it is not yet clear inhow many cases this would result in the frustration of the radio source (van Breugel et al.1984; Fanti et al. 1990; De Young 1993; Pihlström et al. 2003). Outflows are often seenin the kinematics of the H I in these objects, as traced by blueshifted, broad wings ofabsorption (see e.g. Morganti et al. 2013 and ref. therein). These examples show that,at least in some cases, the jets are clearing their way through the surrounding gas.

It is clear that a better understanding of the gas properties of different types of radiosources (e.g. compact, extended) is crucial in explaining the observed characteristics ofAGN. It is also of interest to investigate whether H I has a similar detection rate andmorphology in early-type galaxies with/without AGN and hence to learn more aboutthe interplay between AGN and the surrounding gas. Previous studies of H I in AGNare limited in number and sensitivity, not ideally suited to study weak structures likeoutflows. Some of these limitations will be overcome by surveys performed with new tele-scopes, such as Apertif (Oosterloo et al. 2010b), the Australian Square Kilometre ArrayPathfinder (ASKAP, DeBoer et al. 2009), and MeerKat (Booth et al. 2009). However,even with current telescopes progress can be made using stacking techniques. H I emissionstacking techniques have been used effectively to detect the global signal from faint H Iemitters (Lah et al. 2007; Verheijen et al. 2007; Lah et al. 2009; Fabello et al. 2011a,b;Delhaize et al. 2013; Geréb et al. 2013). More recently, Murray et al. (2014) performeda spectral stacking analysis using H I absorption residual spectra to detect warm neutralmedium in the Milky Way.

Here we present the results of a snap-shot survey we have performed with the maingoal of observing as many objects as possible, to a continuum flux density limit thatis lower than the earlier studies. For the first time, we stack extragalactic data of H Iabsorption to lower the optical depth detection limit, and to study the gas properties ofdifferent types of AGN. In an upcoming paper we will present the data and the discussionof the single objects and their H I profile parameters.

In this paper the standard cosmological model is used, with parameters ΩΛ = 0.3, Λ= 0.7 and H0 = 70 km s−1 Mpc−1.

4.2 Sample selection and observationsFor our sample selection, we used the cross-correlation of the Sloan Digital Sky Survey(SDSS, York et al. 2000) and Faint Images of the Radio Sky at Twenty-cm (FIRST,Becker et al. 1995) catalogs. In total, 120 sources were selected with peak flux S1.4 GHz> 50 mJy in the FIRST catalog, in the redshift range 0.02 < z < 0.231, above thedeclination δ > 15 deg. The interval 0.136 < z < 0.175 was excluded because it isstrongly affected by radio frequency interference (RFI). Additional information on theradio properties of the sample is extracted from the 1.4 GHz NRAO VLA Sky Survey(NVSS, Condon et al. 1998).

The SDSS spectra were visually inspected to ensure that accurate optical redshifts canbe derived for the galaxies. For nine sources the spectra appear without any well-defined1 We start from redshift z = 0.02 in order to exclude nearby, well-studied galaxies.

Page 69: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

60 chapter 4: HI absorption stacking of radio galaxies

Figure 4.1: The 3-σ rms noise distribution in the 101 H I data cubes.

emission/absorption lines, meaning that for these sources the optical line identification,and hence the redshift in SDSS, may not be reliable. For this reason, we excluded thesesources from our sample. For the rest of the sample, the redshifts errors extracted fromSDSS are lower than ∼ 60 km s−1.

The observations were carried out with the Westerbork Synthesis Radio Telescope(WSRT) in the period Nov. 2012 - Nov. 2013. Using the WSRT East-West array, theshort integration time of 4 hours for each source means that the synthesised beam ofthe data cubes is very elongated, with a typical angular size of 75 × 11 arcsec. Theobservational setup consists of 1024 channels in a bandwidth of 20 MHz. We discard 10more objects because their spectra are dominated by RFI. As a result, our final AGNsample includes 101 sources.

The data were reduced using the MIRIAD package (Sault, Teuben, & Wright 1995).The H I data cubes were Hanning smoothed over 3 channels, yielding a final velocityresolution of ∼ 16 km s−1. The 3-σ rms noise distribution of the H I data cubes arepresented in Fig. 4.1.

We also created continuum images using the line-free channels to measure the con-tinuum flux density of the sources. The corresponding radio power distribution rangesbetween 1023 - 1026 W Hz−1.

4.2.1 Characterization of the AGN sampleThe shortlisted target sample was separated into compact and extended radio sources.Our sample is small enough that this classification could be done visually. However,we also explored automated procedures based on objective parameters, which could beused in larger surveys of radio sources in the future. We found the best method to bethe classification based on the NVSS major-to-minor axis ratio vs. the FIRST peak-

Page 70: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.2: Sample selection and observations 61

0.2 0.4 0.6 0.8 1.0FIRST peak/integrated flux

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8NVSS m

ajo

r/m

inor axis

Figure 4.2: Radio morphological classification of the observed sample (101 sources).Red circles indicate compact sources, and extended sources are marked by blue squares.H I detections (both compact and extended) are marked by filled symbols, while emptymarkers stand for non-detections.

to-integrated flux ratio. As illustrated in Fig. 4.2, and after a final visual inspection,we find that except for one source, all compact sources are located in the region withNVSS major-to-minor axis ratio < 1.1 and FIRST peak-to-integrated flux ratio > 0.9(marked by the box). Most of the extended sources have NVSS major-to-minor axisratio > 1.1 and FIRST peak-to-integrated flux ratio < 0.9. However, even this methodis not perfect and we identify a few more extended sources inside the box, which seemto be misclassified because these sources tend to show strong core emission compared tothe faint extended emission coming from the lobes. Furthermore, following a literaturesearch of the detections we find that one source is classified as an extended source in ourselection, however it is a 6 kpc Compact Steep Spectrum (CSS) source (Saikia & Gupta2003). This object was misclassified because the structure of the radio source is slightlyresolved at the 5 arcsec resolution of FIRST. Thus, this source is added to the sample ofcompact sources.

The observed sources are a mix of different types of AGN. Because our sample issolely flux selected, the sample contains radio galaxies (the majority of the sources),QSOs (Quasi-Stellar Objects), Seyfert galaxies, gas-rich mergers. Radio galaxies aretypically found in red (g − r > 0.7) early-type hosts (Bahcall et al. 1997; Best et al.2005), while the other three groups are typically blue (g − r < 0.7) objects and richerin H I (Geréb et al. 2013). In order to make the sample more homogeneous, we haveexcluded blue galaxies with g − r < 0.7 from our analysis. The color information neededfor this exercise, i.e. the g and r band optical magnitudes, are extracted from the SDSSdatabase. As a result of this selection, we exclude gas-rich mergers, QSO-s, Seyfert

Page 71: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

62 chapter 4: HI absorption stacking of radio galaxies

Figure 4.3: Peak optical depth and N(H I) distribution of the detections (filled bars)and 3-σ upper limits (empty bars).

galaxies, such that the remaining sample of 93 AGN mainly consists of radio galaxies.

4.3 Results

4.3.1 H I detectionAfter the above selection, our sample contains 93 AGN with g − r > 0.7, of which wedetected 27 sources above the 3-σ level. The remaining 66 were non-detections. A firstresult is therefore that, despite the short integration time, we obtained a high detectionrate of ∼ 29%, comparable with deeper studies of radio galaxies.

Page 72: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.3: Results 63

The absorbed flux depends both on the optical depth and on the covering factor (cf )of the H I, such that Sabs = Scf(1 − e−τ ). The column density (in cm−2) is related tothe integrated optical depth (in km s−1) by N(H I) = 1.823 × 1018(Tspin/cf)

∫τ(v)dv

(Wolfe & Burbidge 1975). As we mention in Sec. 4.2, we derive the optical depth foreach spectrum using the WSRT continuum flux. As illustrated in Fig. 4.3, the detectedτ distribution is very broad, with the peak τ of the direct detections ranging between0.2% and 30%. The corresponding column densities are N(H I) = 5 × 1017 (Tspin/cf)cm−2 and N(H I) = 3 ×1019 (Tspin/cf) cm−2 respectively. In Sec 4.3.2 we expand onthese results using stacking techniques.

In Fig. 4.4 we compare the flux density distribution of detections and non-detectionsusing the Kolmogorov-Smirnov (K-S) test. The probability that the two distributions aredifferent is only 10%, implying that, statistically, detections and non-detections have asimilar flux distribution. H I detections are found down to the lowest fluxes in our sample.This is an important result, because it shows that our detections are not biased towardbrighter sources and suggests that systematic H I absorption studies can be carried outat even lower radio fluxes.

The detections in the g − r > 0.7 AGN sample reveal a wide variety of H I profiles. InFig. 4.5 we show that we detect profiles where the peak of the absorption is redshiftedup to +300 km s−1, and blueshift up to −200 km s−1 compared to the systemic velocity.The analysis of the individually detected profiles shows that the median FWHM of theH I main absorption components is ∼ 62 km s−1. A detailed discussion of the analysis ofthe complex profiles will be given in an upcoming paper presenting the full dataset. Asa general conclusion, the diversity of the detected profiles suggests that the kinematicsof H I is quite complex in radio AGN.

4.3.2 H I stacking: the H I distribution of AGN at low opticaldepth

One of the main goals of the study is to explore the presence of H I at lower opticaldepth by using stacking techniques. We adapted the script used for H I emission studiesin Geréb et al. (2013) to perform stacking of H I absorption. The spectra are extractedat the peak of the continuum source, assuming that the sources are unresolved. Thestacking is done in optical depth, by aligning the spectra based on their SDSS redshift.

As shown in Fig. 4.6, stacking a large number (66) of undetected sources results inan average non-detection with a relatively sensitive 3-σ upper limit of τ < 0.002 (solidline), while the sample of the detected sources gives an average τpeak = 0.02 (dashedline). From Gaussian fitting we measure FWHM = 203 ± 7 km s−1 in the profile ofstacked detections. The integrated optical depth of this profile is relatively high, N(H I)= (7.4 ± 0.2) × 1018 (Tspin/cf) cm−2. Using the median FWHM of ∼ 62 km s−1 andthe 3−σ upper limit τ = 0.002, we derive N(H I) < (2.26 ± 0.06) × 1017 (Tspin/cf) cm−2

for the upper limit of the stacked non-detections. In Fig. 4.7, the red line marks theupper limit reached after stacking, showing the sensitivity improvement compared tosingle detections and upper limits.

We have checked that the width of the co-added profiles is not affected by redshiftinaccuracy during stacking in the following way. Maddox et al. (2013) show that for H Iemission, if the redshift errors are smaller than the median width of the co-added H Iprofiles, the resulting width is not much affected by redshift inaccuracy. In our sample

Page 73: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

64 chapter 4: HI absorption stacking of radio galaxies

101 102 103 104

S1.4 GHz [mJy]

0.0

0.2

0.4

0.6

0.8

1.0cu

mulative fraction

DetectionsNon-DetectionsD

Figure 4.4: Kolmogorov-Smirnov test for the flux distribution of detections (red line)and non-detections (black line). D is the maximum distance of the set of distancesbetween the cumulative fraction of detections and non-detections. We measure D = 0.117for the sample of 27 detections and 66 non-detections.

−200 −100 0 100 200 300 ∆v (km/s)

0

1

2

3

4

5

6

Count

Detections

Figure 4.5: Velocity offset of the H I peak from the systemic velocity in the 27 detections.

the 3-σ SDSS redshift errors are lower than the median FWHM = 62 km s−1 (see Sec4.2), suggesting that the width of the stacked profile is not affected by redshift inaccuracy.

Page 74: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.3: Results 65

−1000 −500 0 500 1000Velocity (km/s)

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02τ

Detections, stacking with z SDSSNon-DetectionsDetections, stacking with HI peak

Figure 4.6: The sample of 93 red AGN, the stacked profile of 27 detections (dashedand dotted lines, see explanation in the legend) and 66 non-detections (solid line)

However, Maddox et al. (2013) also show that the error distribution of the SDSS redshiftshas broad, non-Gaussian tails to large values, which may have an effect on the stacking.

In Fig. 4.5 we show that a significant fraction of our sources are offset from thesystemic velocity. This suggests that the width of the stacked profile can be affectedby the velocity offset distribution of the detections. In order to test the possible effectof this, instead of using the SDSS redshift, we stack the H I detections by shifting thespectra with the frequency of the H I peak from Fig. 4.5. The resulting profile is plottedin Fig. 4.6 with a dotted line, showing that both the width and the peak τ of the stackedprofile change, however, as expected, the integrated optical depth remains the same.From Gaussian fitting we measure FWHM = 107 ± 4 km s−1. The stacked FWHM ishigher than the median FWHM = 62 km s−1 of the individual H I detections (see Sec.4.3.1). However, as we mentioned earlier in Sec. 4.3.1, only the FWHM of the main H Icomponent is estimated in the individual detections because of the complexity of theH I profiles. Therefore, the larger stacked FWHM must be the result of additional H Icomponents. At the 3-σ significance level, a redshifted feature is present in the stackedspectra at ∼ +250 km s−1, however after inspecting the profiles we find that this featureis dominated by only a few (about 2) sources with complex H I morphology.

The interesting result from our stacking experiment is that the average H I contentof the non-detected sources is well below that of the detected sources, thus a strongdichotomy is found in the presence of H I between detected sources and non-detections.The dichotomy is also present in Fig. 4.7, where some sources have fairly high H Iintegrated optical depths, however in many objects the N(H I) is much lower. As alreadymentioned in Sec. 4.3.1 and shown in Fig 4.3, the individually detected τ distributionin our sample is very broad, which is consistent with the broad range of H I masses andcolumn densities that is seen in ATLAS3D. Hence, we cannot disentangle whether the

Page 75: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

66 chapter 4: HI absorption stacking of radio galaxies

Figure 4.7: The H I column density versus the 1.4 GHz flux density for detections (filledcircles) and non-detections (empty triangles). The red line marks the N(H I) upper limitof non-detections after stacking.

0.2 0.4 0.6 0.8 1.0FIRST peak/integrated flux

0.0

0.2

0.4

0.6

0.8

1.0

Cumulative fraction

DetectionsNon-DetectionsD

Figure 4.8: Compactness distribution of detections and non-detections. We measure D= 0.319 for 27 detections and 66 non-detections.

Page 76: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.3: Results 67

dichotomy in the stacked profiles is due to a bimodal or a power-law distribution of theH I column densities. This is, in fact, a long-standing problem that has been investigatedfor early-type galaxies for the past 30 years (Knapp et al. 1985).

Van Gorkom et al. (1989) - albeit using fewer objects and not using stacking techniques- already estimated the expected number distribution of τ in a sample of radio galaxiesand concluded that, in general, only about 30 percent of the radio galaxies (with compactcores) have H I with τ > 0.01. Our result expands on this as we reach much lower values ofτ after stacking. The fact that the stacked spectrum of those sources that are undetectedindividually does not show a detection implies that more sensitive observations wouldnot yield many more new individual detections. Therefore we expect ∼ 30% to be arepresentative detection rate for H I in radio AGN.

One of the goals of our absorption stacking is to investigate the presence and signa-tures of blueshifted wings, indicative of outflows driven by jet-cloud interactions. Thepeak optical depth of such absorption features is very low, below 1% in most of thecases (see references from the Introduction). Therefore only in the brightest objects itis possible to reach such detection limits with current observations. Stacking makes itpossible to decrease the optical depth limit and to test whether the detection of outflowsis a matter of sensitivity.

4.3.3 H I and radio morphologyDifferent types of AGN, and in particular young radio sources, have been suggestedto show different detection rates of H I and to be surrounded by a different gaseousmedium. Thus, we have separated our sample into compact and extended radio sourcesas described in Sec. 2. In Fig. 4.8, the probability that the two data samples come fromthe same distribution is 2.5%. This result implies that, statistically, detections and non-detections have a different compactness distribution. We find that compact sources (withg − r > 0.7) show a ∼ 42% detection rate (20 detections, 28 non-detections), whereasonly ∼16% of extended sources are detected in H I (7 detections, 38 non-detections).

From our compact sources 11 are identified with a match in the COmpact RAdiosources at Low redshift (CORALZ) sample (Snellen et al. 2004; de Vries et al. 2009),and VLBI observations assure us that CORALZ sources are intrinsically small (CSS andGPS), likely young AGN. For CORALZ sources we find an even higher detection rate of55%.

The result of stacking in Fig. 4.9 shows that the optical depth dichotomy of detectionsand upper limits also holds for compact and extended sources, in agreement with thedichotomy between detections and non-detections reported in Sec. 4.3.2. Furthermore,stacking reveals dissimilar profiles both regarding the peak optical depth and the width ofthe profiles. Gaussian fitting yields a FWHM of 203 ± 7 km s−1 for the stack of compactsources (similar to the total sample of detections), and 120 ± 13 km s−1 for the extendedsources. The corresponding column density in compact sources is N(H I) = (8.5 ± 0.3) ×1018 (Tspin/cf) cm−2, whereas we detect lower columns of N(H I) = (2.9 ± 0.2) × 1018

(Tspin/cf) cm−2 for extended sources.In order to test the effect of redshift differences on the measured profile widths, we

repeat the stacking exercise from Sec. 4.3.2 by shifting the spectra with the H I peak.The stacked profiles become narrower, however the FWHM of the profiles of the compactand extended sources remains different. From Gaussian fitting we find FWHM = 115 ±

Page 77: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

68 chapter 4: HI absorption stacking of radio galaxies

−1000 −500 0 500 1000Velocity (km/s)

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02τ

Compact detections, stacking with z SDSSCompact non-detectionsCompact detections, stacking with HI peak

−1000 −500 0 500 1000Velocity (km/s)

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02

τ

Extended detections, stacking with z SDSSExtended non-detectionsExtended detections, stacking with HI peak

Figure 4.9: The stacked profiles of compact (20 detections, 28 non-detections) andextended (7 detections, 38 non-detections) sources above and below respectively. Detec-tions are indicated by dashed and dotted lines like in Fig. 4.6, and non-detections bysolid lines.

4 km s−1 for compact sources and FWHM = 75 ± 5 km s−1 for the extended sample.However, the integrated optical depth of compact sources is hardly affected, we measureN(H I) = (8.4 ± 0.2) × 1018 (Tspin/cf) cm−2. For extended sources the N(H I) increasesby 34% to N(H I) = (3.9±0.2) × 1018 (Tspin/cf) cm−2. The difference in N(H I) betweencompact and extended is still a factor of 2, meaning that our conclusions do not change.

From our WSRT observations we do not know the location of the absorption, northe structure of the radio continuum. Higher resolution FIRST observations (5 arcsecangular resolution) are a better measure of the core brightness; therefore, we repeat the

Page 78: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.3: Results 69

stacking exercise by deriving τ from the FIRST peak flux density (and by shifting thespectra with the optical redshift). For compact sources we measure N(H I) = (8.1 ± 0.2)× 1018 (Tspin/cf) cm−2 (5% change) and for extended sources N(H I) = (4.8 ± 0.3) ×1018 (Tspin/cf) cm−2 (65% change) in the stacked profiles. These differences are a resultof the different resolution of our WSRT observations and the FIRST survey, and theyare more prominent for the extended sources, as expected. However, even if all of theH I absorption is against the core, the difference between the integrated optical depth ofcompact and extended sources is still a factor of ∼2.

The role of cf , i.e. the gas fraction covered by the radio source, can also be importantfor the interpretation of the observed optical depth. Curran et al. (2013) pointed out thatthe covering factor (cf) is proportional to the τobs, and inversely proportional with thesize of the radio source. Hence, a systematic difference in the covering factor of compactand extended sources could be influencing the measured optical depths.

In principle, the integrated optical depth difference between compact and extendedsources in Fig. 4.9 could be explained by a systematic difference in the covering factor.For example, only 30 percent of the continuum may be covered by the H I screen (cf = 0.3)in extended sources, while 100% (cf = 1) may be covered in compact sources. VLBIstudies available in the literature can be inspected to verify if such a trend exists. Thecovering factor is close to cf = 1 in the compact source 1946+708 (Peck et al. 1999),however one can also find examples of compact sources with lower covering factor of cf ∼0.2 in B2352+495 (Araya et al. 2010) and in 4C 12.5 (Morganti et al. 2013). In extendedsources cf = 1 is estimated in the re-started source 3C 293 (Beswick et al. 2004). However,a lower covering factor of a few percent is measured in NGC 315 (Morganti et al. 2009),and cf = 0.5 is estimated in the extended source 3C 305 (Morganti et al. 2005). Thus,from VLBI observations it is not at all clear that such a systematic difference couldrealistically be present, as both compact and extended sources show a large range ofcovering factors. Thus, we conclude that at least part of the difference in optical depthbetween compact and extended sources is a real effect. However, VLBI observationsof homogeneous AGN samples are needed to measure the cf and verify the differencebetween compact and extended sources at high resolution.

4.3.4 The nature of the H I absorbing systemsWe can use the detailed studies of the H I content of local early-type galaxies as a referencesample to investigate the nature of the H I gas in our sample. Serra et al. (2012) detectedH I emission in about 40% of ATLAS3D early-type galaxies in the field. In 25% of the totalsample, i.e. in roughly half of the detected cases, the gas is distributed in a flatteneddisk/ring morphology. In the other half of the detected sample, H I has an unsettledmorphology.

In radio AGN the detection of H I may strongly depend on the relative orientationof the H I gas and the background continuum source. If some of the H I structures haveflattened morphology, the detection of H I in these sources would be limited by orientationeffects. To test this effect, we have inspected the inclination of our galaxies using theminor-to-major axis ratio in the r band (expAB_r), and using the optical images of thegalaxies from the SDSS. The fact that the most highly inclined objects at b/a < 0.6 inFig. 4.10 all show H I absorption suggests that orientation plays a role and that the H Iis likely in a flattened structure, such as a disk, in many sources. In the optical images

Page 79: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

70 chapter 4: HI absorption stacking of radio galaxies

we indeed see the presence of (nearly) edge-on disk morphologies in these objects.However, we do detect some galaxies at b/a > 0.6. H I disks with low inclination are

less likely to be detected, therefore the detection of H I in the high b/a regime impliesthat at least in some galaxies the H I is not in a flattened structure, but it possibly has amore unsettled distribution. This is also supported by the fact that flattened structuresin the b/a < 0.6 region all have relatively high integrated optical depth, while at higherb/a detections span a wider N(H I) range.

Considering the 25% detection rate of disks in ATLAS3D galaxies, the 16% detectionrate in the sample of extended sources is comparable to what one would expect to seeif all detections of extended sources have a disk morphology. However, compact sourceshave a higher detection rate of 42%, which is about the same as the total detection ratefor H I emission in the ATLAS3D sample. Given that the detection of H I absorptionrequires a certain geometry (i.e. the H I has to be in front of the continuum source), thissuggest that compact sources are at least as H I-rich as the general population of early-type galaxies, but the total detection rate (H I emission and absorption) could be evenhigher in compact sources. The larger FWHM and higher H I detection rate of compactsources suggests that their H I gas often has more spherical morphology, less affectedby orientation, e.g. the H I can be present in an unsettled distribution or thicker disks.Furthermore, the high integrated optical depths in compact sources, and in particularthe young radio sources in our sample, indicates that young AGN are particularly rich ingas. However, due their small size, compact sources might have higher covering factorsand therefore higher observed optical depths than extended sources (Curran et al. 2013).Hence, high-resolution observations are needed to estimate the effect of the coveringfactor on the measured optical depth.

Our stacking exercise does not reveal a significant detection of outflowing gas at ourcurrent detection limit. As we will discuss in a forthcoming paper, about 10% of thedirect detections show a blueshifted wing, possibly due to an outflow with a typical peakoptical depth of τ ∼ 0.01. The 10% detection rate implies that we will need to reach atleast a sensitivity of τ ∼ 0.001 in order to detect the contribution of such outflows in thestacked spectrum. This is below our detection limit and, therefore, it may not be toosurprising that we do not see a blue wing in the stacked profile. Upcoming surveys withthe SKA Pathfinders will result in a few hundred thousand new H I detections, most ofwhich will be above z = 0.1. Radio AGN are more numerous in the distant Universe,therefore we can expect to have new detections of H I absorbers associated with radiocontinuum sources. Furthermore, surveys specifically designed for H I absorption search,e.g. the ASKAP FLASH (The First Large Absorption Survey in H I, Johnston et al.2008), will deliver several hundred new detections, allowing one to reach the necessarysensitivity to detect, statistically, the jet-driven H I outflows.

4.4 Conclusions and implications for the futureIn this paper we carried out a systematic study of the H I properties of radio AGN.Despite our shallow observations, we obtain a detection rate of 30 percent, similar todeeper studies. Our sample is larger than previous studies, allowing us to carry out H Iabsorption stacking and to confirm that in the low-τ limit 30 percent is a representativeH I detection rate for the general population of radio AGN. We find a dichotomy in thepresence of H I in the sense that even when a large number of spectra are averaged,

Page 80: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

4.4: Conclusions and implications for the future 71

Figure 4.10: Column density vs. the minor-to-major axis ratio from the exponential fitin r band for detections and non-detections. The red line marks the N(H I) upper limitof non-detections after stacking.

galaxies that do not show H I absorption in their individual spectra remain undetectedto an average column density of N(H I) < (2.26 ± 0.06) × 1017 (Tspin/cf) cm−2. Weargue that in many galaxies, the H I must be in a flattened structure so that orientationeffects play a role. This result is in good agreement with Curran & Whiting (2010), whosuggest that galactic disks are the bulk of H I absorption in all types of AGN. However,orientation effects alone can not fully explain the dichotomy that we see in our sample,suggesting that some fraction of our galaxies must be depleted of cold gas.

Upcoming surveys will observe AGN over a large flux density range. Our resultssuggest that the detection rate does not depend on the apparent flux of a source andthis has positive implications for future, deeper surveys. These large-area surveys willuncover a very large number of H I absorptions systems. According to the 1.4 GHzradio luminosity function, star forming galaxies become dominant at radio power lowerthan < 1023 W/Hz (Mauch & Sadler 2007), therefore at lower fluxes we expect a mix ofstar-forming/AGN populations.

The detection of H I absorption highly depends on the strength of the underlying con-tinuum; therefore, it is possible to detect absorption up to high redshift, if the continuumis strong enough. Unlike H I absorption, H I emission studies are limited by sensitivityin the higher redshift Universe. However, if emission and absorption are tracing similarmorphological structures, H I absorption studies can be just as efficiently used to findcold gas not just in AGN, but also in star-forming galaxies at higher redshift. The in-creased number of sources will provide enough data to perform H I stacking experimentsand, hence, to probe the highest redshift regime of the observed radio sky at low opticaldepth. Thus, even though H I absorption only traces the cold gas component with Tspinup to a few × 1000 K, it can still provide important information (including redshift) for

Page 81: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

72 chapter 4: HI absorption stacking of radio galaxies

the detected objects.Compact sources show higher detection rates, optical depths and FWHM than ex-

tended sources, strongly suggesting that different gas conditions exist in these two typesof radio sources; however, high resolution observations and a better measure of the cov-ering factor are needed to confirm this result. It seems that a large fraction of compactsources reside in a gas rich environment, and the nuclear activity in most of the youngAGN is connected with the presence of unsettled gas.

