Characterizing Dwarf and Diffuse Galaxy Populations in the Local Universe

Characterizing Dwarf and Diﬀuse Galaxy Populations in the Local Universe

Ananthan Karunakaran

A thesis submitted to the

Department of Physics, Engineering Physics & Astronomy

in conformity with the requirements for

the degree of Doctor of Philosophy

Queen’s University

Kingston, Ontario, Canada

August 2021

Extragalactic astronomy in the last decade has greatly beneﬁted from signiﬁcant improvements to (1) astronomical instrumentation and (2) data reduction and analysis techniques.

These improvements afford the ability to reveal ever fainter low surface brightness (LSB) dwarf galaxies at optical wavelengths. Thoroughly characterizing these dwarf galaxies is essential in furthering our understanding of how galaxies form and evolve. Accordingly, there have been several concerted efforts to compile and characterize the properties of (1) satellite dwarf galaxies around Milky Way-like systems beyond the Local Group and (2) a subset of large, LSBs known as Ultra-Diffuse Galaxies (UDGs) whose properties fall at the extreme end of the dwarf galaxy population. In this thesis, I will present studies that aim to illumi- nate both of these sub-fields by exploiting their neutral hydrogen (HI) gas and star-forming properties. The first of these studies focus on the HI gas content of LSB dwarf galaxy satellite candidates around the nearby spiral, M101, with the aim of obtaining spectroscopic distance estimates to them via their HI content and investigating any potential environmental trends related to their HI content. The second study compares the quenched and star-forming satellite populations from the Satellites Around Galactic Analogs (SAGA) survey to two state-of-the-art zoom-in hydrodynamical simulation suites, APOSTLE and Auriga, in order to understand the relatively low observed quenched satellite fraction. The second half of this thesis presents an extensive HI follow-up survey of optically-detected UDG candidates from the Systematically Measuring Ultra-Diffuse Galaxies (SMUDGes) survey. This follow-up

i survey aims to conﬁrm SMUDGes UDG candidates as bonaﬁde UDGs by obtaining spectroscopic distances estimates and to place constraints on their formation mechanisms using their HI properties.

ii Acknowledgements

These last 6 years would not have been possible without the amazing and constant support of my supervisor. Kristine, thank you for teaching and mentoring me in both academic and personal matters and for giving me the opportunity to pursue my passion.

I have had to pleasure of getting to know many graduate students in the department at

Queen’s and I am grateful for the friendships that have blossomed during my time here. It would have been tremendously more diﬃcult and miserable without you all.

To my boys, time has ﬂown by and I can always count on you for reminding me where it all started. Thank you.

It seems impossible to thank my amazing family in words for their neverending love, support, and understanding while I pursue my dream. I will forever be indebted to the sacriﬁces you have all made for me. Thank you so much and I love you all.

Finally, to my partner, Jacquie. Thank you for being there for me through my highs and lows, for reminding me that it is okay to step back and take a break, and, of course, for your help with picking out colour palettes for my graphs. I am excited for what our future holds and I love you so much.

iii Statement of Co-Authorship

The original work presented in this thesis was completed by the author under the supervi- sion of Dr. Kristine Spekkens at Queen’s University. Chapters2,3, and4 are versions of accepted/published journal articles that comply with the copyright permissions of the AAS journals. A brief description of the work done by the author for each article is provided at the beginning of each of these chapters and any contributions from collaborators are explicitly stated. These chapters have not been organized chronologically and instead are organized such that there is a more coherent ﬂow between them.

iv Contents

Abstract i

Acknowledgments iii

Statement of Co-Authorship iv

Table of Contentsv

List of Tables ix

List of Figuresx

List of Acronyms xii

List of Symbols xiv

Chapter 1 Introduction1

1.1 Dwarf Galaxies...... 5

1.1.1 A Brief Census of the Local Group Satellites ...... 5

1.1.2 Gas, Star Formation, and the Role of Environment...... 6

1.1.3 Dwarf Galaxies in Simulations ...... 9

1.2 Low Surface Brightness Galaxies and Modern LSB Surveys ...... 13

1.2.1 Satellite Systems around Massive Galaxies...... 16

1.2.2 Ultra-Diﬀuse Galaxies and their Formation Mechanisms ...... 19

v CONTENTS CONTENTS

1.2.3 LSB and UDG Follow-up Observations...... 22

1.3 This Work...... 24

Chapter 2 Neutral Hydrogen Observations of Low Surface Brightness Galax-

ies around M101 and NGC 5485 26

2.1 Abstract...... 27

2.2 Introduction...... 27

2.3 Sample Selection ...... 29

2.4 Observations and Data Reduction...... 31

2.5 Results...... 35

2.5.1 HI Detections ...... 35

2.5.2 HI Non-detections...... 39

2.5.3 Optical and Gas Properties...... 39

2.6 Discussion...... 42

2.6.1 Associations of HI Detections...... 43

2.6.2 HI non-detections and the NGC 5485 group ...... 46

2.6.3 Satellite Gas Richness...... 49

2.7 Conclusions...... 50

Chapter 3 Satellites Around Milky Way Analogs: Tension in the Num-

ber and Fraction of Quiescent Satellites Seen in Observations

Versus Simulations 53

3.1 Abstract...... 54

3.2 Introduction...... 54

3.3 Satellite Samples...... 56

3.3.1 Observed sample: SAGA-II...... 56

3.3.2 Simulated samples: APOSTLE and Auriga ...... 57

3.4 Identifying Star-Forming Satellites ...... 59

vi CONTENTS CONTENTS

3.4.1 Observed Star-Forming Satellites...... 60

3.4.2 Simulated Star-Forming Satellites...... 61

3.4.3 Comparing Star-Forming Satellites...... 63

3.5 Observed and Simulated Quenched Fractions ...... 64

3.6 Discussion and Conclusions ...... 66

3.7 Appendix: Testing Resolution and Star-formation Tracers in Simulations . . . 73

Chapter 4 Systematically Measuring Ultra Diﬀuse Galaxies in HI: Results

from the Pilot Survey 75

4.1 Abstract...... 76

4.2 Introduction...... 76

4.3 Sample Selection ...... 80

4.4 Observations and Data Reduction...... 81

4.5 Results...... 85

4.5.1 Properties of HI Detections...... 85

4.5.2 HI Non-detections ...... 92

4.6 Discussion...... 97

4.6.1 Comparing UDGs with HI Detections and Non-detections ...... 98

4.6.2 Constraining Formation Mechanisms...... 104

4.6.3 Disk Geometry and the BTFR ...... 105

4.7 Conclusions...... 112

Chapter 5 An Update On and the Outlook of the SMUDGes in HI Survey115

5.1 Introduction...... 115

5.2 Survey Progress...... 116

5.2.1 SMUDGes in the Legacy Surveys...... 116

5.2.2 SMUDGes in HI ...... 117

5.2.2.1 Campaigns 2 and 3 ...... 118

vii CONTENTS CONTENTS

5.2.2.2 Campaigns 4 and 5 ...... 118

5.3 Preliminary Results ...... 122

5.4 Outlook...... 129

Chapter 6 Summary and Conclusions 133

6.1 Summary ...... 133

6.1.1 Satellite Galaxies around Milky Way-like Systems ...... 133

6.1.2 Ultra-Diﬀuse Galaxies...... 136

6.2 Future Work ...... 139

6.2.1 Satellites around Milky Way-like Systems ...... 139

6.2.2 Ultra-Diﬀuse Galaxies...... 140

6.2.3 Upcoming Surveys...... 140

6.3 Conclusions...... 142

Bibliography 144

viii List of Tables

2.1 Target LSB Dwarf Candidate Properties ...... 34

2.2 Properties of LSB Dwarf Candidates with HI detections ...... 38

2.3 HI Upper Limits for Non-detections...... 42

3.1 UV properties of Observed satellites ...... 72

4.1 Target UDG Candidate Properties ...... 83

4.2 Properties of UDG with HI detections ...... 89

4.3 HI Properties of Dwarfs ...... 90

4.4 HI Properties of Non-detections...... 95

4.5 Inclinations and Rotation Velocities...... 111

5.1 Summary of SMUDGes in HI campaigns...... 120

ix List of Figures

1.1 Compilation of galaxy images demonstrating their diversity ...... 4

1.2 HI masses of Local Group dwarf galaxies as a function of galactocentric distance8

1.3 Comparison of modern simulation projects...... 12

1.4 Sample of LSB cutouts from HSC-SSP...... 15

1.5 The Robert C. Byrd Green Bank Telescope ...... 24

2.1 HI detection spectra of LSB satellites around M101 and NGC 5485 ...... 37

2.2 HI non-detection spectra of LSB satellites around M101 and NGC 5485 . . . 40

2.3 HI mass as a function of luminosity for LSB dwarfs around M101 and NGC

5485...... 43

2.4 HI mass to V-band luminosity ratio verse g r colour ...... 44 − 2.5 Distribution of LSBs in the region around NGC 5485...... 48

2.6 Luminosity versus projected separation for M101, Local Group, and SAGA

satellites...... 51

3.1 Curve-of-growth GALEX UV photometry...... 58

3.2 Comparison of Observed and Simulated Star-forming Satellite Samples . . . . 62

3.3 Comparison of Satellite Quenched Fractions in Observed and Simulated Samples 67

3.4 Stellar Mass as a function of Projected Distance for Star-forming and Quenched

Satellites...... 71

3.5 Testing eﬀects of Resolution and Star-formation Tracers in Simulations on

Satellite Quenched Fractions...... 74

x LIST OF FIGURES LIST OF FIGURES

4.1 Projected sky distribution of UDG candidates around Coma Cluster . . . . . 82

4.2 HI detections of UDG and dwarf galaxies in SMUDGes in HI Pilot Campaign 86

4.3 Comparing HI-derived properties to other HI samples...... 91

4.4 Gas-richness as a function of size for UDGs and dwarf galaxies ...... 92

4.5 HI mass to g-band luminosity ratio versus g r colour for the UDG sample . 94 − 4.6 Optical image cutouts of HI-detected UDGs...... 101

4.7 Optical image cutouts of UDG candidates with HI-nondetections ...... 102

4.8 Optical image cutouts of HI-detected dwarf galaxies ...... 103

4.9 Baryonic Tully-Fisher relation for UDGs and dwarfs from our follow-up sam-

ple alongside nearby galaxy samples ...... 108

5.1 Comparison of colour distributions across the diﬀerent SMUDGes in HI cam-

paigns...... 121

5.2 HI Detections of UDGs and Dwarfs in Campaigns 2 and 3 of SMUDGes in HI 123

5.3 HI mass to g-band luminosity ratio as a function of optical colour for Cam-

paigns 1 to 3 of SMUDGes in HI ...... 125

5.4 Comparison of HI properties from Campaigns 1 to 3 of SMUDGes in HI with

other HI samples...... 126

5.5 Gas-richness as a function of size for UDGs including new UDG detections . . 129

xi List of Acronyms

ΛCDM Λ- Cold Dark Matter

ALFALFA Arecibo Legacy Fast ALFA Survey

APOSTLE A Project Of Simulating The Local Environment

BCD blue compact dwarf

CFHT Canada-France-Hawaii Telescope

dE dwarf elliptical galaxy

DECam Dark Energy Camera

DES Dark Energy Survey

DGSAT Dwarf Galaxy Survey with Amateur Telescopes

dIrr dwarf irregular galaxy

dSph dwarf spheroidal galaxy

DSS Dynamic Schedueling System

ELVIS Exploring the Local Volume in Simulations

FUV Far-ultraviolet

FWHM full-width at half maximum

GALEX Galaxy Evolution Explorer

GBO Green Bank Observatory

GBT Green Bank Telescope

HIPASS HI Parkes All-Sky Survey

HSC Hyper Suprime-Cam

HST Hubble Space Telescope

IF intermediate frequency

JWST James Webb Space Telescope

xii Jy Jansky (10−26 W m−2 Hz−1) kpc kiloparsec (3.086 1019 m) × LAB Leiden/Argentine/Bonn Survey

LMC Large Magellanic Cloud

LSB low surface brightness

Mpc megaparsec (3.086 1022 m) × NED NASA/Extragalactic Database

NGC New Galactic Catalog

NSA NASA-Sloan Atlas

NUV Near-ultraviolet

Pan-STARRS Panoramic Survey Telescope and Rapid Response System pc parsec (3.086 1016 m) × RFI radio frequency interference

RMS root mean square

SAGA Satellites around Galactic Analogs

SDSS Sloan Digital Sky Survey

SEP Source Extractor Python

SMC Small Magellanic Cloud

SMUDGes Systematically Measuring Ultra-Diﬀuse Galaxies

UDG ultra-diﬀuse dalaxy

UV ultraviolet

VEGAS Versatile GBT Astronomical Spectrometer

VLT Very Large Telescope

VST VLT Survey Telescope

xiii List of Symbols

b/a axial ratio

c speed of light

D distance

DEC declination

fgas gas fraction

H0 Hubble constant

Hα hydrogen emission, Balmer series: n=3 to n=2 transition

HI neutral atomic hydrogen emission

H2 molecular hydrogen

i inclination

L solar Luminosity

Lx luminosity at in band x, e.g. g-band luminosity is Lg

Λ cosmological constant

λ wavelength

M solar mass

M∗ stellar mass

mx apparent magnitude in band x, same example above

Mbary total baryonic (gas+stellar) mass

Mgas total gas mass

xiv MHI HI gas mass

Mx absolute magnitude in band x, see example above

MHI Lx HI mass to luminosity ratio in band x

Mstellar Lx Stellar mass to luminosity ratio in band x µ surface brightness

ν frequency

RA right ascension reff eﬀective radius in angular units

Reff eﬀective radius in physical units rvir virial radius

SHI integrated HI ﬂux S N signal-to-noise ratio

σ∆V RMS noise in spectra at velocity resolution

∆V velocity resolution

Vrot rotation velocity

Vsys systemic velocity

W50 velocity width at 50% peak ﬂux density

W50,c corrected velocity width at 50% peak ﬂux density z redshift

xv Chapter 1

Introduction

Galaxies are composed of varying amounts of stars, gas, and dust all of which are embedded within the center of a massive cloud (or halo) of dark matter. Partially as a result of this varying combination of components, galaxies are diverse – they are observed to have a variety of morphologies (i.e. shapes) and sizes (see Figure 1.1). Within our current theory of galaxy formation, the smallest galaxies, referred to as dwarf galaxies, are suggested to merge together at early times in the Universe to form the much larger galaxies observed today. Dwarf galaxies are dominated by their dark matter content and have as few as several hundreds of stars to as many as a few billions of stars. Whereas, larger galaxies, like the

Milky Way, have well over 10 billion stars and their matter distributions are not dominated by dark matter. Dwarf galaxies are the most abundant type of galaxy in the Universe and are found in a variety of environments from isolated, ﬁeld regions to the dense centers of galaxy clusters.

The environment in which a galaxy resides also plays a crucial role in their evolution and, therefore, in dictating their physical properties. This is particularly true for dwarf galaxies as their low total masses make them more susceptible to the eﬀects of more massive nearby galaxies. As a result, dwarf galaxies have a diverse range of morphologies, sizes, and masses.

Those in less crowded environments are able to retain their star-forming gas and continue 1 2

forming stars undisturbed, while those in more crowded environments have had their gas reservoirs stripped away and the premature cessation of their star formation. Dwarf galaxies that fall into the former category are typically classified as dwarf irregulars (dIrr) for their irregular morphology due to patchy star-forming regions, which also produce bluer colours at optical wavelengths. In contrast, the latter category of dwarf galaxies fall into the dwarf elliptical (dE) or dwarf spheroidal (dSph) classification and display smoother, more uniform morphologies with redder colours at optical wavelengths reflecting their much older stellar populations.

The investigation of dwarf galaxies at the lowest masses have long been limited to the

Local Group of galaxies consisting of the Milky Way, Andromeda (M31), their respective satellite dwarf galaxies, and a few other nearby dwarf galaxies. The eﬀects mentioned above are clearly demonstrated within the Local Group, with the vast majority of nearby satellites having no on-going star formation. With recent technological advances, however, detections of much fainter dwarf galaxies at much larger distances have become feasible.

Despite this progress, these newly detected of dwarf galaxies have not been examined at the same level of detail as their counterparts in the Local Group. These dwarf galaxies fall into the low surface brightness regime and are classiﬁed as low surface brightness dwarf galaxies or LSB dwarfs. These galaxies have been found across a vast range of environments in large quantities. There have been several eﬀorts to detect LSBs in isolation, in Milky

Way/Local Group-like systems, and in dense galaxy groups and clusters.

There are many LSBs that resemble the faint dSph population of Local Group satellites that have been detected around nearby massive hosts. These systems can be used to test whether or not the trends (i.e. environmental dependencies) and properties (i.e. stellar population, gas content, etc) observed in the Local Group appear in other systems. Additionally, careful analysis of these systems may also provide useful insight for theoretical models of galaxy formation and evolution.

While some LSBs possess similar properties to Local Group dwarf galaxies, there are 3

some that are more extreme with very low surface brightness and extended sizes. These extreme LSBs have been recently dubbed Ultra-Diﬀuse Galaxies or UDGs. The majority of

UDGs have been detected in dense environments such as galaxies clusters, however, many more have also recently been discovered in isolation. The origin of UDGs is an active area of study on both observational and theoretical fronts.

In this thesis, we characterize the neutral atomic hydrogen (HI, “H One”)1 gas and star- forming properties of (1) dwarf galaxies around Milky Way/Local Group-like systems and

(2) UDGs across a variety of environments. We describe how we use these properties to derive useful quantities such as distances, star-formation rates, rotational velocities, among others. These derived quantities are critical in understanding the role of dwarf galaxies in the formation of more massive galaxies and can help place observational constraints on proposed theoretical formation mechanisms for UDGs. This thesis consists of two studies that focus on satellite dwarf galaxies (Chapters2 and3) and a summary of an on-going survey of UDGs

(Chapters4 and5). We summarize, present potential paths for future work, and conclude in Chapter6.

1HI emission results from the spin-ﬂip transition of Hydrogen atom’s electron and the spectral line is emitted at a wavelength of λ ≈21 cm or a frequency of ν ≈1.420406 GHz 4

Figure 1.1: A compilation of images showing the diversity within the vast galaxy population. The individual images have been resized to compare the sizes of each galaxy assuming they are at the same distance. It is interesting to note the diﬀerence in size between a massive spiral galaxy like M31 (Andromeda, left), a typical dwarf galaxy Fornax (top middle), and an Ultra-Diﬀuse Galaxy (middle). Image credit: Pieter van Dokkum 1.1. DWARF GALAXIES 5

1.1 Dwarf Galaxies

Within our current cosmological framework, known as Lambda Cold Dark Matter or ΛCDM,

the baryonic (i.e. luminous) components of galaxies are embedded within signiﬁcantly more

massive dark matter halos. The latter of which is a more signiﬁcant contributor to the

energy density of the Universe (Planck Collaboration et al., 2020). Crucially, within the

ΛCDM framework structure grows hierarchically: smaller dark matter halos merge over

time to build larger ones within which more massive galaxies reside in the Universe today

(e.g. Bullock and Boylan-Kolchin, 2017). Furthermore, these smaller dark matter halos,

within which dwarf galaxies reside, vastly outnumber more massive galaxies in the Universe.

Therefore, it is imperative to carefully study their properties to better understand their role

in galaxy formation and evolution.

1.1.1 A Brief Census of the Local Group Satellites

10.5−11 The Local Group of galaxies centers around the two massive (M∗ 10 M ) spiral ∼ 9.5 galaxies, the Milky Way and M31, each of which host a number of smaller (M∗ . 10 M ) satellite dwarf galaxies. Given their proximity, these satellite dwarf galaxies provide the best

opportunity to conduct detailed observations of galaxies in this mass regime. The brightest

(MV < 9) of these satellite galaxies are commonly referred to as the “Classical” satellites. − The SDSS (Sloan Digital Sky Survey, Adelman-McCarthy et al., 2007) era doubled the

number of known satellites around the Milky Way, bringing the total to 23 (Koposov et al.,

2008). There have since been numerous surveys which have detected many more satellites

at even fainter magnitudes (MV > 5, e.g. ATLAS, Torrealba et al. 2016a,b; Dark Energy − Survey, Bechtol et al. 2015; Drlica-Wagner et al. 2015; Kim et al. 2015; Kim and Jerjen

2015; Koposov et al. 2015; Luque et al. 2016; Drlica-Wagner et al. 2020; Hyper Suprime-

Cam Subaru Strategic Program, Homma et al. 2016, 2018; Gaia, Torrealba et al. 2019),

bringing the total number of satellites to 50. There have been similar eﬀorts to compile ∼ 1.1. DWARF GALAXIES 6

fainter satellites around M31 (e.g. Martin et al. 2009; Bell et al. 2011; Slater et al. 2011;

Martin et al. 2013b,a, 2016, 2017; Weisz et al. 2019) which have brought the total number of satellites around M31 to over 30.

1.1.2 Gas, Star Formation, and the Role of Environment

As previously mentioned the dwarf galaxies in the Local Group have vast range of physical properties and this diversity extends to their neutral atomic hydrogen (HI) content and the star-forming properties. Studying these properties from a population perspective is readily accomplished through large sky surveys.

The Arecibo Legacy Fast ALFA (ALFALFA) Survey is a blind HI survey covering a large fraction of the sky ( 7000 deg2) and has detected over 30,000 extragalactic HI sources ∼ (Haynes et al., 2018). There have been many follow-up studies of dwarf galaxies which initially stem from the ALFALFA survey. It has been used to characterize satellite galaxies in the Local Group (e.g. Grcevich and Putman, 2009; Spekkens et al., 2014), detect gas-rich dwarf galaxies at the edge of the Local Group such as Leo P (Giovanelli et al., 2013), and characterize the broader dwarf galaxy population (see below).

There are several methods that can be used to tell if a galaxy is star-forming and how quickly they are doing so. An extremely useful space observatory in this regard is GALEX

(Galaxy Evolution Explorer) which carried out an expansive imaging survey at ultraviolet

(UV) wavelengths along with several deeper, smaller spatial coverage surveys (Morrissey

et al., 2007; Bianchi, 2009). These UV data are a direct tracer of on-going and recent star

formation in galaxies as the emission stems from the young, high and intermediate mass (O

and B) stars. The GALEX All-Sky Imaging Survey (AIS) covered an immense fraction of the

sky ( 27000 deg2) in the Near-UV (NUV) and Far-UV (FUV) wavelengths (1344-2831Å). ∼ The combination of the sensitivity of these data with their spatial resolution is important for completely understanding the star-forming nature of dwarf galaxies in particular. Their intrinsic faintness makes it diﬃcult to use other star-formation tracers (i.e. Hα) as they 1.1. DWARF GALAXIES 7

can require prohibitive integration times (i.e. via long-slit spectroscopy) or they may miss one of the few star-forming regions within the dwarf galaxy (i.e. via multi-object ﬁbre spectroscopy). Given that star formation proceeds through the eventual cooling and collapse of HI gas, it should be expected and has been shown that most galaxies, including dwarf galaxies, are detected in both HI and UV (e.g. Huang et al., 2012a). Therefore, there is great utility in exploiting the complementarity of these tracers when attempting to understand the dwarf galaxy population.

A key factor that dictates the diversity of Local Group dwarf galaxies is their environment. The vast majority of satellites in the Local Group that fall within the virial radius of the Milky Way or Andromeda have signiﬁcantly depleted HI reservoirs and are quenched

(i.e. not actively forming stars), while those beyond the virial radius are typically rich in HI gas and are star-forming (Grcevich and Putman, 2009; Spekkens et al., 2014; Putman et al.,

2021). This trend can be clearly seen in Figure 1.2 which shows the most recent investigation of the HI properties of dwarf galaxies in the Local Group using HI data from the GALFA survey (Peek et al., 2011, 2018). Of course, there are a few exceptions to this trend, notably the gas-rich satellites: the Large and Small Magellanic Clouds (LMC and SMC) around the

Milky Way and IC10, LGS3, NGC185, and NGC205 around M31. However, most of these gas-rich satellites are much more massive than the remainder of the satellite population and are less susceptible to the eﬀects of the Milky Way or M31 as will be described below.

The primary mechanism that can readily remove HI from these satellites as they fall into the potential of their massive hosts is through ram-pressure stripping by the host’s hot gas corona. The seminal work of Gunn and Gott(1972) lays out the framework for ram- pressure stripping in a galaxy cluster environment and this work has since been explored in the context of Milky Way-like systems by Gatto et al.(2013). The standard criterion for ram-pressure stripping to occur is

2 2 ρcorona(vorbital) & ρdwarf (σdwarf ) , (1.1) 1.1. DWARF GALAXIES 8

Figure 1.2: HI masses (ﬁlled symbols) and HI mass upper limits (open symbols) of Local Group satellites as a function of distance from either the Milky Way (red symbols) or M31 (blue symbols). The shaded regions of the same colours correspond to the range of adopted virial radii for the Milky Way and M31. Figure from Putman et al.(2021).

where ρcorona and ρdwarf are the densities of the massive galaxy’s (e.g. Milky Way or M31’s) corona and the dwarf galaxy’s gas. vorbital and σdwarf are the infalling dwarf galaxy’s orbital velocity and velocity dispersion, the latter of which can be used as a proxy for the dwarf galaxy’s total mass. Following this, it is evident that lower mass galaxies are more susceptible to ram-pressure stripping in dense environments. Infalling satellites will also feel tidal (i.e. gravitational) forces from their massive hot galaxy. While ram-pressure plays a signiﬁcant role in stripping gas, it has been shown that a combination of ram-pressure and tidal forces are more eﬀective at stripping the gas from a satellite than either force alone (Emerick et al., 2016). Another external mechanism that can readily remove gas from dwarf galaxies

6 is reionization which is only eﬀective for particularly low mass dwarf galaxies (M∗ . 10 M , Rodriguez Wimberly et al., 2019). 1.1. DWARF GALAXIES 9

In less dense environments, however, dwarf galaxies tend to thrive with sizeable HI reservoirs and are almost uniformly star-forming. An essential and comprehensive study of dwarf galaxies from the ALFALFA 40% catalog was presented in Huang et al.(2012b). In

7.7 addition to the HI properties of all low HI mass (MHI < 10 M ) sources, Huang et al.

(2012b) used optical photometric and spectroscopic properties from SDSS in combination with UV properties from GALEX. By deﬁnition, all dwarf galaxies in this sample will be more gas rich by virtue of being detectable in HI. While a detailed environmental analysis is not conducted, they do separate the dwarfs in their sample into those that are members of the Virgo cluster and those that are not. On average, those outside of the Virgo cluster have lower stellar masses, higher gas fractions (MHI /M∗), and have broadly comparable

star-formation rates. The sample as a whole demonstrates that dwarf galaxies that are

selected in HI surveys have higher gas fractions (by deﬁnition), bluer colours, and higher

star-formation rates.

An alternative approach to understanding the role of environment on dwarf galaxies is to

investigate their star-forming properties. Using the NASA-Sloan Atlas (Blanton et al., 2011),

Geha et al.(2012b) estimate the number and fraction of isolated (“ﬁeld”) dwarf galaxies that

are quenched as a function of stellar mass. In their analysis, they deﬁne an isolated dwarf

10.5 galaxy as one that has no massive (M∗ & 10 M ) neighbour within 1.5 Mpc and deﬁne a galaxy as quenched using a combination of its Hα equivalent width and the strength of its

Dn4000 break. The key result from their investigation is the absence of quenched ﬁeld dwarf

9 galaxies below M∗ < 10 M . Commensurate with the results from Huang et al.(2012b), it

is clear that dwarf galaxies in low density environments are able to retain their HI reservoirs

and continue forming stars unperturbed.

1.1.3 Dwarf Galaxies in Simulations

Over the past two decades there have been exceedingly more detailed large-scale cosmologi-

cal hydrodynamical simulations used to test and constrain our theories of galaxy formation 1.1. DWARF GALAXIES 10

and evolution. As their name describes, these simulations include both dark matter and baryons with diﬀerent implementations of astrophysical processes (i.e. star formation and evolution, feedback processes, active galactic nuclei, magnetic ﬁelds, etc). A few notable large-scale simulation projects are EAGLE (Evolution and Assembly of GaLaxies and their

Environments, Schaye et al., 2015), Illustris (Vogelsberger et al., 2014) and its successor

IllustrisTNG (TNG, Pillepich et al., 2018) which are unprecedented in their size and resolution. The largest of the TNG simulation volumes is 300 Mpc3 with dark and baryonic ∼ 7 particle resolutions on the order of 10 M . The combination of volume and particle reso- ∼ lution results in hundreds of thousands of galaxies within the simulation. The sizes of these diﬀerent simulation projects is displayed in Figure 1.3. The vertical axis shows the baryonic particle resolution while the lower and upper horizontal axes show the number of resolved,

9 relatively massive (M∗ & 10 M ) galaxies and the simulation volume, respectively. The TNG project is unmatched in its combination of size and resolution (see green circle), especially when compared to its predecessor Illustris (orange circle) and EAGLE (red circle).

These projects lie in the “box regime” shown as the shaded region in the right side of the

ﬁgure. Although, ﬁner particle and spatial resolutions are required to test and constrain the formation and evolutionary processes and properties of dwarf galaxies.

Simulation suites capable of such resolution reside in the “zoom regime” in the top-left of Figure 1.3. As their name suggests, these are zoomed-in regions of interest from larger, lower resolution simulations that are re-run at higher resolution (baryonic particle mass

6 < 10 M ). Many of these suites typically focus on speciﬁc environments such as galaxy clusters (e.g. RomulusC, Tremmel et al., 2019), the Local Group (e.g. APOSTLE, Sawala et al. 2016; Fattahi et al. 2016; FIRE-2, Garrison-Kimmel et al. 2019a; NIHAO, Buck et al.

2019), and individual Milky Way-like galaxies (e.g. Auriga, Grand et al. 2017; Latte, Wetzel et al. 2016). These simulation suites are able to reproduce many observed properties on these vast scales and allow us to better understand the mechanisms that aﬀect the evolution of dwarf galaxies in diﬀerent environments. For example, all of the aforementioned Local Group 1.1. DWARF GALAXIES 11

and Milky Way simulations are able to accurately reproduce the broad distribution of star- formation histories of dwarf galaxies in the Local Group despite being run using diﬀerent implementations of subgrid physics and hydrodynamic methods. An intriguing next step forward is demonstrated by the intermediate volume ( 50 Mpc3), high resolution (baryonic ∼ 5 particle mass 10 M ) TNG50 simulations (blue circle in Figure 1.3, Nelson et al., 2019) ∼ which provides “zoom-in” resolution at a “box” volume. These eﬀorts show a promising future for characterizing dwarf galaxies in simulations. 1.1. DWARF GALAXIES 12

Figure 1.3: A comparison of the modern simulation projects with the vertical axis showing the mass resolution of baryons and the lower and upper horizontal axes showing the number of resolved galaxies and the simulation volume, respectively. The shaded regions in the top-left and middle-right display the “zoom regime” and “box regime” category that most simulation projects fall into. The diagonal dotted lines show constant resolution element counts. The grey vertical and horizontal arrows in the top-right respectively represent a factor of 20 and 10 increase in computational cost. Figure from Nelson et al.(2019) 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 13

1.2 Low Surface Brightness Galaxies and Modern LSB Sur-

veys

The low surface brightness (LSB) galaxy population was long postulated to be a signiﬁcant contribution to the total galaxy population and were not present in early galaxy surveys due to observational biases (Zwicky, 1957). LSB galaxies in early studies were typically deﬁned as

2 having particularly low central B band surface brightness, µ0,B > 23 mag arcsec , relative − to the broader galaxy population (Impey and Bothun, 1997). These early surveys employed

photographic plates and early iterations of CCDs to search for LSB galaxies primarily in

group and cluster environments (e.g. Binggeli et al., 1984; Davies et al., 1988; Impey et al.,

1988; Irwin et al., 1990; Bothun et al., 1991), though some began to explore lower density

environments (e.g. Roukema and Peterson, 1995). Given their faint magnitudes, many LSB

galaxies were classiﬁed as dwarf galaxies, speciﬁcally dE. However, extreme examples were

also discovered: Malin 1 is a giant low surface brightness spiral galaxy with a disk scale

length greater than the total size of the Milky Way (Bothun et al., 1987). Much work on LSB

galaxies was also done in HI using interferometric observations from the Very Large Array

(e.g. de Blok et al., 1996) and single-dish observations from facilities such as Arecibo (e.g.

O’Neil et al., 2004; Trachternach et al., 2006). These studies hinted at a surface brightness

dependence on HI content as most LSB galaxies were found to have relatively large HI

reservoirs.

More recently, there have been eﬀorts to either use dedicated instrumentation suitable

for low surface brightness observations or exploit the depth and coverage of large, wide-ﬁeld

sky surveys. The most notable of the dedicated instrumentation eﬀorts is the Dragonﬂy

Telephoto Array (Abraham and van Dokkum, 2014). The Dragonﬂy Telephoto Array is a

pair of telescopes that are is composed of 24 high-end consumer telephoto camera lenses

coupled to astronomical grade CCDs which, when combined, are particularly sensitive to

low surface brightness emission. This sensitivity is largely owning to the minimization of 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 14

internal reﬂections within the telephoto lenses. van Dokkum et al.(2014) demonstrate that

2 Dragonﬂy can readily detect extremely low surface brightness (µg 32 mag arcsec ) features ∼ that have previously been out of reach. On the other hand, Fliri and Trujillo(2016a) present the IAC Stripe 82 Legacy Project which used careful data reduction techniques of deep SDSS

Stripe 82 data to retain as much low surface brightness emission as possible. The IAC Stripe

82 Legacy Project is able to reach low surface brightness ( 28.5 mag arcsec2) while covering ∼ a large area of the sky, 275 deg2. Greco et al.(2018) presented a sample of over 750 new LSB galaxies detected in the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP, Aihara et al., 2018) and demonstrated a clear diversity in the optical properties of LSB galaxies at the high resolution aﬀorded by the 8.2-meter Subaru Telescope (see Figure 1.4). These are just a few of the many eﬀorts to detect ever fainter emission and galaxies in the local

Universe. 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 15

Figure 1.4: gri composite colour images (55 55 arcseconds) of blue (left panels) and red − × (right panels) LSBs from the HSC-SSP. Angular sizes of the LSBs decrease from top-left to bottom-right of each group. The large diversity in their properties is clearly seen from this subsample alone. Figure from Greco et al.(2018) 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 16

1.2.1 Satellite Systems around Massive Galaxies

The Milky Way and the Local Group provide the ideal laboratory to study the properties of satellite dwarf galaxies and how they are aﬀected by their local environment. However, if we aim to interpret dwarf galaxy properties and their environment in the broader context of cosmological galaxy formation, evolution, and the intrinsic cosmic variance, then we must consider moving to systems beyond the Local Group. While we may not be able to observe satellites as faint at larger distances, we will be able to make statistical statements with respect to their properties.

Systematic searches for satellite dwarf galaxies around Milky Way analogs has been an on-going eﬀort since the early work of Zaritsky et al.(1993). The push toward lower surface brightness in more recent searches enables the detection of satellites further down the stellar mass function and out to larger distances. However, before proceeding, it should be noted that these low surface brightness searches are by deﬁnition done through integrated light, as opposed to resolved star searches in much closer systems (i.e. . 5 Mpc, Müller et al. 2018b) which allow for distance measurements through standard candles (e.g. Tip of the Red

Giant Branch, TRGB). Therefore, unless otherwise speciﬁed, LSBs satellites in the following studies should be considered LSB satellite candidates when lacking distance measurements.

Certain groups employ a similar approach to the Dragonﬂy team and made use of small aperture refractor telescopes in order to image nearby massive galaxies. The Dwarf Galaxy

Survey with Amateur Telescopes (DGSAT, Javanmardi et al., 2016) builds upon the early work of Martínez-Delgado et al.(2008) to detect LSB candidates around half of a dozen

Milky Way-like galaxies in their ﬁrst survey. The second run expanded the sample to three additional hosts (Henkel et al., 2017). The DGSAT survey has detected dozens of LSBs candidates around these hosts to-date. In the same vein as DGSAT, Tief Belichtete Galaxien

(TBG, “very long exposed galaxies”) is another group that has put together a concerted eﬀort to use small telescopes to detect LSB candidates around massive hosts in the Local Volume, 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 17

detecting over three dozen LSB candidates in the Local Volume (Karachentsev et al., 2014a,

2015, 2020).

Easily one of the most extensively investigated Milky Way-like hosts is M101. Merritt et al.(2014) used the deep Dragonﬂy Telephoto Array data ﬁrst used to probe M101’s stellar halo (van Dokkum et al., 2014) to search for nearby LSB candidates and detect 7 new LSB candidates. In addition to the 7 LSB candidates from Merritt et al.(2014), Karachentsev et al.(2015) presented 4 new LSB candidates. Moving beyond obtaining new observational data, Bennet et al.(2017a) used existing deep optical imaging from the Canada France

Hawaii Telescope Legacy Survey (CFHTLS) to search for new LSB candidates in a 9 deg2

region around M101 using a novel, semi-automated detection algorithm to detect 38 new

LSB candidates. Interestingly, most LSB candidates were located to the East of M101 and

project near a background group of galaxies centered on the massive elliptical galaxy, NGC

5485. Müller et al.(2017) extended their search for LSB candidates to a much broader region

(including M101, M51, and M63) using SDSS data and their own implementation of a LSB

detection algorithm to 15 new LSB candidates in a 300 deg2 region.

Subsequent follow-up observations of this new, rather extensive, sample of LSB candidates set out to confirm their association. Merritt et al.(2016) and Danieli et al.(2017) and presented TRGB distances from Hubble Space Telescope (HST) observations of the LSB candidates detected by Dragonfly. These efforts found that only three are true satellites of

M101, while the remainder are likely members of the NGC 5485 background group. A similar, though more extensive eﬀort to obtain TRGB distances with HST was presented in Bennet et al.(2019a, 2020), who found that of the 23 LSB candidates only two have distances consistent with M101. Carlsten et al.(2019) took an alternative approach to estimate distances to the many LSB candidates around M101. They used the Surface Brightness Fluctuations

(SBF)2 method to estimate distances to all of the known LSB candidates around M101

2A secondary distance estimation method that exploits the inverse relationship between pixel-to-pixel ﬂux variations of extended sources and their distance. 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 18

using the same CFHTLS data as Bennet et al.(2017a). The vast majority of the distance estimates they presented had wide ranges that could place them as near as M101 or in the distant background. While this method is useful to employ with extant data and provide a

ﬁrst-order estimate on distances, it is inherently more uncertain than resolved star distance estimates (see Bennet et al., 2020) or other follow-up methods (see below).

Using archival CFHT data, Carlsten et al.(2020) performed a systematic search for LSB candidates around 10 hosts in the Local Volume detecting 93 new satellite candidates in these systems. They used their established SBF technique to estimate distances for the satellite systems around these hosts (Carlsten et al., 2021b) and found that a signiﬁcant portion of these candidates are likely background galaxies. Their most recent eﬀorts have expanded their sample to over 200 low-mass satellites in the Local Volume as part of the Exploration of Local VolumE Satellites (ELVES) Survey (Carlsten et al., 2021a) which includes their sample of satellites from their earlier works.

Arguably one of the most ambitious recent eﬀorts to compile, conﬁrm, and characterize satellite galaxies around Milky Way analogs is the Satellites Around Galactic Analogs

(SAGA) Survey (Geha et al., 2017b). The SAGA Survey aims to compile satellite candidates within the projected virial radii of nearby ( 25 40 Mpc) Milky Way-like galaxies, the ∼ − latter of which are selected primarily by their K band luminosity. At the completion of − the survey, they aim to compile the satellite systems of 100 Milky Way-like galaxies with

satellites brighter than the Leo I dwarf (Mr < 12.3). The basic strategy of this survey − is to select all potential satellite candidates from a number of existing galaxy catalogs, se-

lect a subset that satisﬁes several optical property criteria (e.g. surface brightness, colour,

magnitude), and conduct spectroscopic follow-up observations to obtain redshifts to conﬁrm

them as true satellites. In their recent Stage-II release, Mao et al.(2021) present an updated

catalog of 127 spectroscopically-conﬁrmed satellites around 36 Milky Way-like hosts. The

intriguing result of this sample is that the vast majority of satellites have signiﬁcant Hα

emission and are, therefore, actively star-forming. This result is directly at odds with the 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 19

well-established trend seen with satellites of the Milky Way and M31 (see Figure 1.2).

1.2.2 Ultra-Diﬀuse Galaxies and their Formation Mechanisms

Large LSB dwarf galaxies have been known to exist in cluster environments from early

LSB surveys (e.g. Sandage and Binggeli, 1984). It was not until the deep optical survey with the Dragonfly Telephoto Array of the Coma Cluster that a significant population of extended LSB dwarf galaxies were detected (van Dokkum et al., 2015a). They dubbed these particularly large objects “Ultra-Diffuse Galaxies” (UDGs) and provided an informal definition of physical properties that has now been widely adopted: UDGs are defined

−2 as having low central surface brightness, µ0,g & 24 mag arcsec , and large physical sizes, reff & 1.5 kpc. This original study around the Coma Cluster was the spark to search for UDG candidates in existing and new imaging data in not only the Coma Cluster but several other nearby galaxy clusters such as Virgo and Fornax (e.g. Koda et al., 2015a; Mihos et al., 2015a; Davies et al., 2016; Venhola et al., 2017). The number density of UDG candidates decreases from cluster environments to group environments (e.g. Román and Trujillo, 2017; Shi et al., 2017;

Merritt et al., 2016) to lower density, ﬁeld environments (e.g. Martínez-Delgado et al., 2016;

Prole et al., 2019b). This trend is conﬁrmed by van der Burg et al.(2017), who show that there is a correlation between the number of UDG candidates and the halo masses of the groups/clusters in which they reside.

In terms of their properties, UDGs span a broad range. Various photometric and spectroscopic studies of UDGs have been conducted to understand their chemical compositions, stellar populations, and velocity dispersions (van Dokkum et al., 2016; Pandya et al., 2018;

Martín-Navarro et al., 2019; Chilingarian et al., 2019). While the majority of UDG candidates appear to be red, round, and passive there is a growing population of UDG candidates that appear blue, irregular, and gas-rich/star-forming (Román and Trujillo, 2017; Leisman et al., 2017; Greco et al., 2018; Prole et al., 2019b). The former are typically discovered in 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 20

higher density environments while the latter are found in lower ones, and the relationship between these blue and red objects is not entirely clear. There is at least some evidence that they are related: Román and Trujillo(2017) discovered an interesting population of 11 blue and red UDG candidates around multiple Hickson Compact Groups (HCG)3. They showed

that the blue UDG candidates were located at larger projected distances from the group

centers than the red ones and that passively evolving the colours and surface brightness pro-

ﬁles of the blue objects produced systems similar to the red ones. Blue UDGs may therefore

be the progenitors of red UDGs, the latter being quenched when they fall into dense envi-

ronments. The field would benefit from a coherent effort to detect and characterize UDGs

across a variety of environments.

The need for large, systematic searches for UDG candidates is clear. Zaritsky et al.

(2019) have initiated an eﬀort to achieve this, called Systematically Measuring Ultra-Diﬀuse

Galaxies (SMUDGes). The goal of SMUDGes is to obtain a signiﬁcant sample of large UDGs

(reff & 2.5kpc) across all environments, from the ﬁeld to dense clusters. This project mines the optical data from the DESI (Dark Energy Spectroscopic Instrument) Legacy Imaging

Surveys which cover 14, 000 deg2 of the sky (Dey et al., 2019b). The initial SMUDGes ∼ search focused on a 300 deg2 region centered on the Coma Cluster, where 275 new large UDG

candidates were discovered. Extrapolating this detection rate to the entire survey region,

they estimate a potential order of magnitude increase in the sample size.

All of these observational eﬀorts aim to shed light on the physical properties of UDGs

and UDG candidates to help understand how these UDGs formed in the ﬁrst place. There

are generally two categories for these formation mechanisms: external and internal. In ex-

ternal mechanisms, UDGs found in dense environments are more susceptible and likely to

experience the eﬀects of their environment and their passive properties are readily explained

by eﬀect such as ram-pressure stripping and/or tidal heating (Yozin and Bekki, 2015; Car-

leton et al., 2019; Tremmel et al., 2020; Sales et al., 2020). The internal UDG formation

3HCGs are composed of a few (3-5) galaxies in a compact conﬁguration and are relatively isolated. 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 21

mechanisms are more applicable to those which formed in less dense environments. These mechanisms have their own challenges and a few of them are outlined below.

The ﬁrst formation mechanism posits that UDGs are embedded in high-spin dark matter halos, resulting in very extended disks at a given stellar mass. Using their own self-consistent simulations, Yozin and Bekki(2015) showed that dwarf galaxies that possess large disks will reproduce the properties of the red UDGs when they fall into a cluster potential. To look into this further, Amorisco and Loeb(2016) aimed to show that UDGs are part of the high-spin tail of the regular dwarf galaxy population. The spin is a measure of the angular momentum of a galaxy’s dark matter halo and is known to dictate the size of the baryonic distribution within it (Mo et al., 1998). Using a simple analytical model they not only replicate the properties of typical galaxies in the Virgo cluster but are able to show that the primary distinguishing factor between UDGs and the broader galaxy population is their spin distribution. Constraining this mechanism is diﬃcult given the fact that the measurement of the spin (i.e. a highly uncertain estimator based on observational properties) would have to have been early in the UDG’s formation and would have likely changed over time throughout the UDG’s evolution.

Another mechanism through which UDGs may form is via stellar feedback. Using the

NIHAO zoom-in cosmological simulations, di Cintio et al.(2017) show that UDGs can form in the ﬁeld via a cycle of gas outﬂows with subsequent stellar and dark matter expansion caused by bursty star formation at redshifts between z=4 and 1. This process essentially

“puffs up” the baryons and results in distinctly extended, low-surface brightness field galaxies with cored dark matter profiles. Notably, their findings also predict that field UDGs with larger effective radii have larger HI masses. These isolated, gas-rich UDGs are then stripped and quenched when they fall into a cluster potential and they would eventually resemble the passive UDG population.

A third interesting mechanism through which UDGs in low density environments may form is through mergers. Using the ROMULUS25 simulations, Wright et al.(2021) present 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 22

a large sample of UDGs in low density environments and show that they are predominantly formed through major mergers. These events occur at very early times in their formation which cause star formation to dwindle near their centers and continue at the edges. They also find that while both UDGs and dwarfs experience mergers, the effects are more drastic on UDGs and, interestingly, that these effects are temporary with similar spin distributions between the two populations at present-day.

These mechanisms suggest that there are potentially multiple mechanisms through which

UDGs can form. The path forward to placing constraints on these mechanisms requires not only a sample of UDG candidates across a variety of environments but a sample of conﬁrmed

UDGs with distance measurements.

1.2.3 LSB and UDG Follow-up Observations

While imaging surveys are now discovering large numbers of LSB dwarfs and UDG candidates, distances are needed to conﬁrm their associations to any potential hosts and to measure their physical properties. In particular, distances are needed to convert the angular sizes of UDG candidates to physical sizes, and therefore verify that they satisfy the UDG size criterion. As the majority of UDG candidates have been discovered in the cluster environment, it is reasonable to assume they are associated with the cluster and lie at a similar distance. Although, around individual host galaxies, careful considerations need to be made regarding the broader projected environment. Recall that the vast majority of the satellite candidates around M101 are not true satellites. Accordingly, this “distance by association” method cannot be used when there is no nearby massive host (i.e. ﬁeld UDG candidates) and therefore a direct distance measure is needed.

By virtue of their low surface brightness, however, obtaining distances to faint LSB and UDG candidates is a diﬃcult and observationally expensive process. van Dokkum et al.

(2015b) obtained ground-based observations of DF44, the second brightest and largest UDG 1.2. LOW SURFACE BRIGHTNESS GALAXIES AND MODERN LSB SURVEYS 23

in their sample, with the Keck I 10-m Telescope using the Low Resolution Imaging Spectrom- eter. Even this relatively bright target required a 1.5-hour integration to obtain a redshift.

Other groups have also conducted spectroscopic follow-up of bright LSB and UDG candidates with similar or longer required integration times (Kadowaki et al. 2017; Ferré-Mateu et al. 2018; Ruiz-Lara et al. 2018; Chilingarian et al. 2019). Measuring stellar kinematics of red UDGs is an even more taxing endeavour. van Dokkum et al.(2016) used the Deep

Imaging Multi-Object Spectrograph on Keck II to study the stellar dynamics of DF44 for a total integration time of 33.5 hours. Alternative approaches such as SBF also have their pitfalls (see above and Bennet et al. 2020). With hundreds of new satellite LSB candidates and potentially thousands of UDG candidates to study, it is an infeasible task to conduct follow-up studies with typical optical spectroscopic follow-up.

