An X-Ray Study of Massive Star Forming Regions With

Home , NGC 1987, NGC 2244, NGC 6334, NGC 6357, Photoevaporation, RCW 120

The Pennsylvania State University

The Graduate School

Department of Astronomy and Astrophysics

AN X-RAY STUDY OF MASSIVE STAR FORMING REGIONS

WITH CHANDRA

A Thesis in

Astronomy and Astrophysics

Junfeng Wang

c 2007 Junfeng Wang

Submitted in Partial Fulﬁllment of the Requirements for the Degree of

Doctor of Philosophy

December 2007 The thesis of Junfeng Wang was read and approved∗ by the following:

Eric D. Feigelson Professor of Astronomy and Astrophysics Thesis Co-adviser Chair of Committee

Leisa K. Townsley Senior Scientist of Astronomy and Astrophysics Thesis Co-adviser Special Member

Kevin Luhman Assistant Professor of Astronomy and Astrophysics

Hiroshi Ohmoto Professor of Geochemistry

Donald Schneider Professor of Astronomy and Astrophysics

Richard Wade Associate Professor of Astronomy and Astrophysics

Lawrence Ramsey Professor of Astronomy and Astrophysics Head of the Department of Astronomy and Astrophysics

∗Signatures on ﬁle in the Graduate School. iii Abstract

Massive stars are characterized by powerful stellar winds, strong ultraviolet (UV) radiation, and consequently devastating supernovae explosions, which have a profound influence on their natal clouds and galaxy evolution. However, the formation and evolution of massive stars themselves and how their low-mass siblings are affected in the wind-swept and UV-radiation-dominated environment are not well understood. Much of the stellar populations inside of the massive star forming regions (MSFRs) are poorly studied in the optical and IR wavelengths because of observational challenges caused by large distance, high extinction, and heavy contamination from unrelated sources. Al- though it has long been recognized that X-rays open a new window to sample the young stellar populations residing in the MSFRs, the low angular resolution of previous generation X-ray telescopes has limited the outcome from such studies. The sensitive high spatial resolution X-ray observations enabled by the Chandra X-ray Observatory and the Advanced CCD Imaging Spectrometer (ACIS) have significantly improved our ability to study the X-ray-emitting populations in the MSFRs in the last few years. In this thesis, I analyzed seven high spatial resolution Chandra/ACIS images of two massive star forming complexes, namely the NGC 6357 region hosting the 1 Myr old Pismis 24 cluster (Chapter 3) and the Rosette Complex including the 2 Myr old NGC 2244 cluster immersed in the Rosette Nebula (Chapter 4), embedded clusters in the Rosette Molecular Cloud (RMC; Chapter 5), and a triggered cluster NGC 2237 (Chapter 6). The X-ray sampled stars were studied in great details. The unique power of X-ray selection of young stellar cluster members yielded new knowledge in the stellar populations, the cluster structures, and the star formation histories. The census of cluster members is greatly improved in each region. A large fraction of the X-ray detections have optical or near-infrared (NIR) stellar counterparts (from 2MASS, SIRIUS and FLAMINGOS JHK images), most of which are previously uncatalogued young cluster members. This provides a reliable probe of the rich intermediate- mass and low-mass young stellar populations accompanying the massive OB stars in each region. For example, In the poorly-studied NGC 6357 region, our study increased the number of known members from optical study by a factor of 40. As a result, ∼ normal initial mass functions (IMFs) for NGC 6357 and NGC 2244 were found, inconsistent with the top-heavy IMFs suspected in previous optical studies. The observed X-ray luminosity functions (XLFs) in NGC 6357 and NGC 2244 are compared to the Orion Nebula Cluster XLF, yielding the first estimate of NGC 6357’s total cluster population, a few times the known Orion population. For NGC 2244, a total population of 2000 ∼ X-ray-emitting stars is derived, consistent with previous estimate from IR studies. The morphologies and spatial structures of the clusters are investigated with absorption-stratified stellar surface density maps. Small-scale substructures superposed on the spherical clusters are found in NGC 6357 and NGC 2244. Both of their radial stellar density profiles show a power-law cusp around the density peak surrounded by an isothermal sphere. In NGC 2244, the spatial distribution of X-ray stars is strongly concentrated around the central O5 star, HD 46150. The other O4 star HD 46223 has iv few companions. The X-ray sources in the RMC show three distinctive structures and substructures within them, which include previously known embedded IR clusters and a new unobscured cluster (RMC A). We do not find clear evidence of sequentially triggered formation. The concentration of X-ray identified young stars implies that .35% of stars could be in a distributed population throughout the RMC region and clustered star formation is the dominant mode in this cloud. The NGC 2237 cluster, similar to RMC A, may have formed from collapse of pre-existing massive molecular clumps accompanying the formation of the NGC 2244 cluster. The spatial distribution of the NIR counterparts to X-ray stars in the optical dark region northwest of NGC 2237 show little evidence of triggered star formation in the pillar objects. The observed inner disk fraction in the MSFRs as indicated by K-band excess appears lower than the IR-excess disk fractions found in the nearby low-mass star formation regions of similar age. An overall K-excess disk frequency of 6% for X-ray ∼ selected stars in the intermediate- to high-mass range in the NGC 6357 region (Chapter 3), and 10% for stars with mass M & 2M⊙ in NGC 2244 (Chapter 4) are derived, which ∼ indicates that the inner disks around higher-mass stars evolve more rapidly. The X-ray stars in these regions provide an important new sample for studies of intermediate-mass PMS stars that are not accreting, in addition to the accreting HAeBe stars. The low K-excess disk frequency for X-ray selected stars in the solar mass range in NGC 2244 is intriguing, which may be attributed to different sensitivities to disk materials, selection effects between X-ray samples and IR samples and/or faster disk dissipation due to photoevaporation in the MSFRs. X-ray properties of stars across the mass spectrum are presented. Diversities in the X-ray spectra of O stars are seen, both soft X-ray emission consistent with the microshocks in stellar winds and hard X-ray components signifying magnetically confined winds or close binarity. X-ray luminosities for a sample of stars earlier than B4 in NGC 6357, NGC 2244, and M 17 confirm the long-standing log(L /L ) 7 relation, x bol ∼− although larger scatter is seen among the Lx/Lbol ratios of B-type stars. Low-mass PMS stars frequently show X-ray flaring, including intense flares with luminosities above 32 1 L 10 ergs s− . Diffuse X-ray emission is present in the NGC 6357 region and in the x ≥ NGC 2244 cluster. The derived luminosity of diffuse emission in NGC 6357 is consistent with the integrated emission from the unresolved PMS stars. The NGC 2244 diffuse emission is likely originated from the wind termination shocks, and hence is truly diffuse in nature. In summary, Chandra X-ray observations offer multifaceted approaches to study the young stellar clusters in MSFRs in depth. Future perspectives with the Spitzer Space Telescope mid-IR observations for a systematic measurement of disk frequencies in X-ray sampled massive clusters and X-ray observations of the earliest phases of massive star formation are discussed. v Table of Contents

List of Tables ...... viii

List of Figures ...... x

Acknowledgments ...... xii

Chapter 1. Introduction ...... 1 1.1 Observational Overview of Star Forming Regions ...... 2 1.1.1 Low-mass Star Forming Regions ...... 2 1.1.2 Massive Star Forming Regions ...... 4 1.1.3 X-ray Observations of Massive Star Forming Regions . . ... 5 1.2 MotivationsforthisThesis...... 7 1.2.1 Advantages of Chandra Observations of the MSFRs . . . . . 7 1.2.2 Some Outstanding Issues to Be Addressed in Understanding MSFRs in the Chandra Era...... 8 1.3 OverviewofthisThesis ...... 10 1.3.1 Chapter 3: An X-ray Census of Young Stars in the Massive Southern Star-Forming Complex NGC 6357 ...... 11 1.3.2 Chapter 4: A Chandra Study of the Young Open Cluster NGC 2244intheRosetteNebula ...... 11 1.3.3 Chapter 5: A Chandra Study of the Stellar Populations in the Rosette Molecular Cloud ...... 12 1.3.4 Chapter 6: A Chandra Study of the Triggered Cluster NGC 2237...... 13

Chapter 2. Methods ...... 16 2.1 TelescopeandInstrument ...... 16 2.2 BasicACISDataReduction ...... 17 2.3 DataAnalysis...... 18 2.3.1 ACISExtract...... 18 2.3.2 SourceFinding ...... 18 2.3.3 Source Variability and Spectral Fitting ...... 20 2.3.4 Simulation of Extragalactic Contamination ...... 21 2.3.5 Simulation of Stellar Contamination in the Galactic Disk . . 22 2.3.6 NIR Color-Color and Color-Magnitude Diagrams ...... 23 2.3.7 Stellar Surface Density Map ...... 23

Chapter 3. An X-ray Census of Young Stars in the Massive Southern Star-Forming Complex NGC 6357 ...... 26 3.1 Introduction...... 26 3.2 Observational Overview of NGC 6357 and Pismis 24 ...... 27 vi

3.3 Observations and Data Reduction ...... 29 3.3.1 Chandra Observation and Data Selection ...... 29 3.3.2 Image Reconstruction and Source Finding ...... 29 3.3.3 Photon Extraction and Limiting Sensitivity ...... 30 3.3.4 Source Variability and Spectral Fitting ...... 32 3.3.5 SIRIUSNIRObservation ...... 33 3.4 Identification of the Chandra Sources ...... 33 3.4.1 X-ray Sources with Stellar Counterparts ...... 33 3.4.2 Extragalactic Contamination ...... 34 3.4.3 Field Star Contamination ...... 35 3.4.4 Likely New Stellar Members ...... 36 3.4.5 Classification of the X-ray Sample ...... 36 3.4.6 EGGsandProtostars ...... 39 3.5 Properties of the Stellar Cluster and Its Environment ...... 39 3.5.1 X-ray Luminosity Function ...... 39 3.5.2 Spatial Distribution of the Stellar Cluster ...... 41 3.5.3 The Morphology of the Ring-like Nebula ...... 43 3.6 X-raysacrosstheMassSpectrum ...... 44 3.6.1 X-raysfromknownmassivestars ...... 44 3.6.2 Newly-discovered Candidate O Stars ...... 46 3.6.3 Intermediate Mass Stars ...... 47 3.6.4 FlaringPMSStars ...... 48 3.6.5 X-ray Selected Deeply Embedded Population ...... 48 3.7 DiffuseX-rayEmission ...... 49 3.8 Summary ...... 50

Chapter 4. A Chandra Study of the Young Open Cluster NGC 2244 in the Rosette Nebula ...... 80 4.1 Introduction...... 80 4.2 Chandra Observations and Data Reduction ...... 83 4.2.1 Source Finding and Photon Extraction ...... 83 4.2.2 Source Variability ...... 84 4.2.3 SpectralFitting...... 85 4.3 Identiﬁcation of Stellar Counterparts and Their Properties ...... 85 4.3.1 Stellar Counterparts Matching and IR diagrams ...... 85 4.3.2 Extragalactic Contaminants ...... 88 4.3.3 Galactic Stellar Contamination ...... 88 4.4 Global Properties of the Stellar Cluster ...... 90 4.4.1 X-ray Luminosity Function and Initial Mass Function . . . . 90 4.4.2 Initial Mass Function and K-band Luminosity Function . . . 91 4.4.3 Morphology and Substructures ...... 93 4.4.4 RadialDensityProﬁle ...... 94 4.4.5 Mass Segregation ...... 95 4.4.6 X-ray Stars with Infrared Excess Disk ...... 96 4.5 X-raysacrosstheMassSpectrum ...... 98 vii

4.5.1 X-raysfromMassiveStars...... 98 4.5.2 X-rays from Intermediate Mass Stars ...... 99 4.5.3 X-rays from Other Interesting Sources ...... 100 4.6 SoftDiﬀuseEmission...... 102 4.7 Summary ...... 102

Chapter 5. A Chandra Study of the Stellar Populations in the Rosette Molecular Cloud ...... 132 5.1 Introduction...... 132 5.2 Chandra Observations and Data Reduction ...... 133 5.3 Identification of Stellar Counterparts ...... 134 5.4 X-rays from Known Massive stars, Protostars, and Other Interesting Sources ...... 135 5.5 Spatial Distribution of the Stellar Population ...... 136 5.6 Properties of the X-ray Sampled Stellar Clusters ...... 138 5.6.1 SensitivityLimits...... 138 5.6.2 NIR Color-Color and Color-Magnitude Diagrams ...... 139 5.6.3 Comparison with CO Emission Morphology and Román-Zúñiga et al. (2007a) NIR Clusters ...... 139 5.7 Formation Modes of Clusters and Stars ...... 141 5.7.1 Sequential Star Formation and Spontaneous Star Formation . 141 5.7.2 Clustered Star Formation vs. Distributed Star Formation . . 143 5.8 Summary ...... 143

Chapter 6. A Chandra Study of the Triggered Cluster NGC 2237 ...... 182 6.1 Introduction...... 182 6.2 Chandra Observations and Data Reduction ...... 183 6.3 Stellar Counterparts to Chandra Sources...... 185 6.4 Properties of the Stellar Population ...... 187 6.4.1 Spatial Appearance of the NGC 2237 Cluster ...... 187 6.4.2 Distribution of the NIR Counterparts to the Chandra Stars . 187 6.4.3 Notes on Interesting Sources ...... 189 6.5 Summary ...... 189

Chapter 7. Conclusions and Future Prospects ...... 213 7.1 Summary ...... 213 7.2 Implications for Star Formation ...... 219 7.3 FutureWork ...... 220 7.3.1 A Synergy Study of Young Stellar Clusters with Two Great Observatories ...... 220 7.3.2 X-ray Studies of Structures of Young Stellar Clusters . . . . . 221 7.3.3 X-ray Observations of Precursors to Star Clusters ...... 221

Bibliography ...... 226 viii List of Tables

3.1 Main Chandra Catalog: Basic Source Properties ...... 53 3.2 Tentative Source Properties ...... 54 3.3 X-ray Spectroscopy for Brighter Sources: Thermal Plasma Fits . . . . . 55 3.4 X-ray Spectroscopy for Less Bright Sources: Thermal Plasma Fits . . . 56 3.5 Stellar Counterparts and Classiﬁcations ...... 57 3.6 KnownOandEarlyBstarsinPismis24 ...... 58 3.7 X-ray and IR luminous stars as Candidate O stars ...... 59 3.8 X-ray Selected Candidate Intermediate-mass Stars ...... 60 3.9 X-ray Selected Heavily Obscured Sources ...... 61

4.1 Log of Chandra Observations ...... 104 4.2 Chandra Main Catalog: Basic Source Properties ...... 105 4.3 Chandra Secondary Catalog: Tentative Source Properties ...... 106 4.4 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 107 4.5 X-ray Spectroscopy for Photometrically Selected Sources: Power Law Fits ...... 108 4.6 StellarCounterparts ...... 109 4.7 X-ray properties of cataloged OB stars in NGC 2244 ...... 110

5.1 Chandra Catalog: Primary Source Properties ...... 145 5.1 Chandra Catalog: Primary Source Properties ...... 146 5.1 Chandra Catalog: Primary Source Properties ...... 147 5.1 Chandra Catalog: Primary Source Properties ...... 148 5.1 Chandra Catalog: Primary Source Properties ...... 149 5.1 Chandra Catalog: Primary Source Properties ...... 150 5.1 Chandra Catalog: Primary Source Properties ...... 151 5.1 Chandra Catalog: Primary Source Properties ...... 152 5.2 Chandra Catalog: Tentative Source Properties ...... 153 5.2 Chandra Catalog: Tentative Source Properties ...... 154 5.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 155 5.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 156 5.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 157 5.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 158 5.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 159 ix

5.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 160 5.4 X-ray Spectroscopy for Photometrically Selected Sources: Power Law Fits ...... 161 5.5 StellarCounterparts ...... 162 5.5 StellarCounterparts ...... 163 5.5 StellarCounterparts ...... 164 5.5 StellarCounterparts ...... 165 5.5 StellarCounterparts ...... 166 5.5 StellarCounterparts ...... 167 5.5 StellarCounterparts ...... 168 5.5 StellarCounterparts ...... 169 5.5 StellarCounterparts ...... 170 5.6 X-raySampledStellarClusters ...... 171

6.1 Primary Chandra Catalog: Basic Source Properties ...... 191 6.1 Primary Chandra Catalog: Basic Source Properties ...... 192 6.1 Primary Chandra Catalog: Basic Source Properties ...... 193 6.2 Tentative Chandra Catalog: Basic Source Properties ...... 194 6.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 195 6.3 X-ray Spectroscopy for Photometrically Selected Sources: Thermal Plasma Fits ...... 196 6.4 X-ray Spectroscopy for Photometrically Selected Sources: Power Law Fits ...... 197 6.5 StellarCounterparts ...... 198 6.5 StellarCounterparts ...... 199 6.5 StellarCounterparts ...... 200 6.5 StellarCounterparts ...... 201 x List of Figures

1.1 Evolution of a Young Stellar Object ...... 14 1.2 Comparision between the NIR image and the X-ray image of the Orion NebulaCluster ...... 15

2.1 The Chandra Observatory ...... 24 2.2 The Chandra/ACISfocalplane ...... 25

3.1 Multiwavelength images of the NGC 6357 region ...... 62 3.2 ACIS-I images of the NGC 6357 region ...... 63 3.3 ExamplesofcloseX-raysourcepairs ...... 65 3.4 HST/WFPC2 image with 2MASS sources and Chandra sources overlaid 66 3.5 A cartoon illustration of the source classification ...... 67 3.6 2MASS color-color diagram of NGC 6357 X-ray stars ...... 68 3.7 2MASS J H vs. J color-magnitude diagram of NGC 6357 X-ray stars 69 − 3.8 2MASS H K vs. K color-magnitude diagram of NGC 6357 X-ray stars 70 − 3.9 X-rayluminosityfunctions...... 71 3.10 Spatial distribution of the X-ray stars ...... 72 3.11 Projected stellar surface density maps ...... 73 3.12 Radial density profiles for the COUP and the Pismis 24 cluster . . . . . 74 3.13 SIRIUS K image of the central 50′′ region in Pismis 24 ...... 75 3.14 X-rayspectraoftheO3starsandWR93...... 76 3.15 The LX vs. Lbol relation for the X-ray detected OB stars ...... 77 3.16 X-ray properties of flaring source #672 (CXOU J172457.87-341203.9) . 78 3.17 Spatial distribution of the deeply embedded sources ...... 79

12 4.1 COJ = 1 0 emission map and DSS2 R-band image of the Rosette − complex ...... 111 4.2 ACIS-I images of the NGC 2244 cluster ...... 112 4.3 ACIS image and 2MASS-Ks image of the central region around the O5V starHD46150 ...... 114 4.4 Lightcurves of signiﬁcantly variable sources ...... 115 4.5 NIR color-color diagram for NGC 2244 Chandra stars ...... 116 4.6 NIR color-magnitude diagram for NGC 2244 Chandra stars ...... 117 4.7 Cumulative distribution of the median photon energies ...... 118 4.8 Spatial distribution of Chandra sources without ONIR counterparts . . . 119 4.9 X-ray luminosity function for the unobscured NGC 2244 population and comparisonwithotherclusters ...... 120 4.10 Comparison between IMFs of the NGC 2244 X-ray stars and COUP ONC stars...... 121 4.11 Stellar surface density map for the unobscured population in NGC 2244 122 4.12 Hα image and 2MASS Ks image of the secondary overdensity seen in Figure4.11 ...... 123 xi

4.13 The observed radial density profile of the NGC 2244, the ONC, and the NGC6357clusters ...... 124 4.14 The radial density profiles in physical scales ...... 125 4.15 Spatial distribution of the massive stars ...... 126 4.16 Spatial distribution of sources with significant NIR excess ...... 127 4.17 X-ray spectra of six O and early B stars in NGC 2244 ...... 128 4.18 The Lx vs. Lbol relation for X-ray detected O and early B stars in NGC 2244 ...... 129 4.19 Hα image and ACIS image of # 919 neighborhood ...... 130 4.20 X-ray light curve of the eclipsing binary V578 Mon ...... 131

12 5.1 The distribution of the ISM in the RMC outlined by CO continuum emission...... 172 5.2 A Chandra mosaicoftheRMCregion ...... 173 5.3 X-ray light curve of ACIS source #119 ...... 174 5.4 X-ray spectra of sources #119 and #164 ...... 175 5.5 Spitzer mid-IR image of source #89 neighborhood ...... 176 5.6 Stellar surface density maps for the RMC region ...... 177 5.7 NIR color-color diagrams for RMC Chandra stars in three regions . . . . 178 5.8 NIR color-magnitude diagrams for RMC Chandra stars in three regions 179 5.9 Spatial distribution of the X-ray-sampled Class I, II, and III sources . . 180 5.10 Surface density contours for the Chandra RMC sources using Nearest Neighboranalysis...... 181

6.1 DSS2 R-band image of the NGC 2237 region ...... 202 6.2 ACIS-I image of the NGC 2237 region ...... 203 6.3 Smoothed ACIS-I image of the NGC 2237 ...... 204 6.4 NIR color-color diagram for NGC 2237 Chandra stars ...... 205 6.5 NIR color-magnitude diagram for NGC 2237 Chandra stars ...... 206 6.6 The spatial distribution of hard X-ray sources ...... 207 6.7 The stellar surface density for all NGC 2237 sources ...... 208 6.8 Spatial distribution of X-ray-selected Class II and III IRstars...... 209 6.9 Spatial distribution of stars in diﬀerent mass range ...... 210 6.10 X-ray spectra of sources #149 and #54 ...... 211 6.11 Hα image of a possible HH-ﬂow object ...... 212

7.1 Disk fraction as a function of cluster age ...... 224 7.2 A global view of the Rosette complex ...... 225 xii Acknowledgments

I would like to gratefully acknowledge the many people who made this thesis possible after I changed my research topic in the late stage of my graduate study. I cannot overstate my gratitude to my thesis advisors, Prof. Eric Feigelson and Dr. Leisa Townsley. Without their enthusiasm, inspiration and advice throughout the thesis, I would have been lost. I benefit a lot from their good ideas and constant encouragement. I wish to thank my thesis committee members, Profs. Kevin Luhman, Don Schneider, Richard Wade, and Hiroshi Ohmoto, who are helpful and caring. I would like to thank our group members Pat Broos, Kosta Getman, and Masahiro Tsujimoto for providing sound advice and patiently helping with various technical and scientific problems. I am especially grateful to Pat Broos for developing the software tool ACIS Extract, without which it will be “mission impossible” for me to complete analyzing the large datasets in time. I sincerely thank Prof. Gordon Garmire, the Principle Investigator of ACIS, for building a wonderful instrument and generous support to me. I acknowledge our collaborators, Carlos Roman-Zuniga and Elizabeth Lada for providing FLAMINGOS near-IR catalogs and helpful discussions. I thank Prof. Jian Ge for training me in both infrared instrumentation and on-site observations. His support is deeply appreciated. I thank my fellow graduate students for providing a friendly atmosphere in which I learned astronomy as well as American and international culture. I am indebted to my parents, Yongxiang Wang and Caien Song, who raised me and educated me with all they have. I cannot end without thanking my wife, Lei Lin, on whose constant encouragement, support, and love I have relied throughout my graduate life here far from our motherland. My research work has been financially supported by NASA contract NAS8-38252 and Chandra contract SV4-74018 issued by the Chandra X-ray Observatory Center under NASA contract NAS8-03060 awarded to Gordon Garmire, and Chandra X-ray Obser- vatory grants GO1-2008X, GO3-4010X, and GO6-7006X awarded to Leisa Townsley. I thank the generous travel fund from the Zaccheus Daniel Foundation for Astronomical Sciences. xiii

Twinkle twinkle little star, how I wonder what you are. –English lyrics 1

Chapter 1

Introduction

Our night sky is decorated with stars. Despite the improved understanding of how they evolve and consequently die, the formation of stars has remained one of the holy grails that human beings have been chasing after for centuries. However, the birth sites of stars are buried deep within the cold dense dark clouds consisting mostly molecular hydrogen accompanied by large amount of dust, which are inaccessible in visual light. Due to this fact, the origin of stars remain mysterious despite the long history of stellar observations. With the advent of new technologies in the infrared (IR) and radio band in the last half century, astronomers finally are able to probe the inside of star-forming clouds through the high extinction materials. In radio and millimeter/submillimeter, atomic/molecular lines (e.g., HI, CO, NH3, HCN) and masers (e.g., H2O, OH, CH3OH) are often observed in star forming regions and used to trace structures, physical conditions, and even chemical characteristics (e.g., Jessop & Ward-Thompson 2001). The role of molecular clouds as the sites of active star formation in the Galaxy is firmly established (see reviews by Blitz 1993; Williams et al. 2000). In IR, an optically invisible star was discovered in the Orion Nebula with a group of stars in the dark clouds (Becklin & Neugebauer 1967) and confirmed as the first observational evidence of forming stars. Since then, large NIR surveys utilizing JHK arrays have routinely discovered many more young stellar objects (YSOs) in nearby star forming clouds. For example, the Two Micron All Sky Surveys (2MASS) is widely used to identify and study star forming regions (e.g., Bica et al. 2003; Dutra et al. 2003). Modern mid-IR telescopes such as the Spitzer Space Telescope have revealed stunning details of star formation regions by efficient mapping of large areas of obscured environments (e.g., Evans et al. 2003a; Benjamin et al. 2003). For decades, IR and longer wavelength observations have been the dominant approaches to identify clusters and study star formation. Young stars possess circumstellar disks and envelopes, which are bright at IR wavelengths and much elevated from the level of stellar photosphere emission. There have been several widely used methods in identifying YSOs in these wavelengths, mainly based on spectral energy distributions (SEDs; Adams et al. 1987). For Spitzer/IRAC mid-IR observations, Allen et al. (2004) developed a classification scheme based on the mid-IR colors. Lada et al. (1991) identified star clusters in the Orion B cloud by K-band star counting. Adopting an average density of background stars, the cluster membership can be evaluated statistically by subtracting the background contamination from all detections. This star counts technique is proven very useful in study embedded clusters (Gutermuth et al. 2005). In optical, the most commonly adopted method in identifying young stars in spectroscopic surveys is through detection of the Hα emission line, which is produced in high level 2 accretion activity and indicates youth of a star (e.g. Herbig & Bell 1988; Hillenbrand et al. 1993). Last but not least, X-ray emission has long been recognized as an effective indicator of youth of stars (see review by Feigelson & Montmerle 1999). Given the low temperatures ( 10 100 K) in molecular cloud cores where the ∼ − gravitational collapse and consequently star formation occurs, how do X-rays that require presence of high temperature ( 10 100 MK) plasma come into the picture of star ∼ − forming clouds? Surprisingly, stars in a wide range of masses, O stars ( 20 150M⊙) ∼ − and brown dwarfs ( 20 80M ) alike, are X-ray emitters throughout the formation ∼ − Jup process and the early stages of the stellar evolution. The discovery of extended X-ray emission associated with the Orion Nebula by several X-ray satellites (Uhuru, ANS, and SAS-3) first established a firm link between X-ray emission and star forming regions (Giacconi et al. 1972, 1974; den Boggende et al. 1978; Bradt & Kelley 1979). It was not until the advent of the first imaging X-ray satellite Einstein Observatory, were the stars identified to be the sources that are producing the X-rays in the Orion Nebula (Ku & Chanan 1979). With their large field-of-view, Einstein and ROSAT detected a number of X-ray emitting low mass stars called T Tauri stars in nearby star forming regions (e.g., Feigelson & Decampli 1981). We now know that the presence of hot plasma is almost ubiquitous in X-ray observations of young stars (see review by Feigelson et al. 2007). Low-mass stars are magnetically active and the X-ray producing plasmas are generated through the same mechanism as in the Sun. They exhibit X-ray flares that are orders of magnitude more luminous than the contemporary solar flares. Following magnetic reconnection events, hot gas confined along the magnetic field lines can be instantly heated to X-ray emitting temperature. In massive O stars, the X-rays are produced by small-scale shocks in radiatively-driven stellar winds (Lucy & White 1980; Owocki et al. 1988; Owocki & Cohen 1999). Large-scale diffuse X-ray emission originated from the thermalization of high velocity O star winds has also been found in some massive star-forming regions (e.g., M17, Townsley et al. 2003). In rare cases, diffuse X-ray emission from previous undetected supernovae events could have been detected (e.g., Wolk et al. 2002; Townsley et al. 2006a). In the next section, I will give a general overview of the formation of low-mass stars and massive stars. I also briefly review multiwavelength observations of representative star forming regions, with an emphasis on the X-ray observations and the massive star forming regions (MSFRs).

1.1 Observational Overview of Star Forming Regions

1.1.1 Low-mass Star Forming Regions Boldly speaking, the broad picture of low mass star formation is considerably 1 clear. Now it is generally accepted that low-mass stars, including our Sun, form by the gravitational collapse of rotating clumps within molecular clouds, as supported by

1The formation of very low mass objects–brown dwarfs–is reviewed in Whitworth et al. (2007) and Luhman et al. (2007) from a theoretical view and an observational perspective, respectively. 3 numerous observations. It is worth noting that Immanuel Kant (1755) proposed the Neb- ular Hypothesis, one of the oldest surviving theories on how our solar system formed. Figure 1.1 illustrates the evolution scenario of a YSO. Gas and dust contract onto the central protostar, accompanied by jets and outflows to remove angular momentum. When the main accretion phase is nearly over, the YSO with an optically thick accretion disk, known as a classical T Tauri star (CTTS), continues to contract along the Hayashi track with roughly constant effective temperature and decreasing luminosity. At a later stage when the accretion disk is optically thin or gone, the YSO becomes a weak-lined T Tauri star (WTTS), which consequently moves on to the Main-Sequence (MS). Due to their proximity and rich samples, several nearby low-mass star forming regions (LMSFRs) have been well studied in multiwavelength, including Chamaeleon, Taurus-Auriga, and ρ Ophiuchius. Some observational highlights include, the Hubble Space Telescope (HST) direct imaging of circumstellar disks and envelopes around a number of YSOs in the Taurus-Auriga molecular cloud (Padgett et al. 1999), and the detection of Herbig-Haro outflows (see review by Reipurth & Bally 2001). These observations provide strong support to the Nebular Hypothesis. It is not exaggerating that the foundations for the current picture of low mass star formation and evolution are based on the extensive studies of the nearby regions. Vast amount of literature exists on the LMSFRs. Lada et al. (1993) gives a review of clouds and star formation. Based on studies of Taurus, Ophiuchus, and Orion, they suggest that there are at least two modes or environments of star formation: isolated and clustered; the clustered mode of star formation may be the dominant mode of star formation in our Galaxy. Zinnecker et al. (1993) provides a thorough review on the major star forming regions and the theories for the origin of stellar masses and the initial mass function (IMF). Low mass stars in OB associations are reviewed in Walter et al. (2000) and Briceño et al. (2007). Using recent groundbased and spacebased infrared surveys, they find the spatial distribution of low- mass stars in OB associations contain significant substructure and the disk dissipation timescales are very different for stars of a given mass in OB associations. Previous X-ray studies of nearby LMSFRs with Einstein, ROSAT, and ASCA are reviewed comprehensively by Feigelson & Montmerle (1999). An important point is raised: the X-ray surveys effectively improve our samples of young stars (particularly the weak-lined T Tauri stars). Other excellent reviews on high energy processes in YSOs include Favata & Micela (2003), Güdel (2004). The latest review of Güdel et al. (2007) discuss the entire young stellar population and the X-ray astrophysics revealed by a recent large multiwavelength survey of Taurus Molecular cloud, namely the XMM- Newton Extended Survey of the Taurus Molecular Cloud (XEST). One of the closest LMSFRs to the Sun is the ρ Ophiuchius cloud ( 145 pc), and ∼ the embedded cluster is also among the youngest in the nearby star forming regions. The stellar content in the cluster were used for definition of Class 0, I, II, III sources for low-mass YSOs based on the SEDs (Andre & Montmerle 1994; Greene et al. 1994). Chandra observations of this cloud reported X-ray detections from brown dwarfs, Class I sources, and protostars in a younger phase than Class I (Imanishi et al. 2001). The Taurus-Auriga complex has long been intensively studied where many T Tauri stars were found (e.g., Kenyon & Hartmann 1995). The cloud has filamentary appearance and star formation is ongoing inside some of the filaments. X-ray surveys 4 are very useful in identifying new WTTS that are largely missed by longer wavelength surveys (e.g., Feigelson et al. 1987; Walter et al. 1988; Neuhaeuser et al. 1995). The L1551 cloud region was observed with Chandra and XMM-Newton (Bally et al. 2003; Favata et al. 2003). The latest XEST survey covers 5 square degrees of Taurus region ∼ to study the variability and evolution of X-ray emitting plasma and to search for new members of the association. The results have been reported in a series of papers (see Guedel et al. 2007 for a review). From the accumulated X-ray observations in these LMSFRs, some of the X-ray properties of low-mass YSOs are well characterized. For example, the typical X-ray 31 1 luminosity is L < 10 erg s− with a thermal plasma spectrum of kT 1 5 keV. An x ∼ − empirical relation is found between the X-ray luminosity and the bolometric luminosity, 3 5 L /L 10− 10− . X-ray flaring is commonly observed among low-mass YSOs. The x bol ∼ − X-ray emission properties are similar for CTTS and WTTS.

1.1.2 Massive Star Forming Regions

Massive stars are those with initial masses 8M⊙ and eventually will end their ≥ lives in supernovae explosions and become black holes or neutron stars. They are very luminous and characterized by powerful stellar winds, high mass loss rate, and strong ultraviolet (UV) radiation. Undoubtedly the winds from the massive stars and the supernovae explosion affect the chemistry and physical conditions of the natal cloud and the interstellar media (ISM) by both returning processed materials and kinematically create cavities, therefore contribute to the evolution of the galaxies. New generation of low-mass stars are also triggered to form by the presence of massive stars. There are accumulating evidence from short-lived radionucleides in meteorites implying that our Sun formed near a massive star (Hester & Desch 2005a; Tachibana et al. 2006). Williams & Gaidos (2007) evaluated the likelihood of supernova enrichment of the protoplanetary disks and found that the solar system most likely formed in one of the richest clusters. The key to understand the origin of the solar system now requires more knowledge of the formation and evolution of massive stars themselves and how their low mass siblings are affected in the wind and UV radiation dominated environment. Massive O stars typically are not isolated (except the so-called run-away O stars, e.g., Blaauw 1961; Gies 1987; Ma´ız-Apellániz et al. 2004); they live in dense OB clusters or unbound OB associations. Thus we cannot fully understand massive star formation without studying the stellar clusters in the MSFRs. MSFRs represent “mini starburst galaxies”, where rich young stellar clusters (YSCs) are formed in massive cores in the GMCs. Lada & Lada (2003) give a comprehensive review on the embedded young clusters and present the first extensive catalog of embedded clusters in Galactic molecular clouds. A large fraction of all stars in the Galactic disk form in these regions. The YSCs are ideal laboratories to study star formation and stellar evolution (Clarke et al. 2000; Allen et al. 2007). Because they are unevolved massive systems with many young stars, the YSCs allow us to measure the IMFs reliably, examine their radial variations, and test whether the IMFs are universal. Unlike the low-mass stars, the story of massive star formation is much more complicated and inconclusive (see reviews by Stahler et al. 2000; Bally & Zinnecker 2005; 5

Beuther et al. 2007; McKee & Ostriker 2007; Zinnecker & Yorke 2007). Massive stars are rare compared to the large number of low-mass stars. They are more distant, heavily obscured and short-lived. The formation process is rather quick, and the mechanism remains highly debated given the observational challenges to constrain the theories. Observations in mid-IR and radio have established a few class of objects in the embedded phase of massive star formation, representing possibly different evolutionary stages. A group of newly identified cold dense molecular cores known as the infrared dark clouds (IRDCs) probably represents the earliest phase of formation of massive stars and massive star clusters (Perault et al. 1996; Egan et al. 1998; Rathborne et al. 2006). The density and temperature of the massive cores match the expected initial conditions of massive star formation. The “hot cores” were known as the earliest stage in the massive star formation process, but probably later than the IRDC phase. The gas in hot molecular cores is still dense but warmer, typically observable in methanol maser emission. A detailed review on the hot molecular cores and the earliest phases of massive star formation is presented in (Kurtz et al. 2000). When ionized gas is present and highly confined to the massive protostar, it is known as a hypercompact HII region and ultracompact HII (UCHII) region (Wood & Churchwell 1989; Kurtz & Hofner 2005; Churchwell 2002; Hoare et al. 2007). When the gas is largely ionized and the HII expands to disrupt the natal molecular cloud, it is known as the compact and classical HII region (Mezger et al. 1967). There are two major competing theories of massive star formation, which are thoroughly reviewed in Bally & Zinnecker (2005); Zinnecker & Yorke (2007). In analogue to low-mass stars, some researchers believe that massive stars form in a scaled-up version of low-mass star formation inside the hot cores (e.g., McKee & Tan 2002; Yorke 2004). This will require very high accretion rates (see work by Wolfire & Cassinelli 1987; Stahler et al. 2000). There was concern that the intense radiation field from the massive stars can significantly suppress the gas accretional growth. However this problem can be solved by accretion from the disk onto the star, instead of direct infall from envelope (Yorke & Sonnhalter 2002). McKee & Tan (2002) also shows that the accretion flow continues despite the presence of high radiation pressure in high-density cloud cores. The other competing theory is proposed by Bonnell et al. (1997, 2007), suggesting that massive stars form through accreting gas competitively against other stars within a clustered environment and through merging in dense protostellar clusters. The most massive star forms under the most favorable conditions. Both theories have some observational grounds (Zinnecker & Yorke 2007). Further observational constraints are needed to rule out any of them, and apparently part of this effort must be devoted to MSFRs.

1.1.3 X-ray Observations of Massive Star Forming Regions Traditional optical studies on MSFRs suffer tremendously from extinction. In some distant obscured regions, only the most massive members are known in optical. NIR and mid-IR observations are much less vulnerable to dust extinction, but the con- taminations from older Galactic populations and background galaxies are overwhelming. When studying the IMFs in these regions, identifications of individual members are challenging, so the measurements are typically obtained statistically with controlled 6 background subtraction. In the bright nebulosity around the photodissociation region created by soft UV radiation or in the vicinity of bright stars, source finding become extremely challenging for optical and IR observations. In the X-ray band, observations with the early generation X-ray satellites, such as the ROSAT All Sky Survey (RASS), made great contribution to our understanding of star forming regions. However, they lack of two critical capabilities, that is, the arcsec level spatial resolution to resolve the cluster members in the typically crowded and distant star forming regions and the sensitivity to hard band X-rays to penetrate deep in the obscuring star forming clouds (see examples in Feigelson & Montmerle 1999). The advent of the two new generation X-ray telescopes, namely Chandra and XMM-Newton, opens a new era for the study of star forming regions, especially the distant MSFRs that contain massive clusters. Feigelson et al. (2007) provides the latest review of the X-ray properties of young stars and young stellar clusters based on recent Chandra and XMM-Newton surveys, including some MSFRs. Other reviews on Chandra observations of MSFRs are Townsley et al. (2003, 2005, 2006a), but with the emphasis on diffuse emission. The increasing number of publications on X-ray observations come out recently indicates this is a booming research field. Lists of related publications are provided in Townsley (2006) and Tsujimoto et al. (2007). Other related recent papers include but not limited to NGC 2024 (Ezoe et al. 2006a), NGC 2362 (Damiani et al. 2006), Mon R2, W3, and NGC 6334 (Persi & Rosa Marenzi 2006), Cyg OB2 (Albacete Colombo et al. 2007; Yanagida et al. 2007), Serpens (Winston et al. 2007), and more observations are underway. X-ray images of these regions show astonishingly rich stellar content and complex structure: low-mass PMS stars around massive stars, many of which are previously unknown; a few regions show diffuse X-ray emission, either from stellar winds or supernovae remnants. When combined with long wavelength surveys, these studies provide important new insights into star and cluster formation. As the prototypal stellar nursery, the well-known Orion Nebula Cluster is the nearest rich cluster with massive stars and has been studied intensively in multiwavelength. One of the highlight in X-ray studies of ONC is the Chandra Orion Ultradeep Project (COUP) known for its unprecedented depth, which uses the Chandra X-ray Ob- servatory to point at the Orion Nebula region nearly continuously for 2 weeks (Getman et al. 2005a; see Feigelson et al. 2007 review; Figure 1.2). COUP investigates the X-ray luminosity (XLF) of the Orion Nebula Cluster (Feigelson et al. 2005), X-ray flaring in the PMS stars (Wolk et al. 2005; Favata et al. 2005), and the dependence of X-ray properties on the stellar properties such as mass and age (Preibisch & Feigelson 2005; Preibisch et al. 2005). COUP also nicely demonstrates how the X-ray detections correspond to the optical and infrared stars, with little contaminants. Although still in the early stages, the emerging X-ray studies of MSFRs have provided abundant new information about X-ray properties of massive stars. The typical 32 33 1 X-ray luminosity for an O star is L 10 10 erg s− . Some O stars exhibit soft x ∼ − thermal plasma spectrum of kT 0.5 keV, consistent with an origin from microshocks ∼ in O star winds (e.g., Rauw et al. 2002, Rho et al. 2004, Skinner et al. 2005, Sana et al. 2006, Albacete-Colombo et al. 2007). Anomalously hard spectra are also seen in some early O stars, indicating possible binarity with presence of colliding-winds. An empirical 7 relation L /L 10− exists for OB stars. However large deviations from this relation x bol ∼ 7 are also noticed among B-type stars (Stelzer et al. 2005; Sana et al. 2006; Wang et al. 2007a; Broos et al. 2007; Wang et al. 2007b). In the next section I first summarize advantages of X-ray observations of the MSFRs, then demonstrate how Chandra observations can help address some outstanding science issues in understanding MSFRs that are observationally challenging to investigate in the optical and IR wavelengths, which motivates this thesis.

1.2 Motivations for this Thesis

1.2.1 Advantages of Chandra Observations of the MSFRs X-ray photons have low extinction cross sections in the ISM, therefore X-ray observations are much less vulnerable to dust extinction than the optical observations. Chandra observations routinely penetrate AV > 100 mag of visual extinction. The deep COUP studies show that Chandra images can overcome up to A 500 mag of V ≃ extinction into the cloud (Grosso et al. 2005). As shown by the COUP observation (see Feigelson et al. 2005 for the global X- ray properties of ONC), the XLF of a young stellar cluster spans 4 orders of magnitude (log L 28 32 erg/s). About half of this spread is attributed to a poorly-understood x ∼ − link to stellar mass (Preibisch & Feigelson 2005). Thus a flux-limited X-ray survey of a star forming region approximately gives a complete sample of PMS stars down to a corresponding mass limit. Powerful X-ray flares are present equally in stars through the PMS phases; no evidence is found for a X-ray-quiet, non-flaring PMS population (Preibisch et al. 2005). These flares thus reveal heavily obscured, as well as lightly obscured, low-mass cloud populations at all PMS phases. Chandra observations effectively discriminate all PMS phases from older contaminating Galactic field stars. COUP study shows the general decay of the XLF as stars age using complete, mass-stratified samples (Preibisch & Feigelson 2005). PMS stars are, on average, 3 orders of magnitude more X-ray-luminous than random 1 10 Gyr-old − disk stars. X-ray surveys are thus remarkably effective in eliminating foreground and background stars from the cluster sample. The main contaminant (usually < 10%) is quasars, most of which are faint in optical and IR. This allows us to identify the possible quasars among the detections. X-ray observations are also capable of detecting low-mass companions around high-mass stars and are not hampered by the bright nebulosity seen in optical/IR. The X-ray luminosity of a low mass PMS star is only 100 times fainter than that of ∼6 a massive O star on average, in contrast to the 10 difference between their optical luminosities (Cox 2000). Also the extended diffuse X-ray emission in MSFRs are mostly faint compared to the X-ray emission from stars (e.g., Townsley et al. 2003; Ezoe et al. 2006b), therefore the identification of point sources are still reliable in the presence of diffuse emission. 8

1.2.2 Some Outstanding Issues to Be Addressed in Understanding MSFRs in the Chandra Era Chandra X-ray Observatory and the Advanced CCD Imaging Spectrometer (ACIS) onboard offer superior spatial resolution, sensitivity, and broad energy bandpass, while simultaneously provide coordinates, lightcurves, and spectra of the X-ray detections (Weisskopf et al. 2000; Garmire et al. 2003). Given the difficulties presented in optical and IR studies of MSFRs, these characteristics make a Chandra/ACIS study of MSFRs unique and important in many aspects. Census of the Stellar Populations • Many distant and obscured MSFRs are poorly studied in the optical and IR. Especially, the lower mass populations are rarely identified. The observational difficulties caused by dust extinction in the optical and by heavy contamination from unrelated old stars and galaxies in the IR severely hurt the science yield from direct imaging. To confirm young ages of such a large number of stars, multi-object spectroscopy is needed but considerably time-consuming. Further studies of the massive young clusters require a good census of their stellar populations. X-ray surveys can effectively improve the census of the young stellar populations in the rich massive clusters. With low fraction of contamination from non-members and the ability to detect obscured sources, an X-ray survey of a star forming region approximately gives a complete and clean sample of PMS stars which is nearly complete to a mass limit corresponding to the depth of the observation. Relatively short Chandra exposures of close (< 1 kpc), compact ( 1 pc), young (1 3 Myr) clusters are sufficient ∼ − to obtain a complete unbiased sample of cluster members down to a sub-solar mass range (Getman et al. 2006). For more distant regions, the sample of cluster members is complete to a comparable mass limit with deeper observations. A 94 ks Chandra ∼ observation of massive star forming complex Rosette (d 1.4 kpc) is roughly complete ∼ to 0.5 M⊙ (Wang et al. 2007b). Based on the X-ray and IR photometry, large number of uncatalogued low-mass stars can be identified with superior astrometry. A number of candidate OB stars and intermediate-mass stars can be identified in the obscured regions, with positions ready for optical and NIR followup to measure their spectral types and ages. Structures in Young Clusters and Modes of Star Formation • The morphology of a YSC provides clues for cluster formation and dynamical evolution. Morphological studies based on optical or infrared samples are complicated by patchy extinction, nebular contamination, confusion with field stars, and bias towards stars retaining protoplanetary disks. High spatial resolution Chandra observations provide positions of the cluster members with sub-arcsecond precision, again with little contamination and no bias towards disky stars. The spatial distributions of these X-ray identified stars provide excellent ways to explore the different modes of star formation and dynamical evolution. If there is primordial structure or a complicated star formation history, it is reflected as substructures in the large-scale cluster structure. For a largely spherical cluster, radial density profiles can be derived and used to diagnose whether the cluster has reached dynamical 9 equilibrium. Clustering and subclustering can be revealed by smoothed source distributions (stellar surface density maps). If the cluster has patchy absorption that significantly affects the X-ray sensitivities, absorption-stratified distributions can be used to study the unobscured and obscured populations separately. With the X-ray identified young stars, we are able to examine the fraction of stars that are formed in clusters and in a distributed way, and find clues for which mode of star formation (distributed vs. clustered) is dominant in the GMCs. The large-scale distribution of star clusters also allows investigation of cluster formation in a sequential manner (Elmegreen & Lada 1977) as well as triggering at the peripheries of the HII region. Mass Segregation • In some clusters, the massive stars are concentrated near the cluster center while the low mass stars are more radially dispersed. This is known as mass segregation, and is observed in well-known clusters such as the Orion (Hillenbrand 1997) and the NGC 3603 (Sung & Bessell 2004). With the good census of the cluster members, the spatial distribution determined by the Chandra detections, and the stellar mass information derived from their ONIR counterparts, mass segregation can be investigated in the YSCs. By estimating the relaxation time of the YSC and comparing it with the age of the observed cluster, primordial mass segregation and dynamical mass segregation may be distinguished. In this way some constraints can be placed on the cluster formation process. The IMF and the Total Population • The upper IMF is found to be a power law with a common slope dN/d(log m) ∼ 1.35 observed in many clusters, known as the Salpeter slope (Massey 1998). Possible − exceptions are the cluster formed close to the Galactic center (Scalo 1998). Whether the IMF is universal for massive stars in OB clusters and OB associations has been extensively discussed. The competitive accretion theory actually predicts a steep power law IMF slope similar to the currently observed universal value in young clusters (Bonnell et al. 2007). If the formation mechanisms for stars of different masses are indeed different, a break in the IMF slope is expected at a mass in the intermediate mass to high mass range (Zinnecker & Yorke 2007). There is no report of such feature seen in the upper IMF. Chandra observations can contribute to the IMF study as well. The XLF (which is directly measured) can be considered to be the convolution of the IMF (which is unknown) and the X-ray–Mass Luminosity (L M) correlation (which is measured in x − both COUP and XEST studies; Preibisch et al. 2005, Telleschi et al. 2007). Using the best-studied Orion Nebula Cluster XLF (COUP XLF) and IMF as a calibrator, the XLF of a YSC can be used to probe the IMF of this stellar cluster and to estimate the total X-ray emitting population (e.g., Getman et al. 2006). In the IR band, this is equivalent to the traditional K-band luminosity function (KLF) method to study populous clusters (e.g., Lada & Lada 1995). However, while the KLF is derived statistically by star counting and subtracting a background population from the observed total population, the XLF is based on identifications of individual cluster members. The Prevalence of Disks in the MSFRs • 10

Most stars are born in wind-swept and radiation-dominated OB clusters, which are in great contrast to relatively isolated environments like the nearby Taurus-Auriga molecular cloud (Walter et al. 2000; Lada & Lada 2003). The latest evidence that the solar system formed near a massive star (Hester et al. 2004) requires a richer understanding of the variety of disk structures and timescales of disk dissipation in MSFRs. Besides their own evolution from accreting onto the central stars, disks of low mass stars can be photoevaporated by the intense ultraviolet (UV) light from nearby OB stars (Hollen- bach et al. 2000). Examples of this effect are seen as the proplyds in the Orion Nebula (O’dell et al. 1993). Apparently the evolution of these disks, and planet formation, is largely constrained by the disk mass loss due to external irradiation. A systematic study focusing on protoplanetary disk evolution in the wind and radiation-dominated MSFRs is needed. While dusty protoplanetary disks are intensively observed in IR, most of the targeted samples suffer from a critical limitation: they lack a complete, unbiased representation by the diskless stars that cannot be identified with IR-color. This is effectively done by X-ray selection. A key advantage of X-ray surveys is that the X-ray sampled young stars are not biased towards PMS stars with disks: they trace magnetic activity rather than photospheric or disk blackbody emission (Preibisch et al. 2005). Diffuse X-ray Emission • Last but not least, with sufficient resolving power to separate point sources from true diffuse emission and the ability to estimate the contribution from unresolved faint sources from the XLF, the level of true diffuse X-ray emission from thermalization of fast O-star winds in the MSFRs can be carefully evaluated with Chandra observations. As a result of joint effort between Chandra/ACIS GTO and GO observations at Penn State, significant amount of data on MSFRs are available and advanced data reduction tools have been developed to handle these complex data sets. I am equipped to investigate the X-ray properties of the large sample of massive stars and the vastly uncatalogued accompanying low mass PMS stars, greatly improve the census of cluster members, probe the XLFs and the IMFs, structure and dynamics in the young clusters, and measure fraction of young stars that possess IR-excess disks in a disk-unbiased X- ray-selected samples. All these questions were difficult to tackle before the Chandra era.

1.3 Overview of this Thesis

In general, this thesis portraits two massive star forming regions in great details, namely the NGC 6357 complex and the Rosette complex. Chapter 2 brieﬂy describes the observatory and instrument, and introduce some methods used in the data reduction and analysis. The subsequent science chapters have been adapted from standalone published papers and drafts written for future publication; related publications will be noted in footnotes. 11

1.3.1 Chapter 3: An X-ray Census of Young Stars in the Massive Southern Star-Forming Complex NGC 6357 In this chapter we present the first high spatial resolution X-ray study of the 2 massive star forming region NGC 6357, obtained in a 38 ks Chandra/ACIS observation . Inside the brightest constituent of this large HII region complex is the massive open cluster Pismis 24. It contains two of the brightest and bluest stars known, yet remains poorly studied; only a handful of optically bright stellar members have been identified. We investigate the cluster extent and Initial Mass Function and detect 800 X-ray 30 −1 ∼ sources with a limiting sensitivity of 10 ergs s ; this provides the first reliable ∼ probe of the rich intermediate-mass and low-mass population of this massive cluster, increasing the number of known members from optical study by a factor of 40. The −1 ∼ high luminosity end (log L [2-8 keV] 30.3 ergs s ) of the observed X-ray luminosity h ≥ function in NGC 6357 is clearly consistent with a power law relation as seen in the Orion Nebula Cluster and Cepheus B, yielding the first estimate of NGC 6357’s total cluster population, a few times the known Orion population. We investigate the structure of the cluster, finding small-scale substructures superposed on a spherical cluster with 6 pc extent, and discuss its relationship to the nebular morphology. The long-standing 7 L 10− L correlation for O stars is confirmed. The X-ray spectra of two O3 stars X − bol show soft component consistent with the wind-shock theory, although the best fit to Pis 24-1 spectrum requires a hard X-ray emission component. Twenty-four candidate O stars and one possible new obscured massive YSO or Wolf-Rayet star are presented. Many cluster members are estimated to be intermediate-mass stars from available infrared photometry (assuming an age of 1 Myr), but only a few exhibit K-band excess. We ∼ report the first detection of X-ray emission from an Evaporating Gaseous Globule at the tip of a molecular pillar; this source is likely a B0-B2 protostar.

1.3.2 Chapter 4: A Chandra Study of the Young Open Cluster NGC 2244 in the Rosette Nebula In Chapter 4, we present the ﬁrst high spatial resolution X-ray study of the stellar population of the 2 Myr old NGC 2244 cluster immersed in the Rosette Nebula using the 3 Chandra X-ray Observatory . Over 900 X-ray sources are detected; 77% have optical or FLAMINGOS NIR stellar counterparts that are mostly previously uncatalogued young stellar cluster members. All known OB stars with spectral type earlier than B1 are detected, and the X-ray selected stellar population is estimated to be nearly complete −1 between 0.5 and 3 M⊙. The XLF ranges from 29.4 < log Lx < 32.0 ergs s in the hard (2 8 keV) band. By comparing the NGC 2244 and Orion Nebula Cluster XLFs, we − estimate a total population of 2000 stars in NGC 2244. A number of further results ∼ emerge from our analysis: The XLF and the associated K-band luminosity function

2The work in this Chapter has been previously published as Wang, J., Townsley, L. K., Feigelson, E. D., Getman, K. V., Broos, P. S., Garmire, G. P., and Tsujimoto, M., 2007, the Astrophysical Journal Supplement Series, 168, 100. 3The work in this Chapter has been accepted as Wang, J., Townsley, L. K., Feigelson, E. D., Getman, K. V., Broos, P. S., Roman-Zuniga C., and Lada, E., 2007 for publication in the Astrophysical Journal. 12 indicate a normal Salpeter IMF for NGC 2244. This is inconsistent with the top-heavy IMF reported from earlier optical studies which lacked a good census of < 4M⊙ stars. The spatial distribution of X-ray stars is strongly concentrated around the central O5 star, HD 46150. The O4 star HD 46223 has few companions. The cluster stellar radial density profile shows two distinctive structures: a power-law cusp around HD 46150 out to 0.7 pc surrounded by an isothermal sphere extending out to 4 pc with core ∼ radius 1.2 pc. This double structure, combined with the absence of mass segregation, indicates that this 2 Myr old cluster is not in dynamical equilibrium. Our results will strongly constrain models of the cluster formation process. The spatial distribution of X-ray selected K-excess disk stars and embedded stars is asymmetric with an apparent deficit towards the north. The fraction of X-ray-selected cluster members with K-band excesses due to inner protoplanetary disks is 6%, slightly lower than the 10% disk fraction estimated from the FLAMINGOS study based on the NIR-selected sample. This is due to the high efficiency of X-ray surveys in locating disk-free weak-lined T Tauri stars. X- ray luminosities for 24 stars earlier than B4 confirm the long-standing log(L /L ) 7 x bol ∼− relation. The Rosette OB X-ray spectra, unlike other samples, are soft and consistent with the standard model of small-scale shocks in the inner wind. About 50 intermediate- mass (2

1.3.3 Chapter 5: A Chandra Study of the Stellar Populations in the Rosette Molecular Cloud Chapter 5 continues our study in the Rosette complex to explore the young stellar populations in the Rosette Molecular Cloud (RMC) region, mostly weak-lined T Tauri stars that are effectively identified by X-rays. This is achieved by analyzing high spatial resolution Chandra/ACIS images of this region. A total of 395 X-ray point sources are detected, 299 of which (76%) have an optical or NIR counterpart identified from deep FLAMINGOS images. Based on smoothed stellar surface density maps, we investigate the spatial distribution of the X-ray sources and define three distinctive structures and substructures within them. Structures B and C are associated with previously known embedded IR clusters (Phelps & Lada 1997 clusters 2 and 4), while structure A is a new X-ray-identified unobscured cluster. NIR color-color and color-magnitude diagrams for these X-ray sampled clusters are constructed. Three Class I sources are found within region B. We compare the distribution of Class I, II, and III stars with the morphology of molecular content as outlined by the CO emission and find that the Class II/I sources are more confined to the midplane of the cloud. The X-ray properties of a few interesting X-ray sources are presented, along with their IR characteristics. Our X-ray-defined regions confirm the structures reported in the NIR study by Román-Zúñiga et al. (2007a). Cluster A, similar to NGC 2237 (Chapter 6), represents the early epoch of star formation when NGC 2244 formed. The shock front from the Rosette Nebula may have triggered the collapse of pre-existing clumps inside the molecular cloud and form PL4, although 13 spontaneous formation due to large scale cloud turbulence is also possible for the formation of PL4 and other clusters further from the nebula. The PL2 cluster is young, and may represent triggered star formation at the irradiated edge of the cloud through the triggering process known as “collect-and-collapse”. The data are consistent with speculations that triggered star formation is present in the RMC but not in the original sequential triggering proposed by Elmegreen & Lada (1977). The concentration of X-ray identified young stars implies that .35% of stars could be in a distributed population throughout the RMC region and clustered star formation is the dominant mode in this region.

1.3.4 Chapter 6: A Chandra Study of the Triggered Cluster NGC 2237 We investigate yet another star forming region in the Rosette complex to identify the previously unknown stellar population and find clues of triggered star formation here. We present the first high spatial resolution X-ray images of the NGC 2237 cluster in the Rosette Nebula obtained via a 20 ks Chandra observation. 168 X-ray point sources are detected with a limiting X-ray sensitivity nearly complete to the solar mass range. 80% of the X-ray sources have matched optical and near infrared counterparts from USNO, 2MASS and deep FLAMINGOS images. We estimate at most 10% of the X-ray emitting population are extragalactic and galactic contamination. Our X-ray sample provides the first probe of the low mass population in this recently discovered satellite cluster to the massive OB cluster NGC 2244. The locations of most ACIS sources in the color-magnitude diagram are consistent with a small population of PMS low mass stars (M . 2M⊙) in the NGC 2237 region with a visual extinction of 1 . AV . 2 at 1.4 kpc. We derive an overall K-excess disk frequency of 13% for stars with masses M & 1M⊙ using the X-ray selected sample, consistent with the reported disk fraction from NIR study. Combining the spatial distribution and properties of the NIR counterparts to our Chandra stars do not provide a simple and coherent star formation picture of this region. The NGC 2237 cluster, similar to RMC A, may have formed from collapse of pre-existing massive molecular clumps accompanying the formation of the NGC 2244 cluster. For the X-ray stars to the northwest of NGC 2237, there is only tentative evidence suggestive of a triggered formation process in the optical dark pillars by the NGC 2244 O stars. Spitzer mid-IR observations and spectroscopy are needed to determine the age and evolution status of the embedded stars in the pillar region. Finally in Chapter 7, we summarize our findings in this thesis. Based on these new results from Chandra observations of the MSFRs, we attempt to draw some conclusions and speculate on some of the important issues in massive star and cluster formation. Substantial amount of exciting future work in studies of massive star formation and MSFRs is outlined. 14

Fig. 1.1 The evolution scenario of a YSO with the classiﬁcations and the spectral energy distributions shown. Adapted from Smith (2004). 15

Fig. 1.2 A comparison between the NIR image and the X-ray image of the central region in the Orion Nebula Cluster. The Trapezium stars are seen near the center of the images. The left panel is the NIR image taken with VLT/ISSAC: K-band image is shown here in red, the H-band image in green, and the J-band image in blue. The right panel is the deep X-ray image taken with Chandra/ACIS. Red represents soft X-ray emission (0.5-2 keV), while blue represents hard X-ray emission (2-8 keV). Both panels have 5.′5 5.′5 ﬁeld of × view. Credit: ESO/VLT/M.McCaughrean et al./NASA/CXC/Penn State/E.Feigelson et al. 16

Chapter 2

Methods

In this chapter, we briefly review the telescope and instrument used for the observations in this thesis. The observations are described individually in each Chapter. We also describe in details some procedures adopted and tools developed to facilitate the data reduction and analysis. Because of the different physical conditions in each regions, these methods and the region-specific numbers will be described in the subsequent individual chapters.

