List of Figures

(1)

University of Groningen

Nowhere to hide: identifying AGN in the faint radio sky Radcliffe, Jack Frederick

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:

2019

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Radcliffe, J. F. (2019). Nowhere to hide: identifying AGN in the faint radio sky. University of Groningen.

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: 28-06-2021

(2)

Nowhere to Hide:

Identifying AGN in the faint radio sky

Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen

op gezag van de

rector magnificus prof. dr. E. Sterken en in overeenstemming met

de beslissing van het College voor Promoties.

en

ter verkrijging van de graad van Doctor of Philosophy aan de Faculty of Engineering and Physical Sciences van de

University of Manchester

De openbare verdediging zal plaatsvinden op dinsdag 23 april 2019 om 11:00 uur

door

Jack Frederick Radcliffe geboren op 27 oktober 1991 te Shrewsbury, Verenigd Koninkrijk

(3)

Prof. dr. P.D. Barthel Prof. dr. M.A. Garrett

Co-promotores Dr. R.J. Beswick Dr. T.W.B. Muxlow

Beoordelingscommissie Prof. dr. C. Jackson

Prof. dr. J.P. McKean Prof. dr. R.A. Windhorst Prof. dr. A.A. Zijlstra

(4)

ISBN: 978-94-034-1506-2 (electronic version)

The work described in this thesis was performed in the research group at the Kapteyn Astro- nomical Institute at the University of Groningen, the Netherlands, and the Jodrell Bank Centre for Astrophysics at the University of Manchester, United Kingdom.

Cover design: J.F. Radcliffe

Cover image credits: (AGN inset) Centaurus A - (Optical) ESO / WFI; (Submillimetre) MPIfR / ESO / APEX / A.Weiss et al.; (X-ray) NASA / CXC / CfA / R.Kraft et al.

(AGN inset) 3C66B - NRAO / AUI 2006.

(AGN inset) 3C31 - NRAO / AUI 2006.

(AGN inset) 3C296 - NRAO / AUI 2006.

(Lovell telescope) - Pete Birkinshaw, cc-by-2.0 (edited by J.F. Radcliffe).

(WSRT) - ASTRON 2017 (edited by J.F. Radcliffe).

(Background) GOODS-N - NASA, ESA, G. Illingworth (University of California, Santa Cruz), P.

Oesch (University of California, Santa Cruz; Yale University), R. Bouwens and I. Labbé (Leiden University), and the Science Team.

Printed by: Gildeprint

(5)

problems’

— Epictetus

(6)

7.1 Extending multi-source self-calibration: CASA and direction dependent calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 7.2 VLBI-selected AGN in the microJy flux regime . . . . . . . . . . . . . . 213 7.3 Identifying radio AGN for the MIGHTEE survey . . . . . . . . . . . . . 215 7.4 AGN feedback using integrated EVN and e-MERLIN observations . . . 215 7.5 SKA-VLBI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

Acknowledgements 219

(10)

List of Figures

1 The faint radio sky above the Green Bank radio telescope. . . . . xv

2 The 76m Lovell Telescope . . . . xvi

3 The Great Observatories Origins Deep-North (GOODS-N) as seen by the Hubble Space Telescope . . . xviii

4 De zwakke radiohemel boven de Green Bank-telescoop in de VS. . . . . xix

5 De Westerbork Synthesis Radio Telescoop (WSRT) . . . . xx

6 Het veld Great Observatories Origins Deep-North (GOODS-N) . . . xxii

1.1 Composition of the extragalactic radio source populations from Padovani et al. (2015) . . . . 2

1.2 The individual telescopes that make up the European VLBI Network and Very Long Baseline Array . . . . 4

1.3 A diagram of a two element radio interferometer . . . . 7

1.4 Illustration of time smearing. . . . 14

1.5 An example of interferometer non-coplanarity. . . . 15

1.6 An example wide-field VLBI survey using the multiple phase centre correlation technique to target the starburst galaxy M82. . . . 18

1.7 Simplified diagram illustrating the main components of Active Galactic Nuclei from Heckman & Best (2014) . . . . 23

1.8 Scaling relations between the central supermassive black hole and host galaxy properties which infers a co-evolution. . . . . 24

1.9 An example of negative AGN feedback in MS0735.6+7421 (McNamara et al., 2005) . . . . 26

1.10 AGN selection using emission line ratios and infra-red selection techniques. . . . . 31

1.11 The star formation rate density (SFRD) history from Novak et al. (2017). 34 2.1 The point source model of J123646+621405 and the_uv-stacked 1 Jy source used for self-calibration . . . . 46

2.2 Noise distribution profiles used to determine the detection threshold. . 48

ix

(11)