It is possible to exploit the neutral atomic hydrogen (HI) gas in these targets to obtain distance estimates through the redshift of the HI spectral line. Spekkens and Karunakaran

(2018) demonstrated this to be the case by detecting the HI reservoirs of the blue UDG candidates discovered by Román and Trujillo(2017) as an alternative to optical spectroscopy.

A mere 5 hours of observing time was used to observe 5 blue UDG candidates with the Robert

C. Byrd Green Bank Telescope (GBT). This suggests that following up LSB and UDG candidates in HI with a single-dish radio telescope is an eﬃcient alternative for obtaining distance and dynamical mass estimates when an HI reservoir is present. Even if they are not detected in HI, valuable information about their gas properties can still be obtained as the

MHI gas richness, MStellar , is a distance-independent quantity. This thesis heavily relies on the facilities of the Green Bank Observatory (GBO) and its primary telescope the GBT. The GBT is a single-dish radio telescope with a 100 110 meter × parabolic dish with an oﬀset secondary. Not only is this the world’s largest fully steerable

telescope, it is also one of the most sensitive and has a well-characterized beam response

pattern owing to its oﬀset design. The full-width at half maximum of the GBT is 90 and while

this is a large resolution element, the steep fall oﬀ of the beam response pattern from the 1.3. THIS WORK 24

center of its pointing allows for the accurate identiﬁcation and measurement of our target of interest’s HI reservoirs with minimal confusion of other sources (Spekkens et al., 2013). The observations presented here use the VErsatile GBT Astronomical Spectrometer (VEGAS) as the backend for the L-band receiver. A crucial advantage of these observations are their wide bandwidth. The resulting spectra are sensitive to velocities up to 14000 km s−1 or ∼ 200 Mpc. ∼

Figure 1.5: The Robert C. Byrd Green Bank Telescope (GBT). The oﬀset secondary and the receiver cabin are clearly visible at the top of the support structure which seemingly hovers over the 100 110 m2 parabolic dish below. Image courtesy of NRAO/AUI ×

1.3 This Work

In this thesis, I will present my eﬀorts to further our understanding of satellite dwarf galaxies around Milky Way-like systems and characterize the properties of UDGs in order to place 1.3. THIS WORK 25

constraints on their formation mechanisms. Chapter2 presents a study which uses GBT observations of a magnitude-selected sample of LSB satellite candidates around M101 to conﬁrm their association with it and explore the role of environment. Chapter3 presents an investigation into the star-forming properties of satellite galaxies from the SAGA survey and a comparison to comparably sized satellite samples from two modern simulation suites, APOSTLE and Auriga. The second half of this thesis transitions to investigations of UDGs. Chapter4 presents the ﬁrst results from the GBT HI follow-up survey of UDGs.

The preliminary results from the on-going analysis of the full GBT HI follow-up survey is presented in Chapter5. Finally, Chapter6 contains a summary of this work, along with a brief discussion of future work and my conclusions. Chapter 2

Neutral Hydrogen Observations of Low Surface Brightness Galaxies around M101 and NGC 5485

Statement of Co-Authorship

This chapter consists of a version of a publication in the Astronomical Journal (A. Karunakaran et al 2020 AJ 159 37, ADS link). The work presented in this publication stems from my

MSc thesis, however, several advancements have been made. In particular, the inclusion of distance estimates from Hubble Space Telescope data analysed by our collaborators (Paul

Bennet and David Sand) and surface brightness fluctuations in the literature afforded a significant change in the interpretation beyond that presented in my MSc thesis.

As the ﬁrst author of this paper, I carried out all of the GBT observations and the subsequent reduction and analysis. I also wrote all of the text in the manuscript with Dr. K.

Spekkens providing signiﬁcant comments and edits, while the remaining co-authors provided general comments to help improve the quality of the text and the interpretation of the results.

26 2.1. ABSTRACT 27

2.1 Abstract

We present atomic hydrogen (HI) observations using the Robert C. Byrd Green Bank Tele- scope along the lines-of-sight to 27 low surface brightness (LSB) dwarf galaxy candidates discovered in optical searches around M101. We detect HI reservoirs in 5 targets and place stringent upper limits on the remaining 22, implying that they are gas poor. The distances to our HI detections range from 7 Mpc –150 Mpc, demonstrating the utility of wide-bandpass

HI observations as a follow-up tool. The systemic velocities of 3 detections are consistent with that of the NGC 5485 group behind M101, and we suggest that our 15 non-detections with lower distance limits from the optical are associated with and have been stripped by that group. We ﬁnd that the gas richnesses of conﬁrmed M101 satellites are broadly consistent with those of the Milky Way satellites, as well as with those of satellites around other hosts of comparable mass, when survey completeness is taken into account. This suggests that satellite quenching and gas stripping proceeds similarly around halos of similar mass, in line with theoretical expectations.

2.2 Introduction

Recent improvements in astronomical instrumentation (e.g. Abraham and van Dokkum,

2014; Aihara et al., 2018) and novel image searching algorithms (e.g. Bennet et al., 2017b;

Müller et al., 2018a; Zaritsky et al., 2019) have revitalized studies of the low surface brightness (LSB) universe. Many of these LSB features detected resemble dwarf galaxies (Merritt et al., 2014), while some have extreme properties relative to the high surface brightness galaxy population (e.g. van Dokkum et al., 2015a).

Wide-ﬁeld LSB searches for faint companions of nearby galaxies allow comparisons with

Local Group satellite populations, and there have been concerted eﬀorts to obtain a census of the satellite populations around such hosts (Chiboucas et al., 2009, 2013b; Karachentsev 2.2. INTRODUCTION 28

et al., 2014b; Merritt et al., 2014; Karachentsev et al., 2015; Crnojević et al., 2016b; Javan- mardi et al., 2016; Bennet et al., 2017b; Müller et al., 2017; Smercina et al., 2018; Crnojević et al., 2019). One focus is to measure the satellite populations of Milky Way-like hosts

(Javanmardi et al., 2016; Geha et al., 2017a) in order to constrain cosmic variance among halos of similar mass (Moster et al., 2011; Fielder et al., 2019). In this context, the Satellites

Around Galactic Analogues survey (SAGA, Geha et al., 2017a) has revealed a population of star-forming satellites which project within the virial radii of Milky Way-like hosts. The on-going star formation in the bulk of the SAGA detections stands in contrast to the largely quiescent, gas-poor satellite population of the Milky Way itself (e.g. McConnachie, 2012b;

Spekkens et al., 2014). This raises the possibility that satellite gas stripping and star formation quenching proceed diﬀerently around halos of similar mass, providing an important new observational constraint on the underlying physics (e.g. Wetzel et al., 2015b; Fillingham et al., 2018b; Simpson et al., 2018b; Garrison-Kimmel et al., 2019b).

−1 The nearby Milky Way-like spiral M101 (DM101 = 7.0 Mpc, VM101 = 241 km s adopted

in this work; Lee and Jang, 2012; Tikhonov et al., 2015; Mihos et al., 2013) has been the

target of several LSB searches aiming to characterize its satellite populations (Karachentsev

et al., 2014b; Merritt et al., 2014; Karachentsev et al., 2015; Javanmardi et al., 2016; Bennet

et al., 2017b; Müller et al., 2017). Many of the detected LSB objects have morphologies con-

sistent with satellites of that host, although their asymmetric sky distribution is challenging

to explain if they are all bona-ﬁde companions (Bennet et al., 2017b).

This tension is alleviated somewhat by the presence of a background group in the vicin-

−1 ity of M101 (DBG 27 Mpc,VBG 1961 km s ; Merritt et al., 2016; Tully, 2015), which ∼ ∼ includes the massive ellipticals NGC 5485 and NGC 5473 among its 25 members (Saulder ∼ et al., 2016; Karachentsev and Makarova, 2019). Indeed, follow-up observations to charac-

terize the properties of the LSB dwarf candidates around M101 suggest that many lie at the

distance of the NGC 5485 group (Merritt et al., 2016; Bennet et al., 2019b; Carlsten et al.,

2019). The distances to many others remain unconstrained, and there is a need for additional 2.3. SAMPLE SELECTION 29

follow-up observations to constrain both their locations and their physical properties.

Deep searches for the atomic hydrogen (HI) reservoirs of LSB detections are a powerful tool for constraining their physical properties (Spekkens and Karunakaran, 2018): detections provide estimates of distance and HI mass, while non-detections yield upper limits on the gas richnesses of objects along the line-of-sight (LOS). Since the HI content of ﬁeld dwarfs

(Huang et al., 2012a; Bradford et al., 2015) as well as the environmental dependence of that content (Grcevich and Putman, 2009; Spekkens et al., 2014; Brown et al., 2017) are well-characterized in the local universe, HI follow-up observations provide a mechanism for constraining the physical properties of the LSB dwarf candidates and the impact of their environments on those properties.

We present HI follow-up observations along the LOS to 27 LSB dwarf candidates in the

M101 region using the Robert C. Byrd Green Bank Telescope (GBT). We aim to characterize their gas properties and to use that information to place the M101 and NGC 5485 systems into context with other Milky Way-like hosts and nearby galaxy groups.

The structure of this paper is as follows. In Section 2.3, we describe our HI target selection. We outline our observations and data reduction procedure in Section 2.4. In

Section 2.5, we present the properties of our detections as well as upper limits on the gas content of our non-detections. In Section 2.6, we discuss the implications of these ﬁndings for the membership of the M101 and NGC 5485 group systems, and for the inﬂuence of these hosts on the gas content of the satellites. We summarize in Section 2.7.

2.3 Sample Selection

The M101 satellite system consists of 4 previously known, more luminous satellites (MV <

14: NGC 5474, NGC 5477, Holm IV, and DDO 194; Tikhonov et al., 2015). In their HI − mapping study of the M101 region, Mihos et al.(2012) detected two low-mass HI features.

We do not consider these in our sample selection: one does not have an optical counterpart, 2.3. SAMPLE SELECTION 30

while the other has a recessional velocity that is not consistent with M101. Recent optical

LSB searches, described in the previous section, have identiﬁed fainter satellite candidates

(5 of which have been recently conﬁrmed with MV < 8.2: DF-1, DF-2, DF-3, DwA, Dw9; − Danieli et al., 2017; Bennet et al., 2019b, see below). As described below, we select our

sample of HI targets from M101 satellite candidates identiﬁed in those LSB searches.

Merritt et al.(2014, hereafter M14) carried out one of the ﬁrst modern LSB searches

around M101 using the Dragonﬂy Telephoto Array, discovering 7 dwarf candidates in a

9 deg2 ﬁeld. In their survey around nearby spirals, Karachentsev et al.(2015, hereafter ∼ K15) discovered an additional 4 dwarf candidates in the region around M101, one of which

(DwA/DGSAT1) was separately discovered by Javanmardi et al.(2016). LSB searches were

also conducted using extant data. Bennet et al.(2017b, hereafter B17) conﬁrmed the M14

and K15 detections and discovered 38 new LSB dwarf candidates in a 9 deg2 ﬁeld around ∼ M101 in CFHT Legacy Survey data using a semi-automated algorithm tuned to reveal LSB

features. While the preceding searches focused in the immediate region around M101, Müller

et al.(2017, hereafter M17) searched through SDSS data for new LSB dwarf candidates over

330 deg2 in the broader M101 group complex to reveal 6 new candidates beyond the M101

virial radius. In combination, these 4 studies revealed a total of 55 unique LSB dwarf

candidates in the region around M101.

We select all LSB dwarf candidates from the combined samples of M14, K15, B17, and

M17 with apparent magnitudes mV < 19.5 mag for HI follow-up with the GBT. The optical

properties of the resulting 27 targets are given in Table 2.1. Since we determine observing

times from a gas richness scaling relation for local dwarfs (Bradford et al., 2015, see Section

2.4), our adopted mV threshold implies a follow-up time for each target of a few hours at

most (see column 10 of Table 2.1). Column 7 of Table 2.1 gives the references for the optical

properties that we adopt in this study.

A variety of studies have constrained distances to the M101 LSB dwarf candidates since

their discovery, and columns 8 and 9 of Table 2.1 list the value that we adopt for our 2.4. OBSERVATIONS AND DATA REDUCTION 31

follow-up targets. HST campaigns reported by Merritt et al.(2016, hereafter M16), Danieli et al.(2017, hereafter D17) and Bennet et al.(2019b, hereafter B19) have either conﬁrmed a dwarf candidate’s association with M101 from Tip of the Red Giant Branch (TRGB) distances from resolved star colour-magnitude diagrams, or reported lower distance limits derived from the lack of resolved stars. Distance constraints are also reported by Carlsten et al.(2019, hereafter C19) via a surface brightness ﬂuctuation (SBF) technique. We adopt distances or lower limits from M16, D17 or B19 when available, and otherwise we adopt the

C19 estimates. Table 2.1 shows that of the 27 M101 dwarf candidates that we targeted 4 have distance estimates consistent with M101, 18 have distance lower limits that place them in the M101 background, and the remaining 5 have no prior distance information.

We note that although a combination of stellar mass and colour is a more accurate predictor of gas content than stellar mass alone for the high surface brightness galaxy population

(e.g. Catinella et al., 2012; Brown et al., 2015), we do not use g r as a selection criterion − for our HI follow-up sample. Instead, we investigate the utility of the available colours for

LSB dwarf candidates as a predictor of gas richness in Section 2.6.

2.4 Observations and Data Reduction

We performed 63 hours of observations between 2016 August and 2019 August using the

GBT to determine the HI content along the LOS to the 27 LSB dwarf candidates in Table

2.1 (programs AGBT16B-046, AGBT17A-188, and AGBT17B-235). We used the L-band receiver and the Versatile GBT Astronomical Spectrometer (VEGAS) with a bandpass of

100 MHz and spectral resolution of 3.1 kHz, dumping data every 5 seconds to build up the requisite integration time. This wide bandpass allows for the HI spectral line to be detected at heliocentric velocities up to 14000 km s−1. This observational setup was used for all ∼ but one candidate, DF-3, which had a distance estimate that is consistent with that of

M101 (D17) at the time it was observed. For this target, we used a narrower bandpass, 2.4. OBSERVATIONS AND DATA REDUCTION 32

11.72 MHz, and a higher spectral resolution, 0.4 kHz, centered at the heliocentric velocity

−1 of M101, VM101 = 241 km s .

Our observations spent an equivalent amount of time on the target of interest (i.e. the

“ON") and on a set of reference positions (that collectively constitute the “OFF"). We estimate the integration times for our targets using mV calculated from mg and g r − M following the relations in Jester et al.(2005) to reach a gas richness of MHI 1 ( 0.5 LV ∼ L ∼ 9 dex below the Bradford et al. 2015 scaling relations at LV . 10 L ) with S/N = 5 in a −1 single 25 km s channel. Gas richness is a distance-independent quantity since both MHI and LV scale with distance squared. Therefore, a single spectrum allows us to meaningfully search for an HI reservoir in our targets anywhere within the wide bandpass.

The data were reduced using the standard GBTIDL1 procedure getps. Before smoothing the raw spectra to our desired resolutions, we ﬁrst removed narrow-band and broadband radio frequency interference (RFI). The latter is primarily present as a strong, intermittent

GPS signal at ν 1.38 GHz that is best removed by ﬂagging the entire 5-second data ∼ dumps in which the RFI is present. On average, we ﬂagged 20% of the data to remove this feature. Narrow-band RFI presents itself throughout the entire spectrum as spurious, strong signals that span a few raw spectral channels. As scans and their constituent integrations are co-added and eventually smoothed, these narrow-band features may resemble the expected

HI proﬁle of a dwarf if not excised. To remove narrow-band RFI, we searched through all channels of each integration for signals that exceed 5 times the median absolute deviation.

This threshold value was found to be the most eﬀective at identifying RFI spikes while avoiding noise ﬂuctuations. The values in these channels were then replaced with the median of the 200 surrounding channels.

The calibrated, RFI-excised spectra were smoothed to multiple resolutions from 5 − 50 km s−1 and examined by eye to search for statistically signiﬁcant emission. A represen-

tative RMS noise for each spectrum at ∆V = 25 km s−1 resolution is given in column 11 of

1http://gbtidl.nrao.edu/ 2.4. OBSERVATIONS AND DATA REDUCTION 33

Table 2.1. We detect HI emission along the LOS to 5 targets (column 12 of Table 2.1). Their spectra are shown in Figure 2.1 at ∆V given in Table 2.2, which lists all other properties derived from these HI detections. We find no emission associated with the 22 remaining targets. These spectra, with velocity resolution, ∆V = 25 km s−1, are shown in Figure 2.2, and the corresponding upper limits on HI mass and gas richness are in Table 2.3. 2.4. OBSERVATIONS AND DATA REDUCTION 34 HI 25 σ Ref Int. Time B19 2.5 0.40 B19 2.9 0.42 Y B19 5.5 0.21 B19 1.9 0.49 B19 1.7 0.39 B19 3.6 0.25 B19B19B19B19 5.7B19 1.6 4.4 1.6 0.22 3.6 0.38 0.29 0.34 0.29 C19 2.0 0.47 C19C19 1.0C19 1.0 3.3 0.61 0.76 0.62 C19 5.4 0.35 C19 0.2 0.52 Y C19 0.7 0.52 D17D17 1.4 4.5 0.49 0.23 D17 1.3 0.39 M16 2.7 0.27 Y 9 4 4 4 35 30 27 26 35 21 25 27 . . . . 3 8 6 1 2 5 1 1 1 1 1 1 1 1 1 ...... 0 ...... 2 2 0 0 0 0 . +3 − +3 − 9 opt +0 − +0 − +0 − +0 − 26 10 12 15 15 15 15 15 10 17 15 15 15 15 15 6 2 . . D > 37 87 52 83 > > > > > > > > > > > > > > > . . . . 18 11 6 6 6 6 Ref B17, B17 B17, B17 B17, B17 B17, B17 B17, B17 B17, B17B17, B17 - - 3.4 0.29 Y B17, B17 B17, B17 B17, B17 B17, B17 B17, B17 B17, B17 K15, B17 K15, B17 K15, B17 K15, B17 M16, B17 M16, B17 M16, B17 M16, B17 M17, M17M17, M17 - - - - 0.2 0.5 2.35 0.79 Y M17, M17 M17, M17 M17, M17 - - 0.8 0.66 M17, M17 - - 0.3 0.85 25 3 3 1 3 3 3 3 29 68 6 6 23 21 16 10 23 23 20 54 6 47 42 33 86 37 ...... 20 ...... 1 1 1 1 1 1 0 0 . 0 0 0 2 2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± eff ± 6 0 0 8 9 . . . . 70 44 08 91 35 97 59 35 92 7 . 59 33 08 91 56 95 16 11 88 64 34 70 ...... col.(12): Flag indicating whether or not we detect HI emission associated with dwarf 1 − 14 20 . 42 28 4242 7 11 42 8 42 7 42 8 22 6 28 7 14 9 22 7 14 6 1422 10 14 4 4 22 7 1414 8 4 1414 5 5 141414 5 3 4 1414 13 10 28 10 14 10 ...... 0 r r 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ± − ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± g 10 . 67 51 71 23 56 70 30 50 20 60 30 37 20 10 40 60 50 70 50 60 60 60 60 20 50 80 = 25 km s ...... 0 V − ∆ 111 0 30 0 0 0 11 0 0 3030 0 0 111 0 1 1 1 0 0 0 30 0 30 0 30 0 11 0 1 0 2 0 1 0 1 0 2 0 1 0 1 0 1 0 2 0 ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 g ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± Table 2.1: Target LSB Dwarf Candidate Properties m 1 0 8 2 4 9 1 8 4 4 4 8 3 8 5 8 5 5 8 1 2 ...... 54 76 01 46 06 46 ...... colours. col.(6): Effective radius of LSB dwarf candidate. col.(7): First reference lists origin of candidate name 20 20 19 19 20 19 19 20 20 19 20 19 21 19 20 20 19 19 20 20 19 19 18 15 18 18 19 r − g H:M:S D:M:S (mag) (mag) (arcsec) (Mpc) (hours) (mJy) Det? (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Dw6Dw7Dw8 14:02:20.1 14:07:21.0 55:39:17 14:04:24.9 55:03:51 55:06:13 Dw3 14:08:45.8 55:17:14 Dw4 14:13:01.7 55:11:16 Dw5 14:04:13.0 55:43:34 DwB 14:08:43.1 55:09:57 DwC 14:05:18.0 54:53:56 DF-1DF-2DF-3 14:03:45.0 14:08:37.5 53:56:40 14:03:05.7 54:19:31 53:36:56 DF-5 14:04:28.1 55:37:00 DwD 14:04:24.6 53:16:19 Dw38 14:01:17.6 54:21:14 Dw32Dw33 14:07:46.4 14:08:33.8 54:15:26 55:26:49 Dw26Dw31 14:08:50.4 14:07:41.7 53:27:24 54:35:18 Dw13Dw19 14:08:01.2 14:10:20.1 54:22:30 54:45:50 Name RA Dec DwA* 14:06:49.8 53:44:29 band apparent magnitudes and dw1343+58 13:43:07 58:13:40 dw1355+51dw1408+56 13:55:11 14:08:41 51:54:29 56:55:38 dw1412+56 14:12:11 56:08:31 dw1416+57 14:16:59 57:54:39 dw1446+58 14:47:00 58:34:04 − candidate. References:Danieli M14 et = al. ( 2017 ),Merritt M17LOS et = corresponds al. ( 2014 ),Müller to K15 et position = al. ( 2017 ), from KarachentsevJavanmardi B19 et et = al. ( 2016 ); al. ( 2015 ),Bennet given M16 et the = al. ( 2019b ), 9.1Merritt C19 arcminute = et GBTCarlsten beam, al. ( 2016 ), et an B17 offset al. ( 2019 ). = of *Bennet a - et few Target al. ( 2017b ), arcseconds initially is D17 reported negligible. = by K15 but GBT col.(1): Adopted LSBg dwarf candidateand name. position in cols.(2)col.(10): cols.(1)-(3), and second Total (3): effective lists GBTRMS source J2000 integration noise of position time, of photometric including of the measurements the spectrum optical in ON+OFF at centroid, cols. positions a which and velocity (4)-(6). subtracting resolution corresponds col.(8): any of to time Adopted our lost distance due GBT constraint to LOS. from RFI cols.(4) reference flagging. in and col.(11): col.(9). (5): Representative 2.5. RESULTS 35

2.5 Results

2.5.1 HI Detections

We detect HI along the LOS to 5 of the LSB dwarf candidates in our sample, and their spectra are shown in Figure 2.1. At our observing frequency of 1.4 GHz, the full-width ∼ half maximum of the GBT beam, 9.10, encompasses the entire stellar component of our ∼ dwarf targets. The GBT beam response is well understood down to 30dB (e.g. Spekkens ≈ − et al., 2013), and we can assess the extent to which gas-rich sources near the LOS contaminate our spectra. For all of our targets, we search through NED2 and SDSS imaging catalogs

(Aguado et al., 2019) for objects within 300 of the LOS that may present themselves as gas-rich interlopers in our spectra (the spectrum of one interloper, ASK 301585, is visible in panel (b) of Fig. 2.1). We ﬁnd no such interlopers for our HI detections, and conclude that they are the HI counterparts to the LSB dwarf candidates along the LOS. The HI properties that we derive for the detections are given in Table 2.2.

We first derive distance-independent quantities, systemic velocities (Vsys) and velocity widths (W50), following the methods of Springob et al.(2005). We fit first-order polynomials to both edges of each HI profile in Figure 2.1 between 15% and 85% of the peak flux value.

To determine Vsys and W50, we first determine the velocities corresponding to the 50% flux level for each polynomial fit. The mean of these two values provides Vsys, while the difference provides W50 which is also corrected for instrumental and cosmological redshift broadening to produce W50,c. We do not correct for the effect of inclination on W50. The uncertainties on the instrumental broadening correction, which we take to be 50%, dominate those on

Vsys and W50 (see Springob et al. 2005).

We estimate kinematic distances, DHI , for DF-5, DwB, dw1408+56, and Dw26 using

−1 −1 Vsys of our detections and assuming H0 = 70 km s Mpc . The value of DHI for DF-5 is

2The NASA/IPAC Extragalactic Database (NED) is operated by the Jet Propulsion Laboratory, Cali- fornia Institute of Technology, under contract with the National Aeronautics and Space Administration. 2.5. RESULTS 36

consistent with the lower limits on Dopt from M16 and C19. The values of Vsys that we derive for DwB and dw1408+56 are similar to those of the background group containing NGC 5485, implying an association. The SBF lower limit on Dopt from C19 for DwB is consistent with our distance estimate, while their Dopt estimate for dw1408+56 is not. C19 note that their distance for this latter object may be unreliable due to its unusual morphology, and they conclude that it likely lies in the background of M101 as we ﬁnd here. Our measurement of

DHI for Dw26 is the ﬁrst distance estimate for this system (see also B17). For the purposes of this work, we assign a distance of DHI = DM101 = 7.0 Mpc to dw1343+58 because of its similar recessional velocity to the broader M101 group complex.

We calculate the HI ﬂux, SHI = SδV , by integrating over the line proﬁle. HI masses, ´ MHI , are determined using the standard equation for an optically thin gas (Haynes and

Giovanelli, 1984): 5 2 MHI = 2.356 10 (DHI ) SHI M , (2.1) ×

−1 where DHI is in Mpc and SHI is in Jy km s . Uncertainties are determined following the

methods of Springob et al.(2005). Using mg and g r from Table 2.1, the relations of Jester − et al.(2005), and DHI we estimate the V band luminosities, LV , and the gas richnesses, − MHI /LV . These values are tabulated in Table 2.2. 2.5. RESULTS 37

20 (a) dw1343+58 (b)DF-5 5 (c) DwB 16 ASK 1 301585 12 3 8 0 4 1 0 4 1 1 − −100 300 500 700 − 600 1100 1600 2100 1200 1700 2200 2700

(d)dw1408+56 1.4 (e)Dw26 6 Flux (mJy) 0.7

2 0.0

2 0.7 − − 1200 1700 2200 2700 10250 10750 11250 11750 VHelio(km/s)

Figure 2.1: HI detections along the LOS to LSB dwarf candidates in the region around M101. Target names are in the top-left corner of each panel and the spectral resolutions ∆V of the plotted spectra are in Table 2.2. The spectrum in (a) is oﬀ-center as Milky Way emission dominates at lower velocities. The dash-dotted feature in (b) is the HI emission associated with ASK 301585 (see text). The vertical, black lines below the spectra indicate the Vsys of M101 for (a) and the NGC 5485 group for (b), (c), and (d). The derived properties of the HI detections are provided in Table 2.2. 2.5. RESULTS 38 , 1 − in 08 5 4 3 . . . . ) 2 HI ) 0 1 0 1 Mpc D

V 1 HI ± ± ± ± ± L L − M M 4 3 2 ( ( . . . in col.(2). 23 . V ∆ = 70 km s ) ]) 09 0 18 3 05 14 10 1 12 4 0 . . . . .

in col.(6) and 0 0 0 0 0 HI H M ± ± ± ± ± M HI S and 85 79 81 82 94 . . . . . sys V 04 8 12 6 08 6 04 7 12 7 ) log( ]) (log[ . . . . .

V 0 0 0 0 0 L L ± ± ± ± ± 48 26 65 70 32 log( . . . . . 7 6 6 7 8 ∗ 0 HI . 156 20.3 27.3 27.2 D ) (Mpc) (log[ 1 12 7 02 04 08 04 . . . . . − 0 0 0 0 0 HI ± ± ± ± ± S km s 62 06 37 38 15 . . . . . ) (Jy 1 1 0 6 0 5 0 4 0 16 0 ,c − ± ± ± ± 50 ± W km s )( 4 36 4 69 2 55 12 81 1 37 1 − ± ± ± ± ± sys V km s 195 from col.(7). col.(9): Logarithm of HI mass calculated from Eq.(1) using 1424 1913 1904 10972 HI D col.(7). col.(10): HI-mass to V-band luminosity ratio (gas richness). V ∆ Table 2.2: Properties of LSB Dwarf Candidates with HI detections ) (mJy) ( 1 − V σ ∆ in Table 2.1 and km s r ( − g and g (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) m DF5 20 0.3 DwB 15 0.4 Dw26 30 0.2 Name col.(4): Heliocentric systemic velocity. col.(5): Velocity width of the HI detection, corrected for cosmological redshift and instrumental col.(2): Velocity resolution of spectrum used to compute HI properties (see Figure 2.1 ). col.(3): RMS noise of spectrum at dw1343+58 10 2.3 dw1408+56 15 1.0 using except for dw1343+58(*) for which the M101 distance is adopted; see text for explanation. col.(8): Logarithm of V-band luminosity calculated broadening. col.(6): Integrated HI ﬂux. col.(7): Kinematic distance estimated using the Hubble-Lemaître Law, 2.5. RESULTS 39

2.5.2 HI Non-detections

We ﬁnd no HI signal along the LOS to the remaining 22 LSB dwarf candidates in our sample that we can attribute to these objects. We show their spectra in Figure 2.2, normalized by the RMS values σ25 listed in Table 2.1. We search through NED and SDSS imaging catalogs for potential interlopers within 300 of our targets. In Figure 2.2, HI emission from nearby objects in the spectra are shown by green dashed lines with object names given above them.

We note that the features are positive if the corresponding object is located in the ON, and negative if they are located in an OFF.

−1 We use σ25 to place single, 25 km s -channel 5σ upper limits on the HI mass of the LSB

lim dwarf candidate non-detections, MHI , using a modiﬁed version of Eq.(1),

lim 7 2 M = 2.945 10 D (σ25)M , (2.2) HI ×

−1 where σ25 is in km s and D is in Mpc. For objects with Dopt estimates or lower limits in

lim Table 2.1, we use that value as D in Eq.(2). For objects lacking a Dopt we compute MHI by

setting D to the M101 distance, DM101 = 7.0 Mpc. We also estimate the HI mass of these

lim objects if they are located at the distance of the NGC 5485 group, (MHI )BG, by setting lim lim D = DBG = 27 Mpc in Eq.(2). Both MHI and (MHI )BG are given in Table 2.3, along with lim the corresponding (distance-independent) upper limit MHI /LV on gas richness.

2.5.3 Optical and Gas Properties

With the derived HI properties for our sample in-hand we can make comparisons to their op-

lim tical properties. Figure 2.3 plots MHI /LV for detections and (MHI )/LV for non-detections

as a function of LV . We separate our sample into four groups: objects with HI detections

(blue stars), non-detections with conﬁrmed distances, Dopt (red circles), non-detections with

lower limits on Dopt (orange squares), and non-detections with no distance measure (ma-

genta diamonds). Since MHI /LV is distance independent, the location of the points along 2.5. RESULTS 40

DF1 150 DF2

DwA

DwC

100 DwD

VGS38 Dw3

Dw4

50 Dw5 ) Dw6 25 σ ASK 302203 NGC 5486 NSA 053178 Dw7

Dw8 0 Dw13

Dw19

Normalized Flux (S/ Dw31 50 − Dw32

Dw33

Dw38

100 dw1355+51 − dw1412+56

NSA 045249 IC 0995 dw1416+57

150 dw1446+58 − 1000 2000 3000 4000 5000 VHelio(km/s)

Figure 2.2: Spectra along the LOS to LSB dwarf candidates for which no associated HI emission was found, at a resolution of ∆V = 25 km s−1 and normalized by the RMS values listed in Table 2.1. The spectra are presented in the same order as in Table 2.3 but with DF- 3 omitted because the spectral setup for that target was different; see text for explanation. Target names are shown above their respective spectrum at the right side of the figure. The vertical offset between the spectra is arbitrary to better display them. Parts of the spectra where an interloping HI signal of a nearby gas-rich galaxy is present in the ON or OFF scans are shown as green dashed lines. The vertical orange lines indicate the velocity region within which HI emission from M101 itself may be present. Two spectra (DwC and Dw38) have had M101 emission removed and replaced with black dotted lines for aesthetic purposes. 2.5. RESULTS 41

the y-axis is known for all of the points. For non-detections with lower limits on Dopt or no available estimates, we place symbols at both the Dopt lower limit and DBG (ﬁlled and empty orange squares) or DM101 and DBG (ﬁlled and empty magenta diamonds) along the x-axis and connect the two with a horizontal dashed line. For clarity, we do not include

HI detection Dw26, with log(LV ; L ) = 8.3. Considering this object in addition to those plotted in Fig. 2.3, over half of the HI detections correspond to objects with higher LV than the non-detections even if the latter are placed at the distance of the NGC 5485 group.

Figure 2.4 shows gas richness versus g r for the sample, separated into HI detections − (blue stars) and HI non-detections (red circles with downward arrows). The horizontal dotted line indicates MHI /LV = 1M /L which is typical for dwarf galaxies (e.g. Huang et al., 2012a; Bradford et al., 2015). Both our detections and non-detections have a wide range of colours. We discuss this further in the following section. 2.6. DISCUSSION 42

Table 2.3: HI Upper Limits for Non-detections lim lim lim Name log(MHI ) log(MHI )BG (MHI /LV ) (log[M ]) (log[M ]) (M /L ) (1) (2) (3) (4) DF-1 5.77 - 0.70 DF-2 5.51 - 0.49 DF-3 5.69 - 1.86 DwA 5.75 - 0.68 DwC 6.16 6.67 0.81 DwD 6.52 7.02 0.91 Dw3 6.43 6.93 0.87 Dw4 6.22 6.73 0.84 Dw5 6.18 6.68 0.88 Dw6 6.41 6.92 0.92 Dw7 6.29 6.80 1.12 Dw8 6.36 6.87 0.75 Dw13 6.30 6.81 1.08 Dw19 6.04 6.88 1.27 Dw31 6.33 7.12 0.89 Dw32 6.55 7.22 0.94 Dw33 6.69 7.00 0.99 Dw38 7.11 7.13 2.01 dw1355+51 6.06 7.23 0.61 dw1412+56 6.58 7.05 0.74 dw1416+57 5.98 7.15 0.85 dw1446+58 6.09 7.26 0.53

col.(2): 5σ upper limit on MHI calculated from Eq.(2) using D = Dopt and σ25 from Table 2.1. col.(3): Same as col.(2) but instead D = DBG = 27 Mpc of the NGC 5485 group for systems without Dopt estimates or with lower limits. An exception is Dw33 which has a Dopt estimate, however, we suggest it may reside in the background group (see Section 2.6.2).col.(4): Upper limit on the gas richness (which is distance independent).

2.6 Discussion

With the gas properties of our HI detections and non-detections in hand, we consider their most likely physical associations along with their implications for the gas richnesses of satellite systems on both the galaxy and group scales. 2.6. DISCUSSION 43

1.25

1.00

0.75 )

⊙ 0.50 /L ⊙ M

; 0.25 V /L

HI 0.00 M log( 0.25 −

0.50 HI detections − HI non-detections, conﬁrmed optical distances HI non-detections, optical distance lower limits 0.75 − HI non-detections, unconstrained distances

5.5 6.0 6.5 7.0 7.5 log(LV /L ) ⊙

lim Figure 2.3: MHI /LV (blue stars) and MHI /LV (red, orange, and magenta downward arrows) as a function of LV for LSB dwarf candidates in our sample. Among HI non-detections, red circles show objects with measured Dopt, orange squares show objects with lower limits on Dopt, and magenta diamonds show objects with no distance constraints. We place orange squares at LV corresponding to the Dopt lower limits (ﬁlled symbol and rightward arrows) and DBG (empty symbol) of the NGC 5485 group for the corresponding objects, and magenta diamonds at LV corresponding to DM101 (ﬁlled) and DBG (empty). Pairs of symbols for the same objects are connected by horizontal dashed lines. We omit one gas-rich target for aesthetic purposes, Dw26 (log(MHI /LV ) = 0.62, log(LV ) = 8.32).

2.6.1 Associations of HI Detections

The systemic velocities of our 5 detections allow us to determine whether they are associated with M101 or the NGC 5485 group in the background (see Section 2.2), or whether they are at some other location along the LOS.

−1 The systemic velocity of the HI detection of dw1343+58, Vsys = 195 km s (Table 2.2), 2.6. DISCUSSION 44

1.25 Gas-poor Gas-rich 1.00

0.75 )

⊙ 0.50 /L ⊙ M

; 0.25 V /L

HI 0.00 M log( 0.25 −

0.50 −

0.75 −

0.2 0.0 0.2 0.4 0.6 0.8 1.0 − g r (mag) −

lim Figure 2.4: MHI /LV (blue stars) and MHI /LV (red circles with downward arrows) versus g r colour for our sample. The horizontal dotted line shows MHI /LV = 1M /L , typical − for dwarf galaxies (e.g. Huang et al., 2012a; Bradford et al., 2015). We omit one gas-poor lim outlier in colour for aesthetic purposes, Dw7 (log(M /LV ) = 0.05, g r = 1.7). HI −

conﬁrms that it is a nearby dwarf. Its projected separation from M101, dproj 580kpc, and ∼ −1 similar Vsys to M101 (VM101 = 241 km s ) suggests that it is likely a member of the M101

−1 group. The SDSS spectrum for dw1343+58, VSDSS = 165 33 km s , is consistent with the ± Vsys that we derive, although the SDSS pipeline mis-classiﬁed it as a star. Given its optical morphology, M17 suggest that dw1343+58 is a blue compact dwarf (BCD). The relatively low gas richness of dw1343+58 compared to typical star-forming dwarfs (e.g. Huang et al.,

2012a; Bradford et al., 2015), and the UV counterpart that we identify in our search of 2.6. DISCUSSION 45

archival GALEX data provide further evidence in support of this classiﬁcation (Thuan and

Martin, 1981; Gil de Paz et al., 2003).

On the other hand, DwB and dw1408+56 have Vsys consistent with that of the NGC 5485 group. Using their kinematic distances and projected separations, we estimate that the physical separations of DwB and dw1408+56 from NGC 5485 are 450 kpc and 940 ∼ ∼ kpc, respectively. Given its kinematic distance, the properties of dw1408+56 resemble those

−2 of a gas-rich ultra-diﬀuse galaxy (UDG; Reff > 1.5 kpc, µ0 > 24 mag arcsec ; van Dokkum

−2 et al., 2015a) progenitor with Reff = 1.45 0.17 kpc and µr,0 = 23.28 0.06 mag arcsec ± ± (Spekkens and Karunakaran, 2018). This ﬁnding is broadly consistent with UDG number- halo mass relation estimates (e.g. van der Burg et al., 2017), which imply that the number of UDGs in a group like that containing NGC 5485 should be of order 1.

DF-5 was suggested to be a member of the NGC 5485 group since HST imaging does not resolve any stars (M16), and our HI detection conﬁrms that it is in the background of

−1 −1 M101 with Vsys = 1424km s . While this Vsys diﬀers by 500km s from that of the ∼ group (Tully, 2015), the small projected separation between DF-5 and NGC 5485 implies that a physical association is nonetheless feasible. We return to this issue in Section 2.6.2.

Dw26 is our most distant HI detection with DHI = 156 Mpc. At such a distance, Dw26 is a star-forming galaxy well in the background of both M101 and the NGC 5485 group.

The vast range of distances that we derive from our HI detections in the M101 region demonstrates the need for follow up observations of optical LSB features to determine their physical properties and associations.

The relationship between stellar mass, colour and gas richness in the high surface brightness galaxy population is well-studied (Catinella et al., 2012; Huang et al., 2012a; Bradford et al., 2015; Brown et al., 2015): at ﬁxed stellar mass, bluer systems are more gas-rich. By contrast, Fig. 2.4 illustrates that our HI detections exhibit the same broad range in g r − as our non-detections, and therefore that the reported colours do not predict gas richness. 2.6. DISCUSSION 46

While it is possible that star formation is regulated diﬀerently in these faint systems compared to the broader galaxy population (Wheeler et al., 2015; El-Badry et al., 2016; Di

Cintio et al., 2019), we deem it more likely that the large photometric uncertainties at such low surface brightness (B17, M17), evidenced by the sizeable error bars in Fig. 2.4, preclude the use of colour as an indicator of the presence of gas. This further underscores the utility of follow-up HI observations to constrain the gas content and star formation activity in the

LSB galaxy population.

2.6.2 HI non-detections and the NGC 5485 group

The majority of our follow-up HI observations resulted in non-detections, implying that the

LSB dwarf candidates along those LOS are gas poor. Among them, 15/22 have distance constraints or lower limits from optical follow-up that place them in the background of

M101 (see Table 2.1). Because ﬁeld galaxies are almost universally star forming and HI rich

(Geha et al., 2012b; Bradford et al., 2015) whereas our non-detections are clearly gas poor, we consider the likelihood that all of our non-detections are associated with the NGC 5485 background group and have been stripped of their HI reservoirs through environmental processes (e.g. Gatto et al., 2013; Wetzel et al., 2015b; Simpson et al., 2018b).

Figure 2.5 shows all of the targets in our sample that have no known association with

M101 and project within 2 degrees of NGC 5485 (= 1 Mpc at DBG; dashed circle). These ∼ targets are either confirmed to be in the background of M101 via our HI detections (stars), are undetected in HI but have optical distance measurements or lower limits that place them beyond M101 (circles). Fig. 2.5 suggests that, if our HI non-detections are at the NGC 5485 group distance and have been stripped of their gas due to environmental effects, then the sphere of influence of that group has a radius of at least 1 Mpc. Observations show ∼ that the fraction of gas-rich or star-forming satellites depends on group mass (Brown et al.,

2017; Schaefer et al., 2019), while simulations suggest that the environmental inﬂuence of a group extends beyond its virial radius Rvir (Bahé et al., 2013; Cen et al., 2014). The 2.6. DISCUSSION 47

latter suggest that the HI non-detections in Fig. 2.5 could plausibly stem from gas stripping within 2 3 Rvir of the NGC 5485 group, which is broadly consistent with literature ∼ − estimates of its Rvir (solid circle in Fig. 2.5, Tully, 2015; Saulder et al., 2016; Karachentsev and Makarova, 2019).

Indeed, the properties of both our detections and non-detections are consistent with a scenario in which almost all are physically associated with the NGC 5485 group. In this picture, the non-detections correspond to galaxies that have had their gas stripped by the

NGC 5485 intra-group medium or by tides, as the gas in satellite galaxies is not expected to survive after ﬁrst infall (Wetzel et al., 2015b). The HI detections correspond to galaxies on the group outskirts that have not lost their gas or have yet to be processed by the group environment. NGC 5485 has stellar stream-like features which are likely due to past interactions (Karachentsev and Makarova, 2019), further supporting this scenario.

We consider each of the HI detections in turn to assess whether or not their properties are consistent with this scenario. The large projected separation between dw1408+56 and the group center as well as their correspondence in Vsys suggest that this object is indeed on the group outskirts. Similarly, the large difference in Vsys between DF-5 and the NGC 5485 group ( 500km s−1) suggests that it is on its first infall with a large peculiar velocity. By ∼ contrast, DwB projects near the group center and has a similar Vsys; if it is indeed near the group center it should be gas-poor. This can be reconciled with our scenario if DwB is separated from the NGC 5485 group along the line-of-sight. In this case, the correspondence between the NGC 5485 group Vsys and that of DwB would stem from the peculiar velocity of the latter because it is on first infall.

We note that the only HI follow-up target plotted in Fig. 2.5 that is known not to be associated3 with the NGC 5485 group is Dw26, which our HI detection places far in the

3We note that the C19 SBF distance for Dw33 places it in the background of M101 but also in the foreground of NGC 5485. However, since C19 caution that the uncertainties on their Dw33 distance measurement may be underestimated, we deem it plausible that Dw33 is in fact associated with NGC 5485. 2.6. DISCUSSION 48

background (D 156 Mpc). However, it is unlikely that a signiﬁcant number of our non- ∼ detections are at such large distances: a mechanism for stripping them in a lower density environment needs to be found. We therefore posit that the most plausible origin for the majority of our non-detections is that they belong to the NGC 5485 group.

HI non-detections with distance lower limits HI detections 57 dw1408+56 M101 NGC 5485 300 kpc

1 Mpc dw1412+56 56

Dw5 DF-5 Dw6

Dw33 Dw3 Dw4 DwB Dw7 Dw8 55 DwC

Dec (deg) Dw19

Dw31

Dw13 Dw38 Dw32

Dw26

DwD

53 215 214 213 212 211 210 209 RA (deg)

Figure 2.5: Targets in our sample that project within 2 deg (= 1 Mpc at the NGC 5485 ∼ distance; dashed circle) of NGC 5485 (solid grey square) and that have no known association with M101. Objects conﬁrmed to be in the background of M101 via our HI detections are shown as stars: ﬁlled in blue are those associated with the NGC 5485 group and the dark red star is Dw26 in the distant background. Objects that are undetected in HI but have optical distance measurements or lower limits that place them beyond M101 are shown as yellow circles. For reference, the open square shows the location of M101 in the foreground. The open grey circle centered on NGC 5485 has a radius of 300 kpc at its distance, representative of literature estimates for the virial radius of the corresponding group (Tully, 2015; Saulder et al., 2016; Karachentsev and Makarova, 2019). 2.6. DISCUSSION 49

2.6.3 Satellite Gas Richness

Our observations aﬀord comparisons between the gas content of conﬁrmed M101 satellites with that of the satellites of the Milky Way as well as with the populations around galaxies of similar mass. Figure 2.6 makes such a comparison: we plot MV as a function of projected separation, Dproj, at the host distance for all known M101 satellites (maroon symbols; Tikhonov et al., 2015; Danieli et al., 2017; Bennet et al., 2019b) down to the B17 completeness limit of MV = 7.5 (horizontal dotted line), all known Milky Way/Local Group satellites down − to that same limit (yellow symbols; 2015 update of McConnachie, 2012b), and the 27 satellites detected within the projected virial radii of 8 Milky Way mass hosts from the SAGA survey. The SAGA satellite populations are spectroscopically complete down to a limit of

Mr = 12.3 , or MV = 12.1 for a median g r = 0.4 (horizontal dash-dotted line, Geha − − − et al., 2017a). The stars denote satellites that are star forming or gas rich, and the circles denote satellites that are quiescent or gas poor. M101 satellites with HI observations from this work are enclosed by black boxes, and the vertical dashed line indicates Dproj = Rvir.

We adopt Rvir = 260 kpc for M101 (Merritt et al., 2014) and Rvir = 300 kpc for the Milky

Way and the SAGA hosts (see Geha et al. 2017a for more detail).

Fig. 2.6 illustrates that the satellites within Rvir of M101, the Milky Way or the SAGA hosts that are brighter than MV 12 are typically star forming or gas rich, as are most ' − companions beyond Rvir. By contrast, the satellites within Rvir that are fainter than MV ' 12 are typically quiescent or gas poor. This diﬀerence in gas content and star formation − activity between bright and faint satellites likely results from environmental eﬀects: the gas reservoirs of low-mass subhalos are more easily stripped, and their star formation quenched, due to their shallower potential wells relative to high-mass subhalos (Wetzel et al., 2015b;

Emerick et al., 2016; Fillingham et al., 2016b). Fig. 2.6 therefore implies that the high fraction of star-forming satellites detected within Rvir by SAGA and the low fraction of HI-

rich satellites within the Rvir of M101 and the Milky Way can be explained by diﬀerences 2.7. CONCLUSIONS 50

in survey sensitivity rather than inherent diﬀerences in the star formation activity among satellite populations. Instead we ﬁnd that, when completeness is taken into account, the gas richnesses of the satellite populations of M101, the Milky Way, and the SAGA hosts within

Rvir are broadly consistent with one another. This suggests environment has a similar eﬀect on the satellite populations of diﬀerent Milky Way-mass hosts, in line with expectations from simulations (e.g. Fillingham et al., 2018b).

2.7 Conclusions

We have presented HI follow-up observations using the GBT along the LOS to 27 optically detected LSB dwarf candidates in the M101 region (see Table 2.1). We ﬁnd the HI counterparts of 5 targets (Figure 2.1) and derive their gas properties such as HI masses and velocity widths (Table 2.2). We ﬁnd no HI emission along the LOS associated to the remaining 22

(Figure 2.2) and place stringent 5σ single, 25 km s−1-channel upper limits on their HI masses and gas richnesses (Table 2.3).

Among detections, we ﬁnd that dw1343+58 is likely a member of the M101 group while

Dw26 is far in the background with DHI 150Mpc. We ﬁnd that the remaining 3 LSB ∼ dwarfs with HI detections (DF-5, DwB, and dw1408+56) are likely members of the background NGC 5485 group. We show that detections and non-detections span a similar range in measured colour (Figure 2.4), implying that at these low surface brightnesses, measured colours are not a good indicator of gas content. These ﬁndings demonstrate the utility of

HI follow-up of optically detected LSB features to constrain their physical properties.