2.1 Telescope and Instrument

The Chandra X-ray Observatory was launched on July 23, 1999. The orbit is an elliptical, with the perigee and the apogee distance of 10000 km and 140161 km after its launch, respectively. A schematic view of the Chandra Observatory is shown in Figure 2.1. The key elements of the observatory consist of the High Resolution Mirror Assembly (HRMA), the Pointing Control and Aspect Determination System (PCAD), the focal plan science instruments which include the Advanced CCD Imaging Spectrom- eter (ACIS) and the High Resolution Camera (HRC), and the objective transmission gratings which include the High Energy Transmission Grating (HETG) and the Low Energy Transmission Grating (LETG). Details of Chandra are described in the Chandra Proposer’s Observatory Guide (http://cxc.harvard.edu/proposer/POG/). We introduce the basics of HRMA and ACIS, since these are the instruments that oﬀer the superior spatial resolution and high sensitivity X-ray images, which moves our sciences in star forming regions forward. There are 4 pairs of concentric thin-walled mirrors in the HRMA. These are designed as X-ray grazing-incidence Wolter Type-I mirrors of 10m focal length, with ∼ paraboloid mirrors in the front and hyperboloid mirrors in the back. These mirrors are the most precisely shaped and aligned with the smoothest surface ever constructed in the X-ray history. As a result, Chandra has the ability to obtain unprecedently sharp images. The spatial distribution of the X-ray photons at a nominal energy on the detector after the optics is known as the point spread function (PSF) for that given energy. The encircled energy radius of the PSF gives the sharpness of the focused image. At 1.496 keV, the encircled energy of the on-axis PSF contains 80% of the incident photons inside a circle of 0.5′′ radius. For comparison, to achieve the same fraction ∼ of the incident photons, the ROSAT/HRC and the modern European satellite XMM- Newton both require a encircled energy radius of 6′′. In other words, Chandra’s ∼ angular resolution is over 10 times better than that of the ROSAT and XMM-Newton. For reader’s reference, the recently launched Japanese X-ray satellite Suzaku has an angular resolution of 2′. 17

ACIS contains ten 1024 1024 pixel X-ray CCDs (Figure 2.2). Four CCDs in a × 2 2 array comprise the ACIS-I that is used for imaging. Another six chips arranged in × a 1 6 array form the ACIS-S used either for imaging or readout for the gratings. The × ′ ′ ′ ′ field of view for ACIS-I is 16. 9 16. 9, and 8. 3 50. 6 for ACIS-S. Two CCDs (ACIS- × × S1 and ACIS-S3) are back-illuminated (BI) and the rest are front-illuminated (FI). FI CCDs are “conventional,” where photons illuminate the frontside and the gates above the semiconductor layers reduce the sensitivity to soft X-rays. BI CCDs are backside- illuminated (“flipped” with respect to the direction of incoming photons), therefore the path of X-ray photons is not blocked by the gates. The ACIS CCDs also work as a spectrometer with a moderate resolution R = E/∆E 10 50. The pre-launch calibrated energy resolutions with FI CCDs almost ∼ − achieved the theoretical values over the working energy range. However, after the orbital activation the CCDs were unfortunately damaged by low energy (100-200 keV) protons encountered when passing the radiation belt. The increased charge transfer inefficiency (CTI) caused the energy resolution of the FI CCDs substantially degraded. This problem is treated by lowering the CCDs’ operating temperature to 120◦C and introducing cus- − tomized detector response functions (Townsley et al. 2002). The Principal Investigator for the ACIS instrument is Prof. Gordon Garmire.

2.2 Basic ACIS Data Reduction

The data reduction procedures described in Townsley et al. (2003) as developed by the Penn State ACIS team are closely followed. The Level 2 files distributed by the Chandra X-ray Center (CXC) are highly processed and some useful information are not available with the generic pipeline reduction. To obtain an improved Level 2 event list, data reduction starts with filtering the Level 1 event list (“evt1” file) processed by the CXC pipeline. First, parameter files like the observation bad pixels list and the PCAD aspect file are properly linked. In order to improve absolute astrometry, preliminary X-ray positions of ACIS-I sources are obtained by running the wavdetect wavelet-based source detection algorithm (Freeman et al. 2002) within the Chandra Interactive Analysis of Observations (CIAO) package on the original Level 2 event list, using only the central 8′ 8′ of the field. The resulting X-ray sources are matched to the 2MASS point source × catalog. The mean position offsets between X-ray sources and their NIR counterparts are then calculated. The astrometry is improved by applying the offsets in right ascension (RA) and declination (Dec) to the X-ray coordinates. This is achieved by reprojecting the event data after adjusting the aspect file and the nominal pointing coordinates. Software randomization is also taken out at this point. The data is examined skeptically for processing error using CIAO command dm- list. The number of flight grades events is counted, and this number is expected to be few since they should have been removed during the in-flight processing. For data taken in “Timed Events Very Faint” mode (5 5 pixel event island instead of just a 3 3 event × 1 × island), a cleaning algorithm developed by Vikhlinin is implemented in CIAO command

1See http://hea-www.harvard.edu/∼alexey/vf bg/vfbg.html 18 acis process event, which can use the outer pixels of the 5 5 event island to help identify × bad events from cosmic rays and further reduce background. An acis process event call updates the status bit for events flagged by this routine. The CCD charge transfer inefficiency (CTI) is corrected using the Penn State CTI corrector developed by Townsley et al. (2002). CTI-corrected ACIS Quantum Efficiency Uniformity (QEU) files must be used after this step for each CCD that is corrected for CTI. The next step is to filter the events with different filters. The data is cleaned with standard grade filter (grade=0, 2, 3, 4, 6). Our customized procedure choose to apply only the non-controversial status flags (ignore status bits 16-19 and 23). Then the data is filtered with Good Time Intervals (GTI) tables provided in the CXC distributed data. Glitches in the aspect solution (most of which are already included in the CXC GTI tables) are searched to be excluded. CCD background flares can be present because of space weather. We examine the lightcurves generated with CIAO command dmextract and lightcurve. The time intervals containing flares are evaluated based on count rates. GTI tables are then modified accordingly. Any remaining hot columns and hot pixels that are not included in the CXC bad pixel list are searched and removed. Finally an energy filter is applied to remove the high energy background events. After all these steps to create Level 2 event files, the flaring pixels in the data are flagged but not removed, because significant number of good events can be removed by applying the flaring pixel filter. Our procedure applies this filter and an energy filter to minimize background events when images for source searching are created. For spectral analysis, the energy filter is implicitly applied by the fitting software package. For further reference, Appendix B of Townsley et al. (2003) and Getman et al. (2005a) describe the Penn State ACIS data reduction procedure in details.

2.3 Data Analysis

2.3.1 ACIS Extract Our data analysis procedures heavily use an IDL software package ACIS Extract (AE), written by P. Broos et al. at Penn State (Broos et al. 2002). It is a versatile tool that works with Level 2 event files that allow users to extract point sources, derive source photometry in different energy bands, perform timing analysis, model the spectra of large numbers of point and diffuse emission observed with the ACIS instrument. It can even handle multiple observations that may not have the same aim point or roll 2 angle. Details of AE can be found in the ACIS Extract User’s Guide .

2.3.2 Source Finding A multifaceted source ﬁnding procedure developed by the Penn State ACIS team (Townsley et al. 2003; Getman et al. 2005a, 2006) is summarized here, which is designed to locate all potential sources, even in the presence of crowding or a broad PSF, with the

2http://www.astro.psu.edu/xray/docs/TARA/ae users guide/ 19 recognition that features due to detector noise may also enter the source list. First, we assemble a large number of candidate sources using a variety of techniques and criteria (wavdetect, reconstruction, visual selection, etc., as described below), which includes a number of possible false detections. These spurious sources are identified and removed after the photon extraction step, where our customized tool evaluates the statistical significance of each detection above background, rejecting likely false detections when significances fail to meet a chosen threshold. These photons are then returned to the background and we re-evaluate the source significance for the surviving candidates, finally defining reliable detections and marginal detections for inclusion in tables of source properties. The details of this procedure are described below. Source detection was carried out with the improved Level 2 event list that is band- limited to 0.5–7.0 keV with cosmic ray afterglows removed. Effective exposure maps and images for the full field, the 17′ 17′ I array, the central 8′ of the I array, and the inner 4′ × of the field were made with 4-pixel, 2.6-pixel, 1-pixel, and half-pixel binning, respectively (an ACIS sky pixel is 0.5′′ on a side). Adaptive-kernel smoothed flux images were created with the CIAO tool csmooth (Ebeling et al. 2006) to help identify additional potential faint sources and any possible diffuse emission. 5 First the wavdetect program was run with source significance threshold 1 10− × on each of the four binned images described above, in three energy bands (0.5–2.0 keV, 2.0–7.0 keV, and 0.5–7.0 keV). This source significance threshold may give some false detections, but experience has shown us that this threshold is appropriate for crowded fields that might contain complex backgrounds due to diffuse emission. Merging the detections from the 12 runs resulted in a list of potential point sources on ACIS-I. Careful visual inspection revealed that wavdetect still missed several apparent sources in the central region of the cluster. To take advantage of the sub-arcsecond PSF at positions around the aimpoint, we applied a subpixel positioning code (Mori et al. 2001) to improve spatial resolution in the inner part of the field and performed an image reconstruction with the Lucy-Richardson maximum likelihood algorithm (Lucy 1974). Our image reconstruction technique for crowded stellar clusters is described in detail in Townsley et al. (2006b) and a similar procedure is followed here. The maximum 3 likelihood image reconstruction code was run in the crowded region using a 50′′ 50′′ ∼ × image made with 0.2-pixel binning, using the 1.0 keV PSF at that on-axis location from the PSF library in the CIAO calibration database. We chose the 200th iteration of the algorithm as the most appropriate image for source searching, as it showed good PSF removal without over-resolving the data. We adopted the find procedure in the DAOPHOT package (Stetson 1987) to obtain the centroid, shape, and brightness parameters of the resolved peaks. Several combinations of flux (f), roundness (r), and sharpness (s) limits were tested to minimize the number of peaks from one photon, and the adopted parameters to reject spurious peaks were: r > 1.0 or r < 1.0; or f < 0.8; or f < 1.6 − and s > 0.98. Using SAOImage DS9 (Joye & Mandel 2003), world coordinate system (WCS) region files were made, centered on the unrejected peaks, based on 90% PSF contours. The final reconstructed source list was made with the aid of the I-array image

3From the IDL Astronomy User’s Library maintained by Wayne Landsman at http:// idlastro.gsfc.nasa.gov/homepage.html. 20 overlaid with these regions where any remaining reconstructed peaks that contain only single photon events within 90% PSF contours were rejected. This image reconstruction procedure needs to be performed separately because Chandra’s PSF varies at different locations and an appropriate local PSF must be used for the reconstruction. The list of potential sources from the maximum likelihood reconstruction was merged with the wavdetect source catalog. By overlaying regions representing the potential sources onto the original and smoothed ACIS-I images, additional sources were added based on spatial concentrations of 3 photons and proximity of near-IR sources ( 1′′). ≥ ≤ The above source finding procedure ended with a list of potential sources identified on the ACIS-I array. A preliminary event extraction for these candidate sources is made with AE, which computes a wide variety of statistical properties of the sources. Using the AE-calculated probability PB that the extracted events are solely due to Poisson fluctuations in the local background, source validity can be statistically evaluated while taking account of the large distorted PSF at far off-axis locations and spatial variations in the background. The traditional source significance, defined as the photometric signal to noise ratio, is calculated for every source as well. Empirically we reject sources with PB > 1% likelihood of being a background fluctuation. This threshold can vary in different observations, and is determined by careful review of the neighborhood images of the extracted sources with high PB and counts distribution with respect to the off- axis angles. We performed source extraction with this trimmed source list using AE as described in detail by Townsley et al. (2006b).

2.3.3 Source Variability and Spectral Fitting A band-limited, adaptively smoothed light curve for each source was generated by AE together with a median energy time-series. The variability of the source was evaluated by the significance of a one-sided Kolmogorov-Smirnov statistic PKS, comparing the source events arrival times to that of a uniform light curve model. If PKS > 0.05, the null hypothesis of a constant source was accepted. If PKS < 0.005, the source was classified as “variable.” Otherwise, the source was “possibly variable.” 4 The extracted source spectra were fit for relatively bright sources, the extracted spectra were fit using single temperature and two-temperature apec thermal plasmas (Smith et al. 2001) and power law models subjected to an absorbing column (NH ) of 5 interstellar material with the XSPEC package (Arnaud 1996), based on source spectra, background spectra, ancillary response functions (ARFs) and redistribution matrix functions (RMFs) from ACIS Extract. The best-fit model was achieved by the maximum 2 likelihood method (Cash 1979), although χ fitting was also used in the NGC 6357 data analysis. Abundances of 0.3 Z⊙ for the X-ray emitting plasma were assumed for the automated fitting performed by AE. This is based on observational results that the mean

4There is no definite criteria of bright sources in our observations with moderate exposure time. In NGC 6357, we choose the 40 sources with >80 net counts. In the latest papers, our group decide to adopt photometric significance to select sources for spectral fitting, instead of the arbitrary net counts. 5http://heasarc.gsfc.nasa.gov/docs/software/lheasoft/xanadu/xspec 21 chemical composition of X-ray emitting plasma in young stars is generally lower with respect to the solar photospheric value. The topic of coronal abundances is thoroughly reviewed in Drake (2002), Favata & Micela (2003), and Güdel (2004). The abundances of the stellar corona are found different from the photospheric values depending on the first ionization potential (FIP) of the relevant elements. In the solar corona, elements with high FIP ( 10 eV) appear depleted relative to the solar photospheric values (the ≥ FIP effect; Feldman 1992). In some stellar coronae, however, the abundances of high FIP elements are overabundant, known as the inverse FIP (IFIP) effect. The IFIP effect is commonly observed in X-ray active stars (e.g., Brinkman et al. 2001; Huenemoerder et al. 2001; Kastner et al. 2002; Maggio et al. 2007); coronal abundance of Fe appears depleted by factors of a few in these stars. In general we prefer unconstrained fits (including power law fits) than constrained fits. The single temperature thermal plasma apec model is the default model used for spectral fitting. For sources brighter than 100 counts, if a one-temperature thermal plasma model did not fit the data well, a two-temperature thermal plasma model or variable abundance vapec thermal plasma model was invoked. A power law model was adopted if it represented the data more adequately than the thermal model (visually or with improved statistics) or the thermal model required nonphysical parameters (e.g., kT 15 keV). For a number of sources that required a very hard thermal plasma ≫ and were identified with known stellar counterparts or exhibited flaring light curves that were suggestive of PMS stars, we truncated the plasma temperature at kT = 15 keV and adopt the thermal plasma model. When no model was acceptable, we froze the parameter kT = 2 keV in the thermal model, a typical value for young PMS stars (Getman et al. 2005b; Preibisch & Feigelson 2005), and then fit for the absorbing column density NH and the normalization parameters. Counts per CCD frame was checked for the brightest source in each observation. No source shows high count rate to corrupt our spectral fitting, thus no correction for photon pile-up is performed(Townsley et al. 2002).

2.3.4 Simulation of Extragalactic Contamination We estimate the contamination from the extragalactic population following the method described in Getman et al. (2006). Monte-Carlo simulations of the expected extragalactic source contamination are constructed by placing fake AGNs randomly across the ACIS-I field superposed on the ACIS background. The X-ray flux distribution follows the logN–logS distribution from Moretti et al. (2003) and the corresponding power law photon index for individual sources follows the flux dependence found by Brandt et al. (2001). Synthetic source spectra were generated with the XSPEC fakeit function and convolved with an absorption column NH . We ran simulations with three typical 22 −2 NH : the total Galactic absorption towards NGC 6357 (NH = 1.2 10 cm , Dickey & × 22 −2 Lockman 1990); Galactic absorption plus an additional NH = 1.0 10 cm typical of × 22 −2 Pismis 24 OB stars; and Galactic absorption plus an additional N = 3.0 10 cm H × typical of reddened 2MASS stars. Then we identified the closest real X-ray detection to the fake source. Source significance for each fake AGN was then calculated, using the local background of a real source that is closest to the fake source. We set a detection 22 threshold based on source photometric significance (Signif). This limiting source significance is determined by the median Signif of those hard X-ray sources without infrared counterparts and lying far off-axis, which are the most likely candidates for AGNs in the field.

2.3.5 Simulation of Stellar Contamination in the Galactic Disk To estimate possible contamination from foreground field stars, we follow the techniques described in the Chandra Orion Ultradeep Project (COUP) membership study and the Chandra study of the Cep B star forming region (Getman et al. 2005b, 2006). We use simulations based on the stellar population synthesis model of Galactic disk stars by the Besan¸con group (Robin et al. 2003). The models are calculated using their 6 on-line service . Synthetic catalogues of stars generated from these models give stars 2 with V < 22 mag in a solid angle of 0.08 deg (ACIS-I FOV) to the distance of ∼ Chandra observed star forming region. These stars include main sequence (MS) stars of late spectral types, giants, and a few subdwarfs. We then adopt the XLFs for MS K-M stars (Schmitt et al. 1995), F-G stars (Schmitt 1997), and late type giants (Huensch et al. 1996; Pizzolato et al. 2000) derived from a complete volume-limited sample in the solar neighborhood. The conversion of intrinsic X-ray flux in the ROSAT PSPC 0.1 2.4 keV band to Chandra ACIS-I 0.5 8.0 − 2− keV count rate is from PIMMS, using a thermal plasma model assuming kT (4πd FX −26 0.2 ≈ × 5.5 10 ) /11.34 keV from Güdel et al. (1998), where d is the stellar distance and × F is the X-ray flux in ROSAT PSPC 0.1 2.4 keV band. Only Galactic absorption X − is assumed to be present for stars in the foreground of the region of interest and the amount of absorption to the fake star is calculated based on its distance when the simulated catalog is generated. Following the same procedures used in the simulations for extragalactic contamination, a simulated field of X-ray stars was added to the ACIS background map, and the star locations were examined for source significance. For a given detection threshold Signif, 10000 simulation runs give a typical number of detectable foreground field stars. These sources should be readily distinguished from extragalactic AGN and most young cluster members by the brightness of their stellar counterparts and (when sufficient X-ray counts are present to measure median energies) low absorption in their spectra. When the Chandra observed star forming region does not have any thick molecular cloud in the background, the contamination from background stars also needs to be considered. The Besan¸con model is run again to obtain the distribution of background stellar population. The XLFs for MS stars and giants are used again to simulate the number of stars behind the observed star forming region that may enter our X-ray sample. A complete removal of all non-members is impossible without further spectroscopic information, thus we ignore the remaining stellar and AGN contamination in further analysis given that the fraction is statistically very small.

6http://bison.obs-besancon.fr/modele/ 23

2.3.6 NIR Color-Color and Color-Magnitude Diagrams The identification of young stars detected in X-rays needs to be confirmed in optical, IR, and longer wavelengths. We match our X-ray detections to cataloged stars in other bands and study their properties. The H K vs. J H color-color diagram and − − the J H vs. J color-magnitude diagramof NIR counterparts to Chandra sources are − used as diagnostic tools of extinction and age for each cluster. We use NIR counterparts 7 with high-quality photometry . The locus for main-sequence stars of different spectral types from early O to late M is obtained from Bessell & Brett (1988). The locus of CTTS is taken from Meyer et al. (1997) and the locus for Herbig Ae/Be stars is from Lada & Adams (1992). We use transformation equations presented in Carpenter (2001) to convert colors and magnitudes among 2MASS photometric system, CIT system, and Bessell & Brett (1988) photometric system. The J H vs. J color-magnitude diagram gives both the rough mass distribution − and the absorption distribution of the ACIS sources with NIR counterparts. Zero age main sequence (ZAMS) track is taken from Cox (2000). Pre-MS isochrones of different ages are adopted from the generally accepted models by Baraffe et al. (1998); Siess et al. (2000). Reddening vectors are calculated assuming the standard interstellar reddening law (Rieke & Lebofsky 1985).

2.3.7 Stellar Surface Density Map To identify large structure and substructures, we perform a simple sliding-cell smoothing to create the stellar surface density map of the X-ray sampled cluster. A 20′ 20′ grid is created to cover the stellar positions of the ACIS-I array, and at each ∼ × position the total number of detected X-ray sources within a sampling kernel of radius r0 (arcsec) is counted to calculate the local average stellar density. The extent of the resolved spatial structure largely depends on the adopted r0 and a number of r0 are used to check whether any substructure persists and is indeed significant. Note that no X-ray luminosity information is needed. A bright source and a faint detection both count as a detection equally. In the case of multiple observations, only sources with similar exposure time were considered to guarantee roughly equal X-ray sensitivity throughout the field. Because of the energy information embedded in the X-ray data, this can be done separately for the unobscured population and obscured population in the cluster of interest. A similar smoothing technique is used in 2MASS data (Li 2005; Li & Smith 2005a,b). This simple procedure enhances visual inspection of substructures and the identified structures are confirmed with the results of more sophiscated cluster finding algorithm.

7In case of 2MASS data, we require the 2MASS counterparts here and in the J H vs. J diagram to have signal-to-noise ratios greater than 6 (photometric quality higher than− “C”) and are free of confusion ﬂags and warning of contamination ﬂags. See Explanatory Supplement to the 2MASS All Sky Data Release (Cutri et al. 2003; http://www.ipac.caltech.edu/2mass/ releases/allsky/doc/) for details. 24

Fig. 2.1 A schematic drawing of the Chandra Observatory is shown with the modules labeled (the Chandra Proposer’s Observatory Guide 2006). 25

ACIS FLIGHT FOCAL PLANE

~22 pixels ~11" not constant with Z

I0 I1 w203c4r w193c2 0 1 ACIS-I 22 pixels x (aimpoint on I3 = (949, 978)) = 11" I2 I3 } w158c4r w215c2r 2 3 330 pixels = 163"

S0 S1 S2 S3 S4 S5 + w168c4r w140c4r w182c4r w134c4r w457c4 w201c3r ACIS-S 4 5 6 7 8 9 } (aimpoint on S3 = (235, 497))

∆ + Y Target 18 pixels = 8".8 +Z +Z Offset Coordinates BI chip indicator Top

column Pointing one two three + ∆ Z

node zero row Coordinates Image Region Bottom . . -Z Pixel (0,0) +Y Sim Motion Frame Store Coordinate CCD Key Node Row/Column Definitions Definition Orientations

Fig. 2.2 A schematic drawing of the ACIS focal plane (the Chandra Proposer’s Obser- vatory Guide 2006). The top panels show the ACIS-I and ACIS-S arrays, and the lower panel explain some terminologies. Our observations typically utilize ACIS-I0, I1, I2, I3, S2, and S3, with the aimpoint at the corner of the I3 chip denoted by “x”. 26

Chapter 3

An X-ray Census of Young Stars in the Massive Southern Star-Forming Complex NGC 6357

3.1 Introduction

NGC 6357 (= W 22 = RCW 131 = S 11) is a large H ii region complex in the southern sky excited by the OB association Pismis 24, at a distance of 2 kpc (Neckel ∼ 1978; Felli et al. 1990; Massi et al. 1997; Bohigas et al. 2004). This young stellar cluster and its surrounding gaseous environment are relatively poorly studied; only a handful of stellar members have been identified and characterized, and the molecular cloud has not been well-mapped. In optical and near-infrared images, the exciting cluster is not prominent against the Galactic field population except for a few bright members (Figure 3.1). Yet, two of these stars are remarkably luminous and massive with spectral types O3 If (M 200 300 M⊙) and O3 III(M 100 M⊙; Massey et al. 2001; Walborn et al. 2002). ∼ − ∼ Such massive stars are usually found in only the richest Galactic clusters and the giant starburst regions such as Carina, 30 Doradus, and NGC 3603. This raises the question is Pismis 24 much richer than it appears, or is it a poor cluster with a top-heavy IMF? This would be highly unusual, as top-heavy IMFs are seen only in extreme starburst environments such as the super-star clusters of M 82 (Elmegreen 2005). We investigate this question by measuring, for the first time, the low-mass pre- main sequence (PMS) population of the Pismis 24 cluster and its environs. This is achieved with a sensitive X-ray observation with the Chandra X-ray Observatory. Chan- dra observations can penetrate heavy absorption up to A 500 (Grosso et al. 2005) V ≃ and resolve point sources with sub-arcsecond resolution. In most cases, the X-ray detected stars appear on existing near-infrared (NIR) images; the advantage is that the 1 4 strong X-ray emission of PMS stars (elevated 10 10 above MS stars Preibisch & − Feigelson 2005) traces magnetic activity rather than photospheric or circumstellar disk emission, therefore effectively discriminating cluster members from older Galactic field stars. We estimate that the cluster has 10, 000 stars, 5 times richer than the known ≃ ≃ Orion Nebula Cluster. The presence of two O3 stars is compatible with a cluster of this population. The known massive stars and the large number of low mass X-ray stars suggest that Pismis 24 is a very rich cluster with a standard IMF. Our study also gives considerable new information about the cluster and its X-ray properties. We provide a census of 750 cluster members with sub-arcsecond positions, ∼ 613 of which have NIR counterparts. Although not complete, the survey includes PMS stars with masses extending down to 0.3M⊙; most do not have circumstellar disks ∼ revealed by K-band excess. We study the spatial distribution of the cluster members and identify a possible subcluster. We examine the ringlike morphology of the NGC 6357 nebula and investigate the possible feedback from the massive stars to their environs. 27

We measure X-ray properties of a dozen known O stars, and list over 20 new candidate O stars. A faint X-ray source is found associated with a protostar in an Evaporating Gaseous Globule (EGG) for the first time. A few PMS stars are seen with very powerful 32 1 X-ray flares with peak L 10 ergs s− . x ∼ Section 3.2 reviews past study of the Pismis 24 cluster and NGC 6357 region. The Chandra observation and its analysis are described in 3.3, and a NIR observation using § the Simultaneous 3 color InfraRed Imager for Unbiased Surveys (SIRIUS) camera on the Infrared Survey Facility (IRSF) telescope is summarized in 3.3.5. The association § between X-ray sources and optical and near-infrared (ONIR) stars, as well as the infrared colors of X-ray counterparts are presented in 3.4. The cluster population inferred from § the X-ray emitting stars, their XLF, the spatial distribution of the cluster members, and the morphology of the nebula are discussed in 3.5. X-ray properties of O stars and § intermediate- to low-mass PMS stars are presented in 3.6. The possibility of diffuse § X-ray emission is examined in 3.7. A brief summary of our findings is given in 3.8. § § This study is part of a series of X-ray investigations of high-mass star formation regions that includes the Rosette Nebula (Townsley et al. 2003; Wang et al., in preparation), M 17 (Townsley et al. 2003; Broos et al., in preparation), the Orion Nebula (Feigelson et al. 2003; Getman et al. 2005 and associated articles), 30 Doradus (Townsley et al. 2006a, 2006b), and Cepheus B/OB3b (Getman et al. 2006).

3.2 Observational Overview of NGC 6357 and Pismis 24

NGC 6357, a large H ii region complex showing an annular morphology in the radio and optical (Haynes et al. 1978; Lortet et al. 1984), is located in the Sagittarius spiral arm and spans 60′ 40′ in the southern sky. NGC 6334, another prominent star- × ′ forming complex likely associated with NGC 6357, lies 60 away (Neckel 1978; Ezoe ∼ et al. 2006a). The massive open cluster Pismis 24 (= C 1722-343 = OCl 1016) lies in NGC 6357’s central cavity (Pismis 1959; Alter et al. 1970). The kinematic distance to NGC 6357, d = 1.0 2.3 kpc, is given by Wilson et al. (1970). Neckel (1978) derives ± d = 1.74 0.31 kpc with improved accuracy based on UBV and Hβ data. Spectroscopic ± parallax of 10 high-mass members in Pismis 24 gives a larger distance of 2.56 0.10 kpc ± (Massey et al. 2001), which we adopt in this chapter. We should note that recent work (Kharchenko et al. 2005; Prisinzano et al. 2005; Arias et al. 2006) has found smaller distances to other high mass star forming regions (e.g., M8) than previous studies; we will discuss the uncertainties in distance-related stellar properties wherever applicable. Figure 3.1b shows the large-scale morphology in the mid-infrared, observed with the Midcourse Space Experiment (MSX) satellite. A large ring-like structure extends 60′ in diameter and bright nebulosity is seen at the northern rim. The central cavity ∼ appears free of large amounts of warm dust. Far-infrared (FIR) continuum revealed 5 several luminous (L 10 L⊙) embedded sources coincident with CO line and radio ∼ continuum emission peaks (McBreen et al. 1983). G353.2+0.9, which we study here, is 5 the youngest (t 10 yr) and brightest of NGC 6357’s three H ii regions. It lies at the ∼ northern interface between the cluster and molecular cloud and contains a number of embedded sources (Figure 3.1a; Frogel & Persson 1974; Sakellis et al. 1984; Persi et al. 1986; Massi et al. 1997; Bohigas et al. 2004). In contrast with NGC 6334, almost no water 28 maser emission is found in NGC 6357 (Healy et al. 2004a). This may either indicate that massive star formation has ceased in the region (Persi et al. 1986) or that it has been overrun by the ionization front (Healy et al. 2004a). G353.1+0.6 is a more evolved H ii region viewed edge-on, bounded on the northern side by a different molecular cloud. It contains a single O5 star, several B stars and embedded IR sources, but no indication of current star formation can be found (Felli et al. 1990; Massi et al. 1997). G353.2+0.7, which is also a FIR peak, does not contain early type stars (Felli et al. 1990; Massi et al. 1997), but it hosts the only water maser found in NGC 6357 (Sakellis et al. 1984). However, Persi et al. (1986) identifies this water maser source as an evolved late M star. Despite a number of optical, infrared, and radio studies on NGC 6357, no X- ray observation of this field has been reported. Figure 3.1f shows an unpublished 9 ks ROSAT PSPC observation revealing a few strong X-ray sources, including NGC 6357’s three H ii regions. The brightest X-ray emission coincides with the core of the Pismis 24 cluster, but the ROSAT instrumentation was unable to resolve the source population. The OB cluster Pismis 24 is centered 1′ south of the G353.2+0.9 ionization ∼ front. Early optical study of NGC 6357 and Pismis 24 revealed 20 O-type and early ∼ B-type stars (Moffat & Vogt 1973; Neckel 1978, 1984; Lortet et al. 1984), including the unusally late-type WC7+O7-9 Wolf-Rayet (WR)/O star binary HD 157504 (= WR 93; van der Hucht 2001). Two of the cluster members, namely Pismis 24-1 (=HDE 319718) and Pismis 24-17, were recently classified as spectral type O3.5, some of the brightest and bluest stars known (Massey et al. 2001; Walborn et al. 2002). Only a dozen other O3 stars are known in the Galaxy (Massey & Johnson 1993; Walborn 1994; Ma´ız-Apellániz et al. 2004). The two luminous O3 stars are easily identified close to the sharp boundary of the bright optical nebula in the optical image, accompanied by several other bright cluster members extending 4′ north-south (Figure 3.1e), but are not as apparent in the ∼ Two Micron All Sky Survey (2MASS) K-band image (Figure 3.1c). Figure 3.1d shows a SIRIUS JHK composite image, which has higher resolution and penetrates deeper into the cloud than 2MASS. The two luminous O stars are evident in the center of the SIRIUS image, which contains thousands of stars. Other noticeable features are the bright nebula with tenuous diffuse filaments in the north and the infrared dark column occupying the south-east corner of the image, previously noted as the south eastern complex (S.E.C.) in CO mapping (Massi et al. 1997). Several heavily reddened stars can be seen in the S.E.C. region. However, most NIR stars are foreground and background objects, not cluster members. Although most researchers believe G353.2+0.9 is excited by the hot stars in Pismis 24 (e.g., Lortet et al. 1984), Felli et al. (1990) argued that Pismis 24 is projected onto the H ii region by chance, and the embedded massive stars (IRS 1, 2, and 4; Persi et al. 1986) supply the ionizing radiation. Recent studies do not support this claim. Massi et al. (1997) have shown that most molecular gas lies behind the H ii region and the cluster location is consistent with a face-on blister-type H ii region morphology. Optical spectroscopic study by Bohigas et al. (2004) indicates that the hot O stars of Pismis 24 50 1 emit 10 UV photons s− , sufficient to ionize the H ii region. Massey et al. (2001) consider the W-R/O binary WR 93, lying 4′ away from the cluster core, to be a likely member of Pismis 24 and caution that the cluster might be larger than the central concentration of O stars. Spectroscopic study indicates a mass 29

−5 −1 loss rate for WR 93 of M˙ = 2.5 10 M⊙ yr using a clumped wind model with × −1 terminal wind velocity v∞ = 2290 km s (Prinja et al. 1990; Nugis et al. 1998; van der Hucht 2001). The nearby radio continuum and CO peaks suggest that WR 93 ionizes and heats the surrounding cloud (McBreen et al. 1983). The visual extinction towards WR 93 is high, with E(B V ) = 1.82 or A 5.8 mag (Nugis et al. 1998; van der − V ≃ Hucht 2001). Healy et al. (2004b) and Hester & Desch (2005b) present an HST/WFPC2 image of the G353.2+0.9 H ii region in NGC 6357, illustrating a sequence of formation of low- mass stars and describing their evolution from molecular cores to EGGs to proplyds in the vicinity of massive stars. Very recently, De Marco et al. (2006) analyze HST observations of eight HII regions, including Pismis 24, to look for protoplanetary disks. In Pismis 24, they ﬁnd a jet-like feature, similar to a Herbig-Haro object in their source # 14. However, no central star is seen in the jet candidate.

3.3 Observations and Data Reduction

3.3.1 Chandra Observation and Data Selection NGC 6357 was observed on July 9, 2004 with the Imaging Array of the Advanced CCD Imaging Spectrometer (ACIS-I) on board Chandra. Detailed description of the instrument can be found in Weisskopf et al. (2002). Four front illuminated (FI) CCDs form the ACIS-I which covers a field-of-view (FOV) of 17′ 17′. Two CCD chips ∼ × on the ACIS Spectroscopic Array (ACIS-S) were also set to be functioning during the observation, although the mirror point spread function (PSF) degrades significantly far (> 15′) off-axis. Here the ACIS-S data will not be quantitatively discussed. The observation was made in the standard Timed Exposure, Very Faint mode, with 3.2-second integration time and 5 pixel 5 pixel event islands. The total exposure time is 38 ks × and the satellite roll angle is 289 degrees. The aim point is centered on the O3 If star Pis 24-1, the heart of OB association Pismis 24. The observation ID is 4477. Chapter 2 provides detailed explanations of the Penn State ACIS data reduction procedure. To improve astrometry, we calculated the position offsets between 277 X- ray sources and their 2MASS counterparts, and applied an offset of +0′′.02 in right ascension (RA) and 0′′.33 in declination (Dec) to the X-ray coordinates. The CCD − charge transfer inefficiency (CTI) was corrected using the Penn State CTI corrector (version 1.16) developed by Townsley et al. (2000, 2002).

3.3.2 Image Reconstruction and Source Finding Figure 3.2a shows the raw ACIS I-array image at reduced resolution. Adaptive- kernel smoothed flux images were created with the CIAO tool csmooth (Ebeling et al. 2006). The resulting full field two-color smoothed image and a zoom-in on the cluster core are shown in Figure 3.2; red and blue represent the soft band emission (0.5–2.0 keV) and hard band emission (2.0–7.0 keV), respectively. 5 The wavdetect program was run with source significance threshold 1 10− on × each of the four binned images described above, in three energy bands (0.5–2.0 keV, 2.0–7.0 keV, and 0.5–7.0 keV). Merging the detections from the 12 runs resulted in a 30 list of 614 potential point sources on ACIS-I. To take advantage of the sub-arcsecond PSF at positions around the aimpoint, we applied a subpixel positioning code (Mori et al. 2001) to improve spatial resolution in the inner part of the field and performed an image reconstruction with the Lucy-Richardson maximum likelihood algorithm (Lucy 1974). Our image reconstruction technique for crowded stellar clusters was applied to the central region using a 50′′ 50′′ image made with 0.2-pixel binning and centered ∼ × on Pis 24-1, using the 1.0 keV PSF at that on-axis location from the PSF library in the CIAO calibration database. This image reconstruction procedure was performed separately using another image centered on Pis 24-17. Examples of valid sources from image reconstruction are shown in Figure 3.3. The list of potential sources from the maximum likelihood reconstruction was merged with the wavdetect source catalog. By overlaying regions representing the potential sources onto the original and smoothed ACIS-I images, an additional 40 sources were added based on spatial concentrations of 3 photons and proximity of 2MASS near-IR sources ( 1′′). The source finding ≥ ≤ procedure described here results in a total of 910 potential sources identified on the ACIS-I array.

3.3.3 Photon Extraction and Limiting Sensitivity A preliminary event extraction for the 910 potential X-ray sources was made with ACIS Extract (version 3.79; hereafter AE), a versatile script written in IDL that performs source extraction, fits X-ray spectra, creates light curves, and computes a wide variety 1 of statistical properties of the sources (Broos et al. 2002) . Using the AE-calculated probability PB that the extracted events are solely due to Poisson fluctuations in the local background, source validity can be statistically evaluated while taking account of the large distorted PSF at far off-axis locations and spatial variations in the background. The traditional source significance, defined as the photometric signal to noise ratio, is calculated for every source as well. We rejected sources with PB > 1% likelihood of being a background fluctuation. The trimmed source list includes 779 sources, with full-band (0.5–8.0 keV) net (background-subtracted) counts ranging from 1.7 to 1837 counts. We performed source extraction with this list using AE as described in detail by Townsley et al. (2006a). The 779 valid sources are purposely divided into two lists: the 665 sources with PB < 0.1% make up our primary source list of highly reliable sources (Table 3.1), and the remaining 114 sources with P 0.1% likelihood of being spurious background B ≥

1http://www.astro.psu.edu/xray/docs/TARA/ae users guide.html 31

2 fluctuations are listed as tentative sources in Table 3.2. We believe that most of those 3 tentative sources are likely real detections . We should note that the separation of reliable sources, tentative sources, and invalid sources is based on a single simple statistic, the Poisson probability of observing the extracted counts given the local background density. This permits scientists to evaluate the reliability of each source individually. A high cutoff (around 4 counts for a typical Chandra exposure on-axis) reduces spurious noise-based sources (false positives) but misses some real sources (false negatives). A low cutoff reduces the false negatives while increasing the false positives. In the absence of a consensus statistical procedure for the Poisson weak-signal-in-background problem, there is no way to clean the list in an objective fashion without losing real sources. Since different scientists make different decisions on the balance between false positives and false negatives, and there are good scientific reasons to believe many astronomically interesting faint sources lie in Chandra images of rich stellar clusters, we have opted to report marginal sources in Table 2. The reader is clearly warned that the reality of some of these weakest sources is suspect, and each source’s PB can be examined individually. These choices concerning how we choose to report faint sources do not affect our science results. The weakest X-ray sources are only used for X-ray luminosity function estimation ( 5.1), contributing to the faintest § two bins of the XLF, which do not impact on our conclusions regarding the NGC 6357 stellar population. We further note that every detection scheme must – explicitly or implicitly – estimate the local background, and every such estimate must implicitly identify the data that are guessed not to be background. For crowded young stellar clusters, this can never be accomplished in a unique, objective manner as it depends on the unknown distribution of true faint sources (including thousands of stars which lie below any detection threshold) and possibly diffuse emission from O star winds (Townsley et al. 2003). Table 3.1 and Table 3.2 have a format that closely follows the tables in Getman et al. (2006) and are identical to Tables 1 and 2 in Townsley et al. (2006a). Column 1 gives the ACIS running sequence number, and column 2 provides the IAU designation. Columns 3–6 show RA and Dec in degrees, positional error in arcseconds, and off-axis angle in arcminutes. Column 7–11 give net counts in the full band, the associated 1σ- equivalent uncertainty based on Poisson statistics, background counts in the extraction region, net counts in the hard band, and the fraction of the PSF used for source extraction

2 A few PB values are greater than 0.01, our rejection threshold in the preliminary run. This is due to the feedback to the background from the rejected sources when we re-extract the trimmed source list. As described in detail in 2.1 of Townsley et al. (2006a), our catalog cutting scheme is iterative: the initial set of 910 potential§ sources is extracted and then cut at the threshold of 1%. The rejected sources are allowed to contribute to the background and the trimmed list of 779 valid sources is then re-extracted, resulting in different local background estimates and different source validity metrics PB. Thus, some PB values in Table 2 are > 1%. 3 The fraction of tentative sources (Table 2) that have optical and near-infrared counterparts is 50%, suggesting that they are not random spurious detections ( 4.1). For the 64 weakest X-ray sources that have 3 counts or less, 10 of them have been cautioned§ with a ”U” flag (unreliable sources that have large probability of being background, 4.5) in the table, while 27 of them have optical and near-infrared identifications. Most of the§se faint sources are clustered within ′ the central 2 of the field and are likely low-mass cluster members. 32

(reduced values for this quantity indicate source crowding). Columns 12 and 13 present two measures of the source validity: the photometric significance, and the probability that the source is a spurious background fluctuation PB, used for establishing source validity as described above. Column 14 lists possible anomalies in the observation and Column 15 checks the source variability. Table footnotes give details for Column 13– 15. Column 16 gives the “effective” exposure time, defined as the amount of exposure time needed for the source to accumulate the current number of counts if placed at the Chandra aimpoint. Column 17 gives the source’s median energy in the full spectral band after background subtraction. The on-axis limiting sensitivity of this 38 ks observation can be estimated using the 4 Portable Interactive Multi-Mission Simulator (PIMMS ). Assuming an on-axis detection 22 of 3 counts (0.5–8.0 keV) in 38170 seconds and an absorbing column NH = 1.0 10 −2 × cm (estimated from an average A 6.0 mag to Pismis 24 in Bohigas et al. 2004, V ∼ adopting the conversion between NH and AV from Vuong et al. 2003), the observed X- ray flux for a thermal plasma with temperature of 1 keV (appropriate for a PMS star’s 16 1 2 X-ray spectrum) is 5.6 10− ergs s− cm− . Correcting for absorption and a distance × −1 of 2.56 kpc to Pismis 24, this gives a limiting luminosity of log Lt,c = 30.2 erg s , where Lt,c is the intrinsic (absorption-corrected) luminosity in the total Chandra band 0.5–8.0 keV.

3.3.4 Source Variability and Spectral Fitting The variability of the source was evaluated by the significance of a one-sided Kolmogorov-Smirnov statistic PKS, comparing the source events arrival times to that of a uniform light curve model. If PKS > 0.05, the null hypothesis of a constant source was accepted. If PKS < 0.005, the source was classified as “variable.” Otherwise, the source was “possibly variable.” For the 40 sources with >80 net counts, the extracted spectra were fit to models of optically thin thermal plasmas using the XSPEC package (Arnaud 1996), based on source spectra, background spectra, ancillary response functions (ARFs) and redistribution matrix functions (RMFs) constructed by the AE script. The data were grouped in 2 bins with 10 counts and the χ statistic was adopted to evaluate the goodness-of-fit. A ≥ single apec (Smith et al. 2001) thermal plasma model with a single absorption component was considered. Interactive fitting with alternative models (e.g., vapec thermal plasma with variable abundances; two-temperature apec thermal plasma model) was performed for a few strong sources that did not achieve a satisfying fit with the automated process. Table 3.3 gives the best fit results using the thermal plasma model; its format closely follows Table 3 in Townsley et al. (2006a). Uncertainties representing 90% confidence intervals are given; incomplete confidence levels imply the parameters are not well constrained, or XSPEC may have encountered some abnormality in the error calculations. For the 144 sources that have net counts between 20 and 80, we fit ungrouped spectra with a single temperature apec thermal plasma model using the likelihood ratio test (C statistic); the fitting results are presented in Table 3.4.

4 http://heasarc.gsfc.nasa.gov/Tools/w3pimms.html 33

None of the NGC 6357 sources are bright enough to warrant correction for photon pile-up. The brightest source in the field is the W-R/O binary WR 93 which has 1837 ACIS counts in the 38 ks exposure, or a count rate of 0.15 counts per CCD frame. As it lies 5′.4 off-axis where the PSF is somewhat degraded, the photon pile-up effect will not be significant enough to distort the spectral fitting, according to previous simulation work (Townsley et al. 2002).

3.3.5 SIRIUS NIR Observation To identify infrared counterparts with images that are deeper and of higher spatial resolution than 2MASS, J, H, and K band images were obtained on 30 September 2005 using the wide-ﬁeld NIR SIRIUS camera on the 1.4 m IRSF telescope at the South Africa Astronomical Observatory. Details of the instrument can be found in Nagayama et al. 1 (2003). The pixel scale of the detector is 0.45 arcsec pixel− with a 7′.7 7′.7 ﬁeld- × of-view. Three second exposures were made at each of 20 dither positions for a total exposure of 60 seconds. A typical limiting magnitude for such SIRIUS observations is Ks=15.7. Unfortunately, the observing conditions were non-photometric and the seeing was 1.1′′ FWHM in the K band. We nonetheless found these images to be very valuable for identifying infrared counterparts (as described below) because of the higher spatial resolution compared to 2MASS.

3.4 Identiﬁcation of the Chandra Sources

3.4.1 X-ray Sources with Stellar Counterparts Identifying stellar counterparts for X-ray sources in massive star forming regions beyond 1 kpc is challenging. In regions such as the Chamaeleon I cloud (0.2 kpc) or Cepheus B region (0.75 kpc), the cluster members are sufficiently bright and well- separated that the 2MASS catalog provides adequate identifiers for nearly all Chandra stellar sources (Feigelson & Lawson 2004; Getman et al. 2006). However, for NGC 6357 at 2.5 kpc, deeper and higher resolution observations such as the SIRIUS images are needed for source identification. In massive star forming regions such as NGC 6357, optical and NIR imaging have the additional challenge of detecting faint stars amidst highly structured background emission from the H ii region or deeply embedded in the molecular cloud. Thus we cannot arrive at a quantitative statement regarding the completeness of the search for stellar counterparts of Chandra sources. But we can say that, with very few exceptions, e.g., infrared luminous active galactic nuclei (AGNs), the counterparts which we do identify are indeed cluster members. For sources with off-axis angle θ 3′, the Chandra sourcelist was matched to ≤ ′′ the 2MASS Catalog of Point Sources with a 1 matching radius. The search radius was enlarged to 2′′ for θ > 3′ in consideration of the degraded off-axis Chandra PSF (Getman et al. 2005a). We find that 445 of our X-ray sources (58%) have 2MASS counterparts. The matching results are presented in Table 3.5. Four of the ACIS sources (#88, 517, 688, 774) have two counterparts within 2′′ and we assign the star with the smaller separation as the real counterpart. Three pairs of sources – 252 and 257, 401 and 407, 404 and 406 – share a single 2MASS counterpart within 1′′. Given the high density 34 of both X-ray sources and 2MASS objects around cluster center, we follow the simple Monte Carlo technique from Alexander et al. (2003) to evaluate the number of chance superpositions between 2MASS sources and X-ray detections. A random offset between 5′′ and 10′′ is applied to the catalog to remove the true associations, and then the same cross-identification radii are used to match X-ray positions to 2MASS positions. This procedure is repeated many times and the resulting false detection rate is 2 3%. − For the 564 X-ray sources covered by SIRIUS imaging, we visually identified 476 infrared counterparts with separations 1′′ from X-ray sources. Of these, 308 are known ≤ 2MASS counterparts that have been matched to ACIS sources. Therefore the combined matches from 2MASS and SIRIUS yield 613 NIR counterparts for the 779 X-ray point sources. Of the 166 X-ray sources without NIR counterparts, 89 lie in the SIRIUS FOV. In spite of its small FOV, an archival HST WFPC2 (F814W) image of the H ii region G353.2+0.9 in NGC 6357 (Figure 3.4; Healy et al. 2004b) was visually examined for optical counterparts. By registering to a 2MASS image, an astrometric correction of ∆RA = 0.9′′ and ∆Dec = 0.4′′ was applied to the HST image. There are 154 X- − ray sources covered by the WFPC2 imaging, 99 of which have HST counterparts. The WFPC2 image demonstrates details of this interface between the hot tenuous H ii region and the cold structured molecular cloud. Some of the optically bright sources are not X-ray emitters and are probably foreground stars; a few stars appear very faint and are probably deeply embedded in the molecular cloud. All but three of the HST counterparts (corresponding to ACIS #359, 375, 390) have matched infrared counterparts. This increases the number of X-ray sources with ONIR counterparts to 616. Thanks to HST’s high resolution imaging, those ACIS sources are resolved from the crowded field where infrared images fail to detect them. Even though optical imaging suffers heavily from extinction, we speculate that 10 20 more HST counterparts for the less-embedded ∼ ′′− ACIS sources with separations 1 from other sources would have been recovered due ≤ to the high spatial resolution, if the HST spatial coverage of Pismis 24 were complete.

3.4.2 Extragalactic Contamination We estimate the contamination from the extragalactic population following the method described in Getman et al. (2006). Monte-Carlo simulations of the expected extragalactic source contamination are constructed by placing fake AGNs randomly across the ACIS-I field superposed on the ACIS background. The X-ray flux distribution follows the logN–logS distribution from Moretti et al. (2003) and the corresponding power law photon index for individual sources follows the flux dependence found by Brandt et al. (2001). Synthetic source spectra were generated with the XSPEC fakeit function and convolved with an absorption column NH . We ran simulations with three typical 22 −2 NH : the total Galactic absorption towards NGC 6357 (NH = 1.2 10 cm , Dickey & × 22 −2 Lockman 1990); Galactic absorption plus an additional NH = 1.0 10 cm typical of × 22 −2 Pismis 24 OB stars; and Galactic absorption plus an additional N = 3.0 10 cm H × typical of reddened 2MASS stars. Then we identified the closest real X-ray detection to the fake source. Source significance for each fake AGN was then calculated using the local background of the nearby real source in the NGC 6357 background map. We set a detection threshold of Signif 2.3, where Signif is the quantity given in column 12 ≥ 35 of Table 3.1. This limiting source significance is the median Signif of those hard X-ray sources without infrared counterparts and lying far off-axis, which are the most likely candidates for AGNs in the field. The number of detectable AGNs from typical simulations with the three absorption column densities ranges from 14 to 29. The level of AGN contamination for the NGC 6357 region (Galactic coordinates l = 353.2, b = +0.9) is similar to the contamination calculated for the Cepheus B (l = 110.2, b = +2.6) observation which had a similar Chandra exposure (Getman et al. 2006). The true extragalactic contamination may be lower due to localized regions of very high absorption by the molecular cloud behind the OB association. We conclude that no more than 4%, and more likely 2%, of the ≃ 779 X-ray sources are extragalactic contaminants. Among possible AGN candidates, the brightest in X-rays is source #1. Based on its flat X-ray spectrum, large separation from the central cluster, non-variable lightcurve, and lack of optical or near-IR counterpart, ACIS #1 is considered to be a likely AGN. Using a power law model with a single absorption component, its spectrum can be adequately fit using a power law photon index 21 2 Γ = 1.4 and an absorption column N = 8.0 10 cm− . H × 3.4.3 Field Star Contamination To estimate possible contamination from foreground field stars, we follow the techniques described in the Chandra Orion Ultradeep Project (COUP) membership study and the Chandra study of the Cep B star forming region (Getman et al. 2005b, 2006). We use simulations based on the stellar population synthesis model of Galactic disk stars by the Besan¸con group (Robin et al. 2003). The models are calculated using their on-line 5 service . Synthetic catalogues of stars generated from these models predict 5600 field 2 ∼ stars with V < 22 mag in a solid angle of 0.08 deg (ACIS-I FOV) to the distance ∼ of NGC 6357. About 10% are F-type main sequence (MS) stars, 20% are G-type MS ∼ stars, 45% are K-type MS stars, 20% are M-type dwarfs; the rest are giants with a ∼ ∼ few subdwarfs. We then adopt the XLFs for MS K-M stars (Schmitt et al. 1995), F-G stars (Schmitt 1997), and late type giants (Huensch et al. 1996) derived from a complete volume-limited sample in the solar neighborhood. The conversion of intrinsic X-ray flux in the ROSAT PSPC 0.1 2.4 keV band to Chandra ACIS-I 0.5 8.0 keV count − − 2 rate is from PIMMS, using a thermal plasma model assuming kT (4πd FX 5.5 −26 0.2 ≈ × × 10 ) /11.34 keV from Güdel et al. (1998), where d is the stellar distance and FX is the X-ray flux in ROSAT PSPC 0.1 2.4 keV band. Only Galactic absorption is assumed − to be present for stars in the foreground of the NGC 6357 cloud and the amount of absorption to the fake star is calculated based on its distance when the simulated catalog is generated. Following the same procedures used in the simulations for extragalactic contamination, a simulated field of 5600 X-ray field stars was added to the ACIS background ∼ map, and the star locations were examined for source significance. For detection thresh- olds ranging from Signif 1.5 to Signif 3.0, 10000 simulation runs give a typical ∼ ∼

5 http://bison.obs-besancon.fr/modele/ 36 number of 6 to 16 detectable foreground field stars. These sources should be readily distinguished from extragalactic AGN and most young cluster members by the brightness of their stellar counterparts and (when sufficient X-ray counts are present to measure median energies) low absorption in their spectra. Eleven individual stars are identified as likely foregound stars in the field ( 3.4.5). Those contaminants will be removed from § further consideration of the young cluster population. The likelihood of background star contamination is small because of the heavy absorption of the molecular cloud and the large distance. The contamination from Galactic disk stars is thus 1 2% of the observed source − population, about half that of the extragalactic contamination. A complete removal of all non-members is impossible without further information, thus we ignore the remaining foreground star and AGN contamination in our further analysis given that the fraction is statistically very small.

3.4.4 Likely New Stellar Members For the 163 ACIS sources without ONIR counterparts, at most 20–30 of them are possible extragalactic contaminants ( 4.2). The rest of the 140 sources are probably § ∼ new members of Pismis 24, in addition to the 605 likely members (excluding the 11 candidate foreground stars) with ONIR counterparts identified from the thousands of ONIR stars in the field ( 4.1). § ′ Sixty-two ACIS sources without ONIR counterparts are clustered within 2 of the O3 star Pis 24-1. The rest are widely distributed throughout the field, although a few are located to the northwest of the ionization front where there is much less optical nebulosity. Five of them are clustered around the WR star. Most of these likely members are faint X-ray sources; only two of them have more than 10 net counts. More than 100 of them have median photon energy E 2.0 keV and 50 have E 3.0 keV, implying h i≥ ∼ h i≥ heavy obscuration (see Figure 8 from Feigelson et al. 2005). We speculate that, due to the face-on geometric configuration, these represent the more distant cluster members that are located behind those optically visible members, deeply immersed in the molecular cloud. Some may be very young, embedded in molecular cores or protostellar envelopes that have not yet been identified. But many are likely to be older Class II and III PMS stars that have counterparts in Spitzer Space Telescope surveys. The properties and spatial distribution of these likely new members will be further discussed in 3.5.2 and § 3.6.5, after we classify the X-ray sources. § 3.4.5 Classification of the X-ray Sample From the results of the previous sections, we can establish a rough classification of the X-ray sources. These class designations appear in Table 3.5. The numbers in parentheses after the class names are the number of sources in each category. R (600)– Reliable X-ray sources with ONIR stellar identifications. They represent the more massive and less obscured members of the stellar population. C (590)– X-ray sources with off-axis angle Θ 5.0′. They are likely members of the ≤ central cluster. 37

H (109)– Hard X-ray sources with E 3.0 keV. Most of these are heavily obscured h i≥ members of the embedded stellar population. About 20-30 of these are probably extragalactic AGNs.

U (18)– Unreliable X-ray detections. We consider the 18 X-ray sources in Table 2 with large probability of being background (log P 1.8) to be possible spurious B ≥− detections.

F (11)– Foreground field stars. Seven soft X-ray sources ( E < 1.3 keV, corresponding −2 h i to log N 22.0 cm ) with more than 20 net counts in 0.5–8.0 keV (to have H ≤ reliable median photon energy measurements) are identified as foreground sources, excluding any known O stars. One X-ray emitting A8 subgiant, HD 157528, was 6 recovered using this criterion . The limitation for this criterion is that unidentified massive stars with soft X-ray emission may be included. Two of these candidate foreground stars, ACIS #9, #539, are included as candiate O stars (see 3.6.2). § In addition, we select 4 soft X-ray sources ( E < 1.3 keV) as likely field stars h i since their JHK colors are consistent with the locus of Besan¸con simulated field stars. All these sources have visual photometry in the UCAC2 catalogue (Zacharias et al. 2004), which provides additional support for them being classified as likely foreground stars.

The classes R, C, and H are not exclusive and many sources fall in two or three categories as illustrated in Figure 3.5. Thirty-eight sources are not classified because they are either too faint or poorly characterized. Moffat & Vogt (1973) listed 15 bright stars in Pismis 24 as their photometry targets and concluded 12 of them are members. Spectroscopic observation confirmed that seven of these, and four additional stars, are massive OB stars (Massey et al. 2001). Our X-ray sample with infrared counterparts increases the known cluster population 50-fold, from 15 to 750, providing the first opportunity to investigate the full stellar ∼ ∼ population of the region. The H K vs. J H color-color diagram of 142 Chandra sources with high-quality − 7 − 2MASS photometry is shown in Figure 3.6. The typical uncertainty in color is 0.06 ∼ mag. The locus for main-sequence stars of different spectral types from early O to late M (Bessell & Brett 1988) is plotted as the thick line. Stars that are reddened by normal interstellar extinction occupy the color space between the two reddening vectors of an O dwarf and an M giant (dashed lines). The foreground A8 subgiant (# 773) clearly reveals itself, lying on the locus with almost no reddening. The stars to the right of the

6 No distance was reported for this star. Using the NIR color-magnitude diagram and the A8IV spectral type, we estimate a distance of d 100 pc. Adopting this distance, the total absorption ∼ −1 corrected X-ray luminosity is log L 28 erg s . This is in line with the observed X-ray t,c ∼ luminosities of subgiant stars of similar spectral types in the ROSAT all-sky survey catalogue of optically bright subgiant stars (Huensch et al. 1998). 7 The 2MASS sources plotted here and in the J H vs. J diagram have signal-to-noise ratios greater than 6 (photometric quality higher than “C−”) and are free of confusion flags and warning of contamination flags. See Explanatory Supplement to the 2MASS All Sky Data Release (Cutri et al. 2003; http://www.ipac.caltech.edu/2mass/releases/allsky/doc/) for details. 38 reddening band are K-band excess sources. It is clearly seen that the X-ray observation selects far more Class III infrared sources (stars that have simple blackbody spectral- energy distributions, with little or no inner accretion disks) than Class II sources (stars with K-band excess circumstellar disks; Lada 1987; Wolk & Walter 1996). The 2MASS J H vs. J color-magnitude diagram (Figure 3.7) gives both the − rough mass distribution and the absorption distribution of the ACIS sources with 2MASS counterparts. Zero age main sequence (ZAMS) track (Cox 2000) and two PMS isochrones (Baraffe et al. 1998; Siess et al. 2000) are shown with spectral types, stellar masses, and reddening vectors for estimation of the dereddened masses and the associated absorption. We assume 1 Myr for the age of Pis 24 cluster and a distance of 2.56 kpc when estimating stellar properties from the color magnitude diagram. We remind the reader that the estimated mass and the amount of absorption for a given star also relies on the assumption of distance and cluster age besides the uncertainties in the star’s color and magnitude. For example, if we adopt a younger cluster age (0.3 Myr), the PMS isochrone will shift upwards and yield a lower dereddened mass and less amount of absorption as shown in Figure 3.7. Note that the inferred properties for stars with spectral type later than B3 are less dependent on the age and distance assumptions. In addition we show the H K vs. K color-magnitude diagram for the same sample of stars accounting for − some small K-band excess. Known Pis 24 optical members detected in X-ray are located along the reddening vectors from the locus of O stars with a typical A 5 mag. Not V ≃ surprisingly, the foreground star #773 resembles a very massive and luminous object at the distance of NGC 6357. Besides the foreground star, WR 93 is the most luminous object. Most of the other ACIS sources occupy infrared color domains that are consistent with being high-mass stars and PMS intermediate-mass stars (M 2M⊙) with 5 < ≥ AV < 15 at 2.5 kpc. The highly-reddened bright 2MASS sources are possibly obscured massive stars. The new high-mass stars double the known high-mass population of the cluster ( 6.2). The intermediate-mass stars, mostly Class III objects, are newly identified § members of the cluster ( 6.3). The ratio between Class III and Class II objects for the § high-mass and intermediate-mass cluster members is 15 : 1 based on the 140 stars ∼ ∼ with mass M 2M⊙. Although the uncertainties of the estimated stellar masses are not ≥ negligible (as much as a factor of 2; see the discussion of Orion stars in Preibisch et al. 2005; Luhman 1999; Hillenbrand & White 2004), this is a quantitative measurement for the massive end of the cluster members and implies that the fraction of stars possessing inner disks among the intermediate-mass and high-mass stars is low even in such a young stellar cluster. Those intermediate-mass stars with significant IR excess are likely the accreting PMS stars known as Herbig Ae/Be stars (HAeBes; Herbig 1960; The et al. 1994). X-ray properties of the intermediate mass stars with infrared excess will be examined in 3.6. A handful of objects are heavily obscured with A > 20. Due § V to their brightness in K-band (K 9 13), these objects, some of which exhibit K- s ∼ − band excess, are easily accessible for ground based follow-up infrared spectroscopic or photometric study. The remaining 170 ACIS members without 2MASS counterparts ∼ but seen with SIRIUS and/or HST are likely lower mass stars. The close separations to the high/intermediate mass stars suggest they perhaps are companions, but this could well be a spatial projection effect in such a crowded cluster. Similar to those unobscured 39

ACIS sources without ONIR counterparts, they are either too faint or too close to bright stars to be cataloged in 2MASS.