2.3 Compact radio source J123659+621833 before and after MSSC. . . . . . 50

2.4 Diagram of the MSSC algorithm and its implementation in AIPS. . . . . 55

3.1 Sources / sub-fields targeted by these GOODS-N observations. . . . . . 66

3.2 R.m.s. sensitivity for our 1.6 GHz EVN observations after primary beam correction. . . . . 67

3.3 Composite image of 1.4 GHz WSRT radio-KPNO optical overlay of the GOODS-N field, centred on the HDF-N (Garrett et al., 2000), surrounded by postage stamp images of the 1.6 GHz 31 VLBI detected sources presented in this chapter. . . . . 70

3.4 Redshift distribution for our detected VLBI sources. . . . . 78

3.5 Relative astrometric precision between Muxlow et al. (2005) and the EVN observations. . . . . 82

3.6 Radio power vs. redshift for our VLBI sources. . . . . 85

3.7 Brightness temperature distribution with respect to redshift. . . . . 87

3.8 Sky distribution of sources observed with e-MERLIN at 5 GHz. . . . . . 90

4.1 Multi-wavelength coverage of a sub-set of observations in the GOODS- N field . . . 100

4.2 Spitzer IRAC AGN selection criteria for the VLBI detected sample. . . . 111

4.3 The Stern et al. (2005) IRAC selection scheme and the WISE 3-band AGN identification schemes . . . 113

4.4 The KI and KIM selection schemes (Messias et al., 2012). . . . 116

4.5 AGN selection using the monochromatic radio excess parameters,_q₂₄ and_q₁₀₀ . . . 118

4.6 The total radio excess parameter_q_TIRfor those VLBI sources with FIR counterparts. . . 120

4.7 The breakdown of the AGN classification schemes with a VLBI-selected AGN sample. . . 130

4.8 The photon index against redshift for the X-ray detections illustrating the increase in obscuration at the_{z ∼ 2}overdensity. . . 132

4.9 IR colour-colour plots for various AGN samples and classification techniques using a matched area with X-ray, IR and radio coverage. . . 133

4.10 The radio excess parameter (_q₂₄) for all of the 1.5 GHz VLA sources with₂₄_µmcounterparts. . . . 134

4.11 Radio stacking on X-ray bright but radio quiet AGN. . . . 137

4.12 _q₂₄against redshift for a sample of radio-quiet X-ray AGN. . . . 140

4.13 HST F125W near-IR imaging of the host galaxies of the VLBI detected sources . . . 144

(12)

4.14 CFHT observations of the host galaxies of VLBI detected sources in the near-IR (_K_s band) . . . 147 4.15 1.5 GHz observations of FR-I source J123656+615659 . . . 161 4.16 The host galaxy of VLBI variable source J123642+621545 . . . 162 5.1 ₁^◦_{× 1}^◦ image of the 1996 VLA data illustrating the source removal

routine outlined in Section 5.2.3 . . . 172 5.2 Summary of the various source extraction tests performed usingPYBDSF. 177 5.3 Comparison between the 1.4 GHz (corrected using the in-band spectral

indices) and the 1.52 GHz fluxes of the JVLA 2018 epoch. . . 181 5.4 The two epoch comparison of the variability statistic,_V_s, to the modu-

lation index,_m. . . . 183 5.5 Light curves for the variable candidates. . . . 188 5.6 Spitzer IRAC AGN selection criteria for the variable candidate sample. 196 5.7 Type II supernovae / LLAGN candidate. . . . 197 5.8 The transient normalised areal densities, and upper limits, with respect

to flux density. . . 200 7.1 Implementation of MSSC to provide direction-dependent calibration

for VLBI observations. . . 212 7.2 The_uv-coverages for the integrated EVN e-MERLIN observations of

the GOODS-N field (EVN project code EG078G; Radcliffe et al. in prep.). 214 7.3 J123642+621331 (_{z ∼ 2}), a composite star-burst and AGN candidate

imaged using a combination of VLA, e-MERLIN and EVN data. . . . 216 7.4 The angular resolution at 1.5 GHz provided by the current suite of

radio telescopes . . . 218

(13)

(14)