The optical lower distance limits and projected separations of our HI non-detections from the elliptical galaxy NGC 5485 suggest that they all could be satellites of that group, where any gas they once had has been gas stripped by environmental eﬀects (Figure 2.5).

In this context, the sphere of inﬂuence of the NGC 5485 group has a radius R 1 Mpc, ∼ in-line with observational and theoretical constraints. 2.7. CONCLUSIONS 51

Geha+ 17 20 − MW/LG M101 18 Q/Gas-poor − SF/Gas-rich 16 This Work − V

M 14 −

12 −

10 −

8 − 1.0 0.5 0.0 0.5 − − log(Dproj; Rvir)

Figure 2.6: MV as a function of projected separation, Dproj, at the host distance for all known M101 satellites (maroon symbols; Tikhonov et al., 2015; Danieli et al., 2017; Bennet et al., 2019b) brighter than MV = 7.5, all known Milky Way/Local Group satellites down − to that same limit (yellow symbols; 2015 update of McConnachie, 2012b), and satellites brighter than MV = 12.1 from the 8 spectroscopically complete hosts from the SAGA − survey (horizontal dash-dotted line, Geha et al., 2017a). Stars denote satellites that are star forming or gas rich, and circles denote satellites that are quiescent or gas poor. M101 satellites with HI observations from this work are enclosed by black boxes, and the vertical dashed line highlights Dproj = Rvir. We adopt Rvir = 260 kpc for M101 and Rvir = 300 kpc for the Milky Way and SAGA hosts.

We compare the gas richnesses of conﬁrmed M101 satellites to those of the Milky Way satellites and of similar mass hosts in the SAGA survey (Figure 2.6). Accounting for completeness, we ﬁnd general agreement between these populations: satellites within the virial radius brighter than MV 12 are broadly star-forming and gas-rich while those fainter ' − 2.7. CONCLUSIONS 52

than this threshold are broadly quiescent and gas-poor. This suggests that the eﬀect on satellite gas content is similar around hosts of similar stellar mass, in line with theoretical expectations. Chapter 3

Satellites Around Milky Way Analogs: Tension in the Number and Fraction of Quiescent Satellites Seen in Observations Versus Simulations

Statement of Co-Authorship

This chapter consists of the accepted version of a publication that is in press at the Astro- physical Journal Letters (ADS Link). I am the ﬁrst author of this paper which was written by myself and Dr. K. Spekkens. I carried out all of the GALEX UV analysis and the comparisons to the satellite properties from the APOSTLE and Auriga cosmological zoom-in simulations which were extracted by Drs. K. A. Oman and C. M. Simpson.

53 3.1. ABSTRACT 54

3.1 Abstract

We compare the star-forming properties of satellites around Milky Way (MW) analogs from the Stage II release of the Satellites Around Galactic Analogs Survey (SAGA-II) to those from the APOSTLE and Auriga cosmological zoom-in simulation suites. We use archival

GALEX UV imaging as a star-formation indicator for the SAGA-II sample and derive star- formation rates (SFRs) to compare with those from APOSTLE and Auriga. We compare our detection rates from the NUV and FUV bands to the SAGA-II Hα detections and find that they are broadly consistent with over 85% of observed satellites detected in all three tracers. We apply the same spatial selection criteria used around SAGA-II hosts to select satellites around the MW-like hosts in APOSTLE and Auriga. We find very good overall agreement in the derived SFRs for the star-forming satellites as well as the number of star- forming satellites per host in observed and simulated samples. However, the number and fraction of quenched satellites in the SAGA-II sample are significantly lower than those

8 in APOSTLE and Auriga below a stellar mass of M∗ 10 M , even when the SAGA- ∼ II incompleteness and interloper corrections are included. This discrepancy is robust with

respect to the resolution of the simulations and persists when alternative star-formation

tracers are employed. We posit that this disagreement is not readily explained by vagaries

in the observed or simulated samples considered here, suggesting a genuine discrepancy that

may inform the physics of satellite populations around MW analogs.

3.2 Introduction

Characterizing the satellite populations around Milky Way–like hosts is a key component

of understanding galaxy formation and evolution. Owing to the unrivaled depth and com-

pleteness of observations of its satellite population, the Milky Way (MW) has been the

default test-bed for simulations aiming to probe the underlying physics of dwarf galaxy

evolution. The environmental dependence of the satellite star forming fraction therein is 3.2. INTRODUCTION 55

8 well-established: with few massive (M∗ & 10 M ) exceptions, satellites are quiescent within the virial radius and star forming farther out (Grcevich and Putman, 2009; Spekkens et al.,

2014; Putman et al., 2021), implying that satellites are quenched by their hosts. Indeed, it

has been demonstrated that the observed LG satellite quenched fraction transitions from

6.5 9 100% 0% between 10 M∗/M 10 , suggesting a mass dependence to the un- ∼ − . . derlying mechanisms (Fillingham et al. 2015; Wetzel et al. 2015a, see also Slater and Bell

2014; Wheeler et al. 2014). Whether or not the star-forming properties of satellites around

MW-mass hosts beyond the Local Group (LG) are consistent with those within the LG can

provide important constraints on cosmological galaxy formation simulations.

Observationally, searches for the satellite populations around nearby MW anologs were

pioneered by Zaritsky et al.(1993, 1997) and are now being pursued by various groups (e.g.

Crnojević et al., 2016a; Javanmardi et al., 2016; Bennet et al., 2019a; Carlsten et al., 2020).

One of the most extensive of these campaigns is the ongoing Satellites Around Galactic

Analogs (SAGA, Geha et al., 2017b) survey, which aims to detect and characterize all

6.6 satellites brighter than the Leo I dwarf (M∗ 10 M ) around 100 MW-like hosts. The ∼ SAGA Stage II (SAGA-II) release presented in Mao et al.(2021, hereafter M21) shows that

the vast majority of conﬁrmed satellites within the virial radii of the 36 hosts surveyed so

far are star forming rather than quenched. While this result is commensurate with some

earlier surveys of brighter satellites across a broader host mass range (e.g. Guo et al., 2013;

Phillips et al., 2015; Davies et al., 2019), it is in strong contrast to the fainter satellites in

the LG. This discrepancy between the LG and other observed systems may have important

implications for models that aim to replicate the trends seen in the LG.

Theoretically, a variety of cosmological zoom-in simulation suites can now probe star

formation and quenching physics in satellites around MW analog hosts down to Leo I masses

(Wetzel et al., 2016; Garrison-Kimmel et al., 2019b; Akins et al., 2021). In particular, galaxy

properties in APOSTLE (Sawala et al., 2016; Fattahi et al., 2016) and Auriga (Grand et al.,

2017) are interesting to contrast given their similar resolutions and suite sizes but diﬀerent 3.3. SATELLITE SAMPLES 56

host selection, hydrodynamical schemes (SPH vs. moving-mesh), and evolutionary models.

One focus has been the comparison of the simulated satellite quenched fraction to that in the LG. The agreement between simulations is generally good, with most studies suggesting

8−9 a characteristic mass (M∗ 10 M ) below which satellites are more readily quenched by ∼ their hosts (e.g. Fillingham et al., 2016a; Simpson et al., 2018a; Akins et al., 2021; Joshi et al., 2021).

The consistency of simulated satellite populations with those in the LG combined with the stark contrast between the quenched fractions in the MW and SAGA-II strongly motivate direct comparisons between theory and other MW analogs in order to build robust models of galaxy formation. This requires selecting star-forming objects consistently across observed and simulated samples.

In this letter, we compare star-forming satellites and quenched fractions in the SAGA-II observations to those in the APOSTLE and Auriga simulations. The observations and simulations have similar host numbers, host masses, and satellite selection functions (§3.3). We use archival UV imaging and simulated star formation rates to select star-forming satellites in SAGA-II and APOSTLE/Auriga, respectively, comparing star formation rates (SFRs) to gauge consistency across samples (§3.4). We then compare quenched fractions in the observed and simulated samples (§3.5) and discuss possible explanations for the signiﬁcant discrepancies we ﬁnd (§3.6).

3.3 Satellite Samples

3.3.1 Observed sample: SAGA-II

We adopt the “complete systems" in the SAGA-II release as our observed sample, which

12 1 consists of 127 conﬁrmed satellites across 36 surveyed hosts with Mhalo (0.7 2) 10 M . ∼ − × 1Estimated in SAGA-II following Nadler et al.(2020) where virial parameters are estimated at ' 99.2 times the critical density of the Universe, ρcrit. 3.3. SATELLITE SAMPLES 57

SAGA-II hosts are selected primarily on luminosity ( 23 > MK > 24.6), are largely in − − the ﬁeld with a few that are members of LG-like pairs (see M21 for details), and are mostly star-forming galaxies. We use the SAGA-II optical properties, stellar masses and distances derived for all observed hosts and satellites, which are reproduced in Table 3.1.

As explained in detail in M21, imaging catalogs are used to build satellite candidate lists around each host, and candidates without archival redshifts are targeted spectroscopically to conﬁrm an association. Sample-wide, 80% (100%) of candidates with extinction-corrected

(designated with subscript “o") r-band absolute magnitudes Mr,o 12.3 ( 15.5) are ≤ − − M targeted spectroscopically in SAGA-II. We convert these limits to stellar masses log ∗ M ∼ M 6.6 log ∗ 7.8 using the relations in M21 and an average satellite sample color of M ∼ (g r)o 0.39. Since star-forming satellites are easier to detect than quiescent ones in Hα the − ∼ spectroscopic coverage is only indirectly related to completeness, particularly for quiescent

systems. M21 undertake detailed modelling to estimate the impact of incompleteness and

interlopers on the sample quenched fraction, which we adopt here (see §3.5).

The single-ﬁbre Hα measurements in SAGA-II provide an estimate of star formation

activity that is not amenable to direct comparisons with simulations (M21). We therefore

make use of data from the Galaxy Evolution Explorer (GALEX, Martin et al., 2005) to search

for UV emission in SAGA-II satellites to provide a homogeneous, global star formation

activity indicator for each system. In total, 119/127 SAGA-II satellites have archival NUV

and/or FUV coverage, and we select the deepest available imaging for the search (26/58

tiles have depths greater than GALEX AIS data, i.e. integration times 300 seconds; Table 3.1).

3.3.2 Simulated samples: APOSTLE and Auriga

We adopt hosts and satellites from APOSTLE (Sawala et al., 2016; Fattahi et al., 2016) and

Auriga (Grand et al., 2017) to deﬁne simulated samples. The APOSTLE suite traces the formation and evolution of LG-like environments with MW-M31 pairs (selected by halo mass, 3.3. SATELLITE SAMPLES 58

LS-312514-628 NUV FUV

Mr = -12.9 mag reff = 3.6 arcsec S/N = 3.7 S/N = 5.1 mag = 24.82 M = -10.1 mag M = -10.5 mag eff arcsec2 NUV FUV g r = 0.32 mag rcog = 4.8 arcsec rcog = 4.8 arcsec DES-206747419 NUV FUV

Mr = -19.7 mag reff = 22.9 arcsec mag S/N = 23.0 S/N = 178.0 = 21.95 eff arcsec2 MNUV = -17.5 mag MFUV = -17.2 mag g r = 0.39 mag rcog = 56.4 arcsec rcog = 56.4 arcsec

Figure 3.1: Optical (left, composite grz from the DESI Legacy Surveys Imaging DR9; Dey et al. 2019a), masked NUV (middle), and masked FUV (right) image cutouts of a small, faint (top) and large, bright (bottom) observed satellite to illustrate our curve-of-growth UV measurement method. The cyan circles represent the aperture with radius rCOG, within which the UV emission is measured. Optical properties from M21 and UV properties that we measure are shown in the bottom-left of the corresponding panel. The scale bar at the bottom of the left panels represents 30 arcseconds.

separation, and kinematics) and their surrounding environment. In contrast, the Auriga

project simulates isolated MW-like halos. Both suites invoke diﬀering models for galaxy

formation and evolution which include prescriptions for all relevant physical processes (i.e.

gas cooling, stellar and AGN feedback, UV background, etc.). For more details on the

EAGLE model used in APOSTLE, see Schaye et al.(2015) and Crain et al.(2015); for

Auriga details, see Grand et al.(2017).

For APOSTLE, we consider the 12 intermediate-resolution (L2) MW-M31 analog pairs

12 2 for a total of 24 distinct satellite systems around hosts with Mhalo (0.5 2.4) 10 M . ∼ − × 2Halo masses in both APOSTLE and Auriga are calculated within the radius that encompasses a mean matter density equal to 200 times ρcrit. 3.4. IDENTIFYING STAR-FORMING SATELLITES 59

For Auriga, we consider the satellite systems of 37 standard resolution (Level 4) non-merging

12 hosts with Mhalo (0.4 2) 10 M (Simpson et al., 2018a; Grand et al., 2019). The ∼ − × 5 adopted simulations have comparable dark matter particle (mDM 5.9 10 M vs. ∼ × ∼ 5 5 4 3 10 M ) and stellar/baryon (mstar 1.2 10 M vs. 5 10 M ) resolutions, respectively. × ∼ × × We test convergence with higher resolution volumes available for both APOSTLE and Auriga

in Appendix 3.7 and ﬁnd no signiﬁcant deviation in the estimated satellite quenched fractions

from the standard resolution volumes.

We deﬁne the simulated satellite population by selecting from the set of SUBFIND

(Springel et al., 2001) subhalos that have embedded galaxies with stellar masses within two

6 spherical stellar half-mass radii of M∗ 10 M and are within an aperture of radius 400 kpc ≥ around each host. We note that we tested smaller and larger spherical apertures (i.e. 300

kpc and 1 Mpc) around Auriga hosts and find a minimal difference (< 5%) on our final

results, likely due to the application of the SAGA-II selection function (see below).

We take a single random projection (diﬀerent projections produce nearly identical results

on the whole) for each host to deﬁne the sample, although we orient the line connecting

APOSTLE host pairs away from the line of sight to minimize the eﬀects of interlopers

from the other host. For a given simulation volume orientation, the sample selection criteria

mimic those of SAGA-II: we choose the set of these subhalos with projected separations

−1 10 kpc Dproj 300 kpc and relative line-of-sight velocities ∆Vsys 275 km s of their ≤ ≤ | | ≤ host. This produces a simulated APOSTLE sample of 229 satellites, and a simulated Auriga

sample of 411 satellites. We discuss the similarities and diﬀerences between these simulated

samples in §3.4.2.

3.4 Identifying Star-Forming Satellites

With the satellite samples established in §3.3, we now outline our method to select observed

(§3.4.1) and simulated (§3.4.2) star-forming satellites within them. We check for consistency 3.4. IDENTIFYING STAR-FORMING SATELLITES 60

of our star-forming satellite deﬁnitions across the observations and the simulations in §3.4.3.

3.4.1 Observed Star-Forming Satellites

We use UV emission as the primary indicator of star formation in the observed satellites from SAGA-II with archival GALEX imaging (see Table 3.1). We take a curve-of-growth approach using the Astropy Photutils package (Bradley et al., 2020) to detect statistically signiﬁcant UV emission. Our method is illustrated in Figure 3.1, and the corresponding measurements are in Table 3.1.

We start by masking bright sources near the satellite targets in each 1.2 degree-wide

GALEX tile and measuring the mean and standard deviation of the flux within 1000 randomly-placed circular regions across them. The region radius is the satellite effective radius reff from the SAGA-II photometry. Working from the (generally deeper) NUV tile, we measure background-subtracted fluxes within circular apertures about the optical position of each satellite starting from r = 0.5reff . We increase the aperture size in steps of 0.75“

(3“) for less (more) extended sources until the background-subtracted ﬂuxes in adjacent apertures change by less than the noise diﬀerence between them. We compute the signal-to- noise S/NNUV in the smaller of these regions (with a radius rCOG reported in Table 3.1), and place an identical region on the FUV tile to measure S/NFUV .

We consider measured ﬂuxes with S/N > 2 as detections in a given band. By this deﬁnition, 115/119 satellites with GALEX coverage have associated NUV emission, and

104/113 have associated FUV emission. We use standard equations (Morrissey et al., 2007) to convert to apparent AB magnitudes mNUV and mFUV (see Table 3.1), correcting for foreground extinction using E(B V ) from Schlaﬂy and Finkbeiner(2011) with RNUV = − 8.2,RFUV = 8.24 (Wyder et al., 2007).

Not only do the vast majority of the SAGA-II satellites with UV coverage show emission in one or both bands, but the correspondence between satellites with Hα equivalent widths

EW 2Å(M21) is also very high: 98/113 satellites with observations in NUV, FUV and ≥ 3.4. IDENTIFYING STAR-FORMING SATELLITES 61

Hα are detected in all three tracers. We posit that the majority of non-detections stem from observational limitations (such as image depth/sensitivity combined with satellite distances or Hα fiber position) rather than physical differences. The strong correlation between UV and Hα star formation tracers, despite the difference in the timescales they probe, is common in dwarf galaxies (e.g. Lee et al., 2011), and suggests that quenching is rapid at these masses

(e.g. Wetzel et al., 2015a).

We deﬁne an observed satellite to be star forming either if it is detected in the UV or if it has EW 2Å as reported by M21. Since we ﬁnd 12 (6) satellites with NUV (FUV) ≥ emission that do not satisfy the Hα criterion but only 2 satellites (1 NUV, 1 FUV) for which the inverse is true, the fraction of star-forming satellites in Table 3.1, 120/127, is marginally higher than that reported by M21. We discuss the implications of these numbers for the quenched fraction in §3.5.

3.4.2 Simulated Star-Forming Satellites

We consider two SFR measures to identify star-forming satellites in the simulations. Our

ﬁducial metric, the “instantaneous" rate, SFRsim, is estimated using the gas particles associated with the satellite subhalo determined by SUBFIND at z = 0 and corresponding star-formation rate relations for APOSTLE (Schaye et al., 2015) and Auriga (Springel and

Hernquist, 2003; Grand et al., 2017), with the former using a gas pressure threshold and the latter using a gas density threshold. These SFRs have previously been shown to reproduce observed trends (Vogelsberger et al., 2013; Furlong et al., 2015; Schaye et al., 2015). We also consider the average mass of star particles formed over the last gigayear as a measure of

SFR. Like SFRsim this metric is less susceptible to shot noise than estimates over shorter time intervals, but, unlike the observational tracers, it averages over a signiﬁcant fraction of a satellite orbit. We demonstrate in Appendix 3.7 that both metrics produce similar results, and adopt SFRsim to select simulated star-forming satellites throughout.

−1 We deﬁne a simulated satellite as star-forming if SFRsim > 0 M yr . A total of 54/229 3.4. IDENTIFYING STAR-FORMING SATELLITES 62

APOSTLE UV SF, Hα Q. SAGA-II 80% SAGA-II 100% Spec. Cov. APOSTLE-SF Auriga UV Q, Hα SF 7 Auriga-SF 0 UV + Hα SF UV+Hα Q. APOSTLE-Q Auriga-Q 6 SAGA-II-SF

) 1 1 − − 5 yr 2 M − 4

3 SFR/ − 3 log( 4 2 −

5 Cumulative #1 of Satellites per Host − Nq/Ntot = 175/229 Nq/Ntot = 259/411 0 6 7 8 9 10 6 7 8 9 10 log(M /M ) log(M /M ) ∗ ∗

Figure 3.2: Left: Simulated satellite SFRsim-M∗ relation derived from APOSTLE (cyan squares) and Auriga (pink triangles). Quenched simulated satellites are represented as short vertical lines at their M∗, along with quenched/total number counts Nq/Ntot to the right. The SFRNUV -M∗ relation for observed satellites is overplotted in green and purple, with the symbol shape and color indicating whether or not the satellite is star-forming (SF) from UV and/or Hα tracers as explained in the legend. Right: Cumulative number of satellites per host for the three samples. The quenched and star-forming satellites from the two simulations are shown as dashed and solid lines, respectively. The solid purple histogram shows star- forming satellites from SAGA-II, with the vertical dotted (dash-dotted) lines showing 100% (80%) spectroscopic coverage. There is very good agreement between the observed and simulated star-forming satellites by these metrics. There is also a population of low-mass quenched satellites in the simulations that has no counterpart in the observed satellite list in Table 3.1.

APOSTLE and 152/411 Auriga satellites meet this criterion, and their properties are illus-

trated in Figure 3.2. The left-hand panel of Figure 3.2 shows the SFRsim M∗ relation of − star-forming satellites and the stellar mass distribution of the quenched ones. The right-hand

panel of Figure 3.2 shows the cumulative number of star forming and quenched satellites per

simulated host. Considering the diﬀerences between the simulations (see §3.3.2), there is good

agreement between them despite the diﬀerent host environments: star-forming satellites in

both APOSTLE and Auriga follow similar SFRsim M∗ relations, and both populations − become increasingly dominated by quenched systems at low M∗. These trends are qualita-

tively similar if a speciﬁc SFR threshold is adopted instead of a non-zero SFRsim (i.e. Akins 3.4. IDENTIFYING STAR-FORMING SATELLITES 63

et al., 2021).

The mild diﬀerence between the APOSTLE and Auriga satellite samples in the right panel of Figure 3.2 likely stems from the diﬀerent galaxy formation prescriptions adopted by the simulations. For example, APOSTLE halos may more readily remove gas from lower- mass subhalos leaving them permanently quenched, while Auriga subhalos may re-accrete expelled gas allowing them to be more long-lived. Additionally, the earlier onset of the

UV background in the APOSTLE simulations may further contribute to fewer star-forming satellites per host at lower M∗.

3.4.3 Comparing Star-Forming Satellites

To check for consistency between our deﬁnition of star-forming satellites in the observed and simulated samples, we estimate SFRs for the SAGA-II satellites with NUV detections,

SFRobs, to compare with SFRsim for simulated objects. We use Equation 3 from Iglesias-

Páramo et al.(2006) and the NUV luminosity, LNUV , calculated from mNUV assuming the satellite to be at the distance of its host (see Table 3.1). We do not perform internal extinction corrections to LNUV in estimating SFRs, since homogeneously-measured infrared (IR) ﬂuxes would be required and since our main interest in the SFRs themselves is diagnostic. The IR correction is likely small at the low-M∗ end of the satellite distribution, but signiﬁcant (and uncertain) at higher masses (e.g. McQuinn et al., 2015). The values of SFRobs in Table 3.1 are therefore approximate, and likely represent lower limits at the high-mass end.

We check for consistency between SFRobs and SFRsim in the left panel of Figure 3.2, where the SFRobs M∗ relation for the observed sample is plotted with the symbol shapes − and colors identifying star-forming satellites according to UV and/or Hα criteria. It is clear that there is broad agreement between the observed and simulated star-forming satellites3,

9 with the lower SFRobs at M∗ & 10 M relative to SFRsim likely stemming from the lack

3 The object with the lowest M∗ in the observed sample (LS-330948-4542) appears to have a size that is severely under-estimated in the M21 catalog, which likely explains its outlying SFRobs in the left panel of Figure 3.2. 3.5. OBSERVED AND SIMULATED QUENCHED FRACTIONS 64

of an IR correction in the former. The cumulative distribution of observed star-forming satellites per host, shown in the right panel of Figure 3.2, also compares favourably to that from the simulations, with the observed distribution falling in between those for APOSTLE

8.5 9 and Auriga for M∗ . 10 M and slightly below both of them for M∗ & 10 M . It is also interesting to brieﬂy consider the eﬀect of the hosts in this comparison. The majority of hosts in the observed sample and all of those in the simulated sample are star-forming galaxies.

Both samples also demonstrate the concept of ’galactic conformity’, where the properties of the satellites match those of their hosts (e.g. Phillips et al., 2014), at higher satellite masses.

Taken as a whole, Figure 3.2 suggests broad consistency between the deﬁnition of a star-forming satellite in the observed sample and in the simulated samples. It is also clear from the paucity of open green circles relative to the short vertical lines in the left panel of Figure 3.2 that there is a population of low-mass quenched satellites in the simulations that has no counterpart in the observed satellite list in Table 3.1; the right panel of Figure

3.2 illustrates that a signiﬁcant fraction of the quenched simulated satellites fall within the

80% – 100% spectroscopic coverage limits for SAGA-II. This suggests a higher number and fraction of quenched satellites in the simulated samples than in the observed one, although incompleteness and interlopers in the latter need to be considered. We compare quenched fractions in the next section.

3.5 Observed and Simulated Quenched Fractions

With star-forming satellites identiﬁed and their consistency checked, we proceed to compare observed and simulated quenched fractions.

Because most SAGA-II candidates are confirmed in the Hα emission line (see §3.3.1), Ta- ble 3.1 is likely missing quenched satellites even in regions where the spectroscopic coverage is high. Interloping field galaxies are also more likely to be star-forming than quenched given their relative ubiquity (Geha et al., 2012a). Correcting for both effects would systematically 3.5. OBSERVED AND SIMULATED QUENCHED FRACTIONS 65

raise the observed quenched fraction relative to that calculated directly from Table 3.1. M21 model them in detail, deriving a (dominant) incompleteness correction by assuming that all spectroscopically targeted but undetected candidates are quenched satellites, and a (sub- dominant) interloper correction by drawing mock samples from gravity-only simulations.

We use the M21 corrections directly from their Figure 11 (in the same M∗ bins), retaining the interloper correction despite comparing to simulations (for which it should not be required). The incompleteness/interloper-corrected quenched fractions we adopt are therefore conservative upper limits on the observed values implied by SAGA-II.

Figures 3.3 and 3.4 plot the observed and simulated quenched fractions in two diﬀerent ways. In Figure 3.3, the SAGA-II quenched fractions (purple stars with dark bars showing random counting uncertainties4 and light bars showing systematic incompleteness/interloper corrections) are compared to those in APOSTLE (cyan band and squares) and Auriga (pink band and triangles). Figure 3.4 plots M∗ as a function of projected host distance Dproj for star-forming (stars) and quenched (circles) satellites in the APOSTLE (cyan) and Auriga

(pink) samples. The M∗ bin deﬁnitions in M21 and Figure 3.3 are shown as a gradient of purple horizontal bands in Figure 3.4. The average simulated quenched fraction in those bins is directly to their right, and the range of observed quenched fractions bracketed by the ratios from Table 3.1 (smaller value) and the incompleteness/interloper-corrected ratios

(larger value) are in parentheses. In both plots, the dotted (dash-dotted) lines show the

SAGA-II 100% (80%) spectroscopic coverage.

Figures 3.3 and 3.4 illustrate that, despite their diﬀerences (i.e. hydrodynamic schemes, galaxy formation and evolution models, and host environments), there is a striking correspondence between the quenched fractions from the APOSTLE and Auriga simulations as

6 10 a function of M∗ and Dproj for 10 . M∗/M . 10 . It is also clear that, even when conservatively accounting for both incompleteness and interlopers as in M21, the observed

468% conﬁdence intervals calculated using the Wilson score interval (Brown et al., 2001) for both the observed and simulated samples. 3.6. DISCUSSION AND CONCLUSIONS 66

satellite quenched fraction is lower than in the simulations across the M∗ range considered.

7 8 The diﬀerence is largest for 10 . M∗/M . 10 , where the SAGA-II spectroscopic coverage is essentially complete and the incompleteness/interloper-corrected observed quenched

fraction is 2–3 times lower than in the simulations. We discuss the implications of this result

in the following section.

Finally, the median projected separation for the SAGA-II quenched and star-forming

objects are shown at the bottom of Figure 3.4 by short, white lines. These separations are

relatively consistent with the simulations where star-forming satellites have larger projected

separations than quenched satellites, however the scarcity and incompleteness of observed

quenched satellites limits this comparison. This perspective illustrates a distance dependence

8 9 in both simulation samples beginning at intermediate masses (10 . M∗/M . 10 ) where quenched satellites are located at lower projected distances and become more ubiquitous at

lower stellar masses, similar to previous trends reported around more massive hosts and in

the LG (Guo et al., 2013; Wang et al., 2014; Fillingham et al., 2018a).

3.6 Discussion and Conclusions

We have identiﬁed star-forming satellites around MW analogs in observed and simulated

samples, which have similar sizes, similar host masses, and which are selected in a simi-

lar manner (§3.3). We used UV emission in conﬁrmed SAGA-II satellites (Figure 3.1) and

an instantaneous SFR in APOSTLE and Auriga satellites to deﬁne observed and simu-

lated star-forming objects, respectively, which were checked for consistency (§3.4 and Fig-

ure 3.2). We compared quenched fractions in the resulting samples, and ﬁnd that the

incompleteness/interloper-corrected observed values are 2–3 times lower than the simu- ∼ 7 8 lated ones for 10 . M∗/M . 10 (§3.5 and Figures 3.3 and 3.4). The observed and simulated quenched fractions are therefore strongly discrepant in a mass range that is well-

probed in both samples. 3.6. DISCUSSION AND CONCLUSIONS 67

1.0 SAGA-II 80% Spec. Cov. SAGA-II 100% APOSTLE Auriga SAGA-II 0.8

0.6

0.4 Quenched Fraction

0.2

0.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 log(M /M ) ∗ Figure 3.3: Satellite quenched fractions as a function of stellar mass. The dark purple stars show the observed quenched fraction, with dark purple bars showing random uncertainties at 68% conﬁdence and the light purple bars showing the systematic incompleteness and interloper corrections from M21 (see text). The dotted (dash-dotted) lines show the SAGA- II 100% (80%) spectroscopic coverage. The pink triangles, cyan squares and corresponding colored bands show the simulated quenched fraction and 68% conﬁdence intervals from APOSTLE and Auriga, respectively. Even accounting for incompleteness, there is a clear 7 discrepancy between the observed and simulated quenched fractions for 10 . M∗/M . 108, where the SAGA-II spectroscopic coverage is high.

The comparisons presented here are broadly consistent with previous investigations of satellite quenched fractions. Observationally, our quenched fractions diﬀer only slightly from those found by M21, which Akins et al.(2021) report to be lower than those of satellites 3.6. DISCUSSION AND CONCLUSIONS 68

around 4 MW-like hosts in the Justice League simulations. Furthermore, we note that this discrepancy extends to other nearby systems that are similar to the MW/M31 with respect to quenched satellites (Chiboucas et al., 2013a; Carlsten et al., 2021a) and it can also be inferred from the star formation histories of satellites from the FIRE-2 simulations (Garrison-

Kimmel et al., 2019b). More broadly, since the quenched fractions in the LG have been shown to agree with simulations (e.g. Fillingham et al., 2016a; Simpson et al., 2018a; Joshi et al.,

2021; Akins et al., 2021) and the LG and SAGA-II have been shown to disagree (Geha et al.

2017b; M21), the discrepancy in quenched fractions between SAGA-II and the APOSTLE and Auriga simulations is unsurprising. Here, we have demonstrated the degree to which the observations and simulations are inconsistent with large, comparably-sized samples, and that the discrepancy is robust against diﬀerent choices of observed or simulated star formation tracers as well as a variety of simulation parameters (§3.4.2 and Appendix 3.7).

The agreement between the SFR–M∗ relations and cumulative M∗ distributions in Fig-

ure 3.2 combined with the discrepant quenched fractions in Figures 3.3 and 3.4 suggest

that the diﬀerence between the observed and simulated samples stems from the number of

6.6 7.8 quenched satellites in each. For 10 . M∗/M . 10 , the APOSTLE and Auriga samples, respectively, have an average of 3.5 and 4.9 satellites per host, comparable to the LG

(McConnachie, 2012a) but greater than the incompleteness/interloper-corrected SAGA-II

value5 of 2 (M21). The star-forming satellite counts, quenched fractions and total satel- ∼ 6.6 7.8 lite counts presented here are therefore all broadly consistent at 10 . M∗/M . 10 if there are 50 100 additional quenched satellites in the simulated samples compared to that − estimated for the incompleteness/interloper-corrected observed sample.

One possibility is that the quenched satellite number diﬀerence is observationally-driven: in this scenario, the SAGA-II incompleteness correction under-estimates the number of

6.6 7.8 quenched satellites with 10 . M∗/M . 10 , with the deﬁcit increasing towards the 5The incompleteness/interloper models of M21 predict that Table 3.1 is missing ∼ 0.7 satellites per host 6.6 7.8 in the 10 . M∗/M . 10 range; see their §5.3. 3.6. DISCUSSION AND CONCLUSIONS 69

low-mass end (c.f. Figure 3.2 right). The nearly complete SAGA-II spectroscopic coverage and the conservative M21 incompleteness correction suggest that the quenched satellites would most likely be missing from the imaging catalogs from which spectroscopic targets are drawn.

It is plausible that low surface brightness (LSB) satellites are missing from the SAGA-II imaging catalogs from which follow-up targets are drawn since they are not developed for

LSB detection. M21’s comparisons to deeper overlapping catalogs argue against this scenario, although quantitative photometric completeness simulations (e.g. Bennet et al., 2017a) have not been carried out. It is also plausible that a larger fraction of the 70 detected diﬀuse LSB galaxies (dLSBGs) without redshifts are actually satellites than the 25% 30% assumed in − the M21 incompleteness correction. If all of these dLSBGs were satellites, it may remedy

the discrepancy at the lowest masses, however, the total number of quenched satellites per

host would still be low compared to that in the simulations. Extending these investigations

of surface brightness eﬀects to simulations may provide some additional insight (e.g. Font

et al., 2020). We conclude that observational eﬀects are unlikely to fully explain the quenched

fraction discrepancy reported here.

A second possibility is that the observed and simulated quenched fraction diﬀerence is

simulation-driven: in this scenario, the simulations over-predict the number of quenched

satellites around MW analogs. The correspondence between APOSTLE and Auriga in Fig-

ures 3.3 and 3.4 as well as similar results from other simulations (Akins et al., 2021; Joshi

et al., 2021) imply that the eﬀect is somewhat model-agnostic. This consistency is not neces-

sarily predictable: while tides are relatively similar across simulation suites, the interstellar

medium (ISM), star-formation feedback-dependent physics and hydrodynamical schemes

7 8 producing the ram pressure that begins to dominate quenching of 10 . M∗/M . 10 satellites (Wetzel et al., 2015a; Fillingham et al., 2016a) are not (e.g. Agertz et al., 2007;

7 Sijacki et al., 2012). Furthermore, Digby et al.(2019) ﬁnd all intermediate-mass ( 10 . 9 M∗/M . 10 ) dwarfs in APOSTLE and Auriga have young (τform . 6 Gyr ago) stellar 3.6. DISCUSSION AND CONCLUSIONS 70

populations. This suggests that any form of quenching in these satellites, as implied in this work, must have occurred recently and rapidly, consistent with previous similar investigations (e.g. Wetzel et al., 2015a; Fillingham et al., 2016a).

Whether or not the agreement between simulated quenched fractions presented here has a common physical origin or stems from a conﬂuence of disparate eﬀects with a similar net outcome is unclear. Nonetheless, it is plausible that the ISM gas densities simulated here with state of the art resolution and star formation feedback physics in the simulations generically produce satellites that are less resilient to ram-pressure stripping than in nature.

A separate, detailed study is required to determine if this mechanism quantitatively explains the discrepancy reported here (e.g. Bose et al., 2019; Digby et al., 2019).

We conclude that the dearth of observed quenched satellites relative to simulated ones in

Figures 3.3 and 3.4 is not readily explained by vagaries in the samples considered here. There is apparently a genuine discrepancy between the satellite populations of the MW, M31 and their simulated analogs on the one hand, and of the SAGA-II host galaxies on the other.

This highlights that while the ability to reproduce the properties of the LG is a necessary feature of any complete model of galaxy formation and evolution, exclusive reliance on the

LG as the benchmark for faint satellites risks introducing severe biases in the models. More detailed comparisons between observed and simulated satellites will further elucidate the origin of this discrepancy. This requires larger, more observationally complete samples that probe even further down the luminosity function (e.g Bennet et al., 2019a; Carlsten et al.,

2020), and large samples of simulated satellites (e.g. this work; Font et al., 2020; Joshi et al.,

2021) at higher resolutions (e.g. Wetzel et al., 2016; Wheeler et al., 2019) and self-consistent star-forming ISMs. Both will be available soon. 3.6. DISCUSSION AND CONCLUSIONS 71

Quenched APOSTLE SAGA-II Star-Forming Auriga Quenched Fraction in Sims. (vs. SAGA-II) 10

0%(0% 6%) ∼ − 9

) 15%(0% 8%) ∼ − /M

∗ 30%(11% 22%) ∼ − M 8 SAGA-II 100% Spec. Cov. 35%(0% 10%)

log( ∼ − 55%(5% 23%) ∼ − 65%(10% 36%) ∼ − 7 90%(0% 58%) SAGA-II ∼ − 80%

Median SAGA-II Q Median SAGA-II SF 0 50 100 150 200 250 300 350 400 (Dproj; kpc)

Figure 3.4: Stellar mass as a function of projected distance Dproj of star-forming (stars) and quenched (circles) simulated satellites from APOSTLE (cyan) and Auriga (pink). The horizontal purple bands show the observed quenched fraction bins from M21 and Figure 3.3. The average simulated quenched fraction in each band is given to the right of it, and the numbers in parentheses show the measured (smaller) and incompleteness/interloper- corrected (larger) quenched fractions in the observed sample. The horizontal dotted (dash- dotted) lines show the SAGA-II 100% (80%) spectroscopic coverage, and the short vertical bars at the bottom show the median quenched (left) and star-forming (right) of the observed satellites in Table 3.1. There are hints of a Dproj dependence of the quenched fraction in the simulations and observations. Even accounting for incompleteness, there is a discrepancy 7 8 between the observed and simulated quenched fractions for 10 . M∗/M . 10 , where the SAGA-II spectroscopic coverage is high. 3.6. DISCUSSION AND CONCLUSIONS 72 SF? GALEX Tile ) 1 NUV − yr SFR

M log ( ∼ 0.140.190.180.06 -2.570.05 -2.460.03 -2.46 -1.810.04 Y0.05 -2.16 Y0.08 -0.82 Y Y AIS_183_50183_0001_sv27 -1.49 AIS_183_50183_0001_sv18 -1.76 Y AIS_183_50183_0001_sv27 -2.22 Y AIS_183_50183_0001_sv26 MISWZS03_27553_0283_17492 Y MISWZS03_27605_0283_17497 Y Y AIS_182_50182_0001_sv68 AIS_182_50182_0001_sv68 AIS_182_50182_0001_sv68 ± ± ± ± ± ± ± ± ± FUV m FUV ) S/N ( 0.200.190.160.08 8.40.08 5.70.05 6.60.05 18.90.13 20.55 0.03 – 20.61 220.05 20.72 32.2 19.01 0.07 1.2 29.2 22.1 19.73 16.36 13 – 17.88 >24.04 18.55 19.55 -2.99 -2.05 Y Y MISWZS03_27552_0283_17564 AIS_284_50284_0001_sv49 ± ± ± ± ± ± ± ± ± ± ± NUV tile used. m GALEX NUV ) S/N ( COG Table 3.1: UV properties of Observed satellites r host D deg deg (Mpc) (arcsec) (mag) (mag) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Name RA Dec LS-432563-224 20.7772 17.8916 38.4 13 13.9 18.88 DES-313240666DES-350665706 39.9254DES-353757883 50.1913DES-353742769 -1.4187 50.4652DES-371747881 -15.5749 50.9464DES-373383928 -15.7104 55.3397DES-373393030 -15.4004 37 34.3 55.5682 -13.1446 34.3 55.5841 34.3 -13.217 -13.4218 31.9 10.6 8.8 61.7 31.9 31.9 9.2 15.3 23.2 9.1 13.9 6.8 21.5 8.6 34.7 19.49 19.40 16.16 21.6 16.2 17.67 21.58 18.35 19.48 LS-429811-3398LS-431187-1672 20.285LS-429812-2469 20.328 20.5362 17.6022 17.7539 17.5279 38.4 38.4 38.4 6.8 9.3 7.2 5.7 5.7 7.2 20.78 20.49 20.50 The first 10 rows of thiscols. table are (1)–(4): showncol. here. (6): TheM21 NUV fullNUV signal-to-noise table starsatellite ratio. is formation col.(7): availablecol.(11): name, online rates NUV Star-forming in apparent computed classification machine J2000 AB using as readable magnitude, fluxes defined centroid format. corrected in in for § 3.4.1 position col. foregroundcol.(12): extinction. (7), and col. name distances (8)-(9): of from host Same col. as distance. (4) cols. and (6) col. Equation and(7) but (3) (5) for from FUV. Curve-of-growthIglesias-Páramo col.(10) et First-order radius al. ( 2006 ) uncorrected around for the internal dust optical attenuation. centroid used to measure UV fluxes. 3.7. APPENDIX: TESTING RESOLUTION AND STAR-FORMATION TRACERS IN SIMULATIONS 73

3.7 Appendix: Testing Resolution and Star-formation Tracers

in Simulations

To test our results for convergence with resolution, we consider 5 high resolution (L1)

4 volumes (10 hosts) from the APOSTLE simulations with mDM 5 10 M , mstar ∼ × ∼ 4 4 1 10 M and 6 high resolution (Level 3) hosts from Auriga with mDM 4 10 M , × ∼ × 3 mstar 6 10 M . The satellites in these sets of simulations are treated as in 3.3.2: sub- ∼ × § halos are selected with SUBFIND, the SAGA-II spatial selection criteria are applied, and all physical properties are deﬁned identically.

We also test for any dependence on our star-formation metric, i.e. SFR estimated from the gas particles/cells. We repeat our quenched fraction estimates using the star-formation rate calculated based on the average number of star particles created over the last gigayear

(SFR-1Gyr). This measure provides a more accurate estimate of a satellite’s SFR compared to SFR dervied from star particles on shorter timescales that are susceptible to shot noise given the time and particle resolutions in the simulations. However, this measure will lead to a marginally higher number of star-forming satellites relative to our ﬁducial as it will include satellites that may have ceased forming stars within the last gigayear.

The results of these tests are shown in Figure 3.5. The left column shows the quenched fractions as a function of stellar mass for the APOSTLE (top) and Auriga (bottom) samples

−1 using our ﬁducial SFR deﬁnition, i.e. SFRsim = SFRgas > 0 M yr . The shaded regions

correspond to the total sample at the standard (ﬁlled; 229 APOSTLE satellites and 411

Auriga satellites), the subset simulated at higher resolution (diagonal hatched pattern; 123

APOSTLE satellites and 92 Auriga satellites), and the matching subset of volumes at the

standard resolution (grid hatched pattern; 98 APOSTLE satellites and 79 Auriga satellites).

The right column of Figure 3.5 plots the same samples except using the alternative SFR

−1 based on the star particles, i.e. SFR1Gyr > 0 M yr . In all 4 panels, we can see that the

discrepancy in quenched fractions as a function of stellar mass, our primary result, remains 3.7. APPENDIX: TESTING RESOLUTION AND STAR-FORMATION TRACERS IN SIMULATIONS 74

between the observed and simulated samples at higher resolution and with an alternative star-formation deﬁnition.

SFR-Gas SFR-1Gyr 1.0 SAGA-II 80% Spec. Cov. SAGA-II 100% 1.0 AP-L2-tot(229) AP-L2-sub(98) AP-L1-sub(123) SAGA-II 0.8 0.8

0.6 0.6

0.4 0.4 Quenched Fraction

0.2 0.2

0.0 0.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 6.5 7.0 7.5 8.0 8.5 9.0 9.5 1.0 1.0 Au-L4-tot (411) Au-L4-sub (79) Au-L3-sub (92) SAGA-II 0.8 0.8

0.6 0.6

0.4 0.4 Quenched Fraction

0.2 0.2

0.0 0.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 6.5 7.0 7.5 8.0 8.5 9.0 9.5 log(M /M ) log(M /M ) ∗ ∗

Figure 3.5: Quenched fraction as a function of stellar mass plotted in the same manner as Figure 3.3. The top and bottom rows show the APOSTLE and Auriga simulations, respectively. The left column plots the quenched fractions calculated using the SFRs derived from the gas, while the right column shows the quenched fractions calculated using the average SFR over the past 1 Gyr based on the stellar particles/cells. The ﬁlled bands show the total simulation samples at the standard resolution, the diagonal-hatched bands, show the higher resolution subset, and the grid-hatched bands show the subset of standard resolution volumes that match the high resolution volumes. For reference, the total number of satellites used in these comparisons is listed in the legend. Chapter 4

Systematically Measuring Ultra Diﬀuse Galaxies in HI: Results from the Pilot Survey

Statement of Co-Authorship

This chapters consists of a version of a publication in the Astrophysical Journal (A. Karunakaran

et al 2020 ApJ 902 39, ADS link). I am the ﬁrst author of this paper. I carried out all of

the GBT observations, the subsequent data reduction and analysis, and produced all of the

ﬁgures. The photometric properties presented in this work were derived by our collaborator

Dr. R. Donnerstein who was also helpful in the additional GALFIT modelling which I con-

ducted. I wrote the bulk of the manuscript with Drs. K. Spekkens and D. Zaritsky providing

more signiﬁcant comments and edits.

75 4.1. ABSTRACT 76

4.1 Abstract

We present neutral hydrogen (HI) observations using the Robert C. Byrd Green Bank Tele- scope (GBT) of 70 optically-detected UDG candidates in the Coma region from the System- atically Measuring Ultra-Diﬀuse Galaxies survey (SMUDGes). We detect HI in 18 targets, conﬁrming 9 to be gas-rich UDGs and the remainder to be foreground dwarfs. None of our

HI-detected UDGs are Coma Cluster members and all but one are in low-density environments. The HI-detected UDGs are bluer and have more irregular morphologies than the redder, smoother candidates not detected in HI, with the combination of optical color and morphology being a better predictor of gas richness than either parameter alone. There is little visual difference between the gas-rich UDGs and the foreground dwarfs in the SMUDGes imaging, and distances are needed to distinguish between them. We find that the gas richnesses of our HI-confirmed UDGs and those from other samples scale with their effective radii in two stellar mass bins, possibly providing clues to their formation. We attempt to place our UDGs on the baryonic Tully-Fisher relation (BTFR) using optical ellipticities and turbulence-corrected HI linewidths to estimate rotation velocities, but the potential systematics associated with fitting smooth Sérsic profiles to clumpy, low-inclination low surface brightness disks precludes a meaningful analysis of potential BTFR offsets. These observations are a pilot for a large campaign now underway at the GBT to use the HI properties of gas-rich UDGs to quantitatively constrain how these galaxies form and evolve.

4.2 Introduction

The study of the low surface brightness (LSB) galaxy population has been reinvigorated as a result of improvements in astronomical instrumentation (e.g., Abraham and van Dokkum,

2014; Aihara et al., 2018) and data reduction methods (e.g., Fliri and Trujillo, 2016b; Trujillo and Fliri, 2016), as well as the use of novel image searching algorithms (e.g., Bennet et al.,

2017b; Müller et al., 2018a; Prole et al., 2018; Zaritsky et al., 2019; Carlsten et al., 2020). 4.2. INTRODUCTION 77

Among this recent surge of LSB detections are populations of extended red LSB galaxies akin to those discovered in early LSB studies (e.g., Sandage and Binggeli, 1984; Impey et al.,

1988; Bothun et al., 1991, see also Conselice 2018).

In their survey of the Coma Cluster, van Dokkum et al.(2015a) presented the first significant sample of these extended LSBs, dubbing them ultra diffuse galaxies (UDGs) and

2 proposing size (Reff > 1.5 kpc) and surface brightness (µ0,g & 24 mag/arcsec ) criteria that have since been widely adopted to deﬁne them. To date, over 1000 UDG candidates have been discovered in subsequent searches of the Coma Cluster (e.g., Koda et al., 2015b; Yagi et al., 2016; Zaritsky et al., 2019) and several other clusters (e.g., Mihos et al., 2015b; Beasley et al., 2016; Shi et al., 2017; Venhola et al., 2017; Mancera Piña et al., 2019a; Lee et al.,

2020), as well as a growing number in lower density environments (e.g., Martínez-Delgado et al., 2016; Bellazzini et al., 2017; Román and Trujillo, 2017; Leisman et al., 2017; Trujillo et al., 2017; Bennet et al., 2018; Román et al., 2019; Barbosa et al., 2020).

Across these environments there exists a large diversity in the physical properties of

UDGs, similar to that seen in the high surface brightness galaxy population. Most UDGs seem to be embedded in dwarf galaxy-mass dark matter halos (e.g., Beasley and Trujillo,

2016; Amorisco et al., 2018; Chilingarian et al., 2019; Prole et al., 2019a), although there is evidence that at least some are in more massive halos (van Dokkum et al. 2016; Zaritsky 2017;

Lim et al. 2018; Forbes et al. 2020, although see Saifollahi et al. 2020). While UDGs found in clusters tend to be red (i.e., quiescent) and smooth, those in lower density environments are bluer (i.e., star forming) and have more irregular morphologies (Román and Trujillo,

2017; Prole et al., 2019b). Some UDGs exhibit extreme properties that pose challenges to proposed galaxy formation mechanisms, such as high dark matter fractions (van Dokkum et al., 2016; Beasley et al., 2016), dark matter deﬁciencies (van Dokkum et al. 2018, 2019; although see Trujillo et al. 2019), and oﬀsets from established galaxy scaling relations such as the baryonic Tully-Fisher relation (Mancera Piña et al., 2019b, 2020).