In a forthcoming paper we will publish the details of our H I detections and the AGNsample. We will explore in more detail the parameter space of our H I profiles (e.g.blueshift/redshift, width, asymmetries) in relation with radio source and host galaxyproperties. We aim to understand if (and to what extent) the properties of the detectionsare related to the strength and morphology of the radio AGN.

4.5 AcknowledgementsThe WSRT is operated by the ASTRON (Netherlands Foundation for Research in As-tronomy) with support from the Netherlands Foundation for Scientific Research (NWO).

RM gratefully acknowledge support from the European Research Council under theEuropean Union’s Seventh Framework Programme (FP/2007-2013) /ERC AdvancedGrant RADIOLIFE-320745.

Page 82: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Allison, J. R., Curran, S. J., Emonts, B. H. C., et al. 2012, MNRAS, 423, 2601

Allison, J. R., Sadler, E. M., & Meekin, A. M. 2014, arXiv:1402.3530

Araya, E. D., Rodríguez, C., Pihlström, Y., et al. 2010, AJ, 139, 17

Bahcall, J. N., Kirhakos, S., Saxe, D. H., & Schneider, D. P. 1997, ApJ, 479, 642

Becker, R. H., White, R. L., & Helfand, D. J. 1995, ApJ, 450, 559

Best, P. N., Kauffmann, G., Heckman, T. M., et al. 2005, MNRAS, 362, 25

Beswick, R. J., Peck, A. B., Taylor, G. B., & Giovannini, G. 2004, MNRAS, 352, 49

Booth, R. S., de Blok, W. J. G., Jonas, J. L., & Fanaroff, B. 2009, arXiv:0910.2935

van Breugel, W., Miley, G., & Heckman, T. 1984, AJ, 89, 5

Chandola, Y., Sirothia, S. K., & Saikia, D. J. 2011, MNRAS, 418, 1787

Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ, 115, 1693

Curran, S. J., & Whiting, M. T. 2010, ApJ, 712, 303

Curran, S. J., Allison, J. R., Glowacki, M., Whiting, M. T., & Sadler, E. M. 2013,MNRAS, 431, 3408

Davis, T. A., Alatalo, K., Bureau, M., et al. 2013, MNRAS, 429, 534

DeBoer, D. R., Gough, R. G., Bunton, J. D., et al. 2009, IEEE Proceedings, 97, 1507

Delhaize J., Meyer M. J., Staveley-Smith L., Boyle B. J., 2013, MNRAS, 433, 1398

Dickey, J. M. 1982, ApJ, 263, 87

Emonts, B. H. C., Morganti, R., Struve, C., et al. 2010, MNRAS, 406, 987

Fanti, R., Fanti, C., Schilizzi, R. T., et al. 1990, A&A, 231, 333

Fanti, C., Fanti, R., Dallacasa, D., et al. 1995, A&A, 302, 317

Page 83: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

74 BIBLIOGRAPHY

Fabello S., Catinella B., Giovanelli R. et al., 2011, MNRAS, 411, 993

Fabello S., Kauffmann G., Catinella B. et al., 2011, MNRAS, 416, 1739

Gallimore, J. F., Baum, S. A., O’Dea, C. P., Pedlar, A., & Brinks, E. 1999, ApJ, 524,684

Geréb, K., Morganti, R., Oosterloo, T. A., Guglielmino, G., & Prandoni, I. 2013, A&A,558, A54

van Gorkom, J. H., Knapp, G. R., Ekers, R. D., et al. 1989, AJ, 97, 708

Gupta, N., Salter, C. J., Saikia, D. J., Ghosh, T., & Jeyakumar, S. 2006, MNRAS, 373,972

Knapp, G. R., Turner, E. L., & Cunniffe, P. E. 1985, AJ, 90, 454

Lah P., et al., 2007, MNRAS, 376, 1357

Lah P., et al., 2009, MNRAS, 399, 1447

Johnston S., et al., 2008, ExA, 22, 151

Maddox, N., Hess, K. M., Blyth, S.-L., & Jarvis, M. J. 2013, MNRAS, 433, 2613

Martel, A. R., Baum, S. A., Sparks, W. B., et al. 1999, ApJS, 122, 81

Mauch, T., & Sadler, E. M. 2007, MNRAS, 375, 931

Morganti, R., Oosterloo, T., & Tsvetanov, Z. 1998, AJ, 115, 915

Morganti, R., Oosterloo, T. A., Tadhunter, C. N., et al. 2001, MNRAS, 323, 331

Morganti, R., Oosterloo, T. A., Tadhunter, C. N., van Moorsel, G., & Emonts, B. 2005,A&A, 439, 521

Morganti, R., de Zeeuw, P. T., Oosterloo, T. A., et al. 2006, MNRAS, 371, 157

Morganti, R., Peck, A. B., Oosterloo, T. A., et al. 2009, A&A, 505, 559

Morganti, R., Fogasy, J., Paragi, Z., Oosterloo, T., & Orienti, M. 2013, Science, 341, 1082

Murray, C. E., Lindner, R. R., Stanimirović, S., et al. 2014, ApJ, 781, L41

Oosterloo, T.A.; Morganti, R.; Crocker, A. et al.; 2010; MNRAS; 409; 500

Oosterloo, T., Verheijen, M., & van Cappellen, W. 2010, ISKAF2010 Science Meeting

Owsianik, I., & Conway, J. E. 1998, A&A, 337, 69

Peck, A. B., Taylor, G. B., & Conway, J. E. 1999, ApJ, 521, 103

Pihlström, Y. M., Conway, J. E., & Vermeulen, R. C. 2003, A&A, 404, 871

Readhead, A. C. S., Taylor, G. B., Xu, W., et al. 1996, ApJ, 460, 612

Page 84: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

BIBLIOGRAPHY 75

Roberts, M. S. 1970, ApJ, 161, L9

Saikia, D. J.,&Gupta, N. 2003, A&A, 405, 499

Sault R. J., Teuben P. J., Wright M. C. H., 1995, ASPC, 77, 433

Serra P., et al., 2012, MNRAS, 422, 1835

Shostak, G. S., van Gorkom, J. H., Ekers, R. D., et al. 1983, A&A, 119, L3

Snellen, I. A. G., Mack, K.-H., Schilizzi, R. T., & Tschager, W. 2004, MNRAS, 348, 227

Verheijen M., van Gorkom J. H., Szomoru A., Dwarakanath K. S., Poggianti B. M.,Schiminovich D., 2007, ApJ, 668, L9

Vermeulen, R. C., Pihlström, Y. M., Tschager, W., et al. 2003, A&A, 404, 861

de Vries, N., Snellen, I. A. G., Schilizzi, R. T., Mack, K.-H., & Kaiser, C. R. 2009, A&A,498, 641

Wolfe, A. M., & Burbidge, G. R. 1975, ApJ, 200, 548

York D. G., et al., 2000, AJ, 120, 1579

De Young, D. S., Roberts, M. S., & Saslaw, W. C. 1973, ApJ, 185, 809

De Young, D. S. 1993, ApJ, 405, L13

Young, L. M., Bureau, M., Davis, T. A., et al. 2011, MNRAS, 414, 940

Page 85: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

76 BIBLIOGRAPHY

Page 86: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Chapter 5The HI absorption ‘Zoo’

– K. Geréb, F. Maccagni, R. Morganti, T. Oosterloo –Submitted to Astronomy & Astrophysics

AbstractWe present the analysis of the H I absorption in a sample of 101 flux-selected radio AGN (S1.4 GHz> 50 mJy) observed with the Westerbork Synthesis Radio Telescope (WSRT). We detect H Iabsorption in 32 objects (30% of the sample). In a previous paper, we performed a spectralstacking analysis on the radio sources, while here we characterize the absorption spectra of theindividual detections using the recently presented busy function (Westmeier et al. 2014), anefficient way of fitting both asymmetric and Gaussian H I profiles.

The H I absorption spectra show a broad variety of widths, shapes and kinematical proper-ties. The Full Width Half Maximum (FWHM) of the detections ranges between 32 km s−1 <FWHM < 570 km s−1, whereas the Full Width at 20% of the peak intensity (FW20) varies be-tween 63 km s−1 < FW20 < 825 km s−1. We quantify the asymmetry of the lines by measuringthe velocity offset |∆vCP| between the H I peak and the line centroid (measured at 20% of themaximum). Based on the different shapes and widths of the H I lines, we separate our sample inthree groups: narrow lines (FWHM < 100 km s−1), intermediate (FWHM < 200 km s−1) andbroad profiles (FWHM > 200 km s−1). In each group we study the kinematical and radio sourceproperties of the detections, with the goal of identifying different morphological structures ofH I. Narrow lines at the systemic velocity are likely produced by regularly rotating H I disks.In the sample of detections, 31% of the lines are narrow. More H I disks can be present amongintermediate FWHM lines with up to 63% detection rate, however the H I in these sources ismore unsettled. Among the broadest lines, 45% of the profiles appear blueshifted, while a lowerfraction, 9% are redshifted. Broad lines show large asymmetries up to |∆vCP| ∼ 250 km s−1,and we note that symmetric broad lines are missing from our sample. The combination of broadwidths and lack of symmetry could suggest that these profiles are tracing unsettled gas.

We find three new cases of blueshifted broad wings (with FW20 & 500 km s−1); along withthe high radio power of their AGN, these detections are good candidates for being H I outflows.Together with the known cases of outflows already included in the sample (3C 293 and 3C 305),the detection rate of H I outflows is 5% in the total radio AGN sample. Three of the broadest(up to FW20 = 825 km s−1) detections are associated with gas-rich mergers.

Page 87: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

78 chapter 5: The HI absorption ‘Zoo’

Using stacking techniques, in Chapter 4 we show that compact radio sources have higherτ , FWHM and column density than extended sources. Here we measure the H I line param-eters individually in compact and extended sources using the busy function. Blueshifted andbroad/asymmetric lines are often present among compact sources. This result, in good agree-ment with the results of stacking, suggests that unsettled gas is responsible for the larger stackedFWHM detected in compact sources. Such H I gas properties may arise due to jet-cloud inter-actions, as young radio sources clear their way through the rich ambient gaseous medium.

5.1 IntroductionNuclear activity in (radio) AGN is though to be connected to the presence and kinemat-ical properties of the gas in the circumnuclear regions. Observational evidence clearlyshows that interactions between AGN and their ambient gaseous medium do occur. Thus,such interplay is thought to be responsible for the balance between the feeding of theblack hole and feedback processes. H I is one of the components that may play a role inthese processes.

Radio AGN are typically hosted by early-type galaxies (Bahcall et al. 1997; Best et al.2005). In the nearby Universe our knowledge of the cold gas properties of early-typegalaxies has increased in recent years thanks to projects like WSRT-SAURON (Morgantiet al. 2006; Grossi et al. 2009; Oosterloo et al. 2010a) and ATLAS3D (Serra et al. 2012;Young et al. 2011; Davis et al. 2013). In radio-loud AGN, H I absorption studies canbe used to explore the presence and kinematics of the gas. A number of H I absorptionstudies from recent years have provided a better understanding of the H I propertiesof radio galaxies (van Gorkom et al. 1989; Morganti et al. 2001; Vermeulen et al. 2003;Gupta et al. 2006; Curran & Whiting 2010; Emonts et al. 2010).

In the mentioned studies, the morphology and kinematics of H I gas is found to be verycomplex in radio galaxies. H I can trace rotating disks, offset clouds, and complex mor-phological structures of unsettled gas, e.g. infall and outflows. van Gorkom et al. (1989)reported a high fraction of redshifted H I detections in compact radio sources, and they es-timated that infalling H I clouds can provide the necessary amount of gas to fuel the AGNactivity. Later work revealed that not just infalling gas, but also blueshifted, outflowingH I is present in many AGN, and in particular in compact Gigahertz-Peaked Spectrum(GPS) and Compact Steep Spectrum (CSS) sources (Vermeulen et al. 2003; Gupta et al.2006). The structure of compact sources often appears asymmetric in brightness, locationand polarization. Such disturbed radio source properties indicate dynamical interactionsbetween the radio jets and the circumnuclear medium, and this process is likely to bethe driver of fast H I outflows that has been detected in a number of radio galaxies. Allthese properties are consistent with a scenario in which interactions between the radiosource and the surrounding gas have an effect both on the gas and on the radio sourceproperties. It is clear that one needs to disentangle all these phenomena in order tounderstand the intricate interplay between AGN and the gas.

Because AGN and their host galaxies are known to have a broad range of complexH I morphologies, kinematics, gas masses and column densities, future large datasetswill require robust methods to extract and analyze meaningful information that can berelevant for our understanding of the amount and conditions of the gas in radio galaxies.Recently, Westmeier et al. (2014) presented the busy function (BF) for parametrizing H Iemission spectra. The BF is efficient in fitting both Gaussian and asymmetric profiles,

Page 88: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.2: Description of the sample and observations 79

therefore it is also suitable for fitting the wide variety of absorption lines in our sample.In this paper, we use for the first time the busy function to parametrize and describe

the complex H I absorption properties of a relatively large sample of 32 radio sourceswith H I detections. The total sample of 101 sources was recently presented in Gerebet al. (2014), hereafter referred to as Chapter 4. The main goal of Chapter 4 was tocarry out a spectral stacking analysis of the H I absorption lines and to measure theco-added H I signal of the sample at low τ detection limit. Stacking is very efficient atreproducing the global profile of the stacked spectra, but it does not provide informationon the underlying profile distribution. Here we present the detailed discussion of the H Iabsorption busy fit parameters in relation to the results of stacking.

One interesting finding of the available H I absorption studies presented above is thatthere appears to be a trend between the H I properties and the evolutionary stage ofthe radio source. CSS and GPS sources have been proposed to represent young ( <∼ 104

yr) radio AGN (Fanti et al. 1995; Readhead et al. 1996; Owsianik & Conway 1998). Thehigh H I detection rate in compact CSS and GPS sources has been interpreted as evidencefor a relation between the recent triggering of the AGN activity and the presence of H Igas (Pihlström et al. 2003; Gupta et al. 2006; Emonts et al. 2010; Chandola et al. 2010).

In Chapter 4 we looked at the H I properties of compact and extended sources usingstacking techniques, and we found that compact sources not only have higher detectionrate and optical depth, but also larger profile width than extended sources. We arguethat such H I properties reflect the presence of rich gaseous medium in compact sources,and that the larger FWHM of compact sources is due to the presence of unsettled gas.In the present paper we use the BF to measure the H I parameters of individual compactand extended detections. We discuss the profile parameters of compact and extendedsources in relation to the results of stacking from Chapter 4.

Several examples from the literature show that H I mass outflow rates of a few ×10M⊙yr−1 are associated with fast (∼ 1000 km s−1), radio jet-driven outflows (Morganti et al.1998, 2005), therefore such feedback effects are considered to have major impact both onthe star formation processes in galaxies and the further growth of the black hole. How-ever, at the moment little is known about the frequency and lifetime of such H I outflowsin radio galaxies, and larger samples are needed to constrain the role and significance ofoutflows in the evolution of galaxies. We have not found signatures of broad, blueshiftedwings in the stacked spectra presented in Chapter 4. Here we use the busy fit parametersto identify and characterize new cases of H I outflows.

In this paper the standard cosmological model is used, with parameters ΩΛ = 0.3, Λ= 0.7 and H0 = 70 km s−1 Mpc−1.

5.2 Description of the sample and observationsAs described in Chapter 4, the sample was selected from the cross-correlation of theSloan Digital Sky Survey (SDSS, York et al. 2000) and Faint Images of the Radio Sky atTwenty-cm (FIRST, Becker et al. 1995) catalogs. In the redshift range 0.02 < z < 0.23,101 sources were selected with peak flux S1.4 GHz > 50 mJy in the FIRST catalog. Thecorresponding radio power distribution of the AGN ranges between 1023 – 1026 W Hz−1.

The observations were carried out with the Westerbork Synthesis Radio Telescope(WSRT). Each target was observed for 4 hours. In the case of 4C +52.37, we carried out

Page 89: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

80 chapter 5: The HI absorption ‘Zoo’

8 hour follow-up observations in order to increase the H I sensitivity in the spectra. Thiswill be discussed in Sec. 5.4.2. A more detailed description of the observational setupand the data reduction can be found in Chapter 4.

Because our sample is solely flux-selected, we can expect to have a mix of radiosources with various host galaxies. The radio galaxy sample consists of compact (CSS,GPS, and unclassified) and extended sources in Table 5.0. The sizes of the radio sourcepopulation vary between 4 pc and 550 kpc. Besides radio galaxies, we also find opticallyblue objects with g − r < 0.7 colors. These blue objects are associated with differenttypes of objects, for example gas-rich mergers (UGC 8387, Mrk 273), Seyfert galaxiesand QSOs (Quasi Stellar Objects). To make the AGN sample more homogeneous, weexcluded these sources from the stacking analysis in Chapter 4. In this paper theseobjects are marked by yellow squares in the figures from Sec. 6.5 and discussed in Sec.4.3.

In Chapter 4 we have divided the sample in compact and extended radio sources basedon the NVSS major-to-minor axis ratio vs. the FIRST peak-to-integrated flux ratio. Thesame classification is used here.

5.3 Results

We detect H I absorption in 32 of the observed galaxies, and 24 of these are new detections(see notes on individual sources in Appendix 5.8.1). The H I profiles in Fig. 5.2 show avariety of complex shapes and kinematics. The H I lines are separated in three groupsbased on the profile analysis that will be discussed in Sec. 5.3.1 and Sec. 5.3.2. As wemention in Chapter 4, the τ and N(H I) range of detections is quite broad, spanning twoorders of magnitude between a few × (1017 – 1019) (Tspin/cf) cm−2, where Tspin is thespin temperature and cf is the covering factor of the gas. In Table 5.0 we summarize thecharacteristics of the detected sources.

Non-detections and the corresponding 3-σ upper limits are presented in Table 5.2 ofAppendix 5.8.2. The N(H I) upper limits for non-detections were calculated by assum-ing FWHM = 100 km s−1, following the analysis of the profile widths in Sec. 5.3.1.Each detected source is given a number identifier in Table 5.0 and Table 5.2, and werefer to the detections based on this sequential number. In Chapter 4 we discussed theoptical depth and column density properties of the sample, whereas here we focus onthe kinematical properties of the H I lines. Below we study in more detail the width,asymmetry parameters, and the blueshift/redshift distribution of the detections usingthe busy function.

In Chapter 4 we show that statistically, detections and non-detections have a similarflux distribution, implying that detections in our sample are not biased toward brightersources. Here, in Fig. 5.1 we present the radio power distribution of detections andnon-detections. According to the Kolmogorov-Smirnov test, the significance level thatthe two distributions are different is only 10%, implying that statistically detections andnon-detections have a similar radio power distribution. The largest difference betweenthe two distributions (D) is measured at ∼24.6 W Hz−1.

Further notes on the individual detections are presented in Appendix 5.8.1.

Page 90: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.3: Results 81T

able

5.0:

Cha

ract

erist

ics

ofth

eH

Ide

tect

ions

.Fo

rea

chde

tect

edso

urce

we

list:

1)id

entifi

erth

atw

illbe

used

inth

ete

xt;

2)sk

yco

ordi

nate

s;3)

alte

rnat

ive

nam

eof

the

sour

ces;

4)SD

SSre

dshi

ft;5

)1.

4G

Hz

flux

dens

ity;6

)1.

4G

Hz

radi

opo

wer

;7)

radi

om

orph

olog

y;8)

radi

oso

urce

size;

9)pe

akop

tical

dept

h;10

)col

umn

dens

ity.

The

radi

om

orph

olog

yab

brev

iatio

nsin

colu

mn

7ar

eas

follo

wsC

:com

pact

inou

rse

lect

ion,

E:ex

tend

edin

our

sele

ctio

n,C

SO:C

ompa

ctSy

mm

etric

Obj

ect,

CSS

:Com

pact

Stee

pSp

ectr

umso

urce

,CJ:

Cor

e-Je

t,C

X:c

ompl

exm

orph

olog

y,U

:unr

esol

ved,

M:m

erge

r,Q

SO:q

uasa

r,FS

RQ

:Fla

tSp

ectr

umR

adio

Qua

sar.

Iden

tifier

RA

,Dec

Nam

ez

S 1.4

GH

zP 1

.4G

Hz

Rad

ioSi

zeτ p

ea

kN

(HI)

mJy

WH

z−1

Mor

phol

ogy

(kpc

)10

18(T

spin

/cf)

cm−

2

#1

07h5

7m56

.7s

+39

d59m

36s

B3

0754

+40

10.

0657

7792

23.9

8C

SS0.

25(K

)0.

042

9.4

#2

08h0

6m01

.5s

+19

d06m

15s

2MA

SXJ0

8060

148+

1906

142

0.09

7882

142

24.5

4E

-0.

099

26.9

#3

08h0

9m38

.9s

+34

d55m

37s

B2

0806

+35

0.08

2493

142

24.3

8C

J(E

)13

9(Bo)

0.00

91.

0#

408

h36m

37.8

s+

44d0

1m10

sB

308

33+

442

0.05

542

134

23.9

9C

SO?

1.7(d

V)

0.01

61.

9#

508

h43m

07.1

s+

45d3

7m43

sB

308

39+

458

0.19

1947

331

25.5

4C

SO0.

14(H

)0.

273

34.5

#6

09h0

9m37

.4s

+19

d28m

08s

Mrk

1226

0.02

7843

6323

.05

FSR

Q?

0.00

8(H)

0.11

923

.3#

709

h35m

51.6

s+

61d2

1m11

sU

GC

0510

10.

0393

8614

823

.73

E(M

)0.

048(L

)0.

073

53.6

#8

10h2

0m53

.7s

+48

d31m

24s

4C+

48.2

90.

0531

5282

23.7

4E

(X-s

hape

d)-

0.05

09.

3#

910

h53m

27.2

s+

20d5

8m36

sJ1

0532

7+20

5835

0.05

2639

7923

.72

C-

0.02

33.

9#

1011

h20m

30.0

s+

27d3

6m11

s2M

ASX

J112

030+

2736

100.

1125

1617

724

.76

C-

0.14

715

.6#

1112

h02m

31.1

s+

16d3

7m42

s2M

ASX

J120

2311

2+16

3741

40.

1195

3282

24.4

8C

-0.

042

7.4

#12

12h0

5m51

.4s

+20

d31m

19s

NG

C40

93-M

CG

+04

-29-

020.

0237

8980

23.0

1C

-0.

034

5.2

#13

12h0

8m55

.6s

+46

d41m

14s

B3

1206

+46

90.

1009

5569

24.2

5E

550(M

)0.

052

2.6

#14

12h3

2m00

.5s

+33

d17m

48s

B2

1229

+33

0.07

8819

9424

.16

FRII

-0.

034

5.6

#15

12h4

7m07

.3s

+49

d00m

18s

4C+

49.2

50.

2069

111

4026

.15

CSS

6(S)

0.00

20.

5#

1612

h54m

33.3

s+

18d5

6m02

s2M

ASX

J125

433+

1856

020.

1154

476

24.4

2C

SO0.

017(H

)0.

068

6.3

(B)

Bes

wic

ket

al.(

2004

);(B

o)B

ondi

etal

.(20

04);

(C)

Car

illi&

Tayl

or(2

000)

;(F

)Fe

rett

i&G

iova

nnin

i(19

94);

(H)

Hel

mbo

ldt

etal

.(2

007)

;(K

)K

uner

t-B

ajra

szew

ska

&La

bian

o(2

010)

;(L)

Lons

dale

etal

.(20

03);

(M)

Mar

chã

etal

.(20

01);

(Ma)

Mac

ket

al.(

1993

);(R

)R

omer

o-C

añiz

ales

etal

.(20

12);

(S)

Saik

ia&

Gup

ta(2

003)

;(T

)Ta

ylor

etal

.(20

07);

(dV

)de

Vrie

set

al.(

2010

).

Page 91: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

82 chapter 5: The HI absorption ‘Zoo’

Tab

le5.

0:-C

ontin

ued.

Iden

tifier

RA

,Dec

Nam

ez

S 1.4

GH

zP 1

.4G

Hz

Rad

ioSi

zeτ p

ea

kN

(HI)

mJy

WH

z−1

Mor

phol

ogy

(kpc

)10

18(T

spin

/cf)

cm−

2

#17

13h0

1m32

.6s

+46

d34m

03s

2MA

SXJ1

3013

264+

4634

032

0.20

552

9725

.07

C-

0.01

83.

2#

1813

h17m

39.2

s+

41d1

5m46

sB

313

15+

415

0.06

6109

246

24.4

2C

X0.

005(d

V)

0.03

17.

0#

1913

h20m

35.3

s+

34d0

8m22

sIC

883,

UG

C83

870.

0230

5897

23.0

7E

(M)

0.98

(R)

0.16

278

.8#

2013

h25m

13.4

s+

39d5

5m53

sSD

SSJ1

3251

3.37

+39

5553

.20.

0755

5137

23.7

1C

SO?

0.01

4(dV

)0.

053

9.3

#21

13h4

0m35

.2s

+44

d48m

17s

IRA

SF1

3384

+45

030.

0654

4236

23.5

7C

J0.

0041

(dV

)0.

260

20.3

#22

13h4

4m42

.1s

+55

d53m

13s

Mrk

273

0.03

734

132

23.6

3E

(M)

0.14

8(C)

0.09

176

.2#

2313

h52m

17.8

s+

31d2

6m46

s3C

293

0.04

5194

3530

25.2

3FR

I28

0(B)

0.05

714

.5#

2414

h22m

10.8

s+

21d0

5m54

s2M

ASX

J142

210+

2105

540.

1914

7784

24.9

4C

-0.

048

10.1

#25

14h3

5m21

.7s

+50

d51m

23s

2MA

SXJ1

4352

162+

5051

233

0.09

9694

141

24.5

5U

0.27

(dV

)0.

013

4.7

#26

14h4

9m21

.6s

+63

d16m

14s

3C30

5-I

C10

650.

0416

7725

0025

.01

FRI

-0.

005

1.5

#27

15h0

0m34

.6s

+36

d48m

45s

2MA

SXJ1

5003

4+36

4845

0.06

6101

6123

.81

QSO

-0.

190

35.3

#28

15h2

9m22

.5s

+36

d21m

42s

2MA

SXJ1

5292

250+

3621

423

0.09

8934

3823

.97

C-

0.07

525

.1#

2916

h02m

46.4

s+

52d4

3m58

s4C

+52

.37

0.10

5689

577

25.2

2C

SO0.

35(d

V)

0.01

56.

7#

3016

h03m

32.1

s+

17d1

1m55

sN

GC

6034

0.03

402

278

23.8

7E

122(M

a)0.

066

5.6

#31

16h0

3m38

.0s

+15

d54m

02s

Abe

ll21

470.

1097

6110

024

.49

FSR

Q0.

02(T

)0.

125

50.7

#32

16h1

2m17

.6s

+28

d25m

47s

2MA

SXJ1

6121

7+28

2546

0.05

3138

7823

.72

C0.