List of Tables

1.1 An overview of key strengths and weaknesses of AGN selection in multiple wavebands. . . . . 29 2.1 Comparison between standard phase referencing, single source self-

calibration and MSSC calibration techniques. . . . . 52 3.1 EG078B observing strategy. . . . 64 3.2 1.6 GHz VLBI and 1.5 GHz VLA properties of the VLBI detected sources 75 3.3 Derived VLA & VLBI radio properties of the 31 GOODS-N AGN. . . . . 77 3.4 Phase calibrators used in the 5 GHz eMERLIN observations . . . . 91 3.5 Positions and relative offsets of J1234+619 . . . . 91 4.1 Multi-wavelength data available in the GOODS-N field that are used

in this analysis. . . . 101 4.2 The number of multi-wavelength VLBI counterparts and the waveband

sensitivities. . . . 107 4.3 IR AGN colour-colour selection scheme definitions. . . . 109 4.4 AGN classification schemes. . . . 123 4.5 A summary of the stacking resulted based upon the weighted mean

radio stack of X-ray selected AGN. . . . 139 5.1 A summary of the individual VLA/JVLA epochs utilised in this chapter 170 5.2 A summary of the multi-wavelength properties and AGN classification

for the variable candidates. . . . 191

xiii

(15)

(16)

Summary

Figure 1 | The faint radio sky above the Green Bank radio telescope in the USA. This composite image reveals processes that are invisible to the human eye. While the sky you see with your own eyes is dominated by stars, most of the objects are radio galaxies or quasars, which are driven by the accretion of matter upon a central super-massive black-hole in the centre of galaxies. These galaxies are often more than 5,000,000,000 light-years away. Image courtesy of NRAO / AUI.

Just as the night sky is strewn with stars, it is also scattered with cosmic radio sources invisible to the naked eye. Our eyes, whether aided by a telescope, are sensitive to the light created by stars and not these radio sources. To see these, we need radio telescopes to collect these radio waves broadcast by these radio sources; which

xv

(17)

Figure 2 | The 76m Lovell Telescope which is the largest telescope in e-MERLIN, the UK’s interferometer network consisting of seven telescopes. This telescope has been pivotal in the detection of pulsars and the first ever tracking of a satellite, the USSR’sSputnik satellite. Credit: ©Mike Peel / CC BY-SA 4.0 (via Wikimedia Commons).

are generally-speaking not stars but galaxies. While the stars we see at night are overshadowed by the sunlight scattered by our own atmosphere, radio sources are

‘visible’ during the day allowing us to observe them at anytime (Figure 1).

Astronomers have been investigating the radio sky ever since the 1950’s using parabolic dishes of a similar vein to those providing your satellite television. Soon after radio interferometers were conceived, providing much more detailed images that were possible with separate dishes. Interferometers led the field of radio astronomy, which resulted in the formation of impressive arrays of telescopes in the United Kingdom, the Netherlands and the USA.

This development, along with the advancement of radio astronomy as a science, resulted in the awarding of the 1974 Nobel prize in Physics to two of the pioneers, Martin Ryle and Antony Hewish. Over the last 70 years, radio astronomy has provided many insights into many astrophysical phenomena for which the United Kingdom and the Netherlands have played a significant role (Figure 2).

In this thesis, we investigate the occurrence of the two main sources of this radio emission within galaxies located at the furthest reaches of the Universe. The radio emission in these objects is formed via the synchrotron mechanism, which is a by-

(18)

product of two distinct processes. The first process is associated with the formation and death of new stars, while the second process relates to the suction of matter into the central super-massive black holes in the centre of all galaxies. The detection of synchrotron radiation from the birth and death of stars within our own galaxy, the Milky Way, led to the birth of radio astronomy in the 1930-40s. In the 1960s, astronomers detected the second process originating from the so-called ‘monsters’

in the centres of distant galaxies which led to the rapid expansion of radio into the mainstream. This rapid development was part in thanks to the Earth-rotation aperture synthesis technique and the subsequent formation of intercontinental Very Long Baseline Interferometry (VLBI) arrays that allowed unparalleled image sharpness (often surpassing the performance of optical instruments) to be achieved.

When we direct our radio telescopes away from the plane of the Milky Way, we see hundreds of bright (those ‘monsters’), millions of weak, billions of weak radio sources. In an area covering the size of the full moon, we can expect to see thousands of these radio sources. The latest surveys in the deepest corners of the Universe revealed that the origin of this emission is driven by star formation, nuclear activity from supermassive black holes or a combination of the two.

This thesis investigates the nature of the radio emission in distant galaxies to the lowest measurable level. This extreme sensitivity allows us to investigate these processes across a large range of intensities and distances.

The central two chapters of this thesis are dedicated to ultra-sharp imaging of a special piece of sky, the Hubble Deep Field-North (HDF-N), which constitutes the central part of the so-called GOODS-North deep field. This piece of sky has provided us with an extremely deep view into the Universe ever since the seminal Hubble Space Telescope observations in 1995. Since then, many different telescopes have focused upon this patch of sky, which has permitted the characteristics of approximately 10,000 galaxies spread across millions and billions of light years to be assessed. In this, and in similar fields, unprecedented insights into the evolution of the Universe as we know it today have been obtained.