Proposed UDG formation mechanisms generally fall into two categories: internally and 4.2. INTRODUCTION 78

externally-driven physics. Isolated (i.e., ﬁeld) UDGs may be formed through multiple internal mechanisms. For example, Amorisco and Loeb(2016) suggest that UDGs formed in dwarf dark matter halos with elevated angular momenta, naturally explaining their extended sizes. Alternatively, using the NIHAO (Numerical Investigation of a Hundred Astrophysical

Objects, Wang et al. 2015) suite of simulations, di Cintio et al.(2017) show that UDG-like objects can form through bursty star-formation early in their evolution resulting in a more extended, diffuse matter distribution. The red, smooth UDGs observed in groups and clusters may represent the population of field UDGs that formed through the aforementioned mechanisms and were subsequently quenched via ram-pressure and/or tidal effects (Yozin and Bekki, 2015; Liao et al., 2019; Jiang et al., 2019; Carleton et al., 2019). However, some may form initially as typical dwarf galaxies that are tidally disturbed after in-fall into a cluster or by a massive companion (Bennet et al., 2018; Sales et al., 2020).

In order to constrain which of these proposed formation mechanisms explains the origin of the detected UDGs, larger samples of UDGs with distance measurements are required, particularly in the ﬁeld where inferring distances by projected separation from clusters or groups is not possible. While some optical distances to UDGs have been obtained (van

Dokkum et al., 2015b; Bellazzini et al., 2017; Kadowaki et al., 2017; Alabi et al., 2018;

Ferré-Mateu et al., 2018; Ruiz-Lara et al., 2018; Chilingarian et al., 2019; Martín-Navarro et al., 2019), sample sizes are limited due the large spectroscopic integration times required at low surface brightnesses.

By contrast, the neutral hydrogen (HI) in gas-rich UDGs can not only provide a distance measure but also help distinguish among formation mechanisms. The HI redshift provides kinematic distances for candidates that can distinguish foreground dwarfs (Reff < 1.5 kpc) from true UDGs (Reff > 1.5 kpc), and linewidths reﬂect their internal dynamics, and the

HI ﬂux provides the gas mass. HI follow-up observations of optically-detected UDG candidates have been demonstrated to be feasible with single-dish radio telescopes (Papastergis et al., 2017; Spekkens and Karunakaran, 2018) and searches through extant blind HI survey 4.2. INTRODUCTION 79

detections for diﬀuse stellar counterparts have also been fruitful (Leisman et al., 2017).

The Systematically Measuring Ultra-Diﬀuse Galaxies (SMUDGes; Zaritsky et al., 2019, hereafter Z19) survey is uniquely positioned to produce samples of UDG candidates for

HI follow-up observations across a range of environments, as the combination of depth and coverage of the DECaLS data used to detect UDG candidates are unmatched. The SMUDGes pilot survey searched publicly available DECaLS data (one of three DESI pre-imaging Legacy surveys, see Dey et al. 2019b for details) for large (reff > 5.3“ = 2.5 kpc at DComa ∼ 100Mpc) UDG candidates in a 290 deg2 region centered on the Coma Cluster. The 275

UDG candidates resulting from that search (Z19) as well as subsequent SMUDGes detections

provide ample targets to pilot a large follow-up campaign.

In this paper, we present pilot HI observations along the lines-of-sight to 70 SMUDGes

UDG candidates, which represent the ﬁrst phase of a large HI follow-up campaign using the

Robert C. Byrd Green Bank Telescope (GBT). We aim to obtain redshift measurements to

UDG candidates and characterize the gas properties of conﬁrmed UDGs to constrain their

formation mechanisms. These observations, which are part of a much larger GBT program

that is currently underway, represent the largest HI follow-up campaign of optically-selected

UDG candidates ever reported.

The structure of this paper is as follows. In Section 4.3, we describe our HI target

selection. We outline our observations and data reduction procedure in Section 4.4. In Section

4.5, we present the properties of our HI detections and non-detections. In Section 4.6, we

discuss the environmental and morphological properties of HI detections and non-detections,

place initial constraints on UDG formation mechanisms, and discuss our UDGs in the context

of the baryonic Tully-Fisher relation. We conclude and outline future work in Section 4.7.

−1 −1 Throughout this work we use DComa = 100 Mpc, H0 = 70km s Mpc , ΩΛ = 0.7, and

Ωm = 0.3. 4.3. SAMPLE SELECTION 80

4.3 Sample Selection

We select HI follow-up targets from the SMUDGes pilot sample (Z19) and subsequent searches of the DECaLS data. Focused on the 290 deg2 region centered on the Coma Cluster, the SMUDGes pilot survey employed a semi-automated UDG candidate identification procedure, described in detail in Z19. Briefly, the DECaLS observations were preprocessed to remove any defects, and then foreground or background sources significantly brighter than

UDGs were replaced with background noise. Next, these processed images were spatially

filtered to various scales using wavelet transforms, and diffuse objects are identified using

SEP (Barbary, 2016; Bertin and Arnouts, 1996). In order to compare results with other studies (e.g., van Dokkum et al., 2015a; Yagi et al., 2016), their photometric properties were then modeled as exponential proﬁles using GALFIT (Peng et al., 2010) and only objects

1 −2 with reff > 5.3“(Reff = 2.5 kpc at DComa ) and µ0,g > 24 mag arcsec were kept. The remaining objects were examined by eye, and 275 classiﬁed as bona-ﬁde UDG candidates.

We select 34 of the 275 SMUDGes UDG candidates with mg . 19.5 mag to follow up in HI (listed as Z19 in column 14 of Table 4.1). This magnitude limit combined with gas richness scaling relations for local dwarfs (e.g., Bradford et al., 2015) implies that integration times of no more than a few hours are required to follow up each source (see Section 4.4).

A subsequent optical search within the same region using an improved SMUDGes pipeline

(Zaritsky et al., in prep) detected an additional 36 UDG candidates that satisfied the above magnitude limit and we include them in our HI follow-up sample as well (listed as K20 in column 14 of Table 4.1). The DECaLS imaging for all targets was subsequently modeled as a Sérsic profile with a variable Sérsic index using GALFIT and the resulting parameters are listed in columns (4)-(11) of Table 4.1. The parameter uncertainties are the GALFIT values which are derived using Poisson pixel noise; a more comprehensive error estimation method for SMUDGes photometery is being developed using simulated UDG recovery for use in the

1 We use reff for angular sizes and Reff for physical sizes throughout this paper. 4.4. OBSERVATIONS AND DATA REDUCTION 81

full survey (Zaritsky et al., 2019).

The optical properties of the 36 previously unpublished UDG candidates are largely consistent with the sample selected from Z19, although there are a few candidates with

−2 smaller sizes (reff > 4.7“) and higher surface brightnesses (µ0,g > 23.7 mag arcsec ). Some of these candidates have mg > 19.5 mag (our HI follow-up criterion) because initial estimates were used during the target selection. There is some overlap of the UDG candidates in

SMUDGes with other UDG samples (see Z19). Of the 36 UDG candidates we present here,

6 have been either presented in other work and/or previously detected in HI. We include references for these objects in column 14 of Table 4.1.

In total, our HI follow-up sample consists of 69 UDG candidates in the Coma Cluster region and 1 outside of it2. Their projected spatial distribution relative to galaxies from the

SDSS DR15 (Aguado et al., 2019) with 5000km s−1 < cz < 9000km s−1 is shown in Figure

4.1.

We do not select on color in this work despite its accuracy for predicting gas richness

in the high surface brightness galaxy population (e.g., Catinella et al., 2012; Brown et al.,

2015). Instead, we investigate the relationship between color and gas-richness in Sections

4.5.2 and 4.6.

4.4 Observations and Data Reduction

We performed 88 hours of position-switched HI observations between 2018 February and

2018 August using the GBT along the lines of sight (LOS) to the 70 UDG candidates

in Table 4.1 (program AGBT18B-239). 9 objects were observed with an oﬀset between the

optical centroid and the LOS in order to minimize contamination from known nearby objects

(see Section 4.5.1). These objects are indicated with an asterisk next to their RA in Table

4.1.

2The exception is SMDG1103517+284118 which falls outside the Coma Cluster region and was nonetheless included as a target of interest 4.4. OBSERVATIONS AND DATA REDUCTION 82

HI detections, UDGs 40 HI detections, dwarfs HI non-detections SDSS R=3 Mpc 35

Dec (deg) 25

15 210 205 200 195 190 185 180 RA (deg)

Figure 4.1: Projected sky distribution of our UDG candidate HI follow-up sample in the Coma Cluster region (colored points), with galaxies from SDSS DR15 with 5000km s−1 < cz < 9000km s−1 plotted as small grey circles. Our sample is subdivided into HI detections of UDGs (blue stars), HI detections of foreground dwarf galaxies (green squares), and HI non-detections (red circles). The orange open circle is centered on the Coma Cluster (α = 12h59m48.7s; δ = 27◦5805000, Kadowaki et al., 2017) and has a radius of 3 Mpc that ∼ represents the virial radius of the Coma Cluster (Kubo et al., 2007). 4.4. OBSERVATIONS AND DATA REDUCTION 83 z − g . col.(14): and 1 Y Y r − − g 98) 18) H M 19 19 19 Ref HI Z19 Z19 Z19 Z19 Z19 Z Z Z = 50 km s 20( 20( K K V ∆ 39 58 69 38 79 85 41 73 84 53 50 ...... σ 55 0 0 4 0 2 0 5 0 346 0 9 0 0 0 2 1 ...... Int. Time 04 0 0204 0 02 0 0 04 0 0203 0 0 04 2 03 0 02 1 ...... 0 0 0 0 0 0 0 0 0 0 ± ± ± ± ± ± ± ± ± ± 71 73 52 56 96 92 65 70 76 67 ...... 3 0 1 0 5 0 1 0 1 0 2 0 1 0 1 0 2 0 7 0 ± ± ± ± ± ± ± ± ± ± 4 29 67 24 − − − − 01 01 01 65 02 9 01 77 01 01 47 01 84 03 0 01 ...... 0 0 0 0 0 0 0 0 0 0 ± ± ± ± ± ± ± ± ± ± b/a θ n 83 70 77 82 63 90 57 63 64 60 ...... 9 0 2 0 7 0 4 0 3 0 6 0 7 0 2 0 3 0 1 0 ...... 0 0 0 0 0 0 0 0 0 0 ± ± ± ± ± ± ± ± ± ± eff 2 2 1 7 1 4 0 3 1 5 ...... 04 14 03 9 05 15 07 16 02 10 06 17 08 7 03 12 04 6 02 14 ...... z r 0 0 0 0 0 0 0 0 0 0 band apparent magnitude and central surface brightness. cols.(6) and (7): index in GALFIT model of UDG candidate. col.(12): Total effective GBT integration time, − ± ± ± ± ± ± ± ± ± ± − g 20 87 60 11 50 88 03 68 91 82 ...... Sérsic 02 0 01 0 03 0 05 1 01 0 02 0 01 1 02 0 02 0 04 0 ...... r g 0 0 0 0 0 0 0 0 0 0 − ± ± ± ± ± ± ± ± ± ± g (mag) (mag) (arcsec) (deg) (hours) (mJy) Det? 32 54 46 66 34 62 64 54 64 59 ...... 06 0 13 0 14 0 05 0 10 0 12 0 09 0 06 0 06 0 11 0 ) ...... 2 0 0 0 0 0 0 0 0 0 0 ,g ± ± ± ± ± ± ± ± ± ± 0 mag µ 95 99 71 38 30 27 55 00 83 25 arcsec ...... ( Table 4.1: Target UDG Candidate Properties 01 25 01 24 01 25 04 25 01 24 01 24 01 25 01 25 01 25 03 25 ...... 0 0 0 0 0 0 0 0 0 0 g ± ± ± ± ± ± ± ± ± ± m 06 98 49 04 44 48 77 01 26 00 ...... 18 18 18 20 18 19 18 19 19 19 H:M:S D:M:S (mag) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Name RA Dec SMDG1220212+290831 12:20:21.0 29:08:34 SMDG1221086+292920 12:21:17.0* 29:29:21 SMDG1221235+303643 12:21:23.4SMDG1221401+284346 30:36:44 12:21:40.0SMDG1221497+283111 28:43:47 12:21:49.7 28:31:12 SMDG1103517+284120 11:03:51.7 28:41:20 SMDG1217378+283519 12:17:38.0SMDG1217443+332043 28:35:20 12:17:44.2SMDG1217451+281724 33:20:44 12:17:45.0 28:17:25 SMDG1220188+280131 12:20:19.0 28:01:34 The first 10 rows of thiscol.(1): table are Adopted shownindicate here. SMUDGes an The offset UDG full (in table candidate RA is name. and/or available online Dec) cols.(2) in in machine and the readable GBT (3): format. pointing position. J2000 cols.(4) position and (5): of optical centroid, which corresponds to our GBT LOS. RA values with an asterisk (*) colors. col. (8)-(11):including Best-fitting the effective ON+OFF radius,Reference positions axial from and subtracting which ratio,HI any UDG position detection? time candidate angle, lost is and due selected, to alternative RFI. references col.(13): in Representative parentheses. RMS noise S97 of = theSchombert spectrum et at al. ( 1997 ); a M98 velocity = resolutionMartin ( 1998 ); of H18 = Haynes et al. ( 2018 ). col.(15): 4.4. OBSERVATIONS AND DATA REDUCTION 84

Our observational setup and data reduction procedure are similar to those used in

Karunakaran et al.(2020), which we brieﬂy outline here. We used the L-band receiver and the Versatile GBT Astronomical Spectrometer (VEGAS) with a spectral resolution of

3.1 kHz and a wide bandpass of 100 MHz which allows for the detection of HI emission

−1 lines out to VHelio 14000 km s . We estimate the integration times for our targets using ∼ MHI M −1 mg to reach a gas richness of 1 with S/N = 5 in a single 50 km s channel. Lg ∼ L Gas richness is a distance-independent quantity since both MHI and Lg scale with distance squared. Therefore, a single spectrum allows us to search for an HI reservoir in our targets anywhere within the wide bandpass.

The data were reduced using the standard GBTIDL3 procedure getps. We remove narrow-band and broadband radio frequency interference before smoothing our spectra to our desired resolutions, following the same procedure as Karunakaran et al.(2020). Further- more, we scale the fluxes in our final spectra up by 20% to account for the systematic offset in the GBT noise diode calibration values reported by Goddy et al.(2020). The RMS noise,

−1 σ50, for each spectrum at ∆V = 50 km s resolution is given in column 13 of Table 4.1.

We examined the calibrated, RFI-excised spectra by-eye after smoothing to multiple resolutions from 5 50 km s−1 for statistically signiﬁcant emission. We detect HI emission − along the LOS to 18 UDG candidates (column 15 of Table 4.1). We show their spectra in

Figure 4.2 at ∆V given in Tables 4.2 and 4.3, which also lists other properties we have derived from these HI detections. We ﬁnd no signiﬁcant HI emission associated with the 52

lim remaining targets and place stringent 5σ upper limits on HI mass, MHI , and gas richness, lim MHI /Lg, which are listed in Table 4.4. 3http://gbtidl.nrao.edu/ 4.5. RESULTS 85

4.5 Results

4.5.1 Properties of HI Detections

We detect HI along the LOS to 18 UDG candidates, and their spectra are shown in Figure

4.2. At our observing frequency of 1.4 GHz, the GBT beam (FWHM 9.10) response ∼ ∼ is well understood down to 30dB (e.g., Spekkens et al., 2013). We therefore search ≈ − through NED4 and the DESI Legacy Imaging Survey Sky Viewer5 for objects within 300 of the LOS that may present themselves as gas-rich interlopers in our spectra for all of our targets. We ﬁnd no such interlopers for any of our HI detections, and conclude they are the

HI counterparts to the corresponding UDG candidates.

We derive distance-independent quantities from the spectra (systemic velocity, Vsys, and velocity width, W50) as described in Karunakaran et al.(2020) and briefly outline the method here. Using a first-order polynomial fit to each edge of the HI profile between 15% and 85% of the peak flux value, we find the velocities corresponding to the 50% flux value. Their mean corresponds to Vsys (column 4 of Tables 4.2 and 4.3) and the difference corresponds to

W50. The latter is corrected for instrumental and cosmological redshift broadening following

Springob et al.(2005) to produce W50,c (column 5 of Tables 4.2 and 4.3). We assume an uncertainty of 50% for the instrumental broadening correction, which dominates the uncertainties on Vsys and W50,c. We note that we are conducting signal recovery simulations, similar to Springob et al.(2005) but tailored to UDG-like HI profile shapes, to more accurately understand how instrumental effects at the GBT affect our HI detections in the full survey.

4The NASA/IPAC Extragalactic Database (NED) is operated by the Jet Propulsion Laboratory, Cali- fornia Institute of Technology, under contract with the National Aeronautics and Space Administration. 5http://legacysurvey.org/viewer 4.5. RESULTS 86

1220188+280131 1225185+270858 5.0 1226040+241802 UDG 2 UDG UDG 10 2.5 0 0 0.0 2 1500 3000 − 5300 6400 11000 12000 1230359+273311 1241424+273353 1248019+261236 UDG 10 UDG 4 UDG 2 5 2 0 0 0 5 6000 7500 − 7250 8250 5400 6400 1301005+210355 5.0 1312223+312320 1315427+311846 UDG UDG UDG 2 10 2.5 5 0 0.0 0 2.5 6000 8000 − 6800 8300 7000 8000 1103517+284120 1223451+283549 1231329+232916 20 Dwarf Dwarf Dwarf 4 10 Flux (mJy) 10 2 5 0 0 0 200 1200 2000 3000 1000 3000 1239050+323016 30 1240017+261919 1253571+291500 5.0 Dwarf Dwarf Dwarf 20 2 2.5 10 1 0.0 0 0 2.5 1 − 1000 2000 200 1200 500 2000 −15 1255412+191221 1306148+275941 1313188+312452 Dwarf Dwarf Dwarf 2 10 100 5 0 0 0 200 1200 2000 3000 500 1500 VHelio(km/s)

Figure 4.2: HI detections along the LOS to UDG candidates in our sample. The first 9 panels show targets that satisfy the UDG size criterion of Reff > 1.5kpc given their redshifts (confirming them as UDGs), while the last 9 panels show targets which do not (confirming them as foreground dwarfs). Target names and classification (UDG or Dwarf) are in the top-right corner of each panel. The black dotted line in each panel represents 0 mJy. The spectral resolutions ∆V of the plotted spectra and the derived properties of the HI detections are in Tables 4.2 and 4.3 for UDGs and foreground dwarfs, respectively. 4.5. RESULTS 87

Distances are required to conﬁrm candidates as true UDGs. Using Vsys and the Hubble-

Lemaître Law, we estimate kinematic distances for all of our HI detections and adopt a

distance uncertainty of 5 Mpc (Leisman et al., 2017; Spekkens and Karunakaran, 2018).

Interestingly, detections are almost equally split between foreground (DHI < 40 Mpc) and

background objects (DHI > 80 Mpc): this emphasizes how gas-rich diﬀuse objects at dif-

ferent distances can look similar on the sky (see also Figures 4.6 and 4.8), an issue that

we discuss further in Section 4.6.1. Based on these distances and the angular sizes listed in

−2 Table 4.1, we conﬁrm 9 new UDGs with Reff > 1.5 kpc and µ0,g & 24 mag arcsec , and give their HI properties in Table 4.2. The remaining detections are dwarfs in the foreground

of Coma; their derived HI properties are in Table 4.3. To the best of our knowledge, the

UDGs in Table 4.2 represent the largest sample of optically-selected UDGs with follow-up

HI detections reported so far.

In the left panel of Figure 4.3, we compare the distribution of W50,c for our HI-conﬁrmed

UDGs (orange) to those of the HI-bearing ultra-diﬀuse sources (HUDs) samples, HUDs-B

(green) and HUDs-R (purple), from Leisman et al.(2017). The HUDs-B and -R samples

are distinguished by their “broad" and “restrictive" optical selection criteria (see Leisman

et al. 2017 for more details) with the latter sample using the same criteria used in this

work. We also include galaxies in the ALFALFA α.40 catalog (Haynes et al., 2011). Our

HI-conﬁrmed UDGs span a broad range in W50,c and are generally more consistent with

the HUDs samples than galaxies in ALFALFA. During our literature search for possible HI

interlopers we discovered that 5 of our 18 HI detections were previously reported as gas-rich

objects. Of these previously detected objects, 1 is a UDG (SMDG1220188+280132) that has

been detected by ALFALFA. It was not included in the HUDs sample, likely due to their

distance criteria (Leisman et al., 2017). Therefore, we present SMDG1220188+280132 as a

UDG here for the ﬁrst time.

We calculate the HI flux, SHI = SδV , by integrating over the line profile, where ´ uncertainties stem mainly from the noise statistics of the profile (Springob et al., 2005) and 4.5. RESULTS 88

a 2% noise diode uncertainty (van Zee et al., 1997). We use these ﬂuxes and our kinematic distances to determine HI masses, MHI , using the standard equation for an optically thin gas (Haynes and Giovanelli, 1984):

5 2 MHI = 2.356 10 (DHI ) SHI M , (4.1) ×

−1 where the distance, DHI , is in Mpc and SHI is in Jy km s . HI masses are listed in column

8 of Tables 4.2 and 4.3. Uncertainties are determined following the methods of Springob et al.(2005) to which distance uncertainties are added in quadrature.

We calculate stellar masses, M∗, for detections (column 9 of Tables 4.2 and 4.3) using

mg and g r from Table 4.1 in the relations of Zhang et al.(2017) and assuming DHI , prop- − agating photometric and distance uncertainties along with those reported on the relations.

Finally, we estimate baryonic masses as Mbary = 1.33MHI + M∗ (column 10 of Tables 4.2

and 4.3).

In the right panel of Figure 4.3, we show the HI-conﬁrmed UDGs in the MHI M∗ plane, − along with the HUDs samples (-B: green circles and -R: purple squares) and galaxies from

the α.40 catalog with SDSS and GALEX coverage from Huang et al.(2012a, grey circles).

We ﬁnd that our UDGs are broadly consistent with both the HUDs and α.40 samples,

although Figure 4.3 illustrates how the HUDs sample as a whole may be more gas-rich

(mean MHI /M∗ 15, Leisman et al., 2017) compared to ours (mean MHI /M∗ 5). This ∼ ∼ may point to a diﬀerence in UDG samples drawn from HI vs. optical searches, although

their selection functions need to be understood before intrinsic population diﬀerences can

be quantiﬁed. 4.5. RESULTS 89 30 18 17 17 20 17 28 15 41 ...... 0 0 0 0 0 0 0 0 0 from Table 4.1 , r eff ± ± ± ± ± ± ± ± ± R − 00 63 48 11 81 76 57 00 85 ...... g and ]) (kpc) 15 2 16 2 08 4 10 3 10 3 09 2 13 3 13 3 08 5 ......

g 0 0 0 0 0 0 0 0 0 bary . col.(11): Eﬀective radius in M m in col.(2). col.(4): Heliocentric ± ± ± ± ± ± ± ± ± ∗ M V M 37 67 41 07 09 82 76 03 32 ...... ∆ + HI ]) (log[ 24 8 21 8 20 9 20 9 20 9 21 8 20 8 20 9 20 9 ) log ...... ∗

0 0 0 0 0 0 0 0 0 M M M ± ± ± ± ± ± ± ± ± 33 . 1 51 01 57 01 36 01 22 18 44 ...... ) log( ]) (log[ 15 7 19 8 08 8 10 8 11 8 10 8 16 8 15 8 08 8 ......

0 0 0 0 0 0 0 0 0 HI M ± ± ± ± ± ± ± ± ± M 19 44 22 91 87 63 49 84 14 in col.(7)...... log( HI 5 9 9 8 7 8 0 8 9 9 . We adopt distance uncertainties of 5 Mpc. col.(8): Logarithm 6 8 1 8 1 8 3 8 . . . . . D . . . . 1 HI − D ) (Mpc) (log[ Mpc 1 05 86 05 100 09 107 08 106 10 32 07 84 05 165 08 97 06 110 1 ...... − − 0 0 0 0 0 0 0 0 0 HI ± ± ± ± ± ± ± ± ± S km s 61 17 26 37 26 24 13 26 51 ...... = 70 km s ) (Jy from Table 4.1 and 5 0 1 0 2 0 7 0 4 0 2 0 7 0 5 0 4 0 8 0 ,c − H ± ± ± ± ± ± ± ± ± 50 eff r W km s and in col.(7). col.(9): Logarithm of stellar mass calculated using 3 34 )( 2 32 5 63 3 102 2 17 5 32 3 31 3 42 6 13 sys HI 1 V ± D − ± ± ± ± ± ± ± ± sys V km s V 86 2283 69 5888 66 11585 56 6794 52 7766 51 6043 53 7051 05 7487 99 7486 ...... ∆ physical units using Table 4.2: Properties of UDG with HI detections in col.(6) and ) (mJy) ( 1 HI − S V σ 10 1 20 0 15 0 25 0 10 1 25 0 25 0 20 1 25 0 ∆ km s ( (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) Name SMDG1220188+280131 SMDG1225185+270858 SMDG1226040+241802 SMDG1230359+273311 SMDG1241424+273353 SMDG1248019+261236 SMDG1301005+210355 SMDG1312223+312320 SMDG1315427+311846 in col.(7), and the corresponding relation from Zhang et al. ( 2017). col.(10): Logarithm of baryonic mass, HI systemic velocity. col.(5): Velocity width of the HI detection, corrected for cosmological redshift and instrumental broadening. col.(6): Integrated HI ﬂux. D of HI mass calculated from Eq. 4.1 using col.(7): Distance estimated using the Hubble-Lemaître Law, col.(2): Velocity resolution of spectrum used to compute HI properties (see Figure 4.2 ). col.(3): RMS noise of spectrum at 4.5. RESULTS 90 24 23 15 62 17 54 02 02 40 ...... 0 0 0 0 0 0 0 0 0 eff ± ± ± ± ± ± ± ± ± R 30 22 75 28 24 77 42 78 75 ...... ]) (kpc) 61 0 72 0 16 1 39 1 30 0 50 0 14 0 67 0 46 0 ......

0 0 0 0 0 0 0 0 0 bary M ± ± ± ± ± ± ± ± ± M 84 53 56 72 99 14 68 71 49 ...... ]) (log[ 42 7 3553 7 6 70 6 23 8 64 6 75 7 50 7 23 7 ) log ...... ∗

0 0 0 0 0 0 0 0 0 M M ± ± ± ± ± ± ± ± ± 68 79 64 19 02 38 58 94 14 ...... ) log( ]) (log[ 2950 6 5 68 5 15 7 61 6 72 6 45 6 17 6 39 7 ......

0 0 0 0 0 0 0 0 0 HI M ± ± ± ± ± ± ± ± ± M 54 38 85 45 97 57 32 67 19 ...... log( 6 7 5 7 1 7 0 7 8 6 5 6 2 6 0 7 5 7 ...... HI D ) (Mpc) (log[ 1 0704 15 8 10 6 05 7 05 34 21 6 13 9 04 36 10 11 ...... − 0 0 0 0 0 0 0 0 0 HI ± ± ± ± ± ± ± ± ± S km s 19 44 72 23 34 37 97 15 50 ...... ) (Jy 1 5 0 2 4 4 0 8 0 2 0 3 0 5 0 3 0 8 0 ,c − ± ± ± ± ± ± ± ± 50 ± W km s )( 3 42 6 8 4 59 5 13 1 32 2 13 4 67 1 19 2 26 1 − ± ± ± ± ± ± ± ± ± Table 4.3: HI Properties of Dwarfs sys V km s All parameters have the same deﬁnitions as in Table 4.2 . V 45 502 76 420 55 2559 25 802 82 613 82 1060 24 452 51 2377 46 668 ...... ∆ ) (mJy) ( 1 − V σ 5 4 10 0 10 2 25 0 15 0 25 1 25 0 20 0 10 2 ∆ km s ( (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) Name SMDG1313188+312452 SMDG1239050+323016 SMDG1231329+232916 SMDG1240017+261919 SMDG1253571+291500 SMDG1255412+191221 SMDG1223451+283549 SMDG1103517+284120 SMDG1306148+275941 4.5. RESULTS 91

This Work 11 0.20 HUDs-B HUDs-R 10 ) 0.15 α.40, H12

/M 9 HI

0.10 M 8 log(

Relative Frequency 0.05 7

0.00 6 1 2 3 6 8 10 log(W50, km/s) log(M /M ) ∗

Figure 4.3: Comparison of derived properties between our HI-conﬁrmed UDGs and other similar samples. Left : Distribution of W50,c for UDGs in our sample (orange), the HUDs-B and -R samples (purple and green, respectively), and galaxies from the α.40 catalog with SDSS and GALEX coverage (grey). Right : MHI M∗ relation for the same samples as in − the left panel.

In Figure 4.4, we show the relationship between gas-richness, MHI /M∗, and size, Reff ,

for our HI-conﬁrmed UDGs and foreground dwarfs (ﬁlled stars). We also include the 6

UDGs from Mancera Piña et al.(2020) and 5 UDGs around Hickson Compact Groups

from Spekkens and Karunakaran(2018) (open symbols). For consistency across samples we

have calculated Reff from the exponential scale-lengths reported by Mancera Piña et al.

(2020). Because of the large systematic diﬀerences between diﬀerent color-mass-to-light-

ratio prescriptions (Roediger and Courteau, 2015), we have also re-calculated stellar masses

for the 5 UDGs followed up by Spekkens and Karunakaran(2018) using the photometry of

Román and Trujillo(2017) and the Zhang et al.(2017) relations. We also propagated the

5 Mpc distance uncertainties into the errorbars on Reff for all samples. The colors of the

symbols in Figure 4.4 represent the stellar masses of the object. There is some evidence that

larger UDGs are more gas-rich within each stellar mass bin but little evidence for a similar

trend among the foreground dwarfs; we discuss possible implications of this in Section 4.6.2. 4.5. RESULTS 92

log(M /M ) < 6.6 7.5 < log(M /M ) 8.2 6.6 < ∗log(M /M ) 7.2 log(M /M )∗> 8 .2 ≤ ∗ ≤ ∗ 1.5 ) ∗ 1.0 /M HI M 0.5 log(

0.0 This Work HUDS HCG UDGs Dwarfs UDGs 0.75 0.25 0.25 0.75 − − log(Reff/kpc)

Figure 4.4: Gas-richness as a function of size for our 9 HI-conﬁrmed UDGs and 9 foreground dwarfs as ﬁlled stars. The gray dashed line shows the Reff = 1.5 kpc size boundary between dwarfs and UDGs. We also include the 5 UDGs around Hickson Compact groups from Spekkens and Karunakaran(2018, open circles) and 6 UDGs from Mancera Piña et al. (2020, open diamonds). The colors of the symbols represent the stellar mass bin of the objects.

4.5.2 HI Non-detections

We ﬁnd no statistically signiﬁcant HI signals along the LOS to 52/70 targeted UDG candidates that can be attributed to these objects. We smooth their spectra to ∆V = 50 km s−1

and list their representative RMS noise, σ50, in column 13 of Table 4.1. We modify equation

4.1 to place stringent, 5σ HI-mass upper limits

lim 7 2 M = 5.89 10 (Dlim) σ50 M , (4.2) HI × 4.5. RESULTS 93

where Dlim is the adopted distance in Mpc. In most cases we assume the Coma Cluster distance of Dlim = DComa = 100 Mpc, aside from a few exceptions described below. We also set (distance-independent) upper limits on the ratio of HI-mass to g band luminosity, − lim MHI /Lg, and list all of the calculated properties for non-detections in Table 4.4. We brieﬂy highlight a few of the HI non-detections in our sample. SMDG1221577+281436 was reported as the marginal HI detection of a nearby, gas-rich dwarf galaxy with VHelio =

450 8 km s−1 in Huchtmeier et al.(2009, d1221+2814). However, when we smooth our spec- ± tra to match their velocity resolution we see no signal despite our deeper data. SMDG1253151

+274115 (ﬁrst reported as DF30 in van Dokkum et al. 2015a) and SMDG1251013+274753

were conﬁrmed as UDGs via optical spectroscopy in Kadowaki et al.(2017) and Kadowaki

−1 −1 et al., in prep, with Vopt = 7316 81 km s and Vopt = 6118 45 km s , respectively. The ± ± former was conﬁrmed as a Coma Cluster member and we use Dlim = 100 Mpc to estimate HI

properties. The latter was conﬁrmed to lie outside the Coma Cluster, and therefore we esti-

mate its distance using Vopt and the Hubble-Lemaître Law to be Dlim = 87 Mpc. In addition,

−1 Kadowaki et al. (in prep) ﬁnd velocities for SMDG1217378+283519 (Vopt = 493 69 km s ) ± −1 and SMDG1221086+292920 (Vopt = 1024 66 km s ) that place them well in the foreground ± lim of Coma, and we use the corresponding Dlim to compute MHI . SMDG1302417+215954 (IC 4107) has previously been reported as an HI non-detection (Schombert et al., 1992). One

6 −1 SDSS spectrum of this object classiﬁes it as a star with Vopt = 267 km s (Kim et al., 2014),

7 −1 while another classiﬁes it as a QSO with V > 100, 000 km s . We adopt Dlim = 3.8 Mpc

using the lower SDSS velocity, consistent with both its morphology and the Karachentsev

et al.(2014c) association of this object with the NGC 4826 group.

In Figure 4.5, we show MHI /Lg for our HI detections of UDGs (blue stars) and fore-

lim ground dwarfs (green squares), and MHI /Lg for our HI non-detections (red downward arrows) as a function of g r (left panel) and g z (right panel). The vertical dashed lines − − 6http://cas.sdss.org/dr7/en/get/specById.asp?id=746142786461368320 7http://skyserver.sdss.org/dr12/en/get/SpecById.ashx?id=2983699425022470144 4.5. RESULTS 94

in each panel show the median colors of the follow-up sample as a whole: g r = 0.53 − and g z = 0.79. We note that our upper limits are generally higher than the MHI /Lg = − 1M /L used to estimate the required integration times. There are three potential reasons

for this reduction in sensitivity: integrations/scans ﬂagged due to RFI, 20% calibration ad-

justment, and/or noisier than expected data. By and large, our HI detections have colors

that are bluer than the HI non-detections but the scatter is large (see also Figures 4.6- 4.8);

in the left panel, we also show the median g r = 0.35 of the entire HUDs sample as the − vertical dashed-dotted line. Several of our HI detections, including 6/9 UDGs, hover around

this line and the vast majority of our non-detections lie on its redder side. We discuss the

diﬀerences between the optical properties of our HI-conﬁrmed UDGs, foreground dwarfs,

and HI non-detections in Section 4.6.1.

1.0 1.0 ))

0.5 0.5 /L

M ( g

/L 0.0 0.0 HI M

log( 0.5 0.5 HI Non-detections − − HI Detections, Dwarfs HI Detections, UDGs 0.0 0.2 0.4 0.6 0.8 0.00 0.25 0.50 0.75 1.00 1.25 g r(mag) g z(mag) − −

Figure 4.5: MHI /Lg (blue stars and green squares for HI-conﬁrmed UDGs and foreground lim dwarfs, respectively) and MHI /Lg (red downward arrows for non-detections) as a function of g r (left) and g z (right) color for our sample. The dashed vertical lines in each panel − − show the median g r = 0.53 and g z = 0.79 colors for our sample. For comparison, we − − show the median g r = 0.35 color of the HUDs sample (Leisman et al., 2017) in the left − panel with a vertical dash-dotted line. 4.5. RESULTS 95

Table 4.4: HI Properties of Non-detections

lim lim Name Dlim log(MHI )(MHI /Lg)

(Mpc) (log[M ]) (M /L )

(1) (2) (3) (4)

SMDG1217378+283519 7a 6.22 1.27

SMDG1217443+332043 100 8.61 0.96

SMDG1217451+281724 100 8.35 2.22

SMDG1220212+290831 100 8.39 1.43

SMDG1221086+292920* 14.6a 6.96 1.32

SMDG1221235+303643 100 8.69 1.89

SMDG1221401+284346 100 8.49 1.50

SMDG1221497+283111 100 8.7 1.92

SMDG1221577+281436 100 8.8 1.06

SMDG1223448+295949 100 8.54 1.58

SMDG1224082+280544 100 8.41 0.82

SMDG1224166+291506 100 8.18 1.49

SMDG1225265+311646* 100 8.59 1.19

SMDG1231418+264433 100 8.3 2.49

SMDG1237277+333048 100 8.33 1.19

SMDG1239267+274736 100 8.41 1.81

SMDG1239503+244949 100 8.53 1.36

SMDG1240119+251447 100 8.4 2.76

SMDG1247233+180140 100 8.69 1.02

SMDG1249413+270645 100 8.27 1.49

SMDG1251013+274753* 87a 8.5 1.86 4.5. RESULTS 96

Table 4.4 – continued from previous page

lim lim Name Dlim log(MHI )(MHI /Lg)

(Mpc) (log[M ]) (M /L )

(1) (2) (3) (4)

SMDG1253048+253121 100 8.44 1.97

SMDG1253151+274115* 100b 8.28 1.63

SMDG1253489+273934 100 8.79 2.86

SMDG1254252+194332 100 8.61 0.83

SMDG1254556+285846 100 8.45 1.89

SMDG1255336+213035 100 8.12 1.59

SMDG1307464+291230 100 8.44 2.51

SMDG1308296+271354 100 8.35 1.98

SMDG1322561+314804 100 8.61 2.06

SMDG1226306+220532 100 8.63 0.95

SMDG1231070+253508* 100 8.62 2.28

SMDG1232244+274043 100 8.22 1.46

SMDG1233516+234545 100 8.49 0.73

SMDG1234503+293313 100 8.49 1.34

SMDG1235065+263342 100 8.27 1.57

SMDG1240490+254406 100 8.45 3.50

SMDG1241097+221223 100 8.33 1.92

SMDG1245022+230956 100 8.28 1.44

SMDG1246029+255724 100 8.51 1.54

SMDG1248202+183824 100 8.7 0.37

SMDG1249353+253106 100 8.56 0.72

SMDG1251291+284433* 100 8.29 2.80 4.6. DISCUSSION 97

Table 4.4 – continued from previous page

lim lim Name Dlim log(MHI )(MHI /Lg)

(Mpc) (log[M ]) (M /L )

(1) (2) (3) (4)

SMDG1251337+314240 100 8.44 1.12

SMDG1251371+244922 100 8.46 1.28

SMDG1252056+221556 100 8.2 1.71

SMDG1252402+262602* 100 8.48 0.85

SMDG1302417+215954 3.8c 5.94 0.25

SMDG1306158+273459 100 8.35 2.32

SMDG1312226+195525 100 8.27 1.68

SMDG1322538+220445* 100 8.4 1.72

SMDG1333509+275006 100 8.38 1.72 aKadowaki et al., in prep,bKadowaki et al.(2017), cKim et al.(2014) col. (2): Adopted distance in Eq.4.2; see text for details. col. (3): 5σ upper limit on MHI

calculated from Eq.4.2 using Dlim from col. (2) and σ50 from Table 4.1. col.(4): Upper

limit on the gas richness (which is distance independent).

4.6 Discussion

With our pilot sample of HI-conﬁrmed UDGs, foreground dwarfs, and HI non-detections in

hand, we provide some initial insight on three main questions our survey aims to answer: 1.

Are there optical features that distinguish bona-ﬁde gas-rich UDGs from foreground dwarfs

or HI non-detections among UDG candidates? 2. What constraints, if any, do our HI-

conﬁrmed UDGs place on formation mechanisms? 3. How unusual are UDGs in the context

of local galaxy scaling relations? We address these questions in Sections 4.6.1, 4.6.2, and 4.6. DISCUSSION 98

4.6.3, respectively.

4.6.1 Comparing UDGs with HI Detections and Non-detections

Our follow-up HI observations of 70 SMUDGes UDG candidates have revealed 9 gas-rich

UDGs and 9 gas-rich foreground dwarf galaxies, while the remaining 52 targets were not detected in HI. In this section, we explore diﬀerences between the environment and optical/NUV properties of these subsamples both to improve our detection eﬃciency in the full survey as well as to constrain the properties of HI-rich and HI-poor objects in the LSB regime.

We ﬁrst revisit the spatial distribution of the follow-up targets shown in Figure 4.1.

The projected distribution of our sample spans both high and low density regions around

Coma, with no obvious diﬀerence in location relative to the large-scale ﬁlamentary structure

(grey circles) between HI detections (blue stars and green squares) and non-detections (red circles). This qualitatively suggests that there is no strong correlation between HI content and projected environment, implying that sky location is not a good predictor of gas richness among pilot sample galaxies.

Quantitatively, we ﬁnd that none of the HI-conﬁrmed UDGs are likely to be gravitation- ally bound to the Coma Cluster based on their redshifts and projected spatial separations.

Furthermore, only one of these objects (SMDG1248019+261236) has at least one massive

−1 companion (Mg < 19 mag) that projects within 300 kpc and within 500 km s (Kad- − ± owaki et al., in prep). While not obvious from Figure 4.1, our HI-conﬁrmed UDGs reside in sparse environments. These ﬁndings are generally consistent with previous work that has investigated the environmental dependence of gas content (Brown et al., 2017).

We next investigate whether or not discernible NUV emission in archival GALEX imaging predicts a detectable HI reservoir among UDG candidates. The vast majority of pilot survey targets that are in the GALEX footprint do not have detectable NUV emission, which is commensurate with the ﬁndings of Singh et al.(2019) for the broader SMUDGes sample. 4.6. DISCUSSION 99

This is also the case for our HI detections with GALEX All-sky Imaging Survey (AIS;

Morrissey et al., 2007; Bianchi, 2009) coverage, raising the possibility that AIS-depth NUV imaging is not suﬃcient to detect ongoing star formation in UDGs. We therefore examine the subset of pilot survey targets with GALEX NUV exposures of at least 1000 seconds, i.e.,

Medium Imaging Survey (MIS) depth or 5-10 times deeper than the AIS. Of the 32 pilot ∼ survey targets in this category, 14 have discernible GALEX emission. All of these objects for which our HI spectra are sensitive to at least MHI /Lg = 2 M /L across the observed

band have been detected in HI, while many of the objects for which deep GALEX images

lim reveal no NUV emission are HI non-detections with MHI /Lg < 1.5 M /L . This suggests that MIS-depth NUV imaging is a good predictor of gas richness among SMUDGes UDG

candidates.

Finally, we compare the DECaLS optical morphologies of the HI-conﬁrmed UDGs, the

UDG candidates that we did not detect in HI, and the HI-detected foreground dwarfs.

Figures 4.6-4.8 show color and grayscale grz cutout pairs for these three subsets of our

sample, where the adjusted contrast and brightness of the color image highlights the brighter

emission in the object and the histogram equalization of the grayscale image highlights

fainter emission. In all panels, the dashed, white ellipses have the disk geometry and semi- a major axis, 2 = reff , of the best-fitting GALFIT models reported in Table 4.1. We note that Figures 4.6 and 4.8 show all of the HI-confirmed UDGs and foreground dwarfs in our sample, while Figure 4.7 shows images of a subset of the 52 HI non-detections with similar reff and mg to the HI-confirmed UDGs.

Figures 4.6 and 4.7 demonstrate that on the whole, the HI-detected UDGs are bluer

than the UDG candidates that we do not detect, although as illustrated in Figure 4.5

the scatter in color is large (c.f. SMDG1301005+210355 which we do detect in HI, and

SMDG1223448+295949 which we do not). This is consistent with the clear trends seen at

higher surface brightnesses (Huang et al., 2012a; Catinella et al., 2012; Brown et al., 2015)

as well as in other LSB studies (Leisman et al., 2017; Greco et al., 2018; Prole et al., 2019b), 4.6. DISCUSSION 100

suggesting that star formation proceeds similarly in high and low surface brightness galaxies

(Bell et al., 2000).

Figures 4.6 and 4.7 also illustrate that the HI-detected UDGs are more irregular in morphology both within and beyond reff than the undetected UDG candidates, although there is some scatter (c.f. SMDG1225185+270858 which we do detect in HI, and SMDG1253151+274115 which we do not). On the other hand, the combination of DECaLS-depth color and morphology does appear to predict gas richness: blue and irregular objects in our pilot sample are almost invariably gas-rich, while red and smooth objects are invariably gas-poor. The eﬃciency of future HI follow-up UDG campaigns can therefore be increased relative to the statistics presented here by preferentially targeting candidates that are both blue and irregular.

Do our HI-confirmed UDGs differ in optical morphology from our gas-rich foreground dwarfs? Comparing Figures 4.6 and 4.8 reveals that, among gas-rich objects, the foreground dwarfs tend to have larger angular sizes than the confirmed UDGs, consistent with Z19’s hypothesis using a clustering analysis. The bluest gas-rich objects that we detect are also foreground dwarfs and not confirmed UDGs. While some stars in the very nearby dwarf

SMDG1255412+191221 begin to appear resolved in the DECaLS imaging, we find no clear difference in optical morphology between bona-fide UDGs and foreground dwarfs in the pilot sample, making the two difficult to distinguish among follow-up targets. Blue foreground dwarfs are therefore an important potential contaminant among gas-rich UDG candidates identified by their optical colors and morphologies alone. Distance information is required to identify UDGs in the field. 4.6. DISCUSSION 101

Figure 4.6: 55“ 55“ color and grayscale grz image cutouts of HI-detected UDGs shown in × pairs with the color image on the left and the grayscale image on the right. The adjusted contrast and brightness of the color images highlights brighter emission in each object, while the histogram equalization of the grayscale images highlights the lower surface brightness emission. In all panels, the dashed, white ellipses have the disk geometry and semi-major a axis, 2 = reff , of the best-ﬁtting GALFIT models reported in Table 4.1. The object’s color from Table 4.1 is in the top-right corner of each image pair, and a scale bar that is 1 kpc across at the UDG distance is in the bottom-right corner. For a subset of the objects, we also overlay red ellipses corresponding to the disk geometry of GALFIT models with lower inclinations as detailed in Section 4.6.3. The inclinations of the corresponding disk, computed using Equation 4.4, are in the top-left corner of each image pair. 4.6. DISCUSSION 102

Figure 4.7: Same as Figure 4.6, but for HI non-detections. 4.6. DISCUSSION 103

Figure 4.8: Same as Figure 4.6, but for HI-detected foreground dwarf galaxies and the scale bar represents 200 pc. 4.6. DISCUSSION 104

4.6.2 Constraining Formation Mechanisms

The stellar masses and velocity widths of our HI-conﬁrmed UDGs are commensurate with them being dwarf galaxies, in line with other estimates for UDGs in a variety of environments

(e.g., Sifón et al., 2018; Pandya et al., 2018; Zaritsky et al., 2019; Barbosa et al., 2020). How

UDG-like field dwarfs could form is an active area of research (see Section 4.2), and the small size of the HI-confirmed UDG sample from this pilot survey is too small for quantitative comparisons with theory. Nonetheless, we briefly consider the gas richnesses and sizes of the

HI-conﬁrmed UDGs in the context of formation model predictions.

The star-formation feedback model presented by di Cintio et al.(2017) predicts that

UDGs in the ﬁeld today have gas richnesses that scale with their sizes at ﬁxed stellar mass.

As shown in Figure 4.4, we ﬁnd evidence for a trend between MHI /M∗ and Reff when

the gas-rich UDGs are subdivided into two stellar mass bins. This trend persists when the

gas-rich UDG samples from Spekkens and Karunakaran(2018, who noted this trend in their

smaller sample) and Mancera Piña et al.(2020) are also considered, but it is not evident in

the foreground dwarf sample also plotted in Figure 4.4. The correlation between gas richness

and size for UDGs is qualitatively consistent with the predictions of di Cintio et al.(2017),

although a similar trend may also emerge from other UDG formation scenarios.

It is also possible that the correlations between gas richness and size exist in the broader

galaxy population, and therefore that the trends in Figure 4.4 do not constrain UDG forma-

tion mechanisms at all. That the foreground dwarfs in our sample do not follow this trend

argues against this possibility. Examining gas richnesses and sizes for a larger sample of

galaxies might further clarify this issue, as might be obtained by homogenizing measured

properties across the SPARC (Lelli et al., 2016), SHIELD (Cannon et al., 2011), and LIT-

TLE THINGS (Hunter et al., 2012) samples along with samples of gas-rich UDGs. More

data are needed to quantify comparisons between gas richnesses and sizes predicted by UDG

formation models and other mechanisms, which we anticipate undertaking with data from 4.6. DISCUSSION 105

the full survey.