6(F)

0.06

17.

8

(B)

Bes

wic

ket

al.(

2004

);(B

o)B

ondi

etal

.(20

04);

(C)

Car

illi&

Tayl

or(2

000)

;(F

)Fe

rett

i&G

iova

nnin

i(19

94);

(H)

Hel

mbo

ldt

etal

.(2

007)

;(K

)K

uner

t-B

ajra

szew

ska

&La

bian

o(2

010)

;(L)

Lons

dale

etal

.(20

03);

(M)

Mar

chã

etal

.(20

01);

(Ma)

Mac

ket

al.(

1993

);(R

)R

omer

o-C

añiz

ales

etal

.(20

12);

(S)

Saik

ia&

Gup

ta(2

003)

;(T

)Ta

ylor

etal

.(20

07);

(dV

)de

Vrie

set

al.(

2010

).

Page 92: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.3: Results 83

22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0 26.5log(P1.4 GHz) [W/Hz]

0.0

0.2

0.4

0.6

0.8

1.0

Cum

ula

tive fra

ctio

n

DetectionsNon-DetectionsD

Figure 5.1: The 1.4 GHz radio power cumulative fraction of detections and non-detections in the sample of 101 AGN. We measure D = 0.228 for 32 detections and69 non-detections. The source parameters for detection and non-detections are listed inTable 5.0 and Table 5.2, respectively.

5.3.1 Fitting complex H I absorption profiles with the busy func-tion

The sample of absorption lines in Fig. 5.2 is very heterogeneous in terms of line shapeand widths. Thus, it is crucial to develop a uniform method to characterize the proper-ties of this variety of H I profiles. So far, Gaussian fitting has been widely used to deriveabsorption line properties, e.g. the width of the profile, and to determine the pres-ence of multiple components (Vermeulen et al. 2003; Gupta et al. 2006; Curran et al.2011). When multiple peaked profiles occur, like in our absorption sample, Gaussianfitting methods have the disadvantage of having to make an a priori assumption on thephysical conditions of the gas, by choosing the number of components to be fitted. Inthe case of H I integrated emission profiles, an alternative solution has been proposedby Westmeier et al. (2014): the busy function fitting method. The busy function is anheuristic, analytic function, given by the product of two error functions and a polynomialfactor:

B(x)= a4 × (erf[b1w + x − xe] + 1)×(erf[b2w − x + xe] + 1) × (c|x − xp|n + 1)

(5.1)The main advantage of this function is that a proper combination of the parameters

can fit a wide variety of line profiles. When c = 0, the busy function may well approximatea Gaussian profile, while if c = 0, for different values of b1 and b2, an asymmetricdouble horn profile can be reproduced. With the same function it is then possible to

Page 93: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

84 chapter 5: The HI absorption ‘Zoo’

−1000 −500 0 500 1000Velocity [km/s]

−0.008

−0.006

−0.004

−0.002

0.000

0.002

0.004

τ

3

χ2 = 0.64

−1000 −500 0 500 1000Velocity [km/s]

−0.015

−0.010

−0.005

0.000

0.005

τ

4

χ2 = 0.0

−1000 −500 0 500 1000Velocity [km/s]

−0.30

−0.25

−0.20

−0.15

−0.10

−0.05

0.00

0.05

τ

5

χ2 = 1.57

−1000 −500 0 500 1000Velocity [km/s]

−0.14

−0.12

−0.10

−0.08

−0.06

−0.04

−0.02

0.00

τ

10

χ2 = 4.14

−1000 −500 0 500 1000Velocity [km/s]

−0.03

−0.02

−0.01

0.00

0.01

τ

12

χ2 = 1.56

−1000 −500 0 500 1000Velocity [km/s]

−0.05

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02

τ

13

χ2 = 1.33

−1000 −500 0 500 1000Velocity [km/s]

−0.06

−0.04

−0.02

0.00

0.02

τ

16

χ2 = 1.21

−1000 −500 0 500 1000Velocity [km/s]

−0.30

−0.25

−0.20

−0.15

−0.10

−0.05

0.00

0.05

τ

21

χ2 = 1.24

−1000 −500 0 500 1000Velocity [km/s]

−0.07

−0.06

−0.05

−0.04

−0.03

−0.02

−0.01

0.00

τ

30

χ2 = 0.84

−1000 −500 0 500 1000Velocity [km/s]

−0.06

−0.04

−0.02

0.00

τ

32

χ2 = 0.95

Figure 5.2: (a) - H I profiles (10 detections) in the narrow region with FWHM < 100km s−1. Red lines represent the BF fits.

Page 94: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.3: Results 85

−1000 −500 0 500 1000Velocity [km/s]

−0.04

−0.03

−0.02

−0.01

0.00τ

1

χ2 = 1.05

−1000 −500 0 500 1000Velocity [km/s]

−0.08

−0.06

−0.04

−0.02

0.00

0.02

0.04

τ

2

χ2 = 0.69

−1000 −500 0 500 1000Velocity [km/s]

−0.10

−0.08

−0.06

−0.04

−0.02

0.00

0.02

0.04

τ

6

χ2 = 1.3

−1000 −500 0 500 1000Velocity [km/s]

−0.05

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02

τ

8

χ2 = 1.48

−1000 −500 0 500 1000Velocity [km/s]

−0.020

−0.015

−0.010

−0.005

0.000

0.005

0.010

0.015

τ

9

χ2 = 1.34

−1000 −500 0 500 1000Velocity [km/s]

−0.03

−0.02

−0.01

0.00

0.01

τ

14

χ2 = 0.88

−1000 −500 0 500 1000Velocity [km/s]

−0.035

−0.030

−0.025

−0.020

−0.015

−0.010

−0.005

0.000

τ

18

χ2 = 1.44

−1000 −500 0 500 1000Velocity [km/s]

−0.04

−0.02

0.00

0.02

0.04

τ

20

−1000 −500 0 500 1000Velocity [km/s]

−0.05

−0.04

−0.03

−0.02

−0.01

0.00

0.01

τ

23

χ2 = 57.92

−1000 −500 0 500 1000Velocity [km/s]

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02

τ

24

χ2 = 1.09

−1000 −500 0 500 1000Velocity [km/s]

−0.15

−0.10

−0.05

0.00

τ

27

χ2 = 1.42

Figure 5.2: (b) - H I profiles (11 detections) in the intermediate width region at 100km s−1< FWHM < 200 km s−1.

Page 95: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

86 chapter 5: The HI absorption ‘Zoo’

−1000 −500 0 500 1000Velocity [km/s]

−0.07

−0.06

−0.05

−0.04

−0.03

−0.02

−0.01

0.00

τ7

χ2 = 3.6

−1000 −500 0 500 1000Velocity [km/s]

−0.04

−0.03

−0.02

−0.01

0.00

0.01

0.02

0.03

τ

11

χ2 = 1.21

−1000 −500 0 500 1000Velocity [km/s]

−0.0015

−0.0010

−0.0005

0.0000

0.0005

0.0010

τ

15

χ2 = 1.18

−1000 −500 0 500 1000Velocity [km/s]

−0.015

−0.010

−0.005

0.000

0.005

0.010

0.015

τ

17

χ2 = 0.91

−1000 −500 0 500 1000Velocity [km/s]

−0.15

−0.10

−0.05

0.00

τ

19

χ2 = 1.97

−1000 −500 0 500 1000Velocity [km/s]

−0.08

−0.06

−0.04

−0.02

0.00

τ

22

χ2 = 1.45

−1000 −500 0 500 1000Velocity [km/s]

−0.010

−0.005

0.000

0.005

0.010

τ

25

χ2 = 0.8

−1000 −500 0 500 1000Velocity [km/s]

−0.005

−0.004

−0.003

−0.002

−0.001

0.000

0.001

τ

26

χ2 = 0.78

−1000 −500 0 500 1000Velocity [km/s]

−0.06

−0.04

−0.02

0.00

0.02

0.04

0.06

τ

28

χ2 = 1.0

−1000 −500 0 500 1000Velocity [km/s]

−0.010

−0.005

0.000

0.005

τ

29

χ2 = 0.77

−1000 −500 0 500 1000Velocity [km/s]

−0.14

−0.12

−0.10

−0.08

−0.06

−0.04

−0.02

0.00

τ

31

χ2 = 1.56

Figure 5.2: (c) - H I profiles (11 detections) in the broad width region at FWHM > 200km s−1.

Page 96: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.3: Results 87

fit single and double peaked profiles, while with Gaussian fitting more functions areneeded for the fitting of multiple lines. Hence, using the busy function it is possibleto derive the characteristics of very different lines in a uniform way, without any pre-defined assumptions on the number of gas components that may produce the line profile.Westmeier et al. (2014) applied the busy function fit to integrated emission lines of theHI Parkes All Sky Survey (HIPASS) sample, but such a method has never been appliedto absorption lines before. We used the C++ code provided by (Westmeier et al. 2014)to fit our absorption profiles and estimate the width, asymmetry and blue/red-shift ofthe profiles. On average, the chi-square test on the fitted lines is χ2 ∼ 1.1, and thefit is successful in parametrizing 30 out of 32 profiles of our sample. The BF fails tofit the profiles in source #20, where two lines are separated in velocity, and source #4,where both emission and absorption is present in the spectra. For #4 we extract a newspectrum from the cube at a location still close to the nuclear region, where the emissionis not very strong. This spectrum can be successfully fitted by the BF. For source #20,we evaluate the profile parameters of the main (stronger) component using Gaussianfitting.

The FWHM and the Full Width at 20% Maximum (FW20) of the lines are estimatedby measuring the width of the fitted busy function profiles at 50% and at 20% of thepeak intensity. The line centroid is measured at the middle point of the width at 20%.These parameters are summarized in Table 5.1 for all the absorption lines of our sample.

5.3.2 Characterization of the profiles with BF parametersThe profile width distribution of our sample is presented in Fig. 5.3. The sample ofgas-rich mergers and blue (g − r < 0.7) objects, which were excluded from Chapter4, are marked by yellow squares. We detect a broad range of widths between 32 kms−1 < FWHM < 570 km s−1 and 63 km s−1 < FW20 < 825 km s−1. Following avisual inspection we find that broader profiles are more complex than narrow lines. Weclearly see a separation in shape with increasing width, it appears that we can separatethree groups, representing physically different H I structures. The first group consistsof narrow single components, the second group of two (or more) blended components,whereas profiles in the third group include well separated double components. Thesegroups are presented and separated at the dashed lines in Fig. 5.3. The grouping of theH I profiles in Fig. 5.2 is also based on this selection. This separation is further supportedby the asymmetry analysis below.

In order to quantify the asymmetry of the detected lines, we derive the asymmetryparameter as the ∆vCP = vCentroid - vHI Peak velocity offset between the centroid andthe peak intensity of the H I line. In Fig. 5.4 (top panel) we show the absolute valueof the asymmetry distribution as function of the FW20 profile width. We find that innarrow profiles at FW20 < 200 km s−1, the offset between the centroid and the H I peakis < 50 km s−1. In the group with 200 km s−1 < FW20 < 300 km s−1, the asymmetryparameters are larger, with up to 100 km s−1 difference between the line centroid and theH I peak. Broad detections at FW20 > 300 km s−1 have the most asymmetric profiles,with |∆vCP| parameters larger than a few × 100 km s−1. Thus, the grouping of objects inFig. 5.3 is further supported by the increasing asymmetry observed in the three groups,and this grouping will be used in the further analysis.

Among broad lines with FW20 > 300 km s−1 there are almost no symmetric pro-

Page 97: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

88 chapter 5: The HI absorption ‘Zoo’

0 100 200 300 400 500 600 700 FWHM (km/s)

0

200

400

600

800

1000 FW20 (km

/s)

1:1ExtendedMergers+BlueCompact

Figure 5.3: The FWHM and FW20 width distribution of the 32 detections derived withthe busy function

files detected in Fig. 5.4 (top panel). Narrow lines cannot yield large asymmetries byconstruction, hence we normalize the asymmetries by the FW20 of the lines in Fig. 5.4(bottom panel), obtaining a more uniform distribution. Nevertheless, we confirm thetrend that symmetric broad lines (with FW20 > 300km s−1) are missing.

To expand on the analysis of the line asymmetries, we derive the velocity offset ofthe H I peak (with respect to the systemic velocity) to quantify the blueshift/redshiftdistribution of the main, deepest H I component. In Fig. 5.5 no clear correlation is seenbetween the velocity offset and the width distribution of the lines. A detection is classi-fied as blueshifted/redshifted if the velocity offset of the line is larger than ± 100 km s−1.

Using the line parameters presented above of the three width regions (introduced inFig. 5.3), we aim to identify H I structures belonging to different morphological groups:disks, clouds in radial motion (outflows, in fall), and other unsettled gas structures, forexample gas-rich mergers.1 In general, narrow lines of a few × 100 km s−1 at the systemicvelocity can be due to gas regularly rotating in a disk-like structure, e.g. high resolutionobservations show that the main (deep) absorption component in 3C 293 is associatedwith an H I disk (Beswick et al. 2004). However, the origin of H I profiles with broaderwidth of the order of > 500 km s−1 has to involve other physical processes, e.g. disturbedkinematics due to mergers, outflows, in order to accelerate the gas to such high velocities.We expect to find H I disks in the narrow region, while broader, asymmetric profiles, forexample outflows must belong to the broadest group with large FW20 in our sample.The nature of the gas structures in the three groups is discussed in Sec. 5.4.

1 Following what has been found by the detailed (and spatially resolved) studies of single objects

Page 98: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.3: Results 89

Figure 5.4: 1. (Top panel): asymmetry vs. FW20 distribution of the 32 detections.2. (Bottom panel): Normalized asymmetry parameter vs. FW20 width distribution.The classification of the groups is based on the width regions from Fig. 5.3. Blackdiamonds mark narrow lines with FWHM < 100 km s−1, black crosses mark the middlewidth region with 100 km s−1< FWHM < 200 km s−1, empty circles indicate broadline detections with FWHM > 200 km s−1, and yellow squares indicate blue (g − r <0.7) objects and mergers, which objects were excluded from our discussion in Chapter4. In the histograms grey bars mark narrow lines, the crossed hatched region indicatesintermediate width profiles, the dotted hatched region marks broad lines, whereas yellowbars mark mergers and blue objects.

Page 99: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

90 chapter 5: The HI absorption ‘Zoo’

Table 5.1: Busy fit parameters of the H I absorption lines. The horizontal lines separatethe three groups from Fig. 5.2 (a), (b), (c).

Source ID FWHM FW20 Centroid vHI Peak(km s−1) (km s−1) (km s−1) (km s−1)

# 3 82 ± 24 108 ± 26 -243 -209# 4 80 ± 13 135 ± 13 44 78# 5 79 ± 37 127 ± 20 58 89# 10 62 ± 1 96 ± 4 -108 -83# 12 90 ± 9 122 ± 10 18 25# 13 32 ± 1 69 ± 1 34 80# 16 60 ± 10 141 ± 14 356 406# 21 43 ± 6 63 ± 7 -23 19# 30 47 ± 3 72 ± 3 -16 4#32 77 ± 11 115 ± 12 -21 2# 1 122 ± 15 245 ± 16 -22 -20#2 190 ± 9 266 ± 10 104 54#6 119 ± 6 182 ± 6 -39 -34#8 125 ± 4 179 ± 11 -67 -12#9 156 ± 15 190 ± 21 -58 -51#14 146 ± 7 175 ± 9 -50 -81#18 134 ± 294 245 ± 176 59 148#20 138 ± 53 211 ± 80 -134 -134#23 100 ± 1 231 ± 1 -142 -71#24 180 ± 14 275 ± 23 -196 -171#27 101 ± 15 161 ± 54 18 33#7 536 ± 10 825 ± 11 -78 26#11 175 ± 38 301 ± 50 -148 -183#15 370 ± 152 586 ± 72 -285 -114#17 172 ± 42 584 ± 85 -309 -132#19 272 ± 4 416 ± 5 27 28#22 570 ± 2 638 ± 2 85 -95#25 286 ± 56 422 ± 53 -67 -196#26 162 ± 10 461 ± 91 20 139#28 232 ± 32 360 ± 38 -29 -19#29 464 ± 30 674 ± 22 -256 -26#31 358 ± 34 500 ± 7 -36 -149

Page 100: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.4: The nature of H I absorption in flux-selected radio galaxies 91

5.4 The nature of H I absorption in flux-selected radiogalaxies

Using stacking techniques, in Chapter 4 we show that in some of our galaxies H I mustbe distributed in a flattened (disk) morphology, whereas H I has a more unsettled distri-bution in other galaxies of our sample. This is in good agreement with the ATLAS3D

study (Serra et al. 2012) of field early-type galaxies (ETGs). ATLAS3D has shown thatroughly half of the H I detections in ETGs are distributed in a disk/ring morphology,and H I has an unsettled morphology in the other half of the detected cases.

Here, our main goal is to use the BF parameters to identify such disks and unsettledstructures. As mentioned in Sec. 5.3.2, based on the different shapes of the profiles weexpect to find different morphological structures in the three groups separated by thedashed lines in Fig. 5.3.

In Fig. 5.5, 80% of the narrow lines with FWHM < 100 km s−1 are detected closeto the systemic velocity with vHI Peak < ± 100 km s−1. Narrow profiles at the systemicvelocity are most likely produced by large scale disks, as seen in the case of the ATLAS3D

sample of early-type galaxies, where typical FWHM < 80 km s−1 have been found for theH I absorption lines. Previously, H I disks with similar profile characteristics have beenobserved in radio galaxies, e.g. in Cygnus A (Conway 1999; Struve et al. 2010), HydraA (Dwarakanath et al. 1995). Besides disks at the systemic velocity, for narrow lineswe also see one case where the H I peak is redshifted by +406 km s−1 (in source #16).Such narrow lines can be produced by infalling gas clouds with low velocity dispersion.Similar cases of highly redshifted lines have been detected before, e.g. in NGC 315 thenarrow absorption is redshifted by +500 km s−1. Morganti et al. (2009) found that theredshifted H I line in NGC 315 is produced by a gas cloud at a few kpc distance from thenucleus. In 4C 31.04, a neutral hydrogen cloud is detected with 400 km s−1 projectedvelocity towards the host galaxy (Mirabel 1990), whereas in Perseus A the H I absorptionis redshifted by ∼3000 km s−1 (van Gorkom & Ekers 1983; De Young et al. 1973), andits nature is still unclear.

For intermediate widths in Fig. 5.3, H I is still detected close to the systemic ve-locity in most (73%) of the cases, while 27% of the lines are blueshifted/redshifted (2blueshifted, 1 redshifted). In Fig. 5.2 (b) we see that multiple H I components of un-settled gas make the H I kinematics more complex in this group. These are indicationsthat relatively large widths of 100 km s−1 < FWHM < 200 km s−1 can be producedby similar gas structures as narrow detections (disks, clouds), but with more complexkinematics.

Among the broadest lines with FWHM > 200 km s−1, the main H I component isblueshifted/redshifted in 55% of the cases (5 blueshifted and 1 redshifted source). Asmentioned earlier in Sec. 5.3.2, in Fig. 5.4 there are no symmetric lines detected atFW20 > 300 km s−1. The combination of broad widths and lack of symmetry couldsuggest that indeed these profiles are the result of unsettled gas.

Indeed, for these widths we find blueshifted, broad wings, e.g. in 3C 305, where boththe kinematical and spatial properties of the H I indicate the presence of fast, jet-drivenoutflows (Morganti et al. 2005). In fact, when broad and blueshifted wings occur, thecentroid velocity is a better measure of the line offset than the H I peak (with respectto the systemic velocity). To test any connection of the radio power with the H I gasmotions in our sample, in Fig. 5.6 we plot the velocity offset of the H I centroid against

Page 101: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

92 chapter 5: The HI absorption ‘Zoo’

the radio power of the AGN. Below, in Sec. 5.4.1 we discuss the gas and radio sourceproperties of such blueshifted, broad lines.

In the broadest group we find three cases with very broad, multi-peaked H I profiles.In the SDSS images these galaxies are gas-rich mergers and we discuss the nature of theseobjects in Sec. 5.4.3. Despite not being a merger, #31 also has a very broad H I profilewith multiple peaks. In fact, source #31 is an early-type galaxy in the Abell cluster. Thebroad and multi-peaked nature of the profile of #31 is indicative of complex gas motionswithin the galaxy cluster. H I in absorption in clusters has been detected before in Abell2597 (O’Dea et al. 1994; Taylor et al. 1999) and in Abell 1795 (van Bemmel et al. 2012).

5.4.1 Are powerful AGN interacting with their ambient gaseousmedium?

In our sample, we find that the detection rate of blueshifted (vCentroid - vSystemic < -100 km s−1) absorption is relatively high, 29% (9 sources) in Fig. 5.6, whereas a lowerfraction, 6% (2 sources) of the detections are redshifted (vCentroid - vSystemic > +100 kms−1). The blueshift/redshift distribution of H I absorption lines was previously studiedby Vermeulen et al. (2003), who found a similar trend in a sample of GPS and CSS

Figure 5.5: Blueshift/redshift distribution of the H I peak with respect to the systemicvelocity vs. the FWHM of the 32 detected lines. The symbols are the same as in Fig.5.4

Page 102: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.4: The nature of H I absorption in flux-selected radio galaxies 93

Figure 5.6: Blueshift/redshift distribution of the H I line centroid with respect to thesystemic velocity vs. the radio power in the sample of 32 detections. The symbols arethe same as in Fig. 5.4

sources. In the sample of Vermeulen et al. (2003), 37% of the profiles are blueshifted,and 16% are redshifted with respect to the systemic velocity. A later study confirmed thistrend, Gupta et al. (2006) reported a high, 65% detection rate of blueshifted H I profilesin GPS sources. These studies speculate that interactions between the radio source andthe surrounding gaseous medium is the cause of the outflowing gas motions in higherluminosity sources.

The three most blueshifted (vCentroid - vSystemic < -250 km s−1) profiles in Fig. 5.6are broad lines with FW20 > 500 km s−1. By number identifier these are #15, #17 and#29. These detections show similar kinematical properties as the outflows in 3C 293 and3C 305 by displaying broad wings of blueshifted absorption.

The H I outflows in 3C 305 and 3C 293 are driven by powerful radio sources withlog(P1.4 GHz) > 25 W Hz−1 (see Table 5.0). It was estimated that in 3C 305 and 3C293, the kinetic energy output of the jets is high enough to accelerate the gas to highvelocities of about 1000 km s−1 (Morganti et al. 2003, 2005; Mahony et al. 2013). InFig. 5.6 it appears that blueshifted, broad detections in our sample (in #15, #17 and#29) are likely to occur in high power radio galaxies with log(P1.4 GHz) > 25 W Hz−1,suggesting that their energy output is similar to that of 3C 305 and 3C 293. These areindications that interactions with the powerful radio source may be driving H I outflows

Page 103: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

94 chapter 5: The HI absorption ‘Zoo’

Figure 5.7: Blueshift/redshift vs. FWHM distribution of the H I peak with respectto the systemic velocity in compact and extended sources. Yellow squares indicate blue(g − r < 0.7) objects and mergers, as in the previous figures.

in these sources, and we discuss this possibility in more detail in Sec. 5.4.2.Even though the velocity offset of H I lines can be due to infalling/ouflowing gas,

we should not rule out the possibility that in some cases we may be looking at theline-of-sight rotational motion of the gas (Morganti et al. 2001; Vermeulen et al. 2003).

5.4.2 Fraction and time-scale of candidate H I outflowsRadio AGN are thought to be able to drive fast gas outflows through jet-cloud interac-tions. Because such H I outflows are very faint, with typical optical depth of τ = 0.01,until now only a handful of confirmed H I outflows are known (Morganti et al. 1998,2003, 2005, 2013). Besides 3C 305 (also detected here) and 3C 293 (where the broad,blueshifted component is only barely detected here, and not fitted by the BF for lack ofsensitivity), in our sample we have three other cases, where in addition to the main H Icomponent (deep H I detection close to the systemic velocity) a blueshifted shallow wingis seen.

Source 4C +52.37 (source #29) is a CSS source from the CORALZ sample (see morein Sec. 5.5), and we find a broad blueshifted wing in this galaxy with FW20 = 674km s−1 (see Table 5.1). The dataset of the first set of observations of this source ishighly affected by RFI. Thus, in order to verify our detection, we carried out follow-up

Page 104: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.4: The nature of H I absorption in flux-selected radio galaxies 95

Figure 5.8: 1. (Top panel): Asymmetry parameter vs. FW20 distribution of com-pact (red circles) and extended (blue squares) sources. 2. (Bottom panel) Normalizedasymmetry parameter vs. FW20 width distribution of the same groups. Yellow squaresindicate blue (g − r < 0.7) objects and mergers, as in the previous figures. In thehistograms compact sources are marked by red bars, the blue hatched region indicatesextended sources, and yellow bars indicate mergers and blue objects.

Page 105: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

96 chapter 5: The HI absorption ‘Zoo’

observations of 4C +52.37, and confirm the presence of the wing by the second set ofobservations. Source #15 and #17 share similar kinematical properties, showing broadlines of almost ∼590 km s−1 FWHM. Even though in #15 the main, deeper componentis not as prominent as in other two cases, Saikia & Gupta (2003) detected higher degreeof polarization asymmetry in this object (4C +49.25). Saikia & Gupta (2003) argue thatsuch polarization properties can indicate interactions of the radio source with clouds ofgas which possibly fuel the AGN.

Based on their H I kinematical and radio source properties (see Sec. 5.4.1), we considerthese sources the best candidates for hosting jet-driven H I outflows. However, moresensitive and higher resolution observations are needed to verify that these detectionsare indeed jet-driven, and to estimate how much of the energy output is concentratedin the jets. Including the two already known outflows in 3C 305, 3C 293, in our flux-selected sample the detection rate of outflows at the sensitivity of our observations is∼15% among detections, or 5% in all observed radio sources.

Considering the 5% detection rate and the typical lifetime of radio sources (between107 - 108 yr, Parma et al. 1999, 2007), if every radio source goes through a phase ofoutflow, it means that H I outflows last (on average) not more than a few Myr. Thus, theoutflow would appear as a relatively short phase in the life of the galaxy. This is similarto that which is derived from observations of the molecular gas. Using CO observations,Cicone et al. (2014) estimated a gas depletion time of a few million years in a sampleof galaxies hosting powerful AGN. In the case of NGC 1266 from the ATLAS3D sample(Alatalo et al. 2011), the mass outflow rate is 13 M⊙ yr−1, and if the gas in the nucleusis the source of the molecular outflow, the estimated depletion time scale is < 85 Myr.

5.4.3 Gas rich mergers

Among our detections we find three cases of broad (with FW20 of 416 km s−1, 638 kms−1 and 825 km s−1 in increasing order), multi-peaked profiles in UGC 8387, Mrk 273and UGC 05101. The host galaxies of these broad H I detections are gas-rich mergingsystems. In these sources, a combination of AGN and enhanced star-forming regions islikely to be the origin of the radio emission.