3.4.6 EGGs and Protostars Two-thirds of the IR counterparts – those with low photometric quality – do not appear in Figure 3.6. Although candidate protostars may be present, the large uncertainty in their IR color and the upper limits in K-band do not allow us to confidently classify class II/I objects. Precise infrared photometry will be the key to establishing whether they are true embedded protostellar objects. When compared to an archival HST WFPC2 image (Figure 3.4; Healy et al. 2004b), the X-ray source #461 (CXOU J172445.74-341106.9) is spatially coincident with the tip of an evaporating clump extending from the wall of the molecular cloud, which is known as an evaporating gaseous globule (EGG; Hester et al. 1996; Hester & Desch 2005b). It has only 4.6 net counts but the X-ray photons are spatially concentrated; it is a reliable Table 1 source. The X-ray emission appears to be moderately absorbed with E = 2.5 keV. The colors of its infrared counterpart show large NIR excess (J H = h i − 1.06, H K = 1.42). The 2MASS counterpart was previously detected in the Persi et al. − (1986) infrared observation (=IRS 4) and its JHKLMNQ spectral energy distribution (SED) gives a 1–20 µm luminosity of 630 L⊙ (Persi et al. 1986). This luminosity and its infrared colors are consistent with that of a protostar with spectral type B0–B2 (Persi et al. 1986; Bohigas et al. 2004). In addition, it also coincides with the compact radio peak of a 6 cm VLA observation (=VLA A; Felli et al. 1990). CXOU J172445.74-341106.9 may be the first detection of X-ray emission from an EGG, which represents one important evolutionary stage of star formation. Linsky et al. (2003) identified deeply embedded hard X-ray sources in the “pillars of creation” of the Eagle Nebula (M 16; Hester et al. 1996) but were unable to find any such sources associated with the EGGs. The non-detection of X-ray emission in M 16 EGGs may be related to their high threshold of detection ( 6 counts), heavy extinction, or the ≥ spectral type and evolutionary phases of the embedded stars. Although the uncertainty in the X-ray luminosity is potentially very large for such a faint source, the estimated 1 L 30.1 ergs s− from PIMMS is somewhat consistent with other X-ray detected high- h ∼ mass protostars (NGC 2024 IRS 2b, O8–B2, Skinner et al. [2003]; S106 IRS4, O7–B0, Giardino et al. [2004]). It is likely that the actual X-ray source is the embedded protostar instead of the gaseous globule. No X-ray detection is associated with the HST jet candidate, Pismis 24 # 14 in De Marco et al. (2006). We examined the X-ray image at the HST position and no photons can be found within a 90% PSF contour.

3.5 Properties of the Stellar Cluster and Its Environment

3.5.1 X-ray Luminosity Function We construct the observed hard band (2–8 keV) XLF for all X-ray sources, excluding sources classified as U and F . Based on the fact that the extragalactic contamination is small, we only exclude ACIS # 1 that is very likely an AGN. We do not attempt to 40 study the observed total band (0.5–8.0 keV) XLF or absorption corrected total band luminosities (Lt,c; 0.5–8.0 keV), since we do not have reliable measurements of NH for faint sources. In the case of heavily absorbed X-ray sources, the unknown soft component can introduce a large uncertainty in the observed total band X-ray luminosity. For sources with more than 20 net counts, the hard band (2.0–8.0 keV) flux Fh and absorption-corrected hard band flux Fh,c (and therefore the luminosities Lh and Lh,c) are available from XSPEC spectral fitting. For fainter sources, we define the following procedure to estimate Fh and also Fh,c consistently. First, if the faint X-ray source has a 2MASS counterpart with reliable photometry, we use the AV derived from the color magnitude diagram to obtain the NH , assuming a 1 Myr age for the cluster members 21 2 1 N /A = 1.6 10 cm− mag− (Vuong et al. 2003). We note that adopting a younger H V × age or a smaller distance will shift the PMS isochrone upwards in the color magnitude diagram, causing generally lower values for the derived AV . However for these faint sources that are likely low-mass cluster members, the decrease in A is very small, 1 V ∼ mag (Figure 7). Otherwise, as shown in Feigelson et al. (2005), empirical conversion from median photon energy E provides us with an effective approximate log NH value h i 2 −2 for sources with only a few counts: log NH = 9.96 + 13.62 E 3.86 E cm × h i− × h i −2 for relatively soft sources (1.0 E 1.7 keV); log N = 21.22 + 0.44 E cm for ≤ h i≤ H × h i harder ones ( E > 1.7 keV). We estimate that the uncertainties in log N are 0.3 h i H ∼ ± and can be as large as 0.8 for soft sources that have E 1.3 keV. We assign plasma ± h i≤ temperature kT based on E to reflect the observed trend of increasing temperature h i with increasing median energy: typically kT = 1 keV for E < 1.3 keV, kT = 2 keV h i for 1.3 E 2.0 keV, kT = 3 keV for 2.0 E 4.0 keV, and kT = 5 keV for ≤ h i ≤ ≤ h i ≤ E 4.0 keV (see Figure 8 in Feigelson et al. 2005). With the N and kT parameters h i≥ H set and ACIS-I count rates known for each source, we run PIMMS simulations to derive the observed (absorbed) and absorption corrected flux in the hard band. The XLFs for the lightly obscured ( E < 3.0 keV) and heavily obscured ( E h i h i≥ 3.0 keV) stellar populations in NGC 6357 are plotted in Figure 3.9. For comparison, XLFs derived for the lightly obscured COUP Orion Nebula population (Feigelson et al. 2005) and for the Cep B unobscured and obscured populations (Getman et al. 2006) are also shown. The shapes of the high luminosity portions of the lightly obscured and obscured COUP XLFs (see Figure 3 in Feigelson et al. 2005) are very similar, but the weaker sources in the obscured COUP sample are missing. Therefore we adopt the lightly obscured COUP XLF as the template in this population calibration exercise. At the high LX end, the NGC 6357 XLF is clearly consistent with the power law relation as seen in COUP and Cep B, allowing us to scale to the COUP XLF and make the first reliable estimate of Pismis 24’s total cluster population. 1 At log L 30.3 ergs s− , the NGC 6357 XLF begins to flatten and decline h ∼ dramatically. We examine the completeness in the source detection to determine whether the decline is due to a real deficiency of cluster members in this luminosity range or is simply an expected effect when reaching the faint source detection limit. If the observed X-ray flux from a cluster member with such a luminosity is well above our sensitivity limit, the deficiency of stars in this luminosity range would imply a different X-ray luminosity distribution for the population. Given that sources with 10 net counts in the full-band at any off-axis angle are confidently detected, we adopt two different methods 41 to estimate what is the corresponding flux limit to 10 net counts. First, assuming a 2 −1 −2 keV thermal plasma, PIMMS simulation gives log Fh = 14.6 ergs s cm for a count −1 − rate of 10/38 counts ks . Independently, using all sources with more than 20 net counts that have flux estimates from spectral fitting, we derive an empirical conversion factor between flux and net counts for this observation, and the 10 net counts also converts −1 −2 to log Fh 14.6 ergs s cm . Therefore the two approaches give the same Fh, ∼ − −1 which converts to log Lh 30.2 ergs s . The decline in the NGC 6357 XLF below −1 ∼ log L 30.3 ergs s is thus consistent with incomplete detection of the less luminous h ∼ cluster members. We do not have sufficient sensitivity to search for a difference in XLF 1 shape compared to COUP below log L 30 ergs s− , as seen in Cep B by Getman et al. h ∼ (2006). The scaling factor to match the COUP XLF to the NGC 6357 XLF is 5 6 at ∼ − higher values of log Lh for the lightly obscured population. Using the XLF constructed with Lh,c gives the same scaling factor. The XLF from Lh for the obscured stellar population in NGC 6357 is comparable to the COUP XLF, while the absorption corrected XLF seems to be consistently 5 times higher than the COUP XLF for the high ∼ luminosity bins. Altogether, this implies that the X-ray emitting stellar population 8 in NGC 6357 is roughly 5 10 times richer than the lightly obscured Orion Nebula − Cluster (839 unobscured cool stars; Feigelson et al. 2005). Therefore the X-ray emitting population in Pismis 24 is 5000. Given that the ONC has 2000 known members ∼ ≃ from the deep optical study by Hillenbrand (1997) and COUP detects more than half of the stellar population in X-rays, the total stellar population in Pismis 24 is estimated to be two times larger than the population detected in X-rays, 10, 000 members. It ∼ is not unexpected, assuming the standard IMF, to find massive stars with > 100M⊙ in a cluster with such a large population. A note of caution here is that, even in such a young star-forming region, the stellar population in consideration is not perfectly coeval. It perhaps includes a younger population triggered to form by the ionizing front from the massive stars, but the small population of the second generation stars has little impact on the IMF of this rich cluster.

3.5.2 Spatial Distribution of the Stellar Cluster Aside from several dozen known and new OB stars (see 3.6), the locations of § over 700 new lower-mass cluster members are now determined with sub-arcsec precision. Figure 3.10 shows the spatial distribution of the cluster members. The large scale optical nebulosity contour from the DSS R-band image, which also traces the mid-IR ring-like morphology, is shown together with the outlines of the ionization front in yellow and the CO “South-Eastern Complex” in green (S.E.C. in the CO emission map in Massi et al. 1997). The geometry of the cluster appears spherically symmetric: the source density in the cluster core near the ACIS-I aimpoint is the highest, and declines radially with no apparent discontinuity. There seem to be fewer X-ray detections northwards of

8 If the true distance to the cluster is a smaller value, e.g. 1.7 kpc, all the derived luminosities, including log Lh and log Lh,c, will be systematically lower by 0.35 dex and the estimated population will be 3 times the Orion Nebular Cluster. ∼ 42 the optical nebula, although there is neither an enhancement nor a deficit of stars at the bright photodissociation region (PDR). Although the reduced sensitivity at the chip gaps between the ACIS-I CCDs is not taken into account, we doubt that the number of missing sources in the chip gaps is large enough to alter the spatial distribution of the stellar cluster. 2 Projected stellar surface density (number of stars arcmin− ) contour maps for the unobscured and obscured populations are shown in Figure 3.11, in logarithmic scale. The stellar density is smoothed using a 0.5′ radius kernel. As expected, the highest concentration of stars is at the core of the cluster, where the massive O stars are located. The distribution of stars is rather spherically symmetric centered at the O3 stars, except for a density enhancement centered at the tip of the S.E.C. (centered at RA=17:24:48.3, Dec=-34:15:05). Figure 3.12 further elucidates the structure of the stellar distribution by showing the radial profile of the stellar surface density. The stellar distribution appears to consist of two components: a compact cluster within the central 1.5 pc with a log-normal distribution, and a cluster halo extending to 8 pc with an approximately exponential distribution. The number of X-ray stars in the central region is 333 while the extended region contains 446 stars. Because these stars are poorly studied, we can only speculate that the stars in the central region represent the coeval dense cluster core of Pismis 24, while the apparent extended wing is a mixture of evaporating cluster members and newly formed stars distributed throughout the molecular cloud enveloping the cluster. Figure 3.12 also compares the radial profile to the profile of the Orion Nebula Cluster within 1.5 pc for Chandra stars (Getman et al. 2005a; Feigelson et al. 2005). ∼ 1 The ONC is only extensively studied within a projected radius of 2.5 pc from θ Ori C ∼ (see Hillenbrand & Hartmann 1998 for a review of ONC structure and dynamics). It is strongly concentrated, with a central density > 200 times that of the widespread cloud stellar population distributed at radii > 2.5 pc. The ONC could possess an extended halo of stars as we see in the more massive Pismis 24 cluster but it would be confused with the surrounding stellar population. Thus we limit the comparison of the two X-ray selected clusters to radii <1.5 pc. The stellar density of the ONC core appears to be 5 times higher than the X- ∼ ray population in the Pismis 24 cluster core, but this can be largely attributed to the difference in the number of faint stars detected: the ONC sample from the 800 ksec COUP observation extends into the brown dwarf regime, while the Pismis 24 sample becomes incomplete around M 3M⊙ ( 5.1). Extrapolating to brown dwarf masses ≤ § using the COUP mass distribution (Getman et al. 2005a), we estimate that the NGC 6357 cluster central density would be 20 times higher than the currently observed one ∼ if it were as sensitive as the COUP observation. We thus conclude that the NGC 6357 cluster central density is probably 4 times higher than the ONC central density and ∼ similar to the central density of the NGC 3603 super star cluster (see footnote 1 in Hillenbrand & Hartmann 1998). Figure 3.13 zooms in to the core region of the cluster using the SIRIUS K image as background. The ONIR identified X-ray sources (circles) and unidentified X-ray sources (crosses) are shown with the known massive members highlighted in white. Only a few of the unidentified sources appear isolated, without infrared sources nearby. Most are 43 clustered in the 15′′ neighborhood of the two O3 stars; 24 are located 1.5 2′′ from ∼ − near-IR sources, therefore the available NIR images may not be able to resolve them. These unidentified sources very likely are real low-mass members concentrated around the massive stars.

3.5.3 The Morphology of the Ring-like Nebula A puzzling feature arises when we compare the core of the massive cluster centered on the two O3 stars to the large ring-like nebula with a large cavity that makes up the wider view of NGC 6357 (Figure 1): the cluster is significantly off-center to the north. The “diamond ring” morphology of NGC 6357 is apparent in the Hα and mid-IR images, with the bright point along the edge brightened torus. In contrast, the Rosette Nebula has a similar annular nebula morphology, but the massive OB stars in NGC 2244 exciting the nebula are located right in the center. The hollow HII morphology is attributed to strong OB stellar winds excavating molecular materials and the depletion of gas by newly formed stars (Dorland et al. 1986; McCullough 2000; Townsley et al. 2003). Therefore the nebular shape is generally accepted as spherical or cylindrical. In NGC 6357, the most massive members are so close to the northern edge of the nebula that it seems implausible that the cavity is created by the strong winds of the current generation of hot stars. In general, supernovae explosions and strong stellar winds can shape the natal cloud by creating wind-blown “bubbles”–cavities. However, there is no evidence for a recent supernova: neither diffuse X-ray emission from a hot tenuous plasma nor non- thermal radio emission is detected in this region. We further note that the presence of WR 93, an evolved star, is circumstantial evidence that some population of massive stars in NGC 6357 is old enough to produce supernovae. It is possible that a previous-generation of hot stars created the large cavity and formation of the current ionizing population was triggered at the densest edge of the excavated cavity. Combined with different viewing angles, this can result in diverse morphologies such as an egg, or a diamond ring, rather than a symmetrical annulus. Hα images of quite a number of regions where the stars are displaced with respect to the center of an evacuated cavity have been shown in the Large Magellanic Cloud (Gaustad et al. 1999) and modeled as modified Strömgren spheres (McCullough 2000). The key for testing this hypothesis is to identify the older cluster. X-ray sampling provides a critical test as X-ray luminosities decay only slightly over 10 Myr (Preibisch & ∼ ′ Feigelson 2005). The subcluster seen in the surface density map 4 southeast of the ∼ Pis 24 cluster could be the remnant core of the previous generation of stars. ONIR photometric and spectroscopy study of the 30 stars in this subcluster, permitting their ∼ placement on the HR diagram, might reveal whether they are older than the main cluster concentrated off the center of the nebula. An alternative scenario is that the cavity is indeed created by the massive Pis 24 stars that are emitting ionizing photons. In that case, the displacement from the center could be explained with spatial projection and an inhomogeneous molecular cloud, as first proposed by Bohigas et al. (2004). The O stars were close to the geometrical center of the Pismis 24 cluster and created an expanding HII bubble, which may have encountered a much denser interstellar medium in the northern part than in other directions. The 44 apparent deficit of X-ray sources to the north may be evidence for the absence of star formation in the densest region, where a new generation of stars may emerge eventually.

3.6 X-rays across the Mass Spectrum

3.6.1 X-rays from known massive stars OB stars have been known to be X-ray sources since early observations from the Einstein satellite (Harnden et al. 1979; Seward et al. 1979). Models were developed where instabilities in radiatively-driven stellar winds from massive stars produce shocks and heat the gas to X-ray emitting temperatures (Lucy & White 1980; Owocki et al. 1988; Owocki & Cohen 1999). However, recent studies indicate that more complex models are needed to account for unexpected X-ray emission line profiles and hard, variable continuum emission (e.g., Waldron et al. 2004; Stelzer et al. 2005). We detect X-ray emission from all known early type stars ranging from O3 to B0.5 in this 38 ks ACIS observation. Table 3.6 summarizes ACIS detections and non- detections of X-ray emission from O and early B stars in the Pismis 24 cluster that are classified in literature. Two of the newly classified O3 stars Pis 24-1 (O3.5If) and Pis 24-17 (O3.5IIIf; 33 1 Walborn et al. 2002) exhibit X-ray luminosities L 10 ergs s− . This is consistent t,c ∼ with HD 93128 in the Carina Nebula cluster Tr 14, which has similar spectral type (Evans et al. 2003b). Their lightcurves remain constant through the observation, which is typical for O stars, although variability in X-ray emission from O stars like ζ Ori is well known (Berghofer & Schmitt 1994). Figure 3.14 presents their X-ray spectra and spectral fits (assumed abundances of 0.3 Z⊙ unless otherwise noted). Compared to the best fit using a single temperature plasma model, the X-ray spectrum of Pis 22 2 24-1 (#344) is better fit by a two temperature plasma model (N = 1.3 10 cm− , H × kT1 = 0.5 keV, and kT2 = 1.7 keV). Pis 24-17 (#420) can be adequately fitted by a 22 2 single soft component with N = 1.4 10 cm− and kT = 0.7 keV. The fit is improved H × by adopting enhanced Mg (0.6 Z⊙), Si (0.6 Z⊙), and Ar abundances (2.7 Z⊙). × × × The Si xiii line at 1.86 keV in Pis 24-17 is exceptionally strong, comparable to the same line seen in the O star IRS 2 in RCW 38 (Wolk et al. 2002). The derived temperatures for soft and hard components for these O stars are similar to the X-ray temperatures of other single massive stars: some O stars exhibit a simple single temperature plasma 1 keV (e.g., Macfarlane et al. 1993; Rho et al. 2004; Townsley et al. 2006a), while a ∼ few require an additional high energy component (e.g., Corcoran et al. 1994; Kitamoto & Mukai 1996; Evans et al. 2004; Stelzer et al. 2005). Gagnéet al. (2005) shows that the magnetically channeled wind shock model (Babel & Montmerle 1997b,a) with strong line-driven winds in a single O star can adequately reproduce both the soft and the hard 1 compents in the recent phase-resolved Chandra grating spectra of θ Ori C (O5.5V). In some cases, the high energy component may in fact be an indication of binarity, as powerful winds in two massive components shock to produce hard X-rays (e.g., Portegies Zwart et al. 2002; Albacete Colombo et al. 2003; Townsley et al. 2006b). The brightest X-ray point source in the ACIS field of view is the WC7+O7-9 33 1 binary WR 93 (#747) with a L = 1.6 10 ergs s− , roughly consistent with the t,c × 45

33 1 earlier ROSAT PSPC value L = 1.25 10 ergs s− (0.2–2.4 keV; Pollock et al. 1995). X × It is one of the spectroscopic WR binaries that are bright enough to be detected by previous generation X-ray telescopes; single WC stars are much fainter (Oskinova et al. 2003). Thus the strong X-ray emission from binary WR stars is likely generated in the region where the massive stellar winds collide. However, while archetype colliding-wind binaries such as WR 140 (WC7+O4V) display characteristic non-thermal emission in the radio band (Dougherty et al. 2005), WR 93 shows thermal radio emission (Abbott et al. 1986). Also the X-ray lightcurve appears to be constant through the entire observation; 9 no orbital X-ray variability is seen . The X-ray spectral fit for WR 93 is shown in 21 2 Figure 3.14 with an absorption column N = 7.4 10 cm− and a rather hard kT = 2.4 H × keV. The large absorption column densities towards the two O3 stars and the WR star derived from X-ray spectral fits match well with their large visual extinction (A 5 7) V ∼ − and the reddening from JHK colors. Five of the bright Pismis 24 stars in Moffat & Vogt (1973) – Pismis 24-4, 7, 8, 9, 11 – were not included in the spectral classification observation of Massey et al. (2001). Based on their strong X-ray emission (except for Pis 24-9 which is undetected), absorption column derived from X-ray spectral fitting, UBV brightness, and JHK colors, we suggest that they are also young OB stars and cluster members (see Table 3.7). X-ray emission from OB stars was reported from ROSAT observations to display a 7 characteristic efficiency L /L 10− (Chlebowski et al. 1989; Berghoefer et al. 1997), X bol ∼ and this empirical ratio has been examined in recent X-ray observations of massive stars (e.g., Corcoran 1999; Evans et al. 2004; Stelzer et al. 2005). From the COUP observation, the emission ratio for late O and B stars is found to scatter around the canonical LX /Lbol value by three orders of magnitude. However, as they cautioned (Stelzer et al. 2005), this may not represent the most luminous early O stars due to the absence of spectral types earlier than O7. Therefore the Pismis 24 cluster, with two O3 stars, is particularly valuable for assessing this relation in a single massive cluster. Using Mbol reported in 10 Massey et al. (2001) , Figure 3.15 shows the LX –Lbol relation for our detections of cataloged O and B0 stars together with the strong wind sample and weak wind sample from the COUP study (Stelzer et al. 2005). The LX /Lbol ratio for O stars in the Pismis 24 7 cluster is consistent with the 10− value with an order of magnitude scatter. Although ∼ there are only 9 data points and one upper limit (Lbol for WR 93 is not well determined due to the poorly characterized bolometric correction for late-type WC stars; Massey et al. 2001), it appears that the LX /Lbol correlation has less scatter for stars of the

9 No orbital period for WR 93 is reported in literature. 10 It should be noted that Heap et al. (2006) have proposed a new calibration between the spectral type and eﬀective temperature of O stars, based on analysis of high resolution HST, FUSE, and optical spectra. The spectral calibration in general use is by Vacca et al. (1996), which Massey et al. (2001) used to derive bolometric luminosity for Pismis 24 O stars. The derived log Lbol using the Heap et al. (2006) calibration is 0.4 dex lower than log Lbol using the Vacca et al. (1996) value for an O4V star, and 0.15 dex lower for an O8V star. Here we use the new bolometric corrections from Heap et al. (2006) to derive bolometric luminosities for Pismis 24 stars. The bolometric correction value for the two O3.5 stars is not provided by Heap et al. (2006), and we adopt 0.6. COUP data points are from Stelzer et al. (2005) and originally derived from Hillenbrand− (1997). 46 earliest spectral types and the physics behind this relation is likely to be physically related to stellar wind properties.

3.6.2 Newly-discovered Candidate O Stars We assembled a list of candidate O stars in Table 3.7 to facilitate future spectroscopic follow-up via two approaches. Table 3.7a lists the 13 sources brighter than 80 counts (to guarantee a reliable X-ray spectral fit) with 0.5–8.0 keV absorption corrected −1 X-ray luminosities log Lt,c 32.0 ergs s as potential new high-mass stars, since PMS ≥ 32 1 stars rarely reach X-ray luminosities 10 ergs s− (Favata & Micela 2003; Getman ≥ et al. 2005b). Then we examined their X-ray light curves and observed optical and infrared properties to exclude possible AGNs. Since certain O stars are observed to exhibit powerful flares (Feigelson et al. 2002a), we do not exclude the X-ray luminous sources that may be extraordinarily bright flaring PMS stars. Their lightcurve variability is noted in Table 3.7. O stars may reveal themselves as luminous in infrared but appear relatively faint in X-rays due to heavy obscuration in their early evolutionary stages. For example, an −1 O9V star emitting log Lt,c = 32.0 ergs s behind 15 mag of visual extinction would have K = 10 and 40 counts in our ACIS image when observed at a distance of ∼ 2.5 kpc (assuming a kT 0.6 keV and without considering a possible K-excess). We ∼ selected bright near-IR sources (K 10.0 mag) detected in the X-ray image to examine s ≤ whether their JHK colors and X-ray properties are suggestive of obscured early type stars. This K brightness selection is used in Hanson et al. (1997) when they select the spectroscopy sample of massive YSOs in M 17, and we further adapt it to X-ray emitting samples. All 9 known members with spectral type earlier than B0.5, the WR+O7 binary, and the foreground A8IV star were recovered using this critieria. Table 3.7b gives 11 additional candidates that have colors consistent with early-type stars and are not known foreground stars. Figure 3.7 and Figure 3.8 convincingly suggest that highly reddened bright infrared source #654 and #694 are indeed very obscured early O stars with some IR excess. Source #140 is almost certainly a high-mass star since it qualifies using both the IR and the X-ray selection criteria. As a cautionary note for this method, the presence of K-band excess may mislead us to include some accreting intermediate-mass stars as candidate O stars. Given the uncertainties in distance, age, and absorption, readers are cautioned that ONIR spectroscopy on these objects (Table 7) should yield more appropriate classifications for them (e.g., Massey et al. 2001; Walborn et al. 2002). The X-ray sample suggests that the optical sample of Massey et al. (2001) may have significantly underes- timated the O star population. However, using a smaller distance of 1.7 kpc will result in a 0.35 dex decrease in the derived X-ray luminosity. Some of the X-ray selected candidates may indeed be lower mass flaring PMS stars, which are readily distinguishable from their ONIR spectra. If a large fraction of these 24 candidate O stars are confirmed, the OB population of Pis 24 (Table 6) is doubled to tripled. 47

On the top right corner of the infrared color-magnitude diagram (Figure 3.7 and Figure 3.8), there is another mysterious bright source, #19 (CXOU J172413.60- 341456.7), whose color and brightness (J H = 2.006, H K = 0.991, K = 5.334) sug- − − gest that it is a very luminous object subjected to signiﬁcant absorption (A 15 20). V ∼ − It has 17 net counts in X-ray and the median photon energy is 2.5 keV, which is consistent with being obscured and not a foreground star. However, if the star is located at 2.5 kpc, the intrinsic K-band magnitude and infrared luminosity will be exceptionally luminous. Thus it is plausible that source #19 is a massive YSO or a young post main-sequence star like WR 93. Spectroscopy is highly warranted and should reveal the nature of this unusually luminous but obscured star.

3.6.3 Intermediate Mass Stars The large population of the Pismis 24 cluster offers an exceptionally rich sample of intermediate mass stars. Of particular interest are those with infrared excess, possibly due to the presence of protoplanetary disks. We note that the estimated stellar masses from the infrared color-magnitude diagrams rely on the uncertain age and distance, which are determined with improved precision in recent literature. The ambiguity is more severe among the early B stars (Figure 3.7; for example, the inferred stellar mass for a 14 M⊙ star from the 1 Myr isochrone can be as low as 4 M⊙ when estimated from a 0.3 Myr isochrone. However, the X-ray selected sample is important for future follow-up for this poorly studied region. We list in Table 3.8 100 X-ray selected candidate intermediate- mass stars, which have dereddened spectral types A0 B0 estimated from the NIR color- − magnitude diagram (Figure 3.7 and Figure 3.8) assuming 1 Myr for age and 2.56 kpc for distance. Optical spectroscopy is needed to give accurate spectral classifications. Among them, four stars exhibit significant K-band excesses in the NIR color-color diagram (Figure 3.6), which are good candidates for HAeBe stars. Hα spectroscopy on this sample might further reveal the existence of accretion. Their X-ray luminosities are 1 on the order of log L 30.0 31.5 ergs s− , consistent with previous ROSAT PSPC X ∼ − (0.1-2.4 keV) observation of HAeBe stars (Zinnecker & Preibisch 1994). One interesting result has been mentioned in 4.5 – the available IR photometry § data suggest that the ratio between the number of Class II and Class III PMS stars with mass > 2M⊙ is low, 1 : 15. Over the mass range 2 16M⊙, the ratio between the ∼ − number of Class II and Class III PMS stars is even lower, 4%. It is important to note ∼ that such a ratio is derived from X-ray selected stars, without any prior knowledge of disk indicators. The low fraction of the cluster members possessing disks is rather surprising since apparently these X-ray emitting PMS stars are very young (in the company of O3 stars). However, the low disk frequency is consistent with the previous studies in NGC 6611 by Hillenbrand et al. (1993). They find that optically thick circumstellar disks are already rare among the intermediate-mass PMS stars with ages less than 1 Myr and suggest the disk lifetimes are much shorter for the massive stars than those of solar type stars. It seems also consistent with the suggestion that disk evolution happens rapidly in clusters based on recent Spitzer observations (e.g., Hartmann et al. 2005; Sicilia-Aguilar et al. 2006; Lada et al. 2006). The drastic radiation environment may play a bigger role here in the dissipation of disks, rather than mass loss via accretion. Further Spitzer 48 study on this statistically significant sample will provide constraints on the timescale of the circumstellar disk dissipation and the formation of planetary systems.

3.6.4 Flaring PMS Stars There are 31 sources identified as highly variable (P 0.005; variability flag “c” KS ≤ in Table 3.1 and Table 3.2); their lightcurves exhibit flaring activity as most frequently seen in PMS stars when magnetic reconnection events occur in their coronae. Eleven of them have more than 80 net counts (0.5–8.0 keV), so we can reliably derive their X-ray luminosities through spectral fits. Seven of the flaring sources exhibit luminosities above 1 log L 32.0 ergs s− , which is exceptionally high for PMS stars in their flaring phase t,c ≥ (Feigelson & Montmerle 1999; Favata & Micela 2003). One extraordinarily intense flare is seen in source #672 (CXOU J172457.87- 341203.9), comparable to some of the strongest X-ray flares known in PMS stars (Grosso et al. 2004; Favata et al. 2005; Getman et al. 2006). The time-energy diagram, binned lightcurve, and X-ray spectrum for #672 are shown in Figure 3.16. In contrast to the typical impulsive fast rise phase of 2 hr observed in the X-ray flares of PMS stars, this flare shows a 4 hr rising phase to reach its highest flux and still remains at the peak ∼ level at the end of our observation. This is similar to the powerful flare seen in the non-accreting PMS star LkHα 312 in the Orion B cloud (Grosso et al. 2004). The count rate is 15 times higher during flaring compared to the quiescent level before flaring. ∼ The spectrum is hard ( E = 3.0 keV), as seen in many PMS stars in COUP during h i flares (Favata et al. 2005). The spectral fitting gives a rather hard kT > 10 keV, corresponding to thermal plasma of > 120 MK. The infrared countpart for this X-ray source shows K = 10.2 and its location in the color-magnitude diagram suggests a high-mass star. The SIRIUS images were visually examined and there is a possibility that the flare originates from a poorly resolved low mass companion to the high-mass star. If a future spectrum confirms that it is an early type star, the flaring behavior is similar to 2 that observed in the Orion Nebula O9.5 star θ Ori A (Feigelson et al. 2002a), either from an extraordinary magnetic reconnection flare in a low mass companion, or from non-standard wind shocks in the massive stellar winds (Feigelson et al. 2002a; Stelzer et al. 2005).

3.6.5 X-ray Selected Deeply Embedded Population In Table 3.9 we assemble a sample of 16 heavily obscured objects. These are sources that have E 3.0 keV (with 20 net counts to be reliable) or log NH,X 22.5 2 h i≥ ≥ ≥ cm− derived from spectral fitting. This subsample of the hardest sources (classified H in Table 5) likely contains deeply embedded protostars in addition to older stars heavily obscured by molecular material. Their spatial distribution is shown in Figure 3.17, overlaid on a DSS optical image. Several of them appear to be clustered in the optical/infrared dark column (S.E.C. in the CO map) clearly seen in visible and 2MASS images (Figure 1). Others are widely distributed along the rim of the ringlike nebula. A plausible explanation is that the nebulosity traces the rim of a bowl-shape shell facing us, thus the column density in our line-of-sight to those objects is much higher. This 49 geometric configuration is consistent with the previous conclusion that G353.2+0.9 is a blister HII region viewed face-on (Massi et al. 1997).

3.7 Diﬀuse X-ray Emission

With two of the most massive stars known and a luminous WR star in the field, it is reasonable to expect diffuse X-ray emission in NGC 6357, from strong stellar winds and parsec-scale wind-wind collisions (Townsley et al. 2003). A more recent study of X-ray diffuse emission in the 30 Doradus star-forming complex is reported in Townsley et al. (2006b), which provides our technical procedures to extract diffuse emission. From the point source locations and extraction regions (Table 1 and 2), we can mask the point sources and extract the spectrum of possible remaining diffuse emission. Two regions of different scales are used for extraction, which yield consistent measurements of the emission. They are outlined by the white and the cyan polygons in Figure 3.2b. The background spectrum was scaled by the effective area and by the ratio of the geometric areas to be appropriate for the region of diffuse emission being fit; see Section 5.1 of Townsley et al. (2003) for details. The spectra can be fit using a soft optically thin ther- −2 mal plasma with kT 0.6 keV and log NH 22.1 cm . The corresponding observed ∼ ∼ −1 −1 soft band X-ray luminosity is log L 32.4 ergs s (log L 33.6 ergs s if corrected s ∼ s,c ∼ for absorption). This is comparable to the observed diffuse X-ray emission in M 17 and the Rosette Nebula (Townsley et al. 2003). However, the detected emission does not necessarily arise from wind-generated plasma. Under the assumption that the gas temperature within the cluster is relatively 7 constant and hot (> 10 K), Stevens & Hartwell (2003) show that the luminosity level of the expected diffuse emission in a stellar cluster can be estimated using a scaling factor (equation 10), which Getman et al. (2006) adopt to evaluate the nature of X-ray diffuse emission in the Cepheus B region. We follow the same method to obtain a crude 11 estimate of diffuse emission from the massive star winds . Two O3 stars dominates the cluster core and two parameters are needed to predict the X-ray luminosity: the total stellar mass loss rate M˙ ∗, and the mean weighted terminal velocity for the stars in the cluster v∗. For the O3 star Pis 24-1, Benaglia et al. (2001) derived a terminal −1 wind velocity of v∞ 2295 km s and an upper limit for the mass loss rate of M˙ −6 −1 ∼ ≤ 4.5 10 M⊙ yr . No measurements for Pis 24-17 is found in the literature, and we × assume v∞ and M˙ are comparable to Pismis 24-1. Adopting these wind parameters, and an unrealistically optimistic 100% efficiency for conversion from stellar wind to thermalization of the cluster, and 2 pc for the radius of the cluster core, we estimate ∼ 31 −1 the diffuse X-ray luminosity LX < 10 ergs s (equation 8 and 10). This is apparently insufficient to account for the observed luminosity. An alternative explanation is that the diffuse emission is mainly the contribution from unresolved X-ray emission from thousands of low mass PMS members in Pismis 24. Assuming that Pismis 24 and the ONC have identical IMFs with identical XLFs, the only

11 The WC+O star WR 93 is not in the cluster core region, thus we do not include it in the calculation. For reader’s reference, the mass loss rate and the terminal wind velocity for WR 93 is given in 2 § 50 difference here is that Pismis 24 cluster has 5 times the total known population of the ONC (see 5.1). We focus on the lightly obscured populations since they are less affected § by the different amount of absorption along the two sightlines. Scaling from COUP (Feigelson et al. 2005), the total expected soft-band X-ray luminosity is log Ls,c 33.8 1 ∼ ergs s− from the entire Pismis 24 population and the actual observed soft luminosity 1 from resolved cluster members is log L 33.2 ergs s− (since absorption correction to s ∼ individual stars is infeasible, we consider this to be a lower limit to the intrinsic soft-band luminosity Ls,c). Thus the integrated X-ray emission from the unresolved X-ray emitting 1 PMS population in Pismis 24 is estimated to be log L 33.7 ergs s− , after deducting s,c ≤ the observed Ls from the expected Ls,c. This is fully consistent with the absorption- 1 corrected soft diffuse X-ray luminosity log L 33.6 ergs s− . Therefore we suggest s,c ∼ that most of the diffuse emission is probably the combined contribution from individually undetected X-ray-faint PMS cluster members. However, the spatial distribution of the extended X-ray emission traces the cavity seen in optical and infrared (in particular, dust emission seen in MSX image; see Figure 3.1). No extended emission is seen towards the dark column. This may suggest the presence of diffuse emission is real. We further note that X-ray emission is clearly present on the ACIS-S chips. Sev- eral point sources are detected, with very large PSFs due to large off-axis angles. The X-ray emission captured at the lower edge of the S-array field corresponds to the location of the H ii region G353.2+0.7 (see Figure 2b and Figure 3.1f). But due to the poor PSF, no further conclusion can be drawn on whether the emission is diffuse in nature. Nevertheless, this detection confirms the ROSAT detection of G353.2+0.7 and implies that it contains lots of X-ray-emitting young stars.

3.8 Summary

This 38 ks ACIS observation of the NGC 6357 field has provided the first high spatial resolution X-ray image of this massive star-forming complex. We summarize the main results of our study as follows: 30 1. We detect 779 X-ray sources with a limiting X-ray sensitivity of 10 ergs −1 ′′∼ s . There are 445 ACIS sources matched to 2MASS point sources with 2 separation. ≤ Further visual inspection of an archival HST image and SIRIUS JHK images yield a total of 616 ONIR counterparts. We estimate the extragalactic contaminants via a careful simulation of detection efficiency of the background AGN population, and conclude that no more than 4% of our detections are AGNs. Similarly we adopt the Besan¸con stellar population sythesis model of the Galactic disk and suggest 11 stars as foreground objects based on their X-ray softness and NIR color. Excluding 20 X-ray ∼ sources without ONIR identifications that are possible AGNs, the rest of the 140 X-ray ∼ sources without ONIR counterparts are likely new Pismis 24 members that are clustered around the massive stars and/or deeply embedded in the cloud. The X-ray detected population provides the first deep probe of the rich population of this massive cluster, increasing the number of known members from optical study by a factor of 50. ∼ 2. On the NIR color-color diagram (Figure 3.6), it is clear that ACIS detects more Class III sources than Class II/I sources. In the color-magnitude plot (Figure 3.7), the locations of most of the ACIS sources are consistent with being high mass stars 51 and PMS intermediate mass stars (M 2M⊙), subjected to a visual extinction of ≥ 5 < AV < 15 at 2.5 kpc. The new high mass stars, if spectroscopically confirmed, double the known high mass population of the cluster. Dozens of intermediate mass stars are newly recognized members of the cluster. The ratio between Class III and Class II objects in the intermediate- to high-mass range is 15 : 1. This is an important new sample ∼ for stellar and disk studies of intermediate mass PMS stars. The brightness of several K-excess sources makes them extremely favorable for ground based follow-up. We find a very luminous X-ray emitting infrared source subjected to significant absorption, which may be a new obscured massive YSO or a Wolf-Rayet star. 3. We determine the spatial distribution of over 700 new members to sub- arcsecond precision for the first time. The cluster is roughly spherically symmetric, with the highest stellar density around the massive O3 stars. A density enhancement centered at the tip of the molecular filament deserves further investigation (Figure 3.11). A radial profile of the stellar surface density shows that the physical size of the cluster spans over 2 pc in radius and extends beyond 6 pc (Figure 3.12). 4. XLFs are derived for lightly obscured and heavily obscured stellar populations 1 in NGC 6357 (Figure 3.9). The high X-ray luminosity end (log L 30.3 ergs s− ) of X ≥ the NGC 6357 XLF is clearly consistent with a power law relation as seen in COUP and Cep B, allowing a scaling comparison with the COUP XLF. The first estimate of the total cluster population of Pismis 24 is a few times the known Orion population. The presence of two O3 stars is consistent with the standard IMF. 5. We detect all 10 known OB stars with spectral type earlier than B1 in the Pismis 24 cluster, including the WC7+O7 binary WR 93. Their LX /Lbol ratios are 7 consistent with the canonical 10− value with less than an order of magnitude scatter. The O3 stars Pis 24-1 (O3.5If) and Pis 24-17 (O3.5IIIf) exhibit X-ray luminosities LX 33 −1 ∼ 10 ergs s . The X-ray spectrum of Pis 24-1 shows a soft emission component (kT 0.5 ∼ keV) and a harder component (kT 1.7 keV), while Pis 24-17 exhibits an unusally ∼ strong Si xiii line with a soft 0.7 keV thermal plasma. The WC7+O7 binary WR93 is 33 the brightest X-ray source in the field (L 2 10 ). Its hard spectrum suggests that X ≃ × the X-ray emission is generated from colliding winds. The lightcurves of these massive stars are constant through the observation. We assembled a list of candidate O stars from sources with high X-ray and infrared luminosity to facilitate future spectroscopy follow-up. Four candidate HAeBe stars are also identified. 6. We report the detection of X-ray emission from an EGG at the tip of a molecular pillar, which was previously found to be an IR source and the peak of the radio continuum. It is in the early evolutionary of stages of star formation. The spectral type estimated from its SED is B0–B2. The non-detection of EGGs in the Eagle Nebula may be the result of lower stellar mass and earlier evolutionary phases. We assemble a sample of 16 deeply embedded objects. 1 7. Several flaring sources exhibit luminosities above log L 32.0 ergs s− , which t,c ≥ is exceptionally high for PMS stars. One powerful flare seen in ACIS #672 (CXOU J172457.87-341203.9) is comparable to the strongest X-ray flares known in PMS stars 1 (L 32.5 ergs s− ). Its lightcurve rises 15 times above the quiescent level during the t,c ∼ flare, with a slow rise phase that is distinct from the fast rise flares seen PMS stars. 52

8. Soft unresolved X-ray emission in the NGC 6357 region is present. However, the luminosity derived from spectral fitting is consistent with the estimated level of integrated emission from the unresolved PMS stars. 53 is out streak. (ks) (keV) (16) (17) EffExp Med E e Var d edition of the Astrophysical Journal Anom Characteristics c B P -5 .... a 32.7 2.4 -5 .... b 33.1 2.6 -5 .... a 33.0 2.0 -5 .... a 34.2 3.7 -5 .... a 34.2 1.9 -5 .... a 34.7 3.9 -5 .... a 33.6 1.1 -5 .... a 31.6 2.1 < < < < < < < < or sources with fewer than 4 total full-band counts. No value a 9 ; e = source on field edge; p = source piled up; s = source on read . 0 < ) is ). 6 12.8 12.4 0.91 3.2 38 8.3 7.6 7.2 20.0 0.89 0.90 2.4 3.1 -4.3 .... a 33.7 2.1 7 9.4 8.0 0.90 2.7 9 6.8 58.3 0.89 6.7 3 7.8 1.4 0.90 4.8 .8.1 7.0 13.2 13.7 9.3 0.90 0.90 3.7 2.1 -3.5 .... a 34.9 3.8 Extracted Counts mkarf 17.7 22.8 160.0 0.91 14.7 11.5 12.0 71.7 0.91 8.9 sources is available in Wang et al. (2007a) in the electronic Catalog: Basic Source Properties Net ∆Net Bkgd Net PSF Signif log Chandra of 665 Chandra b ) Full Full Full Hard Frac ′ θ /ApJS/v168n1/65169/65169.html ackground. ector (FRACEXPO from ) ( ′′ Err ibly variable; c = definitely variable. No test is performed f dicates that the source is in a crowded region. J2000 δ Position Table 3.1. Main J2000 (deg) (deg) ( α http://www.journals.uchicago.edu/ApJ/journal/issues 3.3 for a detailed description of the columns. The full table § Source = probability extracted counts (full band) are solely from b B Off-axis angle. Source anomalies: g = fractional time that source was on a det Full band = 0.5–8 keV; Hard band = 2–8 keV. 1 172353.98-340850.8 260.974932 -34.147448 0.4 10.7 267.2 2 172358.42-340802.6 260.993436 -34.134056 0.5 10.2 107.0 3 172359.11-341217.6 260.996328 -34.204895 0.8 9.1 23.2 6. 45 172401.20-340928.0 172402.27-341110.4 261.005039 261.009476 -34.157783 -34.186237 0.9 0.8 9.1 8.5 13.7 19.4 5. 5. 6 172403.19-341402.2 261.013313 -34.233962 1.1 8.5 16.6 5. 8 172405.42-341257.5 261.022606 -34.215981 0.4 7.9 63.2 8. 9 172405.64-340710.0 261.023532 -34.119459 0.7 9.3 38.2 7. P Source variability: a = no evidence for variability; b = poss 1012 172406.70-341305.0 172407.65-341751.9 261.027933 261.031884 -34.218067 -34.297766 0.8 0.8 7.6 9.3 11.0 27.8 4 7 # a b c d e Note. — See (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq CXOU J Note that a reduced PSF fraction (significantly belowreported 90%) for in sources in chip gaps or on field edges. Supplement Series ( 54 is out streak. (ks) (keV) (16) (17) 33.6 1.4 EffExp Med E e ··· Var d edition of the Astrophysical Journal Characteristics Anom c B P or sources with fewer than 4 total full-band counts. No value a 9 ; e = source on field edge; p = source piled up; s = source on read . 0 < ) is ). 796 3.60 4.2 3.8 4.0 2.3 3.7 0.90 3.8 0.90 0.5 0.90 0.90 1.5 1.5 1.3 -2.4 1.0 -2.4 -1.8 -1.5 ...... g... a a a 33.0 33.1 32.2 2.0 2.1 2.6 527 2.70 2.5 3.5 6.2 1.2 5.9 0.90 2.2 0.89 2.4 0.89 0.90 1.5 1.2 1.5 -2.7 1.3 -1.8 -2.5 -2.8 ...... a a a a 34.9 35.3 34.7 36.4 3.5 4.5 1.1 1.1 .1 9.6 2.6 0.91 1.8 -2.7 .... a 32.4 1.6 .3 13.1 1.2 0.91 1.4 -1.6 .... a 28.1 1.4 Extracted Counts mkarf sources is available in Wang et al. (2007a) in the electronic Net ∆Net Bkgd Net PSF Signif log Chandra of 114 b ) Full Full Full Hard Frac ′ θ /ApJS/v168n1/65169/65169.html ackground. ector (FRACEXPO from ) ( ′′ Err ibly variable; c = definitely variable. No test is performed f Table 3.2. Tentative Source Properties dicates that the source is in a crowded region. J2000 δ Position J2000 (deg) (deg) ( α http://www.journals.uchicago.edu/ApJ/journal/issues 3.3 for a detailed description of the columns. The full table § Source = probability extracted counts (full band) are solely from b B Off-axis angle. Source anomalies: g = fractional time that source was on a det Full band = 0.5–8 keV; Hard band = 2–8 keV. 7 172403.64-340634.7 261.015178 -34.109657 1.4 10.0 10.4 5 P Source variability: a = no evidence for variability; b = poss 1116 172406.84-341149.723 172412.29-341500.324 172415.22-340730.0 172415.43-340958.2 261.028513 261.051238 -34.197157 261.063419 -34.250107 261.064294 1.2 -34.125011 1.0 -34.166176 0.9 7.5 0.9 7.0 7.5 6.2 6.4 6.8 5.2 3. 3.7 3. 3. 3. 2527 172415.57-342044.732 172415.91-341441.233 172418.05-340824.537 172418.06-341650.6 261.064909 172419.79-341130.3 261.066326 -34.345752 261.075226 -34.244799 261.075253 1.5 -34.140162 261.082481 0.8 -34.280743 10.3 0.9 -34.191754 1.0 6.2 1.0 6.5 7.0 7.9 4.9 6.3 4.5 5 6.5 3. 4.8 3. 3. 3. # a b c d e Note. — See (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq CXOU J Supplement Series ( Note that a reduced PSF fraction (significantly below 90%) in reported for sources in chip gaps or on field edges. S S S S S SS /ApJS/

corrected 55 c spline fit for a rough osity. The confidence . Luminosities were calculated assuming c 2 ν χ of Fit Goodness Notes -The USNO-B1.0 Catalog (Monet et al. 2003); it USNO http://www.journals.uchicago.edu/ApJ/journal/issues t,c L log t L b are also shown. log l Plasma Fits and the plasma emission measure. Columns 6-10: Observed and minosities are subscripted with a ) 1 l Supplement Series ( − h,c L ion, nevertheless the fit is acceptable to estimate the lumin log (ergs s 6.1 for fit parameters. S means the spectral fit is only used as a § X-ray Luminosities h L log s L for the spectral fit. 2 log χ 2 30.76 31.21 31.28 31.34 31.68 0.47 USNO 0557-0528516+2MAS 1 30.96 31.54 31.60 31.64 31.89 0.66 USNO 0557-0528507+2MAS 2 31.05 31.63 31.69 31.73 32.01 0.80 USNO 0557-0528485+2MAS 2 30.84 31.54 31.60 31.62 31.88 0.46 USNO 0557-0528440+2MAS 06 31.27 31.96 32.00 32.04 32.20 0.71 Pis24-11 2 30.88 31.32 31.37 31.46 31.72 0.27 USNO 0556-0531216+2MAS 3 30.68 31.54 31.65 31.60 31.97 0.93 2MASS 13 30.76 31.22 31.23 31.35 31.44 1.07 USNO 0558-0532169+2MA 05 31.40 31.76 31.77 31.91 32.04 1.06 Pis24-15 05 31.15 31.97 32.00 32.03 32.18 0.92 S; AGN? ...... content. ) 3 − EM log 2 54.7 +0 1 54.8 +0 1 55.0 +0 1 54.8 +0 2 54.7 +0 2 54.9 +0 ...... 05 55.0 +0 09 54.3 +0 05 54.9 +0 05 55.0 +0 0 0 0 0 0 0 . . . . 0 0 0 0 − − − − − − − − − − 3.4. Columns 11: Reduced ariable abundance fit was also performed; see § a 0 3 0 2 0 9 2 3 8 ion. Columns 3-5: Estimated column density, plasma energy, . Pis24 stars and other cataloged names obtained from VizieR ...... –8 keV); t = total band (0.5–8 keV). Absorption-corrected lu 007a) in the electronic edition of the Astrophysical Journa 0 . 10 kT XSPEC eans the fit does not formally satisfy the convergence criter solar abundances. > 6 2.2 +1 9 3.3 +1 9 7.0 +3 8 2.9 +1 1 3.6 +2 7 7.1 +2 8 4.1 +1 8 2.7 +1 1 3.0 +2 ...... × 0 0 1 0 1 3 0 0 1 ral analysis 3 Spectral Fit . d in − − − − − − − − − -MSX6C Infrared Point Source Catalog (Egan et al. 2003). 1 1 08 1 1 3 09 2 2 10 ...... MSX6C ) (keV) (cm 2 H − N (cm 1 22.0 +0 1 22.0 +0 1 22.0 +0 1 22.1 +0 3 21.4 +0 2 21.9 +0 2 22.3 +0 ...... 09 21.9 +0 10 21.5 +0 08 22.0 +0 Table 3.3. X-ray Spectroscopy for Brighter Sources: Therma 0 0 0 0 . 0 0 0 . . 0 0 0 − − − − − − − − − − ). A portion is shown here for guidance regarding its form and Source X-ray luminosities: s = soft band (0.5–2 keV); h = hard band (2 All fits were “wabs(apec)” in XSPEC and assumed 0 2 172358.42-340802.6 1 172353.98-340850.8 2T means a two-temperature fit was also performed; VT means a v 51 172423.93-340816.2 77 172428.94-341450.6 78 172429.03-341813.5 # a Note. — Table 3.3 is published in its entirety in Wang et al. (2 b c Note. — Column 1: X-ray source number. Column 2: IAU designat (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 196 172438.98-341220.6 188 172438.73-341202.9 140 172434.79-341318.0 158 172436.63-341550.7 112 172433.04-341654.5 Seq CXOU J log v168n1/65169/65169.html a distance of 2.56 kpc. estimate of luminosity;intervals actual are fit missing since values no are error unreliable. calculation P is reporte m TYC-The Tycho-2 Catalogue (Hog et al. 2000); for absorption X-ray luminosities, obtained from our spect 56 c and Notes minosities confidence 144 sources is t,c L 1.02 . 31.37 . log t b L log ) 1 − h,c L log h mal Plasma Fits X-ray Luminosities L log s pc. L 30.49 30.55 30.58 30.82 31.05 . 30.30 31.19 31.34 31.24 31.73 . 30.24 30.67 30.79 30.80 31.36 . 30.53 30.74 30.76 30.95 31.08 . 30.36 30.04 30.25 30.53 31.93 . log 2 4 5 3 2 . . . . . +0 +0 +0 +0 +0 ) (ergss ion. Columns 3-5: Estimated column density, plasma energy, 3 − . EM timate of luminosity; actual fit values are unreliable. Some 54.1 54.7 54.4 54.9 30.30 31.41 31.58 31.45 31.92 . 54.0 55.0 53.9 30.62 30.38 30.39 30.82 30.86 . –8 keV); t = total band (0.5–8 keV). Absorption-corrected lu 3 4 6 3 3 2 log ...... 0 0 0 0 0 0 rected for absorption X-ray luminosities. The full table of − − − − − − XSPEC solar abundances. a 8 × 5 3 6 3 1 ...... 3 . d in +3 +4 +2 +0 +0 +14 10 54.7 29.59 31.60 31.76 31.61 31.89 S 1.9 2.3 kT 1.5 1.8 2.7 0.5 2.9 > 9 9 7 3 2 4 5 ...... 0 0 0 0 0 1 1 − − − − − − − 3 2 2 4 2 1 1 ...... +0 +0 +0 +0 +0 +0 +0 ) (keV) (cm H 2 − N 21.6 22.4 22.2 21.5 22.5 22.3 22.7 9 2 4 5 5 2 1 ...... 0 0 0 0 1 0 0 − − − − − − − Table 3.4. X-ray Spectroscopy for Less Bright Sources: Ther . Luminosities were calculated assuming a distance of 2.56 k c Source Spectral Fit All fits were “wabs(apec)” in XSPEC and assumed 0 X-ray luminosities: s = soft band (0.5–2 keV); h = hard band (2 S means the spectral fit is only used as a spline fit for a rough es 9 172405.64-340710.0 21.0 8 172405.42-341257.5 3 172359.11-341217.6 22.0 4.8 53.9 30.01 30.73 30.78 30.81 3 35 172418.81-341204.7 22.1 1.3 54.4 30.43 30.55 30.65 30.79 30 172416.18-341311.3 29 172416.17-341526.6 20 172413.68-341451.8 19 172413.61-340921.8 15 172409.28-341224.3 12 172407.65-341751.9 Note. — Column 1: X-ray source number. Column 2: IAU designat a c b # (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) Seq CXOUJ log intervals are missing since no error calculation is reporte the plasma emissionprovided measure. in Columns the 6-10: electronic edition Observed and of cor the Journal. are subscripted with a 57 s -source is out ··· QFLAG Class 10 SNR); B = high > Explanatory Supplement Explanatory Supplement to 2MASS (J2000) coordinates s K oor; F = reliable photometric errors -no counterpart, s × & Magnitudes Cautionary Flags K − significance detection ( source unaffected by artifacts; b = bandmerge H H − -counterpart. edition of the Astrophysical Journal Supplement Series is available. Column 14: From J √ measurements. Column 13: From earby star; p = persistence contamination from nearby star; s ) K ′′ ( φ , and s K − H , H − J 5.5 for descriptions of the class designations. Column 7-9: § ················································ ). 5 SNR); D = low significance detection; E = point spread fitting p > ion. Columns 3-5: existence of ONIR counterparts; confusion and contamination flags with three characters: 0 = F 17 24 05.62 -34 07 09.5 0.60 0.399 0.172 8.458 AAA 000 photometry quality flags with three characters: A = very high R 17 23 58.36 -34 08 02.5 0.87 2.004 1.251 11.928 AAA 000 RR 17 23 59.16 -34 17 12 24 17.0 03.22 -34 0.80 14 02.1 0.35 0.749 0.487 0.990 10.690 0.369 11.313 AAA AAA 000 000 RR 17 24 01.32 -34 17 09 24 27.3 03.56 -34 1.58 06 36.2 1.77 0.744 0.477 2.092 12.926 0.739 13.505 UAA DBA 000 ccc H H s HR 17 24 06.64 -34 13 04.9 0.74 2.319 1.208 13.316 UCA 000 s sources is available in Wang et al. (2007a) in the electronic urce classifications. See JHK ted); X = source is detected but no valid brightness estimate arcsec. Column 10-12: 2MASS JHK Chandra usion from nearby star; d = diffraction spike confusion from n Table 3.5. Stellar Counterparts and Classifications ··· ··· ······ ··· ··· ··· ··· ··· ··· ······ ··· ··· /ApJS/v168n1/65169/65169.html × ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ×× ··· ··· ··· ··· √ √ √ √ √ √ √ (Cutri et al. 2003). 2MASS (Cutri et al. 2003). 2MASS 7 SNR); C = moderate significance detection ( > Source Counterparts & Class 2MASS Coordinates 2MASS Colors 12 172353.98-340850.8 172358.42-340802.6 3 172359.11-341217.6 6 172403.19-341402.2 9 172405.64-340710.0 4 172401.20-340928.0 7 172403.64-340634.7 5 172402.27-341110.4 8 172405.42-341257.5 10 172406.70-341305.0 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) Note. — Column 1: X-ray source number. Column 2: IAU designat Seq # CXOU J 2MASS SIRIUS HST Flags RA(J2000) DEC(J2000) ∆ http://www.journals.uchicago.edu/ApJ/journal/issues significance detection ( not available; Uto = the upper 2MASS limit All on Sky magnitude Data (source Release not( detec the 2MASS All Sky Data Release of the field of view of the available images. Column 6: X-ray so confusion (possible multiple source); c = photometric conf = electronic stripe from nearby star. The full table of 779 and the offset between X-ray position and near-IR position in h 58 ··· ··· ··· ··· ··· ··· Notes nd Massey et al. using bolometric olors to the 1 Myr PMS b bol L ··· ··· ··· ··· ··· ··· ··· ··· 39.97 HD 157504=[N78] 52 38.05 38.04 > log g H N log g X L ··· ··· ··· ··· ··· ··· ··· ··· X-ray properties log f Seq # e V Catalogue of Stellar Spectral Classifications (Skiff 2005) a .90 #194 31.29 21.9 .61 #527 30.28 22.2 A ric luminosities from Massey et al. (2001) are recalculated d s 14 6.49 #225 30.45 22.0 .00 5.81 No K 9.79 7.03 #140 32.20 21.9 lor-magnitude diagram by de-reddening 2MASS photometric c d s K ). 30 8.47 5.79 #293 31.17 21.0 38.57 2 .21 9.78 5.56 No .24.28 8.35 8.75 5.54 5.84 #77 #151 32.04 31.96 21.5 22.1 38.63 38.36 [N78] 46 .67 5.87 5.86 #747 33.21 21.8 − 0.28 5.89 6.16 #344 33.11 22.1 39.60 HDE 319718=[N78] 35 − H d H − J d J 8 0.48 0.31 7.69 5.81 #331 31.43 21.3 38.81 92 0.57 0.29 7.95 6.45 #464 31.36 21.1 38.61 [N78] 36 − V ONIR photometry b V − 1.39 3.76 0.59 0.47 9.61 6.08 No 1.48 3.97 0.68 0.34 8.97 6.42 #332 31.56 22.1 38.26 1.60 4.57 0.59 0.49 7.37 7.35 #407 31.90 22.0 38.60 [N78] 58 1.49 4.02 0.54 0.31 6.98 6.57 #420 32.83 22.1 39.09 [N78] 57 B b is X-ray derived column density (cm B H − rees, arcminutes, and arcseconds. Coordinates are from the ··· ··· ··· ··· N )). U 01). J 002). Optical photometry from Massey et al. (2001). Bolomet ). 1 − − b V ( ation sample. Tentative spectral types are inferred from co V 996). E 39 (ergs s c . Table 3.6. Known O and Early B stars in Pismis 24 t,c = 1 L V A J2000 δ c J2000 α Star b O9? 17 24 38.81 -34 14 58.2 12.98 0.48 1.44 3.67 0.59 0.21 8.52 5 O9? 17 24 47.81 -34 15 16.5 13.46 0.58 1.68 4.10 0.61 0.33 8.42 6 O9-B1? 17 24 34.68 -34 13 17.1 14.53 0.30:: 1.57 3.79 0.66 0.30 O9-B1? 17 24 39.29 -34 15 26.4 14.26 0.40 1.40 3.48 0.49 0.29 10 O9-B0? 17 24 40.39 -34 12 05.9 13.93 0.53 1.43 3.88 0.60 0.32 9. SpecType i i i i i is absorption corrected full-band luminosity a ID X Other identifications from Neckel (1978). Spectral types from Massey et al. (2001) and Walborn et al. (2 Infrared photometry from 2MASS. L Identifications from Moffat & Vogt (1973) and Massey et al. (20 Visual extinction from Bohigas et al. (2004) ( Units of RA are hours, minutes and seconds; units of DEC are deg ACIS source number for X-ray detections. Not included in Massey et al. (2001) spectroscopic classific h i f g b d e c a Pis 24-12Pis 24-19 B1V B1V 17 24 42.27 17 24 43.69 -34 11 41.2 -34 11 40.7 13.88 14.43 0.38 1.47 3.38 0.51 0 WR 93 WC7+O7-9 17 25 08.85 34 11 12.5 10.68 0.70 1.42 3.65 0.50 0 Pis 24-18 B0.5V 17 24 43.29 -34 11 41.9 13.97 Pis 24-15Pis 24-10 O8V O9V 17 24 28.86 17 24 36.04 -34 14 50.3 -34 14 00.5 12.32 13.02 0.14 0.40 1.27 1.40 3.29 3.51 0.45 0.49 0 0 Pis 24-3 O8V 17 24 42.30 -34 13 21.5 12.75 0.24 1.41 3.46 0.52 0. Pis 24-11 Pis 24-2Pis 24-13Pis 24-16 O5.5V((f)) O6.5V((f)) 17 24 17 O7.5V 43.20 24 45.78 -34 12 -34 43.5 09 17 39.9 24 44.45 11.95 12.73 -34 11 58.8 0.32 0.11 13.02 1.41 1.48 3.4 3. Pis 24-17 O3III(f*) 17 24 44.73 -34 12 02.7 11.84 Pis 24-1 O3If* 17 24 43.49 -34 11 57.0 10.43Pis 24-8 0.40 1.45 3.70 0.55 Pis 24-7 Pis 24-9 Pis 24-4 isochrone. (2001). corrections from Heap et al. (2006) instead of Vacca et al. (1 59