4.6.3 Disk Geometry and the BTFR

We now discuss the HI-conﬁrmed UDGs in the context of the baryonic Tully-Fisher (BTFR) in order to explore the possibility that our sample exhibits an oﬀset from this relation similar to that found by Mancera Piña et al.(2019b, 2020). Because our HI detections stem from spatially unresolved single-dish observations, we must resort to optical measures of the disk geometry to estimate HI disk rotation velocities, Vrot, from the measured velocity widths,

W50,c, in Tables 4.2 and 4.3. We therefore proceed to derive Vrot for our HI detections, examine BTFR oﬀsets in the context of the reliability with which we can estimate the disk geometry, and discuss the implications of these ﬁndings for UDG structure.

We ﬁrst compute rotation velocities for our HI detections using the relation for a ﬂat axisymmetric disk: W50,c,t V GF = , (4.3) rot 2sin(iGF )

where W50,c,t is the proﬁle velocity width that has been corrected for ISM turbulence (see

below) in addition to the instrumental eﬀects discussed in Section 4.5.1 and iGF is the disk

inclination implied by b/a of the best-ﬁtting GALFIT models of the optical UDG morphology

given in Table 4.1 and represented by the white ellipses in Figures 4.6. We calculate iGF via

the standard relation: (b/a)2 q2 cos2(iGF ) = − 0 , (4.4) 1 q2 − 0

where q0 is the intrinsic axial ratio. We adopt q0 = 0.2 in line with many previous studies

(e.g. Giovanelli et al., 1994, 1997; Leisman et al., 2017), although for our intermediate and

low-inclination systems values as large as q0 = 0.5 (Roychowdhury et al., 2013) only impact the derived Vrot at the 10% level.

While the value of q0 does not strongly impact the derived Vrot, we emphasize that there are considerable uncertainties in iGF derived from b/a in Table 4.1. First, if the HI disk 4.6. DISCUSSION 106

is warped (e.g. Kamphuis et al., 2015) or if the HI and optical disks are misaligned (e.g.

Starkenburg et al., 2019; Mancera Piña et al., 2020), iGF will not reﬂect the HI disk geometry.

Second, Figure 4.6 illustrates that the HI-conﬁrmed UDGs have irregular morphologies, while the GALFIT models used to derive b/a in Table 4.1 assume a smooth distribution of light (Z19). This raises the possibility that clumps in the disk systematically pull b/a away from the value that reﬂects the underlying disk geometry, biasing iGF . We therefore consider iGF to be only a rough approximation of the HI disk inclination that are much more uncertain than b/a from the smooth GALFIT models listed in Table 4.1, and list them as such in Table 4.5. We note that, since d(sinx)/dx = cosx is much larger for low x than

when x approaches 90◦, uncertainties in iGF in low- and intermediate-inclination systems

GF GF have a larger impact on Vrot than uncertainties on i in high-inclination systems.

We follow the prescription of Verheijen and Sancisi(2001) to correct W50,c in Tables 4.2

and 4.3 for ISM turbulence to obtain W50,c,t, required in Equation 4.3, for the Gaussian

proﬁles in Figure 4.2:

W W 2 2 −( 50,c )2 −( 50,c )2 W50,c,t = W + W [1 2e 100 ] 2W50,cWT,50[1 e 100 ]. (4.5) 50,c T,50 − − −

The factor of 100 km s−1 in the exponential terms accounts for the proﬁle shapes at 50%

−1 of their peak ﬂux. We set WT,50 = 5 km s in Eq. 4.5, commensurate with estimates for systems with ﬂat rotation curves by Verheijen and Sancisi(2001) and Kirby et al.(2012), since dwarf galaxies rarely have declining rotation curves (Catinella et al., 2006; Lelli et al.,

2016), and the UDG rotation curves from Mancera Piña et al.(2020) are generally ﬂat. We have also not attempted to correct for asymmetric drift in our unresolved data, although

GF −1 this may be signiﬁcant for Vrot . 15 km s (e.g. Iorio et al., 2017; Read et al., 2017). For GF these systems, Vrot is underestimated. If our HI detections have rising rotation curves at the edges of their HI disks as is the case for many dwarfs and some UDGs, then our choice

GF of WT,50 results in an over-correction. The resulting values of Vrot are given in Table 4.5, 4.6. DISCUSSION 107

which we consider highly uncertain due to the uncertainties in iGF discussed above.

In Figure 4.9, we show the BTFR composed of two samples of galaxies with spatially- resolved HI maps: SPARC (purple squares, Lelli et al., 2016) and LITTLE THINGS (black circles, Hunter et al., 2012; Iorio et al., 2017). In those samples, Vrot has typically been measured using a standard tilted-ring approach (Rogstad et al., 1974; Sicking, 1997) that

ﬁts for the disk geometry and rotation simultaneously to break the degeneracy between Vrot

and sini in the line-of-sight velocities. Figure 4.9 also shows the 6 intermediate-inclination

UDGs from the HUDs sample which deviate from the BTFR8 (Mancera Piña et al., 2019b),

and the 11 edge-on (i.e. high-inclination) HUDs (He et al., 2019) which by and large do

not (Mancera Piña et al., 2020). We note that, because the HI maps kinematically modeled

by Mancera Piña et al.(2019b, 2020) do not have suﬃcient spatial resolution to constrain

Vrot and i simultaneously (Di Teodoro and Fraternali, 2015; Kamphuis et al., 2015), a novel

method where i is estimated separately from Vrot is adopted. On the other hand, any value

◦ of i > 75 for the high-inclination UDGs of He et al.(2019) implies the same value of Vrot

since sini 1. ∼ 8 We have calculated the Mbary and its uncertainties using the values from Table 1 of Mancera Piña et al. (2020). We note that the error bars in our Figure 4.9 for the UDGs from Mancera Piña et al.(2019b, 2020) are smaller because they have propagated uncertainties in M∗ and MHI in logarithmic units instead of in linear units. 4.6. DISCUSSION 108

b/a 0.45 0.55 0.65 0.75 0.85 0.95

Inclination 65 55 45 35 25 15 5

10 )

9 /M bary 8 M

log( 7 HI-conﬁrmed UDGs, This Work Foreground dwarfs, This Work HUDs 6 Edge-on UDGs SPARC LITTLE THINGS 5 10 20 50 100 200 300 1 Vrot (km s− )

Figure 4.9: Baryonic Tully-Fisher relation (Mbaryvs.Vrot) formed by the SPARC (blue squares, Lelli et al., 2016) and LITTLE THINGS (green circles, Iorio et al., 2017) samples, where Vrot and i are derived from standard tilted-ring kinematic modeling, with the best-fitting BTFR shown as a dotted black line. The 6 UDGs from the HUDs sample where Vrot is estimated separately using a new method to determine i (cyan diamonds, Mancera Piña et al., 2019b, 2020) lie off the BTFR, while the edge-on HUDs (black triangles, He et al., 2019), by and large, lie within its scatter. When we use the optically-derived inclina- GF tions, i , and turbulence-corrected HI linewidths, W50,c,t, to estimate rotation velocities, GF Vrot , 7/9 HI-confirmed UDGs in our sample (orange and red stars) fall off this relation, as do some of our foreground dwarfs (purple crosses). The UDG symbols are colored according to their axial ratios/inclinations as shown in the colorbar. For the UDGs which fall off the BTFR, colored horizontal lines show how their axial ratios and inclinations change as they are brought onto the relation (from red to yellow). Representations of best-fit GAL- FIT models with the axial ratios corresponding to each pair of stars are shown overlaid on stacked optical images in Figure 4.6. It is plausible that the systematics of fitting smooth photometric models to clumpy, low inclination, LSB objects explains the offsets of the red stars from the BTFR. See text for details. 4.6. DISCUSSION 109

The orange and red stars in Figure 4.9 show the locations of our HI-conﬁrmed UDGs in

GF the Mbary Vrot plane when V is used to estimate rotation velocities, with the symbol − rot colour denoting iGF as given by the colorbar. The two UDGs with the largest iGF fall within the scatter of the relation deﬁned by SPARC and LITTLE THINGS (dotted black line), while the rest do not. Given the uncertainties in iGF particularly at low inclinations, we calculate the inclinations iBTFR required to bring the discrepant points onto the BTFR, connecting pairs of stars corresponding to the same galaxy in Figure 4.9 with a horizontal line. These values of iBTFR are also given in Table 4.5, the median iBTFR = 14◦. As expected from Equation 4.3, the discrepant points move on to the BTFR if the HI disks of the corresponding UDGs have inclinations below iGF . To constrain the plausibility with which a disk with iBTFR can reproduce the optical morphologies of the UDGs, we compute

(b/a)BTFR implied by iBTFR using Equations 4.3-4.5 and overplot ellipses corresponding to the best-ﬁtting GALFIT models obtained with b/a = (b/a)BTFR held ﬁxed in red in Figure

4.6.

In light of the above considerations, we conclude that interpreting the available observations to mean that the HI-conﬁrmed SMUDGes UDGs deviate systematically from the

BTFR is premature. The uncertainties in iGF are large, particularly in low-inclination systems. The white and red ellipses in Figure 6 demonstrate that in many cases, the GALFIT models that generated iGF and those produced holding (b/a)BTFR ﬁxed produce nearly

the same projected disk geometry. Furthermore, the irregular optical morphologies of the

HI-conﬁrmed UDGs in Figure 4.6 relative to our HI non-detections evident in Figures 4.6

and 4.7 raise the possibility that clumpy emission systematically biases the GALFIT ﬁts

that generated iGF . Since there are few clumps in each object and since those clumps are

rarely symmetrically distributed about the object center, it seems plausible that the eﬀect

of ﬁtting these irregular LSB objects with smooth GALFIT models is to systematically

under-estimate b/a such that iGF is biased high. We emphasize that the SMUDGes UDG 4.6. DISCUSSION 110

candidate selection criteria for low surface brightness and high ellipticity (Z19) favors low- inclination disks relative to high-inclination ones with the same M∗ and Reff, and therefore that low-inclination disks should be over-represented in the SMUDGes sample compared to samples with a random distribution of sky orientations with a mean i 60◦. Furthermore, ∼ since our HI follow-up sample is eﬀectively selected on luminosity (see Section 4.3) in a surface brightness-restricted sample, the objects in our sample are more likely still to be at low inclinations. It is therefore possible that most of the HI-conﬁrmed UDGs have low inclinations and that the iGF for those low-inclination systems is biased high.

A detailed investigation of potential biases in iGF for our HI-conﬁrmed UDGs is beyond the scope of this pilot paper, but we are carrying out simulations to quantify biases in smooth

GALFIT models of irregular LSB galaxies as a function of their asymmetry (e.g., Abraham et al., 1996, 2003; Conselice, 2003) for the full survey. As a ﬁrst check on our hypothesis, we

GF estimate Vrot for the foreground dwarfs (which one would expect to lie within the scatter of the extrapolated BTFR, similar to other studies of the dwarf galaxy population; Iorio

GF et al. 2017; Cannon et al. 2011) and overplot them on Figure 4.9. We emphasize that Vrot for both the foreground dwarfs and the UDGs are only order-of-magnitude estimates that

assume the optical and HI disks are aligned, and we do not attempt to quantify these

signiﬁcant uncertainties either in Table 4.5 or in Figure 4.9. Nonetheless, at least some

foreground dwarfs deviate from the BTFR similarly to the HI-conﬁrmed UDGs, lending

credence to our hypothesis that iGF is systematically overestimated.

The gas-rich, intermediate-inclination UDG outliers from the BTFR studied by Mancera

Piña et al.(2019b, 2020) imply that the underlying structure and baryonic composition of

these systems diﬀers fundamentally from that assumed in any of the UDG formation sce-

narios posited so far (see Section 4.2). As proposed by these authors, a high stellar speciﬁc

angular momentum, low star formation feedback scenario is one possible explanation. Ex-

amining Figure 4.9, however, it is curious that the consistency of gas-rich UDG samples

with the BTFR deﬁned by higher surface brightness systems seems to depend on how their 4.6. DISCUSSION 111

Table 4.5: Inclinations and Rotation Velocities GF GF BTFR Name i Vrot i (deg) (km s−1) (deg) (1) (2) (3) (4) HI-conﬁrmed UDGs SMDG1220188+280131 52 19 20 ∼ ∼ ∼ SMDG1225185+270858 53 38 ∼ ∼ − SMDG1226040+241802 37 26 11 ∼ ∼ ∼ SMDG1230359+273311 69 53 ∼ ∼ − SMDG1241424+273353 38 12 6 ∼ ∼ ∼ SMDG1248019+261236 36 25 15 ∼ ∼ ∼ SMDG1301005+210355 51 18 14 ∼ ∼ ∼ SMDG1312223+312320 42 29 17 ∼ ∼ ∼ SMDG1315427+311846 47 8 4 ∼ ∼ ∼ Foreground dwarfs SMDG1103517+284120 35 21 ∼ ∼ − SMDG1223451+283549 54 34 ∼ ∼ − SMDG1231329+232916 63 3 6 ∼ ∼ ∼ SMDG1239050+323016 33 27 ∼ ∼ − SMDG1240017+261919 43 8 17 ∼ ∼ ∼ SMDG1253571+291500 60 37 ∼ ∼ − SMDG1255412+191221 49 11 17 ∼ ∼ ∼ SMDG1306148+275941 65 22 ∼ ∼ − SMDG1313188+312452 39 9 12 ∼ ∼ ∼ col.(2): Inclination calculated using Eq.4.4, b/a from Table 4.1, and an intrinsic axial ratio of q0 = 0.2. col.(3): Rotational velocity calculated using W50,c corrected for turbulence and i in col.(2). Given the systematics associated with measuring inclinations of clumpy GF GF low-inclination objects from smooth models, we consider i and Vrot to be rough estimates (see text). cols.(4): Inclinations required to lie on the BTFR for UDGs and GF dwarfs with Vrot lower than expected from the BTFR at their measured Mbary. 4.7. CONCLUSIONS 112

inclinations were measured: the edge-on systems studied by He et al.(2019) (where inclination uncertainties do not impact estimates of Vrot) are consistent with the BTFR, while the intermediate-inclination systems studied by Mancera Piña et al.(2019b, 2020) (where Vrot

is measured independently from i using a new technique) are outliers. The sensitivity of the

locations of our low- and intermediate-inclination HI-conﬁrmed UDGs in the Mbary Vrot − plane on the adopted inclination suggests that the eﬀect of the viewing geometry should be

carefully considered when inclination-dependent Vrot are used to study the BTFR.

We emphasize that BTFR studies with SMUDGes UDGs that address the possible in-

clination dependence of oﬀsets from this relation require HI imaging with suﬃcient angular

and spectral resolution to simultaneously model Vrot and i using standard tilted ring ap-

proaches. This is feasible for a small subset of the HI detections presented here, and work

in this regard is underway.

4.7 Conclusions

We have presented GBT HI observations of 70 optically-detected SMUDGes UDG candidates

(Z19) with mg . 19.5 mag in the Coma region. We detect HI reservoirs in 18 of them (Figure

4.2), measuring systemic velocities, Vsys, velocity widths, W50,c, and ﬂux integrals, Sdv, ´ directly from the spectra. Using kinematic distances estimated from Vsys, we compute HI masses, MHI , from the spectra as well as stellar masses, M∗, and half-light radii, Reff , from GALFIT models to the deep DECaLS imaging. We use Reff to conﬁrm that 9 of our

HI detections satisfy the size criterion deﬁning UDGs, while the remainder are foreground dwarfs (Tables 4.2 and 4.3). Although only a pilot for a much larger GBT program that is currently underway, these observations already represent the largest HI follow-up campaign of optically-selected UDG candidates ever reported, and the 9 conﬁrmed UDGs are the largest available sample of optically-selected UDGs with HI detections.

Comparing the properties of our HI-detected UDGs, HI-detected foreground dwarfs and 4.7. CONCLUSIONS 113

our HI non-detections, we ﬁnd similar sky distributions relative to the Coma large-scale structure (Figure 4.1) but that 8/9 UDGs are in low-density environments with no massive

−1 (Mg < 19 mag) companions within Rproj = 300 kpc or ∆Vsys = 500 km s . In addition, − ± our HI detections typically have counterparts in the NUV if the exposures are sufficiently deep (& 1000 sec with GALEX). In DECaLS-depth optical imaging, the gas-rich UDGs are bluer and smoother in morphology than the UDG candidates that we do not detect in HI but the scatter is large in both properties (Figures 4.5, 4.6, and 4.7). On the other hand, targets that are both blue and irregular are gas-rich, while those that are both red and smooth are gas-poor: it is the combination of optical morphology and color that best predicts gas richness. Although the angular sizes of the foreground dwarfs are typically larger than those of the HI-confirmed UDGs, there is little difference in optical morphology or color between these subsamples (Figures 4.6 and 4.8). Without distance information, foreground dwarfs contaminate samples of optically blue, irregular UDG candidates.

Commensurate with tentative results for blue UDGs around galaxy groups (Spekkens and

Karunakaran, 2018), we ﬁnd evidence for a correlation between the gas richness, MHI /M∗, and size, Reff , when our HI-conﬁrmed UDGs as well as other gas-rich UDGs are divided into two stellar mass bins (Figure 4.4). The same trend is not obvious for the foreground dwarfs. The correlation between UDG gas richness and size suggested by the data is broadly consistent with predictions from the star formation feedback model for UDG formation (di

Cintio et al., 2017), although other mechanisms may also produce the trend.

We place our HI-conﬁrmed UDGs on the BTFR using best-ﬁtting inclinations, iGF , from smooth GALFIT models of DECaLS imaging and turbulence-corrected velocity widths to

GF GF estimate rotation velocities Vrot . We ﬁnd that the 7/9 objects with the lowest i have lower GF Vrot than expected from the BTFR deﬁned by high surface brightness, gas-rich galaxies with HI rotation curves and disk geometries derived from kinematic models (Figure 4.9), similar to that found by Mancera Piña et al.(2019b, 2020) for a sample of marginally-resolved

HUDs using a new technique for constraining i separately from Vrot via the HI morphology. 4.7. CONCLUSIONS 114

For our sample, however, we find that plausible systematics resulting from the application of smooth GALFIT models to clumpy, low-inclination LSB objects are sufficient to reconcile these discrepancies (Figures 4.6 and 4.9) precluding a meaningful analysis of BTFR offsets.

We plan on investigating this trend and its implications in detail with our full follow-up sample.

The pilot survey results presented here provide some initial insight into the properties of gas-rich UDGs and the mechanisms by which they form. Despite being the largest of its kind, our sample of conﬁrmed gas-rich optically-detected UDGs remains small. A much larger SMUDGes HI follow-up campaign is underway at the GBT. We ultimately plan on targeting over 200 objects, and expect to conﬁrm at least 50 gas-rich UDGs. This larger sample will enable quantitative investigations of the interplay between gas richness and

UDG properties in order to understand how they form and evolve. Furthermore, it will also provide predictive insight into the gas properties of UDG candidates in the eventual

10, 000 deg2 SMUDGes survey. ∼ Chapter 5

An Update On and the Outlook of the SMUDGes in HI Survey

5.1 Introduction

As outlined in the previous chapter, we have initiated an HI follow-up survey of optically- detected UDGs from the Systematically Measuring Ultra-Diﬀuse Galaxies (SMUDGes, Zarit- sky et al., 2019) Survey. We showed that single-dish HI observations are an eﬃcient and useful tool to follow-up UDGs. Similar to the broader galaxy population, the 18 HI-detections from our pilot survey tended to be bluer in colour and more irregular in their morphology.

However, we also highlighted that there may be some differences between UDGs selected from blind HI surveys and those first detected in the optical with subsequent HI follow-up observations. Using the 9 new HI-confirmed UDGs from our pilot data along with other

UDGs from the literature, we found that there was a slight trend between the gas-richness and size of UDGs and no such trend in our HI-detected dwarfs. In this chapter, we provide an update on the optical and HI SMUDGes survey progress, expand on the results from the pilot data using the data collected and reduced thus far, and an outlook of the complete survey and future work. Before proceeding, it should be noted that throughout this chapter 115 5.2. SURVEY PROGRESS 116

“we" is used given standard practice in the ﬁeld and for consistency with previous chapters, however all of the work presented here is my own.

5.2 Survey Progress

5.2.1 SMUDGes in the Legacy Surveys

The SMUDGes survey aims to make great use of the publicly available DESI Legacy Imaging

Surveys (“Legacy Surveys") data which covers 14000 square degrees of the sky (see Section ∼ 1.2). The SMUDGes Pilot survey (see Chapter4) focused on the Coma Cluster and its local environment primarily because of the initial detections of the large population of LSBs and

UDGs in the Coma Cluster (i.e. van Dokkum et al., 2015a; Koda et al., 2015a). Moving beyond the Coma Cluster has been an onerous undertaking due to the extremely large coverage of the Legacy Surveys imaging, which requires an accurate and eﬃcient method to detect UDG candidates.

Our collaborators have developed such a pipeline, the basis of which is outlined in Zarit- sky et al.(2019) and the ﬁnal version consists of several improvements made to increase

UDG candidate detection eﬃciency (i.e. revised masking procedures, infrared maps to exclude dust/galactic cirrus, etc). Prior to being run on the complete Legacy Surveys imaging data, the pipeline was tested using the Legacy Surveys imaging that overlaps the SDSS

Stripe 82 region (S82, a region covering the celestial equator with deep imaging, see Annis et al., 2014; Jiang et al., 2014). As we describe in the following section, we have selected some of our follow-up candidates from this UDG candidate sample from the S82 region which will allow us to probe a range of environments beyond the Coma Cluster.

Before proceeding we note that the Legacy Surveys imaging has been continuously expanding in imaging depth and coverage since the initial SMUDGes Pilot Survey of the Coma

Cluster and its local environment (Zaritsky et al., 2019). This results in a potentially variable sample of UDG candidates in these data and, more crucially, a varying set of properties. 5.2. SURVEY PROGRESS 117

That is to say, as new data are processed through the (also improving) detection pipeline there may be some new candidates that were previously excluded and candidates detected in previous sample iterations that no longer satisfy the SMUDGes UDG candidate criteria.

The number of these candidates that move in and out of the samples is low, however photometric and morphological properties (i.e. size, Sérsic index, surface brightness, and axial ratios) of UDG candidates can vary slightly. As a result of this, our HI follow-up samples, which are selected from what are essentially “work in progress” catalogs, may contain some

UDG candidates which no longer satisfy the UDG criteria. Nevertheless, the majority of the preliminary results presented below using these optical properties will be re-derived as necessary by our collaborators in order to maintain coherence across our HI follow-up samples and the full SMUDGes survey.

5.2.2 SMUDGes in HI

The goal of our HI-follow up survey is to combine deep single-dish HI observations with the deep optical Legacy Surveys imaging to (1) obtain distance measurements for UDG candidates in order to confirm them as bonafide UDGs and (2) to place constraints on the formation mechanisms of UDGs across different environments. In order to accomplish this, we require a statistically significant sample of UDGs with HI detections (& 40 UDGs). Our HI follow-up survey has been conducted over the last 3 years with 5 samples of UDG candidates from the SMUDGes survey. Throughout this chapter a "campaign" will refer to observations of 1 of the 5 aforementioned samples. For example, our pilot campaign (or campaign 1, i.e. Chapter4) focused on the SMUDGes survey of the Coma Cluster and its immediate environment. Our subsequent campaigns have observed SMUDGes UDG candidates selected from different regions including the extended Coma environment, S82, and constant bands of declination. We will briefly describe the targets in each campaign beyond the pilot sample below and have summarized them in Table 5.1. All GBT observations have been completed across the 5 campaigns. The over 500 hours of observations were prepared and 5.2. SURVEY PROGRESS 118

conducted by the author who also completed the subsequent data reduction and analysis.

5.2.2.1 Campaigns 2 and 3

In campaigns 2 and 3, UDG candidates samples were compiled using identical selection

00 criteria (mg . 19.5 mag and reﬀ > 5 ) with the pilot campaign. In addition, observations are conducted using the same conﬁguration as the pilot campaign (see Chapter4, as well as Chapter2) with the GBT’s backend spectometer VEGAS (VErsatile GBT Astronomical

Spectrometer) in Mode 7 which has a bandwidth of 100 MHz ( 21000km s−1) and a spectral ∼ resolution of 3.1 kHz ( 0.6km s−1). Campaign 2 (2019 Sept. to 2020 Jan.) targetted 37 ∼ UDG candidates from the ﬁnal revision of the SMUDGes survey of the Coma Cluster (i.e.

Zaritsky et al., 2019) and 26 UDG candidates from a ﬁrst pass through the Legacy Surveys

imaging data that overlaps the S82 region. The follow-up sample in campaign 3 (2020 Mar.

to 2020 Jul.) includes 42 UDG candidates in the greater Coma cluster environment, between

10 and 20 degrees radially from the Coma cluster core, and 16 additional targets from S82.

We note that we have not selected any UDG candidates that fall within the Virgo Cluster

region given the extensive multi-wavelength (including HI) investigations of this area (e.g.

Davies et al., 2010; Boselli et al., 2011; Ferrarese et al., 2012; Taylor et al., 2012; Boselli

et al., 2018; Brown et al., 2020). The GBT data from these campaigns have been reduced

and we present the preliminary results of the analysis of this data in Section 5.3.

5.2.2.2 Campaigns 4 and 5

Based on the preliminary results from campaigns 1 to 3 and with broader UDG candidate

search capabilities in the Legacy Surveys imaging, we made modiﬁcations to our candidate

selection process. We also made one modiﬁcation to our observational conﬁguration for

campaigns 4 and 5. We discuss them in turn below.

In the pilot campaign, we found that several of our HI detections also have discernible

GALEX UV emission (see Section 4.6.1). With this in mind and in hope of increasing 5.2. SURVEY PROGRESS 119

our HI and UDG detection fraction, for campaign 4, the official final campaign of our survey, we selected objects with discernible UV emission. Here, we have determined what constitutes discernible emission by visual examination of GALEX cutouts. This criterion was in addition to the g band magnitude cut, surface brightness, and angular size selection as − before. This additional selection criterion was afforded by a revised and more capable UDG candidate search pipeline deployed on the ever-expanding Legacy Surveys imaging data. For this campaign, our follow-up sample of 92 targets was drawn from UDG candidates detected across the Legacy Surveys imaging between 23◦ < Dec. < 33◦.

Campaign 5 was submitted and accepted as a “filler" project at the GBT and, by definition, was not assured any significant amount of observing time. However, we were able to obtain data for all targets in this campaign which significantly increases our sample size.

The 95 targets in this follow-up sample were selected from UDG candidates detected in

Legacy Surveys imaging between 2◦ < Dec. < 13◦ (also avoiding the Virgo Cluster, see above) similar to that in campaign 4. For campaign 5 we applied a brighter magnitude cut of mg . 19 mag to minimize the requested observing time while maximizing the sample size. We note that because we have included this additional GALEX selection criteria and are beginning to overlap with the ALFALFA 100% catalog (Haynes et al., 2018) sky coverage, it is likely that some SMUDGes UDG candidates will already be detected in ALFALFA.

While compiling the target samples for campaigns 4 and 5 we cross-matched and excluded approximately 50 UDG candidates in total with existing ALFALFA HI detections. We discuss the feasibility of combining ALFALFA and SMUDGes to complement our HI follow-up survey in the next chapter.

To combat the increasing effects of radio frequency interference (RFI) in our data, we used a different VEGAS configuration for campaigns 4 and 5. While the observations and subsequent reduction of VEGAS in Mode 7 are relatively straightforward, a bandpass of its size is not only prone to effects of RFI at the individual channel and entire bandpass level but is also susceptible to stability issues. In an attempt to mitigate these effects, we 5.2. SURVEY PROGRESS 120

have used VEGAS in Mode 21 configured to use 3 overlapping 23.44 MHz ( 4900km s−1) ∼ bandpasses to cover recessional velocities between 1000 km s−1 and 13000 km s−1 with a − spectral resolution of 2.9 kHz ( 0.5 km s−1), similar to our previous configuration. These ∼ narrower bandpasses will primarily alleviate bandpass stability issues but will also allow for us to selectively flag and exclude bandpasses that are affected by strong, broad RFI. The observations for campaigns 4 and 5 are complete and the data reduction is underway.

As a final note in comparing the follow-up samples of the 5 campaigns, we show the effect of the additional GALEX selection criterion that is clearly visible in the colour distributions of these samples in Figure 5.1. The samples in campaigns 4 and 5 are, on average, bluer in colour than those in campaigns 1 to 3. In addition, we include the mean g r colour for the − HI-bearing Ultra Diffuse galaxies (HUDs) from Leisman et al.(2017) as the vertical dotted grey line. We can see that the campaign 4 and 5 samples are more similar to the HUDs in their optical colour.

Table 5.1: Summary of SMUDGes in HI campaigns Campaign Obs. Region Sample HI UDGs Dates Size Dets. Dets. Pilot (1) 18/Feb-18/Aug Coma Cluster 70 18 9 2 19/Sep-20/Jan Coma Cluster + S82 63 5 4 3 20/Mar-20/Jul Coma Env. + S82 58 8 7 Final (4) 20/Nov-21/May 23◦ < Dec. < 33◦ 92 TBD TBD Filler (5) 21/May-21/Jun 2◦ < Dec. < 13◦ 95 TBD TBD Total 378 & 31 & 20 5.2. SURVEY PROGRESS 121

Campaigns 1-3 Final Campaign 60 Filler Campaign

Count 30

0 0.0 0.2 0.4 0.6 0.8 g-r

Figure 5.1: Comparison of the g r colour distributions for the first three follow-up cam- − paigns (purple histogram), the final campaign (i.e. campaign 4, pink histogram), and the filler campaign (i.e. campaign 5, orange open histogram). A KDE (i.e. smoothed) distribution is shown as the solid lines with the same colours for each sample. For reference, the vertical dotted grey line shows the mean g r colour for the HI-bearing Ultra Diffuse − galaxies (HUDs) from Leisman et al.(2017). 5.3. PRELIMINARY RESULTS 122

5.3 Preliminary Results

With the data collected, reduced, and analyzed from campaigns 2 and 3, we can begin to expand on some of the initial results we presented in Chapter4 with the additional 111

UDG candidates in these two campaigns. Of these 111 UDG candidates, we ﬁnd HI emission associated with 13 of them: 10 of which are classiﬁed as UDGs (blue spectra in Figure 5.2)

−2 based on the surface brightness (µg > 24 mag arcsec ) and physical size (Reff > 1.5 kpc) criteria, with the remaining 3 being LSB dwarfs (green spectra in Figure 5.2). In our visual check of GALEX UV cutouts for our 13 HI detections we found that all but 2 had discernible

UV emission and note that the two without may simply require a re-reduction of archival

GALEX data to conﬁrm them as true UV non-detections. For the remaining 98 HI non- detections, we estimate upper limits on MHI /Lg following the same method in Section 4.5:

5σ HI mass upper-limits using Eq. 4.2 and assuming, for simplicity, a distance of 100 Mpc

−1 and a velocity width of 50 km s and Lg using the g band apparent magnitudes assuming − the same distance. With these 11 new HI-conﬁrmed UDGs along with the 9 from our pilot campaign, we will brieﬂy expand on a few key results from the pilot survey presented in

Chapter4. 5.3. PRELIMINARY RESULTS 123

SMDG0126070-020411 SMDG1159082+321256 20 SMDG1200398+212450 15 UDG Dwarf UDG 10 15 10 10 5 5 5 0 0 0 1000 2000 3000 500 1000 1500 6500 7000 7500 8000 SMDG1205178+233120 4 SMDG1230456+264650 10 SMDG1330214+122904 UDG Dwarf UDG 2 2 5

0 0 0

2 − 3000 4000 2000 3000 7000 8000 1.5 SMDG1340311+282700 SMDG1345111+331129 SMDG1347337+202653 UDG UDG 10 Dwarf 4 1.0

0.5 2 5

0.0 0 0 Flux0 (mJy) .5 2 − − 4500 5000 5500 6000 4000 5000 8000 9000 15 SMDG1406309+191742 SMDG2104498+002509 4 SMDG2113130+014217 UDG UDG UDG 2 10 1 2 5 0 0 0 1 − 3000 4000 6500 7000 7500 8000 4000 5000 6 SMDG2306234+013114 UDG 4

2 − 2500 3000 3500 4000 VHelio(km/s)

Figure 5.2: HI detections of UDGs (blue spectra) and dwarfs (green spectra) from campaigns 2 and 3. Target names and classiﬁcation (UDG or Dwarf) are in the top-right corner of each panel. The black dotted line in each panel represents 0 mJy. The black dashed portions of spectra for SMDG1200398+212450 and SMDG1340311+282700 represent nearby gas-rich galaxies, the latter of which has been truncated for visualization purposes. 5.3. PRELIMINARY RESULTS 124

First, we will address the comparisons of our HI detections and non-detections across campaigns 1-3. With this growing sample we can begin to visualize these subsets (HI detections and non-detections) a bit diﬀerently. In Figure 5.3, we show the HI mass to g band − luminosity ratio (MHI /Lg) for our HI detections of UDGs (blue stars) and dwarfs (green squares), and the corresponding upper-limits on MHI /Lg for our HI non-detections (red

hexagonal bins) as a function of g r (left panel) and g z (right panel). Recall that these − − ratios are distance-independent and, in particular for the non-detections, demonstrate our

sensitivity throughout our entire spectral bandwidth (i.e. out to 14000 km s−1 or 200 ∼ ∼ Mpc). Following the trend found in the pilot campaign, the HI-detections (both UDGs and

dwarfs) in campaigns 2 and 3 tend to be bluer in colour. Interestingly, we can begin to

see the locus of the HI non-detections fall towards redder colours and have lower MHI /Lg.

Together with the sample colour distributions presented in Figure 5.1, we can infer that there may be an increase in our UDG detection rate for campaigns 4 and 5, in which bluer objects were preferentially targetted.

We now turn to our 11 new HI-conﬁrmed UDGs in the context of their derived HI properties. Figure 5.4 presents an updated version of Figure 4.3, where we show the velocity width (W50) distributions (top) and the HI mass-stellar mass plane (bottom) for our now expanded sample of UDGs (orange and blue), the HUDs (green and purple), and the

ALFALFA 40% sample (grey, black dotted line; Huang et al., 2012a). While the W50 distributions are similar between our HI-conﬁrmed UDG sample, the HUDs samples, and the broader galaxy population, we can see that these additional UDGs from our sample are probing a slightly diﬀerent parameter space in the HI mass-stellar mass plane. As described in Section 4.5, the HUDs samples are selected from the ALFALFA HI catalog and have elevated gas richnesses (MHI /M ) for their stellar masses. With the inclusion of the linear

ﬁt from Huang et al. (black dotted line, 2012a), it is clear to see that the majority of the

UDGs from campaigns 2 and 3 fall below the average galaxy from ALFALFA and possess 5.3. PRELIMINARY RESULTS 125

1.25 HI detections, UDGs HI detections, Dwarfs 12 1.00 10 )) 0.75

/L 8 0.50 M ( g 0.25 6 /L

HI 0.00 M 4 # of HI Non-detections

log( 0.25 − 2 0.50 − 0 0.0 0.2 0.4 0.6 0.8 0.00 0.25 0.50 0.75 1.00 1.25 g r(mag) g z(mag) − −

Figure 5.3: An updated version of Figure 4.5 which shows MHI /Lg (blue stars and green squares for HI-conﬁrmed UDGs and foreground dwarfs, respectively) as a function of g − r (left) and g z (right) colour for our sample now including the HI-conﬁrmed UDGs − and dwarfs from campaigns 2 and 3. In addition, the HI non-detections are now shown in hexagonal bins with the colour of the bin corresponding to the count as displayed in the colourbar. As a reminder, the “typical” upper limit on MHI /Lg for the HI non-detections corresponds to the vertical centroid of the bin.

signiﬁcantly smaller HI reservoirs than their HUDs counterparts. This hints that UDG sam-

ples selected in blind HI surveys may be diﬀerent from those that are ﬁrst detected optically

and subsequently followed-up in HI, which we discuss in the next section. The diﬃculty

of disentangling whether this are truly diﬀerent sub-populations lies in understanding the

selection functions of these samples. The results from campaigns 4 and 5 will be crucial to

unraveling this possible diﬀerence given the shift towards selecting UDG candidates from

SMUDGes with properties akin to typical gas-rich galaxies. 5.3. PRELIMINARY RESULTS 126

Pilot 0.20 Cmp. 2+3 HUDs-B HUDs-R 0.15 α.40, H12

0.10

Relative Frequency 0.05

0.00 1.0 1.5 2.0 2.5 3.0 log(W50, km/s) 11

10 )

9 /M HI

M 8 log( 7

6 6 8 10 log(M /M ) ∗

Figure 5.4: A comparison of the velocity width distributions (top) and the MHI M∗ plane (bottom) between our pilot (orange bars and stars) campaign, samples from campaign 2 and− 3 (blue bars and squares), the HUDs-B and -R samples (green and purple), and the ALFALFA 40% sample. Also, included here is the piecewise linear ﬁt (black dotted line) to the ALFALFA 40% sample from Huang et al.(2012a). This is an updated version of Figure 4.3. 5.3. PRELIMINARY RESULTS 127

With our sample of HI-conﬁrmed UDGs now doubled, we revisit the gas richness-size relation that we ﬁrst explored in the context of UDG formation mechanisms (see Figure

4.4 and Section 4.6.2). As a brief reminder, the star-formation feedback UDG formation mechanism from di Cintio et al.(2017) predicts that, at a ﬁxed stellar mass, the gas richness of ﬁeld UDGs scale with their sizes. We show the gas richness-size relationship in Figure

5.5 for the 20 HI conﬁrmed UDGs from campaigns 1-3 along with 5 UDGs around Hickson

Compact groups with HI follow-up from Spekkens and Karunakaran(2018, open circles) and

6 UDGs with HI interferometric imaging from Mancera Piña et al.(2020, open diamonds).

We have separated these UDGs into two stellar mass bins of roughly equal size and coloured them accordingly.

Qualitatively, the UDGs in the lower stellar mass bin (yellow symbols) display a more convincing relationship between gas richness and size than the UDGs in the higher stellar mass bin (red symbols). With the increased sample size available here, however, we can now explore quantifying the relationship between gas richness and size. To this end, we have performed an orthogonal distance regression for each stellar mass bin (yellow and red lines correspond to the same coloured symbols) and the whole sample (purple dash-dotted line). For the individual stellar mass bins, the dotted lines and shaded regions, respectively, correspond to the best fit and 95% confidence intervals from bootstrap resampling. The resulting best fits are

log(MHI /M∗) = 1.580[log(Reff )] + 0.024, 7.2 < log(M∗/M ) 8.0, (5.1) ≤

log(MHI /M∗) = 1.047[log(Reff )] 0.023, log(M∗/M ) > 8.0, (5.2) −

log(MHI /M∗) = 1.457[log(Reff )] 0.098, all UDGs. (5.3) −

These fits are the first of their kind to place any form of observational constraint on the relationship between the gas richness and the size of UDGs. We must, however, also consider the reliability of these fits and to do so we make use of the reduce-chi squared statistic (χ2) 5.3. PRELIMINARY RESULTS 128

which will equal 1 for a good fit. The χ2 for our fit to the the lower stellar mass bin may be considered relatively reliable with a value of 1.4 while less so for the χ2 values for the higher stellar mass bin and the total sample at 1.98 and 2.26, respectively. It is, however, clear that these relatively larger values of χ2, by definition, stem from the large dispersion in gas richness in the high mass bin. In particular, there is one outlying UDG from our HI follow-up sample in the high stellar mass bin, SMDG1345111+331129 (red square, log(MHI /M∗) = 0.32 and log(Reff /kpc) = 0.52), whose relatively low gas richness is − a large driver in increasing the χ2 values. Indeed, repeating our fitting procedure while excluding this outlier results in significantly lower χ2 values for the high and total sample of 1.2 and 1.7, respectively.

Taking a closer look into the derived photometric properties of SMDG1345111+331129 along with a visual inspection of the best-ﬁt GALFIT models leads us toward a similar discussion presented in Section 4.6, where we explored the reliability of the smooth GAL-

FIT models to the irregular, gas-rich UDGs. We leave such a discussion for consideration with the full survey sample and instead brieﬂy consider how this outlier could shift in the gas richness-size plane to obtain a more reliable χ2 value. If we desired to simply bring

SMDG1345111+331129 closer to the scatter of other UDGs in the high stellar mass bin, then we would require a shift in gas richness of approximately 0.35 dex. To accomplish this, we would need an increase in MHI by a factor of 2.2 or an equivalent decrease in M∗, the ∼ latter of which is more readily attained given the possibility for GALFIT models to incor- rectly estimate properties. Since our M∗ estimates are derived from the g r colours and − the mass-to-light colour relations from Zhang et al.(2017), we can estimate that a factor of

2.2 decrease in M∗ can be obtained if the g r decreases from 0.5 to 0.35. Such a shift in − g r colour is not out of the realm of possibility for objects this faint. This brief example − establishes the need for a larger sample size to either better constrain this relation and the scatter within the stellar mass bins. 5.4. OUTLOOK 129

7.2 < log(M /M ) 8.0 log(M /M ) > 8.0 ∗ ≤ ∗ 1.5

) 1.0 ∗ /M

HI 0.5 M log( 0.0

0.5 − Pilot Campaigns 2, 3 HCG UDGs HUDS 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 log(Reff/kpc)

Figure 5.5: A revised version of Figure 4.4 which shows gas richness, log(MHI /M∗), as a function of size, log(Reff ). This figure only shows all HI-confirmed UDGs from our follow-up survey so far and the 5 UDGs around Hickson Compact groups from Spekkens and Karunakaran(2018, open circles) and 6 UDGs from Mancera Piña et al.(2020, open diamonds). The colours of the symbols represent the stellar mass bin of the objects. The dotted lines show the best orthogonal distance regression fit to the stellar mass subsets and the shaded regions shown the 95% confidence intervals. The purple dash-dotted line shows the best fit to the whole sample of UDGs.

5.4 Outlook

It is exciting to see results such as those presented in Figure 5.5 come to fruition with just two additional campaigns and the prospect of adding new HI-conﬁrmed UDGs will only strengthen them. Here, we brieﬂy speculate on the future prospects for constraining UDG properties through the addition of the two campaigns not presented here, with on-going data reduction, the utility of other catalogs and archival data, and future follow-up endeavours also addressed. 5.4. OUTLOOK 130

The Full SMUDGes in HI Survey

As mentioned above, all of the observations are complete with only the data reduction and subsequent analysis (i.e. derived HI properties) of the observations from campaigns 4 and 5 remaining. We anticipate that this will be completed by Fall 2021. Looking toward the full survey, we can consider the increased correlation between HI and GALEX UV detections in campaigns 2 and 3, at over 80%, to speculate that we may have as many as 150 new HI- conﬁrmed UDGs from the 187 UDG candidates in campaigns 4 and 5. This many detections will provide a truly powerful sample to constrain relationships between key UDG properties, such as the gas richness-size relation in the previous section. In addition, we would be in a position to further test for surface brightness, stellar mass, and environmental eﬀects. These investigations will be critical in the interpretation for our full survey sample and will allow for concrete conclusions to be made.

SMUDGes in ALFALFA and GALEX+WISE

We now return to our discussion of the possibility of including ALFALFA HI detections cross- matched with UDG candidates from the full SMUDGes survey in Section 5.2.2.2. Previous

ALFALFA-derived UDG samples (i.e. Leisman et al., 2017; He et al., 2019) mainly used

SDSS imaging which is much shallower than the Legacy Surveys data and may not accurately estimating photometric properties for UDG candidates. The inclusion of these objects would not only provide the opportunity to increase the sample of HI-conﬁrmed UDGs but also allow a direct comparison of UDGs detected in a blind HI survey to our optically-selected

HI follow-up survey. The utility of such a comparison would help investigate any potential evolutionary diﬀerences between the two populations, if any exist at all. As mentioned, our preliminary cross-matching produced 50 targets between SMUDGes and ALFALFA which ∼ is an exceptional increase to our sample size.

Another extremely useful tracer is UV emission to constrain the star-forming properties 5.4. OUTLOOK 131

of UDGs with and without HI detections. We have already established a feasible method of compiling and analysing GALEX UV data in Section3. It would be informative to carry out a similar analysis initially for our HI-conﬁrmed UDGs and subsequently for HI non-detections and other UDG candidates within the SMUDGes survey. A constraint from our methods in

Section3 was the lack of infrared (IR) corrections in star formation rate estimates. It may be possible to perform such corrections within the SMUDGes survey given that the Legacy

Surveys imaging combines ground-based optical photometry with IR photometry from the

Wide-ﬁeld Infrared Survey Explorer (WISE). However, the primary utility of the WISE

imaging is to detect and reject dust in the UDG candidate pipeline and any constraints on

IR photometry will require separate, additional processing.

HI Imaging with Square Kilometer Array Precursors

At the completion of our HI follow-up survey, we may have over 150 new conﬁrmed UDGs.

Such a sample will allow us to carry out statistical comparisons within it and with other

LSB, UDG, and dwarf samples. However, the unresolved HI properties (i.e. MHI , Vsys, and

W50) derived from our single-dish GBT survey will answer only some of the many questions

about these curious galaxies. Moving beyond SMUDGes in HI, we need to use our sample

of gas-rich UDGs to select particularly suitable targets for high-resolution interferometric

HI imaging. This will help answer more detailed questions regarding the distribution and

kinematics of UDGs. While some initial work in this regard is underway (Sengupta et al.,

2019; Mancera Piña et al., 2020), very few UDGs with HI imaging are suﬃciently resolved.

As discussed in Section 4.6, accurately measuring and modelling the dynamics of UDGs is

useful in understanding their relation to the broader galaxy population, especially in the

context of scaling relations (i.e. the baryonic Tully-Fisher relation, BTFR).

Motivated by this need to properly constrain these crucial properties, we were awarded

observations of two gas-rich UDGs with MeerKAT (the ﬁrst time allocation on this facility to

a Canadian Principal Investigator), a newly-commissioned Square Kilometer Array (SKA) 5.4. OUTLOOK 132

precursor facility. MeerKAT is a 64-dish interferometer that will allow for use to obtain suﬃciently high-resolution HI imaging of UDGs. We can use this data to model these UDGs using standard methods (Rogstad et al., 1974; Sicking, 1997; Kamphuis et al., 2015) in contrast to novel techniques (see Section 4.6, Mancera Piña et al., 2020). One of these

UDGs have been observed to-date and we are excited to carry out the data reduction and analysis of these data. Chapter 6

Summary and Conclusions

6.1 Summary

In this thesis, I have presented studies that characterize the HI gas and star-forming properties of galaxies in the dwarf regime in order to further our understanding of galaxy formation and evolution. The first two chapters presented two studies that focus on dwarf galaxy satellites around nearby Milky Way-like galaxies. The final two chapters presented the pilot and preliminary full survey results of the HI follow-up survey of ultra-diffuse galaxy (UDG) candidates from the Systematically Measuring Ultra-Diffuse Galaxies (SMUDGes) Survey. In this final chapter, I will summarize the main results from these works, describe potential avenues forward in each field, and conclude.

6.1.1 Satellite Galaxies around Milky Way-like Systems

This ﬁrst study, presented in Chapter2, investigates the LSB dwarf satellite candidates around the nearby Milky Way-like galaxy, M101. A total of 55 unique LSB satellite candidates were discovered in optical searches in the region around M101 (Merritt et al., 2014;

Karachentsev and Makarova, 2019; Javanmardi et al., 2016; Bennet et al., 2017a; Müller et al., 2017; Carlsten et al., 2019). By observing a magnitude-selected sample of 27 LSB

133 6.1. SUMMARY 134

dwarf satellite candidates with GBT, we aimed to (1) obtain distance measurements via their HI content (i.e. redshifts) and conﬁrm them bonaﬁde satellites and (2) understand whether the environmental dependence seen in the Local Group (i.e. satellites within the virial radii around the Milky Way or M31 are quenched or gas poor, while those beyond it are star-forming and gas rich) is present in other systems.

We detected HI reservoirs associated with 5 of the 27 LSB candidates and we showed that one of them is consistent with being a member of the M101 group, three are likely members of the background group centered around NGC 5485, and one is a distant background spiral galaxy. This demonstrated the utility of wide bandpass HI follow-up observations. While completing this study, several others compiled and presented distance estimates via TRGB

(Merritt et al., 2016; Danieli et al., 2017; Bennet et al., 2019a) or SBF (Carlsten et al., 2019) measurements. Combining these distance measurements with the 22 HI non-detections from our observations proved to be very useful as we were able to show that 15 of these LSBs are likely associated with the NGC 5485 background group.