The results of this thesis are as follows. In Chapter 2, we provide an improved technique in order to conduct deep VLBI observations of fields, such as the HDF-N. The standard calibration techniques that are commonly used turns out to be insufficient to get the ultra sensitive observations required for investigating the faintest radio sources - the crux of this work. The comparison of the classical and new analytical technique,

for the same set of observational data, shows a significant improvement.

In Chapter 3, we apply this new technique to a new, 24-hour VLBI observation of the HDF-N to analyse the compact radio source population. A similar observation from 2004 detects about 12 compact radio sources, whereas these new observations almost triple this number to a total of 31 sources. Additional research shows that the

(19)

Figure 3 | The Great Observatories Origins Deep-North (GOODS-N) as seen by the Hubble Space Telescope. The image reveals over 25,000 galaxies of which 1,000 are young galaxies which are observed when the Universe was less than 2 million years old. The GOODS fields have provided unparalleled insights into the evolution of the Universe. Credit: NASA, ESA, the GOODS Team and M. Giavalisco (STScI).

radio emission from galaxies span a wide range of distance and morphological type.

In Chapter 4, we conduct an in-depth analysis of these galaxies to answer a few key questions: what is the exact nature of radio radiation in relation to other properties of the galaxies? Why do we not detect more or less systems? How does VLBI perform compared to other techniques that study the nature of distant galaxies? We find that alternative methods of identifying these monsters work for some of these sources, but crucially not all thus illustrating the importance of such observations in identifying these active nuclear regions.

The thesis concludes in Chapter 5 with an analysis of a series of observations of the HDF-N spread across 22 years (from 1996 to 2018). This chapter shows that the time variability of the radio emission in distant galaxies is very rare, and often these variations are due to foreground effects. This pattern confirms the general result of this PhD research, which is that the weak radio radiation from distant galaxies is generated mainly by nuclear activity.

(20)

Samenvatting

Figuur 4 | De zwakke radiohemel boven de Green Bank-telescoop in de VS. De samengestelde afbeelding toont processen die onzichtbaar zijn voor het menselijk oog. Toont de hemel zoals je die met je ogen ziet louter sterren, de meeste objecten op deze afbeelding zijn radiogolven uitzendende sterrenstelsels

en quasars. Deze objecten worden aangedreven door superzware zwarte gaten in het centrum van

sterrenstelsels die vaak meer dan 5 miljard lichtjaar van ons verwijderd zijn. Credit: NRAO/AUI.

Net zoals de nachthemel bezaaid is met sterren, is deze bezaaid met kosmische radiobronnen. Ons oog - al dan niet gewapend met een kijker - is gevoelig voor licht, of anders gezegd: is in staat om het licht dat sterren uitzenden te detecteren. Radio- telescopen op hun beurt zijn gevoelig voor de radio-golven die die kosmische bronnen

xix

(21)

Figuur 5 | De Westerbork Synthesis Radio Telescoop (WSRT) werd in 1968 voltooid, naar een plan van prof. Oort. De WSRT heeft gedurende de afgelopen decennia belangwekkende ontdekkingen gedaan.

uitzenden; dat zijn in het algemeen gesproken niet sterren maar sterrenstelsels. Die radiobronnen zijn overigens ook overdag ‘zichtbaar’ (of beter: ‘hoorbaar’). Sterren zijn dat niet omdat het overdag te licht is; we kunnen ook zeggen dat de atmosferische voorgrond (verstrooid, dus blauw zonlicht) het veel zwakkere sterlicht overstraalt (Figuur 4).

Sinds de jaren 1950 van de vorige eeuw onderzoeken astronomen de radio ‘hemel’.

Men ontwikkelde al gauw radio-interferometers, die scherper konden waarnemen dan losse schotels; die interferometers leidden op hun beurt weer tot de indrukwekkende apertuur-synthesetelescopen waarvan de belangrijkste zich in Engeland, Nederland en de VS bevinden. Die ontwikkeling - en daarmee de ontwikkeling van de radioastronomie als wetenschap - leidde in 1974 tot de Nobelprijs voor de Natuurkunde, voor de twee pioniers Ryle en Hewish. De radioastronomie heeft gedurende de afgelopen 70 jaar een enorme vlucht genomen en we danken er enorm veel nieuw kosmisch inzicht aan. Ook Nederland heeft een grote faam in de mondiale radio-astronomie.

Natuurkundige processen in sterrenstelsels vormen de prominente mechanismen die radiostraling genereren. Twee van die processen en hun relatieve voorkomen tot in de verste verten van het heelal komen in dit proefschrift aan de orde. Het betreft hier de zogenaamde synchrotronstraling die een bijprodukt van de vorming van nieuwe

List of Figures

Contents

List of Figures

List of Tables

Summary

Samenvatting