In Table 5.0, gas-rich mergers have the highest column densities in the range (5 -8) × 1019 (Tspin/cf) cm−2, reflecting extreme physical conditions of the gas in merginggalaxies. Even though the presence of AGN is not always clear in merging systems, thereexist tentative signs that the presence of gas has an effect on the growth of mergingBHs. Very Long Baseline Array (VLBA) observations of the H I absorption in Mrk 273by Carilli & Taylor (2000) show that the broad H I profile is the result of several co-added components in this source (see notes on individual sources in Appendix 5.8.1).In particular, Carilli & Taylor (2000) detected an infalling gas cloud towards the South-Eastern component (SE) of Mrk 273, indicative of BH feeding processes. Consideringthe broad and multi-peaked nature of the H I in UGC 8387 and UGC 05101, thesesources likely have similar gas properties to Mrk 273, e.g. H I absorption originatingfrom several unsettled components. With our low-resolution observations we can notdistinguish between the different absorbing regions in merging systems, therefore wedetect the blended, broad H I signal.

Page 106: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.5: The H I properties of compact and extended sources 97

Figure 5.9: Blueshift/redshift distribution of the H I line centroid vs. the radio powerof compact (red circles) and extended sources (blue squares). Yellow squares indicateblue (g − r < 0.7) objectsmergers.

5.5 The H I properties of compact and extended sourcesIn Chapter 4 we found that compact sources have higher τ , FWHM and column densitythan extended sources. Here, using the BF parameters we expand on these results byexamining in more detail the difference in the H I properties of the two types of radiosources. In the following analysis we focus on the sample of 27 red (g−r < 0.7) detections.

In Fig. 5.7 and Fig. 5.8 we present the width distribution of compact and extendeddetections. As expected from the stacking results, the busy function analysis shows thatcompact sources tend to have broader lines. In Chapter 4 we suggested that the largerwidth in compact sources is due to the presence of unsettled gas. In Sec. 5.4 we show thatunsettled gas is typically traced by asymmetric lines, furthermore redshifted/blueshiftedlines can also indicate non-rotational gas motions.

In Fig. 5.8, among broad lines with FW20 > 300 km s−1, a high fraction, 88% of thesources, is compact (7 out of 8), while only one source is extended (17%). As mentionedearlier in Sec. 5.4, the lack of symmetric lines suggests that such broad profiles arisedue to non-rotational gas motions of unsettled gas. The largest asymmetry of |∆vCP|∼ 250 km s−1 is measured in the compact source 4C +52.37, one of our H I outflowcandidates. Furthermore, in Fig. 5.7, almost 90% (7 sources out of 8) of blueshifteddetections with vHI Peak < -100 km s−1 are compact sources, whereas only one source

Page 107: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

98 chapter 5: The HI absorption ‘Zoo’

10-3 10-2 10-1 100 101

Linear size (kpc)

1017

1018

1019

1020N(HI) · c f / T

spin [cm

−2 K

−1]

CORALZ non-detectionsCORALZ detections

Figure 5.10: Radio source size vs. column density in GPS and CSS sources from theCORALZ sample

(∼10%) is extended.Fig. 5.8 and Fig. 5.7 show that the traces of unsettled gas, e.g. blueshifted and

broad/asymmetric lines, are found more often among compact sources. This suggests alink between the morphology of the radio source and the kinematics of the surroundinggas. In fact, all three H I outflow candidates (#15, #17 and #29) from Sec. 5.4.2 areclassified as compact in Fig. 5.9 and Table 5.0. Hence, the presence of unsettled gassuggest that interactions between small (< 10 kpc) radio sources and the rich ambientmedium are likely to occur in the young, compact phase of AGN, providing favourablesites for powerful jet-cloud interactions.

As we mention in Chapter 4, nine of our AGN are part of the COmpact RAdiosources at Low redshift (CORALZ) sample (Snellen et al. 2004; de Vries et al. 2009),a collection of young CSS and GPS sources. The de Vries et al. (2009) observationsprovided high resolution MERLIN, EVN and global VLBI observations of the CORALZsample at frequencies between 1.4 - 5 GHz, along with radio morphological classificationand source size measurements.

Previously, H I observations of 18 CORALZ sources were obtained by Chandola et al.(2010), yielding a 40% H I detection rate. Our sample includes fewer, 11 objects, and wefind a 55% detection rate. We observed three CORALZ sources which were not studiedby Chandola et al. (2010). Among the three sources we have two new detections: #20,#21 and one non-detection: #52.

Pihlström et al. (2003); Gupta et al. (2006); Chandola et al. (2010) reported a col-umn density vs. radio source size anti-correlation for CSS and GPS sources, accountedfor the fact that at larger distances from the nucleus, lower opacity gas is probed in frontof the continuum (Fanti et al. 1995; Pihlström et al. 2003). More recently, Curran et al.(2013) pointed out that the N(H I)-radio size inverse correlation is driven by the fact that

Page 108: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.6: Summary 99

the optical depth is anti-correlated with the linear extent of the radio source. In Fig. 5.10we plot the largest projected linear size (LLS) of the sources reported by de Vries et al.(2009) against the column densities measured from our H I profiles. The column densityof the detections decreases as function of radio source size. However, adding the N(H I)upper limit makes the trend much less clear. In fact, high frequency peakers (HFPs) werealso found not to be following the inverse correlation (Orienti et al. 2006). HFP galaxiesare thought to be recently triggered, 103-105 yr old, small radio sources of a few tens ofpc. Orienti et al. (2006) measured low column densities in these tiny sources, and theyexplain the low column density of HFPs by a combination of orientation effects and thesmall size of the sources. In this scenario, our line-of-sight intersects the inner region ofthe torus against the tiny radio source, therefore in absorption we can only detect thishigh Tspin, (and therefore) low column density gas close to the nucleus.

The above results suggest that both orientation effects and the radio source size canbe affecting the measured optical depths, and the combination of these effects may beresponsible for deviations from the N(H I) to radio size inverse correlation also in oursample in Fig. 5.10.

5.6 SummaryIn this paper we presented the results of an H I absorption study of a sample of 101AGN. The relatively large sample of 32 detections has been parametrized using the busyfunction (Westmeier et al. 2014). The total sample was selected and used for stackingpurposes in Chapter 4, and here we carry out a detailed analysis of the individual profiles.Detections in our sample display a broad range of line shapes and kinematics. The busyfunction is efficient in fitting almost all of the spectra, except for a few peculiar caseswith multiple lines and H I emission features.

In Chapter 4 we find that H I disks and unsettled gas structures are both present inour sample. Here we attempt to disentangle different H I morphological structures usingthe busy function parameters. We find that the complexity of the lines is increasing withincreasing profile width. Based on the line shapes we separate three groups of objectswith different kinematical properties. The narrowest lines with FWHM < 100 km s−1 inour sample are most likely produced by large scale disks or H I clouds. Relatively broadlines (100 km s−1 < FWHM < 200 km s−1) may be produced by similar morphologicalstructures with more complex kinematics. Broad lines with FWHM > 200 km s−1,however, are tracing the most unsettled gas structures, e.g. gas-rich mergers and outflows.

We detect three new cases with broad, blueshifted H I wings. Along with their radiosource properties, i.e. powerful AGN with log(P) > 25 W Hz−1, these sources are thebest candidates for being jet-driven H I outflows. Considering certain and tentative cases,the detection rate of H I outflows is 5% in our total sample. If all radio AGN go throughan outflow phase during their lifetime, the relatively low detection rate suggests thatthe gas depletion timescale of H I outflows is shorter than the typical lifetime of radiogalaxies.

In Chapter 4 we show that the stacked profile of compact sources is broader than thestacked width of extended sources. Here we confirm this result using the BF parametersof the individual detections. H I in compact sources often shows the characteristics ofunsettled gas, e.g. blueshifted lines and broad/asymmetric profiles. Such H I line proper-ties suggest that strong interactions between AGN and their rich circumnuclear medium

Page 109: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

100 chapter 5: The HI absorption ‘Zoo’

are likely to occur in compact AGN, as young radio jets are clearing their way throughthe ambient medium in the early phases of the nuclear activity.

5.7 AcknowledgementsThe WSRT is operated by the ASTRON (Netherlands Foundation for Research in As-tronomy) with support from the Netherlands Foundation for Scientific Research (NWO).We thank Tobias Westmeier and Russell Jurek for the useful suggestions on the busy func-tion fitting module. RM gratefully acknowledges support from the European ResearchCouncil under the European Union’s Seventh Framework Programme (FP/2007-2013)/ERC Advanced Grant RADIOLIFE-320745.

Page 110: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.8: Appendix 101

5.8 Appendix5.8.1 Notes on the individual detections#1: B3 0754+401This source is part of the low luminosity CSS sample presented by Kunert-Bajraszewska& Labiano (2010). The largest linear size of the source is 0.25 kpc, and the morphologyremains unresolved in the multi-element radio linked interferometer network (MERLIN)observations. Kunert-Bajraszewska & Labiano (2010) classified this object as a High Ex-citation Galaxy (HEG). We detect H I in this source at the systemic velocity. The widthand asymmetry parameters of the H I suggest that the gas in this galaxy is not entirelysettled.#2: 2MASX J08060148+1906142This source has not been studied individually before. The relatively large width (FW20= 266km s−1) and double-peaked nature of the H I profile suggests that unsettled gas ispresent in this galaxy.#3: B2 0806+35This source shows radio emission both on the kpc and pc scale. It has been observed withVLBA at 5 GHz as part of a polarization survey of BL-Lac objects (Bondi et al. 2004)as a possible BL-Lac candidate. On the parsec scale (beam size of 3.2 × 1.7) the sourcereveals a radio core with an extended jet towards the south. The jet extends for about10 milliarcseconds (mas) (∼ 15 pc at z = −0.0825). Among the BL-Lac candidates, thissource is the weakest object, and has the steepest radio spectrum. It is also the onlyobject not showing polarized emission neither in the jet or the core. In this galaxy, thenarrow H I line is detected at blueshifted velocities with respect to the systemic.#4: B3 0833+442In the CORALZ sample this source is classified as a CSO. However the 1.6 GHz VLBIimage (de Vries et al. 2009) shows a C-shaped radio structure. The LLS of the source is1.7 kpc. Chandola et al. (2010) did not find H I in this source. In our observations, wedetect a narrow H I profile, slightly redshifted from the systemic velocity. The data cubealso reveals (faint) H I emission.#5: B3 0839+458This source has been observed as part of the CRATES sample (Healey et al. 2007). Ithas been classified as a point source with flat spectrum (spectral index α = −0.396). TheVLBA observations at 5 GHz, as part of the VISP survey (Helmboldt et al. 2007), classifyit as a core-jet source. The lobes have sizes of approximately 3.5 mas and are separatedby 6 mas (= 0.2 kpc at z = 0.1919). The radio power of the source is P1.4GHz ∼ 3·1025 WHz−1. In our work, we detect the H I at the systemic velocity of the galaxy (∆v = 57 kms−1). The line is narrow and deep, suggesting that the H I is settled in a rotating diskaround the host galaxy. The line has a slight asymmetry along the blue-shifted edge notfitted by the busy function.#6: Mrk 1226This source has been observed as part of the CRATES sample (Healey et al. 2007). Ithas been classified as a point source with flat spectrum (spectral index α = 0.284). Theobject has also been observed as part of the VISP survey at 5 GHz (Helmboldt et al.2007). The approximate size of the radio source is 15 mas (∼ 8 pc at z = 0.0279). Wedetect H I absorption close to the systemic velocity of the host galaxy. This, along withthe symmetry of the line, suggests that we are tracing neutral hydrogen rotating in a

Page 111: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

102 chapter 5: The HI absorption ‘Zoo’

disk. The host galaxy may have experienced a gas rich merger which has formed the H Idisk.#7: UGC 05101 - IRAS F09320+6134This radio source is hosted by an Ultra-Luminous far-IR Galaxy. The galaxy is under-going a merger event, as also suggested by its optical morphology. The source has alsobeen observed with a ∼ 11.6 × 9.9 mas resolution using VLBI (Lonsdale et al. 2003).These observations show three compact (. 3−4 pc) cores connected by a fainter compo-nent. The size of the overall structure is 48 × 24 pc. These VLBI observations also showthat the radio continuum is dominated by the AGN and not by the starburst activity.The radio power of the source is P1.4GHz ∼ 3 · 1023 W Hz−1. At the resolution of ourobservations, we are not able to disentangle the absorption seen against the differentcomponents: we detect a broad, blended line. The profile is also multi-peaked, reflectingthe unsettled state of the neutral hydrogen disk and of the overall host galaxy. The H Ihas been detected in emission via Effelsberg Telescope observations, through the studyof Polar-Ring Galaxy candidates (Huchtmeier 1997). The emission line is broad andasymmetric, with a peak flux of +2.2 mJy beam−1. The low sensitivity of the spectrumdoes not allow to set further constraints. The H I emission detection has been confirmedby observations with the Nançay decimetric radio telescope, with have higher sensitivity(van Driel et al. 2000).2#8: 4C +48.29This extended AGN is an X-shaped radio source (Jaegers 1987; Mezcua et al. 2011;Landt et al. 2010). We detect a double peaked H I profile close to the systemic velocity.Before Hanning smoothing, the two peaks are more separated, suggesting the presenceof two H I components (one at the systemic and one blueshifted).#9: J105327+205835In the literature there are no records of individual observations of this radio source. TheNVSS and FIRST images suggest that it is a compact source. In our observations, wedetect a broad profile peaked at the systemic velocity and slightly asymmetric towardsblue-shifted velocities. The SDSS image shows that the host galaxy of this source is veryclose to a companion. Past interaction with a companion could explain the presence ofthe H I in the system.#10: 2MASX J112030+273610There are no individual radio observation of this source reported in the literature. Ac-cording to our classification, it is a compact source. The detected H I profile is narrowand blue-shifted with respect of the optical velocity. This may indicate that we are trac-ing neutral hydrogen which is not settled in a rotating disk.#11: 2MASX J12023112+1637414This source has never been studied individually before. According to our classification itis a compact source, showing a shallow, blueshifted profile, indicative of outflowing gas.#12: NGC 4093 - MCG +04-29-02VLA observations reveal compact radio morphology in this source (Burns et al. 1987;del Castillo et al. 1988). We detect a regular H I component at the systemic velocity,likely tracing the kinematics of a rotating disk.#13: B3 1206+469This radio source has been selected as part of the CLASS survey as a possible BL-Lacobject and then classified as a lobe dominated steep spectrum source. This radio source2 The galaxy belongs to the HYPERLEDA catalogue Paturel et al. (2003)

Page 112: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.8: Appendix 103

is extended with a central core and two symmetric lobes oriented in the north-south di-rection. The distance between the lobes is ∼ 4 arcminutes (∼ 550 kpc at z = 0.100). Thespectral index has been measured in the wavelength intervals 1.4 − 4.8 GHz and 1.4 − 8GHz: α4.8

1.4 = 0.04, α81.4 = 0.39; (Marchã et al. 2001). Being extended, the steepness of

the spectrum may be explained by the fact that some of the flux is missed by the VLAobservations at 8.4 GHz. In our observations we detect a narrow and shallow absorptionline close to the systemic velocity.#14: B2 1229+33This extended source was classified as an FR II by Cohen et al. (2004). Based on theSDSS optical spectrum and image, it is a High Excitation Radio Galaxy (HERG). TheH I profile shows a narrow detection at the systemic velocity. Furthermore, a second,redshifted component is also seen. These H I properties suggest the presence of a diskand infalling gas in this object.#15: 4C +49.25The size of this CSS source is 6 kpc (Saikia & Gupta 2003; Fanti et al. 2000). The 5 GHzVLA map reveals a core and two jets on the opposite sides (Saikia & Gupta 2003). Itwas suggested by Saikia & Gupta (2003) that the higher degree of polarization asymme-try in CSS objects, including 4C +49.25, could be the result of interactions with cloudsof gas which possibly fuel the radio source. Indeed, we find a blueshifted, shallow H Icomponent in this source. This could be the result of outflowing gas, induced by jet-ISMinteractions. At the systemic velocity, however, we do not detect H I.#16: 2MASX J125433+185602The source belongs to the Combined Radio All-Sky Targeted Eight GHz Survey (CRATES)sample (Healey et al. 2007). It has been classified as a point source with flat spectrum(spectral index α = 0.282). Observations at 5 GHz, as part of the VLBA Imaging andPolarimetry survey (VISP, Helmboldt et al. 2007), identify this source as a CSO. Itslobes are separated by 7.3 mas (∼ 15 pc at z ∼ 0.0115). We detect H I at redshiftedvelocities compared to the systemic. The line is narrow and asymmetric, with a broaderwing towards lower velocities. The redshift of the line, along with the compactness ofthe source, suggests that the neutral hydrogen may have motions different from simplerotation in a disk.#17: 2MASX J13013264+4634032According to Augusto et al. (2006), this radio source is a point source. We detect afaint, blueshifted H I profile, which can indicate interactions between the AGN and thesurrounding gaseous medium. The radio source is a Blazar candidate in the CosmicLens All Sky Survey (CLASS) and Combined Radio All-Sky Targeted Eight GHz Sur-vey (CRATES) surveys (Caccianiga et al. 2002; Healey et al. 2007). However, it remainsclassified as an AGN by Caccianiga et al. (2002).#18: B3 1315+415VLBI observations of the CORALZ sample (de Vries et al. 2009) reveal complex radiomorphology in this object. The source has a small size of LLS = 5 pc. From the lobeexpansion speed analysis (de Vries et al. 2010), a dynamical age of 130 yr is estimatedin this source. Chandola et al. (2010) detected H I absorption redshifted by +77 km s−1

relative to the systemic velocity, indicating in-falling gas towards the nuclear region. Ourobservations confirm the H I detection.#19: IC 883 - ARP 193 - UGC 8387The host galaxy of this radio source is undergoing a major merger. The galaxy is a Lumi-

Page 113: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

104 chapter 5: The HI absorption ‘Zoo’

nous far Infra-Red Galaxy (LIRG) where LIR = 4.7·1011L⊙ at z = 0.0233 (Sanders et al.2003). The radio source has been observed with e-Merlin (beam size = 165.23 × 88.35mas) and VLBI e-EVN (beam size = 9.20 × 6.36 mas) (Romero-Cañizales et al. 2012).The radio source consists of 4 knots and extends for about ∼ 750 pc. The innermost100 pc of the galaxy show both nuclear activity and star formation. The nuclear activityoriginates in the central core, while the radio emission from the other knots is attributedto transient sources. This galaxy has already been observed in H I in the study of PolarRing galaxy candidates (Huchtmeier 1997). Two complementary observations have beenperformed using the Green Bank Telescope and the Effelsberg Telescope. Due to thedifferent sensitivity of the instruments, the H I has been detected in emission only in theGreen Bank observations (Richter et al. 1994), with a peak flux = 2.4 mJy. In IC 883,CO(1-0) and CO(3-2) are detected by Yao et al. (2003) in the same range of velocities asthe H I emission. The resolution of our observations does not allow the disentanglementof different absorption components. Hence, the H I line in our observations is blended,spanning the same velocity range of the H I seen in emission, and of the molecular gas.The morphology of the absorption line, along with the overall properties of the cold gasdetected in emission, suggests that in this galaxy the cold gas is rotating in a disk, whichis unsettled due to the ongoing merger event.3#20: SDSS J132513.37+395553.2In the CORALZ sample this source is classified as a compact symmetric object (CSO),with largest (projected) linear size (LLS) of 14 pc (de Vries et al. 2009). Our observa-tions show two H I components, one blueshifted, and the other redshifted relative to thesystemic velocity. The newly detected H I profiles suggest that unsettled gas structuresare present in this galaxy, e.g. infalling clouds, outflowing gas.#21: IRAS F13384+4503This galaxy is optically blue (g − r = 0.6), and the SDSS image revels a Seyfert galaxywith late-type morphology. In the CORALZ sample, the radio source is classified as acompact core-jet (CJ) source with two components which are significantly different influx density and/or spectral index (de Vries et al. 2009). The largest linear size of thesource is 4.1 pc. Against the small continuum source, a very narrow H I absorption profileis detected at the systemic velocity, indicative of a gas disk.#22: Mrk 273This object is the host of an ongoing merger. The optical morphology shows a longtidal tail extending 40 kpc to the south (Iwasawa et al. 2011, and references therein).Low-resolution 8.44 GHz radio maps by Condon et al. (1991) show three radio compo-nents, a northern (N), south-western (SW), and a south-eastern (SE) region. The originof the SE and SW component is unclear (Knapen et al. 1997; Carilli & Taylor 2000).The N radio component is slightly resolved in the observations of Knapen et al. (1997);Carilli & Taylor (2000); Bondi et al. (2005), showing two peaks embedded in extendedradio emission. It is thought that the northern component is hosting a weak AGN,however it is also the site of very active star-formation. Using Very Long Baseline Ar-ray (VLBA) observations, Carilli & Taylor (2000) detected H I absorption against the Ncomponent supposedly coming from a disk (showing velocity gradient along the majoraxis), and estimated an H I gas mass of 2 × 109 M⊙. Molecular CO gas of similar amount(109 M⊙) was also detected by Downes & Solomon (1998). Carilli & Taylor (2000) alsodetect extended gas and an infalling gas cloud towards the SE component, suggesting3 The galaxy belongs to the HYPERLEDA catalogue Paturel et al. (2003)

Page 114: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.8: Appendix 105

that the SE component is indeed an AGN. Our low-resolution observations can not dis-tinguish between the different absorbing regions, we detect the blended signal, comingfrom all the H I absorbing regions. The broad H I absorption was also detected with thesingle dish Green Bank Telescope (GBT) by Teng et al. (2013).#23: 3C 293This object is a Compact Steep Spectrum (CSS) radio source, divided in multiple knots(Beswick et al. 2004). It is a restarted radio source, possibly activated by a recent mergerevent (Heckman et al. 1986). Massaro et al. (2010) classify the radio source as FRI. Arotating H I disk has been detected in absorption by Baan & Haschick (1981). WSRTobservations (Morganti et al. 2005) show an extremely broad absorption component atblue-shifted velocities (FWZI = 1400km s−1). VLA-A array observations, with 1.2×1.3arcsec of spatial resolution, identify this feature as a fast H I outflow pushed by the west-ern radio jet, located at 500 pc from the core (Mahony et al. 2013). The radio jet isthought to inject energy into the ISM, driving the outflow of H I at a rate of 8 − 50 M⊙yr−1. The broad shallow outflowing component is also detected. The fit of the spectrumwith the BF identifies the rotating component, while it fails in fitting the shallow wings,highlighting the different nature of these clouds.#24: 2MASX J142210+210554There are no individual observations of this source in the literature. In our classificationscheme, the radio source is compact. The SDSS observations show that it is hosted by anearly-type galaxy. We detect an absorption line, blue-shifted with respect to the systemicvelocity. The line is broad and asymmetric with a smoother blue-shifted edge.#25: 2MASX J14352162+5051233This is an unresolved (U) CORALZ AGN, the size of the radio source is estimated tobe 270 pc. This galaxy has been observed in H I by Chandola et al. (2010), howeverno components were detected. Our observations show a shallow, broad, blueshifted H Iprofile without deep/narrow component at the systemic velocity. Likely we are seeinggas interacting with the radio source.#26: 3C 305 - IC 1065Massaro et al. (2010) classified this source as a high excitation galaxy (HEG) with FRI radio morphology. The profile shows a deep, narrow component, which could be asso-ciated to rotating gas. Furthermore, Morganti et al. (2005) reported the presence of ajet-driven H I outflow in this galaxy. The outflow is also detected in our observations, andit is successfully fitted by the BF. The column density of the outflow is N(H I) = 2 ×1021,assuming Tspin = 1000 K, and the corresponding H I mass was estimated to be M(H I) =1.3 × 107 M⊙ (Morganti et al. 2005). Molecular H II gas was also detected in this sourceby Guillard et al. (2012). However the molecular phase of the gas is inefficiently coupledto the AGN jet-driven outflow.#27: 2MASX J150034+364845In the literature, there is no record of targeted observations of this radio source. Ac-cording to our classification it is a compact source. We detect a deep absorption line(Speak

abs = −5.3 mJy beam−1). The line lies at the systemic velocity of the host galaxy andtraces a regularly rotating H I disk. The line is slightly asymmetric in the blue-shiftedrange of velocities. This asymmetry is not recovered by the BF fit, suggesting that non-circular motions characterize the neutral hydrogen.#28: 2MASX J15292250+3621423We find no individual observations of this source in the literature. In our sample it is

Page 115: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

106 chapter 5: The HI absorption ‘Zoo’

classified as a compact source. We detect H I in this object close to the systemic velocity.However, similarly to the case of source #9 the profile is not entirely smooth.#29: 4C +52.37This source is classified as a compact symmetric object (CSO) in the CORALZ sample.High-resolution observations reveal a core, and jet-like emission on the opposite sides(de Vries et al. 2009). The main H I absorption component in 4C +52.37 was detectedby Chandola et al. (2010), using the Giant Metrewave Radio Telescope (GMRT). Besidesthe main H I line, we detect a broad, shallow profile of blueshifted H I absorption. Thebroad component was not detected by Chandola et al. (2010), most likely because of thehigher noise of the GMRT spectra. The kinematical properties of the newly detectedblueshifted wing are indicative of a jet-driven H I outflow in this compact radio source.#30: NGC 6034This radio source is hosted by a S0 optical galaxy, which belongs to cluster A2151 of theHercules Supercluster. The radio source is extended, with two jets emerging toward thenorth and the south (Mack et al. 1993). The spectrum is flat with no variation of thespectral index (α = −0.65). The line is very narrow and it is centred at the systemicvelocity. This suggests that the H I may form a rotating disk in the host galaxy. Theneutral hydrogen in NGC 6034 has been first detected in absorption by VLA observations(Dickey 1997).4#31: Abell 2147Based on the SDSS optical images, the host galaxy of this source is an early-type galaxywith a very red bulge. Taylor et al. (2007) classified this object as a flat-spectrum radioquasar. The size of the radio source is about 10 mas (∼20 pc at z = 0.1), and themorphology remains unresolved in the 5 GHz VLBA images. Therefore, it is intriguingthat we find a broad H I detection against this very compact radio source. It is likelythat along the line of sight the H I has non-circular motions.#32: 2MASX J161217+282546The radio source is hosted by an S0 galaxy and has been observed with the VLA-A con-figuration by Feretti & Giovannini (1994). At the resolution of the VLA-A observations(1.4 × 1.1 arcseconds, ∼ 0.7 kpc at z = 0.0320), the radio source is unresolved. Wedetect an absorption line at the systemic velocity of the host galaxy, indicative of neutralhydrogen rotating in a disk.

5.8.2 Summary table of non-detections

4 the galaxy belongs to the HYPERLEDA catalogue Paturel et al. (2003)

Page 116: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.8: Appendix 107In

dex

RA

,Dec

zO

ther

nam

eS 1

.4G

Hz

P 1.4

GH

zR

adio

Mor

phol

ogy

τ pea

kN

(HI)

mJy

WH

z−1

1018

(Tsp

in/c

f)cm

−2

#33

07h5

6m07

.1s

+38

d34m

01s

0.21

5605

B3

0752

+38

770

24.9

8C

<0.