In our comparison of the distance dependence on satellite gas richness showed a clear consistency between the Milky Way and M101. We extended this comparison to include the systems from the larger SAGA (Geha et al., 2017a) survey which found that most of the satellites around 8 Milky Way-like systems are star-forming via Hα measurements, seemingly in contrast to the Milky Way satellites. We demonstrated that the SAGA satellite systems are consistent with the Milky Way and M101 when considering satellites within the magnitude limit used in the SAGA survey (i.e. Mr < 12.3 MV < 12.1). − ∼ − The second study, presented in Chapter3, continues investigating the environmental dependence on satellites around Milky Way-like systems from the SAGA survey. The SAGA

Stage II results (SAGA-II, Mao et al., 2021) around the expanded sample of 36 Milky Way- like hosts again ﬁnd the vast majority of satellites are star forming. Not only is this in contrast to the Milky Way/Local Group, it was also shown to contrast high-resolution simulations of Milky Way-like galaxies (Akins et al., 2021). Motivated by these intriguing results from 6.1. SUMMARY 135

SAGA-II, our study aimed to (1) confirm the star-forming classification used by the SAGA survey and obtain the first estimates of SFRs using archival GALEX UV data and (2) compare the SAGA-II satellite systems to comparable samples from the APOSTLE (Sawala et al., 2016; Fattahi et al., 2016) and Auriga (Grand et al., 2017) zoom-in hydrodynamical simulations to understand the similarities and differences between the star-forming and quenched satellites in these samples.

The ﬁrst goal of this study was motivated by the nature of the optical spectroscopic observations carried out by the SAGA survey. These single-ﬁbre observations placed a small

(1-2 arcsec) fibre at the optical centroid of the satellite candidates which miss star-forming regions which, in turn, resulted in the target being classified as quenched instead of star- forming. This was alleviated by our use of archival GALEX FUV and NUV imaging data and our curve-of-growth analysis to measure any emission from the satellites. We were able to confirm that over 80% of SAGA-II satellites with FUV and NUV imaging had statistically significant FUV, NUV, and Hα emission. Crucially, we detected NUV emission in 12 satellites that were undetected in Hα by SAGA-II, demonstrating the utility of the GALEX data.

We suggested that any lack of emission in either of the UV bands or Hα is likely due to the

observational reasons (i.e. data depth/sensitivity, ﬁbre positioning, etc.). However, it should

be noted that while the archival GALEX data were useful for the purpose of this study,

the varying imaging depth did preclude a quantitative statistical comparison between the

diﬀerent samples and is something that could be considered for future work.

For our comparison with the simulated satellite samples, we selected satellites from the

APOSTLE and Auriga simulations which provide two distinct yet comparable samples of

satellites around 24 and 37 Milky Way-like host galaxies from each respective simulation

suite. We selected the satellite samples following the same spatial and velocity selection

6 criteria from SAGA-II and only selected satellites with stellar masses greater than 1 10 M . × From the derived NUV magnitudes, we estimated the SFRs of the SAGA-II satellites and

compared them to the simulated satellites. We found broad agreement between the observed 6.1. SUMMARY 136

and simulated samples in the SFR-stellar mass plane and in their distributions of star- forming satellites.

Following this comparison, we turned to comparing the satellite quenched fractions in the observed and simulated samples. The key result is that the satellite quenched fractions in the

7 8 range of 10 . M∗/M . 10 from SAGA-II are signiﬁcantly lower than those from both APOSTLE and Auriga, even when incompleteness/interloper corrections are applied. We explored observational and simulation-related factors that could explain these contrasting results, and while a detailed investigation of them was beyond the scope of our study, the tension between them appears genuine. However, without a more detailed investigation into both of these aspects, it is unclear how much each one contributes to the diﬀerences in quenched satellite number and fractions we presented.

6.1.2 Ultra-Diﬀuse Galaxies

The second half of this thesis, Chapters4 and5, presented a substantial observational effort to conduct HI follow-up observations of optically-detected UDG candidates from the

SMUDGes (Zaritsky et al., 2019) survey with the GBT. The goals of this follow-up survey, dubbed SMUDGes in HI, were to (1) conﬁrm UDG candidates as bonaﬁde UDGs, (2) characterize their HI and optical properties, and (3) place constraints on their formation mechanisms.

Chapter4 presented the results from the pilot campaign targetting 70 UDG candidates in the Coma Cluster region. There were 18 HI detections from the pilot campaign, 9 of which are bonaﬁde UDGs and the remaining 9 are foreground dwarfs. We showed that their derived velocity widths and their positions in the HI mass-stellar mass, MHI M∗, plane − are consistent with the broader galaxy population from ALFALFA and the HI-bearing Ultra

Diﬀuse galaxies (HUDs, Leisman et al., 2017). We found that both UDGs and dwarfs with

HI detections tend to have bluer optical colours and more irregular morphologies compared

to the HI non-detections. Using our HI-conﬁrmed UDGs along with a other UDGs from the 6.1. SUMMARY 137

literature, we showed that there is a tentative trend between the gas richness, MHI /M∗, and size, Reff , of UDGs at a given stellar mass, as predicted by the star-formation feedback formation mechanism from di Cintio et al.(2017). However, we concluded that a larger sample size in required to make concrete conclusions about this trend. The ﬁnal key result from the pilot campaign is with respect to UDGs and the baryonic Tully-Fisher relation (BTFR).

While some (7/9) of our HI-confirmed UDGs are offset from the BTFR, commensurate with other UDGs with marginally-resolved HI imaging (i.e. Mancera Piña et al., 2020), their offsets could plausibly be explained by the use of potentially overestimated optical inclinations to estimate their rotation velocities. We argued that the smooth GALFIT models applied to clumpy UDGs could systematically underestimate (overestimate) the axial ratios (inclinations) and showed that best-fit GALFIT models with lower inclinations could feasibly be applied to the optical data and, ultimately, reconcile their offset from the BTFR.

In Chapter5, we provided an update on the SMUDGes in HI survey. Observations with the GBT for the the remaining campaigns have been completed, and the data reduction and analysis are complete for the first 3 of 5 observing campaigns (campaign 1 being the pilot survey presented in Chapter4). Campaigns 2 and 3 targetted 111 UDG candidates from an extended region around the broader Coma environment and from Stripe 82 using the same selection criteria and observational configuration as in the pilot campaign. From these campaigns a total of 13 UDG candidates were detected in HI and all but 3 satisfy the UDG surface brightness and size criteria. The total number of HI-confirmed UDGs from campaigns 1-3 is 19.

The selection criteria for campaigns 4 and 5 required that all 187 targets have visually discernible UV emission, in addition to optical selection criteria (i.e. g band magnitude, − surface brightness, and angular size), with the goal of increasing the number of HI detections.

Furthermore, these two campaigns also used a modified observation configuration to mitigate the increasing effects of RFI in our spectra. The data reduction for these campaigns is underway with the goal of all spectral data reduction and analysis being completed by Fall 6.1. SUMMARY 138

2021. Based on the frequency of our HI detections also having UV emission, we estimated that campaigns 4 and 5 should provide as many as 150 new HI detections, a signiﬁcant fraction of which would be UDGs.

The preliminary results from our analysis of the UDGs from campaigns 2 and 3 largely resemble those from our pilot campaign (see Chapters4 and5). The HI detections across the campaigns 1-3 are, on average, bluer in optical colour compared to the UDG candidates with HI non-detections. The distribution of velocity widths of the HI-conﬁrmed UDGs from campaigns 2 and 3 is consistent with those from the pilot campaign and the HUDs sample.

While the HI-confirmed UDGs from campaigns 2 and 3 are also broadly consistent with the larger galaxy population from ALFALFA, it is clear that these 10 UDGs are typically less gas rich than the HUDs sample. This result provides further credence to our suggestion that there may be differences in the populations of optically- and HI-selected UDGs. We concluded this section by revisiting the gas richness-size plane with the addition UDGs from campaigns 2 and 3. Qualitatively, we found that the potential trend between these two properties identified in the pilot study persists with this larger sample, as predicted by the star-formation feedback UDG formation mechanism (di Cintio et al., 2017). However, in our quantitative comparison, we found a more convincing trend in the lower stellar mass bin, as determined by the calculated χ2 value. The relatively large scatter in gas richness in the higher stellar mass bin along with a particular outlier drive up the calculated χ2 value. We explored the feasibility of shifting this outlier back into the scatter of its stellar mass bin and show that a small change in optical g r colour can sufficiently increase the implied − gas richness of this object to reconcile it with the relation implied by the other objects.

The results presented in Chapter5 display just some of the many investigative possibilities aﬀorded by increasing our sample of HI-conﬁrmed UDGs. 6.2. FUTURE WORK 139

6.2 Future Work

The novel studies presented in this work provide a glimpse into just a few of the many possibilities for constraining galaxy formation and evolution LSB dwarf galaxies. Prior to concluding, this section will outline a couple of interesting paths for each part of this thesis

(dwarf satellites and UDGs) that could be investigated further, as well as some of the broader avenues within these ﬁelds that can and will be explored with upcoming facilities and surveys.

6.2.1 Satellites around Milky Way-like Systems

The investigation presented in Chapter3 examined the apparent disconnect between observations and simulations of Milky Way-like galaxies and their satellite systems. Following one of the discussion points from this chapter, a thorough search of the ﬁelds around the SAGA systems are a necessity to truly grasp the population of satellite candidates around these hosts. This is especially important as the SAGA survey selects objects from multiple optical catalogs across diﬀerent surveys. It would be informative to create a detection pipeline similar to or modify that from the SMUDGes survey to search for any and all potential companions around the SAGA hosts.

A crucial observational step forward would be to obtain HI properties of these satellites.

These data will enable extremely valuable insight into the evolutionary state of the star- forming and quenched satellites from SAGA. For example, these data could demonstrate that some of these galaxies have not yet experienced the effects of their hosts and possess large, pristine HI reservoirs. On the other hand, some of these galaxies classified as star forming could be experiencing their final star formation episode and are devoid of HI. Furthermore, if these observations were conducted with sufficient spatial resolution and to sufficient depths, then they may reveal the interaction histories of these systems. Given the ubiquity of star- forming satellites around these systems one might expect there to be minimal, if any, sign 6.2. FUTURE WORK 140

of interaction in the form of HI clouds, streams, or tails. An ideal facility to conduct such observations is MeerKAT because of its sensitivity to faint HI emission and, crucially, its wide one degree ﬁeld of view allows for the entire virial radius of the SAGA systems to be observed in a single-pointing.

6.2.2 Ultra-Diﬀuse Galaxies

Beyond the future work that was presented in Section 5.3, a careful and detailed analysis into the modelling and characterization of blue, irregular population of LSBs and UDGs.

A key result in the pilot campaign (Section 4.6) showed that the smooth GALFIT models applied to these irregular UDGs may not accurately reflect their true morphologies which are crucial in the calculation of their rotation velocities. A possible first step would be to take a non-parametric approach, such as isophotal fitting, to model these objects instead of using a smooth Sérsic model as is universally done at present. Isophotal fitting implementations will allow for the isophotes to vary in size, eccentricity, and position angle. Modern iterations of this method make use of machine learning techniques to better optimize solutions (e.g. Stone et al., 2021). Taking this approach provides a useful first step beyond the fixed morphological parameters of the Sérsic model and a more accurate representation of these UDGs.

6.2.3 Upcoming Surveys

While much of the work presented here focuses on the HI and/or star-forming properties of UDGs, future of dwarf and LSB studies at optical wavelengths will oﬀer new avenues of exploration. One of the most anticipated observatories that will see its ﬁrst light as early as October 2022 is the Vera C. Rubin Observatory. The Rubin Observatory will carry out the Legacy Survey of Space and Time (LSST) which will observe the entire visible sky from its Chilean site every few days for 10 years (Ivezić et al., 2019). This unprecedented coverage and depth will provide an entirely new perspective for all aspects of observational astronomy. In particular, the Rubin Observatory and the LSST will enable the detection and 6.2. FUTURE WORK 141

characterization of a plethora of LSB dwarf galaxies out to greater distances than before, while also providing a more complete picture of more nearby ones down to even fainter magnitudes (Simon, 2019). In particular, the ﬁnal LSST combined imaging is predicted to eﬀectively double the number of satellites around the Milky way (Newton et al., 2018) but also detect new faint dwarf galaxies at distances as far as 5 Mpc (Mutlu-Pakdil et al., 2021).

If distance estimates are obtained for the unresolved LSB objects, then an analysis of LSB populations at diﬀerent redshifts may lead to interesting insight regarding their evolution.

The deeper imaging from LSST will be able to reveal any additional features of or around previously detected LSBs and may point towards any environmental eﬀects if they reside near dense environments.

Similar to the Rubin Observatory and the LSST, the Dragonfly Telephoto array (Abra- ham and van Dokkum, 2014) will, in its own right, continue its influence in the LSB regime with a recently announced1 expansion to its capabilities. The Dragonfly Telescope will expand from its current pair of 24 telephoto lens arrays (the equivalent to a 1 meter aperture) to 168 lenses. Already receiving recognition for its role in the detection of the numerous

UDGs in the Coma Cluster, the Dragonﬂy array will certainly continue to reach ever fainter depths and reveal interesting new features. Additionally, the planned application of an Hα

ﬁlter will enable searches for faint ionized gas clouds around massive and dwarf galaxies alike.

Returning to the radio regime, pilot observations using ASKAP (Australian Square Kilo- meter Array Pathﬁnder) are underway for the Wideﬁeld ASKAP L-band Legacy All-sky

Blind surveY (WALLABY Koribalski et al., 2020). WALLABY will cover a large fraction of the sky below declinations of 30 degrees and will have similar sensitivity to ALFALFA with much ﬁner spatial resolution. As a result, similar analyses as with ALFALFA can be conducted to match and characterize any new optically-detected LSB, UDG, and satellite dwarf galaxy candidates. While the spatial resolution of WALLABY may not suﬃciently resolve

1University of Toronto Press Release 6.3. CONCLUSIONS 142

most of these types of objects, it will allow for the detection and classiﬁcation of objects within the proximity of gas-rich neighbours without becoming confused by their emission as is the case for several targetted observations with single-dish telescopes.

6.3 Conclusions

I have presented 3 novel studies that aim to better understand satellites around Milky Way- like systems and the properties of UDGs in the context of their formation mechanisms.

The HI follow-up observations of LSB satellites around M101 demonstrated the utility of deep, targetted observations of these objects by detecting objects across a wide range of distances. Additionally, the sensitivity of these HI data combined with optical distance measurements were able to place strong constraints on the potential environments of these objects. These investigations were expanded to the satellites around Milky Way-like systems from the SAGA survey. Using archival GALEX data to characterize the star-forming properties of the satellites from this survey and comparing them to those from two distinct hydrodynamical zoom-in simulations (APOSTLE and Auriga), we presented a potential tension between the observed and simulated number and fraction of quenched satellites. These studies place an important foundation for future studies of HI and star-forming properties around larger samples of satellites around Milky Way-like systems. The ambitious efforts of the SMUDGes in HI survey exploit the HI reservoirs of optically-detected UDG candidates from the SMUDGes survey to estimate their distances and constrain their formation mechanisms. These data confirmed that gas-rich UDGs are typically blue and irregular in morphology. Crucially, the increased sample of HI-confirmed UDGs were able to, for the first time, quantitatively estimate the relationship between the gas richness and size for UDGs.

However, concrete conclusions require an even larger sample of UDGs.

It is clear that the combination of improvements in optical astronomical instrumentation 6.3. CONCLUSIONS 143

and subsequent analysis methods have reinvigorated studies of the LSB regime (e.g. Abra- ham and van Dokkum, 2014; Fliri and Trujillo, 2016a; Bennet et al., 2017a; Geha et al.,

2017b; Zaritsky et al., 2019). These data have revealed a plethora of new satellite dwarf galaxies in the local Universe and uncovered a significant population of LSBs, including the interesting subset of UDGs. Future surveys will only continue to add to these efforts and probe event fainter depths. Additionally, ever larger, high-resolution hydrodynamical simulations have provided, and will continue to provide, important theoretical constraints on the effect of massive galaxies on their low-mass neighbours as well as clues to the origin of UDGs. Continuing to compare these observational and simulation samples is a critical path forward to truly understand the evolutionary mechanisms at play in the low-mass regime. The novel work presented in this thesis has highlighted the pivotal role that HI and star-forming properties play in solidifying the interpretation of dwarf galaxy formation and evolution in a cosmological context. Bibliography

R. G. Abraham and P. G. van Dokkum. Ultra-Low Surface Brightness Imaging with the

Dragonﬂy Telephoto Array. Publications of the Astronomical Society of Paciﬁc, 126:55,

Jan. 2014. doi:10.1086/674875.

R. G. Abraham, N. R. Tanvir, B. X. Santiago, R. S. Ellis, K. Glazebrook, and S. van den

Bergh. Galaxy morphology to I=25 mag in the Hubble Deep Field. Monthly Notices of the

Royal Astronomical Society, 279(3):L47–L52, Apr. 1996. doi:10.1093/mnras/279.3.L47.

R. G. Abraham, S. van den Bergh, and P. Nair. A New Approach to Galaxy Morphology. I.

Analysis of the Sloan Digital Sky Survey Early Data Release. Astrophysical Journal, 588

(1):218–229, May 2003. doi:10.1086/373919.

J. K. Adelman-McCarthy, M. A. Agüeros, S. S. Allam, K. S. J. Anderson, S. F. Anderson,

J. Annis, N. A. Bahcall, C. A. L. Bailer-Jones, I. K. Baldry, J. C. Barentine, T. C.

Beers, V. Belokurov, A. Berlind, M. Bernardi, M. R. Blanton, J. J. Bochanski, W. N.

Boroski, D. M. Bramich, H. J. Brewington, J. Brinchmann, J. Brinkmann, R. J. Brunner,

T. Budavári, L. N. Carey, S. Carliles, M. A. Carr, F. J. Castander, A. J. Connolly, R. J.

Cool, C. E. Cunha, I. Csabai, J. J. Dalcanton, M. Doi, D. J. Eisenstein, M. L. Evans, N. W.

Evans, X. Fan, D. P. Finkbeiner, S. D. Friedman, J. A. Frieman, M. Fukugita, B. Gillespie,

G. Gilmore, K. Glazebrook, J. Gray, E. K. Grebel, J. E. Gunn, E. de Haas, P. B. Hall,

M. Harvanek, S. L. Hawley, J. Hayes, T. M. Heckman, J. S. Hendry, G. S. Hennessy,

144 BIBLIOGRAPHY 145

R. B. Hindsley, C. M. Hirata, C. J. Hogan, D. W. Hogg, J. A. Holtzman, S.-i. Ichikawa,

T. Ichikawa, Ž. Ivezić, S. Jester, D. E. Johnston, A. M. Jorgensen, M. Jurić, G. Kauﬀmann,

S. M. Kent, S. J. Kleinman, G. R. Knapp, A. Y. Kniazev, R. G. Kron, J. Krzesinski,

N. Kuropatkin, D. Q. Lamb, H. Lampeitl, B. C. Lee, R. F. Leger, M. Lima, H. Lin, D. C.

Long, J. Loveday, R. H. Lupton, R. Mandelbaum, B. Margon, D. Martínez-Delgado,

T. Matsubara, P. M. McGehee, T. A. McKay, A. Meiksin, J. A. Munn, R. Nakajima,

T. Nash, J. Neilsen, Eric H., H. J. Newberg, R. C. Nichol, M. Nieto-Santisteban, A. Nitta,

H. Oyaizu, S. Okamura, J. P. Ostriker, N. Padmanabhan, C. Park, J. Peoples, John, J. R.

Pier, A. C. Pope, D. Pourbaix, T. R. Quinn, M. J. Raddick, P. Re Fiorentin, G. T.

Richards, M. W. Richmond, H.-W. Rix, C. M. Rockosi, D. J. Schlegel, D. P. Schneider,

R. Scranton, U. Seljak, E. Sheldon, K. Shimasaku, N. M. Silvestri, J. A. Smith, V. Smolčić,

S. A. Snedden, A. Stebbins, C. Stoughton, M. A. Strauss, M. SubbaRao, Y. Suto, A. S.

Szalay, I. Szapudi, P. Szkody, M. Tegmark, A. R. Thakar, C. A. Tremonti, D. L. Tucker,

A. Uomoto, D. E. Vanden Berk, J. Vandenberg, S. Vidrih, M. S. Vogeley, W. Voges, N. P.

Vogt, D. H. Weinberg, A. A. West, S. D. M. White, B. Wilhite, B. Yanny, D. R. Yocum,

D. G. York, I. Zehavi, S. Zibetti, and D. B. Zucker. The Fifth Data Release of the Sloan

Digital Sky Survey. Astrophysical Journal Supplement Series, 172(2):634–644, Oct. 2007.

doi:10.1086/518864.

O. Agertz, B. Moore, J. Stadel, D. Potter, F. Miniati, J. Read, L. Mayer, A. Gawryszczak,

A. Kravtsov, Å. Nordlund, F. Pearce, V. Quilis, D. Rudd, V. Springel, J. Stone, E. Tasker,

R. Teyssier, J. Wadsley, and R. Walder. Fundamental diﬀerences between SPH and grid

methods. Monthly Notices of the Royal Astronomical Society, 380(3):963–978, Sept. 2007.

doi:10.1111/j.1365-2966.2007.12183.x.

D. S. Aguado, R. Ahumada, A. Almeida, et al. The Fifteenth Data Release of the Sloan

Digital Sky Surveys: First Release of MaNGA-derived Quantities, Data Visualization

Tools, and Stellar Library. Astrophysical Journal Supplement Series, 240(2):23, Feb 2019. BIBLIOGRAPHY 146

doi:10.3847/1538-4365/aaf651.

H. Aihara, N. Arimoto, R. Armstrong, et al. The Hyper Suprime-Cam SSP Survey: Overview

and survey design. Publications of the Astronomical Society of Japan, 70:S4, Jan 2018.

doi:10.1093/pasj/psx066.

H. B. Akins, C. R. Christensen, A. M. Brooks, F. Munshi, E. Applebaum, A. Engelhardt,

and L. Chamberland. Quenching Timescales of Dwarf Satellites around Milky Way-mass

Hosts. Astrophysical Journal, 909(2):139, Mar. 2021. doi:10.3847/1538-4357/abe2ab.

A. Alabi, A. Ferré-Mateu, A. J. Romanowsky, et al. Origins of ultradiﬀuse galaxies in the

Coma cluster - I. Constraints from velocity phase space. Monthly Notices of the Royal

Astronomical Society, 479(3):3308–3318, Sept. 2018. doi:10.1093/mnras/sty1616.

N. C. Amorisco and A. Loeb. Ultradiﬀuse galaxies: the high-spin tail of the abundant dwarf

galaxy population. Monthly Notices of the Royal Astronomical Society, 459:L51–L55, June

2016. doi:10.1093/mnrasl/slw055.

N. C. Amorisco, A. Monachesi, A. Agnello, and S. D. M. White. The globular

cluster systems of 54 Coma ultra-diﬀuse galaxies: statistical constraints from HST

data. Monthly Notices of the Royal Astronomical Society, 475(3):4235–4251, Apr. 2018.

doi:10.1093/mnras/sty116.

J. Annis, M. Soares-Santos, M. A. Strauss, A. C. Becker, S. Dodelson, X. Fan, J. E. Gunn,

J. Hao, Ž. Ivezić, S. Jester, L. Jiang, D. E. Johnston, J. M. Kubo, H. Lampeitl, H. Lin,

R. H. Lupton, G. Miknaitis, H.-J. Seo, M. Simet, and B. Yanny. The Sloan Digital

Sky Survey Coadd: 275 deg2 of Deep Sloan Digital Sky Survey Imaging on Stripe 82.

Astrophysical Journal, 794(2):120, Oct. 2014. doi:10.1088/0004-637X/794/2/120. BIBLIOGRAPHY 147

Y. M. Bahé, I. G. McCarthy, M. L. Balogh, and A. S. Font. Why does the environmental in-

ﬂuence on group and cluster galaxies extend beyond the virial radius? Monthly Notices of

the Royal Astronomical Society, 430(4):3017–3031, Apr 2013. doi:10.1093/mnras/stt109.

K. Barbary. SEP: Source Extractor as a library. The Journal of Open Source Software, 1

(6):58, Oct. 2016. doi:10.21105/joss.00058.

C. E. Barbosa, D. Zaritsky, R. Donnerstein, et al. One Hundred SMUDGes in S-PLUS:

Ultra-diﬀuse Galaxies Flourish in the Field. Astrophysical Journal Supplement Series,

247(2):46, Apr. 2020. doi:10.3847/1538-4365/ab7660.

M. A. Beasley and I. Trujillo. Globular Clusters Indicate That Ultra-diﬀuse Galaxies Are

Dwarfs. Astrophysical Journal, 830(1):23, Oct. 2016. doi:10.3847/0004-637X/830/1/23.

M. A. Beasley, A. J. Romanowsky, V. Pota, et al. An Overmassive Dark Halo around an

Ultra-diﬀuse Galaxy in the Virgo Cluster. Astrophysical Journal Letters, 819(2):L20, Mar.

2016. doi:10.3847/2041-8205/819/2/L20.

K. Bechtol, A. Drlica-Wagner, E. Balbinot, A. Pieres, J. D. Simon, B. Yanny, B. Santiago,

R. H. Wechsler, J. Frieman, A. R. Walker, P. Williams, E. Rozo, E. S. Rykoﬀ, A. Queiroz,

E. Luque, A. Benoit-Lévy, D. Tucker, I. Sevilla, R. A. Gruendl, L. N. da Costa, A. Fausti

Neto, M. A. G. Maia, T. Abbott, S. Allam, R. Armstrong, A. H. Bauer, G. M. Bern-

stein, R. A. Bernstein, E. Bertin, D. Brooks, E. Buckley-Geer, D. L. Burke, A. Carnero

Rosell, F. J. Castander, R. Covarrubias, C. B. D’Andrea, D. L. DePoy, S. Desai, H. T.

Diehl, T. F. Eiﬂer, J. Estrada, A. E. Evrard, E. Fernandez, D. A. Finley, B. Flaugher,

E. Gaztanaga, D. Gerdes, L. Girardi, M. Gladders, D. Gruen, G. Gutierrez, J. Hao,

K. Honscheid, B. Jain, D. James, S. Kent, R. Kron, K. Kuehn, N. Kuropatkin, O. La-

hav, T. S. Li, H. Lin, M. Makler, M. March, J. Marshall, P. Martini, K. W. Merritt,

C. Miller, R. Miquel, J. Mohr, E. Neilsen, R. Nichol, B. Nord, R. Ogando, J. Peoples,

D. Petravick, A. A. Plazas, A. K. Romer, A. Roodman, M. Sako, E. Sanchez, V. Scarpine, BIBLIOGRAPHY 148

M. Schubnell, R. C. Smith, M. Soares-Santos, F. Sobreira, E. Suchyta, M. E. C. Swanson,

G. Tarle, J. Thaler, D. Thomas, W. Wester, J. Zuntz, and DES Collaboration. Eight New

Milky Way Companions Discovered in First-year Dark Energy Survey Data. Astrophysical

Journal, 807(1):50, July 2015. doi:10.1088/0004-637X/807/1/50.

E. F. Bell, D. Barnaby, R. G. Bower, et al. The star formation histories of low surface

brightness galaxies. Monthly Notices of the Royal Astronomical Society, 312(3):470–496,

Mar. 2000. doi:10.1046/j.1365-8711.2000.03180.x.

E. F. Bell, C. T. Slater, and N. F. Martin. Andromeda XXIX: A New Dwarf Spheroidal

Galaxy 200 kpc from Andromeda. Astrophysical Journal Letters, 742(1):L15, Nov. 2011.

doi:10.1088/2041-8205/742/1/L15.

M. Bellazzini, V. Belokurov, L. Magrini, F. Fraternali, V. Testa, G. Beccari, A. Marchetti,

and R. Carini. Redshift, metallicity and size of two extended dwarf Irregular galaxies: a

link between dwarf Irregulars and ultra diﬀuse galaxies? Monthly Notices of the Royal

Astronomical Society, 467:3751–3758, May 2017. doi:10.1093/mnras/stx236.

P. Bennet, D. J. Sand, D. Crnojević, K. Spekkens, D. Zaritsky, and A. Karunakaran. Dis-

covery of Diﬀuse Dwarf Galaxy Candidates around M101. Astrophysical Journal, 850(1):

109, Nov 2017a. doi:10.3847/1538-4357/aa9180.

P. Bennet, D. J. Sand, D. Crnojević, K. Spekkens, D. Zaritsky, and A. Karunakaran. Dis-

covery of Diﬀuse Dwarf Galaxy Candidates around M101. Astrophysical Journal, 850(1):

109, Nov 2017b. doi:10.3847/1538-4357/aa9180.

P. Bennet, D. J. Sand, D. Zaritsky, D. Crnojević, K. Spekkens, and A. Karunakaran. Ev-

idence for Ultra-diﬀuse Galaxy Formation through Galaxy Interactions. Astrophysical

Journal Letters, 866:L11, Oct. 2018. doi:10.3847/2041-8213/aadedf.

P. Bennet, D. J. Sand, D. Crnojević, K. Spekkens, A. Karunakaran, D. Zaritsky, BIBLIOGRAPHY 149

and B. Mutlu-Pakdil. The M101 Satellite Luminosity Function and the Halo-Halo

Scatter among Local Volume Hosts. Astrophysical Journal, 885(2):153, Nov. 2019a.

doi:10.3847/1538-4357/ab46ab.

P. Bennet, D. J. Sand, D. Crnojević, K. Spekkens, A. Karunakaran, D. Zaritsky, and

B. Mutlu-Pakdil. The M101 Satellite Luminosity Function and the Halo to Halo Scatter

Among Milky Way Analogues. arXiv e-prints, art. arXiv:1906.03230, Jun 2019b.

P. Bennet, D. J. Sand, D. Crnojević, K. Spekkens, A. Karunakaran, D. Zaritsky, and

B. Mutlu-Pakdil. The Satellite Luminosity Function of M101 into the Ultra-faint Dwarf

Galaxy Regime. Astrophysical Journal Letters, 893(1):L9, Apr. 2020. doi:10.3847/2041-

8213/ab80c5.

E. Bertin and S. Arnouts. SExtractor: Software for source extraction. Astronomy & Astro-

physics Supplement, 117:393–404, June 1996. doi:10.1051/aas:1996164.

L. Bianchi. The Ultraviolet sky surveys: ﬁlling the gap in our view of the Universe. Astro-

physics and Space Science, 320(1-3):11–19, Apr. 2009. doi:10.1007/s10509-008-9761-3.

B. Binggeli, A. Sandage, and M. Tarenghi. Studies of the Virgo Cluster. I. Photometry

of 109 galaxies near the cluster center to serve as standards. Astronomical Journal, 89:

64–82, Jan. 1984. doi:10.1086/113484.

M. R. Blanton, E. Kazin, D. Muna, B. A. Weaver, and A. Price-Whelan. Improved Back-

ground Subtraction for the Sloan Digital Sky Survey Images. Astronomical Journal, 142

(1):31, July 2011. doi:10.1088/0004-6256/142/1/31.

S. Bose, C. S. Frenk, A. Jenkins, A. Fattahi, F. A. Gómez, R. J. J. Grand, F. Marinacci,

J. F. Navarro, K. A. Oman, R. Pakmor, J. Schaye, C. M. Simpson, and V. Springel. BIBLIOGRAPHY 150

No cores in dark matter-dominated dwarf galaxies with bursty star formation histo-

ries. Monthly Notices of the Royal Astronomical Society, 486(4):4790–4804, July 2019.

doi:10.1093/mnras/stz1168.

A. Boselli, S. Boissier, S. Heinis, L. Cortese, O. Ilbert, T. Hughes, O. Cucciati, J. Davies,

L. Ferrarese, R. Giovanelli, M. P. Haynes, M. Baes, C. Balkowski, N. Brosch, S. C. Chap-

man, V. Charmandaris, M. S. Clemens, A. Dariush, I. De Looze, S. di Serego Alighieri,

P. A. Duc, P. R. Durrell, E. Emsellem, T. Erben, J. Fritz, D. A. Garcia-Appadoo,

G. Gavazzi, M. Grossi, A. Jordán, K. M. Hess, M. Huertas-Company, L. K. Hunt, B. R.

Kent, D. G. Lambas, A. Lançon, L. A. MacArthur, S. C. Madden, L. Magrini, S. Mei,

E. Momjian, R. P. Olowin, E. Papastergis, M. W. L. Smith, J. M. Solanes, O. Spec-

tor, K. Spekkens, J. E. Taylor, C. Valotto, W. van Driel, J. Verstappen, C. Vlahakis,

B. Vollmer, and E. M. Xilouris. The GALEX Ultraviolet Virgo Cluster Survey (GUViCS).

I. The UV luminosity function of the central 12 sq. deg. Astronomy & Astrophysics, 528:

A107, Apr. 2011. doi:10.1051/0004-6361/201016389.

A. Boselli, M. Fossati, L. Ferrarese, S. Boissier, G. Consolandi, A. Longobardi, P. Am-

ram, M. Balogh, P. Barmby, M. Boquien, F. Boulanger, J. Braine, V. Buat, D. Bur-

garella, F. Combes, T. Contini, L. Cortese, P. Côté, S. Côté, J. C. Cuillandre, L. Drissen,

B. Epinat, M. Fumagalli, S. Gallagher, G. Gavazzi, J. Gomez-Lopez, S. Gwyn, W. Har-

ris, G. Hensler, B. Koribalski, M. Marcelin, A. McConnachie, M. A. Miville-Deschenes,

J. Navarro, D. Patton, E. W. Peng, H. Plana, N. Prantzos, C. Robert, J. Roediger,

Y. Roehlly, D. Russeil, P. Salome, R. Sanchez-Janssen, P. Serra, K. Spekkens, M. Sun,

J. Taylor, S. Tonnesen, B. Vollmer, J. Willis, H. Wozniak, T. Burdullis, D. Devost, B. Ma-

honey, N. Manset, A. Petric, S. Prunet, and K. Withington. A Virgo Environmental Sur-

vey Tracing Ionised Gas Emission (VESTIGE). I. Introduction to the survey. Astronomy

& Astrophysics, 614:A56, June 2018. doi:10.1051/0004-6361/201732407. BIBLIOGRAPHY 151

G. D. Bothun, C. D. Impey, D. F. Malin, and J. R. Mould. Discovery of a Huge Low-Surface-

Brightness Galaxy: A Proto-Disk Galaxy at Low Redshift? Astronomical Journal, 94:23,

July 1987. doi:10.1086/114443.

G. D. Bothun, C. D. Impey, and D. F. Malin. Extremely Low Surface Brightness Galaxies

in the Fornax Cluster: Properties, Stability, and Luminosity Fluctuations. Astrophysical

Journal, 376:404, Aug. 1991. doi:10.1086/170290.

J. D. Bradford, M. C. Geha, and M. R. Blanton. A Study in Blue: The Baryon Content of

Isolated Low-mass Galaxies. Astrophysical Journal, 809:146, Aug. 2015. doi:10.1088/0004-

637X/809/2/146.

L. Bradley, B. Sipõcz, T. Robitaille, E. Tollerud, Z. VinÃcius, C. Deil, K. Barbary, T. J.

Wilson, I. Busko, H. M. Günther, M. Cara, S. Conseil, A. Bostroem, M. Droettboom,

E. M. Bray, L. A. Bratholm, P. L. Lim, G. Barentsen, M. Craig, S. Pascual, G. Per-

ren, J. Greco, A. Donath, M. d. Val-Borro, W. Kerzendorf, Y. P. Bach, B. A. Weaver,

F. D’Eugenio, H. Souchereau, and L. Ferreira. astropy/photutils: 1.0.1, Sept. 2020. URL

https://doi.org/10.5281/zenodo.4049061.

L. D. Brown, T. T. Cai, and A. DasGupta. Interval estimation for a binomial proportion.

Statistical science, pages 101–117, 2001. doi:10.1214/ss/1009213286.

T. Brown, B. Catinella, L. Cortese, V. Kilborn, M. P. Haynes, and R. Giovanelli. The eﬀect

of structure and star formation on the gas content of nearby galaxies. Monthly Notices of

the Royal Astronomical Society, 452:2479–2489, Sept. 2015. doi:10.1093/mnras/stv1311.

T. Brown, B. Catinella, L. Cortese, C. d. P. Lagos, R. Davé, V. Kilborn, M. P. Haynes,

R. Giovanelli, and M. Raﬁeferantsoa. Cold gas stripping in satellite galaxies: from pairs

to clusters. Monthly Notices of the Royal Astronomical Society, 466(2):1275–1289, Apr

2017. doi:10.1093/mnras/stw2991. BIBLIOGRAPHY 152

T. Brown, C. Wilson, M. J. Jiménez, D. Bisaria, I. Roberts, J. Waldsley, K. Spekkens,

L. Parker, M. Thorp, R. Chown, S. Ellison, A. Boselli, C. Cashmore, E. Roediger, N. Zabel,

T. Davis, Y. Bahé, B. Lee, A. Bolatto, A. Mok, B. Diemer, C. Welker, K. P. Olsen, S. Ton-

nesen, S. Vogel, V. Villanueva, H. Yoon, R. Smith, A. Chung, A. Stevens, B. Catinella,

C. Lagos, L. Cortese, and P. Elahi. VERTICO: Virgo Environment Traced In CO. In Con-

tributions to the XIV.0 Scientiﬁc Meeting (virtual) of the Spanish Astronomical Society,

page 13, July 2020.

T. Buck, A. V. Macciò, A. A. Dutton, A. Obreja, and J. Frings. NIHAO XV: the

environmental impact of the host galaxy on galactic satellite and ﬁeld dwarf galax-

ies. Monthly Notices of the Royal Astronomical Society, 483(1):1314–1341, Feb. 2019.

doi:10.1093/mnras/sty2913.

J. S. Bullock and M. Boylan-Kolchin. Small-Scale Challenges to the ΛCDM

Paradigm. Annual Reviews of Astronomy and Astrophysics, 55(1):343–387, Aug. 2017.

doi:10.1146/annurev-astro-091916-055313.

J. M. Cannon, R. Giovanelli, M. P. Haynes, S. Janowiecki, A. Parker, J. J. Salzer, E. A. K.

Adams, E. Engstrom, S. Huang, K. B. W. McQuinn, J. Ott, A. Saintonge, E. D. Skillman,

J. Allan, G. Erny, P. Fliss, and A. Smith. The Survey of H I in Extremely Low-mass Dwarfs

(SHIELD). Astrophysical Journal Letters, 739(1):L22, Sept. 2011. doi:10.1088/2041-

8205/739/1/L22.

T. Carleton, R. Errani, M. Cooper, M. Kaplinghat, J. Peñarrubia, and Y. Guo. The for-

mation of ultra-diﬀuse galaxies in cored dark matter haloes through tidal stripping and

heating. Monthly Notices of the Royal Astronomical Society, 485:382–395, May 2019.

doi:10.1093/mnras/stz383. BIBLIOGRAPHY 153

S. G. Carlsten, R. L. Beaton, J. P. Greco, and J. E. Greene. Using Surface Brightness Fluctu-

ations to Study nearby Satellite Galaxy Systems: The Complete Satellite System of M101.

Astrophysical Journal Letters, 878(1):L16, Jun 2019. doi:10.3847/2041-8213/ab24d2.

S. G. Carlsten, J. P. Greco, R. L. Beaton, and J. E. Greene. Wide-ﬁeld Survey of Dwarf

Satellite Systems around 10 Hosts in the Local Volume. Astrophysical Journal, 891(2):

144, Mar. 2020. doi:10.3847/1538-4357/ab7758.

S. G. Carlsten, J. E. Greene, J. P. Greco, R. L. Beaton, and E. Kado-Fong. ELVES I:

Structures of Dwarf Satellites of MW-like Galaxies; Morphology, Scaling Relations, and

Intrinsic Shapes. arXiv e-prints, art. arXiv:2105.03435, May 2021a.

S. G. Carlsten, J. E. Greene, A. H. G. Peter, R. L. Beaton, and J. P. Greco. Luminos-

ity Functions and Host-to-host Scatter of Dwarf Satellite Systems in the Local Volume.

Astrophysical Journal, 908(1):109, Feb. 2021b. doi:10.3847/1538-4357/abd039.

B. Catinella, R. Giovanelli, and M. P. Haynes. Template Rotation Curves for Disk Galaxies.

Astrophysical Journal, 640(2):751–761, Apr. 2006. doi:10.1086/500171.

B. Catinella, D. Schiminovich, G. Kauﬀmann, S. Fabello, C. Hummels, J. Lemonias, S. M.

Moran, R. Wu, A. Cooper, and J. Wang. The GALEX Arecibo SDSS Survey. VI. Second

data release and updated gas fraction scaling relations. Astronomy & Astrophysics, 544:

A65, Aug. 2012. doi:10.1051/0004-6361/201219261.

R. Cen, A. Roxana Pop, and N. A. Bahcall. Gas loss in simulated galaxies as they fall into

clusters. Proceedings of the National Academy of Science, 111(22):7914–7919, Jun 2014.

doi:10.1073/pnas.1407300111.

K. Chiboucas, I. D. Karachentsev, and R. B. Tully. Discovery of New Dwarf Galaxies

in the M81 Group. Astronomical Journal, 137:3009–3037, Feb. 2009. doi:10.1088/0004-

6256/137/2/3009. BIBLIOGRAPHY 154

K. Chiboucas, B. A. Jacobs, R. B. Tully, and I. D. Karachentsev. Conﬁrmation of Faint

Dwarf Galaxies in the M81 Group. Astronomical Journal, 146(5):126, Nov. 2013a.

doi:10.1088/0004-6256/146/5/126.

K. Chiboucas, B. A. Jacobs, R. B. Tully, and I. D. Karachentsev. Conﬁrmation of

Faint Dwarf Galaxies in the M81 Group. Astronomical Journal, 146:126, Nov. 2013b.

doi:10.1088/0004-6256/146/5/126.

I. V. Chilingarian, A. V. Afanasiev, K. A. Grishin, D. Fabricant, and S. Moran. Internal

Dynamics and Stellar Content of Nine Ultra-diﬀuse Galaxies in the Coma Cluster Prove

Their Evolutionary Link with Dwarf Early-type Galaxies. Astrophysical Journal, 884(1):

79, Oct. 2019. doi:10.3847/1538-4357/ab4205.

C. J. Conselice. The Relationship between Stellar Light Distributions of Galaxies and Their

Formation Histories. Astrophysical Journal Supplement Series, 147(1):1–28, July 2003.

doi:10.1086/375001.

C. J. Conselice. Ultra-diﬀuse Galaxies Are a Subset of Cluster Dwarf Elliptical/Spheroidal

Galaxies. Research Notes of the American Astronomical Society, 2:43, Mar 2018.

doi:10.3847/2515-5172/aab7f6.

R. A. Crain, J. Schaye, R. G. Bower, M. Furlong, M. Schaller, T. Theuns, C. Dalla Vecchia,

C. S. Frenk, I. G. McCarthy, J. C. Helly, A. Jenkins, Y. M. Rosas-Guevara, S. D. M.

White, and J. W. Trayford. The EAGLE simulations of galaxy formation: calibration of

subgrid physics and model variations. Monthly Notices of the Royal Astronomical Society,

450(2):1937–1961, June 2015. doi:10.1093/mnras/stv725.

D. Crnojević, D. J. Sand, K. Spekkens, N. Caldwell, P. Guhathakurta, B. McLeod, A. Seth,

J. D. Simon, J. Strader, and E. Toloba. The Extended Halo of Centaurus A: Uncovering

Satellites, Streams, and Substructures. Astrophysical Journal, 823(1):19, May 2016a.

doi:10.3847/0004-637X/823/1/19. BIBLIOGRAPHY 155

D. Crnojević, D. J. Sand, K. Spekkens, N. Caldwell, P. Guhathakurta, B. McLeod, A. Seth,

J. D. Simon, J. Strader, and E. Toloba. The Extended Halo of Centaurus A: Uncov-

ering Satellites, Streams, and Substructures. Astrophysical Journal, 823:19, May 2016b.

doi:10.3847/0004-637X/823/1/19.

D. Crnojević, D. J. Sand, P. Bennet, S. Pasetto, K. Spekkens, N. Caldwell, P. Guhathakurta,

B. McLeod, A. Seth, J. D. Simon, J. Strader, and E. Toloba. The faint end of the

centaurus a satellite luminosity function. The Astrophysical Journal, 872(1):80, feb

2019. doi:10.3847/1538-4357/aafbe7. URL https://doi.org/10.3847%2F1538-4357%

2Faafbe7.

S. Danieli, P. van Dokkum, A. Merritt, R. Abraham, J. Zhang, I. D. Karachentsev, and L. N.

Makarova. The Dragonﬂy Nearby Galaxies Survey. III. The Luminosity Function of the

M101 Group. Astrophysical Journal, 837:136, Mar. 2017. doi:10.3847/1538-4357/aa615b.

J. I. Davies, S. Phillipps, M. G. M. Cawson, M. J. Disney, and E. J. Kibblewhite. Low

surface brightness galaxies in the Fornax cluster : automated galaxy surface photom-

etry. III. Monthly Notices of the Royal Astronomical Society, 232:239–258, May 1988.

doi:10.1093/mnras/232.2.239.

J. I. Davies, M. Baes, G. J. Bendo, S. Bianchi, D. J. Bomans, A. Boselli, M. Clemens,

E. Corbelli, L. Cortese, A. Dariush, I. De Looze, S. di Serego Alighieri, D. Fadda, J. Fritz,

D. A. Garcia-Appadoo, G. Gavazzi, C. Giovanardi, M. Grossi, T. M. Hughes, L. K. Hunt,

A. P. Jones, S. Madden, D. Pierini, M. Pohlen, S. Sabatini, M. W. L. Smith, J. Verstap-

pen, C. Vlahakis, E. M. Xilouris, and S. Zibetti. The Herschel Virgo Cluster Survey. I.

Luminosity function. Astronomy & Astrophysics, 518:L48, July 2010. doi:10.1051/0004-

6361/201014571.

J. I. Davies, L. J. M. Davies, and O. C. Keenan. Probing the low surface brightness dwarf BIBLIOGRAPHY 156

galaxy population of the virgo cluster. Monthly Notices of the Royal Astronomical Society,

456:1607–1617, Feb 2016. doi:10.1093/mnras/stv2719.

L. J. M. Davies, A. S. G. Robotham, C. d. P. Lagos, S. P. Driver, A. R. H. Stevens,

Y. M. Bahé, M. Alpaslan, M. N. Bremer, M. J. I. Brown, S. Brough, J. Bland-Hawthorn,

L. Cortese, P. Elahi, M. W. Grootes, B. W. Holwerda, A. D. Ludlow, S. McGee, M. Ow-

ers, and S. Phillipps. Galaxy and Mass Assembly (GAMA): environmental quenching of

centrals and satellites in groups. Monthly Notices of the Royal Astronomical Society, 483

(4):5444–5458, Mar. 2019. doi:10.1093/mnras/sty3393.

W. J. G. de Blok, S. S. McGaugh, and J. M. van der Hulst. HI observations of low surface

brightness galaxies: probing low-density galaxies. Monthly Notices of the Royal Astro-

nomical Society, 283(1):18–54, Nov. 1996. doi:10.1093/mnras/283.1.18.