Table 3.7. X-ray and IR luminous stars as Candidate O stars

Seq # CXOU J J − HH − Ks Ks log LX log NH kT Notes (a) Stars with log LX > 32.0 140 172434.79-341318.0 0.660 0.296 9.789 32.04 21.9 7.0 Pis 24-11 158 172436.63-341550.7 1.194 0.504 12.270 32.01 22.0 2.9 V 227 172440.67-341403.8 1.203 0.176 14.324 32.06 22.1 7.6 ﬂaring 379 172443.95-341145.6 -0.572: 0.651: 12.298 32.65 22.0 10.0: HST 399 172444.37-341039.8 1.063 0.957 10.972 32.05 22.0 4.1 HST 471 172446.01-341407.3 1.636 0.802 10.674 32.32 21.9 5.7 ... 519 172447.56-341048.6 1.466 0.706 10.167 32.38 22.1 7.5 V;HST 577 172450.15-341243.4 1.240 0.509 10.788 32.25 22.0 4.1 ... 649 172455.55-341631.3 1.286 0.510 12.064 32.18 22.1 4.3 V 672 172457.87-341203.9 1.331 0.576 10.256 32.52 22.3 10.0: ﬂaring 692 172459.74-340958.7 ··· ··· ··· 32.17 22.5 2.2 faint in SIRIUS 746 172508.76-341115.2 ··· ··· ··· 32.18 22.0: 1.9: close to WR star 753 172510.93-340843.2 ··· ··· ··· 32.08 22.0 3.8 V; IR nebulosity

(b) Stars with K < 10 9 172405.64-340710.0 0.399 0.172 8.458 31.20 21.5 0.7 IRAS 17207-3404 19 172413.61-341656.5 2.006 0.991 5.334 31.0 21.9 2.8 MSX6C G353.0401+00.9326 120 172433.46-341344.9 0.602 0.333 9.649 31.29 21.7 10.0: ... 194 172438.90-341459.0 0.587 0.208 8.519 31.29 21.9 1.7 Pis24-8 225 172440.50-341206.3 0.595 0.315 9.138 ··· ··· ··· Pis24-4 312 172442.88-340911.6 0.931 0.502 9.231 ··· ··· ··· ... 527 172447.87-341517.0 0.608 0.327 8.424 ··· ··· ··· Pis24-7 539 172448.44-341743.7 0.468 0.270 8.026 31.14 21.5 0.9 ... 643 172455.09-341111.5 0.798 0.427 8.866 32.15 22.4 0.6 ... 654 172455.85-341234.0 1.297 0.952 8.389 ··· ··· ··· ... 694 172459.86-341610.8 1.645 1.304 8.137 31.34 22.1 2.5 MSX6C G353.1413+00.8083

Note. — Column 1: X-ray source number. Column 2: IAU designation. Columns 3-5: 2MASS colors and Ks magnitudes. Columns 6-8: X-ray luminosities (0.5–8.0 keV) corrected for absorption, derived column densities, and thermal plasma energy obtained from our spectral analysis. Columns 9: Related notes: ACIS source number; V=Variable lightcurves; HST=HST counterparts found; Other catalog names. Stars in the top panel are selected from X-ray criteria, and those in the bottom panel are selected from infrared criteria. Missing log LX , log NH , and kT values mean that the source has < 20 net counts. 60

Table 3.8. X-ray Selected Candidate Intermediate-mass Stars

˜ a ˜ a b Seq# CXOUJ JHKs AV M ∆K (mag) (mag) (mag) (mag) (M⊙) (mag) 2 172358.42-340802.6 14.99 13.16 11.93 18 13 3 172359.11-341217.6 11.86 11.16 10.69 8 14 · · · 6 172403.19-341402.2 12.60 11.67 11.31 10 13 · · · 12 172407.65-341751.9 14.32 13.03 12.46 8 3 · · · 18 172413.61-340921.8 13.51 12.65 12.34 4 3 · · · 20 172413.68-341451.8 13.12 12.13 11.77 7 4 · · · 29 172416.17-341526.6 13.88 12.28 11.45 16 16 · · · 35 172418.81-341204.7 14.80 13.64 13.04 6 2 · · · 38 172420.25-341310.7 13.90 12.54 11.88 14 12 · · · 45 172421.75-341539.4 15.40 13.55 12.82 15 4 · · · 156 172436.45-341236.7 13.30 12.32 11.57 5 3 · ·√ ·

Note. — Column 1: X-ray source number. Columns 2: IAU designation. Columns 3-5: 2MASS JHKs-band magnitudes. Columns 6-7: Visual extinction and mass estimated from de-reddened location along the standard interstellar reddening vector to the 1 Myr PMS isochrone in the 2MASS color-magnitude diagram, assuming a distance d 2.56 kpc. Column 8: Flag of signiﬁcant K-band excess. The full table of 100 Chandra≃ sources is available in Wang et al. (2007a) in the electronic edition of the Astrophysical Journal Supplement Series (http://www.journals. uchicago.edu/ApJ/journal/issues/ApJS/v168n1/65169/65169.html). a These quantities assume a cluster age of 1 Myr and d 2.56 kpc. For early B ˜ ˜ ≃ stars, AV and M (estimated visual extinction and mass) are especially dependent on the age and distance assumptions (Figure 7). b √ marks signiﬁcant K-band excess derived from high quality 2MASS photometry where near-IR color excess E(H K) is larger than σ(H K), the uncertainty in H K color index. − − − 61

Table 3.9. X-ray Selected Heavily Obscured Sources

a −2 b c d Seq# CXOUJ Criteria log N (cm ) E (keV) Notes H h i 8 172405.42-341257.5 I;II 22.7 3.9 15 172409.28-341224.3 I 22.5 3.1 V;2MASS· · · 61 172426.47-341850.9 I 22.5 3.3 142 172434.95-340533.3 I;II 23.2 4.7 · · · 569 172449.82-341455.6 I 22.4 3.4 · · · 598 172451.90-341331.3 I;II 22.6 3.1 2MASS· · · 599 172451.95-341513.0 II 22.6 2.0 2MASS 634 172454.31-341209.4 I 22.3 3.1 VV;2MASS 664 172456.97-341651.8 II 22.6 2.1 2MASS 692 172459.74-340958.7 II 22.5 2.9 693 172459.85-341302.9 II 22.5 2.8 2MASS· · · 702 172501.09-341404.6 I;II 22.7 3.0 713 172502.54-341549.7 I;II 22.7 4.1 · · · 728 172504.93-340410.5 I;II 22.6 3.9· ·e · 753 172510.93-340843.2 I;II 22.5 3.5 VV;f 775 172519.23-341356.6 I;II 22.8 3.7 · · ·

a −2 Selection criteria: I.– E 3.0 keV; II.–N 22.5 cm . The values shown here h i≥ H,X ≥ are rounded. b Column density derived from spectral ﬁtting. c Median photon energy of the X-ray source. d Related notes: V=variable lightcurves; VV=ﬂaring; 2MASS=2MASS counterparts found. e MSX6C G353.3175+00.9075 f MSX6C G353.2651+00.8443 62

Fig. 3.1 Multiwavelength images of the NGC 6357 region from long to short wavelengths. (a) ◦ ◦ 1.5 1.5 view of 6 cm radio continuum contours from Parkes survey (Haynes et al. 1978) with the three× H ii regions labeled. (b) The same region shown as a 3-color composite mid-infrared ′ image from the MSX survey: red is 8.28 µm, green is 12.13 µm, and blue is 21.3 µm. (c) 20 ′ × 20 K-band image from the 2MASS survey showing the H ii region G353.2+0.9, dark clouds, and ′ ′ the stellar ﬁeld. The box shows the SIRIUS FOV. (d) 6 6 JHK composite image from the ∼ × ′ ′ SIRIUS observation. J is shown in blue, H is green, and K is red. (e) 20 20 R-band image from the Digitized Sky Survey emphasizing Hα emission. The box shows HST/WFPC2× coverage. The brightest members WR 93, Pis 24-1, and Pis 24-17 are marked. (f) An unpublished X-ray image from a short ROSAT PSPC observation. The three brightest regions are associated with the NGC 6357 H ii regions. The ACIS-I FOV is shown as a box, with the oﬀ-axis S2+S3 chips represented as a rectangle to the east. Note that S3 captures G353.2+0.7 at its southern edge. 63

Fig. 3.2 (a) The 17′ 17′ ACIS-I image (0.5–7.0 keV) binned by 2′′ 2′′. The grey × × polygons show the source extraction regions that are matched to the local Chandra point spread function. The black contour in the center outlines the ionization front seen in the optical image. (b) The ACIS full-field adaptively smoothed image showing all four I-chips and two S-chips. Red represents the soft band (0.5–2.0 keV) X-ray emission and blue represents the hard band (2.0–7.0 keV) X-ray emission. The cyan and white polygons outline the extraction regions for possible diffuse emission in the central region, and in a larger area, respectively. The yellow polygon outlines the region used for diffuse background estimation. See 4.6. (c) Zooming-in on the inner 4′ 4′ of the ACIS-I ′′ ′′§ ∼ × data, binned by 0.25 0.25 , then adaptively smoothed. × 64

Fig. 2. — Continued. 65

Fig. 3.3 Examples of close X-ray source pairs (ACIS #313, 317; #402, 406; #427, 428; shown ′′ ′′ as dashed circles) recovered from image reconstruction. Each panel includes: 6 6 raw ACIS ′′ × image binned by 0.25 (top left), reconstructed image (top right), HST image (bottom left), and SIRIUS image (bottom right). The white stripe in the HST image is the diﬀraction spike from a nearby bright star. 66

Fig. 3.4 (a) HST/WFPC2 image (F814W) revealing details of the interface between the massive stars, the H ii region, and the molecular cloud; (b) zoom in on the elephant trunk with 2MASS near-IR sources (red circles) and X-ray sources (cyan polygons) overlaid. The cyan polygons are not circles; they represent the X-ray source extraction polygons based on the PSF. 67

Fig. 3.5 A cartoon illustration of the source classification in 3.4.5: R–Reliable X-ray § sources with optical and NIR stellar identifications; C–Clustered X-ray sources with off-axis angle Θ 5.0 ′; H–Hard X-ray sources with E 3.0 keV. The area of each ≤ h i ≥ circle is proportional to the number of sources in the class. Group U (unreliable X-ray detections) and F (foreground field stars) are not shown here. 68

Fig. 3.6 Color-color diagram of NGC 6357 X-ray stars with high quality 2MASS counterparts. Open circles and filled circles represent sources without and with significant K-band excess (E(H K) > 1σ uncertainty in H K color), respectively. Crosses mark − − the known members of the Pismis 24 cluster from Massey et al. (2001). ACIS source numbers and error bars are given for the counterparts with significant K-band excess. The interesting source #19, the Pis24 member #331, and the foreground star #773 are also labeled to guide the reader. The solid and long-dash lines denote the locus of main sequence stars and giants from Bessell & Brett (1988), respectively. The dash dotted line is the locus for T Tauri stars from Meyer et al. (1997) and the dash triple-dotted line is the locus for HAeBe stars from Lada & Adams (1992). The dash lines represent the standard reddening vector for AV = 20 mag, with asterisks marking every AV = 5 mag. Many additional X-ray stars with lower quality 2MASS photometry are omitted. 69

Fig. 3.7 Infrared J H vs. J color-magnitude diagram for NGC 6357 stars. The solid line − is the 1 Myr isochrone for PMS stars from Baraffe et al. (1998) and Siess et al. (2000). A 0.3 Myr isochrone (Siess et al. 2000) is also shown (dotted line) to demonstrate how the estimated masses differ if we adopt a younger age for the cluster. The dash dotted line marks the location of Zero Age Main Sequence (ZAMS) stars. Dashed lines represent the standard reddening vector with asterisks marking every AV = 5 mag; the corresponding stellar masses are marked. ACIS numbers for sources with significant K-excess are shown. The interesting source #19, the Pis24 member #331, and the foreground star #773 are labeled to guide the reader. Error bars are shown for all sources with color uncertainties > 0.1 mag. This diagram uses the same source sample and symbols as Figure 6. 70

Fig. 3.8 Infrared H K vs. K color-magnitude diagram for NGC 6357 stars. The − solid line is the 1 Myr isochrone from Baraﬀe et al. (1998) and Siess et al. (2000). The dash dotted line marks the location of ZAMS stars. Dashed lines represent the standard reddening vector with asterisks marking every AV = 5 mag; the corresponding stellar masses are marked. ACIS numbers for sources with signiﬁcant K-excess are shown. The interesting source #19, the Pis24 member #331, and the foreground star #773 are labeled to guide the reader. Error bars are shown for all sources with color uncertainties > 0.1 mag. This diagram uses the same source sample and symbols as Figure 6. 71

Fig. 3.9 (a): X-ray luminosity function (XLF) constructed from the observed (not corrected for absorption) hard band (2.0–8.0 keV) X-ray luminosity Lh for the lightly obscured population in NGC 6357 (solid line), Orion (dashed line, Feigelson et al. 2005), and Cep B (dotted line, Getman et al. 2006). (b): XLF using Lh for the heavily obscured population in NGC 6357 and Cep B compared to the XLF for the lightly obscured population in Orion. (c): XLF using absorption corrected hard band (2.0–8.0 keV) X-ray luminosity Lh,c for the lightly obscured population in NGC 6357, Orion, and Cep B. (d): XLF using Lh,c for the heavily obscured population in NGC 6357 and Cep B compared to the XLF for the lightly obscured population in Orion. 72

Fig. 3.10 The spatial distribution of the X-ray stellar sources. Members with ONIR counterparts are plotted as diamonds, and likely members without ONIR identiﬁcations are shown with crosses. Obscured sources ( E 3.0 keV) are plotted in blue and less h i ≥ obscured sources ( E < 3.0 keV) are plotted in red. The large scale optical nebulosity h i contour from the DSS image traces the mid-IR ring-like morphology; it is shown in black together with the outlines of the ionization front in yellow and the CO “South-Eastern Complex” in green. 73

2 Fig. 3.11 Projected stellar surface density (number of stars arcmin− ) maps for (a) the unobscured population and (b) the obscured population, shown in logarithmic scale using a 0′.5 radius smoothing kernel. The highest concentration of stars coincides with the core of the cluster, where the massive O stars are located. A density enhancement, or a subcluster is seen centered at RA=17.414 (hours), Dec=-34.24 (degrees) in (b). There is a small overdensity around the WR star, at RA=17.419 (hours), Dec=-34.19 (degrees) in (a). 74

Fig. 3.12 Radial density proﬁles for X-ray stars in the Orion Nebula Cluster (COUP sample, Getman et al. 2005; thin line) and for Pismis 24 (thick line). COUP sample is spatially complete in the central 1 pc of Orion Nebula Cluster. 75

Fig. 3.13 The distribution of X-ray sources overlaid on the central 50′′ SIRIUS K image. The ONIR identiﬁed sources (circles), and unidentiﬁed sources (crosses) are shown, with the known massive members highlighted in white. 76

Fig. 3.14 X-ray spectra of the O3 stars and the brightest source WR93. The left three 22 −2 panels give apec model fits for Pis24-1 (NH = 1.1 10 cm and kT = 1.2 keV), Pis24- 22 −2 × 21 −2 17 (N = 1.2 10 cm and kT = 0.7 keV), and WR 93 (N = 6.9 10 cm and H × H × kT = 2.3 keV). Top right shows an improved fit with a two temperature thermal plasma 22 −2 fit for Pis24-1 (N = 1.3 10 cm , kT1 = 0.5 keV and kT2 = 1.7 keV). Middle right H × gives a slightly improved fit for Pis24-17 with a variable abundance thermal plasma fit 22 −2 (NH = 1.4 10 cm and kT = 0.7 keV). Bottom right presents an improved fit for × 21 −2 WR 93 with the vapec model (N = 7.4 10 cm and kT = 2.4 keV). H × 77

Fig. 3.15 The LX vs. Lbol relation for the X-ray detected OB stars in Pismis 24 (diamonds) compared to COUP massive stars with strong winds (crosses) and weak winds (squares). Four COUP stars with weak winds have upper limits in LX reported in Stelzer et al. (2005). The Lbol for WR 93 is a lower limit due to uncertain bolometric correction for late-type WC stars. The dashed line represents the canonical relation log(L /L ) 7. X bol ∼− 78

Fig. 3.16 Photon arrival diagram, lightcurve (2 ks bins) and X-ray spectrum of source #672 (CXOU J172457.87-341203.9), with a powerful ﬂare detected in this ACIS obser- 22 2 vation. The best spectral ﬁt is obtained with N = 1.7 10 cm− and kT > 10 keV. H × 79

Fig. 3.17 The spatial distribution of deeply embedded sources. Sources that have E h i≥ 3.0 keV are plotted with crosses and sources that have log N 22.5 are plotted with H,X ≥ circles. Diamonds represent sources that satisfy both criteria. The background DSS image is inverted so that brighter regions appear darker. 80

Chapter 4

A Chandra Study of the Young Open Cluster NGC 2244 in the Rosette Nebula

4.1 Introduction

During their evolution from Class I protostars to zero-age-main-sequence (ZAMS) stars, young stellar objects are readily identified in X-rays due to their highly elevated X- ray emission compared to the older Galactic stellar population (see reviews by Feigelson & Montmerle 1999; Favata & Micela 2003; Feigelson et al. 2007). High spatial resolution Chandra observations of well-known Galactic star forming regions (e.g., Chandra Orion Ultradeep Project, hereafter COUP, Getman et al. 2005; RCW 38, Wolk et al. 2006; Cepheus B, Getman et al. 2006), more distant molecular cloud and HII complexes (e.g., NGC 6334, Ezoe et al. 2006; NGC 6357, J. Wang et al. 2007; M 16, Linsky et al. 2007; M 17, Broos et al. 2007), the Galactic Center (e.g., Arches and Quintuplet clusters, Wang et al. 2006; Muno et al. 2006), and extragalactic star forming regions (e.g., 30 Dor, Townsley et al. 2006a,b), have greatly advanced our knowledge of star formation processes in these regions. Moreover, these studies demonstrate the unique power of studying star formation in the X-ray band. Besides the high energy phenomena, other new information about the young clusters such as population, membership, and star formation environs can be obtained when combining X-ray detections with knowledge obtained from longer wavelength studies. For example, despite the presence of a very small fraction of extragalactic sources and Galactic field stars in X-ray-selected samples, optical and infrared (IR) counterparts to X-ray sources are mostly cluster members. This is in contrast to the high percentage of non-members in the optical/IR images where, except for those with massive dusty disks, membership for individual stars generally has to be ascertained via spectroscopy. In the IMF studies of more distant, high mass star forming regions, traditional optical measurements are significantly encumbered by large reddening and membership confusion. In comparison to the massive stars, the lower mass pre-main- sequence (PMS) stellar populations are much less accessible and thus much harder to evaluate. In the past, Hα emission was used in general as a youth and membership indicator. However, in addition to the observational challenges due to prevalent bright Hα nebulosity in HII regions, the emission itself requires accretion activity in protoplanetary disks (Muzerolle et al. 1998, 2001) and thus susceptible to the disk evolutionary stages. It has long been recognized that X-ray emission circumvents these problems and is very effective for securing PMS membership of young clusters (Feigelson & Montmerle 1999). Modern X-ray observatories like Chandra and XMM-Newton enable identification of hundreds of individual members from their X-ray emission. In this work and subsequent chapters, we report Chandra studies of a well-known star forming complex, 81 concentrating here on a new census of the low mass cluster members and new knowledge of the IMF arised from the X-ray perspective. The Rosette star-forming complex, situated in a large star formation site in the Perseus spiral arm, provides an ideal testbed for studying sequential formation of clusters due to the favorable orientation of its morphological components, consisting of an expanding blister HII region on the edge of a giant molecular cloud oriented perpendicular to the line-of-sight. Thorough review of past and present research on this popular and important star formation region appear in Townsley et al. (2003) (hereafter TFM03) and Román-Zúñiga et al. (2007b) (hereafter RL07). Here we highlight what is most relevant to this work. The Rosette Nebula (= Sharpless 275 = W 16 = NGC 2237–2239, NGC 2244, and NGC 2246) is a large HII region at the tip of the Rosette Molecular Cloud (RMC). In both radio and optical images, it shows a prominent ring-like morphology, with a cluster of ionizing young stars located in the central hole (Celnik 1985; Townsley et al. 2003). The large-scale IRAS data (Cox et al. 1990) and CO emission map (Heyer et al. 2006) clearly show a similar annular morphology extending into the molecular cloud (Figure 4.1a). Extending to the southeast of the Rosette Nebula, the RMC is an elongated giant 5 molecular cloud with 10 M⊙ of gas and dust (Blitz & Thaddeus 1980). Multiwave- ∼ length observations in mid-IR and radio show clumpy structure in the RMC (Williams & Blitz 1998). Embedded star clusters have been revealed in the densest parts through near-IR imaging surveys (Phelps & Lada 1997; Román-Zúñiga et al. 2005, 2007a). Fig- ure 4.1a demonstrates the association between Phelps & Lada (1997) IR clusters and the molecular clumps where the CO emission peaks. Lada & Lada (2003) have shown that embedded clusters are physically associated with the most massive and dense cores in the molecular clouds based on systematic and coordinated surveys (e.g., Lada & Adams 1992). In the X-ray band, previous imaging study of the Rosette Complex was hampered by low spatial resolution. An early Einstein Observatory study detected a few individual O stars and extended 2 keV X-ray emission at the center of the nebula (Leahy 1985). ∼ Berghöfer & Christian (2002) considered the integrated contribution from X-ray emitting low mass PMS stars besides the OB stars, and concluded that the diffuse emission could be explained by unresolved point sources. Chen et al. (2004) analyzed the same ROSAT data set, and attributed the brightest X-ray sources to massive stars, active T Tauri stars, and foreground stars. Gregorio-Hetem et al. (1998) studying ROSAT observation of the RMC reported faint X-ray point sources associated with T Tauri stars and Herbig Ae/Be (HAeBe) stars and X-ray “hot spots” from unresolved embedded low mass star clusters. In TFM03 the first high resolution Chandra X-ray image mosaic of this high- mass star forming region was presented and soft diffuse X-ray plasma (kT 0.06 and 32 −1 ≃ 0.8 keV) with luminosity L 6 10 ergs s was detected in the HII region. It ≃ x ≃ × was attributed to combination of fast O star winds and unresolved T Tauri stars. The young stars powering the Rosette HII region are members of the massive open cluster NGC 2244. Figure 4.1b shows a Digital Sky Survey image (59′ 59′) of this × region with a few bright stars and interesting objects labeled. Despite its apparent low concentration of stars in the optical, NGC 2244 contains 30 early-type stars between ∼ 82

O4V and B3V (Table 6 in TFM03), whose cluster membership are secured from deep photometric study along with proper motion data and spectroscopy (Verschueren 1991; Park & Sung 2002). No other known massive young stellar cluster within 2 kpc, other than RCW 38 (Wolk et al. 2006) and M 17 (Broos et al. 2007), is comparably rich. The pre-main-sequence (PMS) members were largely unknown from optical studies; only a handful of Hα emission objects have been identified as young PMS members (Park & Sung 2002; Li & Rector 2004). Berghöfer & Christian (2002) presented 138 sources selected from ROSAT PSPC and HRI detections. Although incomplete, they revealed previously unknown PMS stars to K spectral types. The distance to NGC 2244 has been measured in many visual photometry studies and ranges between 1.4 kpc and 1.7 kpc. Hensberge et al. (2000) derive a distance 1.39 0.1 kpc to an eclipsing binary member V578 Mon, using a novel Fourier spectral ± disentangling technique. The main sequence (MS) turn-off age estimated by Park & Sung (2002) is 1.9 Myr, which is consistent with the inferred age from V578 Mon, 2.3 0.2 Myr ± (Hensberge et al. 2000). This makes NGC 2244 the youngest cluster within the larger Mon OB2 association (Hensberge et al. 2000). The adopted distance and cluster age for NGC 2244 are 1.4 kpc and 2 Myr throughout our studies, which are consistent with TFM03. Note that a larger distance value d = 1.6 kpc has been used by other researchers (Park & Sung 2002; Román-Zúñiga et al. 2007a). The discrepancy in distance affects ′ our derived X-ray luminosities (log Lx) only by 0.1 dex. At the distance of 1.4 kpc, 1 corresponds to 0.4 pc. Whereas the TFM03 paper was dedicated to study the diffuse X-ray emission in the Rosette HII region, we present here the X-ray point source study of the Rosette complex based on TFM03 data and a new 75 ks Chandra image centered on NGC 2244. The field of view of our mosaic Chandra fields are outlined by the polygons into Figure 4.1a. We have separated our study into a series of four papers with different astrophysical emphasis. In this chapter, we report Chandra/ACIS observations of the NGC 2244 cluster and the Rosette HII region, and study the young stellar population in detail. In an upcoming paper (Wang et al., in preparation; Chapter 5), we describe Chandra/ACIS observations of the embedded clusters in the RMC, aiming to investigate cluster formation in a sequential manner and to test whether molecular clumps preferentially forming embedded clusters of low-mass stars make up the fundamental building blocks of star formation in molecular clouds. The westernmost Chandra field is designed to study triggered star formation and the X-ray detection of a twin cluster to NGC 2244 (Li 2005; Román-Zúñiga et al. 2007a) will be presented in Chapter 6 (Wang et al., in preparation). A detailed analysis of the diffuse X-ray emission in the HII region will appear in Townsley et al. (in preparation). A shallow Spitzer Space Telescope survey of NGC 2244 has been reported (Balog et al. 2007) and a deep Spitzer survey for disk emission from Chandra stars in NGC 2244 is also underway (PI: Bouwman). This Chapter is organized as follows. First we describe the Chandra observations and data reduction in 2. In 3 we then identify the X-ray sources with optical and § § infrared counterparts, and evaluate the fraction of contaminants through simulations utilizing stellar population synthesis model and the log N–log S distribution for extragalactic X-ray sources. 4 is devoted to global properties of the NGC 2244 cluster, such § 83 as the X-ray luminosity function, the initial mass function, the K-band luminosity function, spatial structures, mass segregation, and the K-band excess disk fraction among the X-ray detected stars. We present collective properties of interesting X-ray sources in 5, ending with a summary in 6. § § 4.2 Chandra Observations and Data Reduction

The Rosette complex was observed with the Imaging Array of the Chandra Ad- vanced CCD Imaging Spectrometer (ACIS-I). The ACIS-I field of view is 17′ 17′in a × single pointing and a mosaic observation was designed to best image the ionizing cluster, capture the interface between the photoionized gas and the cold neutral material, step into the dense molecular cloud, and study its recently reported secondary cluster. As shown in Table 1, the entire observation consisted of four 20 ks ACIS-I snapshots in ∼ January 2001 (TFM03, Figure 2), a deep 75 ks ACIS-I image in January 2004 centered on the O5 star HD 46150 in NGC 2244 (Figure 4.2a), and one 20 ks ACIS-I pointing at the twin cluster to NGC 2244 (Li 2005) in 2007. The image mosaic covers 1◦ 0◦.25 ∼ × field of the Rosette Nebula and RMC. All images were taken in standard “Timed Event, Faint” mode with 3 pixel 3 pixel event islands except ObsID 3750 and ObsID 8454, × which used the Very Faint mode (5 pixel 5 pixel event islands). × We follow the same customized data reduction and source extraction described in TFM03, Wang et al. (2007a), and Broos et al. (2007). The same reduced dataset used in TFM03 study (Rosette Field 1-4) and the deep observation (Rosette Nebula/NGC 2244) following TFM03 reduction were used here for further analysis. With slightly different roll angles, we reprojected the ObsID 1874 data and merged them with the ObsID 3750 field. Figure 4.2a shows the merged 94 ks ACIS image of NGC 2244 overlaid with source extraction regions (see details below). Many point sources are visible, which are further illustrated in the smoothed X-ray composite image (Figure 4.2b) for the merged fields created with the CIAO tool csmooth (Ebeling et al. 2006). In Figure 4.2c the existence of soft diffuse emission is emphasized in the context of the DSS optical image, where the diffuse X-ray emission nicely fills in the cavity of the HII region. This component will be discussed in Townsley et al. 2007 (in preparation).

4.2.1 Source Finding and Photon Extraction Following the source finding procedure outlined in Chapter 2, twelve source lists using wavdetect program were merged, with the source position in the highest-resolution image retained, to generate a single list of candidate sources for each observation. An image reconstruction with the Lucy-Richardson maximum likelihood algorithm (Lucy 1974) was performed in the central 50′′ 50′′ around HD 46150 (Figure 4.3) in ObsID × 3750. Eighteen additional candidate sources from the image reconstruction were added to the source list. Adaptive-kernel smoothed flux images in the three energy bands were also created with csmooth to help visually identify additional faint potential sources. Source lists from all observations were then merged to form a master candidate detection list. Our source finding procedure results in a total of 1452 potential sources identified for five ObsIDs (omitting the westernmost field ObsID 8454). A preliminary event 84 extraction for the potential X-ray sources was made with our customized IDL script 1 ACIS Extract (version 3.98; hereafter AE, Broos et al. 2002). Using the AE-calculated probability PB that the extracted events are solely due to Poisson fluctuations in the local background, source validity can be statistically evaluated while taking account of the large distorted PSF at far off-axis locations and spatial variations in the background. After a careful review of the net counts distribution and IR counterparts frequency for all candidate sources, we rejected sources with PB > 0.014 likelihood of being a background fluctuation. The trimmed source list includes 1314 valid sources. Since our data analysis involves multiple Chandra observations, for convenience and avoidance of repeating source designation, we divide the X-ray point sources into NGC 2244 sources and RMC sources based on positions. According to the stellar density distribution of 2MASS sources in NGC 2244 (Li 2005), the CO emission maps (Williams et al. 1995; Heyer et al. 2006), the IRAS 60µm emission (Cox et al. 1990), and radio continuum (Celnik 1985), we assign all X-ray sources within 20 arcmin of the cluster central position (R.A.=06h31m59.s9, Dec.=+04◦55′36′′) as potential NGC 2244 sources. This is indicated by the large circle in Figure 4.1. The resulting cluster extent is consistent with the size of this massive open cluster determined by systematic studies from the All-Sky Compiled Catalog of 2.5 Million Stars (Kharchenko et al. 2005). We remind the reader that the dividing line is not unique; ambiguity of exact physical associations certainly exists for sources located in the interface between the HII region and the molecular cloud to the east, and between the main NGC 2244 cluster and secondary NGC 2237 cluster to the west. In this Chapter we focus on a total of 919 sources located within the NGC 2244 cluster region (as defined above); the rest are presented in Chapter 5 and 6. The 919 valid sources are divided into a primary list of 805 highly reliable sources (Table 4.2) and a secondary list of 114 tentative sources with P 0.001 likelihood of B ≥ being spurious background fluctuations. Table 4.2 and Table 4.3 have formats that are identical to Tables 1 and 2 in Townsley et al. (2006a), Wang et al. (2007a), and Broos et al. (2007). A detailed description of the table columns is given in the table footnotes.

4.2.2 Source Variability One of the most notable characteristics of PMS stars is flaring in the X-ray band, and a few extraordinarily powerful X-ray flares have been reported in several PMS stars (e.g., Imanishi et al. 2001; Grosso et al. 2004; Favata et al. 2005; Getman et al. 2006; Wang et al. 2007a; Broos et al. 2007). A Kolmogorov-Smirnov (K-S) test is performed by AE on each observation in order to evaluate X-ray light curve variability within that observation. Seventy-eight sources display significant variability (PKS < 0.005 in column 15 of Tables 4.2 and 4.3) and 14 of them have more than 200 net counts. Six of these highly variable light curves for sources having more than 500 counts are shown in Figure 4.4. Due to the short duration (20 ks) of most of our Rosette observations, few flares are observed in their entirety. However ObsID 3750 is 75 ks long and completely captures −1 a giant flare from source #634 (average luminosity log Lt,c = 31.3 ergs s ). The count

1 http://www.astro.psu.edu/xray/docs/TARA/ae users guide.html 85 rate during the flare peak is 100 times that of the quiescent level. The estimated peak ∼ −1 luminosity is log Lt,c = 32.3 ergs s . The shape of the flare is not symmetric, with a 1.5 hr rising phase and a 4 hr decaying phase, resembling the fast rise and slow ∼ ∼ decay X-ray flares commonly seen in the COUP young stars (Favata et al. 2005). Its IR counterpart ( 4.3) does not show K-band excess, and the color and magnitude are § consistent with a T Tauri star. The spectrum is very hard (kT 7 keV). Multiple flares ∼ are seen in #691 with different intensities. Source #919 is only covered in the 20ks observation, but it shows a “flat-top” flare, characterized by a fast rise from low flux to high flux and a constant high flux level.

4.2.3 Spectral Fitting The brightest source in the field is the O star HD 46150 (ACIS #373), which has 3588 ACIS counts in 94 ks exposure, or a count rate of 0.12 counts per CCD frame. This is not bright enough to corrupt our spectral fitting, thus it does not warrant correction for photon pile-up (Townsley et al. 2002). Spectral analysis results for the 630 sources with Signif & 2.0 are presented in Tables 4.4 (588 sources; thermal plasma fits) and 4.5 (42 sources; power-law fits). The table notes give detailed descriptions of the columns. Best-fit absorbing column 2 densities range from negligible to log N 23.3 cm− , equivalent to a visual absorption H ∼ of A 120 mag (Vuong et al. 2003). Temperatures range from the softest kT V ∼ ∼ 0.2 keV to the hardest truncated at kT = 15 keV. The range of observed total band (0.5 8 keV) absorption corrected luminosities able to be derived from spectral modeling − −1 is 29.1 . log Lt,c . 32.4 ergs s . Assuming a 2 keV plasma temperature and an average −2 2 AV = 1.5 mag visual extinction (log NH 21.4 cm absorbing column), PIMMS gives ∼ −1 an apparent total band luminosity log L 28.7 ergs s estimated for the faintest on- t ∼ axis detection in Table 4.3. A conservative estimate for limiting sensitivity of the entire 1 NGC 2244 observation is log L 29.4 erg s− ; the exact value depends on off-axis t ∼ location and absorption.

4.3 Identiﬁcation of Stellar Counterparts and Their Properties

4.3.1 Stellar Counterparts Matching and IR diagrams We associate ACIS X-ray sources with optical and near IR (ONIR) sources using 3 positional coincidence criteria, as described in the Appendix of Broos et al. (2007) . The optical and infrared catalogs from recent literature and observations that we adopted for counterparts identiﬁcation include: UBV photometry of NGC 2244/Mon OB2 (Massey et al. 1995 =MJD95), UBVIHα photometry of NGC 2244 (Park & Sung 2002 =PS02),

2 Portable, Interactive Multi-Mission Simulator is software for high-energy astrophysicists, written and maintained by Koji Mukai. See http://heasarc.gsfc.nasa.gov/docs/software/ tools/pimms.html 3 Software implementing the matching algorithm is available in the TARA package at http: //www.astro.psu.edu/xray/docs/TARA/ 86

BVIRHα photometry of NGC 2244 (Berghöfer & Christian 2002 =BC02), the Whole- Sky USNO-B1.0 Catalog (Monet et al. 2003 =USNO), 2MASS All-Sky Catalog of Point Sources (Cutri et al. 2003 =2MASS), and the University of Florida FLAMINGOS Sur- 4 vey of Giant Molecular Clouds . The reference frame offsets between the ACIS fields (astrometrically aligned to the Hipparcos frame using 2MASS sources in the data reduction) and the catalogs are 0.4′′ to MJD95, 0.3′′ to PS02, 0.3′′ to BC02, 0.2′′ to USNO, and 0.2′′ to FLAMINGOS. These offsets were applied before matching sources. Likely associations between ACIS sources and ONIR sources are reported in Ta- ble 4.6; 712 of the 919 ACIS sources (77%) have an ONIR counterpart identified. Since Park & Sung (2002) provide the full list of their high precision photometry, visual photometry is reported in the priority order of PS02, BC02, MJD95, and USNO. USNO photometry is photographic with 0.3 magnitude photometric accuracy (Monet et al. ∼ 2003). JHK magnitudes from FLAMINGOS photometry are reported if available for Chandra sources. The FLAMINGOS photometric data was zero-pointed to 2MASS (see Román-Zúñiga et al. 2007 for details on photometry. A photometric catalog for optical stars in NGC 2244 is presented in RL07 and the complete catalog from the FLAMINGOS survey of the Rosette will be reported in Román-Zúñiga et al. 2007). For areas that were not covered by the FLAMINGOS survey and for bright stars that were saturated (H < 11 mag, Román-Zúñiga 2006), 2MASS photometry is reported. The SIMBAD and VizieR catalog services are used for complementary information. Notes to other published characteristics of the selected sources can be found in the table footnotes. Figure 4.5 shows the NIR J H vs. H K color-color diagram for 617 out of − − 919 Chandra stars with high-quality JHK photometry (error in both J H and H K − − colors < 0.1 mag) listed in Table 4.6. Most Chandra sources are located between the left two dashed lines, the color space associated with diskless young stars (Class III objects), which are reddened by interstellar extinction. We emphasize that while this region is usually filled with field star contaminants in NIR-only studies, here nearly all of these stars are cluster members ( 3.2). A concentration of cluster members subjected § to AV 1 2 mag (assuming late type stars) is apparent, centered at J H=0.25, ∼ − −5 H K=0.8. To the right of this reddening band are 38 K-band excess sources , defined − as stars that have colors (J H) > 1.7(H K) + 2σ(H K). All except three occupy − − − the color space between the middle and right-most dashed lines, which are likely PMS stars with circumstellar accretion disks (Class II objects).

4 PI: Elizabeth A. Lada; details about the FLAMINGOS Survey can be found at http: //flamingos.astro.ufl.edu/sfsurvey/sfsurvey.html. The instrument design and performance of FLAMINGOS are described in Elston et al. (2003). The FLAMINGOS observations of the Rosette Complex fields and IR data reduction are described in Román-Zúñiga (2006). 5 Note that the definition of K-band excess slightly varies when considered by different researchers. In Román-Zúñiga et al. (2007a), an IR-excess star is required to have colors that above J H =0.47(H K)+0.46 (locus of the Classical T Tauri Stars; CTTS), in addition to requiring− that the star− is located to the right of the reddening band for zero age main sequence dwarfs. The region below CTTS locus could include detections of unresolved galaxies. But for the IR counterparts to our X-ray selected sample, this color space has little contamination from galaxies, and likely contain Herbig Ae/Be stars in the young cluster. Therefore no further constraint is applied in defining the K-excess. 87

The three stars (#44, #564, and #805) are located beyond the right-most dashed line. Stars in this domain rightwards of the reddening band are likely surrounded by extended envelopes, and hence classified as candidate Class I protostars (Kenyon et al. 1993; Strom et al. 1995). Sources #44 and #564 are particularly interesting for their large color excess (E(H K) 0.8). Source #44 was identified as a member of NGC − ≥ 2244 (Ogura & Ishida 1981) and assigned spectral type B7Ve (Verschueren 1991). It is further classified as a Herbig Be star in Li et al. (2002) based on the tenuous nebula seen in their KPNO Hα image, large [V-25µm] color, and its confirmed late B spectral type. Source #805 is very faint (J 18.5, close to the FLAMINGOS imaging sensitivity ∼ limit) and may be a low mass protostar or an embedded background object with large H K color (Froebrich et al. 2005). − Figure 4.6 shows the NIR J vs. J H color-magnitude diagram for the same − stars shown in Figure 4.5. Known OB stars are located at the top and reddened from the ZAMS with A 1 mag. Unlike NGC 6357 (Wang et al. 2007a) and other more V ∼ heavily obscured clusters (Wolk et al. 2006; Broos et al. 2007), we do not locate any new candidate massive stars earlier than B0. However two likely new B0-B2 stars (#900 and 919) and a dozen new late B candidates are found. Individual masses and reddening of the lower mass stars can be estimated assuming that they are reddened from the 2 Myr isochrone. Adopting a larger distance d = 1.6 kpc does not have a significant effect here; the mass estimates for the low mass stars are only affected by 0.1 M⊙. The majority of ACIS stars appear to be concen- ∼ trated around 0.7 . J H . 1 and 13 . J . 16, consistent with F, G, K, and M stars − (0.1 . M . 2 M⊙) reddened by 1 . AV . 2 mag. Those showing K-band excess seem to have larger reddening compared to the Class III objects. Sources to the left of the 2 Myr isochrone are probably field stars older than the NGC 2244 population (see 4.3.3). In § addition, to the right of ZAMS track, around twenty stars have inferred stellar masses < 0.1 M⊙. The IR photometry for these J > 17 mag stars may be less reliable, especially the fainter ones that are close to the detection limit. These are likely a mixture of the lowest mass cluster members, background stars, and a few IR luminous extragalactic sources. The high percentage of NIR excess sources among the faintest stars may not be real; Froebrich et al. (2005) have shown that background stars embedded in distant clouds have an overall larger H K color. Contamination by faint background IR sources − is also discussed in Román-Zúñiga (2006). The remaining 200 ACIS sources without matched counterparts are likely to ∼ be a mixture of newly discovered embedded members of the Rosette complex and distant background stars and extragalactic sources (see 4.3.2 and 4.3.3 for estimated § § fractions). Cumulative distribution functions of the median photon energy are shown in Figure 4.7 for sources that have matched counterparts and those that do not have counterparts. The harder median photon energy of sources that do not have counterparts indicates that these sources are deeply embedded or behind the cloud (AV & 10 mag). The new low mass cluster stars reveal themselves because of strong X-ray flares due to magnetic activity. Such X-ray discovered stars are commonly found in the molecular clouds surrounding other clusters (Getman et al. 2005a; Wang et al. 2007a; Broos et al. 2007). Some of these may be very young protostars with local absorption in an envelope −2 or disk; Getman et al. (2007) found in IC 1396N that sources with log NH & 23.0 cm 88 are protostars with dense envelopes. Their spatial distribution (Figure 4.8) shows that they are distributed around the bright stars and along the rims of the Rosette nebula where infrared source identification may be challenging. As emphasized in 1, stellar X-ray emission decays rapidly only after 100 Myr § ∼ and show only a small dependence of X-ray luminosity on disk accretion (Preibisch et al. 2005; Preibisch & Feigelson 2005; Telleschi et al. 2007). Therefore X-ray surveys have very high efficiency in detecting disk-free PMS stars over MS stars, complementing traditional optical and infrared surveys of star forming regions. In the next section we evaluate the level of contamination by extragalactic X-ray sources and Galactic disk stars following the simulations described in Getman et al. (2006) and Wang et al. (2007a). The calculations indicate that . 35 sources are extragalactic ( 4.3.2), 20 are foreground § ∼ and 16 are background field stars ( 4.3.3). These 70 contaminants constitute 8% ∼ § ∼ ∼ of the 919 ACIS sources.

4.3.2 Extragalactic Contaminants Following Getman et al. (2006), we construct Monte Carlo simulations of the extragalactic population by placing artificial sources randomly across the detector. We draw incident fluxes from the X-ray background log N log S distribution (Moretti et al. − 2003), and power law photon indices for the sources are assigned consistent with flux- dependencies described by Brandt et al. (2001). The spectrum of each simulated source 2 was passed through a uniform absorbing column density log N 21.9 cm− , the HI H ∼ column density through the entire Galactic disk in the direction of NGC 2244 (Dickey & Lockman 1990). After applying local background levels found in our ACIS image, we calculate the photometric significance of each fake source and then reject the weak extragalactic sources that would have fallen below our source detection threshold. The simulations suggest that 80 extragalactic sources may be detected in our ACIS-I field ∼ and 35 may have photometric significance Signif 2.0 (column 12 in Table 4.2). ∼ ≥ The true number is probably smaller as we did not account for the patchy distribution of molecular cloud material. The best candidates for the extragalactic contaminants are those sources whose X-ray spectra are best fit by a power law, do not have bright ONIR counterparts, and do not display the characteristic PMS X-ray flares.

4.3.3 Galactic Stellar Contamination Monte Carlo simulations of the Galactic stellar population expected in the direction of our ACIS ﬁelds (l = 206.3, b = 2.1) were examined, based on the stellar − 6 population synthesis model of Robin et al. (2003; henceforth the Besan¸con model) . In addition to the smooth absorption component provided in the model, we added a local absorption at d = 1.4 kpc with a low A 1 mag, as inferred from the average NIR V ∼ reddening of Chandra sources shown in Figure 4.6. Within the ACIS ﬁeld-of-view, the

6 These calculations are made with the Web service provided by the Robin et al. (2003) group at Besan¸con at http://bison.obs-besancon.fr/modele. 89

Besan¸con model predicts 850 foreground MS stars (d < 1400 pc) and 4300 back- ∼ ∼ ground MS stars, giants, and sub-giants within the FLAMINGOS imaging sensitivity limit around J < 19 mag. About 25% of the foreground stars in the Besan¸con simulation have ages less than 1 Gyr, the younger population that could be detected in X-ray surveys. Most reside at distances from 0.8 to 1.4 kpc; 54% are M stars, 28% are K stars, 13% are G stars, ∼ ∼ ∼ and 3% are more massive. Following Getman et al. (2006), we convolved the Besan¸con model populations with the X-ray luminosity functions of stars in the solar neighborhood measured from ROSAT surveys (Schmitt et al. 1995; Schmitt 1997; Hünsch et al. 1999). Luminosities were adjusted to account for the different ROSAT and Chandra spectral bands following the stellar hardness-luminosity relation (Güdel et al. 1998). Following the same procedure we used for extragalactic sources ( 4.3.2), we applied our ACIS § detection process to these simulated field stars. A typical Monte Carlo run predicts that 16 20 foreground stars will be detected in the ACIS-I exposures of NGC 2244. − Five optical counterparts to our X-ray stars are identified as foreground stars based on their known spectral types and positions in photometric diagrams. These are #321, 429, 583, 627, and 678. Proper-motion data (Marschall et al. 1982; Dias et al. 2006) also suggest that they have low probability of being member stars of the cluster. In the NIR color magnitude diagram, there are 30 stars located between 2 Myr isochrone ∼ and the ZAMS track. Those close to the 2 Myr isochrones are still consistent with being members due to uncertainty in the age and distance of the cluster. About fifteen stars close to the MS track are more likely to be unrelated foreground main sequence stars. Available X-ray absorption columns derived from spectral fits mostly require −2 log NH = 20.0 cm , suggestive of being foreground candidates. Their probabilities of being members derived from proper motion data range between 0% and 80%, which are less conclusive. With the above five confirmed foreground stars, altogether twenty optical 7 stars are noted as the best candidates for foreground stars in the Table 4.6 footnotes. This number is quite consistent with the Besan¸con simulated foreground population. The Besan¸con model is again convolved with XLFs to simulate the number of stars behind the Rosette star forming region that may enter our X-ray sample. The model predicts 18% F dwarfs, 41% G dwarfs, 33% K dwarfs, 1%M dwarfs, and ∼ ∼ ∼ ∼ 6% giants. We use the dwarf XLFs established for the solar neighborhood (Schmitt ∼ et al. 1995; Schmitt 1997) and adopt the XLF for giants obtained from Table 2 of Pizzolato et al. (2000). Typical runs of simulations result in 11 dwarfs and 5 giants ∼ ∼ that are detectable in our Chandra observation. To the left of the ZAMS track in the 8 CMD and for J > 18, there are 16 stars whose locations match the Besan¸con model predicted background population of MS stars, subgiants, and giants, which we note as best candidates for background stars in Table 4.6.

7 They are associated with Chandra sources #49, 67, 96, 105, 129, 163, 321, 429, 552, 583, 603, 627, 678, 743, 757, 785, 839, 854, 859, and 889. 8 They are #68, 157, 158, 181, 226, 253, 323, 325, 347, 409, 442, 583, 609, 747, 835, and 880. 90

4.4 Global Properties of the Stellar Cluster

4.4.1 X-ray Luminosity Function and Initial Mass Function As noted in Feigelson et al. (2005), the XLF (which is directly measured here) can be considered to be the convolution of the IMF (which is unknown) and the X-ray–Mass Luminosity (L M) correlation (which is measured in the COUP studies; Preibisch et x − al. 2005). Using the best-studied Orion Nebula Cluster XLF (COUP XLF) and IMF as a calibrator, the NGC 2244 XLF can be used to probe the IMF of the stellar cluster and to estimate the total X-ray emitting population. Such a population analysis has been made for Cep OB3b, NGC 6357, and M17 (Getman et al. 2006; Wang et al. 2007a; Broos et al. 2007). In the following XLF analysis, we use the hard band XLF rather than the total band XLFs, since the unknown soft component of heavily absorbed X-ray sources can introduce a large uncertainty in both the observed and absorption corrected total band X-ray luminosity. By counting the number of sources in different X-ray luminosity bins, we construct the absorption corrected hard band (2–8 keV) XLF (Lh,c) for all unobscured NGC 2244 X-ray sources (MedE 2.0 keV) with derived X-ray luminosities in Figure 4.9a. Here ≤ we exclude the five known foreground stars, and OB stars with spectral types earlier than B3 to be consistent with the Orion cool stars sample. The absorption corrected hard band fluxes (Fh,c) derived from XSPEC spectral fitting are used to obtain luminosities assuming a distance of 1.4 kpc. As the template, we also show the XLF of the COUP unobscured population (839 cool stars; Feigelson et al. 2005). The NGC 2244 XLF is largely consistent with the COUP XLF, suggesting an NGC 2244 population comparable to the ONC. However, at log Lh,c & 29.8 the slope of the NGC 2244 XLF seems steeper than the COUP XLF. The apparent steeper slope is not an artifact of our distance estimate or detection completeness limit. We comfortably detect 15 count sources at any off-axis location; the X-ray luminosity of the detected −1 sources is roughly log Lt 29.4 ergs s (corresponding to 0.5M⊙ in mass from the ∼ ∼ −1 L M relation; Preibisch et al. 2005) and L 29.2 ergs s . The luminosity bins x − h,c ∼ with the steep slope seen in the NGC 2244 XLF are much brighter than the completeness limit. There are two possibilities intrinsic to the NGC 2244 cluster that may be responsible for the apparent steeper slope: (i) The NGC 2244 population is the same as the COUP population, but there is an excess of 50 stars in the luminosity range ∼ 30.0 . log L . 30.4 (solar mass stars as inferred from the L M relation); or (ii) The h,c x − NGC 2244 population is 1.2 times larger than the COUP population, but NGC 2244 ∼ is 20 stars deficient in stars with 30.4 . log L . 31.0 (intermediate mass stars). In ∼ h,c either case, the NGC 2244 XLF deviates from a scaled version of the COUP XLF in a manner similar to that seen in the Cep B/OB3b field studied by Getman et al. (2006). To improve the statistics, remove possible binning effects, and further investigate how the NGC 2244 XLF compares to that of the ONC and other clusters, we derive the cumulative distribution of X-ray luminosities for the unobscured NGC 2244 X-ray sources as well as the unobscured COUP, Cep B/OB3b (age 1-3 Myr, Getman et al. ∼ 2006), NGC 6357 (age 1 Myr, Wang et al. 2007a), and M17 (age 1 Myr, Broos et al. ∼ ∼ 91

2007) populations. The resulting cumulative XLFs are shown in Figure 4.9b. To avoid confusion caused by incompleteness in detections, we cut off the cumulative XLFs at the corresponding completeness limit for each region (see Figure 4.9). At a given X-ray luminosity above the completeness limit, the ratio between the cumulative numbers of sources from two populations reflects the relative scaling between the two populations that are more luminous than this limit. As a result, the unobscured population of NGC 6357, M17, and Cep B, is 5 times, 3 times, and 0.4 times of the size of unobscured ∼ ∼ ∼ 9 ONC population, respectively. These are consistent with previously reported values. The cumulative X-ray luminosity function of NGC 2244 closely follows the COUP XLF, although deviation found in Figure 8 (deficit at 30.4 . log Lh,c . 31.0 and excess at log Lh,c . 30.4) can still be clearly seen. Depending on the treatment of this deviation, the NGC 2244 population ranges from 1.0 to 1.2 times the ONC population. A few exercises were done to examine the possible excess of sources with 30.0 . −1 log Lh,c . 30.4 ergs s . To identify a previously unknown cluster in the field that may contribute extra stars to NGC 2244, we inspect the spatial distribution of sources that have luminosities in the excess bins, but do not find any apparent clustering. To test whether this excess comes from contamination of X-ray bright non-members, we remove candidate contaminants (foreground stars, background stars, extragalactic sources) as suggested in 4.3.2 and reconstruct the XLF. The number drop mainly appears in § −1 low luminosity bins while the excess is still significant at 30.0 . Lh,c . 30.4 ergs s . Therefore we conclude that excluding candidate contaminants would not alter the XLF, since they do not contribute much to the high luminosity bins that characterize the X-ray emitting population.

4.4.2 Initial Mass Function and K-band Luminosity Function To examine whether the deviation in the XLF is a reflection of an intrinsic anoma- lous IMF of the NGC 2244 cluster, we perform two tests on the IMFs using NIR data. One experiment is to use the location of X-ray stars in the IR color magnitude diagram to derive their masses and construct an approximate IMF. The exact mass will not be as accurate as measured from the spectral types and an HR diagram, but their statistical distribution should be sufficient for our interest here. If there is indeed an excess of X-ray 1 sources with log L 30.0 30.4 ergs s− (option i in 4.1), from the empirical L M h,c ∼ − § x − relation (Preibisch & Feigelson 2005), we would expect to see an excess of stars around a solar mass. Figure 4.10 shows the IMF constructed from NIR estimated masses for unobscured COUP stars and for NGC 2244 stars. No excess is apparent for stars in the solar mass range, after the ONC IMF (Muench et al. 2002) is scaled to match the NGC 2244 IMF. Instead, a deficit of intermediate mass stars around 2 3 M⊙ in the X-ray − selected sample is apparent (option ii in 4.1). §

9 In NGC 6357, log L is 0.5 dex higher than the other clusters because candidate OB stars h,c ∼ are included in the sample (Wang et al. 2007a). Note that although the unobscured population of NGC 6357 is larger than that of M17, the obscured population in M17 is signiﬁcantly larger than that of NGC 6357 (Broos et al. 2007), which make the estimated total populations of the two clusters comparable. 92

A second test is to obtain a statistical sample of cluster members and construct a KLF (Lada & Lada 1995) following the Cep B/OB3b KLF analysis in Getman et al. (2006). We used 2MASS Ks data to constuct the KLF since it was spatially complete toward NGC 2244 and it is photometrically complete to a similar mass limit as our Chandra data for the Rosette. A control field from 2MASS centered at (α, δ) =(6h30m00s, +3◦45′00′′) [J2000] is used for background population subtraction (extinction toward this control field and in the foreground of NGC 2244 is low. See Li 2005 for its 2MASS color color diagram). The resulting KLF is similar to that derived by Li (2005) with a power law slope (d log N(K )/dK ) 0.3. To get the IMF, we convert s s ∼ Ks magnitude to log M using the 2 Myr theoretical isochrone. The relation between Ks and log M from the theoretical isochrone (Siess et al. 2000) can be approximated well with a power law as demonstrated by Getman et al. (2006). The resulting IMF derived from K-band star counting (dot-dashed line in Figure 4.10) is consistent with the IMF estimated from NIR properties of Chandra sources (solid histogram in Figure 4.10). This KLF suggests that our X-ray sample is largely complete down to 0.5M⊙, ∼ consistent with the mass limit estimated from the X-ray completeness limit. It also suggests that our X-ray selected sample is missing 20 intermediate-mass stars around ∼ 3 M⊙. It is not surprising that the detection efficiency of intermediate-mass stars in the X-ray band is not as deep as in optical or IR (Schmitt et al. 1985). The X-ray production mechanism is not well understood for stars in the intermediate-mass range ( 4.5.2). In § many cases of X-ray detections of intermediate-mass stars, the X-ray emission in fact comes from a low mass companion to the intermediate mass star. Compared to the ONC, some of the intermediate-mass stars in NGC 2244 may be X-ray quiet because of the absence of low-mass companions. This is reflected as the deficit of log Lh,c 30.4 31.0 1 ∼ 1 − ergs s− sources in the XLF. The “excess” of log L 30.0 30.4 ergs s− sources h,c ∼ − would then be the result of a slightly larger population of NGC 2244 stars compared to that of ONC. Based on all the above analysis, we conclude that the unobscured X-ray emitting NGC 2244 population is about 1.2 times larger than the known unobscured population 1 in ONC, or 1000 stars with log L > 27.0 ergs s− (the COUP detection limit). The ∼ t obscured population, estimated from similar XLF scaling, is 500 stars. Given that the ∼ total COUP sample accounts for 75% of the ONIR sample of the ONC in Hillenbrand ∼ & Hartmann (1998), the total number of NGC 2244 X-ray stars is around 2000. Li (2005) gives a census of 1900 NGC 2244 members estimated from the spatially complete ∼ 2MASS analysis. Although the massive end of the stellar content in NGC 2244 was well-known from early studies (e.g., Ogura & Ishida 1981), the low-mass populations were poorly identified. Perez (1991) suggested that there may not be M < 4 M⊙ stars in the NGC 2244 cluster. Perez (1991), Massey et al. (1995) and Park & Sung (2002) investigated the IMF for NGC 2244 in optical, and reported a top heavy IMF with a flat power law slope Γ = d log N(log M)/d log M 0.7, although Park & Sung (2002) are cautious ∼ − due to the incompleteness of their incomplete intermediate- and low-mass population. Indeed, their optical sample becomes incomplete for stars with stellar mass lower than 3M⊙. In comparison to the traditional IMF studies, our high sensitivity X-ray sample ∼ of stars largely benefits from our robust membership criteria and our ability to identify 93 the low-mass members in a reliable manner. The IMF slope obtained from X-ray selected 10 members which are nearly complete to 0.5M⊙ gives Γ 1.1 instead of Γ 0.7, ∼ − ∼ − consistent with the Orion IMF. Indeed, our X-ray sampled cluster yields a few hundred more low mass stars than expected from the previously suspected top-heavy IMF.