041

<7.

4#

3407

h58m

28.1

s+

37d4

7m12

s0.

0408

25N

GC

2484

243

23.9

8E

<0.

009

<1.

7#

3507

h58m

47.0

s+

27d0

5m16

s0.

0987

4569

24.2

3C

<0.

056

<10

.2#

3608

h00m

42.0

s+

32d1

7m28

s0.

1872

39B

207

57+

3210

425

.01

E<

0.02

4<

4.4

#37

08h1

8m27

.3s

+28

d14m

03s

0.22

5235

4724

.84

C<

0.04

7<

8.6

#38

08h1

8m54

.1s

+22

d47m

45s

0.09

5831

194

24.6

5E

<0.

011

<2.

1#

3908

h20m

28.1

s+

48d5

3m47

s0.

1324

4712

424

.76

E<

0.02

6<

4.8

#40

08h2

9m04

.8s

+17

d54m

16s

0.08

9467

190

24.5

8E

<0.

012

<2.

1#

4108

h31m

38.8

s+

22d3

4m23

s0.

0868

8293

24.2

4E

<0.

022

<4.

0#

4208

h31m

39.8

s+

46d0

8m01

s0.

1310

6512

324

.75

C<

0.02

2<

4.1

#43

08h3

4m11

.1s

+58

d03m

21s

0.09

3357

4624

.01

C<

0.05

7<

10.4

#44

08h3

9m15

.8s

+28

d50m

47s

0.07

8961

B2

0836

+29

126

24.2

9E

<0.

023

<4.

2#

4508

h43m

59.1

s+

51d0

5m25

s0.

1263

4479

24.5

2E

<0.

041

<7.

5#

4609

h01m

05.2

s+

29d0

1m47

s0.

1940

453C

213.

116

7026

.25

E<

0.00

1<

0.3

#47

09h0

3m42

.7s

+26

d50m

20s

0.08

4306

8124

.16

C<

0.02

5<

4.5

#48

09h0

6m15

.5s

+46

d36m

19s

0.08

4697

B3

0902

+46

827

324

.69

C<

0.01

3<

2.3

#49

09h0

6m52

.8s

+41

d24m

30s

0.02

7358

5622

.99

C<

0.04

4<

8.0

#50

09h1

2m18

.3s

+48

d30m

45s

0.11

7168

143

24.7

1E

<0.

028

<5.

1#

5109

h16m

51.9

s+

52d3

8m28

s0.

1903

8563

24.8

1E

<0.

054

<9.

8#

5209

h36m

09.4

s+

33d1

3m08

s0.

0761

5261

23.9

4C

<0.

030

<5.

5#

5309

h43m

19.1

s+

36d1

4m52

s0.

0223

36N

GC

2965

104

23.0

7C

<0.

018

<3.

2#

5409

h45m

42.2

s+

57d5

7m48

s0.

2289

389

25.1

4E

<0.

030

<5.

5#

5510

h09m

35.7

s+

18d2

6m02

s0.

1164

7944

24.1

9E

<0.

088

<16

.0

Tab

le5.

2:H

Ino

n-de

tect

ions

.T

heN

(HI)

uppe

rlim

itis

calc

ulat

edfr

omth

e3-

σrm

s,as

sum

ing

100

kms−

1ve

loci

tyw

idth

.

Page 117: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

108 chapter 5: The HI absorption ‘Zoo’

Inde

xR

A,D

ecz

Oth

erna

me

S 1.4

GH

zP 1

.4G

Hz

Rad

ioM

orph

olog

yτ p

ea

kN

(HI)

mJy

WH

z−1

1018

(Tsp

in/c

f)cm

−2

#56

10h3

7m19

.3s

+43

d35m

15s

0.02

4679

UG

C05

771

142

23.3

C<

0.01

3<

2.4

#57

10h4

0m29

.9s

+29

d57m

58s

0.09

0938

4C+

30.1

940

724

.93

C<

0.00

7<

1.4

#58

10h4

6m09

.6s

+16

d55m

11s

0.20

6868

6324

.89

C<

0.03

9<

7.0

#59

11h0

3m05

.8s

+19

d17m

02s

0.21

4267

9925

.12

E<

0.02

4<

4.4

#60

11h1

6m22

.7s

+29

d15m

08s

0.04

5282

7323

.55

C<

0.02

5<

4.6

#61

11h2

3m49

.9s

+20

d16m

55s

0.13

0406

102

24.6

6E

<0.

035

<6.

3#

6211

h31m

42.3

s+

47d0

0m09

s0.

1257

21B

311

28+

472

100

24.6

2E

<0.

027

<4.

9#

6311

h33m

59.2

s+

49d0

3m43

s0.

0316

25IC

0708

168

23.5

9E

<0.

011

<2.

1#

6411

h45m

05.0

s+

19d3

6m23

s0.

0215

96N

GC

3862

777

23.9

2E

<0.

003

<0.

6#

6511

h47m

22.1

s+

35d0

1m08

s0.

0628

94C

GC

G18

6-04

827

624

.42

E<

0.00

6<

1.2

#66

12h0

3m03

.5s

+60

d31m

19s

0.06

5297

SBS

1200

+60

815

324

.2C

<0.

012

<2.

1#

6712

h13m

29.3

s+

50d4

4m29

0.03

0757

NG

C41

8711

023

.38

C<

0.01

9<

3.5

#68

12h2

1m21

.9s

+30

d10m

37s

0.18

3599

FBQ

SJ1

221+

3010

6024

.76

C<

0.04

4<

8.0

#69

12h2

8m23

.1s

+16

d26m

13s

0.22

9959

117

25.2

6E

<0.

023

<4.

3#

7012

h30m

11.8

s+

47d0

0m23

s0.

0390

99C

GC

G24

4-02

594

23.5

2C

<0.

019

<3.

5#

7112

h33m

49.3

s+

50d2

6m23

s0.

2068

4313

525

.22

E<

0.01

5<

2.7

#72

12h5

2m36

.9s

+28

d51m

51s

0.19

5079

B2

1250

+29

430

25.6

7E

<0.

005

<0.

8#

7313

h03m

46.6

s+

19d1

6m18

s0.

0635

1IC

4130

7023

.83

E<

0.03

8<

6.9

#74

13h0

6m21

.7s

+43

d47m

51s

0.20

2562

B3

1304

+44

013

725

.21

E<

0.02

1<

3.8

#75

13h0

8m37

.9s

+43

d44m

15s

0.03

5812

NG

C50

0361

23.2

6C

<0.

031

<5.

6#

7613

h14m

24.6

s+

62d1

9m46

s0.

1308

0572

24.5

1E

<0.

042

<7.

6#

7714

h00m

26.4

s+

17d5

1m33

s0.

0505

59C

GC

G10

3-04

188

23.7

3C

<0.

022

<4.

0#

7814

h00m

51.6

s+

52d1

6m07

s0.

1178

8717

624

.8C

<0.

021

<3.

8

Tab

le5.

2:-c

ontin

ued.

Page 118: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

5.8: Appendix 109In

dex

RA

,Dec

zO

ther

nam

eS 1

.4G

Hz

P 1.4

GH

zR

adio

Mor

phol

ogy

τ pea

kN

(HI)

mJy

WH

z−1

1018

(Tsp

in/c

f)cm

−2

#79

14h0

8m10

.4s

+52

d40m

48s

0.08

2875

176

24.4

8E

<0.

018

<3.

4#

8014

h09m

35.5

s+

57d5

8m41

s0.

1798

7411

425

.02

C<

0.03

5<

6.5

#81

14h1

1m49

.4s

+52

d49m

00s

0.07

6489

393

24.7

5E

<0.

005

<1.

0#

8214

h47m

12.7

s+

40d4

7m45

s0.

1951

46B

314

45+

410

348

25.5

8E

<0.

006

<1.

1#

8315

h04m

57.2

s+

26d0

0m54

s0.

0539

84V

V20

4b19

024

.12

E<

0.01

2<

2.2

#84

15h0

9m50

.5s

+15

d57m

26s

0.18

7373

424

25.6

2C

<0.

005

<0.

9#

8515

h16m

41.6

s+

29d1

8m10

s0.

1298

8272

24.5

1E

<0.

059

<10

.7#

8615

h21m

16.5

s+

15d1

2m10

s0.

2147

7740

525

.74

C<

0.00

7<

1.3

#87

15h2

3m49

.3s

+32

d13m

50s

0.10

999

182

24.7

5C

<0.

016

<3.

0#

8815

h25m

00.8

s+

33d2

4m00

s0.

0815

5959

23.9

9E

<0.

032

<5.

9#

8915

h32m

02.2

s+

30d1

6m29

s0.

0653

3533

23.5

3C

<0.

064

<11

.7#

9015

h39m

01.6

s+

35d3

0m46

s0.

0778

3791

24.1

3C

<0.

019

<3.

5#

9115

h49m

12.3

s+

30d4

7m16

s0.

1115

824C

+30

.29

914

25.4

7E

<0.

004

<0.

8#

9215

h53m

43.6

s+

23d4

8m25

s0.

1176

064C

+23

.42

168

24.7

8E

<0.

034

<6.

2#

9315

h56m

03.9

s+

24d2

6m53

s0.

0425

3892

23.5

9E

<0.

021

<3.

9#

9415

h56m

11.6

s+

28d1

1m33

s0.

2079

2784

25.0

2E

<0.

024

<4.

4#

9515

h59m

54.0

s+

44d4

2m32

s0.

0417

3B

315

58+

448

5823

.37

C<

0.03

6<

6.6

#96

16h0

4m26

.5s

+17

d44m

31s

0.04

0894

NG

C60

40B

7323

.46

C<

0.04

0<

7.3

#97

16h0

6m16

.0s

+18

d15m

00s

0.03

69N

GC

6061

225

23.8

5E

<0.

010

<1.

8#

9816

h08m

21.1

s+

28d2

8m43

s0.

0501

82IC

4590

113

23.8

3E

<0.

016

<3.

0#

9916

h14m

19.6

s+

50d2

7m56

s0.

0602

5881

23.8

5E

<0.

025

<4.

5#

100

16h1

5m41

.2s

+47

d11m

12s

0.19

8625

B3

1614

+47

313

525

.18

E<

0.02

3<

4.2

#10

116

h24m

24.5

s+

48d3

1m42

s0.

0571

0467

23.7

2C

<0.

030

<5.

5

Tab

le5.

2:-c

ontin

ued.

Page 119: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

110 chapter 5: The HI absorption ‘Zoo’

Page 120: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Augusto, P., Gonzalez-Serrano, J. I., Perez-Fournon, I.,& Wilkinson, P. N. 2006, MNRAS,368, 1411

Alatalo, K., Blitz, L., Young, L. M., et al. 2011, ApJ, 735, 88

Allison, J. R., Curran, S. J., Emonts, B. H. C., et al. 2012, MNRAS, 423, 2601

Allison, J. R., Sadler, E. M., & Meekin, A. M. 2014, arXiv:1402.3530

Araya, E. D., Rodríguez, C., Pihlström, Y., et al. 2010, AJ, 139, 17

Assef, R. J., Kochanek, C. S., Brodwin, M., et al. 2010, ApJ, 713, 970

Baan, W. A., & Haschick, A. D. 1981, ApJ, 243, L143

Bahcall, J. N., Kirhakos, S., Saxe, D. H., & Schneider, D. P. 1997, ApJ, 479, 642

Baldwin J. A., Phillips M. M., Terlevich R., 1981, PASP, 93, 5

Becker, R. H., White, R. L., & Helfand, D. J. 1995, ApJ, 450, 559

Beswick, R. J., Peck, A. B., Taylor, G. B., & Giovannini, G. 2004, MNRAS, 352, 49

van Bemmel, I. M., Morganti, R., Oosterloo, T., & van Moorsel, G. 2012, A&A, 548,A93

Best, P. N., Kauffmann, G., Heckman, T. M., et al. 2005, MNRAS, 362, 25

Beswick, R. J., Peck, A. B., Pedlar, A., et al. 2004, The Interplay Among Black Holes,Stars and ISM in Galactic Nuclei, 222, 255

Bolton, R. C., Cotter, G., Pooley, G. G., et al. 2004, MNRAS, 354, 485

Bondi, M., Marchã, M. J. M., Polatidis, A., et al. 2004, MNRAS, 352, 112

Bondi, M., Pérez-Torres, M.-A., Dallacasa, D.,&Muxlow, T. W. B. 2005, MNRAS, 361,748

Booth, R. S., de Blok, W. J. G., Jonas, J. L.,&Fanaroff, B. 2009, arXiv:0910.3935

Page 121: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

112 BIBLIOGRAPHY

van Breugel, W., Miley, G., & Heckman, T. 1984, AJ, 89, 5

Burns, J. O., Hanisch, R. J., White, R. A., et al. 1987, AJ, 94, 587

Caccianiga, A., Marchã, M. J., Antón, S., Mack, K.-H., & Neeser, M. J. 2002, MNRAS,329, 877

Carilli, C. L.,&Taylor, G. B. 2000, ApJ, 532, L95

del Castillo, E., Gavazzi, G.,&Jaffe, W. 1988, AJ, 95, 1340

Chandola, Y., Sirothia, S. K.,&Saikia, D. J. 2011, MNRAS, 418, 1787

Cicone, C., Maiolino, R., Sturm, E., et al. 2014, A&A, 562, A21

Cohen, A. S., Röttgering, H. J. A., Jarvis, M. J., Kassim, N. E., &Lazio, T. J. W. 2004,ApJS, 150, 417

Condon, J. J., Huang, Z.-P., Yin, Q. F.,&Thuan, T. X. 1991, ApJ, 378, 65

Conway, J. E. 1999, Highly Redshifted Radio Lines, 156, 259

Conway, J. E.,&Schilizzi, R. T. 2000, EVN Symposium 2000, Proceedings of the 5theuropean VLBI Network Symposium, 123

Curran, S. J., & Whiting, M. T. 2010, ApJ, 712, 303

Curran, S. J., Whiting, M. T., 2011, MNRAS, 413, 1165

Curran, S. J., Allison, J. R., Glowacki, M., Whiting, M. T.,&Sadler, E. M. 2013, MNRAS,431, 3408

Dasyra, K. M., Combes, F., Novak, G. S., et al. 2014, arXiv:1402.3187

Davis, T. A., Alatalo, K., Bureau, M., et al. 2013, MNRAS, 429, 534

O’Dea, C. P., Baum, S. A., & Gallimore, J. F. 1994, ApJ, 436, 669

Delhaize J., Meyer M. J., Staveley-Smith L., Boyle B. J., 2013, MNRAS, 433, 1398

DeBoer, D. R., Gough, R. G., Bunton, J. D., et al. 2009, IEEE Proceedings, 97, 1507

Dennett-Thorpe, J., & Marchã, M. J. 2000, A&A, 361, 480

Dickey, J. M. 1997, AJ, 113, 1939

Donoso, E., Yan, L., Tsai, C., et al. 2012, ApJ, 748, 80

Downes, D.,&Solomon, P. M. 1998, ApJ, 507, 615

van Driel, W., Arnaboldi, M., Combes, F., & Sparke, L. S. 2000, A&AS, 141, 385

Dwarakanath, K. S., Owen, F. N.,&van Gorkom, J. H. 1995, ApJ, 442, L1

Emonts, B. H. C., Morganti, R., Struve, C., et al. 2010, MNRAS, 406, 987

Page 122: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

BIBLIOGRAPHY 113

Fabello S., Catinella B., Giovanelli R. et al., 2011, MNRAS, 411, 993

Fabello S., Kauffmann G., Catinella B. et al., 2011, MNRAS, 416, 1739

Fanti, R., Fanti, C., Schilizzi, R. T., et al. 1990, A&A, 231, 333

Fanti, C., Fanti, R., Dallacasa, D., et al. 1995, A&A, 302, 317

Fanti, C., Pozzi, F., Fanti, R., et al. 2000, A&A, 358, 499

Feretti, L., & Giovannini, G. 1994, A&A, 281, 375

Gallimore, J. F., Baum, S. A., O’Dea, C. P., Pedlar, A.,&Brinks, E. 1999, ApJ, 524, 684

Geréb, K., Morganti, R., Oosterloo, T. A., Guglielmino, G., & Prandoni, I. 2013, A&A,558, A54

van Gorkom, J. H., & Ekers, R. D. 1983, ApJ, 267, 528

van Gorkom, J. H., Knapp, G. R., Ekers, R. D., et al. 1989, AJ, 97, 708

Grossi, M., di Serego Alighieri, S., Giovanardi, C., et al. 2009, A&A, 498, 407

Guillard, P., Ogle, P. M., Emonts, B. H. C., et al. 2012, ApJ, 747, 95

Gupta, N., Salter, C. J., Saikia, D. J., Ghosh, T.,&Jeyakumar, S. 2006, MNRAS, 373,972

Healey, S. E., Romani, R. W., Taylor, G. B., et al. 2007, ApJS, 171, 61

Heckman, T. M., Smith, E. P., Baum, S. A., et al. 1986, ApJ, 311, 526

Helmboldt, J. F., Taylor, G. B., Tremblay, S., et al. 2007, ApJ, 658, 203

Huchtmeier, W. K. 1997, A&A, 319, 401

Iwasawa, K., Mazzarella, J. M., Surace, J. A., et al. 2011, A&A, 528, A137

Jaegers, W. J. 1987, A&AS, 71, 603

Johnston S., et al., 2008, ExA, 22, 151

Kauffmann, G., Heckman, T. M., Tremonti, C., et al. 2003, MNRAS, 346, 1055

Knapen, J. H., Laine, S., Yates, J. A., et al. 1997, ApJ, 490, L29

Kunert-Bajraszewska, M.,&Labiano, A. 2010, MNRAS, 408, 2279

Kunert-Bajraszewska, M., Gawroński, M. P., Labiano, A., & Siemiginowska, A. 2010,MNRAS, 408, 2261

Labiano, A., Vermeulen, R. C., Barthel, P. D., et al. 2006, A&A, 447, 481

Lah P., et al., 2007, MNRAS, 376, 1357

Page 123: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

114 BIBLIOGRAPHY

Lah P., et al., 2009, MNRAS, 399, 1447

Landt, H., Cheung, C. C.,&Healey, S. E. 2010, MNRAS, 408, 1103

Lonsdale, C. J., Lonsdale, C. J., Smith, H. E., & Diamond, P. J. 2003, ApJ, 592, 804

Mack, K.-H., Feretti, L., Giovannini, G., & Klein, U. 1993, A&A, 280, 63

Mahony, E. K., Morganti, R., Emonts, B. H. C., Oosterloo, T. A., & Tadhunter, C. 2013,MNRAS, 435, L58

Marchã, M. J., Caccianiga, A., Browne, I. W. A., & Jackson, N. 2001, MNRAS, 326,1455

Martel, A. R., Baum, S. A., Sparks, W. B., et al. 1999, ApJS, 122, 81

Mauch, T., & Sadler, E. M. 2007, MNRAS, 375, 931

Massaro, F., Harris, D. E., Tremblay, G. R., et al. 2010, ApJ, 714, 589

Mezcua, M., Lobanov, A. P., Chavushyan, V. H.,&León-Tavares, J. 2011, A&A, 527, A38

Mirabel, I. F. 1990, ApJ, 352, L37

Morganti, R., Oosterloo, T.,&Tsvetanov, Z. 1998, AJ, 115, 915

Morganti, R., Oosterloo, T. A., Tadhunter, C. N., et al. 2001, MNRAS, 323, 331

Morganti, R., Oosterloo, T. A., Emonts, B. H. C., van der Hulst, J. M.,&Tadhunter,C. N. 2003, ApJ, 593, L69

Morganti, R., Oosterloo, T. A., Tadhunter, C. N., van Moorsel, G.,&Emonts, B. 2005,A&A, 439, 521

Morganti, R., de Zeeuw, P. T., Oosterloo, T. A., et al. 2006, MNRAS, 371, 157

Morganti, R., Peck, A. B., Oosterloo, T. A., et al. 2009, A&A, 505, 559

Morganti, R., Fogasy, J., Paragi, Z., Oosterloo, T.,&Orienti, M. 2013, Science, 341, 1082

Murray, C. E., Lindner, R. R., Stanimirović, S., et al. 2014, ApJ, 781, L41

Oosterloo, T.A.; Morganti, R.; Crocker, A. et al.; 2010; MNRAS; 409; 500

Oosterloo, T., Verheijen, M., & van Cappellen, W. 2010, ISKAF2010 Science Meeting

Orienti, M., Morganti, R.,&Dallacasa, D. 2006, A&A, 457, 531

Owsianik, I., & Conway, J. E. 1998, A&A, 337, 69

Parma, P., Murgia, M., Morganti, R., et al. 1999, A&A, 344, 7

Parma, P., Murgia, M., de Ruiter, H. R., et al. 2007, A&A, 470, 875

Paturel, G., Theureau, G., Bottinelli, L., et al. 2003, A&A, 412, 57

Page 124: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

BIBLIOGRAPHY 115

Peck, A. B., Taylor, G. B., & Conway, J. E. 1999, ApJ, 521, 103

Pihlström, Y. M., Conway, J. E.,&Vermeulen, R. C. 2003, A&A, 404, 871

Readhead, A. C. S., Taylor, G. B., Xu, W., et al. 1996, ApJ, 460, 612

Richter, O.-G., Sackett, P. D., & Sparke, L. S. 1994, AJ, 107, 99

Roberts, M. S. 1970, ApJ, 161, L9

Romero-Cañizales, C., Pérez-Torres, M. A., Alberdi, A., et al. 2012, A&A, 543, A72

Rupke, D. S., Veilleux, S., & Sanders, D. B. 2005, ApJS, 160, 115

Saikia, D. J.,&Gupta, N. 2003, A&A, 405, 499

Sanders, D. B., Mazzarella, J. M., Kim, D.-C., Surace, J. A., & Soifer, B. T. 2003, AJ,126, 1607

Sault R. J., Teuben P. J., Wright M. C. H., 1995, ASPC, 77, 433

Serra P., et al., 2012, MNRAS, 422, 1835

Shostak, G. S., van Gorkom, J. H., Ekers, R. D., et al. 1983, A&A, 119, L3

Snellen, I. A. G., Mack, K.-H., Schilizzi, R. T.,&Tschager, W. 2004, MNRAS, 348, 227

Struve, C.,&Conway, J. E. 2010, A&A, 513, A10

Taylor, G. B., O’Dea, C. P., Peck, A. B., & Koekemoer, A. M. 1999, ApJ, 512, L27

Taylor, G. B., Healey, S. E., Helmboldt, J. F., et al. 2007, ApJ, 671, 1355

Teng, S. H., Veilleux, S.,&Baker, A. J. 2013, ApJ, 765, 95

Verheijen M., van Gorkom J. H., Szomoru A., Dwarakanath K. S., Poggianti B. M.,Schiminovich D., 2007, ApJ, 668, L9

Vermeulen, R. C., Pihlström, Y. M., Tschager, W., et al. 2003, A&A, 404, 861

de Vries, N., Snellen, I. A. G., Schilizzi, R. T., Mack, K.-H.,&Kaiser, C. R. 2009, A&A,498, 641

de Vries, N., Snellen, I. A. G., Schilizzi, R. T.,&Mack, K.-H. 2010, A&A, 521, A2

Westmeier, T., Jurek, R., Obreschkow, D., Koribalski, B. S.,&Staveley-Smith, L. 2014,MNRAS, 438, 1176

Yao, L., Seaquist, E. R., Kuno, N., & Dunne, L. 2003, ApJ, 597, 1271

York D. G., et al., 2000, AJ, 120, 1579

De Young, D. S., Roberts, M. S., & Saslaw, W. C. 1973, ApJ, 185, 809

De Young, D. S. 1993, ApJ, 405, L13

Young, L. M., Bureau, M., Davis, T. A., et al. 2011, MNRAS, 414, 940

Veilleux, S., Meléndez, M., Sturm, E., et al. 2013, ApJ, 776, 27

Page 125: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

116 BIBLIOGRAPHY

Page 126: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Chapter 6HI, radio continuum, and opticalproperties of radio galaxies

– K. Geréb, R. Morganti, T. Oosterloo, P. Best –In preparation for publication in Astronomy & Astrophysics

6.1 AbstractWe use WSRT observations to study the H I gas properties of nine Sloan Digital Sky Survey(SDSS) radio AGN at redshifts z < 0.1. Neutral hydrogen absorption is detected towards theoptical core of a compact and a Fanaroff-Riley II type extended radio source. In a third sourcewe find a tentative detection at low optical depth upper limit τ < 0.008.

We detect diffuse extended emission around both sources with certain H I absorption. Weargue that the faint diffuse structure is the residual emission from a previous cycle of AGNactivity. The presence of cold gas in restarted radio sources suggests that H I is (one of) themain fuel for triggering or even rejuvenating the nuclear activity. Thus, signatures of restartedactivity, e.g. faint relic structures, seem to be a good indicator for finding H I absorption inradio galaxies.

In order to increase the statistical significance of our results, based on a literature searchwe construct a larger dataset of radio sources with available H I observations and optical spec-tra from SDSS. Galaxies with young stellar populations tend to show high H I detection rate,suggesting that star formation in radio galaxies is connected with the presence of an H I-richmedium. If gas accretion is a periodic event in radio galaxies, then perhaps the new gas supplycan continuously contribute to the AGN and star formation fuelling processes.

6.2 IntroductionGas accretion onto the central black hole (BH) of galaxies is thought to provide thenecessary fuel supply for (radio) AGN activity. The availability of cold gas in the cir-cumnuclear region is crucial for the evolution of the galaxy as a whole for the followingreasons. Apart from ‘feeding the monster’, the infalling gas may also trigger a central

Page 127: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

118 chapter 6: HI, radio continuum, and optical properties of AGN

starburst and produce young stellar populations. Furthermore, interactions between theradio jets and the ambient medium are likely to result in highly energetic feedback mech-anisms, releasing substantial amounts of energy back into the ISM. Such interactionsoften produce high velocity gas outflows (Morganti et al. 1998, 2005, 2013; Dasyra et al.2014), which may lead to depletion of cold gas reservoirs. Given the importance that theinterplay between the energy released by the active black hole and the gas is consideredto have, it is important to explore the presence of gas and its relation to other propertiesof AGN.

H I absorption offers key diagnostics to study the physical and kinematical conditionof the gas in the circumnuclear region of radio galaxies. The study of the H I has givenin the last years interesting view of the gas properties in one particular type of AGN:the radio loud. Although most of the H I observations were done in quite powerful AGN(i.e., relatively rare AGN) these studies have provided relevant information on the char-acteristics of the radio sources and their relation with the gas content. In particular, aclose connection between the evolutionary stage of the radio source and the H I contenthas been pointed out by several studies (van Gorkom et al. 1989; Pihlström et al. 2003;Gupta et al. 2006; Emonts et al. 2010; Chandola et al. 2011). The detection rate of H Iseems to be particularly high in young radio sources (compact steep spectrum and giga-hertz peaked spectrum sources), and this has been interpreted as a signature that coldgas plays a prime role in the AGN triggering processes.