A. Dey, D. J. Schlegel, D. Lang, R. Blum, K. Burleigh, X. Fan, J. R. Findlay, D. Finkbeiner,

D. Herrera, S. Juneau, M. Landriau, M. Levi, I. McGreer, A. Meisner, A. D. Myers,

J. Moustakas, P. Nugent, A. Patej, E. F. Schlaﬂy, A. R. Walker, F. Valdes, B. A. Weaver,

C. Yèche, H. Zou, X. Zhou, B. Abareshi, T. M. C. Abbott, B. Abolfathi, C. Aguilera,

S. Alam, L. Allen, A. Alvarez, J. Annis, B. Ansarinejad, M. Aubert, J. Beechert, E. F.

Bell, S. Y. BenZvi, F. Beutler, R. M. Bielby, A. S. Bolton, C. Briceño, E. J. Buckley-

Geer, K. Butler, A. Calamida, R. G. Carlberg, P. Carter, R. Casas, F. J. Castander,

Y. Choi, J. Comparat, E. Cukanovaite, T. Delubac, K. DeVries, S. Dey, G. Dhungana,

M. Dickinson, Z. Ding, J. B. Donaldson, Y. Duan, C. J. Duckworth, S. Eftekharzadeh,

D. J. Eisenstein, T. Etourneau, P. A. Fagrelius, J. Farihi, M. Fitzpatrick, A. Font-Ribera,

L. Fulmer, B. T. Gänsicke, E. Gaztanaga, K. George, D. W. Gerdes, S. G. A. Gontcho,

C. Gorgoni, G. Green, J. Guy, D. Harmer, M. Hernandez, K. Honscheid, L. W. Huang,

D. J. James, B. T. Jannuzi, L. Jiang, R. Joyce, A. Karcher, S. Karkar, R. Kehoe, J.-P.

Kneib, A. Kueter-Young, T.-W. Lan, T. R. Lauer, L. Le Guillou, A. Le Van Suu, J. H. BIBLIOGRAPHY 157

Lee, M. Lesser, L. Perreault Levasseur, T. S. Li, J. L. Mann, R. Marshall, C. E. Martínez-

Vázquez, P. Martini, H. du Mas des Bourboux, S. McManus, T. G. Meier, B. Ménard,

N. Metcalfe, A. Muñoz-Gutiérrez, J. Najita, K. Napier, G. Narayan, J. A. Newman, J. Nie,

B. Nord, D. J. Norman, K. A. G. Olsen, A. Paat, N. Palanque-Delabrouille, X. Peng, C. L.

Poppett, M. R. Poremba, A. Prakash, D. Rabinowitz, A. Raichoor, M. Rezaie, A. N.

Robertson, N. A. Roe, A. J. Ross, N. P. Ross, G. Rudnick, S. Safonova, A. Saha, F. J.

Sánchez, E. Savary, H. Schweiker, A. Scott, H.-J. Seo, H. Shan, D. R. Silva, Z. Slepian,

C. Soto, D. Sprayberry, R. Staten, C. M. Stillman, R. J. Stupak, D. L. Summers, S. Sien

Tie, H. Tirado, M. Vargas-Magaña, A. K. Vivas, R. H. Wechsler, D. Williams, J. Yang,

Q. Yang, T. Yapici, D. Zaritsky, A. Zenteno, K. Zhang, T. Zhang, R. Zhou, and Z. Zhou.

Overview of the DESI Legacy Imaging Surveys. Astronomical Journal, 157(5):168, May

2019a. doi:10.3847/1538-3881/ab089d.

A. Dey, D. J. Schlegel, D. Lang, et al. Overview of the DESI Legacy Imaging Surveys.

Astronomical Journal, 157(5):168, May 2019b. doi:10.3847/1538-3881/ab089d.

A. di Cintio, C. B. Brook, A. A. Dutton, A. V. Macciò, A. Obreja, and A. Dekel. NIHAO - XI.

Formation of ultra-diﬀuse galaxies by outﬂows. Monthly Notices of the Royal Astronomical

Society, 466:L1–L6, Mar. 2017. doi:10.1093/mnrasl/slw210.

A. Di Cintio, C. B. Brook, A. V. Macciò, et al. NIHAO XXI: the emergence of low surface

brightness galaxies. Monthly Notices of the Royal Astronomical Society, 486(2):2535–2548,

Jun 2019. doi:10.1093/mnras/stz985.

E. M. Di Teodoro and F. Fraternali. 3D BAROLO: a new 3D algorithm to derive rotation

curves of galaxies. Monthly Notices of the Royal Astronomical Society, 451(3):3021–3033,

Aug. 2015. doi:10.1093/mnras/stv1213.

R. Digby, J. F. Navarro, A. Fattahi, C. M. Simpson, K. A. Oman, F. A. Gomez, C. S. Frenk,

R. J. J. Grand, and R. Pakmor. The star formation histories of dwarf galaxies in Local BIBLIOGRAPHY 158

Group cosmological simulations. Monthly Notices of the Royal Astronomical Society, 485

(4):5423–5437, June 2019. doi:10.1093/mnras/stz745.

A. Drlica-Wagner, K. Bechtol, E. S. Rykoﬀ, E. Luque, A. Queiroz, Y. Y. Mao, R. H.

Wechsler, J. D. Simon, B. Santiago, B. Yanny, E. Balbinot, S. Dodelson, A. Fausti Neto,

D. J. James, T. S. Li, M. A. G. Maia, J. L. Marshall, A. Pieres, K. Stringer, A. R.

Walker, T. M. C. Abbott, F. B. Abdalla, S. Allam, A. Benoit-Lévy, G. M. Bernstein,

E. Bertin, D. Brooks, E. Buckley-Geer, D. L. Burke, A. Carnero Rosell, M. Carrasco

Kind, J. Carretero, M. Crocce, L. N. da Costa, S. Desai, H. T. Diehl, J. P. Dietrich,

P. Doel, T. F. Eiﬂer, A. E. Evrard, D. A. Finley, B. Flaugher, P. Fosalba, J. Frieman,

E. Gaztanaga, D. W. Gerdes, D. Gruen, R. A. Gruendl, G. Gutierrez, K. Honscheid,

K. Kuehn, N. Kuropatkin, O. Lahav, P. Martini, R. Miquel, B. Nord, R. Ogando, A. A.

Plazas, K. Reil, A. Roodman, M. Sako, E. Sanchez, V. Scarpine, M. Schubnell, I. Sevilla-

Noarbe, R. C. Smith, M. Soares-Santos, F. Sobreira, E. Suchyta, M. E. C. Swanson,

G. Tarle, D. Tucker, V. Vikram, W. Wester, Y. Zhang, J. Zuntz, and DES Collaboration.

Eight Ultra-faint Galaxy Candidates Discovered in Year Two of the Dark Energy Survey.

Astrophysical Journal, 813(2):109, Nov. 2015. doi:10.1088/0004-637X/813/2/109.

A. Drlica-Wagner, K. Bechtol, S. Mau, M. McNanna, E. O. Nadler, A. B. Pace, T. S. Li,

A. Pieres, E. Rozo, J. D. Simon, A. R. Walker, R. H. Wechsler, T. M. C. Abbott, S. Allam,

J. Annis, E. Bertin, D. Brooks, D. L. Burke, A. C. Rosell, M. Carrasco Kind, J. Carretero,

M. Costanzi, L. N. da Costa, J. De Vicente, S. Desai, H. T. Diehl, P. Doel, T. F. Eiﬂer,

S. Everett, B. Flaugher, J. Frieman, J. García-Bellido, E. Gaztanaga, D. Gruen, R. A.

Gruendl, J. Gschwend, G. Gutierrez, K. Honscheid, D. J. James, E. Krause, K. Kuehn,

N. Kuropatkin, O. Lahav, M. A. G. Maia, J. L. Marshall, P. Melchior, F. Menanteau,

R. Miquel, A. Palmese, A. A. Plazas, E. Sanchez, V. Scarpine, M. Schubnell, S. Ser-

rano, I. Sevilla-Noarbe, M. Smith, E. Suchyta, G. Tarle, and DES Collaboration. Milky BIBLIOGRAPHY 159

Way Satellite Census. I. The Observational Selection Function for Milky Way Satel-

lites in DES Y3 and Pan-STARRS DR1. Astrophysical Journal, 893(1):47, Apr. 2020.

doi:10.3847/1538-4357/ab7eb9.

K. El-Badry, A. Wetzel, M. Geha, P. F. Hopkins, D. Kereš, T. K. Chan, and C.-A. Faucher-

Giguère. Breathing FIRE: How Stellar Feedback Drives Radial Migration, Rapid Size

Fluctuations, and Population Gradients in Low-mass Galaxies. Astrophysical Journal,

820(2):131, Apr 2016. doi:10.3847/0004-637X/820/2/131.

A. Emerick, M.-M. Mac Low, J. Grcevich, and A. Gatto. Gas Loss by Ram Pressure

Stripping and Internal Feedback from Low-mass Milky Way Satellites. Astrophysical

Journal, 826(2):148, Aug 2016. doi:10.3847/0004-637X/826/2/148.

A. Fattahi, J. F. Navarro, T. Sawala, C. S. Frenk, K. A. Oman, R. A. Crain, M. Furlong,

M. Schaller, J. Schaye, T. Theuns, and A. Jenkins. The APOSTLE project: Local Group

kinematic mass constraints and simulation candidate selection. Monthly Notices of the

Royal Astronomical Society, 457(1):844–856, Mar. 2016. doi:10.1093/mnras/stv2970.

L. Ferrarese, P. Côté, J.-C. Cuillandre, S. D. J. Gwyn, E. W. Peng, L. A. MacArthur, P.-

A. Duc, A. Boselli, S. Mei, T. Erben, A. W. McConnachie, P. R. Durrell, J. C. Mihos,

A. Jordán, A. Lançon, T. H. Puzia, E. Emsellem, M. L. Balogh, J. P. Blakeslee, L. van

Waerbeke, R. Gavazzi, B. Vollmer, J. J. Kavelaars, D. Woods, N. M. Ball, S. Boissier,

S. Courteau, E. Ferriere, G. Gavazzi, H. Hildebrandt, P. Hudelot, M. Huertas-Company,

C. Liu, D. McLaughlin, Y. Mellier, M. Milkeraitis, D. Schade, C. Balkowski, F. Bournaud,

R. G. Carlberg, S. C. Chapman, H. Hoekstra, C. Peng, M. Sawicki, L. Simard, J. E. Tay-

lor, R. B. Tully, W. van Driel, C. D. Wilson, T. Burdullis, B. Mahoney, and N. Manset.

The Next Generation Virgo Cluster Survey (NGVS). I. Introduction to the Survey. Astro-

physical Journal Supplement Series, 200(1):4, May 2012. doi:10.1088/0067-0049/200/1/4.

A. Ferré-Mateu, A. Alabi, D. A. Forbes, A. J. Romanowsky, J. Brodie, V. Pandya, BIBLIOGRAPHY 160

I. Martín-Navarro, S. Bellstedt, A. Wasserman, M. B. Stone, and N. Okabe. Origins

of ultradiﬀuse galaxies in the Coma cluster - II. Constraints from their stellar popu-

lations. Monthly Notices of the Royal Astronomical Society, 479:4891–4906, Oct 2018.

doi:10.1093/mnras/sty1597.

C. E. Fielder, Y.-Y. Mao, J. A. Newman, A. R. Zentner, and T. C. Licquia. Predictably

missing satellites: subhalo abundances in Milky Way-like haloes. Monthly Notices of the

Royal Astronomical Society, 486(4):4545–4568, Jul 2019. doi:10.1093/mnras/stz1098.

S. P. Fillingham, M. C. Cooper, C. Wheeler, S. Garrison-Kimmel, M. Boylan-Kolchin, and

J. S. Bullock. Taking care of business in a ﬂash: constraining the time-scale for low-mass

satellite quenching with ELVIS. Monthly Notices of the Royal Astronomical Society, 454

(2):2039–2049, Dec. 2015. doi:10.1093/mnras/stv2058.

S. P. Fillingham, M. C. Cooper, A. B. Pace, M. Boylan-Kolchin, J. S. Bullock, S. Garrison-

Kimmel, and C. Wheeler. Under pressure: quenching star formation in low-mass satellite

galaxies via stripping. Monthly Notices of the Royal Astronomical Society, 463(2):1916–

1928, Dec. 2016a. doi:10.1093/mnras/stw2131.

S. P. Fillingham, M. C. Cooper, A. B. Pace, M. Boylan-Kolchin, J. S. Bullock, S. Garrison-

Kimmel, and C. Wheeler. Under pressure: quenching star formation in low-mass satellite

galaxies via stripping. Monthly Notices of the Royal Astronomical Society, 463:1916–1928,

Dec. 2016b. doi:10.1093/mnras/stw2131.

S. P. Fillingham, M. C. Cooper, M. Boylan-Kolchin, J. S. Bullock, S. Garrison-Kimmel, and

C. Wheeler. Environmental quenching of low-mass ﬁeld galaxies. Monthly Notices of the

Royal Astronomical Society, 477(4):4491–4498, July 2018a. doi:10.1093/mnras/sty958.

S. P. Fillingham, M. C. Cooper, M. Boylan-Kolchin, J. S. Bullock, S. Garrison-Kimmel, and

C. Wheeler. Environmental quenching of low-mass ﬁeld galaxies. Monthly Notices of the

Royal Astronomical Society, 477(4):4491–4498, Jul 2018b. doi:10.1093/mnras/sty958. BIBLIOGRAPHY 161

J. Fliri and I. Trujillo. The IAC Stripe 82 Legacy Project: a wide-area survey for faint

surface brightness astronomy. Monthly Notices of the Royal Astronomical Society, 456(2):

1359–1373, Feb. 2016a. doi:10.1093/mnras/stv2686.

J. Fliri and I. Trujillo. The IAC Stripe 82 Legacy Project: a wide-area survey for faint

surface brightness astronomy. Monthly Notices of the Royal Astronomical Society, 456:

1359–1373, Feb 2016b. doi:10.1093/mnras/stv2686.

A. S. Font, I. G. McCarthy, and V. Belokurov. Can cosmological simulations capture the

diverse satellite populations of observed Milky Way analogues? arXiv e-prints, art.

arXiv:2011.12974, Nov. 2020.

D. A. Forbes, A. Alabi, A. J. Romanowsky, et al. Globular clusters in Coma cluster ultra

diﬀuse galaxies (UDGs): evidence for two types of UDG? Monthly Notices of the Royal

Astronomical Society, 492(4):4874–4883, Mar. 2020. doi:10.1093/mnras/staa180.

M. Furlong, R. G. Bower, T. Theuns, J. Schaye, R. A. Crain, M. Schaller, C. Dalla Vec-

chia, C. S. Frenk, I. G. McCarthy, J. Helly, A. Jenkins, and Y. M. Rosas-Guevara.

Evolution of galaxy stellar masses and star formation rates in the EAGLE simula-

tions. Monthly Notices of the Royal Astronomical Society, 450(4):4486–4504, July 2015.

doi:10.1093/mnras/stv852.

S. Garrison-Kimmel, P. F. Hopkins, A. Wetzel, J. S. Bullock, M. Boylan-Kolchin, D. Kereš,

C.-A. Faucher-Giguère, K. El-Badry, A. Lamberts, E. Quataert, and R. Sanderson. The

Local Group on FIRE: dwarf galaxy populations across a suite of hydrodynamic simula-

tions. Monthly Notices of the Royal Astronomical Society, 487(1):1380–1399, July 2019a.

doi:10.1093/mnras/stz1317.

S. Garrison-Kimmel, A. Wetzel, P. F. Hopkins, R. Sanderson, K. El-Badry, A. Graus,

T. K. Chan, R. Feldmann, M. Boylan-Kolchin, C. C. Hayward, J. S. Bullock, A. Fitts, BIBLIOGRAPHY 162

J. Samuel, C. Wheeler, D. Kereš, and C.-A. Faucher-Giguère. Star formation histories of

dwarf galaxies in the FIRE simulations: dependence on mass and Local Group environ-

ment. Monthly Notices of the Royal Astronomical Society, 489(4):4574–4588, Nov 2019b.

doi:10.1093/mnras/stz2507.

A. Gatto, F. Fraternali, J. I. Read, F. Marinacci, H. Lux, and S. Walch. Unveiling the

corona of the Milky Way via ram-pressure stripping of dwarf satellites. Monthly Notices

of the Royal Astronomical Society, 433:2749–2763, Aug. 2013. doi:10.1093/mnras/stt896.

M. Geha, M. R. Blanton, R. Yan, and J. L. Tinker. A Stellar Mass Threshold for Quench-

ing of Field Galaxies. Astrophysical Journal, 757(1):85, Sept. 2012a. doi:10.1088/0004-

637X/757/1/85.

M. Geha, M. R. Blanton, R. Yan, and J. L. Tinker. A Stellar Mass Threshold for Quench-

ing of Field Galaxies. Astrophysical Journal, 757:85, Sept. 2012b. doi:10.1088/0004-

637X/757/1/85.

M. Geha, R. H. Wechsler, Y.-Y. Mao, E. J. Tollerud, B. Weiner, R. Bernstein, B. Hoyle,

S. Marchi, P. J. Marshall, R. Muñoz, and Y. Lu. The SAGA Survey. I. Satellite Galaxy

Populations around Eight Milky Way Analogs. Astrophysical Journal, 847(1):4, Sep 2017a.

doi:10.3847/1538-4357/aa8626.

M. Geha, R. H. Wechsler, Y.-Y. Mao, E. J. Tollerud, B. Weiner, R. Bernstein, B. Hoyle,

S. Marchi, P. J. Marshall, R. Muñoz, and Y. Lu. The SAGA Survey. I. Satellite Galaxy

Populations around Eight Milky Way Analogs. Astrophysical Journal, 847(1):4, Sept.

2017b. doi:10.3847/1538-4357/aa8626.

A. Gil de Paz, B. F. Madore, and O. Pevunova. Palomar/Las Campanas Imaging Atlas

of Blue Compact Dwarf Galaxies. I. Images and Integrated Photometry. Astrophysical

Journal Supplement Series, 147:29–59, July 2003. doi:10.1086/374737. BIBLIOGRAPHY 163

R. Giovanelli, M. P. Haynes, J. J. Salzer, et al. Extinction in SC Galaxies. Astronomical

Journal, 107:2036, June 1994. doi:10.1086/117014.

R. Giovanelli, M. P. Haynes, T. Herter, et al. The I band Tully-Fisher relation for

cluster galaxies: data presentation. Astronomical Journal, 113:22–52, Jan. 1997.

doi:10.1086/118233.

R. Giovanelli, M. P. Haynes, E. A. K. Adams, J. M. Cannon, K. L. Rhode, J. J. Salzer,

E. D. Skillman, E. Z. Bernstein-Cooper, and K. B. W. McQuinn. ALFALFA Discovery

of the Nearby Gas-rich Dwarf Galaxy Leo P. I. H I Observations. Astronomical Journal,

146(1):15, July 2013. doi:10.1088/0004-6256/146/1/15.

J. Goddy, D. V. Stark, and K. L. Masters. L-band Calibration of the Green Bank Telescope

from 2016-2019. Research Notes of the American Astronomical Society, 4(1):3, Jan. 2020.

doi:10.3847/2515-5172/ab66bd.

R. J. J. Grand, F. A. Gómez, F. Marinacci, R. Pakmor, V. Springel, D. J. R. Campbell,

C. S. Frenk, A. Jenkins, and S. D. M. White. The Auriga Project: the properties and

formation mechanisms of disc galaxies across cosmic time. Monthly Notices of the Royal

Astronomical Society, 467(1):179–207, May 2017. doi:10.1093/mnras/stx071.

R. J. J. Grand, F. van de Voort, J. Zjupa, F. Fragkoudi, F. A. Gómez, G. Kauﬀmann,

F. Marinacci, R. Pakmor, V. Springel, and S. D. M. White. Gas accretion and galactic

fountain ﬂows in the Auriga cosmological simulations: angular momentum and metal

redistribution. Monthly Notices of the Royal Astronomical Society, 490(4):4786–4803,

Dec. 2019. doi:10.1093/mnras/stz2928.

J. Grcevich and M. E. Putman. H I in Local Group Dwarf Galaxies and Stripping by

the Galactic Halo. Astrophysical Journal, 696:385-395, May 2009. doi:10.1088/0004-

637X/696/1/385. BIBLIOGRAPHY 164

J. P. Greco, J. E. Greene, M. A. Strauss, L. A. Macarthur, X. Flowers, A. D. Goulding,

S. Huang, J. H. Kim, Y. Komiyama, A. Leauthaud, L. Leisman, R. H. Lupton, C. Sifón,

and S.-Y. Wang. Illuminating Low Surface Brightness Galaxies with the Hyper Suprime-

Cam Survey. Astrophysical Journal, 857:104, Apr. 2018. doi:10.3847/1538-4357/aab842.

J. E. Gunn and I. Gott, J. Richard. On the Infall of Matter Into Clusters of Galax-

ies and Some Eﬀects on Their Evolution. Astrophysical Journal, 176:1, Aug. 1972.

doi:10.1086/151605.

Q. Guo, S. Cole, V. Eke, C. Frenk, and J. Helly. Spatial and luminosity distributions of

galactic satellites. Monthly Notices of the Royal Astronomical Society, 434(3):1838–1848,

Sept. 2013. doi:10.1093/mnras/stt903.

M. P. Haynes and R. Giovanelli. Neutral hydrogen in isolated galaxies. IV - Results for the

Arecibo sample. Astronomical Journal, 89:758–800, June 1984. doi:10.1086/113573.

M. P. Haynes, R. Giovanelli, A. M. Martin, K. M. Hess, A. Saintonge, E. A. K. Adams,

G. Hallenbeck, G. L. Hoﬀman, S. Huang, B. R. Kent, R. A. Koopmann, E. Papastergis,

S. Stierwalt, T. J. Balonek, D. W. Craig, S. J. U. Higdon, D. A. Kornreich, J. R. Miller,

A. A. O’Donoghue, R. P. Olowin, J. L. Rosenberg, K. Spekkens, P. Troischt, and E. M.

Wilcots. The Arecibo Legacy Fast ALFA Survey: The α.40 H I Source Catalog, Its Char-

acteristics and Their Impact on the Derivation of the H I Mass Function. Astronomical

Journal, 142:170, Nov. 2011. doi:10.1088/0004-6256/142/5/170.

M. P. Haynes, R. Giovanelli, B. R. Kent, E. A. K. Adams, T. J. Balonek, D. W. Craig,

D. Fertig, R. Finn, C. Giovanardi, G. Hallenbeck, K. M. Hess, G. L. Hoﬀman, S. Huang,

M. G. Jones, R. A. Koopmann, D. A. Kornreich, L. Leisman, J. Miller, C. Moorman,

J. O’Connor, A. O’Donoghue, E. Papastergis, P. Troischt, D. Stark, and L. Xiao. The

Arecibo Legacy Fast ALFA Survey: The ALFALFA Extragalactic H I Source Catalog.

Astrophysical Journal, 861(1):49, July 2018. doi:10.3847/1538-4357/aac956. BIBLIOGRAPHY 165

M. He, H. Wu, W. Du, et al. Edge-on H I-bearing Ultra-diﬀuse Galaxy Candidates in the

40% ALFALFA Catalog. Astrophysical Journal, 880(1):30, July 2019. doi:10.3847/1538-

4357/ab2710.

C. Henkel, B. Javanmardi, D. Martínez-Delgado, P. Kroupa, and K. Teuwen. DGSAT:

Dwarf Galaxy Survey with Amateur Telescopes. II. A catalogue of isolated nearby edge-

on disk galaxies and the discovery of new low surface brightness systems. Astronomy &

Astrophysics, 603:A18, July 2017. doi:10.1051/0004-6361/201730539.

D. Homma, M. Chiba, S. Okamoto, Y. Komiyama, M. Tanaka, M. Tanaka, M. N. Ishigaki,

M. Akiyama, N. Arimoto, J. A. Garmilla, R. H. Lupton, M. A. Strauss, H. Furusawa,

S. Miyazaki, H. Murayama, A. J. Nishizawa, M. Takada, T. Usuda, and S.-Y. Wang. A

New Milky Way Satellite Discovered in the Subaru/Hyper Suprime-Cam Survey. Astro-

physical Journal, 832(1):21, Nov. 2016. doi:10.3847/0004-637X/832/1/21.

D. Homma, M. Chiba, S. Okamoto, Y. Komiyama, M. Tanaka, M. Tanaka, M. N. Ishigaki,

K. Hayashi, N. Arimoto, J. A. Garmilla, R. H. Lupton, M. A. Strauss, S. Miyazaki, S.-Y.

Wang, and H. Murayama. Searches for new Milky Way satellites from the ﬁrst two years

of data of the Subaru/Hyper Suprime-Cam survey: Discovery of Cetus III. Publications

of the Astronomical Society of Japan, 70:S18, Jan. 2018. doi:10.1093/pasj/psx050.

S. Huang, M. P. Haynes, R. Giovanelli, and J. Brinchmann. The Arecibo Legacy Fast ALFA

Survey: The Galaxy Population Detected by ALFALFA. Astrophysical Journal, 756(2):

113, Sept. 2012a. doi:10.1088/0004-637X/756/2/113.

S. Huang, M. P. Haynes, R. Giovanelli, J. Brinchmann, S. Stierwalt, and S. G. Neﬀ. Gas,

Stars, and Star Formation in ALFALFA Dwarf Galaxies. Astronomical Journal, 143:133,

June 2012b. doi:10.1088/0004-6256/143/6/133.

W. K. Huchtmeier, Karachentsev, I. D., and Karachentseva, V. E. Hi-observations of BIBLIOGRAPHY 166

dwarf galaxies in the local supercluster*. A&A, 506(2):677–680, 2009. doi:10.1051/0004-

6361/200911774. URL https://doi.org/10.1051/0004-6361/200911774.

D. A. Hunter, D. Ficut-Vicas, T. Ashley, et al. Little Things. Astronomical Journal, 144

(5):134, Nov. 2012. doi:10.1088/0004-6256/144/5/134.

J. Iglesias-Páramo, V. Buat, T. T. Takeuchi, K. Xu, S. Boissier, A. Boselli, D. Burgarella,

B. F. Madore, A. Gil de Paz, L. Bianchi, T. A. Barlow, Y. I. Byun, J. Donas, K. Forster,

P. G. Friedman, T. M. Heckman, P. N. Jelinski, Y. W. Lee, R. F. Malina, D. C. Martin,

B. Milliard, P. F. Morrissey, S. G. Neﬀ, R. M. Rich, D. Schiminovich, M. Seibert, O. H. W.

Siegmund, T. Small, A. S. Szalay, B. Y. Welsh, and T. K. Wyder. Star Formation in the

Nearby Universe: The Ultraviolet and Infrared Points of View. Astrophysical Journal

Supplement Series, 164(1):38–51, May 2006. doi:10.1086/502628.

C. Impey and G. Bothun. Low Surface Brightness Galaxies. Annual Reviews of Astronomy

and Astrophysics, 35:267–307, Jan. 1997. doi:10.1146/annurev.astro.35.1.267.

C. Impey, G. Bothun, and D. Malin. Virgo Dwarfs: New Light on Faint Galaxies. Astro-

physical Journal, 330:634, July 1988. doi:10.1086/166500.

G. Iorio, F. Fraternali, C. Nipoti, E. Di Teodoro, J. I. Read, and G. Battaglia. LITTLE

THINGS in 3D: robust determination of the circular velocity of dwarf irregular galax-

ies. Monthly Notices of the Royal Astronomical Society, 466(4):4159–4192, Apr. 2017.

doi:10.1093/mnras/stw3285.

M. J. Irwin, J. I. Davies, M. J. Disney, and S. Phillipps. Automated galaxy surface photom-

etry. V. Detection of very low surface brightness galaxies. Monthly Notices of the Royal

Astronomical Society, 245:289, July 1990.

Ž. Ivezić, S. M. Kahn, J. A. Tyson, B. Abel, E. Acosta, R. Allsman, D. Alonso, Y. Al-

Sayyad, S. F. Anderson, J. Andrew, J. R. P. Angel, G. Z. Angeli, R. Ansari, P. Antilogus, BIBLIOGRAPHY 167

C. Araujo, R. Armstrong, K. T. Arndt, P. Astier, É. Aubourg, N. Auza, T. S. Axelrod,

D. J. Bard, J. D. Barr, A. Barrau, J. G. Bartlett, A. E. Bauer, B. J. Bauman, S. Bau-

mont, E. Bechtol, K. Bechtol, A. C. Becker, J. Becla, C. Beldica, S. Bellavia, F. B. Bianco,

R. Biswas, G. Blanc, J. Blazek, R. D. Blandford, J. S. Bloom, J. Bogart, T. W. Bond, M. T.

Booth, A. W. Borgland, K. Borne, J. F. Bosch, D. Boutigny, C. A. Brackett, A. Brad-

shaw, W. N. Brandt, M. E. Brown, J. S. Bullock, P. Burchat, D. L. Burke, G. Cagnoli,

D. Calabrese, S. Callahan, A. L. Callen, J. L. Carlin, E. L. Carlson, S. Chandrasekha-

ran, G. Charles-Emerson, S. Chesley, E. C. Cheu, H.-F. Chiang, J. Chiang, C. Chirino,

D. Chow, D. R. Ciardi, C. F. Claver, J. Cohen-Tanugi, J. J. Cockrum, R. Coles, A. J. Con-

nolly, K. H. Cook, A. Cooray, K. R. Covey, C. Cribbs, W. Cui, R. Cutri, P. N. Daly, S. F.

Daniel, F. Daruich, G. Daubard, G. Daues, W. Dawson, F. Delgado, A. Dellapenna, R. de

Peyster, M. de Val-Borro, S. W. Digel, P. Doherty, R. Dubois, G. P. Dubois-Felsmann,

J. Durech, F. Economou, T. Eiﬂer, M. Eracleous, B. L. Emmons, A. Fausti Neto, H. Fer-

guson, E. Figueroa, M. Fisher-Levine, W. Focke, M. D. Foss, J. Frank, M. D. Freemon,

E. Gangler, E. Gawiser, J. C. Geary, P. Gee, M. Geha, C. J. B. Gessner, R. R. Gib-

son, D. K. Gilmore, T. Glanzman, W. Glick, T. Goldina, D. A. Goldstein, I. Goodenow,

M. L. Graham, W. J. Gressler, P. Gris, L. P. Guy, A. Guyonnet, G. Haller, R. Har-

ris, P. A. Hascall, J. Haupt, F. Hernandez, S. Herrmann, E. Hileman, J. Hoblitt, J. A.

Hodgson, C. Hogan, J. D. Howard, D. Huang, M. E. Huﬀer, P. Ingraham, W. R. Innes,

S. H. Jacoby, B. Jain, F. Jammes, M. J. Jee, T. Jenness, G. Jernigan, D. Jevremović,

K. Johns, A. S. Johnson, M. W. G. Johnson, R. L. Jones, C. Juramy-Gilles, M. Jurić,

J. S. Kalirai, N. J. Kallivayalil, B. Kalmbach, J. P. Kantor, P. Karst, M. M. Kasliwal,

H. Kelly, R. Kessler, V. Kinnison, D. Kirkby, L. Knox, I. V. Kotov, V. L. Krabbendam,

K. S. Krughoﬀ, P. Kubánek, J. Kuczewski, S. Kulkarni, J. Ku, N. R. Kurita, C. S. Lage,

R. Lambert, T. Lange, J. B. Langton, L. Le Guillou, D. Levine, M. Liang, K.-T. Lim, C. J.

Lintott, K. E. Long, M. Lopez, P. J. Lotz, R. H. Lupton, N. B. Lust, L. A. MacArthur, BIBLIOGRAPHY 168

A. Mahabal, R. Mandelbaum, T. W. Markiewicz, D. S. Marsh, P. J. Marshall, S. Mar-

shall, M. May, R. McKercher, M. McQueen, J. Meyers, M. Migliore, M. Miller, D. J.

Mills, C. Miraval, J. Moeyens, F. E. Moolekamp, D. G. Monet, M. Moniez, S. Monkewitz,

C. Montgomery, C. B. Morrison, F. Mueller, G. P. Muller, F. Muñoz Arancibia, D. R.

Neill, S. P. Newbry, J.-Y. Nief, A. Nomerotski, M. Nordby, P. O’Connor, J. Oliver, S. S.

Olivier, K. Olsen, W. O’Mullane, S. Ortiz, S. Osier, R. E. Owen, R. Pain, P. E. Palecek,

J. K. Parejko, J. B. Parsons, N. M. Pease, J. M. Peterson, J. R. Peterson, D. L. Petravick,

M. E. Libby Petrick, C. E. Petry, F. Pierfederici, S. Pietrowicz, R. Pike, P. A. Pinto,

R. Plante, S. Plate, J. P. Plutchak, P. A. Price, M. Prouza, V. Radeka, J. Rajagopal,

A. P. Rasmussen, N. Regnault, K. A. Reil, D. J. Reiss, M. A. Reuter, S. T. Ridgway,

V. J. Riot, S. Ritz, S. Robinson, W. Roby, A. Roodman, W. Rosing, C. Roucelle, M. R.

Rumore, S. Russo, A. Saha, B. Sassolas, T. L. Schalk, P. Schellart, R. H. Schindler,

S. Schmidt, D. P. Schneider, M. D. Schneider, W. Schoening, G. Schumacher, M. E.

Schwamb, J. Sebag, B. Selvy, G. H. Sembroski, L. G. Seppala, A. Serio, E. Serrano, R. A.

Shaw, I. Shipsey, J. Sick, N. Silvestri, C. T. Slater, J. A. Smith, R. C. Smith, S. Sobhani,

C. Soldahl, L. Storrie-Lombardi, E. Stover, M. A. Strauss, R. A. Street, C. W. Stubbs,

I. S. Sullivan, D. Sweeney, J. D. Swinbank, A. Szalay, P. Takacs, S. A. Tether, J. J. Thaler,

J. G. Thayer, S. Thomas, A. J. Thornton, V. Thukral, J. Tice, D. E. Trilling, M. Turri,

R. Van Berg, D. Vanden Berk, K. Vetter, F. Virieux, T. Vucina, W. Wahl, L. Walkowicz,

B. Walsh, C. W. Walter, D. L. Wang, S.-Y. Wang, M. Warner, O. Wiecha, B. Willman,

S. E. Winters, D. Wittman, S. C. Wolﬀ, W. M. Wood-Vasey, X. Wu, B. Xin, P. Yoachim,

and H. Zhan. LSST: From Science Drivers to Reference Design and Anticipated Data

Products. Astrophysical Journal, 873(2):111, Mar. 2019. doi:10.3847/1538-4357/ab042c.

B. Javanmardi, D. Martinez-Delgado, P. Kroupa, C. Henkel, K. Crawford, K. Teuwen, R. J. BIBLIOGRAPHY 169

Gabany, M. Hanson, T. S. Chonis, and F. Neyer. DGSAT: Dwarf Galaxy Survey with Am-

ateur Telescopes. I. Discovery of low surface brightness systems around nearby spiral galax-

ies. Astronomy & Astrophysics, 588:A89, Apr. 2016. doi:10.1051/0004-6361/201527745.

S. Jester, D. P. Schneider, G. T. Richards, R. F. Green, M. Schmidt, P. B. Hall, M. A.

Strauss, D. E. Vand en Berk, C. Stoughton, and J. E. Gunn. The Sloan Digital Sky

Survey View of the Palomar-Green Bright Quasar Survey. Astronomical Journal, 130(3):

873–895, Sep 2005. doi:10.1086/432466.

F. Jiang, A. Dekel, J. Freundlich, et al. Formation of ultra-diﬀuse galaxies in the ﬁeld and

in galaxy groups. Monthly Notices of the Royal Astronomical Society, 487(4):5272–5290,

Aug. 2019. doi:10.1093/mnras/stz1499.

L. Jiang, X. Fan, F. Bian, I. D. McGreer, M. A. Strauss, J. Annis, Z. Buck, R. Green, J. A.

Hodge, A. D. Myers, A. Raﬁee, and G. Richards. The Sloan Digital Sky Survey Stripe

82 Imaging Data: Depth-optimized Co-adds over 300 deg2 in Five Filters. Astrophysical

Journal Supplement Series, 213(1):12, July 2014. doi:10.1088/0067-0049/213/1/12.

G. D. Joshi, A. Pillepich, D. Nelson, E. Zinger, F. Marinacci, V. Springel, M. Vogels-

berger, and L. Hernquist. The cumulative star-formation histories of dwarf galaxies with

TNG50. I: Environment-driven diversity and connection to quenching. arXiv e-prints, art.

arXiv:2101.12226, Jan. 2021.

J. Kadowaki, D. Zaritsky, and R. L. Donnerstein. Spectroscopy of Ultra-diﬀuse Galax-

ies in the Coma Cluster. Astrophysical Journal Letters, 838(2):L21, Apr. 2017.

doi:10.3847/2041-8213/aa653d.

P. Kamphuis, G. I. G. Józsa, S. . H. Oh, and others. Automated kinematic modelling

of warped galaxy discs in large H I surveys: 3D tilted-ring ﬁtting of H I emission

cubes. Monthly Notices of the Royal Astronomical Society, 452(3):3139–3158, Sept. 2015.

doi:10.1093/mnras/stv1480. BIBLIOGRAPHY 170

I. D. Karachentsev and L. N. Makarova. Dwarf Spheroidal Galaxies in the M101 Group and

Behind it. Astrophysics, 62(3):293–299, Sep 2019. doi:10.1007/s10511-019-09582-7.

I. D. Karachentsev, D. Bautzmann, F. Neyer, R. Polzl, P. Riepe, T. Zilch, and B. Mattern.

Three low surface brightness dwarfs discovered around NGC 4631. arXiv e-prints, art.

arXiv:1401.2719, Jan. 2014a.

I. D. Karachentsev, D. Bautzmann, F. Neyer, R. Polzl, P. Riepe, T. Zilch, and B. Mat-

tern. Three low surface brightness dwarfs discovered around NGC 4631. ArXiv e-prints,

arXiv:1401.2719, Jan. 2014b.

I. D. Karachentsev, E. I. Kaisina, and D. I. Makarov. Suites of Dwarfs around nearby Giant

Galaxies. Astronomical Journal, 147(1):13, Jan. 2014c. doi:10.1088/0004-6256/147/1/13.

I. D. Karachentsev, P. Riepe, T. Zilch, M. Blauensteiner, M. Elvov, P. Hochleitner,

B. Hubl, G. Kerschhuber, S. Küppers, F. Neyer, R. Pölzl, P. Remmel, O. Schneider,

R. Sparenberg, U. Trulson, G. Willems, and H. Ziegler. New low surface brightness dwarf

galaxies detected around nearby spirals. Astrophysical Bulletin, 70:379–391, Oct. 2015.

doi:10.1134/S199034131504001X.

I. D. Karachentsev, P. Riepe, and T. Zilch. A Dozen new Dwarf Galaxy Candidates in the

Local Volume. Astrophysics, 63(1):5–14, Mar. 2020. doi:10.1007/s10511-020-09608-5.

A. Karunakaran, K. Spekkens, P. Bennet, D. J. Sand, D. Crnojević, and D. Zaritsky. Neutral

Hydrogen Observations of Low Surface Brightness Galaxies around M101 and NGC 5485.

Astronomical Journal, 159(2):37, Feb. 2020. doi:10.3847/1538-3881/ab5af1.

D. Kim and H. Jerjen. Horologium II: A Second Ultra-faint Milky Way Satellite in

the Horologium Constellation. Astrophysical Journal Letters, 808(2):L39, Aug. 2015.

doi:10.1088/2041-8205/808/2/L39. BIBLIOGRAPHY 171

D. Kim, H. Jerjen, D. Mackey, G. S. Da Costa, and A. P. Milone. A Hero’s Dark Horse:

Discovery of an Ultra-faint Milky Way Satellite in Pegasus. Astrophysical Journal Letters,

804(2):L44, May 2015. doi:10.1088/2041-8205/804/2/L44.

S. Kim, S.-C. Rey, H. Jerjen, T. Lisker, E.-C. Sung, Y. Lee, J. Chung, M. Pak, W. Yi, and

W. Lee. The Extended Virgo Cluster Catalog. Astrophysical Journal Supplement Series,

215(2):22, Dec. 2014. doi:10.1088/0067-0049/215/2/22.

E. M. Kirby, B. Koribalski, H. Jerjen, et al. The Local Volume H I Survey: galaxy kine-

matics. Monthly Notices of the Royal Astronomical Society, 420(4):2924–2943, Mar. 2012.

doi:10.1111/j.1365-2966.2011.20103.x.

J. Koda, M. Yagi, H. Yamanoi, and Y. Komiyama. Approximately a Thousand Ultra-diﬀuse

Galaxies in the Coma Cluster. Astrophysical Journal, 807:L2, Jul 2015a. doi:10.1088/2041-

8205/807/1/L2.

J. Koda, M. Yagi, H. Yamanoi, and Y. Komiyama. Approximately a Thousand Ultra-

diﬀuse Galaxies in the Coma Cluster. Astrophysical Journal Letters, 807:L2, July 2015b.

doi:10.1088/2041-8205/807/1/L2.

S. Koposov, V. Belokurov, N. W. Evans, P. C. Hewett, M. J. Irwin, G. Gilmore, D. B.

Zucker, H. W. Rix, M. Fellhauer, E. F. Bell, and E. V. Glushkova. The Luminosity

Function of the Milky Way Satellites. Astrophysical Journal, 686(1):279–291, Oct. 2008.

doi:10.1086/589911.

S. E. Koposov, V. Belokurov, G. Torrealba, and N. W. Evans. Beasts of the Southern

Wild: Discovery of Nine Ultra Faint Satellites in the Vicinity of the Magellanic Clouds.

Astrophysical Journal, 805(2):130, June 2015. doi:10.1088/0004-637X/805/2/130.

B. S. Koribalski, L. Staveley-Smith, T. Westmeier, P. Serra, K. Spekkens, O. I. Wong, K. Lee-

Waddell, C. D. P. Lagos, D. Obreschkow, E. V. Ryan-Weber, M. Zwaan, V. Kilborn, BIBLIOGRAPHY 172

G. Bekiaris, K. Bekki, F. Bigiel, A. Boselli, A. Bosma, B. Catinella, G. Chauhan, M. E.

Cluver, M. Colless, H. M. Courtois, R. A. Crain, W. J. G. de Blok, H. Dénes, A. R.

Duﬀy, A. Elagali, C. J. Fluke, B. Q. For, G. Heald, P. A. Henning, K. M. Hess, B. W.

Holwerda, C. Howlett, T. Jarrett, D. H. Jones, M. G. Jones, G. I. G. Józsa, R. Jurek,

E. Jütte, P. Kamphuis, I. Karachentsev, J. Kerp, D. Kleiner, R. C. Kraan-Korteweg, Á. R.

López-Sánchez, J. Madrid, M. Meyer, J. Mould, C. Murugeshan, R. P. Norris, S. H. Oh,

T. A. Oosterloo, A. Popping, M. Putman, T. N. Reynolds, J. Rhee, A. S. G. Robotham,

S. Ryder, A. C. Schröder, L. Shao, A. R. H. Stevens, E. N. Taylor, J. M. vanÂ der Hulst,

L. Verdes-Montenegro, B. P. Wakker, J. Wang, M. Whiting, B. Winkel, and C. Wolf.

WALLABY - an SKA Pathﬁnder HI survey. Astrophysics and Space Science, 365(7):118,

July 2020. doi:10.1007/s10509-020-03831-4.

J. M. Kubo, A. Stebbins, J. Annis, I. P. Dell’Antonio, H. Lin, H. Khiabanian, and J. A.

Frieman. The Mass of the Coma Cluster from Weak Lensing in the Sloan Digital Sky

Survey. Astrophysical Journal, 671(2):1466–1470, Dec. 2007. doi:10.1086/523101.

J. C. Lee, A. Gil de Paz, J. Kennicutt, Robert C., M. Bothwell, J. Dalcanton, J. José G.

Funes S., B. D. Johnson, S. Sakai, E. Skillman, C. Tremonti, and L. van Zee. A GALEX

Ultraviolet Imaging Survey of Galaxies in the Local Volume. Astrophysical Journal Sup-

plement Series, 192(1):6, Jan. 2011. doi:10.1088/0067-0049/192/1/6.

J. H. Lee, J. Kang, M. G. Lee, and I. S. Jang. The Nature of Ultra-diﬀuse Galaxies in Distant

Massive Galaxy Clusters: A370 in the Hubble Frontier Fields. Astrophysical Journal, 894

(1):75, May 2020. doi:10.3847/1538-4357/ab8632.

M. G. Lee and I. S. Jang. The Distance to M101 Hosting Type Ia Supernova 2011fe Based

on the Tip of the Red Giant Branch. Astrophysical Journal Letters, 760(1):L14, Nov 2012.

doi:10.1088/2041-8205/760/1/L14.

L. Leisman, M. P. Haynes, S. Janowiecki, G. Hallenbeck, G. Józsa, R. Giovanelli, E. A. K. BIBLIOGRAPHY 173

Adams, D. Bernal Neira, J. M. Cannon, W. F. Janesh, K. L. Rhode, and J. J. Salzer. (Al-

most) Dark Galaxies in the ALFALFA Survey: Isolated H I-bearing Ultra-diﬀuse Galaxies.

Astrophysical Journal, 842(2):133, June 2017. doi:10.3847/1538-4357/aa7575.

F. Lelli, S. S. McGaugh, and J. M. Schombert. The Small Scatter of the Baryonic Tully-

Fisher Relation. Astrophysical Journal Letters, 816(1):L14, Jan. 2016. doi:10.3847/2041-

8205/816/1/L14.

S. Liao, L. Gao, C. S. Frenk, et al. Ultra-diﬀuse galaxies in the Auriga simula-

tions. Monthly Notices of the Royal Astronomical Society, 490(4):5182–5195, Dec. 2019.

doi:10.1093/mnras/stz2969.

S. Lim, E. W. Peng, P. Côté, et al. The Globular Cluster Systems of Ultra-diﬀuse Galaxies

in the Coma Cluster. Astrophysical Journal, 862(1):82, July 2018. doi:10.3847/1538-

4357/aacb81.

E. Luque, A. Queiroz, B. Santiago, A. Pieres, E. Balbinot, K. Bechtol, A. Drlica-Wagner,

A. F. Neto, L. N. da Costa, M. A. G. Maia, B. Yanny, T. Abbott, S. Allam, A. Benoit-

Lévy, E. Bertin, D. Brooks, E. Buckley-Geer, D. L. Burke, A. C. Rosell, M. C. Kind,

J. Carretero, C. E. Cunha, S. Desai, H. T. Diehl, J. P. Dietrich, T. F. Eiﬂer, D. A. Finley,

B. Flaugher, P. Fosalba, J. Frieman, D. W. Gerdes, D. Gruen, G. Gutierrez, K. Honscheid,

D. J. James, K. Kuehn, N. Kuropatkin, O. Lahav, T. S. Li, M. March, J. L. Marshall,

P. Martini, R. Miquel, E. Neilsen, R. C. Nichol, B. Nord, R. Ogando, A. A. Plazas, A. K.

Romer, A. Roodman, E. Sanchez, V. Scarpine, M. Schubnell, I. Sevilla-Noarbe, R. C.

Smith, M. Soares-Santos, F. Sobreira, E. Suchyta, M. E. C. Swanson, G. Tarle, J. Thaler,

D. Tucker, A. R. Walker, and Y. Zhang. Digging deeper into the Southern skies: a compact

Milky Way companion discovered in ﬁrst-year Dark Energy Survey data. Monthly Notices

of the Royal Astronomical Society, 458(1):603–612, May 2016. doi:10.1093/mnras/stw302.

P. E. Mancera Piña, J. A. L. Aguerri, R. F. Peletier, et al. The evolution of ultra-diﬀuse BIBLIOGRAPHY 174

galaxies in nearby galaxy clusters from the Kapteyn IAC WEAVE INT Clusters Sur-

vey. Monthly Notices of the Royal Astronomical Society, 485(1):1036–1052, May 2019a.

doi:10.1093/mnras/stz238.

P. E. Mancera Piña, F. Fraternali, E. A. K. Adams, A. Marasco, T. Oosterloo, K. A.

Oman, L. Leisman, E. M. di Teodoro, L. Posti, M. Battipaglia, J. M. Cannon, L. Gault,

M. P. Haynes, S. Janowiecki, E. McAllan, H. J. Pagel, K. Reiter, K. L. Rhode, J. J.

Salzer, and N. J. Smith. Oﬀ the Baryonic Tully-Fisher Relation: A Population of Baryon-

dominated Ultra-diﬀuse Galaxies. Astrophysical Journal Letters, 883(2):L33, Oct. 2019b.

doi:10.3847/2041-8213/ab40c7.

P. E. Mancera Piña, F. Fraternali, K. A. Oman, et al. Robust H I kinematics of gas-rich

ultra-diﬀuse galaxies: hints of a weak-feedback formation scenario. Monthly Notices of the

Royal Astronomical Society, 495(4):3636–3655, May 2020. doi:10.1093/mnras/staa1256.

Y.-Y. Mao, M. Geha, R. H. Wechsler, B. Weiner, E. J. Tollerud, E. O. Nadler, and N. Kalli-

vayalil. The SAGA Survey. II. Building a Statistical Sample of Satellite Systems around

Milky Way-like Galaxies. Astrophysical Journal, 907(2):85, Feb. 2021. doi:10.3847/1538-

4357/abce58.