4.4.3 Morphology and Substructures The morphology of young clusters, including dependency on stellar mass, provides clues for cluster formation and dynamical evolution. As morphological studies based on optical or infrared samples are complicated by patchy extinction, nebular contamination, confusion with field stars, and bias towards stars retaining protoplanetary disks, the spatial distributions of X-ray identified stars in populous young clusters should be excellent laboratories to explore their origins and dynamical evolution. For example, an aspherical shape or clumpy distribution would reflect unequilibrated initial conditions while a spherical shape with mass segregation would indicate well-developed virializa- tion (Clarke et al. 2000). In the Orion A cloud, the ONC, NGC 2024, and associated molecular filaments have flattened shapes (Lada et al. 1991; Feigelson et al. 2005) which have been attributed to global gravitational collapse of an elongated cloud (Hartmann & Burkert 2007). In contrast, the rich NGC 6357 and M 17 clusters appear spherical but with subclusters that may reflect distinct (perhaps triggered) subcluster formation (Wang et al. 2007a; Broos et al. 2007). The absence of mass segregation can either reflect a young stellar system that has not yet achieved dynamical relaxation, or a mature system where many of its massive members have been ejected by few-body interactions in the core (Pflamm-Altenburg & Kroupa 2006). It has been recognized that the apparent center of the large-scale annulus (see Fig- ure 1b) defining the optical Rosette Nebula at (α, δ) = (6h31m56s, +04◦59′56′′) (Ogura & Ishida 1981) is offset from the center of the IR surface density distribution at (α, δ) = (6h31m59.9s, +04◦55′36′′) in the 2MASS study by Li (2005). He interprets this offset as a projection effect where the Rosette Nebula resembles a tilted cylinderical cavity in the molecular cloud (see models in Celnik 1986). The distribution of the Class II sources in recent Spitzer/IRAC-MIPS survey of NGC 2244 (Balog et al. 2007) shows good agreement with the 2MASS data. Here our Chandra data show that the stellar concentration around HD 46150 is offset from the larger stellar distribution, irrespective of its relation to interstellar matter. The off-center massive star will be further discussed in 4.4.5. § A ‘center’ must be defined for structural analysis. In NGC 2244, the highest concentration of X-ray stars appears around the second most massive star, HD 46150 (O5V), at (α, δ) = (6h31m55s, +04◦56′34′′). We therefore treat HD 46150 to be the center for our radial profile analysis of this high density region in 4.4.4, although we § are aware of the asymmetrical distribution of X-ray stars as noted above: HD 46150 lies about 2′ northwest of the IR center defined in the 2MASS study (Li 2005).

10 Using lower envelope around the correlation between X-ray luminosity and mass in Figure 3 of Preibisch et al. (2005), we estimate that our X-ray completeness limit corresponds to detections of > 90% of the stars with mass > 0.5M⊙. If the median correlation is used, the mass completeness limit is > 0.2M . However, 30% stars may be missing in the detections. ⊙ ∼ 94

Figure 4.11 shows a smoothed map of the stellar surface density for 572 lightly- obscured (median photon energy . 2keV) NGC 2244 X-ray stars. This map of smoothed spatial distribution is constructed following Wang et al. (2007a) and Broos et al. (2007). A similar smoothing technique has been applied to 2MASS data to identify large-scale structure of the entire Rosette Complex (Li 2005; Li & Smith 2005a,b). A 20′ 20′ ∼ × grid is created to cover the stellar positions, and at each position the total number of sources within a 0.5 arcmin radius sampling kernel is counted to estimate the smoothed stellar density. Only sources covered by both ObsID 1874 and ObsID 3750 are considered to guarantee roughly equal X-ray sensitivity throughout the field. The heavily-obscured sources are omitted because many of them are expected to be background AGNs. The cluster shows an approximately spherical structure that extends 8′ (3.2 pc) in diameter, centered at (α, δ)=(06h31m59s, 04◦55′30′′). This center, as well as the large scale structure and substructure seen in the X-ray-sampled cluster (Figure 4.11), are in good agreement with the results derived from the surface density of IR-excess stars in the FLAMINGOS study (Román-Zúñiga et al. 2007a). It also matches the center defined from the 2MASS star-count (Li 2005) and the center defined from the Spitzer Class II sources (Balog et al. 2007). The large-scale north-south asymmetry can be attributed to the off-center placement of HD 46150. The primary concentration is seen around HD 46150. Five of these stars were noted by Sharpless (1954) as a visual compact subcluster, but we find 50 stars extending to a radius of 1′ around this massive star. ∼ 2 The central surface stellar density here is 700 stars per pc ; recall that this value is ∼ restricted to stars with masses above 0.5 M⊙ due to X-ray sensitivity limits. ∼ A secondary density enhancement of 15 X-ray sources (about 3σ enhancement) ′ ∼ ◦ ′ ′′ is seen 3 south of HD 46150 at (α, δ)=(6h31m56s, +04 54 10 ). The local density peak ′′ here has six stars tightly clustered within 20 ; it is also apparent in the optical Hα image and the 2MASS-Ks image (Figure 4.12 and Li 2005). Assuming they are lightly-obscured late-type stars, the reddening indicated by their NIR colors is A 1 mag, similar to V ∼ that of NGC 2244 cluster. Their NIR estimated spectral types range from F to M type if they are located at the same distance as NGC 2244. The existence of both substructures, around HD 46150 and 3′ to the south, is direct evidence that the NGC 2244 cluster has not attained dynamical equilibrium. But perhaps most remarkable is the absence of companions around the most massive cluster member HD 46223 (O4V). Nine X-ray sources lie within 1′ of HD 46223 compared to 50 around HD 46150. One possible explanation for the isolation of HD 46223 is that it was ejected by dynamical interactions within the HD 46150 subcluster. However, it does not not exhibit high proper motion ( 4.4) and it seems unlikely that such a massive § member, rather than less massive members, would be ejected at high velocity. We note, however, that an O4 supergiant has been reported to be probably ejected from Cyg OB2 (Comeron & Pasquali 2007).

4.4.4 Radial Density Profile The radial density profile for the NGC 2244 cluster, centered at the stellar density peak around HD 46150, is shown in Figure 4.13 with comparison profiles from optical/NIR and X-ray studies of the ONC (Hillenbrand & Hartmann 1998; Feigelson 95 et al. 2005) and from our X-ray study of Pismis 24 in NGC 6357 (Wang et al. 2007a). The radial profile of NGC 2244 has two distinctive components: a powerlaw structure around HD 46150 extending 1.5′, and a structure with a flat core and steeper dropoff extending from 1.5′ to 8′. The power law structure is centered on, but is much more extended than, the 20′′ X-ray resolved subcluster around HD 46150 shown in Figure ∼ 3. NGC 6357, and perhaps the ONC, have a similar radial profile with approximately the same powerlaw slope (Figure 4.13). Figure 4.14 shows the inferred radial density profiles in physical size units (parsecs) for the three clusters, and where the star densities are scaled to their estimated true densities based on comparison of the XLFs shown in Figure 9b. The stellar density of NGC 2244 has been scaled to 1.2 times the ONC population ( 4.4.1), and NGC 6357 § −2 to 5 times the ONC population (Wang et al. 2007a). Omitting the central r powerlaw structures, the profiles of these two clusters can be modeled as isothermal spheres with an estimated core radius r = 1.2 1.4 pc. c − 4.4.5 Mass Segregation The concentration of massive cluster members at the center and lower mass members at larger radii from the cluster center is commonly observed in rich young star clusters (e.g. Carpenter et al. 1997; Hillenbrand & Hartmann 1998; Adams et al. 2001). Schilbach et al. (2006) investigate mass segregation in over 600 open clusters with a wide range of ages. For their youngest clusters with ages 5 Myr, some show mass segregation ∼ whereas others do not. Mass segregation can occur as a natural consequence of dynamical relaxation. De- tails of the process have been debated. For the ONC Trapezium, some researchers argue that the dense collection of 10 OB stars are imprints of initial conditions (Binney & ∼ Tremaine 1987; Bonnell & Davies 1998), while others argue that the core has collapsed and many OB stars have been ejected (Pflamm-Altenburg & Kroupa 2006). McMillan et al. (2007) suggest a model of sequential merger of mass segregated subclusters. Bon- atto et al. (2006) found that the M 16 cluster with age 1.3 Myr has an overall relaxation ∼ timescale around 20 Myr, yet shows some degree of mass segregation at this young ∼ age. In NGC 2244 in the Rosette Nebula, the O stars are not highly concentrated, as shown in Figure 4.15a (see also Figure 1a). The earliest O-type (O4V) star in the cluster, HD 46223, has a rather puzzling location in the cluster near the southeast boundary of 1 1 the nebula. Its proper motion is fairly small (µ = 0.2 mas yr− , µ = 0.4 mas yr− ; α − δ Zacharias et al. 2004) which does not suggest ejection from the cluster center. The issue of mass segregation has not been previously investigated for this 2 Myr old cluster in the literature, mainly because the low-mass population was not adequately identified. Figure 4.15b shows the cumulative radial distributions for the massive stars with NIR-estimated masses M & 8M⊙ and for the X-ray identified low mass stars (M . 2M⊙). The distributions appear very similar, and a Kolmogorov-Smirnov test of the two distributions does not show significant difference. Thus, mass segregation is not present in NGC 2244. 96

We estimate a two-body dynamical relaxation time trelax for the NGC 2244 cluster (e.g., Bonatto et al. 2006): t (N/8 ln N) t where t = 2R/v is the relax ≈ × cross cross disp characteristic crossing time for a star to travel through the cluster with radius R and velocity dispersion vdisp. Adopting R 4 pc from the full cluster extent in Figure 4.14, ∼ 1 a rough estimate for the unmeasured velocity dispersion v 3 km s− (Binney & disp ∼ Tremaine 1987), and N 2000 stars ( 4.4.1), we obtain t 30 Myr for NGC 2244. ∼ § relax ∼ As the cluster age is < 10% of this relaxation time, no signiﬁcant mass segregation is expected from two-body dynamical interactions. If we consider only stars within the estimated core radius r = 1.2 pc, then t 9 Myr which is still considerably larger c relax ∼ than the age of the cluster. The absence of mass segregation is thus consistent with standard dynamical theory, and implies that NGC 2244 (unlike some other clusters) was not formed with a central concentration of massive stars. The main challenge for explaining the dynamical state of NGC 2244 is the diﬀerence between the dominant member HD 46150, which has a rich compact subcluster, and HD 46223 which is mostly isolated.

4.4.6 X-ray Stars with Infrared Excess Disk X-ray selected samples have several advantages over optical and IR samples (see review by Feigelson et al. 2007). X-ray emission arises from stellar magnetic activity 1 4 which is enhanced 10 10 above main-sequence levels for stars during the entire age − range of interest (< 0.1 to > 10 Myr), thus X-ray survey suffer only a small number of field and extragalactic contamination, which are usually identifiable ( 3.2). They 11 § naturally deliver a nearly disk-unbiased sample of young stars . The main disadvantage of typical X-ray surveys is their incompleteness in detecting the lowest mass objects, which can be identified in high-sensitivity IR images. The complementary nature of the Chandra and IR data will provide the best census to date for the young stellar population of this region. In this subsection we focus on the X-ray selected sample of stars with IR-excess. The spatial distribution of 38 X-ray stars with K-band excesses attributed to inner protoplanetary disks ( 4.3.1) is shown in Figure 4.16. Ten of these young stars § cluster around HD 46150 (O5V), and a few are around HD 259135 (B0.5V). A deficit of color-excess stars in the northern part of the nebula is seen: none of them are located in the northern part of the cluster where HD 46149 (O8.5V) is located. This region is also void of dust emission in the IRAS image (Cox et al. 1990). The deficit of disk stars cannot be attributed to photoevaporation by OB stellar UV radiation and winds since a grouping is found around HD 46150. We consider two explanations for

11 There may be additional complications: X-ray selected PMS samples suffer a small bias against accreting stars in the 0.5 8 keV band because Class II systems are on average 2 times fainter than Class III systems− (Preibisch et al. 2005; Telleschi et al. 2007), and there∼ may also be a small bias toward accretion systems in the soft < 1 keV band due to emission at the accretion shock. However accretion variations do not cause X-ray variations in the Chandra band (Stassun et al. 2006). These are minor effects considering that the X-ray luminosity function −1 spans 28 < log Lx < 32 ergs s and it is dominated by flare emission. See discussion in Feigelson et al. (2007). 97 this asymmetry. First, Li (2005) suggested that the Rosette Nebula is open 30◦ north of the line of sight. As gas and dust stream away from the HII region (as in M 17, Townsley et al. 2003), the stars in this region may have undergone a faster inner disk dissipation so that their disks no longer show K-band excess. Second, the star formation in NGC 2244 may have proceeded over a considerable time span along the north-south direction with the older population in the northern region. Similar spatial-age patterns have been found in other young clusters (e.g., Cep OB3b, Burningham et al. 2005). For the 2 Myr old NGC 2244 cluster, using our X-ray selected sample with ∼ high JHK photometric quality (largely complete to 0.5 M⊙), we derive an overall disk frequency of 6% for Chandra stars with mass M & 0.5M⊙ (assuming a presence of 20 ∼ ∼ foreground field stars). The disk fraction is 10% for stars with mass M & 2.0M⊙ and ∼ 5% for stars with mass 0.5M⊙ . M . 2.0M⊙. Using a large Chandra sample similar to ∼ that studied here, Wang et al. (2007a) reported a low fraction of K-band excess among intermediate-mass stars in the young massive star forming region NGC 6357 ( 1 Myr ∼ old) and a similar result is reported for M 17 ( 1 Myr old) in Broos et al. (2007). These ∼ are consistent with the findings that optically thick circumstellar disks are already rare among the intermediate-mass PMS stars with ages less than a few Myr and suggest that the disks are short lived for the massive stars (e.g., Hillenbrand et al. 1993; Natta et al. 2000). It has been suggested that the disks around earlier type stars may evolve faster than around later type stars based on studies of the IR-excess fraction as a function of spectral type in a few clusters (Lada et al. 2000; Haisch et al. 2001). Li (2005) used the 2MASS NIR sample after background population subtraction to derive a disk fraction of 20.5% above mass 0.8M⊙ for NGC 2244. The discrepancy ∼ ∼ between our results reflects the different criteria in selecting excess sources: their NIR excess sources included 2MASS sources with large photometric error. With high quality FLAMINGOS JHK photometry data, Román-Zúñiga et al. (2007a) derive a lower IR- excess fraction of 10% for K < 15.75 stars, although it remains slightly higher than the ∼ IR-excess fraction in our X-ray-selected sample. Their sample covers a larger mass range, probably down to 0.1-0.2 M⊙ depending on extinction (Román-Zúñiga et al. 2007a). The Spitzer survey of NGC 2244 covering 3.6 µm to 24 µm by Balog et al. (2007) estimates that the overall disk fraction in the cluster is 44.5%. The discrepancy between our results is mainly due to the fact that mid-IR Spitzer observations are much more sensitive to circumstellar dust than K-band excesses. The overall IR-excess fraction among NGC 2244 Chandra-selected stars will be studied with a deep Spitzer survey (PI: Bouwman) and compared to IR-only determined excess fraction, aiming to evaluate the different samples of young stars selected through IR colors and those identified in X-rays. The overall low disk frequency seen here at a cluster age of 2 Myr may imply a faster disk dissipation while the cluster is immersed in the hostile environment of UV radiation and strong stellar winds of many massive stars. In clusters of similar age but without presence of O stars, the K-band excess fractions appear higher (e.g., 20% in the 2.3 Myr old IC 348, Lada & Lada 1995). A number of observational and theoretical studies have already demonstrated the photoevaporation of disks by external radiation (e.g., O’dell & Wong 1996; Johnstone et al. 1998; Hollenbach et al. 2000; Throop & Bally 2005). Balog et al. (2006) presented 24µm images of three protoplanetary disks being photoevaporated around high mass O stars, including one disk close to the O5 star HD 98

−10 −8 −1 46150 in NGC 2244 with an estimated mass loss rate 10 10 M⊙ yr . We do not − detect X-ray emission from the IR point source in the cometary structure. Based on the Spitzer identified Class II and Class I sources, it is further suggested that the effect of massive stars on the circumstellar disks is significant only in the immediate vicinity of the hot stars (Balog et al. 2007).

4.5 X-rays across the Mass Spectrum

4.5.1 X-rays from Massive Stars One of the important discoveries of early Einstein observations was the soft X-ray emission from individual early type O stars (Harnden et al. 1979). Most O-type stars were 31 33 found to be soft X-ray emitters (kT < 1 keV) with X-ray luminosities Lx 10 10 −1 ∼ − erg s , and a canonical relation between X-ray luminosity and bolometric luminosity 7 of L /L 10− was proposed and confirmed from Einstein and ROSAT observations x bol ∼ (Pallavicini et al. 1981; Chlebowski et al. 1989). Berghoefer et al. (1997) extended the same relation down to stars of later spectral type (B1–B1.5). Recent Chandra studies of O7–B3 stars in Orion (COUP) found both a soft wind-emission component and a hard flaring component in many OB stars, and a larger dispersion was found for late O and early B stars (log(L /L ) 4 to 8; Stelzer et al. 2005). However, when x bol ∼ − − only considering X-ray emission in the 0.5–2.5 keV band, Sana et al. (2006) derived a tight scaling law log(L /L )= 6.91 0.15 for O-type stars with a deep XMM-Newton x bol − ± observation of NGC 6231. Wind-shock models were developed to explain the X-ray emission from massive stars, where small-scale instabilities in radiatively-driven stellar winds from massive stars produce shocks (Lucy & White 1980; Owocki et al. 1988; Owocki & Cohen 1999). To account for the observed X-ray emission line profiles and hard, variable continuum emission (e.g., Corcoran et al. 1994; Evans et al. 2004; Waldron et al. 2004; Stelzer et al. 2005), more complex models were invoked such as the magnetically channeled wind shock model (Babel & Montmerle 1997b,a; ud-Doula & Owocki 2002). Gagnéet al. (2005) shows that the magnetically channeled wind shock model with strong line-driven winds can adequately reproduce both the soft and the hard components in Chandra grat- 1 ing spectra of θ Ori C (O6V). In some cases, the anomalously hard and luminous X-ray 33 1 component (kT > 10 keV and L 10 ergs s− ) implies close binarity, as powerful h ∼ winds in two massive components collide and shock to produce very high energy X-rays (e.g., Pollock et al. 2005; Skinner et al. 2006; Broos et al. 2007). Schulz et al. (2006) 2 observed a large X-ray outburst in θ Ori A, which can be attributed to reconnection events from magnetic interactions between the binary stars. Due to its richness in population, NGC 2244 offers an excellent opportunity to study X-ray emission in OB stars. Table 4.7 summarizes the detection/non-detection of O and early B-type stars in our observation, along with their optical, IR, and X-ray properties. Chandra spectra are shown in Figure 4.17. We detected all 9 OB stars with spectral types B0.5 or earlier that were in the field. The two early O stars in NGC 2244 32 1 exhibit soft (kT < 1 keV) and strong (L 10 erg s− ) X-ray emission as expected in x ∼ the classical wind-microshock regime. However, we only detected 6 out of 14 B stars with 99 spectral types B1–B3. This low X-ray detection rate among B stars is consistent with other recent Chandra and XMM-Newton observations of massive star forming regions (Wang et al. 2007a; Broos et al. 2007; Sana et al. 2006). The X-ray emission from early B stars is consistently harder than that from the O stars (higher kT in Table 4.7), which suggests that unseen late-type companion stars rather than the B star itself is responsible (Stelzer et al. 2005). The earliest exciting star in this complex, HD 46223, is of spectral type O4V (Walborn et al. 2002). Its X-ray spectrum is adequately fit by a soft kT = 0.3 keV single 21 2 temperature plasma subjected to N = 4 10 cm− absorption (Figure 4.17). HD H × 46150 (O5V) is the visually brightest early type star in the cluster. A two temperature plasma model fit (kT1 = 0.2 keV, kT2 = 0.6 keV) is needed to describe the X-ray 21 −2 spectrum, with an absorption column NH = 2.5 10 cm . Their X-ray luminosities 32 −1 × are similar, L 2.5 10 ergs s . The light curves of the O-stars are examined and t,c ∼ × no variability is suggested by the K-S statistics. Berghöfer & Christian (2002) reported the X-ray and optical luminosities of NGC 2244 early type stars from ROSAT PSPC and HRI observations and BVI photometry, 7 and concluded that they are consistent with the canonical relation L /L 10− . The x bol ∼ Lx/Lbol relation for OB stars in NGC 2244 is shown in Figure 4.18. Statistical tests for the correlation between Lx and Lbol were performed using the ASURV survival analysis package (Isobe et al. 1986). To take into account the available upper limits, the general- ized Kendall’s tau correlation test for censored data is adopted. The null hypothesis (a correlation is not present) probability is P < 0.01%, supporting a significant L L x − bol correlation. As shown in Figure 4.18 as well as Table 4.7, the NGC 2244 O stars closely 7 follow the X-ray to bolometric luminosities L /L 10− ratio, although the B spectral x bol ∼ type stars show larger (yet still < 0.5 dex) scatter. For comparison, previously reported Lx vs. Lbol values for additional OB stars from the massive star forming regions Orion (1 Myr; Stelzer et al. 2005), NGC 6357 (1 Myr; Wang et al. 2007a), and M17 (1 Myr; Broos et al. 2007) are also shown in Figure 4.18. The overall scatter in Lx/Lbol is considerably smaller for O stars from different clusters ( 2 orders of magnitude) than for B stars ( 4 ∼ ∼ orders of magnitude).

4.5.2 X-rays from Intermediate Mass Stars X-ray emission from intermediate-mass stars with spectral types mid-B to A is unexpected since no X-ray production mechanism is known; they lack strong stellar winds and convective surfaces (Berghofer & Schmitt 1994; Berghoefer et al. 1997; Stelzer et al. 2003, 2006a). However X-ray detections of PMS intermediate-mass stars known as Herbig Ae/Be stars are widely reported (e.g., Zinnecker & Preibisch 1994; Damiani et al. 1994; Berghoefer et al. 1996; Stelzer et al. 2005; Hamaguchi et al. 2005). A systematic Chandra archival study (Stelzer et al. 2006b) rules out radiative winds as the origin for X-ray emission in HAeBes based on the observed high X-ray temperatures. Thus an X-ray generating mechanism from magnetic ﬂares similar to late type stars or emission from an unknown/unresolved companion is favored, although the role of accretion for the production of X-rays remains unclear (Hamaguchi et al. 2005; Stelzer et al. 2006b). 100

Using masses estimated from NIR color magnitude diagram (Figure 6), we detect around 50 stars in the intermediate mass range 2M⊙ . M . 8M⊙. The absorption- corrected X-ray luminosities of the detected sources in the 0.5-8.0 keV band are in the 1 range of log L 29.5 31.8 ergs s− . This is fully consistent with the level of X-ray t,c ∼ − emission detected in HAeBes from Hamaguchi et al. (2005) and Stelzer et al. (2006b), but also with the level expected from late-type companions. Optical spectral classification of these stars may further clarify the link between the X-ray emission properties and the spectral types (e.g., Li et al. 2002). Eight of these stars show significant temporal variability (P 0.005), which strongly supports emission from flaring, possibly from KS ≤ unresolved low mass companions. For example, source #804 shows a big flare during our long observation (Figure 4.4). The rising and decay time during the flare is rather symmetric. The count rate during the peak of the flare is 18 times higher than that of the quiescent level. As noted in 4.3.1, #44 is a previously identified Herbig Be star § in NGC 2244 (Li et al. 2002). We detected 40 net counts at its optical position in 94 ks. The fit to its X-ray spectrum indicates a low absorption and a hard kT = 2.8 keV plasma with a nonvariable lightcurve.

4.5.3 X-rays from Other Interesting Sources Herbig-Haro Jets and Knots– Two optical jet systems, namely Rosette HH 1 (R.A.=06h32m20.s76, Dec=04◦53′02.9′′) and HH 2 (R.A.=06h32m14.s14, Dec.=05◦02′17.95′′ [J2000]), have been discovered in the Rosette Nebula (Li 2003; Li & Rector 2004; Li 2005; Meaburn et al. 2005; Li et al. 2006). HH 1 consists of a collimated jet originating from −8 a faint optical star with a mass-loss rate M˙ 10 M⊙. We have detected X-ray emis- ∼ sion (#743) coincident with the location of the exciting source of HH 1, a weak-lined T Tauri star (F8Ve, Li et al. 2005). At the location of the base, knot, or terminal shock of the jet, the distribution of X-ray photons is consistent with background counts. No concentration of photons is coincident with the shock structure seen in optical near the end of the collimated jet. The X-ray spectrum of the star is soft with kT 0.8 keV ∼ and a negligible absorption column. It has been suggested that the combination of low extinction and high Lyman photon flux inside the Rosette Nebula makes the jet optically visible (Meaburn et al. 2005). A group of bright ionized knots in the Rosette Nebula were proposed to be colli- sionally ionized, either by bow shocks formed around globules by the strong winds from O stars which are then overrun by an expanding shell, or collimated flows of shocked gas driven by the wind (Meaburn & Walsh 1986). Chen et al. (2004) found an X-ray source in the nebulous region D with high speed knots and identified two stars as counterparts within ROSAT positional errors. We examined our Chandra image with much higher resolution at all knot locations. No X-ray emission was found to be associated with these high-speed knots, but we detected point sources coincident with optical stars embedded in the nebulous region. ACIS #678 (= Chen et al. source 30) matches the position of HD 259210, a likely foreground star with spectral type A1V. Binaries– The eclipsing binary V578 Mon used to determine the age and distance to NGC 2244 is detected in our X-ray image with 186 net counts (#476). It consists of two early B-type stars, one of the very few massive eclipsing systems known (Harries & 101

Hilditch 1998). Its orbital period is precisely determined in the optical at P = 2.40848 ± 0.00001 days (Hensberge et al. 2000). Using the eclipse ephemeris and the date at the beginning of our observation, we examined the X-ray light curve together with the optical light curve with orbital phases (Figure 4.20). No significant variability is suggested by a K-S test for the X-ray light curve, although dips might be seen in the X-ray light curve around phase φ = 0.85 and φ = 1.0, where the primary eclipse is expected. This could be one of the rare cases where X-ray eclipses can be used to constrain emitting geometry (e.g., Schmitt & Favata 1999). However this could simply be a statistical fluctuation given the limited number of counts. The spectral fit gives a plasma temperature of 2 kT 1.6 keV with low absorption, log N = 20.6 cm− . ∼ H Park & Sung (2002) noted a suspected PMS binary system ([PS02] 125 and [PS02] 126). An optical spectrum of the unresolved binary shows Hα in emission and LiI 6708A˚ in absorption (Chen et al. 2004), confirming its youth. We resolve the system as a close pair of ACIS sources (#243 & #242). Their spectral fits give a rather hard kT 2.9 keV for #243 and kT 1.5 keV for #242; they share the same log NH = 21.4 −∼2 ∼ cm . The light curve of #242 is constant while #243 shows possible but not significant variation (PKS = 0.03). Magnetic Star– Bagnulo et al. (2004) discovered an extraordinarily strong magnetic field in the very young cluster member NGC 2244-334 (=[OI81] 334; spectral type B3; R.A.=06h32m51.s79, Dec=+04◦47′16.1′′ [J2000]), ranking the second strongest lon- gitudinal field known among non-degenerate stars (after HD 215441 = Babcock’s star). We detected a 10 count X-ray source #899 (CXOU J063251.79+044715.9) at its position in the 20 ks observation (not covered by the deep observation). Elephant Trunks– Schneps et al. (1980) identified several spectacular elephant trunk globules in the northwest part of the Rosette Nebula. Only one of them, a small isolated globule denoted R1, is in our field of view. One X-ray source (#169), probably by chance superposition, is located 6 arcsec away from the bright rim of this dark globule. No other X-ray/IR source can be found inside the globule. In the southeast quadrant of the nebula towards the RMC, another molecular pillar is prominent. This region is also highlighted in the recent Spitzer survey (Balog et al. 2007), as the size of the pillar is comparable to the largest “pillar-of-creation” in M16. Chen et al. (2004) noted shocked gas near the pillar, perhaps due to strong winds from a star nearby that was matched to one of the ROSAT sources. In the vicinity of this pillar, a luminous X-ray source #919 (CXOU J063309.61+044624.3) is detected in our observation, but it is not located at the tip of the elephant trunk. Its X-ray spectrum −2 can be fit well with log NH = 21.2 cm and a hard kT = 3.3 keV plasma. With an 1 unusually high luminosity of log L 32.2 ergs s− is comparable to the earliest O stars. t,c ∼ The light curve is variable, as shown in Figure 4.4. The count rate doubles after the first 6 ks and remains in a high state for 13 ks. Its IR counterpart is also bright, with a ∼ K-band magnitude of 9.6 mag that is similar to the observed B0-B2 stars in the field. Its location in the color-magnitude diagram also suggests a spectral type of B0-B1. No K-excess is seen. As an early B-type star, its X-ray variability can be explained by an unresolved late type companion, although the lightcurve does not follow the typical fast- 2 rise and slow-decay phase. A similar transition between high and low states seen in θ 102

Ori A is investigated by Schulz et al. (2006) and interpreted as possibly the reconnection from magnetic interactions in a close binary system.

4.6 Soft Diﬀuse Emission

The potential effect of powerful winds of OB stars have long been recognized and discussed (Weaver et al. 1977). Predictions of measurable X-ray emission in wind-blown bubbles around single O stars were made, and observational effort followed to detect diffuse X-ray emission in HII regions unambiguously. TFM03 reported soft diffuse X-ray emission from 10 MK plasma in Rosette and in M17, and provided detailed review on ∼ diffuse X-ray characteristics of other Galactic star forming regions. In this section, we revisit the diffuse X-ray emission in Rosette HII region, given the long integration and therefore much better point source sensitivity ( 94 ks vs. 20 ks). Large scale soft ∼ ∼ X-ray emission can be seen in Figure 4.2c, which resembles the morphology as shown TFM03. Since the observed X-ray emission is rather soft (kT 0.06 0.8keV), we limit ∼ − our discussion to the unobscured populations only. To quantitatively measure the extended emission, we removed the 919 point sources using masks created by AE tool for optimal masking. The same diffuse emission analysis was done on the merged dataset, and the resulting X-ray luminosity Ls,c 32 1 ∼ L = 6 10 ergs s− . To determine whether this soft emission is truly diffuse in nature, t,c × we need to estimate the contribution of unresolved point sources (low mass PMS stars) and compare it to the observed diffuse emission. Among the expected 1000 unobscured stars, we detected 691 sources, among which 512 have derived luminosity from spectral fits. Around three hundred stars remain unresolved/undetected. For the 179 sources that have less photometric significance, we estimate their luminosity by extrapolating from 415 fainter sources that have spectral fits (source net counts 50 and Signif 2). 32 ≤ ≥ 33 Their integrated total luminosity contribution is L 1.3 10 . With L 1.7 10 t,c ∼ × t,c ∼ × contributed by those have luminosities from spectral fits, the estimated total emission 33 from the detected NGC 2244 unobscured population reaches Lt,c is 1.8 10 ergs −1 ∼ × s . The expected total emission from a 1.2 times unobscured ONC population Lt,c is 33 1 32 1 2.3 10 ergs s− (Feigelson et al. 2005, Table 1,). Therefore L 5 10 ergs s− ∼ × t,c ∼ × is expected from the unresolved low mass stars. On the other hand, if we assume the 32 missing 300 stars have mass M 0.5M⊙, the integrated luminosity is Lt,c 2 10 1 ∼ ≤ ∼ × ergs s− . This is consistent from the above estimated contribution from the subtraction of detected X-ray sources. Altogether from the crude estimates, we can estimate that 32 1 L 2 5 10 ergs s− is present in the NGC 2244 cluster from the unresolved point s,c ∼ − × sources. This can account for a significant fraction of the soft diffuse emission observed in the NGC 2244 field. The true diffuse emission produced by the fast stellar winds from 32 1 O stars in the field is on the order of L 10 ergs s− . s,c,diffuse ∼ 4.7 Summary

We present the high spatial resolution X-ray images of the NGC 2244 cluster in the Rosette Nebula obtained via deep Chandra observations. Our main ﬁndings follow: 103

29 1. We detect 919 X-ray sources with a limiting X-ray sensitivity of Lt,c 1 10 1 ∼ × ergs s− . Positional coincidence matching yields a total of 712 ONIR counterparts. We estimate 8% of extragalactic and galactic contamination. The rest of the X-ray sources without ONIR counterparts are likely new NGC 2244 members clustered around the massive stars or deeply embedded in the cloud. The X-ray detected population provides the first deep probe of the rich low mass population in this massive cluster. 2. The locations of most ACIS sources in the color-magnitude plot inidicate a large population of 2 Myr old PMS low mass stars (M . 2M⊙) subject to a visual extinction of 1 . AV . 2 at 1.4 kpc. We derive an overall K-excess disk frequency of 6% for stars with mass M & 0.5M⊙ using the X-ray selected sample, slightly lower than ∼ the 10% K-excess disk fraction using a FLAMINGOS selected sample. Both fractions are significantly lower than the 45% mid-IR disk fraction in a Spitzer sample that is more sensitive to disks. We emphasize that the combination of young stars identified in X-rays (mostly Class III stars) and those selected through IR colors will provide the best census to date for the young stellar population of this region. Three objects have Class I colors. 3. The derived XLF (Lh,c) for NGC 2244 is compared to the XLFs of the ONC, M17, Cep B, and NGC 6357, which indicates that the unobscured population in NGC 2244 is 1.2 times larger than that of the ONC, or 1000 stars detectable in COUP- ∼ sensitivity X-ray observation. Taking into account the obscured population, the total stellar population in NGC 2244 is around 2000. The XLF and KLF suggest a normal ∼ Salpeter IMF for NGC 2244; we do not confirm a top-heavy IMF reported from earlier optical studies. 4. We examine the spatial distribution of the X-ray identified NGC 2244 cluster members; the stellar surface density map suggests a spherical cluster with substructure. We confirm the existence of a subcluster around HD 46150 with 50 members in a ∼ 1 pc region, and a small subcluster consisting of a number of late type stars is found. The other O4 star HD 46223 has few companions. The radial density profile of NGC 2244 shows a larger relaxed structure around the central subcluster. Similar structure is seen in NGC 6357. No evidence for significant mass segregation is found in this cluster. Altogether we suggest that this 2 Myr cluster is not dynamically evolved and has a complex star formation history. Our results will strongly constrain models of the cluster formation process. 5. We detected all 9 OB stars with spectral types B0.5 or earlier, but only 6 out of 14 B stars with spectral types B1–B3 in our field of view. X-ray spectra for the massive stars in NGC 2244 all show soft emission, unlike other O star samples. We confirm the long-standing log(L /L ) 7 relation for the NGC 2244 O stars. Large deviation x bol ∼ − from this correlation was found for the B stars. 6. We report X-ray emission detected from a few interesting individual objects, including the ionizing source of an optical jet Rosette HH1, binary systems, a magnetic star, and a possible X-ray luminous uncataloged massive star. 7. We confirm the previously reported presence of soft diffuse X-ray emission. After considering the contribution from unresolved faint low mass stars based on the XLF analysis, we estimate the soft band luminosity of true diffuse X-ray emission generated 1 by fast O star winds is on the order of log L 32 ergs s− . s,c ∼ 104 a de econds. ints and (deg) J2000 δ 45 +045625.4 351.87 VF 0.40 +045934.0 286.00 VF 152.85240.84 +045542.0317.15 +044245.0417.34 +043442.0 335.86 +042745.9 335.90 335.76 335.85 F F F F J2000 α Observations nds; units of declination are degrees, arcminutes, and arcs Chandra ng steps are applied in the data reduction process. The aimpo before astrometric correction is applied. (UT) (s) Table 4.1. Log of Target Obs ID Start Time Exposure Time Aimpoint Roll Angle Mo The observing mode: F=Faint, VF=Very Faint. Rosette Field 2...... Rosette Field 3...... Rosette Field 4...... 1875 1876 2001 1877 Jan 05 2001 17:46 Jan 05 2001 23:28 Jan 06 05:10 19500 19410 19510 06 3 06 3 06 3 Rosette Field 1...... 1874 2001 Jan 05 11:53 19700 06 3 a Note. — Units of right ascension are hours, minutes, and seco RosetteNebula/NGC2244 3750 2004Jan0102:20 75000 063156. NGC2244SatelliteCluster 8454 2007Feb0902:25 20480 06305 Exposure times are theroll net angles usable are times obtained after from various the filteri satellite aspect solution 105 median Estimated E Column 16: ); b = possibly KS tion iteration after (ks) (keV) 83.9 1.0 (16) (17) 82.6 0.9 15.0 1.4 < P Column 5: below 90%) may indicate 05 . ··· ··· ··· 9 ; e = source on field edge; p . 0 < Estimated net counts extracted in the Characteristics Estimated net counts extracted in total ) is Anom Var EffExp Background-corrected median photon energy mkarf B P -5 .... a 88.5 0.9 -5 .... b 85.0 0.9 -5 .... -5 .... a 90.6 1.2 -5 .... a 85.7 0.9 -5 .... a 17.0 1.4 -5 g... -5 .... a 16.3 1.7 -5 g... -5 .... a 16.8 1.2 -5 .... a 81.8 1.0 < < < < < < < < < < < Column 10: Log probability that extracted counts (total band) are Columns 7,8: Column 17: . The first five sources and the interesting sources shown in s or for sources in chip gaps or on field edges. nts. Column 13: l band): a = no evidence for variability (0 Off-axis angle. . ector (FRACEXPO from Right ascension and declination for epoch J2000.0. 1.9 0.89 2.1 -4.4 .... b 17.0 1.1 e local background estimates can rise during the final extrac Astrophysical Journal Column 6: . tion region. Note that a reduced PSF fraction (significantly 0 1.4 0.90 4.5 6 3.5 0.90 2.5 7 5.8 0.91 3.7 4 0.8 0.89 2.7 1.5 2.2 0.48 4.1 net counts Columns 3,4: 8 33.3 32.1 0.87 22.8 .8 1.4 0.3 0.89 17.2 .2 1.1 27.3 0.89 12.6 .8 0.7 0.6 0.43 16.3 Background counts extracted (total band). 1.6 4.8 17.6 0.69 40.0 0.4 0.5 121.8 0.88 58.9 upper error on net counts Extracted Counts IAU designation. Net ∆Net Bkgd Net PSF Signif log Column 9: Main Catalog: Basic Source Properties ) Full Full Full Hard Frac ′ θ Column 2: # of counts extracted √ Chandra l. (2007b) in the electronic edition of the 005). No value is reported for sources with fewer than 4 count . ) ( 0 Variability characterization based on K-S statistic (tota ′′ Err < ave to be observed on axis to obtain the reported number of cou errors on column 7. d content and for convenience of the reader. σ KS P standard deviation of PSF inside extraction region Source anomalies: g = fractional time that source was on a det Photometric significance computed as J2000 δ Position Column 15: Table 4.2. values above the 1% threshold that defines the catalog becaus B P Fraction of the PSF (at 1.497 keV) enclosed within the extrac Column 12: Column 14: , computed as J2000 (deg) (deg) ( σ α Column 11: 05); c = definitely variable ( . 0 X-ray catalog sequence number, sorted by RA. < KS < P Source Column 1: 005 . 2 063117.04+045228.3 97.821033 4.874535 0.7 9.5 32.0 6.5 4. 3 063118.11+045511.8 97.825464 4.919953 1.0 8.7 12.4 4.4 2. 1 063114.36+045303.0 97.809835 4.884189 0.9 9.9 23.3 5.7 3. 4 063118.29+045223.1 97.826245 4.873106 1.0 9.2 14.6 4.8 3. 58 063118.76+045207.6 063120.89+045003.8 97.828205 97.837054 4.868779 4.834396 1.1 0.2 9.2 10.7 599.7 9.7 25. 4.1 3.3 # Note. — Table 4.2 will be published in its entirety in Wang et a Note. — (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 743 063220.77+045303.5 98.086542 4.884328 0.3 7.0 25.5 5.7 630 063210.47+045759.6 98.043639 4.966572 0.1 4.0 332.6 18 615 063209.32+044924.4 98.038846 4.823469 0.1 7.7 1685.2 4 476 063200.65+045241.2 98.002718 4.878124 0.1 3.8 185.9 14 373 063155.51+045634.2 97.981330 4.942835 0.0 0.4 3588.5 6 311 063152.54+050159.1 97.968929 5.033106 0.1 5.8 299.3 17 Seq CXOU J that the source is in a crowded region. (total band). variable (0 Effective exposure time: approximate time the source would h hard energy band (2–8 keV). sources are removed from the catalog. = source piled up; s = source on readout streak. energy band (0.5–8 keV); average of the upper and lower 1 solely from background. Some sources have Figure 4.1 are listed here for guidance regarding its form an random component of position error, 1 106 ) is mkarf median E Column 5: Column 17: Estimated net nts. s or for sources in (ks) (keV) (16) (17) Log probability that Estimated net counts Column 10: l band): a = no evidence for ector (FRACEXPO from Column 13: Columns 7,8: . Characteristics Anom Var EffExp e local background estimates can rise during B P Off-axis angle. net counts . A portion is shown here for guidance regarding its form tion region. Note that a reduced PSF fraction (significantly upper error on net counts Right ascension and declination for epoch J2000.0. Column 6: . Background counts extracted (total band). Astrophysical Journal 005). No value is reported for sources with fewer than 4 count Columns 3,4: . ave to be observed on axis to obtain the reported number of cou 0 0 3.0 0.9082 2.9 1.6 0.89 4.5 0.90 -2.2 1.7 .... 1.5 -2.9 a -2.7 .... 84.7 .... a 1.1 a 86.1 88.9 1.8 4.9 < .0 0.1 0.90 1.6 -2.1 .... a 82.8 1.5 Variability characterization based on K-S statistic (tota Column 9: 3.75.33.6 2.8 0.0 0.90 0.11.4 0.90 0.891.3 0.7 1.7 0.90 2.5 1.8 1.8 0.90 -2.1 -2.4 -2.2 1.8 ...... 1.8 -2.5 .... a -2.5 a .... a 83.0 .... 83.7 a 83.3 1.5 b 1.3 84.7 1.4 79.9 1.1 1.6 22.7 6.9 0.90 1.7 -2.0 .... a 63.1 2.7 KS P Extracted Counts Source anomalies: g = fractional time that source was on a det IAU designation. Column 15: Photometric significance computed as # of counts extracted Net ∆Net Bkgd Net PSF Signif log values above the 1% threshold that defines the catalog becaus √ errors on column 7. B σ P out streak. Column 14: Column 2: Secondary Catalog: Tentative Source Properties ) Full Full Full Hard Frac ′ θ Column 12: 05); c = definitely variable ( standard deviation of PSF inside extraction region . l. (2007b) in the electronic edition of the Fraction of the PSF (at 1.497 keV) enclosed within the extrac . 0 ) ( ′′ < Err per and lower 1 ome sources have KS Chandra m the catalog. < P Column 11: J2000 , computed as δ 005 Position . σ Effective exposure time: approximate time the source would h Table 4.3. J2000 (deg) (deg) ( α Column 16: X-ray catalog sequence number, sorted by RA. ); b = possibly variable (0 KS < P 05 . Source Column 1: 9 ; e = source on field edge; p = source piled up; s = source on read . 19 063124.47+045306.2 97.851975 4.885077 0.8 8.4 9.0 4.9 10 2832 063126.33+045728.836 063127.43+045036.452 063128.13+045008.061 063130.85+044847.368 97.859728 063132.25+050439.071 97.864312 063133.57+045233.3 4.95802383 97.867223 063134.14+044958.5 4.84345785 97.878575 063135.41+045813.0 0.8 4.835559 97.884410 063135.75+050259.6 0.8 4.813144 97.889886 0.8 5.077512 7.4 97.892258 0.9 4.875920 9.0 97.897576 0.9 4.832939 9.2 97.898985 0.6 4.970293 9.7 10.2 0.7 5.049898 8.0 10.3 0.6 6.7 11.7 0.7 8.3 13.4 4.4 12.3 5.4 5.5 8.4 5.8 8.2 6.8 10.6 1 7. 6.8 1 5.8 10.7 2 4.1 5.3 3.4 5.3 1 4. 1 2. # 0 Note. — Table 4.3 will be published in its entirety in Wang et a Note. — (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq CXOU J Background-corrected median photon energy (total band). chip gaps or on field edges. variability (0 < the final extraction iteration after sources are removed fro extracted counts (total band) are solely from background. S and content. extracted in total energy band (0.5–8 keV); average of the up below 90%) may indicate that the source is in a crowded region counts extracted in the hard energy band (2–8 keV). Estimated random component of position error, 1 d · · · · · · · · · · · · · · · · · · · · · · · · · · · upper Notes

107 t,c L log Columns 5 and 6 t c L . A portion is shown here log ) 1 hat the parameter is effectively − 1 30.06 30.11 2 30.36 30.37 h,c .89 31.13 31.51 H L es of luminosities; actual parameter d (0.5–8 keV). Absorption-corrected ated fit yielded non-physical results. log e intervals. More significant digits are h X-ray Luminosities L log Astrophysical Journal s L 29.73 28.33 28.34 29.75 29.85 30.25 30.18 30.20 30.52 30.73 30.31 30.19 30.20 30.56 30.64 presents the emission measure for the model spectrum, log etric significance data from Table 4.2. 07 2 8 . . . +0 +0 +0 ) (ergss 3 Column 7 − EM d Sources: Thermal Plasma 53.8 53.7 he same number of significant digits is used for both lower and 53.2 29.71 29.70 29.72 30.01 30.16 53.3 29.96 29.85 29.86 30.21 30.30 53.4 30.20 29.96 29.96 30.40 30.40 2 log . 17 0 . 0 − − b 8 . abundances (Imanishi et al. 2001; Feigelson et al. 2002a). +0 5 since the soft band emission is essentially unmeasurable. . ⊙ Z 22 kT 0.6 52.9 1.7 2.8 53.4 29.78 30.03 30.06 30.23 30.39 2.0 2.0 2.0 2.0 4 6 > . . 0 2 Fits Spectral Fit l. (2007b) in the electronic edition of the − − H N 1 1 3 5 temperature parameters. . . . . +0 +1 +1 +0 ) (keV) (cm he convenience of the reader. H 2 − N dered to merely be a spline to the data to obtain rough estimat was unable to compute them or when their values were so large t were frozen in the fit. Uncertainties represent 90% confidenc and assumed 0.3 means the fit was performed by hand, usually because the autom : s = soft band (0.5–2 keV); h = hard band (2–8 keV); t = total ban H XSPEC XSPEC columns 8–12 ; they are omitted when log c reproduce the source identification, net counts, and photom Quantities in italics a 1 in order to avoid large rounding errors; for consistency, t . 0 Table 4.4. X-ray Spectroscopy for Photometrically Selecte < columns 1–4 Source means a two-temperature model was used. All fits used the “wabs(apec)” model in 2T For convenience X-ray luminosities are presented in 89 063120.89+045003.8 063120.95+045322.7 599.7 14.5 22.8 2.6 21.4 21.0 0.6 3.4 54.4 53.0 31.11 29.86 29.70 29 29.81 29.8 7 063120.48+045023.4 98.9 8.5 21.5 6 063119.27+045110.7 22.6 3.6 20.1 1.9 53.4 30.17 29.91 29.9 5 063118.76+045207.6 9.7 2.1 20.8 4 063118.29+045223.1 14.6 2.7 21.1 3 063118.11+045511.8 12.4 2.5 21.6 2 063117.04+045228.3 32.0 4.5 21.1 1 063114.36+045303.0 23.3 3.7 20.0 11 063121.46+045405.3 15.6 2.7 21.4 Note. — Table 4.4 will be published in its entirety in Wang et a c b a d # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) Seq CXOU J Net Signif log unconstrained. Fits lacking uncertainties should be consi luminosities are subscripted with a present the best-fit values for the column density and plasma uncertainties. Uncertainties are missing when used for uncertainties assuming a distance of 1.4 kpc. values are unreliable. for guidance regarding its form and content. Well-known counterparts from Table 4.6 are listed here for t d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes 108 t,c L he same number aw photon index log t L ) . A portion is shown here 1 d (0.5–8 keV). Absorption- − c log s 2 − h,c 91 29.87 30.10 73 30.10 30.11 99 29.95 30.17 L o merely be a spline to the data to log X-ray Fluxes h L (photons cm was unable to compute them or when their were frozen in the fit. Uncertainties represent log Astrophysical Journal s L XSPEC 29.63 30.41 30.42 30.47 30.53 log etric significance data from Table 4.2. 5 . +0 Γ Quantities in italics d Sources: Power Law Fits N . -5.9 -5.8 29.54 30.31 30.33 30.38 30.47 -5.6 29.80 30.45 30.46 30.54 30.62 5 since the soft band emission is essentially unmeasurable. 4 5 4 . . . . 0 0 0 22 − − − 1 in order to avoid large rounding errors; for consistency, t > . b 0 9 present the best-fit values for the column density and power l . H < +0 N Γ log 1.1 1.3 -6.0 29.02 30.08 30.13 30.12 30.27 0.8 -6.8 28.90 29.72 29.72 29.78 29.80 1.3 1.4 1.8 -5.6 29.46 30.15 30.19 30.23 30.43 2.0 -5.7 29.21 29.98 30.03 30.04 30.33 6 8 9 9 2 . . . . . 0 0 0 0 1 − − − − − Spectral Fit l. (2007b) in the electronic edition of the 3 4 5 3 0 5 ...... inties. Uncertainties are missing when trained. Fits lacking uncertainties should be considered t +0 +0 +0 +0 +1 +0 ) r uncertainties H 2 alues are unreliable. − N Columns 5 and 6 . : s = soft band (0.5–2 keV); h = hard band (2–8 keV); t = total ban 21.9 22.1 6 7 . . 0 0 − − XSPEC ; they are omitted when log c columns 8–12 reproduce the source identification, net counts, and photom a presents the power law normalization for the model spectrum Table 4.5. X-ray Spectroscopy for Photometrically Selecte columns 1–4 Source Column 7 X-ray luminosities are presented in All fits used the “wabs(powerlaw)” model in For convenience 91 063136.42+045602.8 12.2 2.5 22.3 1.5 -6.1 28.70 29.84 29. 94 063136.49+045959.0 10.6 2.2 21.1 89 063136.33+045251.1 46.2 5.7 26 063125.51+045252.2 45.5 5.4 20.0 2.0 -5.8 29.86 29.73 29. 3184 063127.01+050205.0 063135.68+045322.2 14.3 26.9 2.3 4.1 22.3 1.5 -6.0 28.79 29.92 29. 15 063123.59+045141.7 60.6 6.3 21.5 10 063121.28+045023.8 50.6 5.7 21.5 c b Note. — Table 4.5 will be published in its entirety in Wang et a a # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 126 063139.92+050059.4 47.1 5.8 21.7 118 063139.24+050244.7 19.5 3.1 22.1 Seq CXOU J Net Signif log obtain rough estimates of luminosities; actual parameter v parameters. 90% confidence intervals. More significant digits are used fo corrected luminosities are subscripted with a of significant digits is used for both lower and upper uncerta values were so large that the parameter is effectively uncons for guidance regarding its form and content. photometry from JHK 109 ontent. (16) (17) (18) ience, [MJD95]=Massey, Johnson, & 11.98 11.77 11.71 AAA000 14.35 13.58 13.29 AAA000 14.53 13.53 13.22 AAA000 14.94 14.17 13.85 AAA000 14.49 13.64 13.45 AAA000 13.42 12.70 12.46 AAAc00 0522 7.84 7.84 7.82 AAA000 ··· ··· ··· ··· ··· ··· provide NIR identifications and 981), [LR04]=Li & Rector (2004), ProbMem=Probability of Columns 13–17 2MASS FLAMINGOS J H K PhCcFlg 06312100+0453238 063120+045323 14.93 14.14 13.97 AAA000 06311925+0451121 06311825+0452234 06311696+0452281 06311429+0453032 06312048+0450239 α . A portion is shown here for guidance regarding its form and c ··· ··· ··· ··· ··· are the catalogs used for counterparts matching. For conven Optical/Infrared Photometry give available optical photometry. 17.52 16.43 16.84 15.52 16.79 15.67 18.32 16.59 17.15 17.37 Astrophysical Journal Columns 3–6 ··· ··· ··· ··· ··· . In the note on individual sources, [OI81]=Ogura & Ishida (1 Columns 7–12 19.50 19.39 19.47 20.75 19.33 m Table 6.1 and Table 6.2. (mag) (mag) (mag) (mag) (mag) (mag) ID ID (mag) (mag) (mag) m available 15 7.64 8.37 8.22 8.16 8.09 8.13 06312087+0450038 063117+05 14 13.36 13.36 12.90 12.38 12.91 12.57 06311881+0452089 Table 4.6. Stellar Counterparts l. (2007b) in the electronic edition of the hall et al. (1982), [D06]=Dias et al. (2006). ··· ··· 2002), [PS02]=Park & Sung (2002). ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··················································· ······························ ··················································· lists the 2MASS photometric quality flags (Cutri et al. 2003) Column 18 a reproduce the sequence number and source identification fro 3.1). § X-ray Source from [OI81] 76=[BC02] 14 ′′ 2 =HD 46056=BD+04 1291=MWC 808=[OI81] 84 (O8V); FUSE spectru 9 063120.95+045322.7 0948-0096264 8 063120.89+045003.8 0948-0096261 454 7 063120.48+045023.4 6 063119.27+045110.7 0948-0096246 5 063118.76+045207.6 0948-0096241 503 34 063118.11+045511.8 063118.29+045223.1 0948-0096235 2 063117.04+045228.3 0948-0096228 1 063114.36+045303.0 0948-0096209 Columns 1–2 10 063121.28+045023.8 # Note. — Table 4.6 will be published in its entirety in Wang et a 5 8 a (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq CXOUJ USNOB1.0 MJD PS BC U B V R I H being a cluster member from proper motion data: [M82]=Marsc FLAMINGOS ( Degioia-Eastwood (1995), & [BC02]=Berghöfer Christian ( 110 ) ing 1 t,c − .98 L 28.7 28.7 28.7 28.7 28.7 28.7 28.7 28.8 < < < < < < < < log eld 4 (RMC ) (erg s 1 h − 5857 30.01 30.03 16 30.45 The stars are listed L .75 31.12 28.3 28.3 28.3 28.3 28.3 28.3 28.3 28.4 8.74 30.65 9.88 30.54 < < < < < < < < fits to the ACIS spectra (: o X-ray sources. Both are 8 keV); inferred total band − nd de Jager & Nieuwenhuijzen 6 30.04 30.57 6 29.86 31.51 +0.9 30.46+1.4 32.05 29.81 30.94 kT log 0.3 28.75 31.11 ) (keV) (erg s 64 0.6+0.2 0.3 30.48 29.71 32.34 32.38 .0 0.7 29.62 31.04 2 H − N X-ray Properties NetCts log Spectral types are from Ogura & Ishida (1981) and Massey og. faintest sources in Table 4.4 and scaled with the correspond ) (cm φ ′′ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· a Column 2: ical cross-identifications, positions and spectral types. ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Seq ∆ ··· the FOV. ...=non detection. HD 46485 is observed in Rosette Fi are assumed); observed hard band luminosity (2 kT 1299s as a detection here. ◦ kground subtraction; column density and plasma energy from ). However the deep observation resolved this source into tw ) # ( with spectral type, Martins et al. (2005) for O3–O9.5 stars a ˜ bol ⊙ L bol . 2005); : in L f . OI= Ogura & Ishida (1981). (mag) (L and 2MASS sources. Chandra 8 keV). Upper limits to luminosities are estimated using the − -band magnitudes are from 2MASS All-sky Point Sources Catal 5 . K Table 4.7. X-ray properties of cataloged OB stars in NGC 2244 Optical/IR Properties This list is obtained from Appendix A of TFM03 which gives opt source number, from Table 4.2. NFOV=object is not covered in from the optical position. Therefore we do not report BD+04 X-ray properties from Table 4.4: extracted counts after bac ′′ Source numbers and 2 ∼ Chandra Offset (in arcseconds) between the Column 1: Bolometric luminosities are estimated from calibrations o 1299s was reported as a detection in TFM03 (20ks observation ◦ values are approximated from median energy (Feigelson et al 1281B B1.5V 06315893+0455398 9.74 3.7 448 0.1 16 21.1 2.2 29. 1295p B2.5V 06313146+0450596 10.24 3.4 53 0.5 28 21.5 2.0 29. 1299s B1III 06320613+0452153 9.38 4.0 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) ◦ ◦ ◦ Name SpTy 2MASS K log H N BD+04 a Note. — HD 46223 O4V((f)) 06320931+0449246 6.68 5.7 615 0.2 1685 21. HD 46150HD 46485 O5V((f)) O7V 06315551+0456343 6.44 06335094+0431316HD 259105 5.5 7.45 5.2 B1V 373 RMC 164 06315200+0455573 0.1 0.3 8.95 3589 310 4.0 21. 21.6 0.3 HD 46056HD 258691HD 46202MJD O8V((f))HD 259135 O9V((f))IRAS 06309+0450 06312087+0450038HD O9V((f)) 259012 06303331+0441276 7.82 B0.5V 06321047+0457597 7.93 B0.5V 06333749+0448470 5.0 7.72 B0.5V 4.8 B1V 06320061+0452410 8.64 4.8 06313708+0445537 8.12 06313346+0450396 4.3 12.20 8 NFOV 8.79 4.3 4.3 630 4.0 0.3 NFOV 0.1 600 476 NFOV 333 66 0.7 21.4 21.2 186 0. 0.6 275 20.6 1. 21.1 2.5 30 HD 46149HD 259238HD 46106 O8.5V((f)) 06315253+0501591 B0V 7.25 B0.2V 06321821+0503216 4.9 06313839+0501363 10.28 7.62 4.5 311 4.4 0.1 727 107 299 1.0 0.2 20 36 186 21.8 21.4 0.4 0.5 2 HD 46484BD+05 B1V 06335441+0439446 6.86 4.0 NFOV HD 259172OI81 345BD+04 OI81 130 B2V B2 06320259+0505086 B2.5V 10.08 06330656+0506034 06314789+0454181 11.20 3.5 10.87 3.5 3.4 NFOV NFOV OI81 172 B2.5V 06320984+0502134 10.43 3.4 BD+04 OI81 190 B2.5Vn 06315891+0456162 10.54 3.4 OI81 274 B2.5V 06322424+0447037 10.58 3.4 HD 259268 B3 06322304+0502457 10.33 3.2 OI81 392OI81 194 B2.5V B3 06335056+0501376 9.99 06321548+0455203 10.96 3.4 3.2 NFOV 697 0.1 62 21.2 3.6 30. MJD95 B3V 06322249+0455342 13.48 3.2 HD 259300 B3Vp 06322939+0456560 9.34 3.2 815 0.3 60 21.5 1.4 2 334 B3 06325179+0447161 11.53 3.2 899 0.2 10 20.9 1.4 29.33 29 MJD95 B3V 06331016+0459499 12.39 3.2 NFOV separated by exposure time. luminosity corrected for absorption (0 source #164, ChapterColumn 5). 7: (1987) for BColumn stars. 6: No use is made of available photometry. in log Columns 8–12: et al.(1995). Columns 3–4: Column 5: first in order of decreasing mass, and then by right ascension 111