H I observations can also help to time AGN triggering events in the following way.Some of the young radio sources were found to contain large amounts of H I (∼1010

M⊙), distributed in many cases in the form of regular structures (e.g. large discs andrings with sizes up to 200 kpc (Struve et al. 2010; Emonts et al. 2010). These discs arebelieved to form by galaxy mergers or by accretion of cold gas from the intergalacticmedium, which processes are also considered to be involved in the triggering mechanismof nuclear activity (Smith & Heckman 1989; Tadhunter et al. 1989; van der Hulst et al.2004). The time scale for the formation of such large discs is about 1 Gyr, whereas AGNactivity in the largest radio-loud galaxies is relatively short, 107 − 108 years. Therefore,the timescales suggest that either there is a reasonable time delay between the formationof these structures and the triggering of AGN, or the two processes are not related. Thelatter case may suggest that radio activity is a common, sometimes recurrent period inthe evolution of (all) early-type galaxies.

One important questions regarding our understanding of active nuclei is whether AGNactivity is usually episodic and if so, what is the cycle of the activity. After the centralnucleus switches off, for lack of fuelling the lobe structure will fade away, however for alimited time we can identify these sources through their relic radio emission. Periodic gasaccretion may lead to episodic activity of the BH and, interestingly, cold gas seems to befrequently present also in restarted radio sources. This is a rather puzzling result, as itis expected that cold gas reservoirs are depleted by feedback effects during the previouscycle of activity. The main limitation of these studies is that at the moment only ahandful of such radio relics have been found, and even fewer cases are known where bothcontinuum and spectral line measurements are available.

As the rejuvenation of the nuclear activity is likely connected with the presence of newgas supply, it is also interesting to investigate whether H I has also other roles in the evo-lution of radio galaxies. Earlier studies suggest that about 30% of powerful radio sourcesshow the presence of young stellar populations (Lilly & Longair 1984; Smith & Heckman

Page 128: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.3: Sample selection 119

1989; Aretxaga et al. 2001; Tadhunter et al. 2002). Therefore, it is interesting to probethe presence of H I in connection with star formation and AGN triggering processes inradio galaxies.

An indicator of the presence of young stars in addition to the passively evolving old(12.5 Gyr) component is the 4000 A break (Tadhunter et al. 2002, 2005). Kauffmann et al.(2003) showed that values of this break in the range D(4000) ∼ 1.4 − 1.5 are indicator ofthe presence of a stellar component with ages less than ∼ 1 Gyr. If the star-formationactivity is triggered by the same event as the AGN, then the age of the stellar populationscan provide important information on the evolution of radio sources and the timescalesof reactivation.

It is clear that the picture of galaxy evolution, including the onset and cycle of nuclearactivity, is not complete without our knowledge of the presence and distribution of gas.Given the growing interest in the role of the cold gas as tracer of AGN feeding andfeedback processes, it is important to explore the relation of H I to other properties, e.g.the cycle of AGN activity, presence of young stellar populations. However, it is also clearthat H I alone cannot give the full view. Sensitive, high resolution continuum observationsare needed for tracing the signatures of past activity. Furthermore, complementaryoptical information needs to be available to characterize the presence of ionized gas andthe stellar properties of galaxies for identifying young stellar populations associated withgas accretion events. Therefore, the next step is the systematic investigation of the gascontent of different types of AGN based on their radio and optical characteristics.

A way to do this, is to use radio sources obtained from the cross-correlation withoptical, spectroscopical catalogs. Best et al. (2005) created a sample of ∼ 3000 radio-luminous galaxies (2215 radio-loud AGN, and 497 star-forming galaxies) brighter than 5mJy by cross-comparing NVSS (National Radio Astronomy Observatories (NRAO) VeryLarge Array (VLA) Sky Survey) and FIRST (Faint Images of the Radio Sky at Twentycentimeters) surveys with the Sloan Digital Sky Survey (SDSS, York et al. 2000) seconddata release.

In this paper we present H I studies of a pilot sample of nine galaxies selected from thesample of Best et al. (2005). We explore H I absorption in relation to the radio continuumand optical characteristics of the host galaxy. In particular, we focus on the connectionof H I with relic radio structures, optical line ratio diagnostics, presence of young stellarpopulations. To increase the significance of our conclusions, we also collect a sample ofradio sources which have available H I observations in the literature. With this studywe aim to understand in advance what are the relevant parameters to be explored withfuture H I absorption surveys.

Throughout this paper the standard cosmological model is used, with parameters ofΩm=0.3, Λ=0.7 and H0 = 70 km s−1 Mpc−1.

6.3 Sample selectionFor this study we have selected a sample of compact and extended sources. The radioclassification of extended sources in the dataset by Best et al. (2005) is based on thestandard FR classification scheme (Fanaroff & Riley 1974). FR II - s are powerful radiosources with relativistic jets and edge-brightened lobes. A source is classified as an FRII if the peak radio flux compared to the flux at the end of the radio lobe is > 50 %.This type of radio galaxy is usually found at relatively high redshifts and, therefore, they

Page 129: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

120 chapter 6: HI, radio continuum, and optical properties of AGN

Sour

ceO

ther

nam

eR

ight

asce

nsio

nD

eclin

atio

nz

log

P1.

4GH

zR

adio

log(

NII

/Hα

)lo

g(O

III/

Hβ)

D(4

000)

num

ber

hm

s

′′′

(W/H

z)cl

assifi

catio

nSD

SS1

B2

0828

+32

0831

27.4

9+

3219

26.6

0.05

115

22.7

5FR

II0.

200.

591.

86SD

SS2

-11

5011

.28

+53

4320

.70.

0603

223

.23

FRII

-99

-99

2.07

SDSS

3-

0846

32.4

4+

2935

55.2

0.07

010

23.5

3FR

II0.

280.

182.

05SD

SS4

-10

2200

.78

+44

5144

.10.

0829

524

.40

Inte

rm.

0.28

0.24

1.98

SDSS

5-

1145

06.5

2+

5338

52.8

0.06

901

23.1

1In

term

.0.

23-0

.03

2SD

SS6

4C29

.30

0840

02.3

5+

2949

02.6

0.06

485

24.3

5FR

II0.

060.

931.

42SD

SS7

-15

4228

.35

+52

5950

.70.

0690

423

.45

Com

pact

-0.0

10.

341.

9SD

SS8

-13

1739

.19

+41

1545

.60.

0661

624

.42

Com

pact

0.20

0.14

2.03

SDSS

9-

1518

38.8

9+

4045

00.2

0.06

516

23.6

5C

ompa

ct0.

110.

281.

68

Tab

le6.

1:O

ptic

alid

entifi

catio

nan

dra

dio

char

acte

ristic

sof

the

obse

rved

sam

ple.

The

radi

opo

wer

ofth

eso

urce

spr

esen

ted

inC

ol.

6is

take

nfr

omth

eN

VSS

cata

log.

Opt

ical

line

ratio

sin

Col

.7

and

Col

.8

extr

acte

dfr

omth

eSD

SSca

talo

g.In

case

alin

eha

sS/

N<

3,th

era

tiois

set

to-9

9.

Page 130: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.3: Sample selection 121

Figure 6.1: Continuum images of the observed sample extracted from the FIRST survey,overlaid with 3σ contours of our WSRT observations. In case of SDSS6, the contour levelsare adjusted to get a better definition of the extended emission. In the first row we showSDSS1, SDSS2, and SDSS3 from the left to the right. The other images follow accordinglyin increasing order.

tend to be underrepresented in studies of H I in radio sources. To facilitate the inclusionof FR II galaxies, we selected our objects at relatively high redshifts 0.05 < z < 0.09.Other selection criteria were: 1011M⊙ < stellar mass < 5 × 1011M⊙, and δ > 25. Theradio sources are brighter than S1.4GHz > 30 mJy.

According to this classification scheme, our sample contains four FR II, and twointermediate type sources with only one hotspot. The remaining three sources showcompact morphology. The continuum images of the sources are presented in Fig. 6.1. Thetargets of the observations and some of their characteristics (including radio morphologyclassification) are listed in Table 6.1.

Page 131: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

122 chapter 6: HI, radio continuum, and optical properties of AGN

Source number Beam Size PA rms continuum rms linecube(′′) () (mJy/beam) (mJy/beam)

SDSS1 27.80 × 13.95 1.1 0.059 0.315SDSS2 17.28 × 13.15 0.7 0.049 0.322SDSS3 30.80 × 13.77 0.8 0.054 0.322SDSS4 17.35 × 10.74 0.1 0.063 0.325SDSS5 19.47 × 14.80 0.8 0.064 0.375SDSS6 32.27 × 13.95 -2.5 0.058 0.351SDSS7 19.75 × 14.90 0.9 0.050 0.315SDSS8 26.49 × 11.19 11.5 0.065 0.331SDSS9 23.65 × 14.02 1.2 0.037 0.339

Table 6.2: Parameters of the continuum images and line cubes. Col. 1 refers to theserial number of the sources in our sample.

6.4 Observations and data reductionThe observations were obtained with the Westerbork Synthesis Radio Telescope (WSRT).The nine sources were observed in April 2006, with 12 hour exposure times for everysource. The observational setup consists of 20 MHz bandwidth covered by 1024 fre-quency channels. All cubes were made by adding 4 channels together, yielding a velocityresolution of 15.6 km/s before, and 27 km/s after Hanning smoothing.

The data was reduced using the MIRIAD package. After initial flagging and bandpasscalibration, the continuum was subtracted by fitting a first or second order polynomialto the line-free channels. In case an H I detection was found, the cubes were cleaned,restored, Hanning smoothed, and the velocity was set to optical. If no sign of H I wasfound, the cubes were only Hanning smoothed but not cleaned.

For the line cubes different weights were used in order to obtain the best sensitivityfor detecting absorption/emission. In order to detect H I in absorption we used robustweighting. To search for emission in the environment of the central sources (see Section6.5) we used natural-weighted cubes corresponding to the highest S/N ratio.

The continuum images of the sources were created using the line-free channels, latercleaned and self-calibrated in order to obtain a satisfactory map. Table 6.2 contains theparameters of the continuum images and line cubes.

6.5 ResultsWe detect H I absorption towards the optical core in two sources. For a third source wehave a tentative detection, in which case the H I profile appears towards the northernlobe. We also detect an emission feature South-East from the core. The H I profiles andparameters of the detected lines are presented in Fig. 6.2, 6.3 and Table. 6.3.

One of the detected sources (SDSS6, 4C29.30) has been reported in the meantimeby Chandola et al. (2010). The other detections are new. The second detection is foundin SDSS8, a compact source. Interestingly, both detected sources show extended lowsurface brightness emission, that will be discussed in Sec 6.6. The tentative detection

Page 132: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.5: Results 123So

urce

S 1.4

GH

z,co

reS 1

.4G

Hz,

lob

eH

Ide

tect

ion

τN

(HI)

∆V

FWH

Mnu

mbe

r(m

Jy)

(mJy

)10

18(T

spin

/cf)

cm−

2(k

m/s

)km

s−1

SDSS

19.

16-

<0.

103

<0.

18SD

SS2

19.7

2-

<0.

049

<0.

09SD

SS3

28.8

0-

<0.

033

<0.

06SD

SS4

149.

83-

<0.

006

<0.

0194

.03

?0.

008

0.85

+45

72SD

SS5

11.1

2-

<0.

101

<0.

183

SDSS

622

1.8

+0.

065

16.3

+62

112

SDSS

724

.43

-<

0.03

8<

0.06

SDSS

825

1.05

+0.

032

7.39

+13

610

4SD

SS9

44.3

6-

<0.

022

<0.

04

Tab

le6.

3:Pa

ram

eter

sof

the

HI

abso

rptio

nin

the

obse

rved

sam

ple.

The

S1.

4GH

z,co

reflu

xde

nsity

ism

easu

red

inth

eop

tical

core

ofth

eso

urce

s.In

case

ofSD

SS4,

the

tent

ativ

ede

tect

ion

appe

ars

agai

nst

the

nort

hern

lobe

,the

refo

reth

eflu

xde

nsity

ofth

elo

beat

the

posit

ion

ofth

eab

sorp

tion

isal

soin

dica

ted

with

the

corr

espo

ndin

gpa

ram

eter

s.T

heup

per

limits

inC

ol.

5ar

eca

lcul

ated

from

3σ/S

1.4G

Hz,

core

.

Page 133: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

124 chapter 6: HI, radio continuum, and optical properties of AGN

Figure 6.2: H I detections. The figures show the H I absorption from the nuclear regionof SDSS6 (above) and SDSS8 (below). The systemic velocity is marked by a verticalsolid line in both figures.

of H I absorption is found in SDSS4. The optical depth of the absorption is 0.0065 inSDSS6 and 0.022 in SDSS8. The optical depth of the tentative detection in SDSS4 is0.008 at the 2.5-σ limit (Table. 6.3).

For the calculation of column densities (Table 6.3) we used the formula:N(H I) (cm−2) = 1.8216 · 1018 × Tspin ×

∫τ(v)dv

where v is the velocity, τ(v) is the optical depth (or in case of non-detections the 3σ upperlimit), and Tspin is the spin temperature. The latter temperature can be effected by threefactors: collisions, Lyα photons, and the absorption of 21 cm continuum radiation. If

Page 134: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.5: Results 125

Figure 6.3: Tentative absorption against the radio lobe in SDSS4. As in Fig. 6.2, thesystemic velocity is marked by a vertical solid line.

the H I absorption arises from cold, dense, atomic regions, the spin temperature shouldbe very similar to the kinetic temperature of such a medium Tspin ≈ Tkinetic ∼ 100 K.

Table 6.3 shows that H I spans a broad range of column densities betweenfew × 1017 − 1019 (Tspin/cf) cm−2, where cf is the covering factor of the gas. Simi-lar H I column densities were detected by previous studies of larger radio galaxy sam-ples (Allison et al. 2012; Gereb et al. 2014). This is likely related to the fact that ra-dio AGN are typically hosted by early-type galaxies. H I emission studies of SAURON(Morganti et al. 2006; Oosterloo et al. 2010a) and ATLAS3D (Serra et al. 2012) early-type galaxies show a broad range of H I masses, column densities, and kinematics, ac-counted for the fact that H I in ETGs is of external origin. H I can be transported byaccretion from the intergalactic medium or by merger events, and this observationalresult is also in good agreement with AGN triggering theories.

The Full Width Half Maximum (FWHM) of the detected lines is 72−112 km s−1 (seeTable 6.3). The absorption peak in SDSS4 and SDSS6 is detected close to the systemicvelocity, with velocity offsets similar or lower than the 3-σ error on the SDSS redshift(∼60 km s−1). As it is explained in Gereb et al. (2014a, 2014b submitted), such narrowlines of the order of 100 km s−1 at the systemic velocity are likely produced by rotatingH I disks. In SDSS8, the peak is redshifted by 136 km s−1, suggesting that an H I cloudis falling in toward the core of this source.

In the remaining sources no H I absorption has been detected optical depth upperlimits around a few percent (see Table 6.3). However, the optical depth upper limit inSDSS1 and SDSS5 is relatively high, 10%. The high upper limits in these sources aredue to the weak cores revealed once imaged at higher spatial resolution than NVSS, seeTable 6.3. This is a important consideration to keep in mind for future studies.

Among the sources undetected in H I, two are worth a few extra comments. The

Page 135: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

126 chapter 6: HI, radio continuum, and optical properties of AGN

lobe structure of SDSS1 (B2 0828+32) has been studied in more detail in previous works(Ulrich & Ronnback 1996). SDSS1 is one of the few X-shaped galaxies found so far,with two radio lobes of different ages and orientation. As a possible explanation for theX-shape, Ulrich & Ronnback (1996) suggest a change in the orientation of the jets, theage of the older lobes being 70 Myr. No sign of H I absorption or emission was foundneither in the nuclear region, nor in the other parts of the galaxy. However, we remindthe reader that high H I detection limit is set in this source by the weak AGN core.

SDSS9 is a flat-spectrum compact radio source selected by Caccianiga et al. (2001)to investigate the hypothesis that low-power AGN, usually classified as radio-quiet, canproduce relativistic jets. They suggest that SDSS9 could be a radio AGN whose rela-tivistic jets are viewed close to end-on. Their finding of a high-brightness temperaturecore partly support this hypothesis. No signs of H I absorption was found in this source,but two companion galaxies contain H I emission (see Sec 6.5.4 and Table 6.4).

Below we briefly comment on the detections.

6.5.1 SDSS6: 4C 29.30

In our sample the extended SDSS6 source is detected with an optical depth of ∼0.06, wellabove the 3σ detection limit. Apart from the H I we detect diffuse continuum emissionaround this source. The diffuse feature was previously observed by Jamrozy et al. (2007),who suggest that 4C 29.30 is a restarted FR II radio source, where the relic emission isthe signature of the past activity cycle. Previously, Chandola et al. (2010) reported thedetection of a redshifted H I component in this source, and they suggest that infalling H Iclouds provide fresh supply to rejuvenate the activity of the AGN. The redshifted line istentatively detected at low significance in our observations.

SDSS6 has the highest column density H I in our sample (see Table. 6.3). The lowD(4000) index (see Table 6.1) and the presence of strong optical lines in the spectrumof SDSS6 (see Fig. 6.10) indicate that the host galaxy is actively forming stars. Thepresence of young stellar populations (YSPs) in this source suggest that in SDSS6 starsare likely to form in dense regions of cold gas, rich in H I similarly to what has beenfound in SAURON and ATLAS3D early-type galaxies.

6.5.2 SDSS8

SDSS8 is the second detection and, again, we detect low brightness radio emission aroundthe central compact source. In SDSS8 the peak is redshifted by 136 km s−1. This cansuggest that H I in this source is being accreted through gas infall, and this processtransports new supply of gas to reactivate the nucleus.

The host galaxy shows an edge-on stellar disk in the SDSS optical image, and alongwith the lenticular appearance of the host, it is likely that SDSS8 belongs to the S0morphological class. The column density of the H I is relatively high, although no clearsignature of YSPs is detected in this source. However, in red galaxies like the host ofSDSS8 it is also possible that the light of young stars is masked by more luminous oldstellar populations (Tadhunter et al. 2002).

Page 136: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.6: Rejuvenated radio AGN 127

6.5.3 SDSS4The source with tentative detection (SDSS4) is an intermediate radio galaxy with onlyone hot spot. In this AGN the H I absorption is seen against the northern lobe (the lobeof the hot spot). Even though smaller patchy structures (e.g. clouds) could producesuch absorption at larger distances from the nucleus, here we consider the existence of anextended (∼ 40 kpc diameter) gaseous circumnuclear disc more likely; H I in this galaxyis detected at the systemic velocity, suggesting co-rotation of the H I with the host. Suchextended discs are quite rare, there are only a few cases reported. Morganti et al. (2002)detected a similar absorption feature against the lobes in Coma A. They explain theoccurrence of H I absorption at such large distances from the central region as beingproduced by a gaseous disk-like structure of at least 60 kpc in diameter, made up byneutral and ionised gas. Similar neutral gas distribution extending up to tens of kpc wasdetected in the source 3C 234 by Pihlström (2001).

We also detect tentative H I emission in this source at the ∼3-sigma level South-Eastfrom the core. The emission line is redshifted by +490 km s−1, suggesting that the sourceof H I emission is possibly not associated with the galaxy.

6.5.4 Galaxies detected in H I emissionIn several cases we detect galaxies with H I emission in the environment of the SDSSsources. The properties of these galaxies are presented in Table 6.4. The presence of H I-rich companion galaxies interacting with the central source is considered in some casesto be responsible for triggering the AGN activity (see e.g. Keel et al. 2006; Emonts et al.2008a for some examples).

For distances larger than 350 kpc the timescale of interactions between sources is& 109 year. The timescale of the radio loud phase is instead shorter, .108 years. Thus,galaxies above this distance can not be responsible for triggering the nuclear activity, andonly surrounding galaxies within 350 kpc are considered as companion in our sample.Only one galaxy can be classified a real companion based on the above criteria. Thisgalaxy is observed in the environment of SDSS3 within only 32 kpc distance, althoughno sign of interaction with SDSS3 is seen in H I. The other galaxies with H I emission areat distances larger than 400 kpc (see Table 6.4).

6.6 Rejuvenated radio AGNWe used our relatively deep WSRT continuum images to search for relic radio emission.Interestingly, the two H I-detected sources in our sample show diffuse emission featuresaround their central radio galaxy (Fig. 6.1).

The presence of the faint continuum feature in SDSS6 is confirmed by other obser-vations (Jamrozy et al. 2007; Chandola et al. 2010). The angular size of the extendedstructure is 520 arcsec (639 kpc), inside which a 29 arcsec (36 kpc) double-lobed FRII galaxy is embedded. Jamrozy et al. (2007) consider this source a rejuvenated radiogalaxy, with the diffuse halo being the residual of an earlier cycle of activity. The innerradio source has an estimated spectral age of .33 Myr, while the spectral age of the dif-fuse extended emission is &200 Myr (Jamrozy et al. 2007). Although we do not have the

Page 137: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

128 chapter 6: HI, radio continuum, and optical properties of AGN

Associated RA Dec z ∆V H I mass Distance fromgalaxy (h m s) ( ′ ′′) (km/s) (109M⊙) central source (kpc)SDSS3 08 46 28.1 29 35 27.1 0.070862 +18.24 8.4 32.08SDSS7 15 42 7.7 53 05 14 0.07094 -315.17 6.1 412.06SDSS8 13 17 47 41 09 17 0.067379 -7.56 2.4 480.91SDSS9 15 19 19 40 51 46.7 0.065252 -28.16 - 490.72

15 18 30.8 40 36 52 0.06555 -85.65 12.3 578.11

Table 6.4: Parameters of H I emission detections in the environment of the SDSS galax-ies. Col. 1 defines the SDSS galaxy from the sample around which the emission isdetected.

same wealth of information for SDSS8, from the morphology of the diffuse, low surfacebrightness continuum we suggest that it has a similar, relic-like origin.

The D(4000) index of SDSS6 is ∼1.4, corresponding to ∼ 1 Gyr in stellar age(Kauffmann et al. 2003). This suggests that apart from feeding the black hole, coldgas also can contribute to the evolution of powerful radio galaxies by producing youngstars. If new stars are produced by the same effect as the triggering of the AGN activity,then the timescales (compared to the estimated age of the lobes) suggest that the YSPin SDSS6 was formed in the previous cycle of activity. In this case, the detection of H Iin SDSS6 would suggests that new supply of cold gas has been accreted over the last 1Gyr.

The detection of H I in both sources with relic diffuse emission supports the sugges-tion made by Saikia & Jamrozy (2009); Jamrozy et al. (2007); Chandola et al. (2010);Shulevski et al. (2012) that episodic activity in radio AGN is connected with the pres-ence of cold gas. Considering previous results (Chandola et al. 2010; Fernández et al.2010; Shulevski et al. 2012), relic structures could be good indicators for the presence ofH I with the possibility that H I gas is connected to the fuelling of the recent phase ofactivity. The number of known sources with similar properties is still low, and in orderto understand the role of H I in the life cycle of radio AGN, large systematic studies areneeded with available H I and sensitive continuum measurements at high resolution.

6.7 H I and optical propertiesTo complement the information about the H I, we have also explored the optical propertiesof the sources. The SDSS database contains information which can be used to investigatethe stellar populations, star-formation properties of the objects, or to distinguish betweennormal star-forming galaxies and optical AGN. The SDSS optical images and spectra ofeach object are presented in Fig. 6.9, 6.10 and in Table 6.1.

Fig. 6.4 shows the location of our galaxies in the BPT diagram(Baldwin, Philips & Terlevich 1981). The dashed line marks the theoretical limit betweenAGN and normal star-forming galaxies, as above this line AGN provide a substanitalcontribution to the line fluxes (Kewley et al. 2001). The horizontal and vertical solidlines separate high ionization galaxies from LINERs (low-ionization nuclear emissionregion) (Kauffmann et al. 2003). The objects from the observed sample are marked in

Page 138: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.7: H I and optical properties 129

Figure 6.4: BPT line ratio diagram. Our sample is marked by red symbols, sourceswhich were collected from the literature are indicated in black (see Sec 6.8). Filledsymbols indicate H I detections (certain and tentative), whereas empty symbols mark H Inon-detections. SDSS5 is marked by an empty triangle in this plot because we only havean upper limit on the [NII]/Hα ratio of this galaxy. Filled triangles mark tentative H Idetections.

red, while the black symbols are objects taken from the literature and will be discussedin the next session.

For lines with S/N < 2, the line ratios are very poorly constrained, therefore shouldbe completely ignored, or used with precaution. For SDSS2 none of the four emissionlines are detected with S/N > 2, therefore this object is not included in the BPT diagram.In case of SDSS5, the [NII], Hβ and [OIII] lines are all detected with a S/N > 2, andalthough its Hα line has S/N just below 2, in the SDSS catalog one can find an upper limitof 0.23 set for its [NII]/Hα line ratio. This way SDSS5 belongs to LINER galaxies in theBPT diagram, though the low value of [OIII]/Hβ already indicated its type. According tothe diagram, SDSS1 and SDSS6 are high ionization AGN belonging to the Seyfert group,the rest are LINERs. It is worth mentioning that at the lower resolution of NVSS,SDSS1 and SDSS6 are the most powerful radio sources in our sample (with log(P1.4GHz)= 25.09 and 24.77 W Hz−1 respectively). This is consistent with the known correlationbetween radio continuum and optical line luminosity (Tadhunter et al. 1998). Accordingto the radio power (Table 6.1) and optical SDSS spectra of the sources, SDSS2 andSDSS3 appear less powerful with fluxes relatively low for FR II-s, and with optical linestrengths more characteristic for FR I-s. These could be examples of weak emission lineradio galaxies, i.e. powerful sources (FRII) with weak lines (like FRI) (Hardcastle et al.2006; Buttiglione et al. 2010).

Page 139: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

130 chapter 6: HI, radio continuum, and optical properties of AGN

0.00 0.05 0.10 0.15 0.20 0.25 0.30z

1.0

1.5

2.0

2.5D(4000)

D(4000) index

Figure 6.5: D(4000) index as function of redshift. Symbols are the same as in Fig. 6.4.The three horizontal lines represent the values of the break for which 100, 50 and 10percent (in order solid, dashed, dotted line) of the light below a rest-frame wavelengthof 4000 A arises from young stars (Tadhunter et al. 2002).

Signatures of star formation activity are often found in H I-detected early-type galax-ies. For example, in a sample of powerful 3CR radio sources Tadhunter et al. (2002) foundthat about 30% of the galaxies contains a young stellar component. Low-luminosity AGNare mostly found to have stellar populations similar to normal early-type galaxies, whilehigh-luminosity AGN have younger stellar populations (Kauffmann et al. 2003). SDSSprovides the D(4000) index for our nine sources, and we use this parameter to investigatethe YSPs in our galaxies in Fig. 6.5 (red symbols). Above the solid line one can onlyfind old stars, at the limit of the dashed line only 50% of the light is produced by oldstellar populations, while below the dotted line only 10 % is represented by old stars.