D. C. Martin, J. Fanson, D. Schiminovich, P. Morrissey, P. G. Friedman, T. A. Barlow,

T. Conrow, R. Grange, P. N. Jelinsky, B. Milliard, O. H. W. Siegmund, L. Bianchi, Y.-I.

Byun, J. Donas, K. Forster, T. M. Heckman, Y.-W. Lee, B. F. Madore, R. F. Malina,

S. G. Neﬀ, R. M. Rich, T. Small, F. Surber, A. S. Szalay, B. Welsh, and T. K. Wyder. The

Galaxy Evolution Explorer: A Space Ultraviolet Survey Mission. Astrophysical Journal

Letters, 619(1):L1–L6, Jan. 2005. doi:10.1086/426387.

M. C. Martin. Catalogue of HI maps of galaxies. I. Astronomy & Astrophysics Supplement,

131:73–75, July 1998. doi:10.1051/aas:1998252. BIBLIOGRAPHY 175

N. F. Martin, A. W. McConnachie, M. Irwin, L. M. Widrow, A. M. N. Ferguson, R. A.

Ibata, J. Dubinski, A. Babul, S. Chapman, M. Fardal, G. F. Lewis, J. Navarro, and R. M.

Rich. PAndAS’ CUBS: Discovery of Two New Dwarf Galaxies in the Surroundings of the

Andromeda and Triangulum Galaxies. Astrophysical Journal, 705(1):758–765, Nov. 2009.

doi:10.1088/0004-637X/705/1/758.

N. F. Martin, R. A. Ibata, A. W. McConnachie, A. D. Mackey, A. M. N. Ferguson, M. J. Ir-

win, G. F. Lewis, and M. A. Fardal. The PAndAS View of the Andromeda Satellite System.

I. A Bayesian Search for Dwarf Galaxies Using Spatial and Color-Magnitude Information.

Astrophysical Journal, 776(2):80, Oct. 2013a. doi:10.1088/0004-637X/776/2/80.

N. F. Martin, C. T. Slater, E. F. Schlaﬂy, E. Morganson, H.-W. Rix, E. F. Bell, B. P. M.

Laevens, E. J. Bernard, A. M. N. Ferguson, D. P. Finkbeiner, W. S. Burgett, K. C.

Chambers, K. W. Hodapp, N. Kaiser, R.-P. Kudritzki, E. A. Magnier, J. S. Morgan, P. A.

Price, J. L. Tonry, and R. J. Wainscoat. Lacerta I and Cassiopeia III. Two Luminous

and Distant Andromeda Satellite Dwarf Galaxies Found in the 3π Pan-STARRS1 Survey.

Astrophysical Journal, 772(1):15, July 2013b. doi:10.1088/0004-637X/772/1/15.

N. F. Martin, R. A. Ibata, G. F. Lewis, A. McConnachie, A. Babul, N. F. Bate, E. Bernard,

S. C. Chapman, M. M. L. Collins, A. R. Conn, D. Crnojević, M. A. Fardal, A. M. N.

Ferguson, M. Irwin, A. D. Mackey, B. McMonigal, J. F. Navarro, and R. M. Rich. The

PAndAS View of the Andromeda Satellite System. II. Detailed Properties of 23 M31 Dwarf

Spheroidal Galaxies. Astrophysical Journal, 833(2):167, Dec. 2016. doi:10.3847/1538-

4357/833/2/167.

N. F. Martin, D. R. Weisz, S. M. Albers, E. Bernard, M. L. M. Collins, A. E. Dolphin,

A. M. N. Ferguson, R. A. Ibata, B. Laevens, G. F. Lewis, A. D. Mackey, A. McConnachie,

R. M. Rich, and E. D. Skillman. A Rogues’ Gallery of Andromeda’s Dwarf Galaxies. I. A BIBLIOGRAPHY 176

Predominance of Red Horizontal Branches. Astrophysical Journal, 850(1):16, Nov. 2017.

doi:10.3847/1538-4357/aa901a.

I. Martín-Navarro, A. J. Romanowsky, J. P. Brodie, A. Ferré-Mateu, A. Alabi, D. A. Forbes,

M. Sharina, A. Villaume, V. Pandya, and D. Martinez-Delgado. Extreme chemical abun-

dance ratio suggesting an exotic origin for an ultradiﬀuse galaxy. Monthly Notices of the

Royal Astronomical Society, 484:3425–3433, Apr 2019. doi:10.1093/mnras/stz252.

D. Martínez-Delgado, J. Peñarrubia, R. J. Gabany, I. Trujillo, S. R. Majewski, and

M. Pohlen. The Ghost of a Dwarf Galaxy: Fossils of the Hierarchical Formation of

the Nearby Spiral Galaxy NGC 5907. Astrophysical Journal, 689(1):184–193, Dec. 2008.

doi:10.1086/592555.

D. Martínez-Delgado, R. Läsker, M. Sharina, E. Toloba, J. Fliri, R. Beaton, D. Valls-

Gabaud, I. D. Karachentsev, T. S. Chonis, E. K. Grebel, D. A. Forbes, A. J. Romanowsky,

J. Gallego-Laborda, K. Teuwen, M. A. Gómez-Flechoso, J. Wang, P. Guhathakurta,

S. Kaisin, and N. Ho. Discovery of an Ultra-diﬀuse Galaxy in the Pisces–Perseus Su-

percluster. Astronomical Journal, 151:96, Apr. 2016. doi:10.3847/0004-6256/151/4/96.

A. W. McConnachie. The Observed Properties of Dwarf Galaxies in and around the Local

Group. Astronomical Journal, 144(1):4, July 2012a. doi:10.1088/0004-6256/144/1/4.

A. W. McConnachie. The Observed Properties of Dwarf Galaxies in and around the Local

Group. Astronomical Journal, 144:4, July 2012b. doi:10.1088/0004-6256/144/1/4.

K. B. W. McQuinn, E. D. Skillman, A. E. Dolphin, and N. P. Mitchell. Calibrating UV

Star Formation Rates for Dwarf Galaxies from STARBIRDS. Astrophysical Journal, 808

(2):109, Aug. 2015. doi:10.1088/0004-637X/808/2/109.

A. Merritt, P. van Dokkum, and R. Abraham. The Discovery of Seven Extremely Low BIBLIOGRAPHY 177

Surface Brightness Galaxies in the Field of the Nearby Spiral Galaxy M101. Astrophysical

Journal Letters, 787:L37, June 2014. doi:10.1088/2041-8205/787/2/L37.

A. Merritt, P. van Dokkum, S. Danieli, R. Abraham, J. Zhang, I. D. Karachentsev, and L. N.

Makarova. The Dragonﬂy Nearby Galaxies Survey. II. Ultra-Diﬀuse Galaxies near the

Elliptical Galaxy NGC 5485. Astrophysical Journal, 833:168, Dec. 2016. doi:10.3847/1538-

4357/833/2/168.

J. C. Mihos, K. M. Keating, K. Holley-Bockelmann, D. J. Pisano, and N. E. Kassim.

The H I Environment of the M101 Group. Astrophysical Journal, 761:186, Dec. 2012.

doi:10.1088/0004-637X/761/2/186.

J. C. Mihos, P. Harding, C. E. Spengler, C. S. Rudick, and J. J. Feldmeier. The Extended

Optical Disk of M101. Astrophysical Journal, 762(2):82, Jan 2013. doi:10.1088/0004-

637X/762/2/82.

J. C. Mihos, P. R. Durrell, L. Ferrarese, J. J. Feldmeier, P. Côté, E. W. Peng, P. Harding,

C. Liu, S. Gwyn, and J.-C. Cuillandre. Galaxies at the Extremes: Ultra-diﬀuse Galaxies in

the Virgo Cluster. Astrophysical Journal Letters, 809:L21, Aug. 2015a. doi:10.1088/2041-

8205/809/2/L21.

J. C. Mihos, P. R. Durrell, L. Ferrarese, et al. Galaxies at the Extremes: Ultra-diﬀuse

Galaxies in the Virgo Cluster. Astrophysical Journal Letters, 809(2):L21, Aug. 2015b.

doi:10.1088/2041-8205/809/2/L21.

H. J. Mo, S. Mao, and S. D. M. White. The formation of galactic discs. Monthly No-

tices of the Royal Astronomical Society, 295:319–336, Apr. 1998. doi:10.1046/j.1365-

8711.1998.01227.x.

P. Morrissey, T. Conrow, T. A. Barlow, T. Small, M. Seibert, T. K. Wyder, T. Budavári,

S. Arnouts, P. G. Friedman, K. Forster, D. C. Martin, S. G. Neﬀ, D. Schiminovich, BIBLIOGRAPHY 178

L. Bianchi, J. Donas, T. M. Heckman, Y.-W. Lee, B. F. Madore, B. Milliard, R. M. Rich,

A. S. Szalay, B. Y. Welsh, and S. K. Yi. The Calibration and Data Products of GALEX.

Astrophysical Journal Supplement Series, 173(2):682–697, Dec. 2007. doi:10.1086/520512.

B. P. Moster, R. S. Somerville, J. A. Newman, and H.-W. Rix. A Cosmic Variance Cookbook.

Astrophysical Journal, 731:113, Apr. 2011. doi:10.1088/0004-637X/731/2/113.

O. Müller, R. Scalera, B. Binggeli, and H. Jerjen. The M 101 group complex: new dwarf

galaxy candidates and spatial structure. Astronomy & Astrophysics, 602:A119, June 2017.

doi:10.1051/0004-6361/201730434.

O. Müller, H. Jerjen, and B. Binggeli. The Leo-I group: new dwarf galaxy and ultra diﬀuse

galaxy candidates. Astronomy & Astrophysics, 615:A105, July 2018a. doi:10.1051/0004-

6361/201832897.

O. Müller, M. Rejkuba, and H. Jerjen. Distances from the tip of the red giant branch

to the dwarf galaxies dw1335-29 and dw1340-30 in the Centaurus group. Astronomy &

Astrophysics, 615:A96, July 2018b. doi:10.1051/0004-6361/201732455.

B. Mutlu-Pakdil, D. J. Sand, D. Crnojević, A. Drlica-Wagner, N. Caldwell, P. Guhathakurta,

A. C. Seth, J. D. Simon, J. Strader, and E. Toloba. Resolved Dwarf Galaxy Searches

within ~5 Mpc with the Vera Rubin Observatory and Subaru Hyper Suprime-Cam. arXiv e-prints, art. arXiv:2105.01658, May 2021.

E. O. Nadler, R. H. Wechsler, K. Bechtol, Y. Y. Mao, G. Green, A. Drlica-Wagner, M. Mc-

Nanna, S. Mau, A. B. Pace, J. D. Simon, A. Kravtsov, S. Dodelson, T. S. Li, A. H. Riley,

M. Y. Wang, T. M. C. Abbott, M. Aguena, S. Allam, J. Annis, S. Avila, G. M. Bernstein,

E. Bertin, D. Brooks, D. L. Burke, A. C. Rosell, M. C. Kind, J. Carretero, M. Costanzi,

L. N. da Costa, J. De Vicente, S. Desai, A. E. Evrard, B. Flaugher, P. Fosalba, J. Frie-

man, J. García-Bellido, E. Gaztanaga, D. W. Gerdes, D. Gruen, J. Gschwend, G. Gutier-

rez, W. G. Hartley, S. R. Hinton, K. Honscheid, E. Krause, K. Kuehn, N. Kuropatkin, BIBLIOGRAPHY 179

O. Lahav, M. A. G. Maia, J. L. Marshall, F. Menanteau, R. Miquel, A. Palmese, F. Paz-

Chinchón, A. A. Plazas, A. K. Romer, E. Sanchez, B. Santiago, V. Scarpine, S. Serrano,

M. Smith, M. Soares-Santos, E. Suchyta, G. Tarle, D. Thomas, T. N. Varga, A. R. Walker,

and DES Collaboration. Milky Way Satellite Census. II. Galaxy-Halo Connection Con-

straints Including the Impact of the Large Magellanic Cloud. Astrophysical Journal, 893

(1):48, Apr. 2020. doi:10.3847/1538-4357/ab846a.

D. Nelson, A. Pillepich, V. Springel, R. Pakmor, R. Weinberger, S. Genel, P. Torrey, M. Vo-

gelsberger, F. Marinacci, and L. Hernquist. First results from the TNG50 simulation:

galactic outﬂows driven by supernovae and black hole feedback. Monthly Notices of the

Royal Astronomical Society, 490(3):3234–3261, Dec. 2019. doi:10.1093/mnras/stz2306.

O. Newton, M. Cautun, A. Jenkins, C. S. Frenk, and J. C. Helly. The total satellite

population of the Milky Way. Monthly Notices of the Royal Astronomical Society, 479(3):

2853–2870, Sept. 2018. doi:10.1093/mnras/sty1085.

K. O’Neil, G. Bothun, W. van Driel, and D. Monnier Ragaigne. A new H I catalog of

Low Surface Brightness galaxies out to z = 0.1. Tripling the number of massive LSB

galaxies known. Astronomy & Astrophysics, 428:823–835, Dec. 2004. doi:10.1051/0004-

6361:20047091.

V. Pandya, A. J. Romanowsky, S. Laine, J. P. Brodie, B. D. Johnson, W. Glaccum, A. Vil-

laume, J.-C. Cuillandre, S. Gwyn, J. Krick, R. Lasker, I. Martín-Navarro, D. Martinez-

Delgado, and P. van Dokkum. The Stellar Populations of Two Ultra-diﬀuse Galaxies

from Optical and Near-infrared Photometry. Astrophysical Journal, 858:29, May 2018.

doi:10.3847/1538-4357/aab498.

E. Papastergis, E. A. K. Adams, and A. J. Romanowsky. The HI content of isolated ultra-

diﬀuse galaxies: A sign of multiple formation mechanisms? Astronomy & Astrophysics,

601:L10, May 2017. doi:10.1051/0004-6361/201730795. BIBLIOGRAPHY 180

J. E. G. Peek, C. Heiles, K. A. Douglas, M.-Y. Lee, J. Grcevich, S. Stanimirović, M. E.

Putman, E. J. Korpela, S. J. Gibson, A. Begum, D. Saul, T. Robishaw, and M. Krčo. The

GALFA-HI Survey: Data Release 1. Astrophysical Journal Supplement Series, 194(2):20,

June 2011. doi:10.1088/0067-0049/194/2/20.

J. E. G. Peek, B. L. Babler, Y. Zheng, S. E. Clark, K. A. Douglas, E. J. Korpela, M. E.

Putman, S. Stanimirović, S. J. Gibson, and C. Heiles. The GALFA-H I Survey Data Re-

lease 2. Astrophysical Journal Supplement Series, 234(1):2, Jan. 2018. doi:10.3847/1538-

4365/aa91d3.

C. Y. Peng, L. C. Ho, C. D. Impey, and H.-W. Rix. Detailed Decomposition of Galaxy

Images. II. Beyond Axisymmetric Models. Astronomical Journal, 139:2097–2129, June

2010. doi:10.1088/0004-6256/139/6/2097.

J. I. Phillips, C. Wheeler, M. Boylan-Kolchin, J. S. Bullock, M. C. Cooper, and E. J.

Tollerud. A dichotomy in satellite quenching around L* galaxies. Monthly Notices of the

Royal Astronomical Society, 437(2):1930–1941, Jan. 2014. doi:10.1093/mnras/stt2023.

J. I. Phillips, C. Wheeler, M. C. Cooper, M. Boylan-Kolchin, J. S. Bullock, and E. Tollerud.

The mass dependence of satellite quenching in Milky Way-like haloes. Monthly Notices of

the Royal Astronomical Society, 447(1):698–710, Feb. 2015. doi:10.1093/mnras/stu2192.

A. Pillepich, V. Springel, D. Nelson, S. Genel, J. Naiman, R. Pakmor, L. Hernquist, P. Tor-

rey, M. Vogelsberger, R. Weinberger, and F. Marinacci. Simulating galaxy formation

with the IllustrisTNG model. Monthly Notices of the Royal Astronomical Society, 473(3):

4077–4106, Jan. 2018. doi:10.1093/mnras/stx2656.

Planck Collaboration, N. Aghanim, Y. Akrami, M. Ashdown, J. Aumont, C. Baccigalupi,

M. Ballardini, A. J. Banday, R. B. Barreiro, N. Bartolo, S. Basak, R. Battye, K. Ben-

abed, J. P. Bernard, M. Bersanelli, P. Bielewicz, J. J. Bock, J. R. Bond, J. Borrill, BIBLIOGRAPHY 181

F. R. Bouchet, F. Boulanger, M. Bucher, C. Burigana, R. C. Butler, E. Calabrese,

J. F. Cardoso, J. Carron, A. Challinor, H. C. Chiang, J. Chluba, L. P. L. Colombo,

C. Combet, D. Contreras, B. P. Crill, F. Cuttaia, P. de Bernardis, G. de Zotti, J. De-

labrouille, J. M. Delouis, E. Di Valentino, J. M. Diego, O. Doré, M. Douspis, A. Ducout,

X. Dupac, S. Dusini, G. Efstathiou, F. Elsner, T. A. Enßlin, H. K. Eriksen, Y. Fantaye,

M. Farhang, J. Fergusson, R. Fernandez-Cobos, F. Finelli, F. Forastieri, M. Frailis, A. A.

Fraisse, E. Franceschi, A. Frolov, S. Galeotta, S. Galli, K. Ganga, R. T. Génova-Santos,

M. Gerbino, T. Ghosh, J. González-Nuevo, K. M. Górski, S. Gratton, A. Gruppuso, J. E.

Gudmundsson, J. Hamann, W. Handley, F. K. Hansen, D. Herranz, S. R. Hildebrandt,

E. Hivon, Z. Huang, A. H. Jaﬀe, W. C. Jones, A. Karakci, E. Keihänen, R. Keskitalo,

K. Kiiveri, J. Kim, T. S. Kisner, L. Knox, N. Krachmalnicoﬀ, M. Kunz, H. Kurki-Suonio,

G. Lagache, J. M. Lamarre, A. Lasenby, M. Lattanzi, C. R. Lawrence, M. Le Jeune,

P. Lemos, J. Lesgourgues, F. Levrier, A. Lewis, M. Liguori, P. B. Lilje, M. Lilley, V. Lind-

holm, M. López-Caniego, P. M. Lubin, Y. Z. Ma, J. F. Macías-Pérez, G. Maggio, D. Maino,

N. Mandolesi, A. Mangilli, A. Marcos-Caballero, M. Maris, P. G. Martin, M. Martinelli,

E. Martínez-González, S. Matarrese, N. Mauri, J. D. McEwen, P. R. Meinhold, A. Mel-

chiorri, A. Mennella, M. Migliaccio, M. Millea, S. Mitra, M. A. Miville-Deschênes, D. Moli-

nari, L. Montier, G. Morgante, A. Moss, P. Natoli, H. U. Nørgaard-Nielsen, L. Pagano,

D. Paoletti, B. Partridge, G. Patanchon, H. V. Peiris, F. Perrotta, V. Pettorino, F. Pia-

centini, L. Polastri, G. Polenta, J. L. Puget, J. P. Rachen, M. Reinecke, M. Remazeilles,

A. Renzi, G. Rocha, C. Rosset, G. Roudier, J. A. Rubiño-Martín, B. Ruiz-Granados,

L. Salvati, M. Sandri, M. Savelainen, D. Scott, E. P. S. Shellard, C. Sirignano, G. Sirri,

L. D. Spencer, R. Sunyaev, A. S. Suur-Uski, J. A. Tauber, D. Tavagnacco, M. Tenti,

L. Toﬀolatti, M. Tomasi, T. Trombetti, L. Valenziano, J. Valiviita, B. Van Tent, L. Vib-

ert, P. Vielva, F. Villa, N. Vittorio, B. D. Wandelt, I. K. Wehus, M. White, S. D. M.

White, A. Zacchei, and A. Zonca. Planck 2018 results. VI. Cosmological parameters.

Astronomy & Astrophysics, 641:A6, Sept. 2020. doi:10.1051/0004-6361/201833910. BIBLIOGRAPHY 182

D. J. Prole, J. I. Davies, O. C. Keenan, and L. J. M. Davies. Automated detection of

very low surface brightness galaxiesin the Virgo cluster. Monthly Notices of the Royal

Astronomical Society, 478(1):667–681, July 2018. doi:10.1093/mnras/sty1021.

D. J. Prole, M. Hilker, R. F. J. van der Burg, et al. Halo mass estimates from the

globular cluster populations of 175 low surface brightness galaxies in the Fornax clus-

ter. Monthly Notices of the Royal Astronomical Society, 484(4):4865–4880, Apr. 2019a.

doi:10.1093/mnras/stz326.

D. J. Prole, R. F. J. van der Burg, M. Hilker, et al. Observational properties of ultra-

diﬀuse galaxies in low-density environments: ﬁeld UDGs are predominantly blue and star

forming. Monthly Notices of the Royal Astronomical Society, 488(2):2143–2157, Sept.

2019b. doi:10.1093/mnras/stz1843.

M. E. Putman, Y. Zheng, A. M. Price-Whelan, J. Grcevich, A. C. Johnson, E. Tollerud,

and J. E. G. Peek. The Gas Content and Stripping of Local Group Dwarf Galaxies.

Astrophysical Journal, 913(1):53, May 2021. doi:10.3847/1538-4357/abe391.

J. I. Read, G. Iorio, O. Agertz, and F. Fraternali. The stellar mass-halo mass rela-

tion of isolated ﬁeld dwarfs: a critical test of ΛCDM at the edge of galaxy forma-

tion. Monthly Notices of the Royal Astronomical Society, 467(2):2019–2038, May 2017.

doi:10.1093/mnras/stx147.

M. K. Rodriguez Wimberly, M. C. Cooper, S. P. Fillingham, M. Boylan-Kolchin, J. S.

Bullock, and S. Garrison-Kimmel. The suppression of star formation on the smallest

scales: what role does environment play? Monthly Notices of the Royal Astronomical

Society, 483(3):4031–4039, Mar. 2019. doi:10.1093/mnras/sty3357.

J. C. Roediger and S. Courteau. On the uncertainties of stellar mass estimates via colour

measurements. Monthly Notices of the Royal Astronomical Society, 452(3):3209–3225,

Sept. 2015. doi:10.1093/mnras/stv1499. BIBLIOGRAPHY 183

D. H. Rogstad, I. A. Lockhart, and M. C. H. Wright. Aperture-synthesis observations of H

I in the galaxy M83. Astrophysical Journal, 193:309–319, Oct. 1974. doi:10.1086/153164.

J. Román and I. Trujillo. Ultra-diﬀuse galaxies outside clusters: clues to their formation

and evolution. Monthly Notices of the Royal Astronomical Society, 468:4039–4047, July

2017. doi:10.1093/mnras/stx694.

J. Román, M. A. Beasley, T. Ruiz-Lara, and D. Valls-Gabaud. Discovery of a red

ultra-diﬀuse galaxy in a nearby void based on its globular cluster luminosity func-

tion. Monthly Notices of the Royal Astronomical Society, 486(1):823–835, June 2019.

doi:10.1093/mnras/stz835.

B. F. Roukema and B. A. Peterson. A survey for low surface brightness galaxies in the ﬁeld.

Astronomy & Astrophysics Supplement, 109:511–521, Mar. 1995.

S. Roychowdhury, J. N. Chengalur, I. D. Karachentsev, et al. The intrinsic shapes of dwarf

irregular galaxies. Monthly Notices of the Royal Astronomical Society, 436:L104–L108,

Nov. 2013. doi:10.1093/mnrasl/slt123.

T. Ruiz-Lara, M. A. Beasley, J. Falcón-Barroso, J. Román, F. Pinna, C. Brook, A. Di

Cintio, I. Martín-Navarro, I. Trujillo, and A. Vazdekis. Spectroscopic characterization of

the stellar content of ultra-diﬀuse galaxies. Monthly Notices of the Royal Astronomical

Society, 478:2034–2045, Aug 2018. doi:10.1093/mnras/sty1112.

T. Saifollahi, I. Trujillo, M. A. Beasley, et al. The number of globular clusters around the

iconic UDG DF44 is as expected for dwarf galaxies. arXiv e-prints, art. arXiv:2006.14630,

June 2020.

L. V. Sales, J. F. Navarro, L. Peñaﬁel, et al. The formation of ultradiﬀuse galaxies in

clusters. Monthly Notices of the Royal Astronomical Society, 494(2):1848–1858, Mar.

2020. doi:10.1093/mnras/staa854. BIBLIOGRAPHY 184

A. Sandage and B. Binggeli. Studies of the Virgo cluster. III - A classiﬁcation system and an

illustrated atlas of Virgo cluster dwarf galaxies. Astronomical Journal, 89:919–931, July

1984. doi:10.1086/113588.

C. Saulder, E. van Kampen, I. V. Chilingarian, S. Mieske, and W. W. Zeilinger. The matter

distribution in the local Universe as derived from galaxy groups in SDSS DR12 and 2MRS.

Astronomy & Astrophysics, 596:A14, Nov 2016. doi:10.1051/0004-6361/201526711.

T. Sawala, C. S. Frenk, A. Fattahi, J. F. Navarro, R. G. Bower, R. A. Crain, C. Dalla Vecchia,

M. Furlong, J. C. Helly, A. Jenkins, K. A. Oman, M. Schaller, J. Schaye, T. Theuns,

J. Trayford, and S. D. M. White. The APOSTLE simulations: solutions to the Local

Group’s cosmic puzzles. Monthly Notices of the Royal Astronomical Society, 457(2):1931–

1943, Apr. 2016. doi:10.1093/mnras/stw145.

A. L. Schaefer, S. M. Croom, N. Scott, S. Brough, J. T. Allen, K. Bekki, J. Bland-Hawthorn,

J. V. Bloom, J. J. Bryant, L. Cortese, L. J. M. Davies, C. Federrath, L. M. R. Fog-

arty, A. W. Green, B. Groves, A. M. Hopkins, I. S. Konstantopoulos, A. R. López-

Sánchez, J. S. Lawrence, R. E. McElroy, A. M. Medling, M. S. Owers, M. B. Pracy,

S. N. Richards, A. S. G. Robotham, J. van de Sande, C. Tonini, and S. K. Yi. The

SAMI Galaxy Survey: observing the environmental quenching of star formation in GAMA

groups. Monthly Notices of the Royal Astronomical Society, 483(3):2851–2870, Mar 2019.

doi:10.1093/mnras/sty3258.

J. Schaye, R. A. Crain, R. G. Bower, M. Furlong, M. Schaller, T. Theuns, C. Dalla Vecchia,

C. S. Frenk, I. G. McCarthy, J. C. Helly, A. Jenkins, Y. M. Rosas-Guevara, S. D. M.

White, M. Baes, C. M. Booth, P. Camps, J. F. Navarro, Y. Qu, A. Rahmati, T. Sawala,

P. A. Thomas, and J. Trayford. The EAGLE project: simulating the evolution and

assembly of galaxies and their environments. Monthly Notices of the Royal Astronomical

Society, 446(1):521–554, Jan. 2015. doi:10.1093/mnras/stu2058. BIBLIOGRAPHY 185

E. F. Schlaﬂy and D. P. Finkbeiner. Measuring Reddening with Sloan Digital Sky Survey

Stellar Spectra and Recalibrating SFD. Astrophysical Journal, 737(2):103, Aug. 2011.

doi:10.1088/0004-637X/737/2/103.

J. M. Schombert, G. D. Bothun, S. E. Schneider, and S. S. McGaugh. A Catalog of

Low Surface Brightness Galaxies. List II. Astronomical Journal, 103:1107, Apr. 1992.

doi:10.1086/116129.

J. M. Schombert, R. A. Pildis, and J. A. Eder. Gas-rich Dwarfs from the Second Palomar

Sky Survey. I. Catalog and Characteristics. Astrophysical Journal Supplement Series, 111

(1):233–244, July 1997. doi:10.1086/313016.

C. Sengupta, T. C. Scott, A. Chung, et al. Dark matter and H I in ultra-diﬀuse galaxy UGC

2162. Monthly Notices of the Royal Astronomical Society, 488(3):3222–3230, Sept. 2019.

doi:10.1093/mnras/stz1884.

D. D. Shi, X. Z. Zheng, H. B. Zhao, Z. Z. Pan, B. Li, H. Zou, X. Zhou, K. Guo, F. X. An,

and Y. B. Li. Deep Imaging of the HCG 95 Field. I. Ultra-diﬀuse Galaxies. Astrophysical

Journal, 846:26, Sept. 2017. doi:10.3847/1538-4357/aa8327.

F. J. Sicking. The thickness of the HI gas layer in spiral galaxies. PhD thesis, University

of Groningen, Dec. 1997. URL https://ui.adsabs.harvard.edu/abs/1997PhDT.....

...40S.

C. Sifón, R. F. J. van der Burg, H. Hoekstra, A. Muzzin, and R. Herbonnet. A ﬁrst

constraint on the average mass of ultra-diﬀuse galaxies from weak gravitational lens-

ing. Monthly Notices of the Royal Astronomical Society, 473(3):3747–3754, Jan. 2018.

doi:10.1093/mnras/stx2648. BIBLIOGRAPHY 186

D. Sijacki, M. Vogelsberger, D. Kereš, V. Springel, and L. Hernquist. Moving mesh cosmol-

ogy: the hydrodynamics of galaxy formation. Monthly Notices of the Royal Astronomical

Society, 424(4):2999–3027, Aug. 2012. doi:10.1111/j.1365-2966.2012.21466.x.

J. D. Simon. The Faintest Dwarf Galaxies. Annual Reviews of Astronomy and Astrophysics,

57:375–415, Aug. 2019. doi:10.1146/annurev-astro-091918-104453.

C. M. Simpson, R. J. J. Grand, F. A. Gómez, F. Marinacci, R. Pakmor, V. Springel, D. J. R.

Campbell, and C. S. Frenk. Quenching and ram pressure stripping of simulated Milky

Way satellite galaxies. Monthly Notices of the Royal Astronomical Society, 478(1):548–567,

July 2018a. doi:10.1093/mnras/sty774.

C. M. Simpson, R. J. J. Grand, F. A. Gómez, F. Marinacci, R. Pakmor, V. Springel, D. J. R.

Campbell, and C. S. Frenk. Quenching and ram pressure stripping of simulated Milky Way

satellite galaxies. Monthly Notices of the Royal Astronomical Society, 478(1):548–567, Jul

2018b. doi:10.1093/mnras/sty774.

P. R. Singh, D. Zaritsky, R. Donnerstein, and K. Spekkens. Ultra-diﬀuse Galaxies at Ul-

traviolet Wavelengths. Astronomical Journal, 157(5):212, May 2019. doi:10.3847/1538-

3881/ab16f2.

C. T. Slater and E. F. Bell. The Mass Dependence of Dwarf Satellite Galaxy Quenching.

Astrophysical Journal, 792(2):141, Sept. 2014. doi:10.1088/0004-637X/792/2/141.

C. T. Slater, E. F. Bell, and N. F. Martin. Andromeda XXVIII: A Dwarf Galaxy More

Than 350 kpc from Andromeda. Astrophysical Journal Letters, 742(1):L14, Nov. 2011.

doi:10.1088/2041-8205/742/1/L14.

A. Smercina, E. F. Bell, P. A. Price, R. D’Souza, C. T. Slater, J. Bailin, A. Monachesi, and

D. Nidever. A Lonely Giant: The Sparse Satellite Population of M94 Challenges Galaxy

Formation. Astrophysical Journal, 863(2):152, Aug 2018. doi:10.3847/1538-4357/aad2d6. BIBLIOGRAPHY 187

K. Spekkens and A. Karunakaran. Atomic Gas in Blue Ultra Diﬀuse Galaxies around

Hickson Compact Groups. Astrophysical Journal, 855(1):28, Mar 2018. doi:10.3847/1538-

4357/aa94be.

K. Spekkens, B. S. Mason, J. E. Aguirre, and B. Nhan. A Deep Search for Extended Radio

Continuum Emission from Dwarf Spheroidal Galaxies: Implications for Particle Dark

Matter. Astrophysical Journal, 773(1):61, Aug 2013. doi:10.1088/0004-637X/773/1/61.

K. Spekkens, N. Urbancic, B. S. Mason, B. Willman, and J. E. Aguirre. The Dearth of

Neutral Hydrogen in Galactic Dwarf Spheroidal Galaxies. Astrophysical Journal Letters,

795(1):L5, Nov 2014. doi:10.1088/2041-8205/795/1/L5.

V. Springel and L. Hernquist. Cosmological smoothed particle hydrodynamics simulations:

a hybrid multiphase model for star formation. Monthly Notices of the Royal Astronomical

Society, 339(2):289–311, Feb. 2003. doi:10.1046/j.1365-8711.2003.06206.x.

V. Springel, S. D. M. White, G. Tormen, and G. Kauﬀmann. Populating a cluster of galaxies

- I. Results at [formmu2]z=0. Monthly Notices of the Royal Astronomical Society, 328(3):

726–750, Dec. 2001. doi:10.1046/j.1365-8711.2001.04912.x.

C. M. Springob, M. P. Haynes, R. Giovanelli, and B. R. Kent. A Digital Archive of H I 21

Centimeter Line Spectra of Optically Targeted Galaxies. Astrophysical Journal Supple-

ment Series, 160:149–162, Sept. 2005. doi:10.1086/431550.

T. K. Starkenburg, L. V. Sales, S. Genel, et al. On the Origin of Star-Gas Counterrotation

in Low-mass Galaxies. Astrophysical Journal, 878(2):143, June 2019. doi:10.3847/1538-

4357/ab2128.

C. Stone, N. Arora, S. Courteau, and J.-C. Cuillandre. AutoProf – I. An automated

non-parametric light proﬁle pipeline for modern galaxy surveys. arXiv e-prints, art.

arXiv:2106.13809, June 2021. BIBLIOGRAPHY 188

R. Taylor, J. I. Davies, R. Auld, and R. F. Minchin. The Arecibo Galaxy Environment

Survey - V. The Virgo cluster (I). Monthly Notices of the Royal Astronomical Society, 423

(1):787–810, June 2012. doi:10.1111/j.1365-2966.2012.20914.x.

T. X. Thuan and G. E. Martin. Blue compact dwarf galaxies. I - Neutral hydrogen observa-

tions of 115 galaxies. Astrophysical Journal, 247:823–848, Aug. 1981. doi:10.1086/159094.

N. A. Tikhonov, V. S. Lebedev, and O. A. Galazutdinova. M 101 group galaxies. Astronomy

Letters, 41:239–251, June 2015. doi:10.1134/S1063773715060080.

G. Torrealba, S. E. Koposov, V. Belokurov, and M. Irwin. The feeble giant. Discov-

ery of a large and diﬀuse Milky Way dwarf galaxy in the constellation of Crater.

Monthly Notices of the Royal Astronomical Society, 459(3):2370–2378, July 2016a.

doi:10.1093/mnras/stw733.

G. Torrealba, S. E. Koposov, V. Belokurov, M. Irwin, M. Collins, M. Spencer, R. Ibata,

M. Mateo, A. Bonaca, and P. Jethwa. At the survey limits: discovery of the Aquarius

2 dwarf galaxy in the VST ATLAS and the SDSS data. Monthly Notices of the Royal

Astronomical Society, 463(1):712–722, Nov. 2016b. doi:10.1093/mnras/stw2051.

G. Torrealba, V. Belokurov, S. E. Koposov, T. S. Li, M. G. Walker, J. L. Sanders,

A. Geringer-Sameth, D. B. Zucker, K. Kuehn, N. W. Evans, and W. Dehnen.

The hidden giant: discovery of an enormous Galactic dwarf satellite in Gaia DR2.

Monthly Notices of the Royal Astronomical Society, 488(2):2743–2766, Sept. 2019.

doi:10.1093/mnras/stz1624.

C. Trachternach, D. J. Bomans, L. Haberzettl, and R. J. Dettmar. An optical search for low

surface brightness galaxies in the Arecibo HI Strip Survey. Astronomy & Astrophysics,

458(1):341–348, Oct. 2006. doi:10.1051/0004-6361:20054545.

M. Tremmel, T. R. Quinn, A. Ricarte, A. Babul, U. Chadayammuri, P. Natarajan, D. Nagai, BIBLIOGRAPHY 189

A. Pontzen, and M. Volonteri. Introducing ROMULUSC: a cosmological simulation of a

galaxy cluster with an unprecedented resolution. Monthly Notices of the Royal Astronom-

ical Society, 483(3):3336–3362, Mar. 2019. doi:10.1093/mnras/sty3336.

M. Tremmel, A. C. Wright, A. M. Brooks, F. Munshi, D. Nagai, and T. R. Quinn.

The formation of ultradiﬀuse galaxies in the RomulusC galaxy cluster simulation.

Monthly Notices of the Royal Astronomical Society, 497(3):2786–2810, Sept. 2020.

doi:10.1093/mnras/staa2015.

I. Trujillo and J. Fliri. Beyond 31 mag arcsec−2: The Frontier of Low Surface Brightness

Imaging with the Largest Optical Telescopes. Astrophysical Journal, 823(2):123, June

2016. doi:10.3847/0004-637X/823/2/123.

I. Trujillo, J. Roman, M. Filho, and J. Sánchez Almeida. The Nearest Ultra Diﬀuse Galaxy:

UGC 2162. Astrophysical Journal, 836:191, Feb. 2017. doi:10.3847/1538-4357/aa5cbb.

I. Trujillo, M. A. Beasley, A. Borlaﬀ, E. R. Carrasco, A. Di Cintio, M. Filho, M. Mon-

elli, M. Montes, J. Román, T. Ruiz-Lara, J. Sánchez Almeida, D. Valls-Gabaud, and

A. Vazdekis. A distance of 13 Mpc resolves the claimed anomalies of the galaxy lacking

dark matter. Monthly Notices of the Royal Astronomical Society, 486(1):1192–1219, June

2019. doi:10.1093/mnras/stz771.

R. B. Tully. Galaxy Groups: A 2MASS Catalog. Astronomical Journal, 149:171, May 2015.

doi:10.1088/0004-6256/149/5/171.

R. F. J. van der Burg, H. Hoekstra, A. Muzzin, C. Sifón, M. Viola, M. N. Bremer, S. Brough,

S. P. Driver, T. Erben, C. Heymans, H. Hildebrandt, B. W. Holwerda, D. Klaes, K. Kui-

jken, S. McGee, R. Nakajima, N. Napolitano, P. Norberg, E. N. Taylor, and E. Valen-

tijn. The abundance of ultra-diﬀuse galaxies from groups to clusters. UDGs are relatively

more common in more massive haloes. Astronomy & Astrophysics, 607:A79, Nov 2017.

doi:10.1051/0004-6361/201731335. BIBLIOGRAPHY 190

P. van Dokkum, R. Abraham, J. Brodie, C. Conroy, S. Danieli, A. Merritt, L. Mowla,

A. Romanowsky, and J. Zhang. A High Stellar Velocity Dispersion and 100 Globular

Clusters for the Ultra-diﬀuse Galaxy Dragonﬂy 44. Astrophysical Journal, 828:L6, Sep

2016. doi:10.3847/2041-8205/828/1/L6.

P. van Dokkum, S. Danieli, Y. Cohen, A. Merritt, A. J. Romanowsky, R. Abraham, J. Brodie,

C. Conroy, D. Lokhorst, L. Mowla, E. O’Sullivan, and J. Zhang. A galaxy lacking dark

matter. Nature, 555:629–632, Mar 2018. doi:10.1038/nature25767.

P. van Dokkum, S. Danieli, R. Abraham, C. Conroy, and A. J. Romanowsky. A Second

Galaxy Missing Dark Matter in the NGC 1052 Group. Astrophysical Journal Letters, 874

(1):L5, Mar. 2019. doi:10.3847/2041-8213/ab0d92.

P. G. van Dokkum, R. Abraham, and A. Merritt. First Results from the Dragonﬂy Tele-

photo Array: The Apparent Lack of a Stellar Halo in the Massive Spiral Galaxy M101.

Astrophysical Journal Letters, 782(2):L24, Feb. 2014. doi:10.1088/2041-8205/782/2/L24.

P. G. van Dokkum, R. Abraham, A. Merritt, J. Zhang, M. Geha, and C. Conroy. Forty-

seven Milky Way-sized, Extremely Diﬀuse Galaxies in the Coma Cluster. Astrophysical

Journal Letters, 798:L45, Jan. 2015a. doi:10.1088/2041-8205/798/2/L45.

P. G. van Dokkum, A. J. Romanowsky, R. Abraham, J. P. Brodie, C. Conroy, M. Geha,

A. Merritt, A. Villaume, and J. Zhang. Spectroscopic Conﬁrmation of the Existence of

Large, Diﬀuse Galaxies in the Coma Cluster. Astrophysical Journal Letters, 804:L26, May

2015b. doi:10.1088/2041-8205/804/1/L26.

L. van Zee, R. J. Maddalena, M. P. Haynes, et al. Obtaining High Precision HI Fluxes for

Galaxies. Astronomical Journal, 113:1638–1647, May 1997. doi:10.1086/118380.

A. Venhola, R. Peletier, E. Laurikainen, et al. The Fornax Deep Survey with VST. III. Low BIBLIOGRAPHY 191

surface brightness dwarfs and ultra diﬀuse galaxies in the center of the Fornax cluster.

Astronomy & Astrophysics, 608:A142, Dec. 2017. doi:10.1051/0004-6361/201730696.

M. A. W. Verheijen and R. Sancisi. The Ursa Major cluster of galaxies. IV. HI synthesis

observations. Astronomy & Astrophysics, 370:765–867, May 2001. doi:10.1051/0004-

6361:20010090.

M. Vogelsberger, S. Genel, D. Sijacki, P. Torrey, V. Springel, and L. Hernquist. A model

for cosmological simulations of galaxy formation physics. Monthly Notices of the Royal

Astronomical Society, 436(4):3031–3067, Dec. 2013. doi:10.1093/mnras/stt1789.

M. Vogelsberger, S. Genel, V. Springel, P. Torrey, D. Sijacki, D. Xu, G. Snyder, D. Nelson,

and L. Hernquist. Introducing the Illustris Project: simulating the coevolution of dark

and visible matter in the Universe. Monthly Notices of the Royal Astronomical Society,

444(2):1518–1547, Oct. 2014. doi:10.1093/mnras/stu1536.

L. Wang, A. A. Dutton, G. S. Stinson, A. V. Macciò, C. Penzo, X. Kang, B. W. Keller, and

J. Wadsley. NIHAO project - I. Reproducing the ineﬃciency of galaxy formation across

cosmic time with a large sample of cosmological hydrodynamical simulations. Monthly No-

tices of the Royal Astronomical Society, 454:83–94, Nov 2015. doi:10.1093/mnras/stv1937.

W. Wang, L. V. Sales, B. M. B. Henriques, and S. D. M. White. Satellite abun-

dances around bright isolated galaxies - II. Radial distribution and environmental ef-

fects. Monthly Notices of the Royal Astronomical Society, 442(2):1363–1378, Aug. 2014.

doi:10.1093/mnras/stu988.

D. R. Weisz, A. E. Dolphin, N. F. Martin, S. M. Albers, M. L. M. Collins, A. M. N.

Ferguson, G. F. Lewis, A. D. Mackey, A. McConnachie, R. M. Rich, and E. D. Skill-

man. A rogues gallery of Andromeda’s dwarf galaxies - II. Precise distances to 17 faint

satellites. Monthly Notices of the Royal Astronomical Society, 489(1):763–770, Oct. 2019.

doi:10.1093/mnras/stz1984. BIBLIOGRAPHY 192

A. R. Wetzel, E. J. Tollerud, and D. R. Weisz. Rapid Environmental Quenching of Satellite

Dwarf Galaxies in the Local Group. Astrophysical Journal Letters, 808(1):L27, July 2015a.

doi:10.1088/2041-8205/808/1/L27.

A. R. Wetzel, E. J. Tollerud, and D. R. Weisz. Rapid Environmental Quenching of Satellite

Dwarf Galaxies in the Local Group. Astrophysical Journal Letters, 808(1):L27, Jul 2015b.

doi:10.1088/2041-8205/808/1/L27.

A. R. Wetzel, P. F. Hopkins, J.-h. Kim, C.-A. Faucher-Giguère, D. Kereš, and E. Quataert.

Reconciling Dwarf Galaxies with ΛCDM Cosmology: Simulating a Realistic Population

of Satellites around a Milky Way-mass Galaxy. Astrophysical Journal Letters, 827(2):L23,

Aug. 2016. doi:10.3847/2041-8205/827/2/L23.

C. Wheeler, J. I. Phillips, M. C. Cooper, M. Boylan-Kolchin, and J. S. Bullock. The surpris-

ing ineﬃciency of dwarf satellite quenching. Monthly Notices of the Royal Astronomical

Society, 442(2):1396–1404, Aug. 2014. doi:10.1093/mnras/stu965.

C. Wheeler, J. Oñorbe, J. S. Bullock, M. Boylan-Kolchin, O. D. Elbert, S. Garrison-Kimmel,

P. F. Hopkins, and D. Kereš. Sweating the small stuﬀ: simulating dwarf galaxies, ultra-

faint dwarf galaxies, and their own tiny satellites. Monthly Notices of the Royal Astro-

nomical Society, 453(2):1305–1316, Oct 2015. doi:10.1093/mnras/stv1691.

C. Wheeler, P. F. Hopkins, A. B. Pace, S. Garrison-Kimmel, M. Boylan-Kolchin, A. Wet-

zel, J. S. Bullock, D. Kereš, C.-A. Faucher-Giguère, and E. Quataert. Be it there-

fore resolved: cosmological simulations of dwarf galaxies with 30 solar mass resolu-

tion. Monthly Notices of the Royal Astronomical Society, 490(3):4447–4463, Dec. 2019.

doi:10.1093/mnras/stz2887.

A. C. Wright, M. Tremmel, A. M. Brooks, F. Munshi, D. Nagai, R. S. Sharma, and T. R.

Quinn. The formation of isolated ultradiﬀuse galaxies in ROMULUS25. Monthly Notices of

the Royal Astronomical Society, 502(4):5370–5389, Apr. 2021. doi:10.1093/mnras/stab081. BIBLIOGRAPHY 193

T. K. Wyder, D. C. Martin, D. Schiminovich, M. Seibert, T. Budavári, M. A. Treyer,

T. A. Barlow, K. Forster, P. G. Friedman, P. Morrissey, S. G. Neﬀ, T. Small, L. Bianchi,

J. Donas, T. M. Heckman, Y.-W. Lee, B. F. Madore, B. Milliard, R. M. Rich, A. S.

Szalay, B. Y. Welsh, and S. K. Yi. The UV-Optical Galaxy Color-Magnitude Diagram.

I. Basic Properties. Astrophysical Journal Supplement Series, 173(2):293–314, Dec. 2007.

doi:10.1086/521402.

M. Yagi, J. Koda, Y. Komiyama, and H. Yamanoi. Catalog of Ultra-diﬀuse Galaxies in the

Coma Clusters from Subaru Imaging Data. Astrophysical Journal Supplement Series, 225:

11, July 2016. doi:10.3847/0067-0049/225/1/11.

C. Yozin and K. Bekki. The quenching and survival of ultra diﬀuse galaxies in the Coma

cluster. Monthly Notices of the Royal Astronomical Society, 452:937–943, Sep 2015.

doi:10.1093/mnras/stv1073.

D. Zaritsky. Clues to the nature of ultradiﬀuse galaxies from estimated galaxy velocity

dispersions. Monthly Notices of the Royal Astronomical Society, 464(1):L110–L113, Jan.

2017. doi:10.1093/mnrasl/slw198.

D. Zaritsky, R. Smith, C. Frenk, and S. D. M. White. Satellites of Spiral Galaxies. Astro-

physical Journal, 405:464, Mar. 1993. doi:10.1086/172379.

D. Zaritsky, R. Smith, C. Frenk, and S. D. M. White. More Satellites of Spiral Galaxies.

Astrophysical Journal, 478(1):39–48, Mar. 1997. doi:10.1086/303784.

D. Zaritsky, R. Donnerstein, A. Dey, J. Kadowaki, H. Zhang, A. Karunakaran, D. Martínez-

Delgado, M. Rahman, and K. Spekkens. Systematically Measuring Ultra-diﬀuse Galaxies

(SMUDGes). I. Survey Description and First Results in the Coma Galaxy Cluster and En-

virons. The Astrophysical Journal Supplement Series, 240:1, Jan 2019. doi:10.3847/1538-

4365/aaefe9. BIBLIOGRAPHY 194

H.-X. Zhang, T. H. Puzia, and D. R. Weisz. The Impact of Star Formation Histories on Stel-

lar Mass Estimation: Implications from the Local Group Dwarf Galaxies. Astrophysical

Journal Supplement Series, 233(1):13, Nov. 2017. doi:10.3847/1538-4365/aa937b.

F. Zwicky. Morphological astronomy. Springer, 1957. URL https://ui.adsabs.harvard.

edu/abs/1957moas.book.....Z.