Fig. 4.1 (a): A large scale ( 2◦ 1.5◦) view of the Rosette star forming complex in 12 ∼ × COJ = 1 0 emission from Heyer et al. (2006). The multiple ACIS FOVs (polygons) − and the extent of NGC 2244 cluster (circle) are shown. Squares mark the embedded clusters in the RMC with Phelps & Lada (1997) sequence numbers. (b): A 45′ 45′ × DSS2 R-band image of the Rosette Nebula. All known O stars in the FOV that belong to the NGC 2244 cluster (TFM03 Table 6) are labeled by circles. Four stars are marked as squares: V578 Mon is an eclipsing binary; HH 1 is a stellar microjet; ACIS #919 is a candidate massive star; the visually brightest star HD 46241 (K0V) is foreground. These objects are described further in the text. 112

Fig. 4.2 (a): A merged 94 ks ACIS-I image of the NGC 2244 cluster from ObsIDs 1874 and 3750 (outlined by two 17′ 17′ boxes) with reduced resolution (binned by 2 pixels). × The two ObsIDs have slightly different roll angles. (b): X-ray composite image created from csmooth for the merged fields. Blue intensity is scaled to the soft (0.5–2 keV) X-ray emission, green intensity is scaled to the hard (2–7 keV) X-ray emission. (c): Same as (b) but the scaling emphasizes soft diffuse emission and red intensity is scaled to the DSS R-band optical emission. 113

Fig. 2. — Continued. 114

Fig. 4.3 30′′ 30′′ ACIS unbinned image (top left), reconstructed image (top right), × the Hα image (bottom left), and the 2MASS-Ks image (bottom right) of the central region around the O5V star HD 46150. Several new sources are resolved within 5′′ of the dominant star by the Chandra observation. 115

Fig. 4.4 Lightcurves of sources with more than 500 counts that are signiﬁcantly variable (P 0.005). The ACIS sequence numbers and binsize are marked. KS ≤ 116

Fig. 4.5 NIR J H vs. H K color-color diagram for 617 Chandra stars with high-quality − − photometry from combined FLAMINGOS and 2MASS data (error in both J H and − H K colors < 0.1 mag). The (light and dark) green circles and red triangles represent − sources with and without significant K-band excess (E(H K) > 2σ(H K)). The − − dark green circles represent three Class I objects and are labeled with their sequence numbers from Table 2. The five blue triangles are foreground stars. Stars using 2MASS photometry are indicated with black circles. The black solid and long-dash lines denote the loci of MS stars and giants, respectively, from Bessell & Brett (1988). The purple dash dotted line is the locus for classical T Tauri stars from Meyer et al. (1997), and the cyan solid line is the locus for HAeBe stars from Lada & Adams (1992). The blue dashed lines represent the standard reddening vector with crosses marking every AV = 5 mag. Most Chandra sources are located in the reddening band defined by the left two dashed lines associated with Class III objects (triangles). To the right of this reddened band are 38 IR-excess sources. 117

Fig. 4.6 NIR J vs. J H color-magnitude diagram using the same sample and symbols − as Figure 4.5, except that known O stars are denoted as the red ﬁlled triangles. The purple solid line is the 2 Myr isochrone for PMS stars from Siess et al. (2000). The dash dotted line marks the location of Zero Age Main Sequence (ZAMS) stars. The blue dashed lines represent the standard reddening vector with asterisks marking every AV = 5 mag and the corresponding stellar masses are marked. 118

Fig. 4.7 The cumulative distribution of the source hardness indicator, median photon energy, for Chandra sources with identiﬁed ONIR counterparts (solid line) and those without counterparts (dashed line). The sources with identiﬁed ONIR counterparts are considerably softer than the latter group. 119

Fig. 4.8 The spatial distribution of Chandra sources without identiﬁed ONIR counterparts. The background is the DSS R-band image. ACIS-I FOVs are shown. 120

Fig. 4.9 (a): X-ray luminosity function (XLF) constructed from the absorption corrected hard band (2.0–8.0 keV) X-ray luminosity Lh,c for the unobscured NGC 2244 population (solid line), and the COUP ONC unobscured cool stars population (dashed line, Feigelson et al. 2005). The vertical line denotes the estimated completeness limit for NGC 2244 population. (b): the cumulative distribution of X-ray luminosities for the unobscured NGC 2244 X-ray sources as well as the unobscured populations in COUP, Cep B/OB3b (age 1-3 Myr, Getman et al. 2006), NGC 6357 (age 1 Myr, Wang et al. 2007a), and ∼ ∼ M17 (age 1 Myr, Broos et al. 2007). The distributions for NGC 2244, NGC 6357, Cep ∼ B, and M17 are truncated at their completeness limits, log Lh,c 29.2, 30.4, 29.3, and 1 ∼ 30.4 ergs s− , respectively. 121

Fig. 4.10 Comparison between IMFs of the NGC 2244 X-ray stars (solid line) and COUP ONC stars (dashed line) using NIR photometry derived stellar masses. The dotted lines show the ONC IMF (Muench et al. 2002) and its scaled version to match the NGC 2244 IMF. The dot-dashed line is the NGC 2244 IMF estimated using the KLF derived from 2MASS data. The arrows indicate the approximate mass completeness limits for the NGC 2244 and ONC X-ray stars, and the mass completeness limit for the KLF derived from 2MASS data. Note that the 2MASS completeness limit is the same as our Chandra completeness limit in stellar mass. The incomplete bins in the NGC 2244 KLF/IMF are omitted. 122

Fig. 4.11 The stellar surface density map for the unobscured stellar population in NGC 2244. The cluster shows a spherical structure that extends 8 arcmin in diameter. The highest concentration of stars is around R.A.=06h31m55s, Dec.=+04◦56′34′′, the location of HD 46150. A secondary density enhancement is seen centered at (α, δ) = (06h31m56s, 04◦54′10′′). The O4 star HD 46223 is mostly isolated. 123

Fig. 4.12 The KPNO Hα image (Li & Rector 2004) and 2MASS Ks image of the secondary overdensity of stars seen in Figure 4.11. The overlaid circles mark the ACIS sources. The 3.5′ 3.5′ images are both centered at RA=06h31m04s,Dec.=+04◦54′26′′. × 124

Fig. 4.13 The observed radial density proﬁles of the NGC 2244 cluster, the ONC from COUP studies (Feigelson et al. 2005), and the NGC 6357 region from our Chandra/ACIS observation (Wang et al. 2007a). The histogram shows the radial density proﬁle of the ONC from ONIR studies (Hillenbrand & Hartmann 1998). 1σ Poisson error is shown for the COUP ONC. 125

Fig. 4.14 The radial density proﬁles (thick lines) in physical scales for the same three clusters. where the stellar density of NGC 2244 and NGC 6357 has been scaled to 1.2 times and 5 times the ONC population, respectively, based on the XLF analysis. The thin line represent a King model proﬁle for the outer portion of NGC 2244. 126

Fig. 4.15 (a): Spatial distribution of the massive stars (NIR estimated mass M & 8M⊙, ﬁlled stars) and the low mass stars (M . 2M⊙, open diamonds)in NGC 2244, using our X-ray-selected sample. (b): The cumulative radial distributions for the massive stars (NIR estimated mass M & 8M⊙, dashed line) and the low mass stars (M . 2M⊙, solid line). 127

Fig. 4.16 The spatial distribution of 38 sources with signiﬁcant NIR color-excess (circles). The northern part of the nebula seems deﬁcient in NIR excess sources. The boxes outline the FOVs of the multiple ObsIDs. 128

Fig. 4.17 Spectral fits to X-ray spectra of six O and early B stars. Source name, source counts, and fit parameters are marked in each panel. The two model components are shown as the dotted lines for spectra that are best fit with two temperature thermal plasma models. 129

) 31 -1 (erg s tc 30 log L

2 3 4 5 6

log (Lbol/LO •)

Fig. 4.18 The Lx vs. Lbol relation for X-ray detected O and early B stars. The samples are from the NGC 2244 cluster (ﬁlled triangles; this work, Table 4.7), the ONC (crosses, Stelzer et al. 2005), and the massive star forming regions M17 (diamonds, Broos et al. 2007) and NGC 6357 (pluses, Wang et al. 2007a). Upper limits are marked as arrows. The bolometric luminosities are adopted from Broos et al. (2007) to be consistent. 130

Fig. 4.19 (a): Hα image of # 919 neighborhood near a molecular pillar. (b): The X-ray image of the same region. The 2.5′ 2.5′ images are both centered at the X-ray bright × star # 919. (c): The spectral ﬁt to the X-ray spectrum of #919 with log NH = 21.2 and a hard kT = 3.3 keV plasma. 131

Fig. 4.20 X-ray light curve of the eclipsing binary V578 Mon (lower panel), together with the optical light curve (b-band photometry) in orbital phases (upper panel). The orbital phase is calculated using eclipse ephemeris reported in Hensberge et al. (2000). 132

Chapter 5

A Chandra Study of the Stellar Populations in the Rosette Molecular Cloud

5.1 Introduction

In the last two decades, developments in infrared (IR) technologies including JHK arrays and the latest mid-IR Spitzer Space Telescope have enabled large scale studies of the early evolutionary stages of young stars, even when they are still embedded in molecular clouds (e.g., Lada et al. 1991; Lada & Lada 1995; Carpenter 2000; Churchwell et al. 2004; Young et al. 2005). One major discovery from such surveys of nearby clouds is that a large fraction ( 70%-90%) of stars form in embedded clusters (Lada & Lada 2003). ∼ It follows naturally that clustered star formation in the molecular clouds may account for much of the star formation activity in the Milky Way. The spatial distribution and hierarchical structure of embedded clusters in the giant molecular clouds (GMCs) are imprinted with the physical processes in cluster formation (e.g., Elmegreen 2000; Bonnell et al. 2001; Lada & Lada 2003; Tan et al. 2006; Ballesteros-Paredes et al. 2007), hence of particular interest to understand their birth and evolution. As one of the most massive Galactic GMCs and prominent star formation complex, the Rosette Molecular Cloud (RMC) is ideal for studying formation of embedded star clusters. Townsley et al. (2003) (hereafter TFM03) and Román-Zúñiga et al. (2007b) (hereafter RL07) provide reviews of past and present research on this popular and important star formation region. Its orientation is perpendicular to the line-of-sight with a large HII region at its tip known as the Rosette Nebula. This HII region is powered 5 by the OB association NGC 2244. The RMC contains 10 M⊙ of gas and dust (Blitz ∼ & Thaddeus 1980), showing highly clumpy structure (Williams & Blitz 1998). Because of the high visual extinction, optical observations are incapable of probing the obscured stellar populations. The embedded star clusters have been revealed in the dense molecular cloud cores through near-IR (NIR) imaging surveys (Phelps & Lada 1997, hereafter PL97; Román-Zúñiga et al. 2007a), which also clearly demonstrate the association between IR clusters and the IRAS emission peaks. Figure 5.1 outlines the ISM content 12 of the RMC and the locations of PL97 clusters, showing the CO continuum emission contours (Heyer et al. 2006) overlaid on an MSX 8.3 µm image (see also Figure 1a in Wang et al. 2007b). Given that it is distant from the Sun (d 1.4 kpc) and close to the Galactic plane ◦ ∼ (b = 2 ), IR imaging surveys in the Rosette region are unavoidably contaminated by the − overwhelming field populations. Therefore identification of young embedded populations relies on selecting stars with IR color excess, which suggests the existence of circumstellar disks. Because many weak-lined T Tauri stars (WTTS) that are cluster members cannot 133 be identified this way, a background subtraction is needed to statistically derive the properties of the embedded population. Besides IR emission, pre-main-sequence (PMS) stars (WTTS included) are well- known to emit X-rays that can penetrate heavy extinction and reveal themselves inside molecular clouds (Feigelson et al. 2007). Most importantly, the level of X-ray emission 4 from the young stellar populations is elevated 10–10 times above the levels of X-ray emission from the old Galactic disk population. As a consequence, X-ray observations are highly efficient in selecting PMS members in stellar clusters (e.g., Townsley et al. 2006a; Getman et al. 2007; Wang et al. 2007a; Broos et al. 2007). In the X-ray band, a previous low spatial resolution ROSAT imaging study of the Rosette Complex (Gregorio- Hetem et al. 1998) reported embedded faint X-ray sources associated with embedded low mass star clusters. However the poor resolution and soft energy pass band of ROSAT limited more conclusive study. Thanks to the advent of the Chandra X-ray Observatory, TFM03 presented the first high resolution X-ray image mosaic of this high-mass star forming region, and studied the soft diffuse emission in detail. Although the embedded clusters in the RMC were extensively studied in the IR band at large scale, the individual cluster members and thus the young stellar populations are largely unknown. In this chapter we present an X-ray point source study of the RMC based on the TFM03 data. A parallel study of the massive OB cluster in the Rosette Nebula was presented in Chapter 4 (Wang et al. 2007b). We describe here Chandra/ACIS observations of the embedded clusters in the RMC, aiming to investigate cluster formation in a sequential manner and to test whether molecular clumps preferentially forming embedded clusters of low-mass stars make up the fundamental building blocks of star formation in molecular clouds.

5.2 Chandra Observations and Data Reduction

The RMC was observed with the Imaging Array of Chandra’s Advanced CCD Imaging Spectrometer (ACIS-I). The observatory and instrument are described by Weis- skopf et al. (2002). Our customized point source detection, extraction, variability analysis, and spectral fitting procedures with ACIS Extract (Broos et al. 2002) for the RMC data are presented as part of the Rosette star-forming complex observations in Wang et al. (2007b). The RMC source list contains 395 X-ray detections, which are divided into a primary list of 347 highly reliable sources (Table 5.1) and a secondary list of 48 tentative sources (Table 5.2) with 0.001 likelihood of being spurious background ≥ fluctuations. This approach has been adopted in our series of papers on Chandra observations of star forming regions; it allows the reader evaluates the validity of faint source individually. Table 5.1 and Table 5.2 have a format that are identical to Tables 1 and 2 in Wang et al. (2007a). A Chandra mosaic of the RMC region (0.5-8 keV) overlaid with source extraction regions (calculated by AE) is shown in Figure 5.2. During the 20 ks observations, eleven sources display significant variability ∼ (PKS < 0.005 in column 15 of Tables 5.1 and 5.2). Only one of them, source #119 have more than 100 net counts due to the short exposure (ObsID 1876). This source is also the brightest source in the field (see 5.4). Its light curve is presented in Figure 5.3, § 134 which does not resemble the characteristic fast-rise and slow-decay X-ray flares seen in PMS stars. Spectral analysis results for the 182 sources with Signif 2.0 are presented in ≥ Tables 5.3 (164 sources; thermal plasma fit) and 5.4 (18 sources; power law fit). Detailed description of the table columns are available in the table footnotes. Best-fit absorbing 2 column densities range from negligible to log N 23.5 cm− , equivalent to a visual H ∼ absorption of A 200 mag (Vuong et al. 2003). Temperatures range from kT 0.3 keV V ∼ ∼ to the hardest measurable by ACIS, truncated at kT = 15 keV. The range of total band (0.5 8 keV) absorption corrected luminosities derived from spectral modeling is − −1 29.6 . log Lt,c . 32.4 ergs s . Assuming a 2 keV plasma temperature and an average −2 1 AV = 5 mag visual extinction (log NH 21.9 cm absorbing column), PIMMS gives an ∼ −1 apparent total band luminosity log L 29.3 ergs s estimated for the faintest on-axis t ∼ detection in Table 5.2. Admittedly the patchy extinction throughout our fields implies variation in the estimated sensitivities. This will be discussed in 5.6.1 for the different § regions.

5.3 Identiﬁcation of Stellar Counterparts

In this section we describe the association of counterparts identified in other wavelengths to our X-ray identified sources, to establish their stellar properties. Optical coverage of the RMC in the literature is sparse because of the high visual extinction towards the molecular cloud and the contamination of foreground stars without measured spectral types. The optical and IR catalogs used here for counterparts identification include the Whole-Sky USNO-B1.0 Catalog (Monet et al. 2003, =USNO), 2MASS All- Sky Catalog of Point Sources (Cutri et al. 2003, =2MASS), and the University of Florida FLAMINGOS Survey of Giant Molecular Clouds (PI: E. Lada). The FLAMINGOS observations of the Rosette Complex fields and IR data reduction are described in great detail in Román-Zúñiga (2006). Positional coincidence criteria are used to associate ACIS X-ray sources with optical and NIR (ONIR) sources, as described in the Appendix 2 of Broos et al. (2007) . Likely associations between ACIS sources and ONIR sources are reported in Table 5.5. All 395 ACIS sources in the RMC are listed, 299 of which (76%) have an ONIR counterpart identified. JHK magnitudes from FLAMINGOS photometry are reported if available for Chandra sources, and 2MASS photometry is adopted for areas that were not covered by the FLAMINGOS survey and for bright stars that were saturated (H < 11 mag, Román-Zúñiga 2006). The SIMBAD and VizieR catalog services were used for complementary information; this is reported in the table footnotes, mainly from the TASS Mark IV Photometric Survey of the Northern Sky (Droege et al. 2006) and two large optical surveys of the Rosette (Massey et al. 1995;

1 Portable, Interactive Multi-Mission Simulator is software for high-energy astrophysicists, written and maintained by Koji Mukai. See http://heasarc.gsfc.nasa.gov/docs/software/ tools/pimms.html. 2 Software implementing the matching algorithm is available in the TARA package at http: //www.astro.psu.edu/xray/docs/TARA/ 135

Berghöfer & Christian 2002) that partly covered our Chandra fields. Spectral types for only a handful of stars in this region have been reported, and are noted in Table 5.5. We have examined the spatial distribution of the 96 X-ray sources that do not have any associated ONIR counterpart, in the context of Digital Sky Survey optical plates and CO emission maps. Over 20 of these sources are located in the brightest nebulosity of the photodissociation region (PDR) of the Rosette Nebula, where source detection becomes extremely difficult in the long wavelengths. Others are concentrated in regions where the dense molecular cores are located and show X-rays with harder median energy (MedE) on average, suggesting that they are heavily obscured new members of the embedded clusters. Only a very small fraction of X-ray sources that do not have ONIR counterparts seem to be randomly distributed across the fields. These are likely to be AGN contaminants (e.g., Wang et al. 2007a; Broos et al. 2007). Their small number indicates very low contamination (3%) from extragalactic sources due to absorption by the giant molecular cloud.

5.4 X-rays from Known Massive stars, Protostars, and Other Interest- ing Sources

The only known O star in our field, HD 46485 (O7V) is detected with 310 counts in 20 ks exposure (#164). The X-ray spectrum (Figure 5.4) can be adequately fit with a two-temperature thermal plasma model, with soft kT that is similar to NGC 2244 O 21 −2 stars and a low absorbing column of NH 4.1 10 cm (AV 2.5). The derived −∼1 × ∼ X-ray luminosity is log L 32.05 erg s , also consistent with the late O stars found t,c ∼ in NGC 2244. The brightest source in the field is #119, which has 369 ACIS counts in 20 ks exposure. The corresponding count rate is too low to cause photon pile-up (Townsley et al. 2002). The X-ray spectrum (Figure 5.4) can be fit with a two-temperature thermal plasma model (apec+apec), although the fitting statistic is rather poor (χν = 3.3). Notably there seems to be some abundance anomaly around 1 keV. The derived X-ray 1 luminosity is log L 31.91 erg s− . This source is not matched to any well known t,c ∼ ONIR star; it has only one entry in SIMBAD as a star in Berghöfer & Christian (2002). Its location in the NIR color-magnitude diagram suggests that it is an intermediate mass star. One star (#89) in the PL97 embedded cluster #2 (hereafter cluster PL2, this naming convention also applies to other PL97 clusters; substructure B3) exhibits unusually large color indices (J H = 3.22, H K = 2.41) that are not due to errors in − − photometric measurements. It appears highly reddened (A 30), yet it is bright in the V ∼ K-band (K = 9.9), indicating it is definitely not a background AGN. Its location in the color magnitude diagram suggests that it is a young massive class I protostar. Although it has only 21 net counts, the X-ray properties are all consistent with a highly absorbed source: the median photon energy is very hard (4.6 keV) and the absorbing column from −2 spectral fitting is high (log NH = 23.2 cm ). The derived X-ray luminosity without 31 1 absorption correction is L L = 10 erg s− . Because of the heavy absorption, the h ≃ t soft X-ray emission component is unconstrained. Since it is definitely a stellar source, 32 the intrinsic X-ray luminosity could be comparable to the NGC 2244 O stars L 10 t,c ∼ 136

1 1 erg s− (an intrinsic soft X-ray luminosity of log L 31.8 erg s− is required). There s,c ∼ is a similar candidate massive YSO reported in the Pismis 24 cluster in the NGC 6357 star forming complex (Wang et al. 2007a). Figure 5.5 shows the mid-IR image of #89 neighborhood, using IRAC and MIPS data from the Spitzer archive. #89 is very luminous in the mid-IR band. Another Class I protostar is seen close to #89, where the dust seems to be condensed. They are situated in the photodominated region (PDR) of the NGC 2244 O stars. Block et al. (1993) listed the five brightest K-band sources in the PL4 region. The two IR stars BG/IRS1 and BG/IRS2 form a close pair, whose spectra show Brγ. They have been suggested to be good candidates of HAeBe stars (Hanson et al. 1993). The position of #209 and #207 matches with BG/IRS2 and BG/IRS4, respectively. Both −2 have 50 counts and the spectral fits yield kT 3 keV and a log NH 22.2 cm ∼ −2 ∼ ∼ for #207 and log N 22.5 cm for #209. This absorption column (A 15 20) H ∼ V ∼ − agrees with the AK 1.5 mag derived in Hanson et al. (1993). The absorption corrected ∼ −1 luminosities are log L 31.5 erg s . #210 is matched to BG/IRS 5, but it only has t,c ∼ 8 counts. No further information can be derived from spectral fitting. Ten features showing excited gas from driving sources within the RMC have been reported in an [SII] survey (Ybarra & Phelps 2004), interpreted as either interaction of winds from YSOs with the molecular cloud or Herbig-Haro outflows. We have examined X-ray emission in the vicinity of these sources. Features RMC-A through RMC-D and RMC-K are not in our field of view. In a region 6′ southwest of PL2, an X-ray emitting star (#51) is found near the extended [SII] feature RMC-F; it may be responsible for the emission feature. RMC-H is an arc-shaped feature outlining a globule of gas northwest of PL2, with a possible HH outflow originating in the globule. Three X-ray sources are seen tracing the edge of the globule. One of them is the Class I protostar #58, which may be powering the jet-like outflow. The bulk of the [SII] emission in RMC-H, as well as in two other features (RMC-I and RMC-J) is suggested to be excited by the massive stars to the northwest in NGC 2244 (Ybarra & Phelps 2004). RMC-L has a bow-shock appearance, which Ybarra & Phelps (2004) have suggested to be associated with an HH outflow driven by a star in PL4 or IRAS 06314+0427. If this interpretation is true, the X-ray detected Class I protostar #315 may be the exciting source as it lies eastwards on the axis of symmetry of the bow-shock feature. The separation is rather large, 3.5 pc, ∼ but the association is not impossible given that large scale collimated flows have been detected (e.g., HH80/81, 5 pc long; Marti et al. 1993). Other [SII] features (RMC-E, ≥ RMC-G, and RMC-M) do not have an obvious correlation with our X-ray sources.

5.5 Spatial Distribution of the Stellar Population

By identifying objects with JHK IR excess and further applying a Nearest Neigh- bors technique (Casertano & Hut 1985), the survey of Román-Zúñiga et al. (2007a) confirmed the content of the seven PL97 clusters, and contributed four more previously unknown young clusters associated with the molecular cores in the RMC. In this section we examine the X-ray identified stellar population and its association with the known embedded IR clusters in the fields of our Chandra observations. 137

Spatial concentration of the X-ray point sources is visually apparent in Figure 5.2 (see also TFM03 Figure 6a). In order to outline the clustering of the stellar populations in the RMC, smoothed stellar surface density maps for the total population, the unobscured population (MedE 2.0 keV) and the obscured population (MedE > 2.0 keV) are ≤ ′ ′ created with smoothing kernels of different sizes (r = 0.5 to 3.0 ) following Wang et al. (2007a,2007b) and Broos et al. (2007). In M17, the X-ray stellar density maps proved helpful in highlighting the region’s complex structure, including the compact NGC 6618 cluster, a triggered population along the edge of the HII region, and a previously unknown embedded cluster (Broos et al. 2007). Infrared studies of the NGC 2244 and RMC region with a similar smoothing approach have been done using the 2MASS data (Li 2005; Li & Smith 2005). Compared to IR studies, X-rays are much less affected by dust extinction and the X-ray selection samples young stellar objects of all PMS classes, not just the subgroup with circumstellar disks. This smoothing analysis is done using sources from the three 20 ks RMC ACIS fields (Fields 2, 3, and 4 in TFM03; see also the summary of observations in Table 1 of Chapter 4) to guarantee a uniform exposure time, thus minimizing sensitivity bias. Figure 5.6a and b show the resulting maps using all RMC sources smoothed with a 3′ and 2′ radius kernel, respectively. Figure 5.6c-f show the spatial distributions of the unobscured and obscured populations at the two smoothing scales. The scaling is the same for maps of the unobscured population and obscured population to ensure a fair comparison. Three large-scale structures are seen as overdensities irrespective of the adopted smoothing kernel size. We define these regions A, B, and C, guided by the contours using the density map for the total population smoothed with 3′ radius kernel. Small-scale substructures within the three large structures are also apparent in the density maps. Based on the 2′ map for the total population, seven smaller regions of overdensities (A1-3, B1-3, C1) are defined. In addition, in structure C for the obscured population, there is a distinct substructure extending from C1. It is not as significant in the total or unobscured populations because of the large contrast to the high stellar density in C1. We outline this substructure as region C2 using the 2′ map for the obscured population. Note that the substructures are independently defined, not simple divisions of the large- scale structures. Therefore the total numbers of sources contained in the substructures do not necessarily add up to the number of sources within the large-scale structures. While the regions are defined based on density contours, we remind the reader that there is no unique boundary as where to draw the lines to group sources. Nevertheless, we are confident that these structures are physical, and not artifacts caused by the slightly degraded X-ray sensitivities at large off-axis angles on the ACIS-I CCDs. The overdensities are not located at the center of the fields and are indeed active star-forming spots in the RMC; structure B and C spatially match the regions where previously identified embedded IR clusters were found (PL97, RL07). Some substructures (e.g., B2, B3, C1) are clearly associated with the dense molecular clumps. The approximate center, extent, the numbers of X-ray sources and IR sources contained, and previously known associations for the structures are summarized in Table 5.6. Structure A is located in the top-right corner of Field 2, showing significant ag- gregation of mostly unobscured X-ray sources. We conclude that an unobscured cluster 138 is present here. This cluster was not identified in any of the previous IR surveys: it is out of the field of view in the PL97 survey, and because of the few IR-excess sources in this region, it is not detected as a cluster in RL07. However the concentration of IR sources can be seen in the spatially complete 2MASS stellar surface density map (see Figure 1 in Li & Smith 2005). Substructure A1 appears as an elongated extension towards the north. A concentration of 5 X-ray stars constitutes the weak feature A2 to the south of the main struture A3. The “L”-shaped structure B is overall more obscured as suggested by the significant fraction of obscured sources. It is resolved to three roughly spherical clumps, which have similar sizes. B1 and B3 have comparable densities in the unobscured and obscured populations, while B2 is more embedded. B1 and B2 both contain bright MSX mid-IR sources. The position of B3 matches the small IR cluster PL2. In the south-east of our fields (lower-left corner of Figure 5.6), one group of X-ray sources (C1) shows the most significant clustering in both the unobscured and obscured populations. This embedded cluster of stars was first noted in K′-band imaging by Block et al. (1993) and further studied spectroscopically in Hanson et al. (1993). By comparing with the PL97 clusters, this X-ray group corresponds to their embedded cluster PL4 although the center of C1 is 1′ north of the IR position for PL4. It is also noted ∼ as a 2MASS IR cluster (Bica et al. 2003; Li & Smith 2005b). C2 is only visible in the density map for the obscured population, indicating its embedded nature. This region also appears much darker in the DSS optical image. It appears as a curvy extension of PL4 in the X-ray source density maps, but may be associated with another embedded IR cluster PL5, which is only partially covered in our ACIS data at the corner of Field 4. One third of the X-ray sources in C1 are hard sources, whereas 76% of C2 sources are hard sources. This indicates that C2 is more embedded than C1.

5.6 Properties of the X-ray Sampled Stellar Clusters

5.6.1 Sensitivity Limits As stated in 5.2, because of the highly clumpy molecular materials distributed § throughout the fields, the sensitivity limits of our X-ray observations are unlikely to be the same. Besides, due to the degraded point spread function (PSF) of the Chandra mirrors at large off-axis angles, the off-axis sensitivity is approximately 2 times worse than the sensitivity limit for on-axis sources based on many previous observations. In order to assess the relative sizes of the X-ray sampled clusters, we need to estimate the X-ray sensitivity limit in the three regions. When inspecting the distribution of the source counts of our detections, we are confident that a point source with 8 net counts would be detected at any large off-axis angles. This gives a lower limit of the count rate that guarantees the source will be picked up by our algorithm. The typical 21 −2 extinction in regions A, B, and C is AV 1 mag (NH 1.6 10 cm ), 5.1 mag 21 −2 ∼ ∼22 −×2 ∼ (N 8 10 cm ), and 9.7 mag (N 1.6 10 cm ) (Román-Zúñiga et al. H ∼ × ∼ H ∼ × 2007a), respectively. Assuming a kT = 2 keV thermal plasma for the spectral model of a typical PMS star, PIMMS gives an absorption corrected hard band (2-8 keV) luminosity 1 log L 29.6, 29.8, and 30.0 erg s− in regions A, B, and C respectively, for the h,c ∼ ∼ ∼ 139 adopted distance d=1.4 kpc. By comparing to the X-ray luminosity distribution of the Chandra Orion Ultradeep Project (COUP) observation of the Orion Nebula Cluster, the completeness limits here imply that our observations probe 30%, 27%, and 20% ∼ ∼ ∼ of the total X-ray emitting population in regions A, B, and C, assuming the RMC young stellar clusters have XLFs that are similar to the COUP XLF.

5.6.2 NIR Color-Color and Color-Magnitude Diagrams To investigate the IR properties of stars in these X-ray sampled clusters, we focus on Chandra-selected stars with high-quality JHK photometry (error in both J H and − H K colors < 0.1 mag). Near-IR color-color and color-magnitude diagrams are created − for sources within the boundaries of regions A, B, and C. Figure 5.7 shows the J H vs. H K color-color diagrams. Between the left − − two dashed lines is the color space associated with Class III objects (diskless WTTS), which are reddened by interstellar extinction. To the right of this reddened band are sources exhibiting significant K-band excess (here we require E(H K) > 2σ(H K)). − − The color space between the middle and right-most dashed lines is occupied by Class II objects (PMS stars with circumstellar accretion disks). Located beyond the right-most dashed line are protostars still possessing thick envelopes (Class I objects). The colors of most sources in region A are consistent with PMS stars with little reddening (A 1, V ∼ typical of NGC 2244 stars; Wang et al. 2007b). A few Class II sources show higher AV . Sources in region B have larger range of AV ; some sources are reddened with AV > 5 and the K-excess sources are among the most reddened. Despite the smaller number of sources in region B, there are more IR-excess sources present in the color-color diagram. Three Class I sources are found in this region, which strongly indicates the youth of this cluster of stars. Sources in region C shows a clump of late-type PMS stars with A 2 V ∼ and a dozen heavily obscured sources with AV > 10. The NIR J vs. J H color-magnitude diagrams (Figure 5.8) provide an approx- − imate range of the mass distribution and local absorptions for the same stars shown in Figure 5.7. This is achieved by deriving the dereddened location of the stars on the PMS isochrones, assuming the standard reddening. The assumption of age (1 or 2 Myr) does not affect the estimated mass significantly for these late-type PMS stars. The distribution for stars in all regions in the color-magnitude diagrams suggests that most X-ray selected stars are PMS stars with masses 0.5 2M⊙. No candidate O star is found. ∼ − 5.6.3 Comparison with CO Emission Morphology and Román-Zúñiga et al. (2007a) NIR Clusters To further investigate the distribution of Class I, II, III objects and the overall association between X-ray selected stars and the molecular clumps, we classify all the IR counterparts to our X-ray sources and show their spatial distribution in Figure 5.9. 12 The dense molecular material is outlined by the CO emission contours. The overall distribution of the Class II/I sources are more confined to the CO molecular ridge than the Class III stars. It could be interpreted that the older Class III stars have drifted farther away from the cluster while the Class II/I sources have not had the time yet to drift away from their birth places. For example, if a star formed close to the center 140

1 of RMC C cluster and had a characteristic transverse speed of 1 km s− , in 2-3 Myr ∼ it would travel 2-3 pc and have been considered as being formed in the distributed population. This velocity is a reasonable assumption compared to the velocity dispersion observed in many clusters. Given the uncertainty in the stellar drifting velocities, the timescale is also compatible with the PMS evolution from Class II to Class III objects. The location of cluster A is within the bright nebulosity of the Rosette HII region, where most of the molecular materials are gone. This group of unobscured stars is very likely among the earliest population that was triggered by the expansion of the HII region. The overdensity of stars in B, including the PL2 cluster, is situated right on the rim of molecular material as outlined by the CO emssion. This region is probably young, as indicated by the presence of three Class I objects. Out of our field towards the southwest there is the PL1 cluster, similarly embedded in a molecular clump. At large scale they all lie on the outermost expanding shell of the interstellar medium swept up by the winds of the massive stars in NGC 2244 and the expanding HII bubbles. The molecular clumps are being photoevaporated and will eventually become unobscured clusters like region A. Region C contains the large and extended population of embedded stars (PL4 and possibly a small portion of PL5). It has been recognized that the most massive clumps (like region C) in the RMC are producing the embedded clusters (Heyer et al. 2006; Román-Zúñiga et al. 2007a). C1 (PL4) is partially emerging from the molecular core and C2 is still embedded. Note that in spite of their separations from the NGC 2244, these clusters are still in interaction with the ionized nebula (Celnik 1983; RL07). A deep FLAMINGOS NIR imaging survey of this region is presented in Román- Zúñiga et al. (2007a). The survey area is much larger than our Chandra survey, aiming to identify young embedded clusters across the Rosette star-forming complex based on the distribution of IR excess sources. By comparing our surface density maps (Figure 5.6) to locations of the independently IR-selected clusters (see Figure 7 in Román-Zúñiga et al. 2007a), our X-ray-defined structures nicely trace the embedded clusters in RL07 studies. As a cross check, we apply the same Nearest Neighbor analysis in Román-Zúñiga et al. (2007a) to our Chandra RMC sources, and obtain the surface density contours of the X-ray stars. To explore the nature of the clusters, Figure 5.10 shows the contours for the total X-ray-selected sample, the unobscured population, and the obscured population. The locations and morphologies of the clusters closely resemble the structures that we defined using the smoothed surface density maps for X-ray sources, indicating the robustness of the clustering irrespective of the methods used to outline the clustering. Not surprisingly, the density of obscured X-ray sources decreases significantly towards the HII region ionized by the NGC 2244 cluster, as the molecular materials become sparse. It is noticeably higher in the cloud cores, and enhanced towards cluster RELF08, a new embedded cluster that was identified in Román-Zúñiga et al. (2007a). This feature confirms that cluster RELF08 is a very young cluster in the cloud core (Román-Zúñiga et al. 2007a). Further X-ray coverage is desired to reveal more of the stellar content in this region. Because of the X-ray sampling of young stars without IR-excess, X-ray maps show more structures than were seen in the IR-excess density contours. In particular, clustering of unobscured sources is significant near the boundary between the NGC 2244 cluster and the PL2 cluster in the molecular cloud in our X-ray surface density map. As 141 noted in 5.5, this small cluster is previously uncataloged in IR due to the small number § of IR-excess stars. Its location and low fraction of IR-excess sources suggest that it may have formed as a result of triggering by NGC 2244, similarly to the small cluster NGC 2237 to the west of NGC 2244 (Li 2005; Román-Zúñiga et al. 2007a).

5.7 Formation Modes of Clusters and Stars

5.7.1 Sequential Star Formation and Spontaneous Star Formation There is no doubt about the manifestations of active cluster formation in the RMC as the NIR studies reveal aggregations of young stars with IR-excess (Phelps & Lada 1997; Román-Zúñiga et al. 2007a). There have been extensive discussions about the possible formation mechanism for the embedded star clusters in the RMC. PL97 suggested that there may be more than one way to form clusters in GMCs: PL1, PL2, PL4 are within the ionization front of the Rosette Nebula and may be triggered in a process known as sequential star formation (Elmegreen & Lada 1977), where OB stars at the edge of a molecular cloud drive ionization fronts and shock fronts into the cloud and consequently a layer of dense neutral material becomes gravitationally unstable to form new stars in a self-propagating manner. In fact, at a larger scale, sequential formation of OB clusters in GMCs has occurred in the Mon OB2 association and NGC 2244 is the youngest OB cluster among the sub-associations (Blaauw 1964). Depending on the inhomogeneities within the molecular cloud (Block 1990; Cox et al. 1990), PL3, PL5, and PL6 could be triggered as well. However, it is very difficult to explain the formation of cluster PL7 that is distant from the ionized nebula (Phelps & Lada 1997). Spontaneous star formation (Elmegreen 2000) along the midplane of the cloud could be also at work here, although it cannot account for the formation of the off-plane clusters PL1 and PL3. Although incomplete (i.e., no coverage of PL1, PL3, PL6, and PL7), our X-ray survey provides an unbiased census of star formation over a large spatial extent of the RMC and confirms the widespread presence of embedded forming clusters. This enables us to revisit the triggered or spontaneous star formation problem observationally with the X-ray structures, complemented by IR data. Without spectroscopic information to determine the ages of cluster members, it is challenging to reconstruct the sequence of star formation events in the RMC. How- ever there are hints for the relative youth among the clusters. The X-ray discovered unobscured cluster (structure A) is probably the oldest, as suggested by the absence of molecular material and low fraction of IR excess sources. This subcluster is very similar to the satellite cluster NGC 2237 to the west of NGC 2244 (Li 2005; RL07; Wang et al. 2007 in preparation), and probably represents the earliest star formation events during the formation of NGC 2244. Region C1 (PL4) has largely emerged from the molecular core and has low fraction of obscured X-ray sources and IR-excess sources, suggesting that it is probably older than the more embedded C2 and B (PL2). Morphologically this is consistent with the Román-Zúñiga et al. (2007a) conclusion that the more prominent and extended clusters may be more evolved and older. 142

Interestingly, the concentration of three Class I protostars in PL2 is right at the edge of the molecular cloud facing the NGC 2244 OB stars, as seen from the large scale map (Figure 1). This strongly indicates that PL2 may represent a recent epoch of triggered star formation at the CO rim, where external stellar radiation irradiates the molecular cloud and stars are forming at the edge of the cloud (Elmegreen & Lada 1977; Dale et al. 2007). A similar case of triggered star formation is reported in a comprehensive multiwavelength study of RCW 108 (Comerón et al. 2005), where a group of IR-excess stars is concentrated towards the edge of the cloud that faces the ionizing stars in NGC 6193. Román-Zúñiga et al. (2007a) report that the fraction of IR-excess sources appears to increase with distance from the Rosette Nebula (especially for clusters separated more than > 20 pc from NGC 2244, not covered in our field). This implies a temporal sequence of star formation across the complex, which they attribute to the formation and evolution of the molecular cloud. The IR excess sources in our sample (Table 6) do not show a clear trend of increasing fraction from RMC A to B to C. One plausible scenario is that, together with the formation of NGC 2244, satellite clusters such as cluster A and NGC 2237 formed from substructures in the primodial cloud. The remaining gas surrounding A quickly dispersed after the HII region developed. Since the region of RMC C is where the cloud is still interacting with the photoionized nebula (Heyer et al. 2006), the pre- existing clumps inside the RMC could have been triggered to collapse by the shock front from the nebula and formed PL4 and PL5. The age of NGC 2244 ( 2 Myr; see the ∼ overview of the NGC 2244 in Chapter 4) could be considered an upper limit for the ages of the embedded clusters. PL2 then was triggered to form near the boundary of the molecular cloud as the surface of the cloud was evaporated by the radiation and neutral gas was compressed. Notably the IR excess fraction in Román-Zúñiga et al. (2007a) is higher in PL2 (33% 6) than in PL4 (24% 3), which is consistent with our X-ray-selected sample and ± ± our speculation of its extreme youth. If further follow-up observations determine that PL2 is the latest epoch of star formation and PL2 is actually younger than PL4, the above triggering sequence is still valid except that it takes longer to form PL2 (after the formation of PL4). In this scenario, the PL2 cluster could be formed through the so-called “collect-and-collapse” process (Elmegreen 1998; Deharveng et al. 2005). Whitworth et al. (1994) studied fragmentation of swept-up shells of interstellar gas by expanding HII regions and stellar wind bubbles. When critical condensation is reached, the layers of neutral gas become gravitationally unstable and eventually fragment to form stars. −1 Assuming an isothermal sound speed as = 0.2km s in the shocked gas, an emitting rate ˙ 49 −1 37 of hydrogen ionizing photons NLyC 10 s for O stars, a wind luminosity Lw 10 −1 ∼ −3 ∼ erg s , and a hydrogen nuclei number density n0 1000 cm in the interstellar ∼ medium, the timescale at which fragmentation occurs is 0.9 1.6 Myr, the radius of ∼ − the shell when fragmentation starts is 6 10 pc, and the expected masses of fragments ∼ − are 10 23M⊙, as predicted by Whitworth et al. (1994) models for expanding HII ∼ − regions and stellar wind bubbles. The relative youth of PL2, its separation from the NGC 2244 O stars, and the presence of a massive Class I protostar (#89) support that collect-and-collapse may be the working mechanism to form a very young cluster at the CO rim. 143

5.7.2 Clustered Star Formation vs. Distributed Star Formation McKee (1989) and Bertoldi & McKee (1996) show in their photoionization regu- lated star formation model that little recent star formation activity should exist in the outer layers of molecular clouds because of the high ionization fraction there. Observa- tions established that the majority of stars in nearby clouds form in embedded clusters, (e.g., L1630 Lada et al. 1991; Li et al. 1997), although difference among regions exist. In part of the Orion complex L1641, Strom et al. (1993) and Allen (1996) discovered an isolated population of PMS stars widely distributed within the cloud using imaging and spectroscopic studies. They concluded that 50% of the stars could be formed via the distributed mode. What is this fraction in more distant GMCs? Because of the heavy contamination from field stars, the fraction of distributed star formation in these clouds has been largely unstudied. With the X-ray identified young stars suffered little contamination, we are able to examine the fraction of stars that are formed in clusters and in lower number density in the RMC, and find clues for which mode of star formation (distributed vs. clustered) is dominant in the RMC. The total number of sources in the three fields is 553. The total number of X-ray sources contained in structures A, B, and C is 303. Given the short exposure time and the presence of the molecular cloud, at most 10% of these could be field and extragalactic contaminants. This implies that .35% of stars could be formed in a distributed fashion throughout the RMC region and clustered star formation is the governing mechanism here. However, further interpretation of this fraction should proceed with caution. There are at least two lines of complexity here. First, observationally a significant number of faint sources are still missing due to the low sensitivity limit. Presumably there is no preference for them to be clustered or distributed. Secondly the distributed stars may not be born in a distributed formation mode. Numerical simulations (Bate et al. 2003) show that the formation process of a cluster of stars is highly dynamic and chaotic; stellar encounters violently eject some of the young stars from multiple systems. As a formed stellar cluster expands, only a fraction of the cluster members can be retained and expulsion of gas brings dissolution of the cluster and release of the stars (Kroupa et al. 2001; Davies et al. 2006; Weidner et al. 2007). Thus the observed distributed populations may not be formed in situ. Note that Kroupa & Boily (2002) studied different regimes of clusters and concluded that the fraction of stars retained during the cluster expansion is 50% for low-mass clusters, close to that seen in distributed populations. ≈ 5.8 Summary

In this chapter, we describe the high spatial resolution Chandra/ACIS imaging observations of the RMC region. A total of 395 X-ray point sources that belong to the RMC are detected in addition to the 919 X-ray sources in NGC 2244 (Chapter 4). Using USNO, 2MASS, and deep FLAMINGOS data, 299 of the RMC X-ray sources (76%) have an ONIR counterpart identiﬁed and most are PMS stars. This study represents one of the comprehensive census of the PMS stellar populations in the RMC region. 144