All our galaxies show UV excess compared to passively evolving elliptical galaxies,SDSS6 having the lowest D(4000) index. Since the high UV excess of SDSS6 indicates thepresence of YSPs in this source, and along with the detections of H I this result supportsa connection between starburst activity and the presence of H I gas. However, not allgalaxies with H I detections show the presence of starbursting YSPs. This is, again, ingood agreement with SAURON and ATLAS3D results. In the latter studies it was suggestedthat high column density gas is needed for star formation to occur in early-type galaxiesand, in fact, SDSS6 shows the highest column density in our sample in Table 6.3.

Page 140: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.8: Results from the literature 131

6.8 Results from the literatureIn order to better investigate the relation between H I, AGN and the host galaxy, wehave constructed a database of radio galaxies with available H I absorption observations.The data are collected from works by Briggs et al. (1993); Conway & Blanco (1995);Dwarakanath et al. (1995); Emonts et al. (2008a,b, 2010); van Gorkom et al. (1989),Gupta et al. (2007); Jaffe (1991); Kanekar & Chengalur (2008); van Langevelde et al.(2000); Orienti et al. (2006); Pihlström et al. (2003); Struve et al. (2010b); Vermeulen et al.(2003); Morganti et al. (2001, 2009); Véron-Cetty et al. (2000) and Privon in prep. Thecollection includes information about the optical depth, flux density, optical ionizationlines, star-formation properties of compact, extended sources. From the collected sam-ple, 19 sources have available line flux and D(4000) index measurements in the SDSSspectroscopic database.

6.8.1 H I and radio propertiesIn Fig. 6.6 we show the optical depth histograms for compact, extended sources. For thiswork only the optical depth of the deepest H I components is considered for every radiosource. The histogram of compact sources shows a wide range of optical depths, revealingsources with certain detection even at very low values (below τ ∼ 0.001). Extendedsources show a distribution with H I detections only above the typical detection limit (τ∼ 0.01) of present-day radio telescopes (see Section 6.5), below this limit one can onlyfind non-detections.

In Fig. 6.7 we plot the flux density distribution against the optical depth for detectionsand non-detections. Compact sources seem to be systematically brighter, allowing forthe detection of faint H I structures at low optical depth. This suggests that the low corebrightness may represent a bias against H I detections in the central region of extendedsources.

Non-detections show a similar flux density distribution as detections. In the figure ofnon-detections, compact sources – despite being stronger in flux density – appear withrather high upper limits. These compact sources were selected from a study at higherredshift (Vermeulen et al. 2003), where insufficient sensitivity could yield increased upperlimits, resulting in a strong bias for non-detections. Taking this in consideration, one cannot make certain conclusions about the lack of H I in these sources. In compact sourcesthe detection rate of H I is ∼37%, however the actual detection rate could be even higher.Although, this could only be tested by more sensitive observations.

6.8.2 Optical propertiesIn Fig. 6.4, the new collection of galaxies for which optical data is available (blacksymbols) are added to the BPT diagram of our observations. Among LINERs we finda mix of detections and non-detections. Galaxies with highly ionized lines in the regionlog([NII]/Hα) > -0.22 and log([OIII]/Hβ) > 0.47 show a high, 66% H I detection rate (4out of 6). This is similar to the H I emission detection rate obtained by Ho & Ulvestad(2001) in a sample of type 1 AGN (broad-line) with Seyfert nuclei, which galaxies areusually located in the same region of the BPT diagram. Ho & Ulvestad (2001) concludedthat type 1 AGN possess a normal H I gas content, as expected from scaling relations.From our plot, this seems to be the case for AGN with highly ionized lines.

Page 141: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

132 chapter 6: HI, radio continuum, and optical properties of AGN

Figure 6.6: Comparison between the optical depth histograms of compact (top), ex-tended (bottom) sources.

Page 142: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.8: Results from the literature 133

Figure 6.7: Flux density vs. optical depth for H I detections (top) and non-detections(bottom) collected from the literature. The solid, dotted and hashed line represent the3σ limits in the absorption line signal corresponding in order to 1σ=0.2, 0.4, 2 mJy

Page 143: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

134 chapter 6: HI, radio continuum, and optical properties of AGN

In Fig. 6.5 we also show the D(4000) index of the galaxies. We find that YSPs atD(4000) < 1.6 are predominantly found in AGN with H I detections, whereas the majorityof AGN with H I non-detections have older stellar populations. This is in good agreementwith our results obtained for SDSS6, supporting the possibility that star formation inradio galaxies is connected with the presence of cold gas. If gas accretion is a commonprocess as it is suggested by the rejuvenation theory of AGN, it is likely that the sameevent can also periodically provide fuel for the production of new stars. The fact thatmost of the radio galaxies with young stellar populations also show H I detection suggeststhat cold gas plays a major role in both AGN and star formation fuelling processes.

6.9 Conclusions, and future perspectivesWe have explored the presence of H I in radio sources typically not considered verypowerful. We detect H I in the two brightest galaxies, a compact, and an FR II radiosource. Furthermore, we find a tentative detection in an intermediate type radio galaxy.We have not explored the significance of the upper limits because the cores turned outto be weak when imaged at higher resolution.

We observe diffuse extended emission around both detected sources, likely due toresidual of a previous cycle of AGN activity. The detection of H I absorption in restartedsources suggests a link between neutral hydrogen gas and rejuvenation of nuclear activity,where possibly H I is the main fuel for feeding the central black hole. Considering alsoprevious studies, restarted activity seems to be a better indicator of H I absorption thanthe radio morphology, e.g. FR I or FR II type.

Furthermore, recent star formation events also show a close connection with thepresence of H I. Galaxies with young stellar populations tend to show high H I detectionrate, suggesting that stars are likely to form in H I-rich gas regions in radio AGN. If gasaccretion is a periodic event in radio galaxies, perhaps this effect will leave an imprinton the star formation history as well.

Our results are based on datasets with small number of galaxies, which are thereforestatistically not representative. We expect that future surveys will provide much largersamples to study, statistically, the H I gas properties of radio galaxies. We use a semi-empirical simulation of the extragalactic radio continuum sky (Wilman et al. 2008) topredict the number of galaxies that will be observed by future surveys. For example,an all-sky survey of 10000 deg2 in the redshift range 0 < z < 0.1 would provide ∼500sources brighter than 30 mJy. Based on our results we expect 20 − 30% of these sourcesto be detected, namely 100 − 150 sources. If we push the redshift limit out to z < 1,one can get the same result with observing only ∼200 deg2. It is clear that in order toget large datasets with a desired brightness limit, we need to observe large areas of skyat shallow redshifts, else the same result could be obtained on smaller areas but reachingto higher redshifts.

Considering the results of simulations, one can understand that future instrumentse.g. Apertif with observations in the redshift range z < 0.2, or, eventually, SKA willsignificantly increase the number of sources available for H I studies. For example, theASKAP-FLASH (The First Large Absorption Survey in HI) all-sky survey is expected toyield ∼1000 intervening absorbers (along the line of sight to the background continuumsource), and several hundred H I absorption systems with associated neutral gas in theredshift range 0 < z < 1. The increased number of H I absorption detections with the

Page 144: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.10: Appendix 135

0

10

20

30

40

50

60

70

80

90

0 1 2 3 4 5 6 7

z

Figure 6.8: Redshift histogram of radio sources brighter than 30 mJy distributedon a 100 deg2 area of sky, based on extragalactic radio continuum sky simulations(Wilman et al. 2008)

next generation of radio telescopes will provide enough sources interpret our H I resultson a much higher confidence level.

As a conclusion we can say that future large H I surveys reaching to cosmologicallyhigh redshifts will represent the key to statistically analyse large samples of radio sources.These surveys will provide large datasets to trace the distribution of neutral hydrogengas in AGN, and to find the link between H I, the evolution of radio sources, and starformation properties.

6.10 Appendix6.10.1 SDSS images and spectra

Page 145: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

136 chapter 6: HI, radio continuum, and optical properties of AGN

Figure 6.9: SDSS optical images of the host galaxies in our sample. In the first row weshow SDSS1, SDSS2, and SDSS3 from the left to the right, and the other images followaccordingly in increasing order as in Fig. 6.1.

Page 146: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

6.10: Appendix 137

Figure 6.10: SDSS optical spectra of our galaxies. The order of the objects is the sameas in Fig. 6.1 and Fig. 6.9.

Page 147: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

138 chapter 6: HI, radio continuum, and optical properties of AGN

Page 148: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Allison, J. R., Curran, S. J., Emonts, B. H. C., et al. 2012, MNRAS, 423, 2601

Aretxaga, I., Terlevich, E., Terlevich, R. J., Cotter, G., & Díaz, Á. I. 2001, MNRAS,325, 636

Baldwin, J. A.; Phillips, M. M.; Terlevich, R.; 1981; PASP; 93; 5

Best, P. N.; Kauffmann, G.; Heckman, T. M. et al.; 2005; MNRAS; 362; 9

Briggs, F. H.; Sorar, E.; Taramopoulos, A.; 1993; ApJ; 415; 99

Buttiglione, S., Capetti, A., Celotti, A., et al. 2010, A&A, 509, A6

Caccianiga, A.; Marchã, M. J. M.; Thean, A.; 2001; MNRAS; 328; 867

Chandola, Y.; D.J. Saikia, D.J.; Gupta, N.; 2010; MNRAS; 403; 269

Chandola, Y., Sirothia, S. K., & Saikia, D. J. 2011, MNRAS, 418, 1787

Conway, J. E.; Blanco, P. R.; 1995; ApJ; 449; 131

Dasyra, K. M., Combes, F., Novak, G. S., et al. 2014, A&A, 565, A46

Dwarakanath, K. S.; Owen, F. N.; van Gorkom, J. H.; 1995; ApJ; 442; 1

Emonts, B. H. C.; Morganti, R.; Oosterloo, T. A.; 2008a; MNRAS; 387; 197

, Emonts, B. H. C.; Morganti, R.; van Gorkom, J. H. et al; 2008b; A&A, 488, 519

Emonts, B.H.C.; Morganti, R.; Struve, C.; 2010, MNRAS, 406, 987

Fanaroff, B.L.; Riley, J.M.; 1974; MNRAS; 167; 31

Fernández, X., van Gorkom, J. H., Schweizer, F., & Barnes, J. E. 2010, AJ, 140, 1965

Gereb, K., Morganti, R., & Oosterloo, T. 2014, arXiv:1407.1799

van Gorkom, J. H.; Knapp, G. R.; Ekers, R. D. et al.; 1989; AJ; 97; 708

Page 149: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

140 BIBLIOGRAPHY

Gupta, N., Salter, C. J., Saikia, D. J., Ghosh, T., & Jeyakumar, S. 2006, MNRAS, 373,972

Gupta, N.; Srianand, R.; Petitjean, P. et al.; 2007; ApJ; 654; 111

Hardcastle, M. J., Evans, D. A., & Croston, J. H. 2006, MNRAS, 370, 1893

Hardcastle, M.J.; Evans, D.A.; Croston, J.H.; 2007; MNRAS; 376; 1849

Ho, L.C.; Ulvestad, J.S.; 2001; ApJS; 133; 77

van der Hulst, J. M.; Sadler, E. M.; Jackson, C. A. et al; 2004; NewAR; 48; 1221

Jaffe, W.; 1991; A&A; 250; 67

Jamrozy, M.; Konar, C.; Saikia, D.J. et al.; 2007; MNRAS; 378; 581

Kanekar, N.; Chengalur, J. N.; 2008; MNRAS; 384; 6

Keel, W. C.; White III, R. E.; Owen, F. N. et al.; 2006; AJ; 132; 2233

Kewley, L. J.; Heisler, C.A.; Dopita, M.A. et al.; 2001; ApJS; 132; 37

Kauffmann, G.; Heckman, T.M.; Tremonti, C. et al.; 2003; MNRAS, 346, 1055

Lilly, S. J., & Longair, M. S. 1984, MNRAS, 211, 833

van Langevelde, H. J.; PihlstrÃűm, Y. M.; Conway, J. E. et al.; 2000; A&A; 354; 45

Morganti, R.; Oosterloo, T.; Tsvetanov, Z.; 1998, AJ; 115; 915

Morganti, R.; Oosterloo, T.A.; Tadhunter, C.N. et al.; 2001; MNRAS; 323; 331

Morganti, R.; Oosterloo, T. A.; Tinti, S. et al.; 2002; A&A; 387; 830

Morganti, R., Oosterloo, T. A., Tadhunter, C. N., van Moorsel, G., & Emonts, B. 2005,A&A, 439, 521

Morganti, R., de Zeeuw, P. T., Oosterloo, T. A., et al. 2006, MNRAS, 371, 157

Morganti, R.; Peck, A. B.; Oosterloo, T. A. et al.; 2009; A&A; 505; 559

Morganti, R., Fogasy, J., Paragi, Z., Oosterloo, T., & Orienti, M. 2013, Science, 341, 1082

Orienti, M.; Morganti, R.; Dallacasa, D. et al.; 2006; A&A; 457; 531

Oosterloo, T.A.; Verheijen M.; van Cappellen, Wim; 2010a; (arXiv:1007.5141v1)

Oosterloo, T.A.; Morganti, R.; Crocker, A. et al.; 2010b; MNRAS; 409; 500

Pihlström, Y. M.; Conway, J. E.; Vermeulen, R. C.; 2003; A&A; 404; 871

Pihlström, Y. M. 2001, Ph.D. Thesis, Chalmers University & Onsala Observatory

Saikia, D. J.; Jamrozy, M.; 2009; BASI; 37; 63

Page 150: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

BIBLIOGRAPHY 141

Serra, P., Oosterloo, T., Morganti, R., et al. 2012, MNRAS, 422, 1835

Shulevski, A., Morganti, R., Oosterloo, T., & Struve, C. 2012, A&A, 545, A91

Smith, E. P.; Heckman, T. M.; 1989; ApJ; 341; 658

Struve, C., Oosterloo, T., Sancisi, R., Morganti, R., & Emonts, B. H. C. 2010, A&A,523, A75

Struve, C.; Oosterloo, T. A.; Morganti, R.; Saripalli, L.; 2010; A&A; 515; 67

Tadhunter, C. N.; Fosbury, R. A. E.; Quinn, P. J.; 1989; MNRAS; 240; 225

Tadhunter, C. N., Morganti, R., Robinson, A., et al. 1998, MNRAS, 298, 1035

Tadhunter, C.; Dickson, R.; Morganti, R. et al.; 2002; MNRAS; 330; 977

Tadhunter, C.; Robinson, T. G.; GonzÃąlez Delgado, R. M. et al.; 2005; MNRAS; 356;480

Ulrich, M-H.; Ronnback, J.; 1996; A&A; 313; 750

Vermeulen, R.C.; Pihlström, Y.M.; Tshager, W. et al; 2003; A&A, 404, 861

Véron-Cetty, M.-P.; Woltjer, L.; Staveley-Smith, L.; Ekers, R. D.; 2000; A&A; 362; 426

Wilman, R. J.; Miller, L.; Jarvis, M. J. et al.; 2008; MNRAS; 388; 1335

York D. G., et al., 2000, AJ, 120, 1579

Page 151: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis
Page 152: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Summary

In this thesis we have used stacking techniques to study the H I gas properties of galaxiesand AGN beyond the local Universe. We explore the possibilities and the science thatcan be addressed by applying stacking to archival datasets and to current/future wide-band observations, but also the technical limitations of such studies. An effective way tomake full use of stacking is to select archival radio datasets in targeted fields with opticalcoverage, where one can combine the obtained H I information with optical data. Severaldatasets exist, e.g. in the WSRT (Westerbork Synthesis Radio Telescope) archive, thatone could use to exploit the many opportunities given by stacking techniques. In thisstudy we were able to understand important advantages of the technique, for examplethat even for datasets which are not entirely optimal for H I spectral line work (e.g.archival data with poor spectral resolution), stacking is giving promising results becausethe errors in the optical velocities are often of the order 50 − 75 km s−1. Less stringentcalibration requirements make it easier to stack datasets originally calibrated for thecontinuum, and can also make it faster to fully calibrate and analyse data with automatedpipelines. These aspects are also important for possible future stacking experiments.

Below we summarize the main conclusions from each chapter.

7.1 Chapter 2: The Lockman Hole project: gas andgalaxy properties from a stacking experiment

• We detect H I in blue and red galaxies at 0.06 < z < 0.09. As expected, bluegalaxies are more H I rich, but also red galaxies which were selected based on g −r optical colors do show relatively large amounts of H I in the stacked profile.

• In our galaxies a connection is found between the presence of neutral and ionizedgas. All galaxies where H I is detected also contain ionized gas, whereas no H I isfound around galaxies without ionized gas.

• Not just normal star forming galaxies, but also LINERs can have H I gas and on-going star formation

• LINERs can be separated into star-forming and non-star-forming groups based onIR colors. LINERs with ongoing star formation also show relatively large amounts

Page 153: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

of H I and are often detected in the radio continuum. Non-star-forming LINERsresemble optically inactive galaxies, as this group is depleted of cold gas. RadioLINERs in the latter group are the best candidates for hosting low luminosity AGN.

7.2 Chapter 3: From star forming to inactive galaxies:the global cold gas content up to z = 0.12

• Galaxies in the green valley are detected with lower amounts of H I than bluegalaxies, but unlike red galaxies (selected based on NUV - r colors), they are notcompletely depleted of cold (H I) gas. This result indicates that green valley objectsare an intermediate population from H I point of view.

• We do not detect H I in red galaxies defined based on NUV - r colors. However,stacking is a promising technique to lower the detection limit and study the rel-atively unexplored < 107 M⊙ H I mass regime of galaxies using large samples ofgalaxies.

• In agreement with previous studies, our results show that the presence of H I isbetter correlated with IR and NUV - r color rather than with ionization properties.We detect H I in AGN which were defined based on their optical emission lines. Ifthey are classified as green/blue, even galaxies with higher ionization properties(LINERs and optical AGN) do contain cold gas. This suggests that optical AGNare not the (main) reason for depleting gas reservoirs, or that AGN-driven gasdepletion is not an instantaneous effect in galaxies.

• In galaxies where cold gas (H I) is present, conditions are favourable for (residual)SF to be seen. Furthermore, in most of the sample, galaxies with more gas alsohave a higher SFR (blue cloud, SF galaxies).

• The H I mass-luminosity ratios do not change significantly as function of redshift,suggesting that the global H I content remains relatively constant up to z = 0.12.Furthermore, the global SFE displays a similar behaviour, remaining relativelyconstant in the covered redshift range. This can be interpreted as an indicationsthat the H I content and SF are regulated by the same process, e.g. feedback effects,galaxy environment.

7.3 Chapter 4: Probing the gas content of radio galax-ies through H I absorption stacking

• In a systematic study of the H I properties of radio AGN, we find that 30 percentis a representative H I detection rate for the general population of these objects.

• The detection rate does not depend on the apparent flux of a source, suggesting thatH I absorption studies of even fainter radio sources will still bring a large numberof detections. This result has positive implications for future surveys, which willobserve AGN over a broad flux density range.

Page 154: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

• We find a dichotomy in the presence of H I. Even when a large number of spectra areco-added, the stacking of undetected sources has resulted in a non-detection of H I.This suggests that there are objects genuinely depleted of H I or that orientationeffects play a role. However, orientation effects alone cannot fully explain thedichotomy that we see in our sample, suggesting that some fraction of our galaxiesmust be depleted of cold gas.

• H I emission and absorption are tracing similar morphological structures. Compact,young AGN are richer in H I than the general population of early-type galaxies,supporting the hypothesis that nuclear activity in radio galaxies is triggered throughfeeding of cold gas.

• Compact sources show higher detection rates, optical depths and FWHM thanextended sources, strongly suggesting that different gas conditions exist in thesetwo types of radio sources; however, high resolution observations and a bettermeasure of the covering factor are needed to confirm this result.

7.4 Chapter 5: The H I absorption ‘Zoo’• H I absorption displays a broad range of line shapes and kinematics. The busy

function (Westmeier et al. 2014) provides a good tool for fitting and parametrizingthe broad variety of absorption profiles.

• We find that the complexity of the lines is increasing with increasing profile width.Based on the line shapes and widths the lines can be separated into three groupsbelonging to different morphological structures. The narrowest lines with FWHM <100 km s−1 in our sample are most likely produced by large scale disks or H I clouds.Relatively broad lines (100 km s−1 < FWHM < 200 km s−1) may be produced bysimilar morphological structures with more complex kinematics. Broad lines withFWHM > 200 km s−1, however, are tracing the most unsettled gas structures, e.g.gas-rich mergers and outflows.

• Broad lines show large asymmetries, and we note that symmetric broad lines areabsent in our sample. The lack of symmetry could suggest that such broad profilesalways arise due to unsettled gas.

• We detect three new cases of broad (FW20 > 500 km s−1), blueshifted H I wings.Along with their radio source properties, i.e. powerful AGN with log(P1.4GHz) >25 W Hz−1, these sources are the best candidates for being jet-driven H I outflows.Considering certain and tentative cases, the detection rate of H I outflows is 5% inour total sample. The relatively low detection rate suggests that, if all radio AGNgo through an outflow phase during their lifetime, then the gas depletion timescaleof H I outflows is shorter than the typical lifetime of radio galaxies.

• H I in compact sources has a more unsettled distribution, e.g. blueshifted lines andbroad/asymmetric profiles are frequent among compact AGN. Such H I line prop-erties suggest that strong interactions between AGN and their rich circumnuclearmedium are likely to occur as young radio jets are clearing their way through theambient medium in the early phases of the nuclear activity.

Page 155: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

7.5 Chapter 6: H I radio continuum, and optical prop-erties of radio galaxies

• We have explored the presence of H I in a sample of SDSS/NVSS selected radiosources. We detect H I in the two brightest galaxies, however faint AGN cores,which are typical for extended sources, represent a bias against H I detection.

• H I in radio galaxies is connected with the radio source and star formation prop-erties. We observe diffuse extended emission around both detected sources, likelydue to the residual emission of a previous cycle of AGN activity.

• The detection of H I absorption in restarted sources suggests a link between neutralhydrogen gas and rejuvenation of nuclear activity, where possibly H I is the mainfuel for feeding the central black hole. Signatures of restarted activity, e.g. relicradio structures seem to be a good indicator for the presence of H I absorption.

• Recent star formation events also show a close connection with the presence of H I.To make our conclusions more robust, based on a literature search we constructa larger dataset of radio sources with available H I observations. Galaxies withyoung stellar populations tend to show high H I detection rate, suggesting that starformation in radio galaxies is connected with the presence of an H I-rich medium.

• If gas accretion is a periodic event in radio galaxies, perhaps this effect will leavean imprint on the star formation history as well.

7.6 Future ProspectsThis study has been limited by the number of objects due to the restricted availability ofspectroscopic redshifts. The next step will be to increase the size and redshift range ofthe selected samples. Large H I surveys planned with new or upgraded radio telescopeswill also improve the limitations of current H I observations due to their wide bandwidth,high spectral resolution and increased field of view. In particular, the combination oflarge surveys and wide bandwidth promised by the future focal-plane array receivers tobe installed for Apertif (Oosterloo et al. 2010b), the Australian Square Kilometre ArrayPathfinder (ASKAP, Johnston et al. 2008; DeBoer et al. 2009), will provide favourableobservational settings to combine the radio continuum and H I spectral line data withoptical information, and thus to trace, for the first time, the cosmological evolution ofneutral hydrogen.

Using stacking techniques, we have reached the MHI detection limit of a few × 108

M⊙ for galaxies which have been observed for 12 hours with the WSRT, in the redshiftrange z < 0.1. Even though this is a relatively low limit, lower H I masses have beendetected before in SAURON and ATLAS3D. Future, large surveys will provide enough datato test the global H I content at earlier epochs of the Universe at lower, currently ratherunexplored H I detection limit of MHI < 107 M⊙. Stacking is a promising techniqueto reach this limit, given that large samples of galaxies are available with spectroscopicredshift measurements. This will be made possible by future H I surveys with the nextgeneration of radio telescopes, e.g. Apertif, ASKAP, and MeerKat (Booth et al. 2009).Currently the studied samples are biased by the optical selection of the SDSS. Future

Page 156: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

surveys will expand on the work presented here by searching for H I in a ‘blind’ mode,allowing for the detection of the H I population without any optical pre-selection.

Large samples with available multiwavelength data also make it possible to studyand compare the H I properties of subsamples with different color, ionization properties,stellar populations, etc. The success of these studies highly depends on the availability ofancillary data. Thus, the most effective way of carrying out such an analysis is to observeH I in the area of well studied fields, for example the Lockman Hole (Fotopoulou et al.2012), the COSMOS field (Fernández et al. 2013). Furthermore, we expect that in alarger sample the number of potential AGN (optical and/or radio) will increase, makingit possible to further investigate the connection between gas and the black hole activity.

Unlike H I emission, H I absorption studies are less limited by sensitivity in the higherredshift Universe. Our results suggest that the detection rate of H I absorption doesnot depend on the apparent flux of a source and this has positive implications for fu-ture, deeper surveys. These large-area surveys will uncover a very large number of H Iabsorptions systems. At radio power lower than < 1023 W Hz−1 we expect a mix ofstar-forming/AGN populations (Mauch & Sadler 2007). If emission and absorption aretracing similar morphological structures, H I absorption studies can be used just as ef-ficiently to find cold gas not just in AGN, but also in star-forming galaxies at higherredshift. The increased number of sources will provide enough data to perform H I stack-ing experiments and, hence, to probe the highest redshift regime of the observed radiosky at low optical depth.

Page 157: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis
Page 158: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Bibliography

Booth, R. S., de Blok, W. J. G., Jonas, J. L., & Fanaroff, B. 2009, arXiv:0910.2935

DeBoer, D. R., Gough, R. G., Bunton, J. D., et al. 2009, IEEE Proceedings, 97, 1507

Fotopoulou S., et al., 2012, ApJS, 198, 1

Fernández, X., van Gorkom, J. H., Hess, K. M., et al. 2013, ApJ, 770, L29

Johnston S., et al., 2008, ExA, 22, 151

Mauch, T., & Sadler, E. M. 2007, MNRAS, 375, 931

Oosterloo, T., Verheijen, M., & van Cappellen, W. 2010, ISKAF2010 Science Meeting

Westmeier, T., Jurek, R., Obreschkow, D., Koribalski, B. S.,&Staveley-Smith, L. 2014,MNRAS, 438, 1176

Page 159: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis
Page 160: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Nederlandse Samenvatting

Door alle chemische elementen in het universum, inclusief neutraal waterstof (H I),worden fotonen geabsorbeerd of uitgezonden met een specifieke frequentie. De bekendeNederlandse astronoom Henk van de Hulst voorspelde dat, vanwege een verandering vande een verandering in the energietoestand van waterstof, atomair waterstof (H I) elektro-magnetische straling uitzendt op 1420,405 MHz. Deze frequentie valt in het bereik datwe radio noemen, en kan dus alleen worden opgevangen door radiotelescopen. Omdatwe in een uitdijend heelal leven, is de straling die uitgezonden wordt door verre ster-renstelsels roodverschoven naar lagere frequenties wanneer we deze waarnemen. Volgensde wet van Hubble is de roodverschuiving proportioneel tot de afstand naar het uitzen-dende sterrenstelsel, waardoor de spetroscopische verkregen roodverschuiving voor eensterrenstelsel als een betrouwbare afstands indicator dient. Dankzij de vooruitgang inde radio astronomie en de bouw van telescopen, zoals de Westerbork Synthesis RadioTelescope (WSRT), de Karl G. Jansky Very Large Array (VLA), en vele andere in dezethesis genoemde faciliteiten, is het mogelijk om grote delen van de hemel tot relatief hogeroodverschuiving in H I te bestuderen.

Sinds het begin van H I waarnemingen, hebben ze geleid tot belangrijke ontdekkingen.Waterstof is het meest voorkomende gas in het universum, en het oergas waaruit sterrenuiteindelijke vormen bestaat er grotendeels uit. Het gas in de schijf van de Melkwegbestaat bijvoorbeeld uit 70% uit waterstof in atomaire vorm. We weten ondertussen datsterrenstelsels miljarden zonnemassa’s (M⊙) aan H I gas bevatten, dat soms gebiedenver buiten de stellaire schijf beslaat. Omdat sterren in kraamkamers van stervormingbestaande uit dicht moleculaire waterstof wolken ontstaan, is stervorming gerelateerdaan de aanwezigheid van gas. Neutraal waterstof biedt ook een goede mogelijkheid omte leren hoe in het universum interacties en fusies sterrenstelsels uit elkaar scheurenof juist bouwen. Verder suggereren waarnemingen dat H I deels verantwoordelijk kanzijn voor het ontsteken van een van de meest energieke phenomenen in het universum,doordat het de super zware zwarte gaten in het centrum van sterrenstels voedt.

Signalen die uitgezonden zijn door verre objecten zijn zwakker door de afstand, en

Page 161: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

H I waarnemingen uit het hoog roodverschoven universum zijn hierdoor gelimiteerd doorde gevoeligheid van de huidige telescopen. Van de volgende generatie radiotelescopen,bijvoorbeeld Apertif, de Australian Square Kilometre Array Pathfinder, MeerKat enlater de Square Kilometre Array (SKA), wordt verwacht dat ze H I studies naar groteen kosmologisch significante afstanden brengen. Het bouwen van deze faciliteiten vereistechter een grote hoeveelheid werk, financiele ondersteuning en tijd, en het zal nog eenaantal jaren duren voordat ze volledig operationeel zullen zijn.

In de laatste jaren zijn er echter ook statistische methodes beschikbaar gekomen omde detectielimiet van huidige radiotelescopen te verlagen, en daardoor kan men nu ookal H I waarnemingen doen tot aan relatief hoge roodverschuiving. Het stapelen van H Iwaarnemingen is effectief gebruikt om globaal de gemiddelde H I inhoud te bestuderen ingrote verzamelingen van sterrenstelsels.

Omdat het doel van deze thesis het statistisch bestuderen van grote verzamelingenvan verre sterrenstelsels is, wordt in dit werk spectroscopische gestapelde H I data geana-lyseerd. Om een beter begrip te krijgen van de rol van H I in de evolutie van sterrenstelselsin de voorbije 1.5 miljard jaar, bestuderen we de H I eigenschappen van honderden ster-renstelsels door middel van stapel technieken. We kijken ook naar de eigenschappenvan het gas bij de accreterende superzware zwarte gaten, ook wel actieve galactischenuclei (AGN) genoemd, om de interactie tussen AGN activiteit en het omliggende gas teonderzoeken.

8.1 H I gas eigenschappen van sterrenstelselsSterrenstelsels kunnen gebaseerd op hun morfologische verschijning worden verdeeld intwee hoofdtypes: spiraal stelsels en elliptischen. Spiralen, ook wel laat-type stelsels ge-noemd, zijn roterende systemen met heldere stellaire schijven, terwijl stelsels met bol ofellips vorm (vroeg-type stelsels) een chaotischere structuur hebben. Deze twee hoofdtypesstelsels blijken ook andere systematische trends te vertonen met andere eigenschappen.Het is al lang bekend dat de kleuren en lichtsterkte van sterrenstelsels verband houdenmet de morfologie. Typische laat-type spiralen zijn blauwer en hebben een lagere op-pervlakte helderheid dan vroeg-type sterrenstelsels, die vooral rood zijn en meestal meerlicht uitzenden. Deze observationele eigenschappen zijn gemakkelijk te interpreteren inde context van de geschiedenis van stervorming en -evolutie. Spiraalvormige stelsels vor-men actief sterren en hebben dus blauwere kleuren door hun jonge stellaire populatie.Vroeg-type stelsels vormen normaal gesproken geen sterren, en hebben oudere en roderestellaire populaties.

Dankzij 21 cm H I waarnemingen hebben de eerste studies aangetoond dat spiraal-vormige stelsels bijna altijd relatief grote hoeveelheden H I bevatten, terwijl vroeg-typestelsels geen tot zeer weinig gas bevatten. Het leek dus alsof vroeg-type stelsels typisch’rode en dode’ systemen zijn. Later, door systematische H I bestuderingen van grotereverzamelingen sterrenstelsels, hebben we geleerd dat vroeg-type stelsels veel interessan-ter zijn dan simpele rode en dode sterrenstelsels. H I bevindt zich in een neergestrekenschijf/ring structuur in ongeveer de helft van de gedetecteerde gevallen. In vroeg-typestelsels spant H I een groot bereik van massa’s en kolomdichtheden, en samen met hetfeit dat de H I schijf vaak niet uitgelijnd is met de stellaire schijf, wijst dit er sterk opdat H I gas in vroeg-type sterrenstelsels een externe herkomst heeft. In ongeveer 70%van de stelsels met H I in het centrum zijn er ook tekenen van voortdurende stervorming

Page 162: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

gezien. Dit betekent dat, ook al is het een gematigd effect, gasaccretie en hierop volgendestervorming een rol hebben in de evolutie van vroeg-type sterrenstelsels tot en met dedag van vandaag.

De genoemde studies zijn gedaan met directe waarnemingen in het nabije universum.Om het gehele proces van evolutie van sterrenstelsels te begrijpen, is het belangrijk omH I waarnemingen naar grotere afstanden te verplaatsen.

8.2 Stapelen van H I spectraHet proces waarbij spectra van individuele sterrenstelsels wordt gecombineerd, heet ookwel ’H I stapelen’, en het wordt gebruikt om een gemiddeld H I signaal te verkrijgen vande gestapelde verzameling. Bij het stapelen worden de spectra van verre stelsels terugge-schoven naar het rustkader en wordt de ruis-gewogen som van de spectra geproduceerd.Op deze manier kan men een gemiddeld H I signaal van de gestapelde spectra meten. Ditheeft als grote voordeel dat bij het combineren van de spectra de signaal-ruis verhoudingwordt verbeterd, omdat de ruis vermindert met de vierkantswortel van het aantal gesta-pelde spectra. Hierdoor is deze methode niet alleen bijzonder handig om de gevoeligheidvoor individueel ongedetecteerde sterrenstelsels te versterken in het nabije universum,maar vooral ook voor het verre universum waar de hoeveelheid licht uigezonden doorindividuele objecten niet groot genoeg is om door de huidige radio telescopen te wordenwaargenomen.

Spectroscopische onderzoeken zoals de Sloan Digital Sky Survey (SDSS) leveren nietalleen de roodverschuivingen die nodig zijn voor het stapelen van spectra, maar geven ookeen grote hoeveelheid informatie over the optische eigenschappen, bijvoorbeeld de magni-tude, kleur en emissielijnflux metingen. Een consequentie hiervan is dat het een effectievemanier is om de verkregen H I en optische informatie te combineren met data over eengroot gedeelte van het elektromagnetische spectrum, zoals infrarood (IR) en ultraviolet(UV), om het maximale uit het stapelen te halen. Om een uitgebreider idee te krijgenvan de aard van de geselecteerde stelsels, kan de verzamelde informatie gebruikt wordenom verschillende eigenschappen van de groepen sterrenstelsels te definiëren. Men kande relatieve H I hoeveelheid van de deelverzamelingen meten door middel van H I stapel-analyse, waardoor het mogelijk wordt om de rol van H I in de evolutie van verschillendetypes sterrenstelsels te bestuderen.

8.3 Gas en de evolutie van radio AGNAGN-activiteit wordt geassocieerd met accreterende superzware zwarte gaten (ZG) in decentrale gebied van sterrenstelsels. Zulke ZG met massa’s varierend van 106 tot 109.5 M⊙,worden verwacht in ieder sterrenstelsel met een ronding te bestaan. Men gelooft dat AGNverantwoordelijk zijn voor het veranderen van de gaseigenschappen en de evolutie vansterrenstelsels door het verhitten of het buiten het sterrenstelsel blazen van het gas. Zulketerugkoppelingseffecten kunnen acteren door straling, accretie gedreven winden en radiojets. Van deze effecten wordt gedacht dat ze verantwoordelijk zijn voor het onderdrukkenvan stervorming in de massieve sterrenstelsels en voor het reguleren van de groei van de

Page 163: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

ZG. De geschatte levensduur van een radio AGN is relatief kort, hoewel de activiteit inhet radio verjongd kan worden en de sterrenstelsels de terugkoppelingseffecten van hunAGN steeds opnieuw kunnen voelen. Dus, een belangrijke vraag betreffende ons begripvan actieve kernen is of AGN activiteit normaal gesproken episodisch is, en zo ja, whatthe cyclus van activiteit is.

Nucleaire activiteit kan zich onthullen op verschillende manieren. Een bijzonder fasci-nerend type AGN zijn radio stelsels, welke bekend zijn om het lanceren van relativistischejets van radio plasma tot grote afstanden in het intergalactische medium. Radiotelesco-pen kunnen worden gebruikt om de radiofase van AGN-activiteit te bestuderen. Omdatnucleaire activiteit en de vorm van accretie in sterrenstelsels is gereguleerd door de be-schikbaarheid van gas, is het cruciaal om een beter begrip van de fysische and kinemati-sche condities van het gas in de circumnucleaire gebied om de AGN te verkrijgen.

Het peilen van de circumnucleaire gebied om de radio AGN via H I absorptieEen belangrijke methode om de circumnucleaire gebied van galaxies te peilen, is via H Iabsorptie. Men kan op een efficiente wijze zelfs kleine hoeveelheden waterstof in ab-sorptie op hoge roodverschuiving detecteren. Door de jaren heen hebben H I absorptiewaarnemingen op grote schaal bijgedragen aan ons begrip van complexe processen dieplaatsvinden in de centrale gebieden van sterrenstelsels. Het is gebleken dat absorptieeen scala aan structuren in AGN volgt: regelmatig ronddraaiende gasschijven, invallendeH I wolken geassocieerd met het voedingsmechanisme van het centrale zwarte gat en H Igas dat uit het sterrenstelsel stroomt door interacties tussen de jets en het omringendemedium. Daarom suggereert de complexiteit van H I kinematica in AGN dat gas veelverschillende rollen kan spelen in AGN.

Brede absorptielijnen geassocieerd met uitgaande H I stromen zijn over het algemeenzwak. Gedurende de laatste jaren is het aantal detecties van uitgaande stromen omhooggegaan, vooral dankzij gevoelige waarnemingen, hoewel het aantal waarnemingen nogsteeds klein is. Het is duidelijk dat voor een volledig begrip van de eigenschappen vanH I in radio AGN, het noodzakelijk is dat er een grote en statistisch representatieveverzameling van sterrenstelsels wordt bestudeerd. Het observeren van een groot aantalobjecten heeft als voordeel dat stapelexperimenten mogelijk zijn.

Twee bijzondere types radio stelsels, compact met een stijl spectrum (CSS) en gi-gahertz piek spectra (GPS) objecten zijn intrinsieke kleine AGN, die jonger zijn dan <104 yr. Dit soort radio objecten lijken rijk te zijn aan H I, terwijl een groot deel van deuitgestrekte objecten nog niet gedetecteerd is. Volgens het tegenwoordig geaccepteerdeparadigma, wordt AGN activiteit ontstoken tijdens de compacte fase van invallend gas,dus zijn CSS and GPS objecten de voorlopers van de uitgestrekte radio stelsels.

Door de jaren zijn er verschillende kenmerken van radio objecten geopperd voor hetkenmerk van activiteit in het verleden. Nadat de nucleus is uitgegaan zal de lobstructuurverdwijnen door een gebrek aan voeding, al zullen gedurende een gelimiteerde tijd dezestructuren herkenbaar zijn aan hun fossiele emissie. In het bijzonder moet worden opge-merkt dat veel van de opnieuw opgestarte objecten ook H I waarnemingen laten zien inde centrale gebieden. De veel voorkomende aanwezigheid van H I in herstarte AGN wordtgeinterpreteerd als een mogelijke verbinding tussen de aanwezigheid van koud gas en hetopnieuw activeren van de AGN. De belangrijkste begrenzing van deze studies is dat erop het moment slechts een handvol van dergelijke radiorelieken bekend zijn. Verjongde

Page 164: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

objecten vormen echter een sleutel element voor ons begrip van de cyclus van AGN ac-tiviteit, waardoor deze objecten steeds meer aandacht krijgen en het aantal objectenverhoogd wordt.

8.4 Deze thesisIn deze thesis gebruiken we stapel technieken om de globale H I inhoud van sterrenstelselstot aan een roodverschuiving z < 0.12 te meten. We hebben vorige studies uitgebreiddoor multikleuren informatie te gebruiken en door de objecten onder te verdelen in ver-schillende deelgroepen, gebaseerd op optische emissie lijnen en AGN diagnostieken. Wehebben geprobeerd om stervormende sterrenstelsels te scheiden van AGN door gebruikte maken van multigolflengte informatie. Specifiek hebben we ons gericht op het verbandtussen H I gas met stervorming en ZG voedingsprocessen. Om de vraag te beantwoor-den of nucleaire activiteit een effect heeft op de gasvoorraad van sterrenstelsels op groteschaal, hebben we op sterrenstelsels gemikt waar terugkoppeling waarschijnlijker is omop heterdaad betrapt te worden, namelijk AGN en sterrenstelsels in de groene vallei (metoudere sterren vergeleken met actief stervormende sterrenstelsels). De belangrijkste re-sultaten zijn:

• De verhouding tussen H I-massa en lichtsterkte veranderen niet significant als func-tion van roodverschuiving, wat suggereert dat de globale H I inhoud relatief constantblijft tot aan z = 0.12. Daarnaast vertoont de globale stervormingsefficientie eenvergelijkbaar gedrag en blijft relatief constant gedurende de waargenomen roodver-schuivingen. Dit kan worden geinterpreteerd als een indicatie dat de H I inhoud enstervorming zijn gereguleerd door hetzelfde proces, bijvoorbeeld terugkoppelings-effecten en de omgeving van het sterrenstelsel.

• We detecteren H I in de blauwe en rode stelsels op 0.06 < z < 0.09. Zoals verwachtzijn blauwe stelsels rijker aan H I, maar ook rode sterrenstelsels die zijn geselecteerdop g - r optische kleuren vertonen relatief grote hoeveelheden H I in de gestapeldeprofielen.

• We detecteren geen H I in de rode stelsels die gedefinieerd zijn op basis van hun NUV- r kleuren. Stapelen is echter een veelbelovende techniek om de detectielimiet teverlagen van het relatief onverkende < 107 M⊙ H I massa regime van sterrenstelselsdoor grote verzamelingen sterrenstelsels te gebruiken.

• Sterrenstelsels die zich in de groene vallei bevinden worden gedetecteerd met klei-nere hoeveelheden H I dan blauwe sterrenstelsels, maar in tegenstelling tot rodesterrenstelsels die geselecteerd zijn vanwege hun NUV - r kleuren, is hun H I voor-raad niet geheel uitgeput.

• In overeenstemming met vorige studies laten onze resultaten zien dat de de aanwe-zigheid van H I beter gecorreleerd is met IR en NUV - r kleuren dan met ionisatie-eigenschappen. Sterrenstelsels geclassificeerd als groen/blauw en zelfs sterrenstel-sels met hogere ionisatie eigenschappen, bijvoorbeeld Lage Ionisatie Nucleaire Emis-sio Regio (LINER) en optische AGN (geclassificeerd op basis van optische emissie

Page 165: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

lijnen), bevatten koud gas. Dit suggereert dat optische AGN niet de (belangrijk-ste) reden zijn om gasreservoirs uit te putten, of dat AGN gedreven gas niet eeninstantaan effect heeft op sterrenstelsels.

• Niet alleen normale stervormende sterrenstels hebben H I gas en doorgaande ster-vorming, maar ook LINERs. LINERs kunnen worden onderscheiden in stervor-mende en niet-stervormende groepen op basis van hun IR-kleuren. LINERs metvoortgaande stervorming vertonen ook vaak relatief grote hoeveelheden H I en wor-den vaak gedetecteerd in het radio continuum. Niet-stervormende LINERs hebbenal hun gas verloren. Radio LINERs in de laatste groep zijn de beste kandidatenwaar lage-lichtsterkte AGN in voorkomen.

• In sterrenstelsels waar koud gas (H I) aanwezig is, zijn condities gunstig om (res-terende) stervorming te zien. Verder, in het grootste deel van de verzameling,hebben sterrenstelsels met meer gas ook een hogere stervormingssnelheid (blauwewolk, stervormende sterrenstelsels).

In deze thesis hebben we voor de het eerst de mogelijkheid verkend om, gebruikmakend van het stapelen van absorptielijnencentraal, H I te detecteren in AGN. Omde vraag te beantwoorden of radio activiteit ontstoken wordt door accretie van koud(H I) gas, hebben we de verschillen in gas eigenschappen bestudeerd tussen jonge engevolueerde radio objecten. De kinematica van het gas verschaft ons informatie overde gas accretie geschiedenis (inwaardse stromen) en terugkoppelingsprocessen vanwegeinteracties tussen het radio object en het gas (uitgaande stromen), en daarom hebbenwe dus twee verschillende type radio bronnen onderzocht. We hebben het voorkomenvan uitgaande stromen onderzocht, onderzoekend hoe de gasuitputtingstijdschaal zichvergelijkt met de levenscyclus van AGN. Dit kan belangrijk zijn voor het begrenzen vanterugkoppelingsmodellen, en voor ons begrip van het samenspel van AGN activiteit enH I gedurende het hele leven van een radio AGN. Op dit moment is het nog steeds nietduidelijk of AGN activiteit voorkomt in ieder sterrenstelsels, en wat de actieve dienstcyclus van de radio fase is. Omdat de beschikbare verzamelingen gelimiteerd zijn, ishet moeilijk om volledig te begrijpen wat de rol van H I is in dit proces. We hebbengezocht naar meer gevallen van herstarte radio objecten en we hebben de eigenschappenvan reliekstructuren onderzocht.

• In een systematische studie van H I eigenschappen van radio AGN, vinden we dat30 procent een representatieve H I detectie kans is voor de algemene populatie vandeze objects. De detectie kans hangt niet af van de schijnbare flux van een bron,suggererend dat H I absorptie studies van zelfs zwakkere radio bronnen nog steedseen groot aantal detecties zal opleveren. Dit resultaat heeft positieve gevolgenvoor toekomstige studies, die AGN zullen observeren over een groot bereik aanfluxdichtheden.

• We vinden een tweedeling in de H I aanwezigheid. Zelfs als een groot aantal spectrazijn gestapeld, resulteert het stapelen ongedetecteerde bronnen in nondetecties vanH I. Dit suggereert dat er objecten zijn die werkelijk uitgeput van H I zijn, of datorientatie een rol speelt.

• Compacte, jonge AGN zijn rijk aan H I gas, wat de hypothese dat nucleaire activiteitin radio stelsels wordt onstoken door voeding met koud gas ondersteunt. Ook heeft

Page 166: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

H I in compacte bronnen een meer verstoorde verdeling, blauwverschoven lijnenen brede asymmetrische profielen komen bijvoorbeeld regelmatig voor in compacteAGN. Zulke H Ilijneigenschappen suggereren dat sterke interacties tussen AGN enhun rijke circumnucleaire medium waarschijnlijk gebeuren als jonge radio jets inde vroegste fases van hun nucleaire activiteit hun weg banen door het omliggendemedium.

• Brede absorptielijnen vertonen grote asymmetriën, en we merken op dat symme-trische brede lijnen afwezig zijn in onze verzameling. Het gebrek aan symmetriekan suggereren dat zulke brede profielen altijd ontstaan door veranderlijk gas.

• Als we zekere en voorzichtige gevallen beschouwen is het detectieniveau van outflows5 % van onze gehele verzameling. Het relatief lage detectieniveau suggereert dat,als alle radio AGN een uitstroom fase meemaken in hun leven, de tijdschaal voorhet uitputten van uitgaande H I stromen korter is dan de typische levensduur vanradio stelsels.

• De detectie van H I absorptie in herstarte bronnen suggereert een verbinding tussenneutraal waterstofgas en verjonging van nucleaire activiteit, waar mogelijk H I debelangrijkste brandstof is om het centrale zwarte gat te voeden.

• Recente gebeurtenissen met stervorming laten ook een nauw verband zien met deaanwezigheid van H I in AGN. Sterrenstelsels met jonge stellaire population nijgennaar een hoge H I detectie niveau, suggererend dat stervorming in radio stelselsverband houdt met de aanwezigheid van een H I-rijk medium. Als gasaccretie eenperiodieke gebeurtenis is in radio stelsels, dan laat dit effect misschien ook eenafdruk achter op de stervormingsgeschiedenis.

Page 167: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis
Page 168: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

Acknowledgements

I think the good thing about astronomy (besides the obvious) is that it brings peopletogether from all over the world, we have the chance to get to know different cultures,and to see places that we never even dreamed of before. I spent four wonderful years inthe Netherlands, and during this time I’ve met great people and made awesome friends.I have to thank many people who made this possible and with whom I shared the Dutchlife.

First of all, I have to thank my supervisors, Raffaella and Tom. When I first visitedAstron to have my interview, you seemed like a really cool couple (with all due respect)and I was happy that you offered me this position. I didn’t know much about H I studies,and only little later did I realize that I was given the chance to learn from two of theleading experts of the field! I remember how impressed I was when I saw H I in a cubefor the first time, and I had this feeling many more times later thanks to our interestingprojects. From you I have learnt that even scientific papers should tell a story (whoknew?), you inspired me to become more independent and gave me a lot of freedom tofollow my own way. You both showed me a way of how to become a better scientist, thankyou for the opportunity! Raffaella, I could always count on your quick answer, thanksfor always being there to guide me. Your cheerful attitude is always very comforting,and I like your jokes. Tom, it’s easy to understand how much experienced someone is, ifjust by glancing at a plot or image he can tell you all the problems with the results. Ihope one day I will master this skill of yours.

Filippo, it was nice to share ideas and to work with you, it’s great that we wrotea paper together! You have been very helpful and supportive during my thesis writingperiod, I really appreciate that. I wish you good luck with you PhD, but you are in goodhands, I do not worry.

Gergö, my half-Hungarian fellow and officemate, I am happy that we shared an officetogether. We could work next to each other in silence for hours, and that was great :)But of course non-silent moments were even better, as we could talk in Hungarian. Bythe way, I think you are a really good dancer! Anke, I think you are just so much fun!Wish you guys great time in Germany!

Eva, it is not a long time that you are my officemate, but I got to know you as asweet and kind person. But I think this is already indicated by the fact that you likeCalvin and Hobbes and Winnie the Pooh (which passions I totally share) :).

‘H I friends’: Anastasia, my girl, thank you for always being there to support and

Page 169: University of Groningen The role of neutral hydrogen in the life of … · 2016. 3. 8. · The Role of Neutral Hydrogen in the Life of Galaxies and AGN A Spectral Stacking Analysis

cheer me up. We had a lot of fun together, hope you will visit me and we’ll go surfing;) Manolis, I promise you can always have one of my kibbelings. Antonino, I wonderif we will ever talk about something important... like AGN for example ;). Mustafa,my friend, your heart is big and your intentions always good, I wish you and Tuba nicetimes in the Netherlands. It was always nice to go to our meetings with the ‘little’ group,including Yiannis, Davide, Nadine, Mpati, Sarrvesh, Joris, Brad, Betsey. I would alsolike to thank Thijs and Mark for being so helpful with us.

The ‘Phd Kapteyn’ group: Giacomo, you always amaze me with all the quotes thatyou know from the top of your head :) Wish you good luck, arrivederci! Stefano, myAclo buddy, good luck with all your goals, but first and foremost, you should eat more!:) Wouter, it was always good that I could drop by your office with all my randomquestions :). Shoko, you surprised me with nice emails sometimes, thanks for being socaring :) Laura, Marisa, Francesco, Aku, Steven, Aaron, Ajinkya, thanks for the nicelunches, crazy conversations and many nice events that we attended together. Take careof each other and hope to see you later.

Jasmina and Aleksandar, thanks for the nice dinners, concerts, you have been verygood friends to me. Wish you and little Filip all the best and the happiest life!

Pratyush, you are a friend with whom it’s easy to have fun, but I like arguing withyou even more. Hope to visit you in India one day.

Rosina, it was always nice to chat and train with you, and I really like your style!Stephanie, Burcu, Stefania, I really like your easy-going personalities, you rock girls ;)!

Bertrand, it was great having you around :).Edwin, your ambition and achievements are truly impressive. You are really cool,

and also a good friend.Hugo (van Woerden), thank you so much for your stories about the war and the early

days of radio astronomy.I have to thank a lot of people for the many conversations at happy hours, coffee

breaks: Patrick, Johan, Veronica, Leon, Hans, Kyle, Hugo, Harish and Ivona, Jan, Joost,Stephen, Chris, and those whom I’m forgetting.

Hennie, Lucia, Jackie, Ginneke, Christa, Gonda, without you I would have beencompletely lost in all the bureaucratic processes. Thank you ladies for your patience andgenerous help over the years.

I would like to thank the Kapteyn Computer Group for their hard work. With specialthanks to Wim for helping me so many times: mulţumesc, la revedere!

Vibor, Paolo, Giuseppe, John, George, thank you guys for the many lifts to Astron,these rides were always fun. Vibor, thanks for teaching me how to play squash.

Carmen, Elizabeth and Tom, Maura, Javier, Mike, Guiffre, thank you guys for theawesome times!

Drága szüleim, köszönöm nektek, hogy mindig mellettem voltatok és támogattatok azutamon. Lehet, hogy most messze költözöm, de jó tudni, hogy van egy hely ahová mindigszivesen hazavárnak, és ahová jó hazatérni. Ági, köszönöm, hogy a kedvenc testvéremvagy, nálad jobb testvért nem is kivánhatna az ember lánya! Kivánom, hogy legyetekSzékellyel nagyon boldogok! Szófa! Endrus, köszönöm a borítot és a türelmedet a munkasorán! :)

Katinka GerébGroningen, September 2014