We investigate the spatial distribution of the X-ray sources and define three distinctive structures based on smoothed stellar surface density maps. Structure B and C are associated with the previously known embedded IR clusters PL2 and PL4, while structure A is a new X-ray-identified unobscured cluster. Near-IR color-color and color- magnitude diagrams are created for these X-ray sampled clusters. Three Class I sources are found within region B, suggesting the relative youth of the PL2 cluster. We compare the distribution of Class I, II, and III stars with the morphology of the molecular content as outlined by the CO emission and find that the Class II/I sources are more confined to the midplane of the cloud. Our X-ray-defined regions confirm the clustering identified in RL07. Cluster A represents the early epoch of star formation when NGC 2244 formed. The shock front from the Rosette Nebula may have triggered the collapse of pre-existing clumps inside the molecular cloud and form PL4, although spontaneous formation due to large scale cloud turbulence is also possible for the formation of PL4 and other clusters further from the nebula. We speculate that the PL2 cluster is very young, and represents triggered star formation at the edge of the cloud due to the collect-and-collapse mechanism. The concentration of X-ray identified young stars implies that .35% of stars could be in a distributed population throughout the RMC region and clustered star formation is the dominant mode. The X-ray properties of a few interesting X-ray sources were presented, along with their IR characteristics. 145 median E 9.8 3.9 (ks) (keV) (16) (17) 10.6 1.4 16.6 1.5 10.6 1.3 16.9 2.0 29.8 5.1 17.1 2.6 16.7 4.4 ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 .... a 35.4 1.5 -5 g... -5 .... a 35.2 2.0 -5 g... -5 .... a 34.2 1.4 -5 .... a 18.7 1.1 -5 g... -5 .... a 35.4 2.1 -5 g... -5 .... a 18.0 1.8 -5 g... -5 .... a 33.6 2.1 -5 .... a 18.5 1.4 -5 .... a 33.3 2.7 -5 .... a 16.2 1.9 -5 .... a 33.2 1.6 -5 .... a 33.6 1.1 -5 .... b 35.6 2.3 -5 .... a 18.1 3.1 -5 .... c 35.6 1.1 -5 .... c 17.5 0.8 -5 .... a 17.7 2.6 -5 .... b 35.1 1.5 -5 .... a 35.3 2.4 -5 g... -5 .... a 34.6 1.2 -5 .... a 33.0 2.2 -5 .... a 33.3 2.5 -5 .... a 17.4 1.1 -5 .... a 35.7 2.7 < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < 6.7 0.91 1.9 4.4 0.89 1.6 -4.6 .... a 17.9 2.6 6.7 0.90 1.7 -3.6 .... a 17.3 3.6 13 0.9 2.8 0.90 0.89 2.1 1.2 -4.7 -3.7 ...... b a 34.8 18.7 1.0 5.2 0 1.3 0.90 1.6 -4.0 g... 1 2.9 0.89 2.2 0 6.6 0.90 1.9 -4.4 .... b 35.7 3.7 2 2.7 0.90 2.0 -4.6 .... a 35.3 1.4 2 0.6 0.91 3.1 4 6.9 0.90 2.2 4 6.0 0.90 1.5 -3.1 g... 5 3.9 0.90 2.0 7 3.5 0.90 2.6 2 5.8 0.89 2.1 2 0.8 0.89 1.6 3 3.9 0.91 2.3 9 0.0 0.89 1.5 -3.1 .... a 33.5 1.0 6 9.6 0.90 2.9 14 0.5 3.0 0.89 0.89 1.7 1.7 -3.5 -3.9 ...... a a 35.5 33.6 1.2 1.8 6 1.9 0.90 3.6 081 7.91 1.0 0.89 0.5 0.89 5.9 0.90 0.89 1.9 1.7 1.8 -3.6 2.2 -3.2 -4.2 ...... a a b 35.4 33.8 35.6 4.3 0.9 1.2 747 0.83 2.0 0.90 0.0 0.90 6.8 0.90 0.89 1.6 1.5 1.6 -3.4 2.1 -3.2 -3.4 ...... a a a 35.7 35.7 33.2 1.2 1.9 1.1 .0 28.7 0.89 6.6 .5 1.8 0.89 2.4 .4 0.0 0.91 2.7 .8 5.8 0.90 2.2 .6 8.9 0.90 2.3 .1 10.5 0.90 4.4 .4 0.0 0.90 2.7 .9 8.7 0.90 2.9 .5 4.0 0.90 3.4 .8 2.8 0.90 3.2 .2 7.7 0.90 2.3 .3 2.1 0.89 2.6 .3 19.1 0.90 4.9 .6 10.9 0.89 3.4 .8 3.4 0.90 2.3 .5 12.0 0.90 3.1 6 2.7 42.2 0.90 11.0 Extracted Counts Catalog: Primary Source Properties Net ∆Net Bkgd Net PSF Signif log ) Full Full Full Hard Frac ′ θ Chandra ) ( ′′ Err J2000 δ Position Table 5.1. J2000 (deg) (deg) ( α Source 1 063232.43+043705.7 98.135158 4.618251 0.7 6.0 7.5 3.4 0.5 34 063236.58+043611.4 063238.08+043250.1 98.152431 98.158706 4.603172 4.547265 0.9 1.2 6.6 9.9 6.2 16.8 3.2 4.9 0.8 2. 5 063239.50+043628.9 98.164603 4.608042 0.6 6.3 12.3 4.1 0. 6 063239.51+043124.9 98.164639 4.523587 1.2 9.9 11.7 4.4 3. 7 063240.73+043653.3 98.169734 4.614833 0.5 5.9 14.4 4.4 0. 89 063242.33+043217.9 063245.25+043206.5 98.176413 98.188575 4.538329 4.535152 1.3 0.7 9.0 8.4 7.0 21.4 3.5 5.3 2.0 1. 5455 063308.46+043731.9 063308.49+043101.3 98.285256 98.285412 4.625528 4.517037 0.9 0.6 6.0 4.3 9.9 3.7 4.1 2.5 3. 0. 49 063306.79+044115.9 98.278330 4.687764 0.3 6.8 58.0 8.3 2 50 063306.82+044347.6 98.278457 4.729895 0.9 8.0 12.5 4.7 4 48 063306.45+043519.8 98.276875 4.588861 1.0 4.8 6.0 3.2 1. 51 063307.48+043048.6 98.281172 4.513505 0.4 4.6 12.6 4.1 0 52 063308.31+043752.2 98.284630 4.631182 0.2 6.0 144.3 12. 47 063306.39+043454.0 98.276647 4.581686 0.2 2.7 8.9 3.5 0. 17 063249.80+043641.9 98.207531 4.611655 0.7 6.8 10.2 4.0 1 2021 063251.49+043801.9 063251.73+043621.4 98.214575 98.215568 4.633882 4.605954 0.8 0.7 6.3 6.7 8.0 10.4 3.7 4.0 2. 1 18 063251.19+043412.1 98.213307 4.570031 0.5 7.7 29.9 6.3 3 1314 063248.45+043539.8 063248.65+043404.4 98.201886 98.202709 4.594400 4.567892 0.8 0.8 7.3 8.0 8.8 14.6 3.8 4.8 2. 3 22 063252.30+043744.3 98.217949 4.628973 0.6 6.3 15.1 4.6 1 46 063305.02+043857.9 98.270957 4.649426 0.6 6.1 9.6 3.8 1. 19 063251.29+043714.0 98.213746 4.620577 0.7 6.4 5.6 3.2 1. 15 063248.96+043027.4 98.204005 4.507629 1.0 8.2 8.5 3.7 1. 23 063253.46+043716.8 98.222774 4.621353 0.5 6.4 19.5 5.1 1 24 063253.93+043622.6 98.224726 4.606290 0.5 6.6 17.2 4.9 1 10 063246.11+043612.9 98.192141 4.603596 0.8 7.3 10.8 4.1 2 45 063304.40+043156.7 98.268367 4.532443 0.4 4.2 8.8 3.5 0. 25 063253.98+043828.4 98.224955 4.641246 0.6 6.2 12.7 4.3 1 11 063246.72+043831.4 98.194699 4.642082 0.5 4.5 5.8 3.0 0. 2728 063254.51+043603.7 063255.18+043651.6 98.227154 98.229955 4.601030 4.614336 0.9 0.4 6.6 6.4 6.1 34.7 3.4 6.5 1. 1 40 063301.92+044040.8 98.258008 4.678006 0.5 6.3 19.4 5.1 1 3031 063256.25+043509.132 063257.10+044001.7 063257.18+043017.4 98.234407 98.237940 4.585862 98.238269 4.667162 1.0 4.504861 0.8 6.8 0.7 6.1 6.6 6.9 6.6 10.2 3.5 3.4 3.8 2. 1. 0 4142 063302.56+043426.743 063302.67+043512.144 063302.74+043639.9 063304.05+043319.9 98.260667 98.261139 4.574111 98.261439 4.586702 98.266914 1.2 4.611095 0.7 4.555538 6.6 0.9 6.3 0.3 6.0 3.5 9.0 7.2 7.9 4.1 8.9 3.7 3.7 4. 3.5 2. 2. 0. 34 063259.40+043857.7 98.247501 4.649379 0.5 6.1 16.5 4.8 1 3537 063259.45+043653.338 063300.96+043843.639 063301.50+044112.8 063301.90+043213.0 98.247742 98.254003 4.614826 98.256253 4.645463 98.257942 0.9 4.686902 0.8 4.536967 6.2 0.8 6.1 0.4 6.5 4.5 6.3 5.6 6.3 3.4 8.7 3.2 3.4 1. 3.5 1. 1. 0. # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq CXOU J 146 median E 9.0 1.5 8.0 1.4 8.3 6.7 (ks) (keV) (16) (17) 17.5 1.4 12.4 2.5 16.8 1.0 16.3 4.4 19.0 1.0 19.0 1.7 32.2 1.1 19.2 1.0 14.1 1.0 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 .... a 17.7 2.4 -5 g... -5 g... -5 g... -5 .... a 17.0 2.1 -5 .... a 19.1 1.2 -5 .... a 18.1 2.3 -5 g... -5 .... a 16.5 3.0 -5 .... a 19.2 2.0 -5 .... a 19.1 2.6 -5 g... -5 .... a 32.5 3.3 -5 .... a 19.2 1.3 -5 .... a 17.9 1.0 -5 .... a 19.0 0.9 -5 .... c 19.1 2.7 -5 .... a 18.1 4.4 -5 .... a 18.0 3.4 -5 .... a 17.5 1.4 -5 g... -5 .... a 18.6 2.4 -5 .... a 18.9 2.8 -5 g... -5 g... -5 .... a 18.8 2.9 -5 .... a 17.1 1.5 -5 .... a 19.3 4.6 -5 .... a 19.3 3.3 -5 .... a 19.0 1.5 -5 .... a 18.8 2.5 -5 .... b 19.3 2.1 -5 .... a 18.8 1.5 -5 .... a 18.5 1.7 -5 .... a 19.2 1.2 -5 .... a 18.5 1.6 -5 .... a 16.8 1.9 < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < 71 0.5 0.0 0.90 0.90 1.3 1.2 -3.2 .... a 18.3 1.7 4 5.7 0.89 2.3 11 0.3 0.89 0.0 0.89 1.8 1.4 -4.8 .... a 16.4 1.1 1 3.9 0.90 1.8 21 0.9 1.9 0.89 0.89 1.2 1.2 -4.6 .... a 18.9 1.2 0 8.3 0.90 1.9 1 4.9 0.89 1.6 1 1.0 0.89 2.0 6 5.6 0.90 2.0 1 0.0 0.89 1.0 -4.4 .... 1 1.0 0.89 1.0 -4.3 .... 1 5.9 0.89 2.0 1 2.9 0.89 1.8 1 5.9 0.90 2.0 1 0.9 0.89 1.6 51 0.7 0.0 0.91 0.90 1.3 1.8 -3.9 .... a 17.1 1.2 2 2.8 0.89 2.1 7 4.5 0.90 2.2 .9 13.3 0.90 3.4 .3 24.8 0.90 8.5 .0 9.0 0.90 2.6 .0 0.9 0.89 2.1 -4.9 .... a 33.2 1.1 .1 11.0 0.90 3.6 .1 18.0 0.89 3.9 .9 12.2 0.90 2.9 .1 4.0 0.89 2.7 .8 3.4 0.90 3.2 .1 14.0 0.90 3.4 .6 18.6 0.90 4.0 .1 0.0 0.90 1.0 -4.1 .... .1 19.9 0.90 4.5 .1 2.9 0.89 1.4 .0 8.0 0.89 5.0 .1 20.9 0.90 3.7 .1 13.9 0.90 3.0 .2 1.9 0.90 1.2 -4.4 .... b 17.8 1.3 .0 2.0 0.90 0.7 -3.4 g... .3 29.8 0.90 8.0 0.6 6.5 0.90 3.9 1.7 10.7 0.91 5.7 0.1 8.9 0.89 2.9 6 18.5 5.6 0.81 3.7 4 4.5 3.2 0.89 5.4 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Table 5.1—Continued ) Full Full Full Hard Frac ′ θ ) ( ′′ Err J2000 δ Position J2000 (deg) (deg) ( α Source 56 063308.51+042832.6 98.285474 4.475746 0.5 6.5 19.1 5.0 0 59 063309.40+043106.7 98.289180 4.518551 0.1 4.1 88.7 9.9 0 61 063310.06+042547.4 98.291920 4.429842 1.4 9.1 13.0 4.5 3 6263 063310.31+042931.8 063311.18+043406.4 98.292974 98.296616 4.492180 4.568464 0.8 0.3 5.4 1.6 4.3 3.9 2.8 2.5 0. 0. 64 063311.23+043007.3 98.296795 4.502034 0.4 4.8 9.6 3.7 0. 6566 063311.25+042732.467 063311.30+044249.1 063311.32+043533.2 98.296898 98.297089 4.459019 98.297186 4.713639 1.0 4.592566 0.9 0.3 7.3 7.9 1.7 6.9 10.0 4.9 3.4 4.1 2.8 1. 3 0. 68 063311.39+043209.5 98.297493 4.535986 0.3 2.9 6.9 3.2 0. 6970 063311.54+043136.5 063311.93+043119.1 98.298114 98.299740 4.526815 4.521989 0.4 0.5 3.4 3.6 3.8 3.9 2.5 2.5 0. 0. 71 063312.44+042739.8 98.301851 4.461063 0.9 7.1 8.0 3.5 1. 72 063312.84+043458.5 98.303509 4.582937 0.1 1.1 19.9 5.0 0 73 063312.89+043552.7 98.303745 4.597975 0.1 1.6 22.9 5.3 0 74 063313.00+043214.1 98.304182 4.537265 0.3 2.7 5.9 3.0 0. 76 063313.38+044323.9 98.305783 4.723316 0.8 8.5 16.1 5.0 3 77 063313.85+043250.9 98.307737 4.547486 0.2 2.0 12.9 4.1 0 79 063313.96+042827.9 98.308180 4.474439 0.5 6.3 17.2 4.8 0 80 063314.34+043612.1 98.309789 4.603385 0.2 1.7 7.9 3.4 0. 81 063314.56+043536.8 98.310693 4.593559 0.1 1.1 17.9 4.8 0 82 063314.57+042910.2 98.310731 4.486167 0.6 5.6 8.4 3.5 0. 83 063314.72+043605.0 98.311375 4.601406 0.3 1.5 2.9 2.3 0. 84 063314.76+042853.4 98.311516 4.481526 0.4 5.8 24.4 5.5 0 85 063314.83+043633.6 98.311810 4.609335 0.4 1.9 2.9 2.3 0. 86 063314.88+043747.3 98.312014 4.629811 0.2 3.1 28.9 5.9 0 87 063315.24+043305.8 98.313513 4.551637 0.1 1.7 35.0 6.4 0 88 063315.33+043125.5 98.313907 4.523760 0.3 3.3 7.9 3.4 0. 89 063315.44+043459.3 98.314372 4.583146 0.1 0.5 20.9 5.1 0 90 063315.60+043504.0 98.315013 4.584455 0.1 0.5 14.9 4.4 0 91 063315.64+043209.8 98.315178 4.536076 0.3 2.6 6.9 3.2 0. 92 063316.19+043452.6 98.317480 4.581295 0.2 0.3 7.9 3.4 0. 94 063316.71+043709.1 98.319658 4.619222 0.3 2.5 5.9 3.0 0. 95 063316.82+043029.2 98.320111 4.508118 0.1 4.2 79.7 9.5 0 9697 063318.10+042927.8 063318.27+043527.6 98.325430 98.326161 4.491066 4.591010 0.7 0.2 5.2 0.8 4.5 6.9 2.8 3.2 0. 0. 98 063319.07+043038.8 98.329468 4.510799 0.4 4.1 8.8 3.5 0. 99 063319.44+044050.7 98.331030 4.680767 0.7 6.2 9.3 3.7 0. # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 106 063322.23+044650.7 98.342639 4.780753 1.1 11.9 30.5 7. 107 063323.21+042444.6 98.346743 4.412406 0.9 10.1 42.5 7. 105 063322.17+042905.6 98.342384 4.484912 0.4 5.7 23.4 5.4 108 063323.22+043522.8 98.346781 4.589693 0.3 1.7 2.9 2.3 0 104 063321.57+043244.2 98.339877 4.545635 0.3 2.3 4.9 2.8 0 103 063320.72+044215.0 98.336363 4.704186 0.5 7.6 45.3 7.4 101102 063319.63+043824.5 063319.77+043703.0 98.331792 98.332377 4.640161 4.617504 0.5 0.2 3.8 2.4 3.8 13.9 2.5 4.3 0 100 063319.56+043358.5 98.331506 4.566264 0.4 0.9 2.0 2.0 0 Seq CXOU J 147 median E 7.4 1.0 7.4 1.3 (ks) (keV) 23.7 2.8 16.3 4.3 13.8 1.2 25.4 1.5 16.3 1.8 10.1 1.7 (16) (17) 18.8 2.8 16.7 1.2 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 .... a 16.4 3.2 -5 .... a 17.6 5.3 -5 .... a 16.4 2.8 -5 g... -5 .... a 17.8 3.0 -5 .... a 17.1 1.7 -5 .... a 17.6 1.6 -5 .... a 18.0 1.7 -5 .... a 17.0 2.0 -5 .... a 16.8 1.1 -5 .... b 17.5 2.8 -5 .... a 18.2 2.8 -5 .... a 17.0 1.6 -5 .... b 15.1 2.9 -5 g... -5 g... -5 .... a 17.1 1.3 -5 .... a 18.1 1.0 -5 .... a 18.1 3.8 -5 g... -5 .... a 18.9 1.0 -5 .... a 16.0 1.4 -5 .... a 17.7 2.9 -5 .... a 18.7 1.2 -5 g... -5 .... a 18.9 2.0 -5 .... a 18.7 3.9 -5 .... c 18.9 1.1 < < < < < < < < < < < < < < < < < < < < < < < < < < < < .5 6.9 0.90 2.1 .4 6.0 0.90 1.5 -3.3 g... .6 1.7 0.90 1.3 -3.4 g... .9 3.4 0.89 2.0 .4.2.4 0.7 6.1 0.91 0.0 0.89 0.89 1.1 1.5 1.1 -3.0 -3.6 -3.0 ...... a a a 18.2 17.3 17.4 1.2 3.7 1.1 .2 5.2 0.90 2.1 .3.3.8 3.2 0.0 0.90 2.4 0.90 0.90 1.7 1.4 1.6 -4.1 -4.8 -4.7 ...... a a a 17.3 18.3 17.6 3.1 1.3 1.6 .4 3.7 0.90 1.4 -4.2 .... a 17.9 2.2 .4 2.1 0.89 1.7 -4.1 .... a 16.1 0.7 .1 1.9 0.90 1.6 .5 0.0 0.89 1.5 .2 0.2 0.89 1.5 -3.5 g... .3 0.8 0.89 1.8 .1 2.9 0.89 1.0 -3.3 .... .0 1.4 0.90 1.6 -4.2 .... a 17.4 1.8 .1 0.0 0.89 1.8 .1 0.0 0.89 1.4 .6.1 0.6 0.9 0.91 0.90 1.3 1.6 -3.5 .... a 17.8 1.6 .1 2.9 0.89 1.4 .7 1.4 0.91 1.7 -4.9 .... a 15.9 1.5 1.8 8.9 0.90 2.2 1.8 30.8 0.90 5.7 3.1 8.8 0.89 2.5 3.74.9 0.0 8.5 0.89 0.90 1.8 2.3 -3.3 -4.9 ...... a a 17.0 31.9 1.0 3.2 0.8 22.4 0.90 4.5 4.8 4.8 0.90 2.3 -5.0 g... 1.0 14.4 0.90 4.9 0.5 7.7 0.89 3.4 1.9 2.7 0.90 2.6 0.3 10.8 0.89 2.7 0.7 6.6 0.91 4.4 1.1 7.2 0.90 3.3 0.4 63.7 0.89 7.3 0.0 3.0 0.89 3.9 0.6 30.5 0.90 5.7 0.1 11.9 0.89 2.6 39 3.6 4.5 6.8 7.3 0.91 0.89 2.1 2.6 -4.6 .... a 17.0 4.2 4 3.3 7.5 0.91 2.3 7 4.1 3.1 0.91 2.5 .7 0.1 47.9 0.89 18.2 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Table 5.1—Continued ) Full Full Full Hard Frac ′ θ ) ( ′′ Err J2000 δ Position J2000 (deg) (deg) ( α Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 158 063348.25+043302.8 98.451058 4.550801 1.0 7.9 9.5 3.8 1 160 063349.17+042306.4 98.454878 4.385124 1.0 8.4 10.2 4.0 157 063347.32+043756.1 98.447203 4.632252 0.5 8.2 45.2 7.4 156 063346.46+042934.9 98.443620 4.493036 1.0 8.2 12.9 4.5 153154 063342.51+042247.3155 063344.49+042905.3 063345.42+043751.1 98.427126 98.435406 4.379820 98.439265 4.484814 1.4 4.630869 1.1 10.0 1.1 8.6 7.7 8.3 12.1 5.6 4.0 4.7 3.2 1 152 063341.66+043427.8 98.423606 4.574405 0.4 6.1 29.2 6.0 151 063341.24+043640.0 98.421856 4.611134 0.9 6.3 4.4 2.8 0 150 063341.19+042646.8 98.421640 4.446361 1.4 9.4 12.2 4.7 145146 063340.06+042333.7 063340.49+042259.9 98.416937 98.418726 4.392704 4.383332 1.4 1.2 10.2 10.3 10.4 14.5 4. 4. 148 063341.15+043231.9 98.421498 4.542207 0.7 6.4 8.1 3.5 0 149 063341.16+043705.1 98.421532 4.618099 0.4 6.4 34.0 6.4 144 063339.48+043439.7 98.414526 4.577695 0.4 5.6 18.5 4.9 140141 063337.58+043446.2142 063337.88+043934.5143 063338.64+043408.9 063339.45+042818.8 98.406602 98.407844 4.579521 98.411025 4.659601 98.414394 0.7 4.569161 1.0 4.471889 0.8 5.1 0.9 7.1 5.4 8.5 3.6 5.8 3.6 13.1 2.5 3.2 2.5 4.4 0 1 0 134 063332.48+042510.2 98.385361 4.419527 1.3 10.3 11.7 4. 135 063333.75+044010.3 98.390658 4.669535 0.8 6.9 8.8 3.7 1 136137 063333.93+044029.4138 063335.59+043515.9139 063336.82+043029.0 063337.03+043407.6 98.391404 98.398292 4.674845 98.403447 4.587760 98.404322 1.0 4.508077 0.6 4.568788 0.8 7.1 0.4 4.6 6.5 5.0 6.7 4.7 6.2 12.7 3.4 2.8 3.2 4.1 1 0 0 132133 063332.15+043101.6 063332.45+043922.5 98.383964 98.385227 4.517113 4.656262 0.7 0.4 5.2 6.0 4.6 28.3 2.8 5.9 0 130131 063331.40+044056.3 063332.05+042832.3 98.380868 98.383579 4.682319 4.475650 1.0 0.6 7.2 7.2 6.6 17.9 3.4 4.9 1 129 063330.96+043240.5 98.379039 4.544605 0.5 4.0 5.9 3.0 0 127 063330.40+043016.5 98.376701 4.504601 0.7 5.5 5.5 3.0 0 128 063330.59+044139.7 98.377497 4.694385 1.3 7.7 5.8 3.2 1 122 063328.59+043052.6 98.369136 4.514626 0.5 4.8 6.7 3.2 0 126 063330.28+043830.3 98.376202 4.641772 0.2 5.0 67.6 8.8 123 063329.13+042448.2 98.371394 4.413396 1.3 10.3 12.9 4. 124 063329.47+043331.8 98.372815 4.558850 0.5 3.3 2.9 2.3 0 121 063328.06+043300.2 98.366943 4.550078 0.2 3.2 23.0 5.3 118119 063326.21+044044.2 063327.50+043556.9 98.359227 98.364619 4.678968 4.599165 0.9 0.0 6.4 2.9 368.9 6.0 19 3.2 1 112 063323.93+043643.6 98.349734 4.612120 0.3 2.6 6.9 3.2 0 120 063327.94+042912.7 98.366417 4.486874 0.3 6.1 44.4 7.2 113 063324.23+043252.2 98.350987 4.547849 0.3 2.5 4.9 2.8 0 114115 063324.49+043950.8 063324.94+043436.5 98.352067 98.353949 4.664119 4.576824 0.8 0.3 5.5 1.9 4.4 5.9 2.8 3.0 0 0 109 063323.46+043650.5 98.347762 4.614028 0.3 2.7 4.9 2.8 0 110111 063323.46+042834.9 063323.78+043154.7 98.347779 98.349118 4.476370 4.531887 0.8 0.2 6.3 3.2 6.3 11.9 3.2 4.0 0 Seq CXOU J 148 median E (ks) (keV) 13.7 1.8 19.0 5.4 17.1 3.0 17.0 2.3 17.1 1.5 11.9 1.0 16.9 3.6 17.0 1.5 (16) (17) 28.9 1.5 ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 .... a 17.7 1.7 -5 g... -5 g... -5 .... a 19.0 1.4 -5 .... a 19.1 1.5 -5 g... -5 .... a 18.6 1.4 -5 .... a 18.7 4.5 -5 .... a 18.5 2.1 -5 .... a 18.9 1.5 -5 .... a 17.8 1.9 -5 .... a 17.5 1.2 -5 g... -5 .... a 16.6 1.4 -5 g... -5 .... a 17.2 1.6 -5 .... a 17.7 1.4 -5 .... a 17.9 1.5 -5 .... a 18.0 1.1 -5 g... -5 .... a 17.9 2.1 -5 .... a 18.7 1.2 -5 .... a 15.3 1.4 -5 .... a 17.8 1.0 -5 .... a 17.5 1.8 -5 .... a 17.8 1.4 -5 .... a 18.0 1.2 -5 .... b 17.3 3.5 -5 .... a 18.0 1.8 -5 .... a 18.0 1.7 -5 .... a 16.9 1.3 -5 .... a 18.2 1.2 -5 .... a 16.1 1.4 -5 .... a 17.6 1.0 -5 .... a 16.2 3.0 -5 g... < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < .2 2.5 0.90 1.9 -3.9 g... .2 2.9 0.86 2.0 .1 2.9 0.89 1.0 -3.7 .... .2 0.0 0.90 1.2 -4.5 .... a 18.7 0.9 .1 1.9 0.89 2.2 .1 3.9 0.89 1.2 -4.8 .... a 18.9 5.2 .7 4.5 0.90 1.7 .4 0.0 0.90 2.0 .2 0.9 0.90 1.2 -4.4 .... a 18.8 1.9 .2 8.8 0.89 2.1 .3 0.8 0.91 1.2 -3.8 .... a 18.6 1.5 .3.2 0.0 0.0 0.90 0.89 1.2 1.6 -3.4 .... a 18.5 1.2 .3 1.8 0.89 1.2 -3.7 .... a 18.5 2.5 .3 1.8 0.89 1.6 .5 7.0 0.89 1.8 -4.5 .... a 16.5 3.6 .3 0.8 0.89 1.4 -4.6 .... a 17.5 1.7 .6 0.0 0.89 1.6 -3.6 .... a 17.4 1.1 .8 1.5 0.90 1.8 0.2 39.9 0.86 6.5 0.3 28.8 0.89 6.1 0.1 5.9 0.89 4.2 0.1 0.9 0.90 2.5 0.2 5.8 0.90 3.9 0.2 10.9 0.90 2.4 0.2 4.9 0.89 2.6 0.1 9.9 0.89 3.6 2.7 1.9 0.89 5.3 0.7 3.6 0.89 3.4 1.8 18.9 0.90 6.8 0.6 2.6 0.90 3.3 0.3 3.8 0.90 4.4 1.6 18.9 0.90 4.9 0.4 4.8 0.91 3.1 1.6 0.9 0.90 2.3 0.4 2.7 0.91 2.8 0.8 1.5 0.90 2.3 1.7 8.9 0.90 2.4 0.8 7.5 0.90 3.6 0.9 13.4 0.90 5.4 0.6 0.7 0.90 3.0 1.9 6.8 0.89 3.7 2.8 9.3 0.89 3.0 4.6 5.3 0.89 4.3 3 5.8 4.1 0.90 4.0 4 1.4 39.1 0.89 8.8 .2 2.1 12.9 0.90 16.6 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Table 5.1—Continued ) Full Full Full Hard Frac ′ θ ) ( ′′ Err J2000 δ Position J2000 (deg) (deg) ( α Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 212 063406.15+041921.4 98.525649 4.322619 1.5 8.9 8.8 4.0 3 210 063406.12+042447.6 98.525531 4.413244 0.4 4.1 7.8 3.4 0 211 063406.13+042914.9 98.525569 4.487498 0.5 3.2 2.9 2.3 0 208209 063405.91+043027.2 063406.04+042451.4 98.524648 98.525174 4.507563 4.414302 0.5 0.1 3.9 4.0 3.8 54.8 2.5 7.9 0 207 063405.67+042453.7 98.523655 4.414939 0.2 4.1 48.7 7.5 206 063405.43+042915.5 98.522628 4.487643 0.2 3.3 25.9 5.6 204 063405.29+042654.9 98.522054 4.448607 0.3 3.1 8.9 3.5 0 205 063405.31+042808.7 98.522126 4.469094 0.2 3.0 10.9 3.8 202203 063404.60+042925.5 063405.27+043038.5 98.519176 98.521978 4.490431 4.510717 0.5 0.2 3.6 4.2 3.9 22.8 2.5 5.3 0 201 063404.55+043009.7 98.518967 4.502696 0.3 4.0 10.8 3.8 198 063404.05+042307.7 98.516913 4.385473 0.7 5.7 6.3 3.2 0 200 063404.32+042529.8 98.518005 4.424969 0.3 4.0 11.8 4.0 197 063403.83+042626.8 98.515972 4.440802 0.2 3.6 19.9 5.0 195 063403.41+043408.6 98.514248 4.569075 0.9 7.3 8.6 3.7 1 196 063403.47+041944.9 98.514498 4.329145 0.8 8.7 40.3 7.1 192194 063403.10+042932.3 063403.34+043327.6 98.512940 98.513952 4.492311 4.557669 0.5 0.5 4.0 6.7 3.8 18.3 2.5 4.9 0 191 063402.96+042512.2 98.512366 4.420056 0.4 4.4 8.8 3.5 0 189190 063402.29+043007.2 063402.83+043450.7 98.509550 98.511825 4.502022 4.580761 0.6 0.4 4.4 7.9 3.7 61.2 2.5 8.5 0 185186 063400.04+043011.2 063400.85+042614.9 98.500201 98.503566 4.503114 4.437498 0.7 0.5 4.9 4.4 3.7 5.8 2.5 3.0 0 0 187188 063401.15+043031.7 063401.58+043241.4 98.504817 98.506612 4.508811 4.544861 0.7 0.5 4.9 6.3 3.7 17.4 2.5 4.8 0 183 063358.95+042651.7 98.495630 4.447716 0.5 4.7 5.7 3.0 0 184 063359.90+042828.5 98.499608 4.474599 0.2 4.4 27.7 5.8 181 063358.59+043807.4 98.494150 4.635415 1.0 10.9 27.2 6. 182 063358.68+042145.1 98.494510 4.362541 0.5 7.6 34.4 6.5 178179 063357.59+043401.6 063358.04+043331.1 98.489980 98.491846 4.567117 4.558655 1.1 0.3 8.0 7.5 7.5 96.6 10. 3.5 1 180 063358.39+042732.5 98.493308 4.459053 0.3 4.7 15.6 4.5 176177 063357.33+042642.1 063357.50+042159.2 98.488886 98.489590 4.445049 4.366449 0.6 0.9 5.1 7.6 4.7 10.4 2.8 4.0 0 174175 063356.84+043346.5 063357.11+042656.7 98.486834 98.487965 4.562941 4.449095 1.1 0.4 7.9 5.1 6.4 13.6 3.4 4.3 1 172 063356.12+043119.0 98.483845 4.521968 0.7 6.4 10.2 3.8 171 063355.99+043350.9 98.483312 4.564165 0.9 8.1 11.3 4.1 170 063355.82+043101.4 98.482612 4.517078 0.5 6.3 20.2 5.1 169 063354.87+043101.8 98.478661 4.517194 0.4 6.5 40.1 6.9 164 063350.97+043131.5 98.462391 4.525431 0.2 7.6 309.9 18 167 063352.46+042913.1 98.468619 4.486988 0.8 6.4 7.2 3.4 0 168 063353.78+042833.1 98.474108 4.475870 0.5 5.9 15.4 4.5 163 063350.42+043658.4 98.460099 4.616227 0.7 8.6 22.1 5.4 162 063350.30+043840.0 98.459612 4.644455 0.9 9.2 16.2 4.9 161 063349.65+043205.3 98.456910 4.534825 0.7 8.3 29.4 6.4 Seq CXOU J 149 median E 8.0 1.4 (ks) (keV) 17.5 1.2 16.9 1.5 17.4 1.4 11.2 1.9 17.1 2.0 17.1 1.5 16.2 2.1 18.4 4.2 17.2 1.7 18.2 1.9 11.9 1.9 18.9 2.3 (16) (17) 12.7 1.0 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 .... a 19.2 1.3 -5 g... -5 g... -5 g... -5 g... -5 .... a 17.9 3.6 -5 .... a 17.9 3.1 -5 .... a 19.1 1.0 -5 .... a 19.2 1.0 -5 .... a 19.4 2.6 -5 .... a 19.3 1.9 -5 .... a 19.2 1.4 -5 .... a 19.5 1.8 -5 .... a 18.2 1.3 -5 g... -5 .... a 17.5 1.5 -5 .... a 19.4 1.3 -5 g... -5 .... a 19.4 1.4 -5 .... a 18.7 2.6 -5 .... a 19.0 1.7 -5 .... a 19.0 1.2 -5 .... a 18.1 1.0 -5 .... b 19.1 1.8 -5 .... a 16.9 1.5 -5 .... a 17.7 1.4 -5 .... b 18.3 1.9 -5 .... a 18.5 1.3 -5 g... -5 g... -5 .... a 17.2 1.6 -5 .... a 17.9 1.7 -5 .... a 19.0 5.9 -5 .... a 18.7 1.0 -5 g... < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < .9 7.5 0.91 1.9 -4.5 .... a 17.0 3.5 .1 2.0 0.90 1.2 .1 0.0 0.90 1.5 .1 3.0 0.89 2.2 .1 2.9 0.90 2.2 .5 3.6 0.90 1.3 -3.7 .... a 18.4 4.1 .1 5.9 0.90 1.6 .1 0.0 0.90 1.2 .0 1.0 0.89 0.7 -3.1 g... .4.1 3.7 0.0 0.90 0.89 1.4 1.7 -4.4 .... a 17.3 3.5 .1 3.9 0.89 1.8 .1 2.0 0.89 1.0 -4.0 g... .1 0.9 0.89 1.4 .1 3.9 0.89 2.2 .1 1.9 0.89 1.4 .1 0.9 0.89 1.2 .1 1.9 0.89 2.2 .1 2.9 0.89 1.6 .3 3.0 0.90 1.7 -4.1 .... a 17.4 2.0 .1.1 2.9 0.9 0.89 0.89 1.2 1.4 -4.9 .... a 18.5 5.8 .1 2.9 0.90 1.0 -4.0 .... .1 4.9 0.90 2.2 .8.1 7.6 2.9 0.90 0.90 1.9 1.8 -4.7 .... a 17.5 5.6 .1 0.9 0.90 1.0 -3.4 g... .1 1.9 0.90 1.6 .3 3.8 0.90 2.1 .2 0.0 0.89 1.4 .1 3.9 0.89 2.2 .1 0.9 0.89 1.2 .1 1.9 0.89 1.0 -3.7 .... .1 2.9 0.90 1.4 0.1 1.9 0.90 2.5 0.5 12.6 0.91 2.8 0.1 12.9 0.89 4.5 0.3 7.7 0.90 2.8 0.1 30.9 0.90 8.2 4.2 2.3 0.90 2.2 -4.5 .... a 15.6 1.6 0.1 2.9 0.89 2.5 0.1 7.9 0.89 3.5 3.0 11.0 0.90 5.3 0.4 4.7 0.90 2.4 0.1 9.9 0.90 4.4 0.2 0.0 0.90 2.7 0.1 0.0 0.90 2.9 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Table 5.1—Continued ) Full Full Full Hard Frac ′ θ ) ( ′′ Err J2000 δ Position J2000 (deg) (deg) ( α Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 265 063413.41+042013.4 98.555876 4.337066 1.0 7.6 8.1 3.7 1 264 063413.23+042825.2 98.555163 4.473685 0.3 1.2 3.9 2.5 0 263 063413.18+042727.1 98.554955 4.457542 0.2 1.1 4.9 2.8 0 261 063412.52+042641.1 98.552169 4.444754 0.2 1.6 8.9 3.5 0 260 063412.30+042720.3 98.551278 4.455642 0.2 1.3 8.9 3.5 0 258259 063412.26+042228.5 063412.28+042408.8 98.551121 98.551192 4.374610 4.402468 0.8 0.3 5.4 3.8 4.5 10.9 2.8 3.8 0 257 063412.12+042241.9 98.550503 4.378327 0.4 5.2 13.5 4.3 256 063411.99+042511.0 98.549972 4.419730 0.3 2.9 5.9 3.0 0 255 063411.96+042942.8 98.549869 4.495231 0.3 2.4 3.9 2.5 0 254 063411.50+042802.7 98.547958 4.467417 0.4 1.5 2.0 2.0 0 251253 063411.30+042255.5 063411.44+042828.9 98.547109 98.547673 4.382095 4.474698 0.7 0.2 5.1 1.6 4.6 5.9 2.8 3.0 0 0 250 063411.25+042635.8 98.546904 4.443281 0.2 1.9 6.9 3.2 0 249 063410.92+042603.5 98.545504 4.434306 0.1 2.3 28.9 5.9 248 063410.83+042652.1 98.545133 4.447810 0.3 1.9 2.9 2.3 0 247 063410.81+042911.8 98.545050 4.486636 0.3 2.2 4.9 2.8 0 246 063410.75+042641.5 98.544828 4.444885 0.2 2.0 8.9 3.5 0 244 063410.51+042526.9 98.543818 4.424162 0.3 2.9 4.9 2.8 0 243 063410.39+042717.3 98.543328 4.454824 0.3 1.8 3.9 2.5 0 242 063410.34+042709.7 98.543116 4.452699 0.2 1.8 8.9 3.5 0 241 063409.83+042637.7 98.540991 4.443816 0.3 2.2 5.9 3.0 0 239240 063409.34+043447.5 063409.34+042310.0 98.538919 98.538938 4.579868 4.386121 1.0 0.4 7.3 5.0 6.7 13.7 3.4 4.3 1 237238 063409.13+042544.5 063409.29+042655.8 98.538080 98.538725 4.429045 4.448860 0.4 0.3 2.9 2.2 3.9 4.9 2.5 2.8 0 0 236 063409.00+042744.6 98.537535 4.462415 0.4 2.1 2.9 2.3 0 235 063409.00+043050.8 98.537535 4.514121 0.3 3.7 8.9 3.5 0 234 063408.99+042944.5 98.537478 4.495707 0.1 2.9 83.9 9.7 231232 063408.86+043712.7233 063408.87+042001.9 063408.90+042938.4 98.536955 98.536992 4.620219 98.537084 4.333888 1.3 4.494024 1.1 9.7 0.3 8.0 2.8 10.8 8.2 6.9 4.4 3.7 3.2 1 0 230 063408.64+042532.2 98.536032 4.425637 0.5 3.1 2.9 2.3 0 229 063408.23+042713.7 98.534333 4.453806 0.2 2.3 10.9 3.8 228 063408.12+042859.0 98.533841 4.483075 0.2 2.6 18.9 4.9 227 063408.00+043607.6 98.533371 4.602131 0.6 8.7 41.0 7.2 226 063407.92+042533.6 98.533021 4.426027 0.3 3.2 5.9 3.0 0 225 063407.78+043206.6 98.532445 4.535193 0.5 5.0 8.7 3.5 0 224 063407.76+043127.4 98.532366 4.524294 0.5 4.4 4.8 2.8 0 222 063407.54+042547.2 98.531450 4.429787 0.3 3.1 8.9 3.5 0 221 063407.50+042829.7 98.531255 4.474941 0.4 2.6 3.9 2.5 0 219 063407.33+043258.9 98.530543 4.549705 0.5 5.8 10.6 3.8 217 063407.15+042838.0 98.529802 4.477231 0.1 2.7 27.9 5.8 216 063406.99+042806.8 98.529135 4.468576 0.4 2.6 2.9 2.3 0 215 063406.84+042939.2 98.528536 4.494223 0.4 3.2 4.9 2.8 0 214 063406.24+043039.7 98.526024 4.511048 0.3 4.0 12.8 4.1 213 063406.16+042818.2 98.525683 4.471725 0.2 2.8 13.9 4.3 Seq CXOU J 150 median E 8.2 1.4 8.8 1.5 7.0 1.4 (ks) (keV) 10.1 1.9 19.5 1.0 14.2 1.4 10.4 1.4 18.7 1.8 16.3 1.4 15.0 1.6 18.9 2.7 18.8 1.0 18.9 1.3 19.1 1.3 18.1 1.6 19.4 1.8 12.4 3.6 15.4 1.8 18.8 5.1 (16) (17) ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 g... -5 g... -5 .... c 17.9 1.8 -5 g... -5 .... a 17.9 2.0 -5 .... a 16.1 2.4 -5 .... c 18.8 2.0 -5 .... a 18.6 2.0 -5 .... a 19.5 1.7 -5 .... a 19.5 1.5 -5 .... a 19.5 1.4 -5 .... a 18.0 2.2 -5 .... a 16.0 2.5 -5 .... a 19.2 2.2 -5 .... a 18.1 3.0 -5 .... a 19.1 1.7 -5 .... a 19.3 2.4 -5 .... b 19.0 3.5 -5 .... a 18.1 1.2 -5 .... a 18.2 2.8 -5 .... a 19.3 3.2 -5 .... b 17.9 2.1 -5 g... -5 .... a 19.4 1.5 -5 .... a 17.4 2.4 -5 .... a 18.8 3.8 -5 g... -5 g... -5 .... a 18.8 3.0 -5 .... a 19.3 1.6 -5 g... < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < .1.0 2.9 3.3 0.89 0.89 1.2 1.4 -4.9 -3.3 ...... a a 18.7 17.7 3.5 2.1 .1 1.0 0.90 1.0 -4.4 g... .1 0.0 0.90 1.0 -4.2 .... .1 0.9 0.89 1.0 -3.6 .... .1 1.0 0.90 1.0 -4.6 g... .1 4.0 0.90 2.3 .1 2.9 0.90 1.0 -4.0 .... .2 3.9 0.90 2.1 .1 3.0 0.90 1.8 .1 0.0 0.90 1.0 -3.4 .... .0 4.3 0.89 1.4 -3.2 .... a 17.4 4.3 .9 5.3 0.90 2.0 .1 0.9 0.89 1.4 .2 0.0 0.90 2.0 .1 0.0 0.89 1.0 -3.3 .... .0 4.0 0.85 1.7 .1 0.0 0.89 1.0 -4.4 .... .0 1.0 0.85 1.0 -4.7 .... .1 5.0 0.90 1.5 .1 6.1 0.90 2.1 .1 1.0 0.90 1.0 -4.4 .... .5 1.0 0.89 1.6 -3.7 .... a 17.2 1.4 .1 1.0 0.90 1.0 -4.6 g... .1 0.0 0.90 1.5 .4 3.8 0.91 1.8 .1 4.9 0.89 1.4 .0 2.0 0.90 1.8 .1 2.9 0.89 1.0 -4.0 .... .0 2.0 0.89 2.2 .0 1.0 0.90 1.8 2.1 2.7 0.89 2.5 0.1 5.0 0.89 3.1 0.1 5.9 0.89 2.7 1.1 8.2 0.90 3.0 1.9 8.6 0.90 3.0 0.1 26.9 0.89 6.1 0.1 4.0 0.90 3.0 0.1 1.0 0.90 2.9 1.2 9.2 0.90 3.3 2.1 6.4 0.90 2.3 0.1 7.9 0.90 2.7 0.1 11.0 0.90 3.5 0.1 15.9 0.89 3.4 2.3 19.4 0.90 3.8 0.1 21.9 0.89 4.4 0.1 5.0 0.90 2.7 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Table 5.1—Continued ) Full Full Full Hard Frac ′ θ ) ( ′′ Err J2000 δ Position J2000 (deg) (deg) ( α Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 315316 063423.78+042444.4 063423.97+042116.0 98.599108 98.599913 4.412359 4.354465 0.4 0.9 3.4 6.7 3.9 5.0 2.5 3.0 0 1 314 063423.52+043012.3 98.598001 4.503424 0.5 2.9 2.9 2.3 0 313 063423.49+042715.1 98.597902 4.454202 0.3 1.6 2.9 2.3 0 312 063423.46+043542.7 98.597785 4.595208 1.0 8.1 11.9 4.3 311 063422.88+042950.7 98.595349 4.497423 0.2 2.5 15.9 4.5 310 063422.72+042444.0 98.594694 4.412226 0.5 3.3 2.9 2.3 0 309 063422.57+042759.9 98.594063 4.466647 0.3 1.3 2.9 2.3 0 307 063422.05+042428.9 98.591895 4.408044 0.2 3.5 12.9 4.1 306 063421.99+042624.3 98.591644 4.440110 0.2 1.8 9.9 3.7 0 304 063421.57+042127.6 98.589891 4.357684 0.5 6.4 15.9 4.6 303 063421.34+042515.0 98.588958 4.420841 0.4 2.7 2.9 2.3 0 302 063421.26+041933.0 98.588611 4.325840 0.8 8.3 16.1 4.8 301 063421.03+042441.6 98.587654 4.411576 0.1 3.2 49.9 7.6 300 063420.43+042404.7 98.585139 4.401322 0.3 3.8 8.8 3.5 0 299 063420.14+042734.5 98.583937 4.459588 0.2 0.7 6.9 3.2 0 297 063419.96+042744.0 98.583172 4.462235 0.1 0.7 14.9 4.4 296 063419.56+042739.2 98.581534 4.460896 0.1 0.6 13.9 4.3 295 063419.46+042131.8 98.581093 4.358838 0.5 6.3 17.8 4.9 294 063418.67+041920.6 98.577815 4.322396 1.0 8.4 10.9 4.1 293 063417.97+042419.5 98.574890 4.405431 0.5 3.4 2.9 2.3 0 291292 063417.90+043438.0 063417.94+042611.6 98.574611 98.574757 4.577232 4.436582 1.0 0.2 6.9 1.6 5.0 12.9 3.0 4.1 1 290 063417.50+042153.9 98.572927 4.364975 0.6 5.9 8.1 3.5 0 289 063417.37+042509.5 98.572381 4.419318 0.3 2.6 4.9 2.8 0 288 063417.31+042811.7 98.572151 4.469934 0.1 0.4 18.9 4.9 287 063417.24+042502.3 98.571850 4.417330 0.2 2.7 17.9 4.8 286 063417.11+043146.6 98.571312 4.529621 0.4 4.0 7.8 3.4 0 285 063416.92+042419.1 98.570535 4.405306 0.5 3.4 2.9 2.3 0 284 063416.63+042938.9 98.569320 4.494147 0.2 1.9 6.0 3.0 0 283 063416.49+042928.6 98.568745 4.491296 0.3 1.7 2.9 2.3 0 282 063416.48+042938.9 98.568704 4.494157 0.3 1.9 3.0 2.3 0 281 063416.29+042816.9 98.567912 4.471368 0.2 0.6 4.9 2.8 0 280 063416.20+042127.1 98.567525 4.357537 0.7 6.3 8.9 3.7 1 279 063416.14+042811.2 98.567261 4.469795 0.3 0.5 2.9 2.3 0 277278 063415.98+043525.5 063416.11+041850.8 98.566614 98.567144 4.590418 4.314133 1.1 0.9 7.7 8.9 6.5 23.7 3.4 5.6 1 276 063415.93+042710.2 98.566397 4.452850 0.3 0.7 2.9 2.3 0 275 063415.77+042812.4 98.565723 4.470114 0.2 0.6 4.9 2.8 0 273 063415.43+042251.4 98.564320 4.380962 0.5 4.9 6.6 3.2 0 272 063415.27+043033.0 98.563634 4.509181 0.3 2.8 4.9 2.8 0 271 063415.07+042725.1 98.562804 4.456998 0.2 0.7 7.0 3.2 0 270 063414.85+043022.5 98.561897 4.506275 0.4 2.7 2.9 2.3 0 269 063414.79+042538.2 98.561652 4.427280 0.2 2.2 9.0 3.5 0 268 063414.67+043022.5 98.561127 4.506275 0.1 2.7 27.9 5.8 267 063413.60+042718.5 98.556704 4.455141 0.2 1.0 12.9 4.1 266 063413.57+042742.6 98.556544 4.461841 0.2 0.9 7.0 3.2 0 Seq CXOU J 151 median E (ks) (keV) 16.0 0.8 18.9 1.8 18.5 1.0 12.6 1.0 16.4 1.9 13.0 0.9 17.3 0.8 16.4 3.4 17.3 1.5 (16) (17) 18.9 3.2 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P -5 .... c 17.2 2.8 -5 .... a 16.6 1.5 -5 .... a 17.8 3.2 -5 .... b 18.1 1.5 -5 .... a 17.0 2.6 -5 .... a 16.8 1.2 -5 .... a 17.7 5.6 -5 .... a 16.9 1.9 -5 .... b 17.8 3.0 -5 .... a 17.0 2.2 -5 .... a 18.7 5.5 -5 .... a 17.1 2.5 -5 .... b 16.8 1.7 -5 .... a 18.1 3.0 -5 .... a 18.1 1.0 -5 .... a 17.8 1.2 -5 g... -5 .... a 16.3 3.4 -5 .... a 18.5 1.4 -5 .... a 15.4 1.3 -5 .... a 18.6 1.7 -5 .... a 18.9 1.4 -5 .... a 17.9 3.7 -5 .... c 17.4 3.3 -5 .... a 17.6 0.9 -5 .... a 19.3 1.5 -5 g... -5 .... a 19.4 1.0 -5 .... c 15.8 3.5 < < < < < < < < < < < < < < < < < < < < < < < < < < < < < .6 1.6 0.90 1.3 -3.5 .... a 18.3 1.9 .5 1.7 0.91 2.2 .2 0.0 0.89 1.2 -4.1 .... a 18.6 1.3 .3 0.0 0.91 2.1 .9 5.4 0.90 2.0 .2 0.0 0.90 1.0 -3.2 g... .7 5.5 0.90 1.8 .2 6.9 0.90 1.8 .2 3.9 0.89 1.2 -4.2 .... a 17.5 3.3 .2 0.9 0.89 1.0 -3.3 .... .1 0.0 0.89 1.0 -3.5 .... .2 0.0 0.90 1.2 -4.5 g... .5 3.0 0.90 1.6 -3.7 .... b 17.1 2.6 .3.1 2.8 0.9 0.90 0.89 1.4 1.2 -4.8 -4.8 ...... a a 18.2 18.6 3.1 1.4 .2 0.0 0.89 1.8 .1 0.0 0.90 1.0 -3.3 g... .2 2.9 0.89 1.6 .1 0.0 0.89 1.2 .1 0.0 0.90 1.0 -3.5 .... .5 5.7 0.89 1.7 .1 0.0 0.90 1.2 .1 3.0 0.90 1.0 -4.3 g... .7 8.2 0.89 2.0 -4.5 .... b 15.7 2.4 .3 2.8 0.89 1.2 -3.5 .... a 18.2 3.0 .1 1.9 0.90 1.0 -3.6 .... 1.8 8.2 0.89 2.9 5.2 8.0 0.87 4.2 0.9 8.4 0.90 2.3 0.2 7.8 0.89 3.0 2.0 12.8 0.76 4.4 1.8 6.0 0.75 2.4 0.6 9.6 0.91 3.1 5.6 10.8 0.89 3.6 0.3 11.8 0.91 2.7 0.4 0.7 0.91 3.2 0.1 1.9 0.90 3.4 0.2 5.9 0.90 2.9 0.8 18.5 0.91 3.5 2.3 4.6 0.89 4.1 1.6 38.9 0.89 5.9 0.1 0.0 0.90 2.9 0.2 4.9 0.89 2.6 0.1 1.9 0.90 3.9 2.2 13.6 0.90 3.1 5 4.5 3.4 0.87 2.2 -4.7 .... a 16.5 1.6 4 4.9 6.7 0.91 2.0 -3.7 .... a 15.4 3.6 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Table 5.1—Continued ) Full Full Full Hard Frac ′ θ ) ( ′′ Err J2000 δ Position J2000 (deg) (deg) ( α Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 374 063437.08+042535.8 98.654535 4.426638 0.8 5.4 4.4 2.8 0 371373 063435.21+041847.4 063436.55+042138.8 98.646747 98.652330 4.313184 4.360789 1.2 0.7 10.0 7.8 11.5 15.2 4. 4.6 370 063434.49+041900.8 98.643750 4.316889 0.8 9.7 28.8 6.4 369 063434.41+042323.9 98.643376 4.389987 0.6 6.1 10.1 3.8 368 063434.39+043059.8 98.643311 4.516633 0.5 5.3 9.5 3.7 0 366367 063433.60+042805.8 063434.27+042916.3 98.640031 98.642806 4.468301 4.487878 0.5 0.3 4.1 4.5 3.8 14.8 2.5 4.4 0 364 063433.47+042543.0 98.639470 4.428616 0.4 4.5 8.7 3.5 0 362 063433.33+042242.0 98.638887 4.378357 0.7 6.4 8.1 3.5 0 360 063433.09+042649.0 98.637883 4.446950 0.6 4.0 2.8 2.3 0 359 063432.85+041944.4 98.636903 4.329010 0.6 8.9 29.0 6.1 357358 063432.17+041750.9 063432.69+042322.3 98.634079 98.636213 4.297474 4.389555 1.4 0.7 10.6 5.8 10.1 7.3 4. 3.4 0 356 063432.10+041946.2 98.633765 4.329509 0.8 8.8 11.2 4.1 355 063431.81+042832.4 98.632557 4.475669 0.4 3.7 6.8 3.2 0 353354 063431.58+042938.1 063431.59+042316.1 98.631600 98.631648 4.493940 4.387809 0.5 0.4 4.0 5.7 3.8 16.4 2.5 4.6 0 352 063431.47+041912.4 98.631153 4.320123 0.8 9.3 23.4 5.9 351 063431.45+042427.4 98.631081 4.407630 0.4 4.8 12.7 4.1 347 063430.13+042342.9 98.625562 4.395275 0.4 5.2 16.6 4.6 346 063429.47+042724.1 98.622793 4.456722 0.2 3.0 17.9 4.8 345 063429.39+042623.2 98.622459 4.439803 0.5 3.3 2.8 2.3 0 343 063428.91+042810.9 98.620463 4.469715 0.4 2.9 2.9 2.3 0 342 063428.83+043103.5 98.620148 4.517643 0.6 4.4 3.8 2.5 0 338340 063427.70+042055.3 063428.22+042953.9 98.615457 98.617610 4.348721 4.498320 0.9 0.2 7.3 3.5 6.5 13.8 3.4 4.3 1 334336 063427.28+042327.5337 063427.48+042517.2 063427.61+043321.0 98.613704 98.614528 4.390983 98.615069 4.421470 0.6 4.555842 0.5 0.5 5.0 3.5 6.1 4.7 3.9 19.2 2.8 2.5 5.0 0 0 333 063427.26+042435.2 98.613624 4.409790 0.4 4.0 6.8 3.2 0 332 063427.22+043515.5 98.613453 4.587665 0.7 7.9 26.7 5.9 331 063426.83+043114.8 98.611821 4.520804 0.7 4.2 2.9 2.3 0 330 063426.62+042507.7 98.610934 4.418814 0.4 3.5 5.8 3.0 0 329 063426.29+042939.1 98.609546 4.494208 0.4 2.9 3.9 2.5 0 328 063426.08+043021.5 98.608680 4.505995 0.5 3.4 2.9 2.3 0 327 063425.90+043241.7 98.607948 4.544935 0.6 5.4 6.5 3.2 0 326 063425.85+042031.6 98.607714 4.342138 0.4 7.5 47.4 7.5 325 063425.83+043018.6 98.607636 4.505189 0.2 3.3 13.9 4.3 324 063425.73+042815.3 98.607227 4.470927 0.3 2.1 3.9 2.5 0 323 063425.46+042731.3 98.606100 4.458710 0.4 2.0 2.9 2.3 0 322 063425.32+042406.1 98.605532 4.401700 0.3 4.2 11.8 4.0 320321 063424.85+041832.8 063425.12+042653.1 98.603565 98.604698 4.309114 4.448103 1.2 0.1 9.4 2.1 9.3 22.9 4.0 5.3 2 318319 063424.51+043155.4 063424.64+041909.3 98.602126 98.602670 4.532082 4.319262 0.7 0.9 4.5 8.8 3.7 16.8 2.5 4.9 0 317 063424.42+043011.1 98.601752 4.503105 0.5 3.0 2.9 2.3 0 Seq CXOU J 152 l ector Column median E Column 13: Column 10: Estimated net . e local background 6.9 1.3 9.6 3.4 (ks) (keV) 12.9 3.0 (16) (17) ··· ··· ··· Columns 7,8: net counts Characteristics tion region. Note that a reduced PSF Anom Var EffExp upper error on net counts B P -5 .... a 17.1 2.7 -5 .... a 15.9 2.4 -5 g... -5 .... a 16.3 2.3 -5 .... a 17.1 1.7 -5 g... -5 .... a 17.3 1.8 -5 .... a 17.2 1.9 -5 .... b 17.1 1.9 -5 .... a 17.2 1.5 -5 .... a 16.9 1.4 Off-axis angle. < < < < < < < < < < < ave to be observed on axis to obtain the reported 005). No value is reported for sources with fewer . 0 < Variability characterization based on K-S statistic (tota KS Background counts extracted (total band). P Column 6: Right ascension and declination for epoch J2000.0. . Source anomalies: g = fractional time that source was on a det Column 15: Column 9: Columns 3,4: Photometric significance computed as values above the 1% threshold that defines the catalog becaus out streak. B Column 14: P .4.7 2.3 7.1 0.91 0.91 1.6 1.9 -3.0 -3.9 ...... a a 16.7 16.6 1.2 2.8 .8 0.7 0.90 1.9 -4.6 .... a 15.8 1.1 .0 9.3 0.90 2.1 .5 1.9 0.89 1.6 -3.7 .... a 15.2 1.7 .8.7 5.4 0.9 0.90 0.90 1.6 2.1 -4.8 .... a 18.0 3.8 .4 0.0 0.89 1.9 .4.0 0.8 3.6 0.89 0.89 1.1 1.7 -3.0 -3.6 ...... a a 18.1 16.8 0.9 3.7 .2 2.2 0.89 1.9 .5 2.7 0.91 1.3 -3.9 g... 1.9 9.8 0.90 2.2 1.8 10.9 0.89 3.3 1.6 6.9 0.90 2.4 1.4 5.1 0.89 3.2 2.0 8.9 0.90 3.5 1.3 9.2 0.90 4.8 0.9 5.3 0.90 4.1 05); c = definitely variable ( . 0 < errors on column 7. Extracted Counts Column 12: KS σ IAU designation. # of counts extracted √ < P Fraction of the PSF (at 1.497 keV) enclosed within the extrac 005 Net ∆Net Bkgd Net PSF Signif log . Table 5.1—Continued Column 2: Effective exposure time: approximate time the source would h standard deviation of PSF inside extraction region ) Full Full Full Hard Frac ′ θ Column 11: ) ( eV). f the upper and lower 1 ′′ Err rce is in a crowded region. Column 16: ly from background. Some sources have er sources are removed from the catalog. , computed as σ ); b = possibly variable (0 J2000 δ Position KS < P 05 . J2000 (deg) (deg) ( α 9 ; e = source on field edge; p = source piled up; s = source on read . Background-corrected median photon energy (total band). 0 < X-ray catalog sequence number, sorted by RA. ) is mkarf Column 17: Column 1: Source Estimated random component of position error, 1 # Note. — (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 394395 063451.49+043110.4 063452.40+043126.3 98.714548 98.718338 4.519564 4.523984 1.4 1.4 9.2 9.5 6.6 8.3 3.5 3.9 2 2 393 063450.44+042702.2 98.710169 4.450623 1.0 8.3 10.1 4.0 391392 063449.51+043105.9 063449.98+042739.4 98.706293 98.708263 4.518307 4.460959 1.2 0.7 8.7 8.1 8.2 18.2 3.7 5.0 1 390 063447.41+042857.8 98.697569 4.482747 0.9 7.6 9.0 3.7 1 388389 063445.58+042423.2 063446.45+043101.8 98.689926 98.693547 4.406461 4.517182 1.2 0.9 7.8 8.0 6.5 11.4 3.4 4.1 1 385387 063439.92+042527.0 063443.25+042351.3 98.666346 98.680226 4.424190 4.397601 0.8 0.9 6.1 7.6 6.2 9.3 3.2 3.8 0 1 384 063439.88+042444.6 98.666196 4.412415 0.8 6.4 7.6 3.4 0 380382 063439.10+042807.3383 063439.35+043348.0 063439.46+042317.9 98.662931 98.663972 4.468712 98.664434 4.563343 0.8 4.388309 1.2 5.4 0.6 8.2 7.1 3.6 7.0 17.6 2.5 3.5 4.9 0 2 379 063438.76+042242.1 98.661512 4.378379 0.9 7.4 7.8 3.5 1 378 063438.55+042501.8 98.660630 4.417187 0.9 6.0 4.5 2.8 0 377 063438.43+042146.2 98.660146 4.362847 0.7 8.0 20.0 5.2 376 063438.09+042209.9 98.658736 4.369439 0.5 7.6 33.7 6.4 375 063437.67+043147.7 98.656994 4.529935 0.4 6.5 25.1 5.6 Seq CXOU J than 4 counts or for sources in chip gaps or on field edges. band): a = no evidence for variability (0 Log probability that extracted counts (total band) are sole number of counts. (FRACEXPO from 5: estimates can rise during the final extraction iteration aft counts extracted in total energy band (0.5–8 keV); average o Estimated net counts extracted in the hard energy band (2–8 k fraction (significantly below 90%) may indicate that the sou 153 median E (ks) (keV) 15.6 1.4 18.5 1.8 17.9 1.5 (16) (17) 15.2 7.6 18.6 1.1 18.8 1.2 17.3 2.5 18.3 2.5 19.2 6.1 18.4 0.7 15.6 3.1 13.0 0.7 17.8 1.1 35.6 3.4 12.0 4.6 17.6 5.6 23.3 0.9 16.0 1.0 16.9 4.0 18.2 1.1 11.8 3.9 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom Var EffExp B P 0.6 0.90 1.5 -2.8 g... 514 4.0 3.7 0.90 3.1 0.90 0.90 1.2 1.2 1.3 -2.3 -1.8 -2.6 ...... a a 17.6 33.1 2.1 7.8 875 0.04 3.53 0.89 0.0 0.89 2.4 0.90 0.0 0.89 0.90 1.3 1.1 1.1 -2.1 1.2 -2.3 1.3 -2.7 -1.9 .... -2.7 ...... a g... a a a 34.6 15.9 17.6 33.9 1.3 5.2 1.3 1.6 163 5.9 4.5 0.89 1.8 0.90 0.89 1.6 1.7 0.9 -2.8 -2.9 -2.6 ...... g... a a 35.4 34.2 3.3 2.9 1 2.0 0.90 0.7 -2.9 g... 5 2.6 0.91 1.1 -2.7 .... a 18.3 5.8 .4.5.0 2.7 0.0 0.91 1.5 0.89 0.89 1.1 1.1 1.3 -3.0 -2.9 -2.4 ...... g... a a 17.0 18.1 2.3 0.8 .2.2 3.3 0.0 0.90 0.89 1.3 0.9 -2.9 -2.7 ...... b 17.6 3.0 .9.0.8 1.4.0 2.4.4 0.90 5.6.9 0.89 3.3.2 0.90 4.4 0.89 0.3 0.89 0.0 1.2 0.90 1.2 0.89 1.6 -2.5 1.2 -2.4 1.5 -2.7 1.0 -2.4 .... 1.0 -2.2 .... -1.9 .... -2.9 .... a .... a .... a .... a 16.6 c 16.7 a 17.1 16.4 1.8 17.1 2.1 17.8 5.6 4.4 2.2 1.5 .1 1.9 0.89 0.7 -2.3 g... .2 0.9 0.89 0.7 -1.8 .... .1 0.9 0.89 0.7 -2.3 .... .8.5 0.4 1.7 0.90 0.91 1.0 1.1 -2.1 -2.8 .... g... a 16.9 1.1 .1 0.9 0.90 0.7 -2.4 .... .1 2.0 0.89 0.7 -2.6 .... .2 0.0 0.89 0.9 -2.8 .... .4 3.7 0.90 1.1 -2.9 g... .3 0.8 0.89 0.9 -2.5 .... .6.6 0.9 5.0 0.89 0.91 1.4 1.4 -2.8 -2.3 .... g... a 17.4 1.7 .6.3 1.6 2.8 0.90 0.90 1.1 0.9 -2.5 -2.5 ...... a 17.2 1.8 .4.8.1 0.9.5 2.2 0.89 0.2 0.90 0.0 0.89 0.89 1.3 1.6 1.4 -2.5 1.2 -2.4 -3.0 -1.9 ...... g... a a a 17.1 32.7 16.7 1.4 1.2 0.9 .6.2 0.5 0.89 0.0 0.89 1.1 0.9 -2.4 -2.8 ...... a 16.9 1.7 5.7 7.5 0.91 1.7 -2.7 .... a 15.3 4.3 Extracted Counts Net ∆Net Bkgd Net PSF Signif log Catalog: Tentative Source Properties ) Full Full Full Hard Frac ′ θ Chandra ) ( ′′ Err J2000 δ Position Table 5.2. J2000 (deg) (deg) ( α Source 2 063232.91+043437.2 98.137138 4.577017 1.4 8.4 5.9 3.4 2.1 1216 063247.43+043150.226 063249.22+043425.9 063254.45+043724.6 98.197650 98.205124 4.530632 98.226879 4.573882 1.3 4.623523 1.1 0.9 7.9 7.8 6.3 4.5 4.9 4.6 3.0 3.4 3.0 1. 3. 1. 2933 063255.82+043419.136 063258.15+043003.753 063300.61+043030.757 063308.34+044203.4 98.232618 063308.52+043634.3 98.242303 4.571979 98.252556 4.501051 98.284762 1.1 4.508536 98.285532 1.1 4.700962 0.9 4.609538 7.2 1.0 6.6 0.3 5.9 7.3 5.2 4.0 3.3 3.5 3.4 4.6 2.5 4.7 2.5 2. 3.2 0. 3.0 0. 2. 1. 5860 063309.33+043804.375 063309.66+044321.0 063313.25+043023.3 98.288891 98.290259 4.634532 98.305242 4.722503 1.1 4.506480 1.0 0.7 6.1 8.0 4.4 6.9 7.4 2.7 3.7 3.9 2.3 3. 3. 0. 78 063313.88+043421.9 98.307857 4.572767 0.4 0.9 1.9 2.0 0. 93 063316.30+042952.7 98.317938 4.497997 0.7 4.8 3.5 2.5 0. # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 372381 063436.02+042532.0386 063439.22+042732.1 063440.41+043344.8 98.650107 98.663423 4.425572 98.668391 4.458921 0.8 4.562469 0.8 1.4 5.2 5.5 8.3 3.6 3.5 5.0 2.5 2.5 3.2 0 0 2 363365 063433.33+042210.9 063433.50+042834.2 98.638902 98.639618 4.369722 4.476188 1.0 0.6 6.9 4.1 4.8 2.8 3.0 2.3 1 0 339341 063427.78+042128.0344 063428.48+042125.2348 063429.17+041926.7349 063430.65+042140.3350 98.615784 063430.66+041932.0361 98.618690 063430.75+042240.3 4.357783 98.621580 063433.26+042751.7 4.357002 98.627727 1.0 4.324095 98.627761 1.1 4.361198 98.628135 1.1 4.325561 6.8 98.638613 1.1 4.377874 6.9 1.2 4.464362 8.8 1.0 6.9 0.6 4.1 8.9 4.0 6.1 7.2 4.0 2.8 4.0 2.8 6.6 3.9 3.1 0 2.8 2.8 1 3.9 3 2.5 1 2.3 4 0 0 335 063427.41+042547.2 98.614227 4.429796 0.6 3.2 1.9 2.0 0 308 063422.05+043051.6 98.591901 4.514355 0.6 3.3 1.8 2.0 0 305 063421.72+042452.3 98.590504 4.414546 0.6 3.1 1.9 2.0 0 274298 063415.57+043423.7 063420.04+042223.5 98.564896 98.583528 4.573273 4.373197 1.1 0.8 6.6 5.4 3.2 3.5 2.5 2.5 0 0 262 063412.55+042711.1 98.552318 4.453111 0.4 1.3 1.9 2.0 0 252 063411.38+042847.2 98.547429 4.479786 0.4 1.8 1.9 2.0 0 245 063410.62+043156.3 98.544271 4.532329 0.7 4.5 2.8 2.3 0 223 063407.71+043307.2 98.532139 4.552012 0.9 5.9 3.6 2.5 0 220 063407.36+042445.1 98.530677 4.412532 0.6 3.9 2.7 2.3 0 199218 063404.23+043433.7 063407.31+041910.9 98.517656 98.530497 4.576055 4.319709 1.1 1.9 7.5 8.9 5.4 5.4 3.2 3.4 1 2 173193 063356.51+042428.8 063403.21+042500.7 98.485471 98.513388 4.408012 4.416877 1.0 0.7 6.1 4.5 3.4 2.7 2.5 2.3 0 0 147159 063341.08+043930.0165 063348.98+043132.2166 063351.58+042504.5 063352.42+043703.6 98.421186 98.454120 4.658342 98.464928 4.525628 98.468432 1.2 4.417921 1.3 4.617672 1.0 7.7 1.5 8.3 7.0 9.1 4.6 7.2 4.9 3.0 4.5 4.0 3.0 1 3.2 4 1 2 116117 063325.04+044018.5125 063325.78+042421.2 063329.71+043126.1 98.354342 98.357433 4.671829 98.373831 4.405906 1.0 4.523922 1.8 0.7 5.9 10.6 4.5 3.4 8.3 2.8 2.5 4.3 2.3 0 0 Seq CXOU J 154 ); b = values B Column KS median Columns P E < P nts. 05 . (ks) (keV) (16) (17) Photometric significance Off-axis angle. Background counts extracted s or for sources in chip gaps or Column 6: Characteristics Column 12: Anom Var EffExp . 9 ; e = source on field edge; p = source . Column 9: 0 eration after sources are removed from the B < region. P ) is Right ascension and declination for epoch J2000.0. ly from background. Some sources have mkarf Fraction of the PSF (at 1.497 keV) enclosed within the l band): a = no evidence for variability (0 errors on column 7. σ Columns 3,4: Column 11: # of counts extracted eV). √ ave to be observed on axis to obtain the reported number of cou ector (FRACEXPO from IAU designation. standard deviation of PSF inside extraction region 005). No value is reported for sources with fewer than 4 count . 0 Column 2: < Table 5.2—Continued KS eV); average of the upper and lower 1 P background estimates can rise during the final extraction it Net ∆Net Bkgd Net PSF Signif log , computed as σ cantly below 90%) may indicate that the source is in a crowded Log probability that extracted counts (total band) are sole ) Full Full Full Hard Frac ′ θ Variability characterization based on K-S statistic (tota ) ( ′′ Err Column 13: . Column 15: J2000 05); c = definitely variable ( . δ 0 < Effective exposure time: approximate time the source would h Estimated net counts extracted in the hard energy band (2–8 k KS J2000 α X-ray catalog sequence number, sorted by RA. < P Source anomalies: g = fractional time that source was on a det net counts 005 . Column 16: Column 10: upper error on net counts Estimated random component of position error, 1 Column 1: Column 14: Source Position Extracted Counts Estimated net counts extracted in total energy band (0.5–8 k Background-corrected median photon energy (total band). # (deg) (deg) ( Note. — (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq CXOU J (total band). 7,8: 17: catalog. above the 1% threshold that defines the catalog because local extraction region. Note that a reduced PSF fraction (signifi on field edges. Column 5: piled up; s = source on readout streak. possibly variable (0 computed as d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes

155 t,c L ··· ··· ··· ··· ··· ··· ··· ··· log t c L log ) 1 − h,c 05 30.06 30.34 .76 30.78 30.94 L log h X-ray Luminosities L 30.44 30.82 30.50 30.48 30.61 30.49 30.22 30.56 30.26 30.06 30.32 30.08 log s L · · · · · · · · · · · · 29.55 28.93 28.93 29.64 29.65 29.76 30.47 30.49 30.55 30.66 29.58 29.68 29.71 29.93 30.15 29.42 30.22 30.29 30.29 30.54 29.14 30.34 30.46 30.37 30.71 29.53 28.83 28.83 29.61 29.62 29.63 29.90 29.91 30.09 30.18 29.58 29.28 29.29 29.76 29.88 29.04 29.89 30.05 29.95 30.49 29.80 27.91 27.91 29.80 29.82 30.53 30.65 30.70 30.89 31.24 29.75 29.10 29.14 29.84 30.25 29.72 30.14 30.21 30.28 30.65 29.36 30.10 30.16 30.17 30.43 29.77 29.10 29.10 29.85 29.86 30.45 27.47 27.47 30.45 30.48 29.48 30.48 30.55 30.52 30.75 29.86 30.02 30.07 30.24 30.60 29.67 29.98 30.02 30.15 30.42 log 2 2 2 7 2 2 3 5 3 2 2 3 6 2 2 7 6 3 8 5 4 2 ...... +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +1 +0 +0 ) (ergss 3 − EM d Sources: Thermal Plasma 52.7 53.5 53.2 53.4 52.7 53.1 53.6 53.0 53.5 54.4 30.64 30.71 30.76 30.98 31.33 53.0 54.3 54.3 53.3 53.7 53.8 53.3 52.9 53.8 53.6 53.7 53.4 3 2 5 4 3 4 4 3 4 4 2 4 4 2 3 3 4 4 2 log ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − − − − − − − b 6 7 2 8 3 6 5 09 9 ...... +0 +0 +0 +0 +4 +0 +0 +0 +2 1.2 1.6 0.4 1.6 1.4 0.9 1.2 0.2 1.7 1.2 1.2 53.1 29.91 29.22 29.22 29.99 30.00 2.3 53.3 29.76 29.87 29.89 30.12 30.28 4.3 2.0 54.1 29.67 30.44 30.59 30.51 31.03 3.9 1.5 4.6 53.6 4.3 52.9 29.68 29.81 29.81 30.05 30.06 kT 4.0 3.4 53.2 29.07 29.91 30.00 29.97 30.29 7.0 2.2 2.0 2.0 2.0 2.0 15.0 15.0 4 5 2 3 9 6 3 9 7 4 3 7 9 ...... 05 0 0 0 0 0 0 0 0 1 3 3 2 2 . 0 − − − − − − − − − − − − Fits − Spectral Fit − 3 2 4 2 1 3 3 4 2 4 4 4 3 7 9 3 4 7 1 3 ...... +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +1 +0 ) (keV) (cm H 2 − N 21.7 21.8 22.1 22.4 21.9 22.4 22.7 22.1 22.7 21.9 22.2 22.2 21.8 20.0 20.0 20.0 20.0 20.0 20.0 6 3 9 7 2 5 4 7 3 6 7 5 9 ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − a Table 5.3. X-ray Spectroscopy for Photometrically Selecte Source 9 063245.25+043206.5 21.4 3.6 7 063240.73+043653.3 14.4 2.9 6 063239.51+043124.9 11.7 2.3 22.1 5 063239.50+043628.9 12.3 2.6 72 063312.84+043458.5 19.9 3.6 77 063313.85+043250.9 12.9 2.7 21.5 66 063311.30+044249.1 10.0 2.1 59 063309.40+043106.7 88.7 8.5 64 063311.23+043007.3 9.6 2.3 61 063310.06+042547.4 13.0 2.6 22.7 1.1 54.7 56 063308.51+042832.6 19.1 3.4 54 063308.46+043731.9 9.9 2.1 51 063307.48+043048.6 12.6 2.7 47 063306.39+043454.0 8.9 2.2 21.4 4950 063306.79+044115.9 063306.82+044347.6 58.0 12.5 6.6 2.4 21.9 21.6 10.0 53.7 29.91 30.72 30 52 063308.31+043752.2 144.3 11.0 46 063305.02+043857.9 9.6 2.2 45 063304.40+043156.7 8.8 2.1 22.4 39 063301.90+043213.0 8.7 2.1 22.6 44 063304.05+043319.9 8.9 2.2 32 063257.18+043017.4 10.2 2.3 25 063253.98+043828.4 12.7 2.6 21.2 24 063253.93+043622.6 17.2 3.2 20.5 23 063253.46+043716.8 19.5 3.4 21.6 22 063252.30+043744.3 15.1 2.9 21 063251.73+043621.4 10.4 2.3 18 063251.19+043412.1 29.9 4.4 17 063249.80+043641.9 10.2 2.2 10 063246.11+043612.9 10.8 2.3 22.1 3.5 53.3 29.28 29.98 30. 14 063248.65+043404.4 14.6 2.7 # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) Seq CXOU J Net Signif log d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes 156 t,c L ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· log t c L log ) 1 − h,c .57 30.77 30.88 L 31.04 31.51 31.91 2T log h X-ray Luminosities L 31.38 31.54 31.38 30.98 31.37 30.98 30.10 30.45 30.15 30.47 31.40 30.47 30.57 30.80 30.58 log s L ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· · · · · · · · · · · · · · · · 29.93 28.65 28.65 29.95 29.97 29.3729.95 30.06 29.45 30.08 29.45 30.14 30.06 30.24 30.08 29.89 30.18 30.24 30.36 30.68 29.83 29.43 29.52 29.98 30.69 30.08 29.91 30.00 30.30 30.95 30.16 30.19 30.22 30.48 30.74 30.66 29.25 29.25 30.68 30.69 30.46 29.85 29.85 30.55 30.56 29.50 30.36 30.48 30.41 30.85 30.59 30.49 30.56 30.85 31.29 30.48 30.17 30.27 30.65 31.38 29.19 30.25 30.31 30.29 30.47 29.67 30.75 30.81 30.79 30.97 30.52 30.81 30.83 30.99 31.15 29.64 29.57 29.57 29.90 29.91 29.39 30.63 30.73 30.65 30.93 log 2 5 3 5 4 4 1 0 5 8 1 1 4 3 7 3 0 3 2 2 2 0 ...... +0 +0 +0 +0 +0 +0 +0 +1 +1 +0 +0 +0 +1 +1 +0 +0 +1 +0 +0 +0 +0 +1 ) (ergss 3 − EM 53.2 29.36 30.17 30.20 30.23 30.37 53.5 53.0 53.2 29.60 30.21 30.22 30.31 30.39 53.2 53.7 53.7 54.5 52.9 54.4 53.8 53.7 53.6 55.4 54.4 54.4 53.3 53.8 53.5 30.03 29.08 29.11 30.08 30.46 54.1 52.9 53.8 3 4 6 1 3 4 2 2 4 2 2 3 3 log ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − b 1 2 5 5 7 4 7 3 3 ...... +5 +0 +0 +0 +4 +4 +3 +0 +0 0.7 1.4 2.0 6.8 53.5 29.91 30.44 30.46 30.55 30.66 1.7 0.6 1.3 2.3 53.1 29.60 29.72 29.74 29.97 30.13 1.2 54.3 0.9 1.0 kT 2.8 54.1 3.1 1.2 2.5 53.8 0.7 2.7 6.8 15.0 53.1 15.0 15.0 15.0 15.0 15.0 15.0 15.0 1 2 3 5 4 4 3 0 6 2 8 9 4 4 ...... 1 0 0 0 0 1 0 1 5 1 0 1 1 2 − − − − − − − − − − − − − − Spectral Fit 4 4 5 5 1 2 2 4 3 3 2 2 3 3 4 ...... +1 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +1 +0 ) (keV) (cm H 2 Table 5.3—Continued − N · · · · · · 21.9 21.7 22.7 22.0 23.2 22.7 21.6 22.4 21.9 22.2 22.2 22.4 20.0 20.0 20.0 20.0 7 3 1 4 4 5 4 6 9 3 2 4 ...... 0 1 0 0 0 0 0 0 0 0 2 0 − − − − − − − − − − − − a Source 98 063319.07+043038.8 8.8 2.1 21.5 92 063316.19+043452.6 7.9 2.0 22.7 90 063315.60+043504.0 14.9 3.0 89 063315.44+043459.3 20.9 3.7 23.2 9.3 99 063319.44+044050.7 9.3 2.2 95 063316.82+043029.2 79.7 8.0 87 063315.24+043305.8 35.0 5.0 84 063314.76+042853.4 24.4 4.0 82 063314.57+042910.2 8.4 2.0 79 063313.96+042827.9 17.2 3.2 21.5 81 063314.56+043536.8 17.9 3.4 80 063314.34+043612.1 7.9 2.0 # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 149 063341.16+043705.1 34.0 4.9 146 063340.49+042259.9 14.5 2.6 145 063340.06+042333.7 10.4 2.1 21.9 144 063339.48+043439.7 18.5 3.4 143 063339.45+042818.8 13.1 2.6 20.3 135 063333.75+044010.3 8.8 2.1 21.7 134 063332.48+042510.2 11.7 2.3 21.5 133 063332.45+043922.5 28.3 4.4 131 063332.05+042832.3 17.9 3.3 21.6 126 063330.28+043830.3 67.6 7.3 123 063329.13+042448.2 12.9 2.5 21.9 0.8 119 063327.50+043556.9 368.9 18.2 21.5 2.1 54.5 31.34 31.01 121 063328.06+043300.2 23.0 3.9 111 063323.78+043154.7 11.9 2.6 107 063323.21+042444.6 42.5 5.4 106 063322.23+044650.7 30.5 3.7 103105 063320.72+044215.0 063322.17+042905.6 45.3 23.4 5.7 3.9 21.4 22.0 3.2 1.0 53.8 54.0 30.36 30.56 30 102 063319.77+043703.0 13.9 2.9 Seq CXOU J Net Signif log d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes 157 t,c L ··· ··· ··· ··· ··· ··· log t c L log ) 1 − h,c L log h X-ray Luminosities L 30.49 30.97 30.49 30.47 30.64 30.48 30.71 31.33 30.71 log s L · · · · · · · · · 29.74 29.82 29.85 30.08 30.29 29.60 29.69 29.71 29.94 30.15 30.02 29.98 30.01 30.30 30.56 30.67 29.25 29.26 30.69 30.70 29.79 30.51 30.53 30.59 30.70 29.64 30.03 30.10 30.18 30.54 29.73 29.64 29.65 29.99 30.09 30.29 30.35 30.35 30.63 30.63 29.91 30.09 30.13 30.31 30.57 30.22 29.84 29.98 30.37 31.50 29.88 29.82 29.83 30.15 30.28 29.86 29.80 29.83 30.13 30.36 29.88 28.63 28.63 29.90 29.92 30.58 31.08 31.11 31.20 31.36 29.94 30.31 30.33 30.47 30.58 30.20 30.65 30.68 30.78 30.95 30.00 29.96 29.99 30.28 30.51 30.08 30.11 30.13 30.40 30.57 29.47 30.20 30.22 30.27 30.39 29.60 29.58 29.74 29.89 30.84 30.03 30.97 31.05 31.02 31.30 log 3 3 3 1 2 0 3 4 2 3 6 4 3 2 2 3 3 9 2 3 2 2 3 ...... +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 ) (ergss 3 − EM 53.3 53.2 53.6 53.7 53.5 53.1 54.0 29.98 29.76 29.87 30.18 30.97 54.4 53.5 53.6 54.5 30.05 30.63 30.82 30.73 31.47 54.6 53.3 53.4 53.0 54.2 53.5 53.9 53.6 53.6 53.2 54.3 54.2 3 4 3 1 2 3 4 1 3 3 4 3 1 3 2 3 3 6 3 log ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − − − − − − − b 3 2 3 4 4 6 3 ...... +8 +0 +9 +0 +1 +1 +6 1.6 0.6 2.0 53.6 0.9 3.7 4.7 53.9 30.46 30.79 30.80 30.95 31.04 1.9 0.7 1.7 1.3 53.3 29.76 29.52 29.54 29.96 30.21 1.4 0.6 4.4 1.6 53.3 29.97 29.67 29.68 30.14 30.21 4.4 3.9 4.3 53.8 kT 0.3 55.2 31.37 30.46 30.50 31.42 32.05 2T 0.9 53.9 4.4 2.7 54.3 29.72 30.80 30.95 30.83 31.30 2.0 2.0 2.0 2.0 2.0 2.0 15.0 15.0 15.0 5 3 8 4 6 2 8 4 1 9 5 9 6 3 4 ...... 0 0 0 0 0 0 1 1 2 2 2 1 3 2 1 − − − − − − − − − − − − − − − Spectral Fit 4 4 4 3 6 1 2 4 6 3 2 2 6 7 2 4 2 2 3 2 6 6 2 2 1 ...... +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +0 ) (keV) (cm H 2 Table 5.3—Continued − N 22.0 21.7 23.0 21.4 21.8 22.4 22.1 21.7 22.6 21.6 23.5 22.3 22.5 22.2 20.0 20.0 20.0 7 6 3 2 5 3 2 3 9 3 2 3 3 7 ...... 0 0 0 1 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − − a Source # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 205 063405.31+042808.7 10.9 2.5 21.6 204 063405.29+042654.9 8.9 2.2 21.6 203 063405.27+043038.5 22.8 3.9 21.6 200 063404.32+042529.8 11.8 2.6 197 063403.83+042626.8 19.9 3.6 196 063403.47+041944.9 40.3 5.3 195 063403.41+043408.6 8.6 2.0 21.2 194 063403.34+043327.6 18.3 3.4 22.1 191 063402.96+042512.2 8.8 2.1 190 063402.83+043450.7 61.2 6.8 188 063401.58+043241.4 17.4 3.3 182 063358.68+042145.1 34.4 4.9 181 063358.59+043807.4 27.2 4.0 184 063359.90+042828.5 27.7 4.4 180 063358.39+042732.5 15.6 3.1 21.3 179 063358.04+043331.1 96.6 8.8 177 063357.50+042159.2 10.4 2.3 175 063357.11+042656.7 13.6 2.8 21.6 172 063356.12+043119.0 10.2 2.3 21.5 171 063355.99+043350.9 11.3 2.4 170 063355.82+043101.4 20.2 3.6 21.5 169 063354.87+043101.8 40.1 5.4 21.7 168 063353.78+042833.1 15.4 3.0 20.9 164 063350.97+043131.5 309.9 16.6 163 063350.42+043658.4 22.1 3.7 21.5 161 063349.65+043205.3 29.4 4.3 21.6 1.7 160 063349.17+042306.4 10.2 2.2 157 063347.32+043756.1 45.2 5.7 156 063346.46+042934.9 12.9 2.5 21.8 152 063341.66+043427.8 29.2 4.5 150 063341.19+042646.8 12.2 2.3 Seq CXOU J Net Signif log d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes

158 t,c L ··· ··· ··· ··· ··· ··· log t c L log ) 1 − h,c L log h X-ray Luminosities L 30.56 31.00 30.56 30.71 30.92 30.72 30.03 30.33 30.11 log s L ··· ··· ··· ··· ··· ··· · · · · · · · · · 29.72 29.96 30.01 30.16 30.45 30.02 30.51 30.52 30.63 30.68 29.53 30.04 30.05 30.15 30.21 29.53 29.95 30.01 30.09 30.40 29.82 29.05 29.10 29.89 30.36 29.57 30.05 30.14 30.18 30.58 29.65 30.43 30.58 30.50 31.02 29.61 30.01 30.01 30.16 30.18 29.64 30.21 30.22 30.31 30.38 29.48 29.92 29.99 30.06 30.40 29.59 30.17 30.24 30.27 30.61 29.93 29.94 30.00 30.24 30.65 29.58 30.86 30.94 30.88 31.10 29.61 29.77 29.87 30.00 30.51 30.47 30.95 30.98 31.07 31.25 30.34 30.33 30.37 30.64 30.92 30.07 30.06 30.12 30.37 30.77 29.98 30.47 30.55 30.59 30.95 29.76 30.41 30.43 30.50 30.61 30.15 30.90 30.99 30.97 31.34 30.01 31.11 31.27 31.15 31.62 log 2 3 2 3 2 5 2 2 2 3 4 3 4 3 3 9 7 5 9 6 8 8 0 ...... +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +1 ) (ergss 3 − EM 53.5 53.0 53.5 53.0 54.1 53.0 53.2 53.4 53.6 53.2 29.89 29.05 29.07 29.95 30.17 53.7 53.9 54.1 54.2 29.48 30.03 30.30 30.14 31.15 53.6 54.0 53.8 54.0 53.4 54.3 54.6 2 3 3 3 3 4 2 2 3 3 4 2 3 4 4 log ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − − − b 3 6 3 7 3 2 3 6 9 9 9 ...... +0 +0 +1 +8 +3 +1 +6 +7 +7 +2 +6 kT 0.9 0.7 1.4 4.0 1.6 1.4 2.2 2.7 2.7 1.1 1.4 0.8 53.5 30.10 29.21 29.24 30.15 30.47 3.7 53.5 29.85 30.25 30.27 30.39 30.55 2.3 53.4 4.3 53.0 29.62 29.83 29.84 30.04 30.09 2.0 54.5 4.3 54.1 2.7 53.5 30.08 30.15 30.16 30.42 30.51 1.1 54.2 2.2 53.4 2.5 53.6 2.0 2.0 2.0 10.9 15.0 15.0 15.0 15.0 15.0 15.0 3 3 6 8 7 4 0 3 1 7 8 2 7 4 4 6 ...... 7 . 0 0 0 1 0 0 1 1 1 0 0 2 3 1 1 1 8 − − − − − − − − − − − − − − − − Spectral Fit − 4 5 6 4 4 4 9 5 2 2 6 6 2 3 3 5 2 4 5 3 4 4 3 1 3 4 2 2 ...... +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 ) (keV) (cm H 2 Table 5.3—Continued − N · · · 21.8 21.7 21.9 22.1 22.0 22.2 22.0 22.1 23.0 21.7 22.1 22.7 22.5 22.5 22.1 22.4 22.6 3 2 5 4 2 3 9 5 4 1 8 4 2 5 9 2 5 ...... 0 1 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 − − − − − − − − − − − − − − − − − a Source # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 234 063408.99+042944.5 83.9 8.2 231 063408.86+043712.7 10.8 2.2 21.8 217 063407.15+042838.0 27.9 4.4 21.2 235 063409.00+043050.8 8.9 2.2 21.3 229 063408.23+042713.7 10.9 2.5 21.6 227 063408.00+043607.6 41.0 5.3 206 063405.43+042915.5 25.9 4.2 214 063406.24+043039.7 12.8 2.7 21.3 249 063410.92+042603.5 28.9 4.5 228 063408.12+042859.0 18.9 3.5 21.7 219 063407.33+043258.9 10.6 2.4 242 063410.34+042709.7 8.9 2.2 225 063407.78+043206.6 8.7 2.1 21.1 260 063412.30+042720.3 8.9 2.2 20.8 267 063413.60+042718.5 12.9 2.7 21.4 207 063405.67+042453.7 48.7 6.1 246 063410.75+042641.5 8.9 2.2 294 063418.67+041920.6 10.9 2.3 257 063412.12+042241.9 13.5 2.8 288 063417.31+042811.7 18.9 3.5 295 063419.46+042131.8 17.8 3.3 22.4 292 063417.94+042611.6 12.9 2.7 287 063417.24+042502.3 17.9 3.4 209 063406.04+042451.4 54.8 6.5 259 063412.28+042408.8 10.9 2.5 21.2 222 063407.54+042547.2 8.9 2.2 261 063412.52+042641.1 8.9 2.2 268 063414.67+043022.5 27.9 4.4 280 063416.20+042127.1 8.9 2.1 269 063414.79+042538.2 9.0 2.2 21.9 213 063406.16+042818.2 13.9 2.9 21.4 Seq CXOU J Net Signif log d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes 159 t,c L ··· ··· ··· ··· ··· ··· ··· ··· log t c L log ) 1 − h,c .27 30.34 30.51 L log h X-ray Luminosities L 30.07 30.40 30.13 30.60 30.74 30.61 30.28 30.66 30.34 30.73 31.12 30.74 log s L ··· ··· ··· ··· ··· ··· · · · ··· ··· ··· ··· ··· ··· · · · · · · · · · 30.17 30.38 30.42 30.59 30.86 29.32 29.69 29.88 29.84 30.70 29.64 29.79 29.83 30.03 30.27 30.01 29.16 29.22 30.07 30.60 30.14 30.48 30.64 30.64 31.39 30.01 30.31 30.42 30.48 31.05 29.59 30.46 30.61 30.51 31.04 29.98 29.68 29.74 30.16 30.59 30.22 30.18 30.20 30.50 30.64 30.01 28.67 28.67 30.03 30.05 29.47 30.69 30.76 30.71 30.93 29.91 29.51 29.59 30.06 30.68 30.11 30.47 30.47 30.63 30.63 30.12 29.25 29.30 30.17 30.68 29.75 29.81 29.93 30.08 30.76 30.17 30.88 30.90 30.96 31.07 29.81 29.81 29.83 30.11 30.27 29.49 29.97 30.00 30.09 30.25 log 4 2 4 3 6 4 2 0 9 5 2 6 2 2 3 2 6 5 1 3 8 ...... +0 +0 +0 +0 +0 +0 +0 +2 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 ) (ergss 3 − EM 53.1 52.9 53.9 53.8 53.3 53.6 54.5 54.1 53.7 53.7 53.7 53.1 54.6 53.7 53.4 53.7 53.8 53.9 53.3 53.1 53.7 29.78 29.89 29.98 30.14 30.62 5 5 4 3 3 3 2 3 2 4 1 3 4 log ...... 0 0 0 0 0 0 0 0 0 0 0 0 0 − − − − − − − − − − − − − b 6 7 5 2 5 5 2 ...... +0 +2 +0 +0 +3 +1 +1 1.2 54.2 1.2 54.2 29.72 30.13 30.32 30.27 31.10 1.4 4.1 53.3 29.82 30.17 30.18 30.33 30.44 1.2 8.9 1.1 2.1 54.1 0.7 8.9 54.2 29.86 31.11 31.20 31.13 31.39 0.7 0.9 53.7 kT 1.7 53.8 29.61 30.08 30.19 30.21 30.69 1.1 4.4 4.8 53.6 29.67 30.42 30.47 30.49 30.71 1.4 2.0 2.0 2.0 2.0 2.0 2.0 15.0 15.0 15.0 15.0 0 4 4 3 4 7 9 2 1 5 5 ...... 1 0 0 0 0 0 0 1 1 3 3 − − − − − − − − − − − Spectral Fit 3 2 2 7 3 7 5 4 2 2 3 5 2 2 3 2 3 2 4 3 ...... +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +1 +0 +0 +0 +0 +0 +0 +0 ) (keV) (cm H 2 Table 5.3—Continued − N · · · · · · 21.8 22.3 22.4 22.4 22.6 22.9 22.4 21.7 22.3 21.9 22.2 22.1 22.0 21.7 22.0 20.0 7 5 5 5 4 2 3 6 4 2 6 7 7 3 5 ...... 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 − − − − − − − − − − − − − − − a Source # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 371 063435.21+041847.4 11.5 2.2 370 063434.49+041900.8 28.8 4.2 369 063434.41+042323.9 10.1 2.3 22.6 368 063434.39+043059.8 9.5 2.2 21.7 364 063433.47+042543.0 8.7 2.1 359 063432.85+041944.4 29.0 4.4 357 063432.17+041750.9 10.1 2.0 22.4 1.1 356 063432.10+041946.2 11.2 2.4 354 063431.59+042316.1 16.4 3.1 352 063431.47+041912.4 23.4 3.6 22.2 351 063431.45+042427.4 12.7 2.7 347 063430.13+042342.9 16.6 3.2 21.7 0.7 346 063429.47+042724.1 17.9 3.4 21.8 340 063428.22+042953.9 13.8 2.9 21.5 337 063427.61+043321.0 19.2 3.5 332 063427.22+043515.5 26.7 4.1 21.3 326 063425.85+042031.6 47.4 5.9 322325 063425.32+042406.1 063425.83+043018.6 11.8 13.9 2.6 2.9 21.8 4.7 53.4 29.65 30.24 30 321 063425.12+042653.1 22.9 3.9 320 063424.85+041832.8 9.3 2.0 22.7 1.1 54.6 319 063424.64+041909.3 16.8 3.1 312 063423.46+043542.7 11.9 2.5 311 063422.88+042950.7 15.9 3.1 20.3 307 063422.05+042428.9 12.9 2.7 306 063421.99+042624.3 9.9 2.3 304 063421.57+042127.6 15.9 3.0 301 063421.03+042441.6 49.9 6.1 297 063419.96+042744.0 14.9 3.0 300 063420.43+042404.7 8.8 2.1 21.7 296 063419.56+042739.2 13.9 2.9 21.4 Seq CXOU J Net Signif log d · · · · · · · · · · · · · · · · · · · · · Notes 160 Quantities in t,c L ··· ··· 1 in order to avoid log . 0 g uncertainties should t < c L ies log ) 1 − eliable. h,c L ies. Uncertainties are missing when d (0.5–8 keV). Absorption-corrected ated fit yielded non-physical results. log ing a distance of 1.4 kpc. h X-ray Luminosities L 30.49 30.81 30.50 log present the best-fit values for the column density s L ··· ··· ··· ··· ··· ··· · · · 29.65 30.44 30.47 30.50 30.63 29.95 30.40 30.40 30.53 30.57 30.17 30.19 30.19 30.48 30.49 29.61 29.77 29.81 30.00 30.25 29.89 30.24 30.31 30.40 30.75 30.24 30.22 30.30 30.53 31.06 29.74 30.55 30.63 30.61 30.92 log etric significance data from Table 5.1 and 5.2. 2 2 3 2 3 4 7 2 ...... +0 +0 +0 +0 +0 +0 +0 +0 ) (ergss 3 − EM Columns 5 and 6 53.4 53.0 53.4 53.4 53.3 53.8 54.3 54.1 53.8 2 2 7 3 3 4 4 log ...... 0 0 0 0 0 0 0 − − − − − − − b 2 . +1 5 since the soft band emission is essentially unmeasurable. hat the parameter is effectively unconstrained. Fits lackin . abundances. 22 kT 3.4 1.2 3.4 2.0 2.0 2.0 ⊙ 15.0 15.0 15.0 4 7 4 Z > . . . e intervals. More significant digits are used for uncertaint 0 1 2 − − − H Spectral Fit N estimates of luminosities; actual parameter values are unr 6 4 4 2 2 3 3 ...... +0 +0 +0 +0 +0 +0 +0 ) (keV) (cm he convenience of the reader. ignificant digits is used for both lower and upper uncertaint H 2 Table 5.3—Continued − N · · · 21.8 22.0 22.0 22.8 22.2 20.0 4 5 4 2 5 means the fit was performed by hand, usually because the autom : s = soft band (0.5–2 keV); h = hard band (2–8 keV); t = total ban . . . . . 1 0 0 0 0 and assumed 0.3 − − − − − H presents the emission measure for the model spectrum, assum XSPEC Column 7 columns 8–12 ; they are omitted when log c reproduce the source identification, net counts, and photom a columns 1–4 Source was unable to compute them or when their values were so large t means a two-temperature model was used. were frozen in the fit. Uncertainties represent 90% confidenc 2T All fits used the “wabs(apec)” model in For convenience X-ray luminosities are presented in d c b a # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 375 063437.67+043147.7 25.1 4.1 373 063436.55+042138.8 15.2 2.9 376 063438.09+042209.9 33.7 4.8 377 063438.43+042146.2 20.0 3.5 21.1 383 063439.46+042317.9 17.6 3.2 389 063446.45+043101.8 11.4 2.4 387 063443.25+042351.3 9.3 2.1 21.7 393 063450.44+042702.2 10.1 2.2 392 063449.98+042739.4 18.2 3.3 Seq CXOU J Net Signif log Well-known counterparts from Table 5.5 are listed here for t luminosities are subscripted with a be considered to merely be a spline to the data to obtain rough large rounding errors; forXSPEC consistency, the same number of s italics and plasma temperature parameters. d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes he same

aw photon 161 t,c L ··· ··· ··· ··· log t L ) 1 d (0.5–8 keV). Absorption- − c log s 2 − h,c 49 30.48 30.54 45 30.46 30.50 .29 30.29 30.44 L were frozen in the fit. Uncertainties ered to merely be a spline to the data was unable to compute them or when log X-ray Fluxes h L (photons cm XSPEC 30.67 30.88 30.67 30.63 30.79 30.64 log s L ··· ··· ··· ··· ··· ··· · · · · · · 29.68 30.39 30.40 30.47 30.53 29.28 30.34 30.35 30.38 30.42 29.58 30.49 30.56 30.54 30.86 29.97 31.02 31.06 31.06 31.17 29.71 30.55 30.58 30.61 30.76 log Quantities in italics etric significance data from Table 5.1 and 5.2. . 3 8 2 6 8 . . . . . +0 +0 +1 +0 +0 Γ 1 in order to avoid large rounding errors; for consistency, t . d Sources: Power Law Fits N 0 -6.2 -5.2 5 since the soft band emission is essentially unmeasurable. < . 22 present the best-fit values for the column density and power l > b 3 0 . . H +1 +1 N Γ log 1.4 -5.6 29.53 30.49 30.53 30.54 30.68 0.6 -5.5 29.56 31.12 31.16 31.13 31.22 1.3 -5.6 29.36 30.52 30.57 30.55 30.71 0.7 1.5 -5.4 29.64 30.57 30.61 30.62 30.78 1.2 -5.8 1.1 -5.5 1.1 1.4 -5.3 29.78 30.79 30.84 30.83 30.99 1.0 -5.5 29.54 30.78 30.83 30.81 30.93 1.4 -5.6 29.41 30.39 30.43 30.43 30.59 2.0 -5.2 1.5 -5.5 Spectral Fit 6 6 5 6 4 5 5 3 3 4 4 4 4 ncertainties. Uncertainties are missing when ...... +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 ) nconstrained. Fits lacking uncertainties should be consid Columns 5 and 6 H 2 r values are unreliable. are used for uncertainties . − N : s = soft band (0.5–2 keV); h = hard band (2–8 keV); t = total ban 22.0 22.1 22.2 22.1 22.2 21.9 7 7 1 3 9 7 ...... 0 0 1 1 0 0 − − − − − − XSPEC ; they are omitted when log c columns 8–12 reproduce the source identification, net counts, and photom presents the power law normalization for the model spectrum a Table 5.4. X-ray Spectroscopy for Photometrically Selecte columns 1–4 Source Column 7 All fits used the “wabs(powerlaw)” model in For convenience X-ray luminosities are presented in 4 063238.08+043250.1 16.8 3.1 20.0 2.2 -5.4 76 063313.38+044323.9 16.1 2.9 22.0 0.5 -6.2 29.08 30.47 30. 86 063314.88+043747.3 28.9 4.5 73 063312.89+043552.7 22.9 3.9 34 063259.40+043857.7 16.5 3.1 21.6 40 063301.92+044040.8 19.4 3.4 28 063255.18+043651.6 34.7 4.9 a b c # Counts (cm (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) 120 063327.94+042912.7 44.4 5.7 240 063409.34+042310.0 13.7 2.8 22.0 278 063416.11+041850.8 23.7 3.8 22.2 139 063337.03+043407.6 12.7 2.7 22.2 302 063421.26+041933.0 16.1 3.0 22.0 162 063350.30+043840.0 16.2 3.0 21.5 158 063348.25+043302.8 9.5 2.1 21.7 0.5 -6.2 29.25 30.43 30. 154 063344.49+042905.3 12.1 2.3 22.1 1.4 -5.8 29.21 30.25 30 201 063404.55+043009.7 10.8 2.4 22.8 367 063434.27+042916.3 14.8 3.0 390 063447.41+042857.8 9.0 2.1 22.8 1.9 -4.9 Seq CXOU J Net Signif log their values were so large that the parameter is effectively u index parameters. number of significant digits is used for both lower and upper u represent 90% confidence intervals. More significant digits to obtain rough estimates of luminosities; actual paramete corrected luminosities are subscripted with a 162 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· .55 AAA000 17.48 17.26 18.34 17.52 18.12 17.00 9.44 8.55 8.23 AAA000 ··· ··· ··· ··· ··· ··· 15.05 14.49 14.23 AAA000 13.57 13.15 13.02 AAA000 12.99 12.55 12.46 AAA000 13.68 12.99 12.79 AAA000 12.44 12.05 11.98 AAA000 15.68 15.08 14.88 AAB000 12.72 12.40 12.35 AAA000 15.36 14.67 14.54 AABccc 15.56 14.90 14.74 AAA000 063239+043629 14.69 13.45 12.54 063246+043610 14.69 14.29 14.06 063252+043744 063259+043654 17.74 17.28 16.81 063259+043857 063301+044113063301+043214 16.78 16.19 15.85 063301+044040 ····················· Optical/Infrared Photometry 06324673+0438322 063246+04383206324866+0434038 15.80 063248+043403 14.81 15.2806325120+0434122 14.41 063251+043412 14.70 AAA000 16.37 14.40 15.53 AAA000 15.17 ABC000 06325587+0434198 063255+043419 16.37 15.54 15.23 ABU000 06330105+0438449 063301+043844 15.46 14.18 13.13 AAA000 14.54 16.26 16.74 06325634+0435080 063256+043507 15.50 14.78 14 Table 5.5. Stellar Counterparts ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0095331 15.41 19.68 16.49 06325345+0437168 063253+043716 0095483 0095335 16.02 15.20 14.62 06325387+0436230 063253+043623 0095339 14.62 12.63 11.02 06325401+0438283 063259+044257 0095344 15.38 14.40 14.06 06325454+0436045 063254+043604 0095525 18.41 16.59 14.86 06324526+0432074 063245+043207 0095611 0095624 14.97 13.90 13.74 06325722+0430186 063257+043018 0095647 20.33 18.36 17.21 06330059+0430318 063300+043031 0095669 14.14 13.63 13.84 06330264+0435123 063302+043512 0095428 20.03 17.82 16.93 06330267+0436406 063302+043640 0095702 19.74 17.55 16.39 06330639+0434543 063306+043454 ··································································································· ······································································································ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ············································································································································································································ ········································································ ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ···································· ··· ··· ··· ································· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ··································································································· − − − − − − − − − − − − X-ray Source 12 063232.43+043705.7 3 063232.91+043437.2 4 063236.58+043611.4 063238.08+043250.1 0945 56 063239.50+043628.9 7 063239.51+043124.9 8 063240.73+043653.3 9 063242.33+043217.9 063245.25+043206.5 0945 23 063253.46+043716.8 0946 10 063246.11+043612.9 11 063246.72+043831.4 1213 063247.43+043150.2 14 063248.45+043539.8 15 063248.65+043404.4 16 063248.96+043027.4 17 063249.22+043425.9 18 063249.80+043641.9 19 063251.19+043412.1 20 063251.29+043714.0 21 063251.49+043801.9 22 063251.73+043621.4 063252.30+043744.3 24 063253.93+043622.6 0946 25 063253.98+043828.4 0946 2627 063254.45+043724.6 063254.51+043603.7 0946 2829 063255.18+043651.6 30 063255.82+043419.1 063256.25+043509.1 0945 3132 063257.10+044001.7 063257.18+043017.4 0945 3536 063259.45+043653.3 063300.61+043030.7 0945 3334 063258.15+043003.7 063259.40+043857.7 3738 063300.96+043843.6 063301.50+044112.8 39 063301.90+043213.0 40 063301.92+044040.8 4142 063302.56+043426.7 063302.67+043512.1 0945 43 063302.74+043639.9 0946 4445 063304.05+043319.9 46 063304.40+043156.7 47 063305.02+043857.9 063306.39+043454.0 0945 # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 163 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· 97 AAA000 .96 AAA000 .40 AAA000 .80 AAA000 .12 AAA000 .78 AAA000 17.47 17.49 16.70 9.83 9.63 9.56 AAA000 ··· ··· ··· ··· ··· ··· 14.47 13.63 13.33 AAA000 12.77 12.03 11.75 AAA000 14.61 13.97 13.70 AAA000 14.65 14.07 13.79 AAA000 14.14 13.60 13.54 AAA000 13.87 12.88 12.37 AAA000 16.23 14.54 12.57 UAA0cc 12.98 11.80 10.98 AAA000 14.56 13.50 12.83 AAA00c 15.27 14.11 13.55 AAA000 14.62 13.80 13.53 AAA000 12.03 11.11 10.75 AAEc00 12.79 12.53 12.39 AAA000 12.43 11.48 11.08 AAA000 13.81 13.26 12.94 AAA000 11.89 11.69 11.64 AAA000 ··· 063314+042853 063314+042909 15.58 14.66 13.94 063309+044323063310+042548 16.15 15.44 15.14 063308+042832 Optical/Infrared Photometry 06331621+0434530 063316+04345306331670+0437098 063316+043709 16.38 16.67 14.92 15.87 12.86 15.51 UAA000 BCC000 06331545+043459606331557+0435041 063315+043459 15.12 11.90 9.49 AAA000 06331456+0435371 063314+043536 17.69 15.87 14.40 UUA00c 06331120+0442498 063311+044249 15.53 14.63 14.21 AAA000 06331116+0434067 063311+043406 15.45 14.60 14.29 AAA000 06330929+0438040 063309+043804 15.21 14.01 12.91 UUA00c 06330846+0431009 063308+043100 16.43 15.56 15.05 BBB000 Table 5.5—Continued 7.93 11.36 06330854+0436345 063308+043634 11.33 11.04 10. 19.07 17.77 06331484+0436340 063314+043634 15.09 13.96 12 19.33 17.57 06331154+0431360 063311+043136 15.68 14.77 14 18.28 16.99 06331131+0435334 063311+043533 14.58 13.47 12 19.47 18.11 06331124+0430068 063311+043006 15.09 13.82 13 18.27 16.32 06330691+0443468 063306+044346 14.90 14.11 13 ··· ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0095804 20.13 17.52 16.06 06331564+0432100 063315+043209 0095801 18.44 15.76 14.33 06331524+0433062 063315+043306 0095550 0095548 20.04 17.35 15.90 06331473+0436056 063314+043605 0095541 20.35 17.36 16.39 06331433+0436122 063314+043612 0094990 17.32 15.60 15.01 06331396+0428285 063313+042828 0095789 20.07 17.53 15.62 06331386+0432513 063313+043251 0095781 20.16 17.31 15.17 06331284+0434589 063312+043458 0095764 20.42 18.53 16.37 06331191+0431193 063311+043119 0095762 0095761 21.24 18.31 17.84 06331137+0432097 063311+043209 0095760 0094969 10.88 10.63 10.52 06331127+0427316 063311+042731 0095757 0094959 19.97 17.63 15.95 06331030+0429312 063310+042931 0095730 17.26 14.98 13.54 06330941+0431071 063309+043107 0095479 0095476 14.36 12.78 13.03 06330841+0437325 063308+043732 0095475 19.10 16.25 14.50 06330829+0437523 063308+043752 0095714 19.87 17.31 15.15 06330748+0430484 063307+043048 0095643 0095703 13.17 12.60 12.83 06330642+0435205 063306+043520 ······································· ··· ··· ··· ··· ··· ··· ······································· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ································· ···································································································································· ································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ································· ································· − − − − − − − − − − − − − − − − − − − − − − X-ray Source 9293 063316.19+043452.6 94 063316.30+042952.7 063316.71+043709.1 91 063315.64+043209.8 0945 8889 063315.33+043125.5 90 063315.44+043459.3 063315.60+043504.0 8687 063314.88+043747.3 063315.24+043305.8 0945 84 063314.76+042853.4 85 063314.83+043633.6 0946 8182 063314.56+043536.8 063314.57+042910.2 83 063314.72+043605.0 0946 80 063314.34+043612.1 0946 7879 063313.88+043421.9 063313.96+042827.9 0944 7374 063312.89+043552.7 75 063313.00+043214.1 76 063313.25+043023.3 77 063313.38+044323.9 063313.85+043250.9 0945 7172 063312.44+042739.8 063312.84+043458.5 0945 70 063311.93+043119.1 0945 69 063311.54+043136.5 0945 68 063311.39+043209.5 0945 66 063311.30+044249.1 67 063311.32+043533.2 0945 65 063311.25+042732.4 0944 6364 063311.18+043406.4 063311.23+043007.3 0945 60 063309.66+044321.0 61 063310.06+042547.4 62 063310.31+042931.8 0944 58 063309.33+043804.3 59 063309.40+043106.7 0945 5556 063308.49+043101.3 063308.51+042832.6 57 063308.52+043634.3 0946 5354 063308.34+044203.4 063308.46+043731.9 0946 52 063308.31+043752.2 0946 51 063307.48+043048.6 0945 4950 063306.79+044115.9 063306.82+044347.6 0947 48 063306.45+043519.8 0945 # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 164 A000 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· .31 AAA000 16.31 17.16 ··· ··· ··· ··· ··· 18.34 17.73 16.69 15.65 ··· ··· ··· ··· 15.94 15.22 14.94 AAB000 14.66 13.84 13.56 AAA000 14.52 13.66 13.39 AAA000 14.74 13.94 13.66 AAA000 14.61 13.53 13.14 AAA000 14.89 13.98 13.81 AAA000 15.62 14.70 14.21 AAAc00 12.45 11.66 11.07 AAA000 11.91 11.41 11.31 AAA000 15.55 14.74 14.44 AAA000 15.52 14.68 14.40 AAA000 16.92 16.27 15.10 CDB000 15.37 14.51 14.24 AAA000 13.03 12.60 12.47 AAA000 14.99 14.41 14.29 AAA000 14.16 13.37 13.15 AAA000 12.23 11.32 10.81 AAA000 13.99 13.10 12.76 AAA000 13.82 13.02 12.54 AAA000 15.41 14.81 14.55 AAA000 15.55 14.95 14.67 AAA000 ··· ··· 063337+043446 18.16 063332+042509 063329+043126 17.95 17.35 17.06 063330+043830 063327+042912 18.78 17.56 16.68 063323+043156 063251+041000 9.95 9.75 9.71 063319+043824 16.90 063311+043136 11.04 10.54 10.32 Optical/Infrared Photometry 06333555+0435152 063335+043515 16.47 15.74 15.32 CBC000 06333212+0431021 063332+043102 16.44 14.98 14.33 BBA000 06332947+0433332 06332500+0440187 063325+04401806332622+0440450 063326+044045 16.57 16.87 15.40 15.97 14.66 15.55 CAA000 DCC000 06332346+0436509 063323+043650 15.04 13.65 12.85 AAA000 06331976+0437034 063319+043703 16.75 15.00 13.89 CAA000 19.06 06332324+0435231 063323+043523 15.90 14.44 13.52 AA Table 5.5—Continued 16.84 16.65 06333214+0428329 19.12 17.93 06333035+0430161 063330+043016 15.65 14.79 14 ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0096012 20.94 18.92 17.23 06333685+0430298 063336+043029 0095815 20.19 17.57 16.15 06333379+0440101 063333+044010 0095797 20.40 17.50 16.12 06333247+0439225 063332+043922 0095164 0095957 20.38 17.62 16.12 06333094+0432409 063330+043240 0095782 21.58 25.85 16.74 06333049+0441406 063330+044140 0095952 0095936 21.11 19.26 17.66 06332861+0430521 063328+043052 0095923 16.58 13.62 14.00 06332806+0433005 063328+043300 0095921 15.12 13.07 13.25 06332751+0435572 063327+043557 0095897 20.40 19.09 17.53 06332492+0434369 063324+043436 0095891 19.99 18.40 17.33 06332423+0432525 063324+043252 0095659 20.41 18.17 17.15 06332391+0436440 063323+043644 0095882 0095076 15.90 14.40 14.02 06332346+0428355 063323+042835 0095811 18.57 17.34 15.79 06332211+0446495 063322+044649 0095065 19.56 16.78 15.47 06332215+0429055 063322+042905 0095795 16.83 11.12 14.20 06332068+0442157 063320+044215 0095592 20.31 17.34 15.92 06331936+0440506 063319+044050 0095834 18.80 16.52 15.41 06331906+0430386 063319+043038 0095827 20.41 18.73 16.47 06331828+0435280 063318+043527 0095035 20.34 18.77 17.16 06331815+0429281 063318+042928 ································· ···································· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ······································· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· − − − − − − − − − − − − − − − − − − − − − − X-ray Source 99 063319.44+044050.7 0946 98 063319.07+043038.8 0945 97 063318.27+043527.6 0945 95 063316.82+043029.2 96 063318.10+042927.8 0944 # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 141 063337.88+043934.5 139140 063337.03+043407.6 063337.58+043446.2 136137 063333.93+044029.4 138 063335.59+043515.9 063336.82+043029.0 0945 135 063333.75+044010.3 0946 134 063332.48+042510.2 132133 063332.15+043101.6 063332.45+043922.5 0946 130131 063331.40+044056.3 063332.05+042832.3 0944 129 063330.96+043240.5 0945 128 063330.59+044139.7 0946 127 063330.40+043016.5 0945 125 063329.71+043126.1 126 063330.28+043830.3 123124 063329.13+042448.2 063329.47+043331.8 122 063328.59+043052.6 0945 120121 063327.94+042912.7 063328.06+043300.2 0945 116117 063325.04+044018.5 118 063325.78+042421.2 119 063326.21+044044.2 063327.50+043556.9 0945 114115 063324.49+043950.8 063324.94+043436.5 0945 113 063324.23+043252.2 0945 111 063323.78+043154.7 112 063323.93+043643.6 0946 107108 063323.21+042444.6 063323.22+043522.8 0945 109110 063323.46+043650.5 063323.46+042834.9 0944 106 063322.23+044650.7 0947 104105 063321.57+043244.2 063322.17+042905.6 0944 100101 063319.56+043358.5 063319.63+043824.5 102103 063319.77+043703.0 063320.72+044215.0 0947 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 165 A000 A000 A000 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· .33 AAA000 .07 AAA000 .15 AAA000 .06 AAAcc0 ··· ··· ··· 18.21 17.16 6 7.51 7.45 AAA000 ··· ··· ··· ··· 14.53 13.68 13.33 AAA000 14.31 13.45 13.19 AAA000 10.90 10.63 10.49 AAA000 12.52 12.06 11.91 AAA000 15.06 14.06 13.77 AAA000 14.85 13.93 13.61 AAA000 14.95 13.82 13.08 AAA000 14.58 13.87 13.69 AAA000 14.58 13.56 13.17 AAA000 14.79 13.76 13.36 AAA000 12.01 11.76 11.69 AAA000 14.45 13.57 13.30 AAA000 14.30 13.46 12.94 UAA00c 14.95 13.92 13.52 AAA000 14.21 13.44 13.03 AAA000 12.92 12.34 12.20 AAA000 14.04 12.93 12.19 AAAccc 13.98 13.06 12.73 AAA000 14.40 13.36 12.96 AAA000 17.18 15.93 15.24 DCC000 ··· 063355+043349 063349+043132 15.30 14.39 14.09 063341+043638 14.23 13.38 13.04 063349+042306 17.63 16.08 17.04 063341+042648 18.21 16.91 063345+043752 063340+042331 17.73 17.14 16.84 063339+042816 14.72 13.64 12.99 ··· ··· Optical/Infrared Photometry 06335866+043806906335872+0421460 063358+043806 063358+042146 14.93 16.59 14.01 15.30 13.54 14.60 AAA000 CAA000 06335642+0424292 063356+042429 16.56 15.39 14.86 BBBc00 06335709+0426566 063357+042656 14.73 13.61 13.29 AAA000 06334844+0433030 06334113+0432315 063341+043231 16.33 15.69 15.25 BCBcc0 16.15 06335682+0433468 063356+043347 14.63 13.82 13.57 AA 15.40 06334265+0422491 063342+042249 14.45 14.00 13.91 AA 12.97 06334043+0422581 063340+042258 13.45 13.01 12.97 AA Table 5.5—Continued ··· ··· ··· 18.92 18.04 06340112+0430318 063401+043031 15.57 14.67 14 19.20 17.84 06335727+0426422 063357+042642 15.41 14.46 14 19.05 16.94 06335587+0431019 063355+043102 14.39 13.47 13 16.78 15.89 18.85 18.01 06334640+0429348 063346+042935 15.06 13.69 13 ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0096230 20.42 18.04 16.45 06340159+0432414 063401+043241 0096228 0095410 19.78 16.72 16.03 06340084+0426150 063400+042614 0096221 13.09 12.04 11.66 06340007+0430114 063401+043349 0095403 15.21 13.87 13.83 06335988+0428287 063359+042828 0095400 21.02 18.20 17.27 06335890+0426529 063358+042653 0095395 20.97 18.24 16.69 06335837+0427328 063358+042732 0096211 20.06 18.15 16.81 06335800+0433313 063358+043331 0095573 19.70 17.24 15.88 06335750+0421598 063357+042159 0096206 19.89 0095389 0096199 20.51 18.14 16.71 06335609+0431192 063356+043119 0096194 0096187 20.41 17.75 16.55 06335485+0431020 063354+043102 0095362 13.54 12.12 13.13 06335375+0428341 063353+042834 0095350 20.23 17.91 16.31 06335246+0429141 063352+042914 0096012 19.75 16.99 15.53 06335246+0437049 063352+043704 0095342 21.52 19.14 17.62 06335157+0425049 063351+042504 0096137 8.44 8.20 8.10 06335094+0431316 063345+041537 7.5 0095986 19.43 16.65 15.87 06335038+0436585 063350+043658 0096115 16.48 14.95 14.03 06334971+0432051 063349+043205 0096109 16.48 17.44 16.40 0095909 0095481 20.36 0095309 0095471 14.80 0095907 17.29 15.33 14.34 06334115+0437051 063341+043705 0096035 20.15 17.59 15.83 06333945+0434402 063339+043440 0096030 20.38 18.06 16.13 06333862+0434081 063338+043408 ······ ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ································· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· − − − − − − − − − − − − − − − − − − − − − − − − − − − − − X-ray Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 188 063401.58+043241.4 0945 187 063401.15+043031.7 0945 186 063400.85+042614.9 0944 185 063400.04+043011.2 0945 184 063359.90+042828.5 0944 181182 063358.59+043807.4 183 063358.68+042145.1 063358.95+042651.7 0944 180 063358.39+042732.5 0944 178179 063357.59+043401.6 063358.04+043331.1 0945 177 063357.50+042159.2 0943 173174 063356.51+042428.8 063356.84+043346.5 0945 175176 063357.11+042656.7 063357.33+042642.1 0944 171 063355.99+043350.9 172 063356.12+043119.0 0945 170 063355.82+043101.4 0945 169 063354.87+043101.8 0945 168 063353.78+042833.1 0944 167 063352.46+042913.1 0944 166 063352.42+043703.6 0946 165 063351.58+042504.5 0944 164 063350.97+043131.5 0945 162163 063350.30+043840.0 063350.42+043658.4 0946 160 063349.17+042306.4 161 063349.65+043205.3 0945 159 063348.98+043132.2 0945 157158 063347.32+043756.1 063348.25+043302.8 150 063341.19+042646.8 151 063341.24+043640.0 0946 152153 063341.66+043427.8 063342.51+042247.3 0943 154155 063344.49+042905.3 063345.42+043751.1 156 063346.46+042934.9 0944 145146 063340.06+042333.7 063340.49+042259.9 0943 147148 063341.08+043930.0 149 063341.15+043231.9 063341.16+043705.1 0946 143144 063339.45+042818.8 063339.48+043439.7 0945 142 063338.64+043408.9 0945 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 166 A000 ··· ··· .57 AAA000 .13 AAA000 .62 AAA000 .66 AAA000 .70 AAA000 13.35 ··· 15.17 14.32 13.97 AAA000 14.88 13.87 13.23 AAA000 15.16 14.06 13.48 AAA00c 14.82 14.03 13.67 AAA000 13.29 12.39 12.10 AAA000 13.78 13.13 12.68 AAA000 14.81 14.01 13.73 AAA000 14.46 13.43 13.06 AAA000 15.17 14.32 14.01 AAAccc 14.79 14.00 13.77 AAA000 13.49 12.97 12.90 AAA000 14.98 14.40 14.13 AAA000 13.66 12.87 12.63 AAA000 13.75 12.87 12.60 AAA000 14.30 13.40 13.10 AAA000 13.76 13.18 13.04 AAA000 14.70 13.46 12.88 AAA000 12.85 12.45 12.38 AAA000 13.24 12.23 11.62 AAA000 15.21 14.53 14.36 AAA000 13.49 12.63 12.14 AAA000 14.40 13.40 12.94 AAA000 063406+042447 16.86 Optical/Infrared Photometry 06340860+0425317 063408+042531 16.33 15.14 14.60 AAAc00 06340773+043127306340777+0432067 063407+043126 063407+043206 15.36 16.78 14.52 16.05 14.19 15.63 AAA000 CCU000 06340738+0424452 063407+042445 18.00 16.09 14.94 UDB0cc 06340597+0419233 063405+041923 17.15 14.95 13.71 UAA000 06340598+0424515 063358+041536 12.64 10.96 9.92 AEE000 06340565+0424543 063405+042454 14.19 12.31 11.48 AAAccc 06340382+042627406340396+0423070 063403+042627 063403+042307 13.71 16.27 12.55 14.64 11.87 13.91 AAA000 AAA000 06340312+042932906340317+0425008 063403+042932 063403+042500 15.56 16.45 14.41 15.05 13.88 14.35 AAA000 BAAc00 18.22 06340791+0425339 063407+042533 15.01 13.63 12.82 AA Table 5.5—Continued 19.27 18.37 06340748+0428301 063407+042830 15.87 14.92 14 19.02 17.67 06340712+0428383 063407+042838 15.44 14.47 14 20.49 18.63 06340694+0428077 063406+042807 16.08 15.04 14 18.82 17.20 06340683+0429394 063406+042939 14.97 14.02 13 18.49 17.45 06340431+0425307 063404+042530 14.50 13.35 12 ··· ··· ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0096288 20.22 18.64 17.44 06340900+0430509 063408+043050 0095468 20.39 18.06 16.53 06340897+0429447 063408+042944 0095466 21.14 17.96 17.05 06340890+0429384 063408+042938 0096127 16.98 17.88 16.54 06340890+0437119 063408+043711 0095462 19.26 16.30 15.19 06340822+0427143 063408+042714 0095461 19.43 16.58 16.55 06340812+0428594 063408+042859 0095460 0096125 20.42 18.01 16.45 06340798+0436087 063407+043608 0095457 21.01 18.15 16.53 06340753+0425472 063407+042547 0095455 0096284 20.40 18.10 17.00 06340736+0432599 063407+043259 0095452 0095450 0095449 0096273 19.85 17.30 16.28 06340622+0430399 063406+043039 0095442 16.39 14.82 14.43 06340615+0428186 063406+042818 0096272 20.71 18.26 16.58 06340589+0430272 063405+043027 0095435 18.74 15.95 15.32 06340541+0429157 063405+042915 0095434 19.24 16.77 15.51 06340530+0428091 063405+042809 0095433 20.07 17.06 16.01 06340527+0426553 063405+042655 0096265 17.33 15.58 15.06 06340524+0430390 063405+043038 0095430 0096257 20.38 17.98 16.73 06340422+0434340 063404+043433 0095615 15.20 14.17 13.92 06340353+0419452 063403+041945 0096248 18.19 15.72 14.69 06340341+0434088 063403+043408 0096247 19.65 17.69 16.32 06340333+0433282 063403+043327 0096240 19.05 16.27 15.28 06340284+0434509 063402+043450 0096238 20.13 16.94 16.17 06340220+0430076 063402+043007 ································· ··· ··· ··· ··· ······································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ···································· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ·································································· ······ ··· ··· ··· ··· ··· ··· ······································· ··· ··· ··· ··· ··· ··· − − − − − − − − − − − − − − − − − − − − − − − − − − − − X-ray Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 235 063409.00+043050.8 0945 234 063408.99+042944.5 0944 232233 063408.87+042001.9 063408.90+042938.4 0944 230231 063408.64+042532.2 063408.86+043712.7 0946 229 063408.23+042713.7 0944 228 063408.12+042859.0 0944 223224 063407.71+043307.2 225 063407.76+043127.4 226 063407.78+043206.6 063407.92+042533.6 0944 227 063408.00+043607.6 0946 222 063407.54+042547.2 0944 220221 063407.36+042445.1 063407.50+042829.7 0944 218219 063407.31+041910.9 063407.33+043258.9 0945 217 063407.15+042838.0 0944 216 063406.99+042806.8 0944 215 063406.84+042939.2 0944 214 063406.24+043039.7 0945 211212 063406.13+042914.9 213 063406.15+041921.4 063406.16+042818.2 0944 209210 063406.04+042451.4 063406.12+042447.6 207208 063405.67+042453.7 063405.91+043027.2 0945 206 063405.43+042915.5 0944 205 063405.31+042808.7 0944 204 063405.29+042654.9 0944 201202 063404.55+043009.7 203 063404.60+042925.5 063405.27+043038.5 0945 200 063404.32+042529.8 0944 197198 063403.83+042626.8 199 063404.05+042307.7 063404.23+043433.7 0945 196 063403.47+041944.9 0943 195 063403.41+043408.6 0945 191192 063402.96+042512.2 193 063403.10+042932.3 194 063403.21+042500.7 063403.34+043327.6 0945 190 063402.83+043450.7 0945 189 063402.29+043007.2 0945 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 167 A000 ··· ··· ··· ··· .05 ABAc00 .56 AAA000 .55 AAA000 .82 AAA000 .02 AAA000 .89 AAA000 .02 AAA000 .37 AAA000 .63 AAA000 .23 AAA000 .93 AAA000 .05 AAAc00 16.04 ··· ··· ··· ··· ··· ··· 14.54 13.69 13.36 AUU000 14.77 14.11 13.95 AAAs00 14.91 13.85 13.07 AAA000 14.24 13.38 13.01 AAA000 14.56 13.67 13.36 AAA000 14.80 13.94 13.61 AAAc0c 16.01 15.00 13.22 ABU000 ··· 063416+042939 16.95 16.22 15.71 063413+042013 063410+042641 14.94 Optical/Infrared Photometry 06341613+0421276 063416+042127 18.01 15.80 14.49 UBA000 06341539+0422512 063415+042251 17.28 14.83 13.55 UAA000 06341260+0427108 063412+042710 16.27 15.36 14.92 BAAcc0 06341229+0427208 063412+042720 12.36 13.54 13.19 UAA0cc 06341198+0429436 063411+04294306341209+042242406341231+0422301 063412+042242 16.31 063412+042230 18.50 15.35 18.51 16.29 14.97 16.70 14.70 BBB000 14.41 UUA000 UUA000 06341089+0426038 063410+042603 14.31 13.04 12.64 AAA00c 06341079+0429122 063410+042912 16.24 15.29 14.93 AAC000 06341033+0427100 063410+042709 16.24 15.19 14.77 AAAcc0 06340911+042545006340928+0426563 063409+042545 063409+042656 16.39 15.79 15.19 14.63 14.17 14.13 BBA000 AAA000 06340931+0423106 063409+042310 15.79 14.09 13.29 AAA000 13.90 06340935+0434471 063409+043446 12.19 11.59 11.23 AA Table 5.5—Continued 18.93 16.99 06341575+0428126 063415+042812 15.31 14.41 14 18.99 18.01 06341358+0427190 063413+042719 15.88 14.91 14 18.54 17.28 06341355+0427428 063413+042742 14.79 13.90 13 19.03 17.56 06341322+0428257 063413+042825 15.10 14.17 13 19.58 18.07 06341251+0426413 063412+042641 15.01 13.64 13 19.31 18.04 06341228+0424090 063412+042409 15.43 14.36 13 19.20 17.68 06341150+0428027 063411+042802 15.37 14.39 14 18.95 17.95 06341142+0428290 063411+042829 15.61 14.69 14 20.15 18.29 06341123+0426361 063411+042636 15.00 13.72 12 19.84 18.38 06341078+0426528 063410+042652 15.82 14.71 14 19.93 18.18 06341045+0425274 063410+042527 15.53 14.43 13 19.10 17.38 06341038+0427178 063410+042717 15.45 14.37 14 20.32 18.16 06340983+0426377 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0095533 20.32 17.86 16.37 06341611+0428115 063416+042811 0096337 19.75 17.42 15.97 06341601+0435257 063416+043525 0095532 20.62 18.13 16.71 06341594+0427110 063415+042710 0095530 0095525 20.21 18.42 16.18 06341505+0427259 063415+042725 0095522 20.36 17.81 16.24 06341479+0425386 063414+042538 0095512 0095511 0095508 0095507 20.37 18.18 16.77 06341315+0427274 063413+042727 0095501 0095500 0095495 0095494 0095493 0095487 0095477 0095476 0096291 0095472 ······································· ··· ··· ··· ··· ··· ··· ··· ································· ····································································· ··· ··· ··· ································· ································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ·········································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ·································································· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· − − − − − − − − − − − − − − − − − − − − X-ray Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 280281 063416.20+042127.1 282 063416.29+042816.9 063416.48+042938.9 278279 063416.11+041850.8 063416.14+042811.2 0944 277 063415.98+043525.5 0945 276 063415.93+042710.2 0944 272273 063415.27+043033.0 274 063415.43+042251.4 275 063415.57+043423.7 063415.77+042812.4 0944 270271 063414.85+043022.5 063415.07+042725.1 0944 268269 063414.67+043022.5 063414.79+042538.2 0944 267 063413.60+042718.5 0944 265 063413.41+042013.4 266 063413.57+042742.6 0944 264 063413.23+042825.2 0944 262263 063412.55+042711.1 063413.18+042727.1 0944 260261 063412.30+042720.3 063412.52+042641.1 0944 255256 063411.96+042942.8 257 063411.99+042511.0 258 063412.12+042241.9 259 063412.26+042228.5 063412.28+042408.8 0944 254 063411.50+042802.7 0944 251252 063411.30+042255.5 253 063411.38+042847.2 063411.44+042828.9 0944 249250 063410.92+042603.5 063411.25+042635.8 0944 245246 063410.62+043156.3 063410.75+042641.5 247248 063410.81+042911.8 063410.83+042652.1 0944 244 063410.51+042526.9 0944 242243 063410.34+042709.7 063410.39+042717.3 0944 238239 063409.29+042655.8 063409.34+043447.5 0945 236237 063409.00+042744.6 063409.13+042544.5 240241 063409.34+042310.0 063409.83+042637.7 0944 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 168 A000 ··· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· .11 AAA000 .28 AAA000 .22 AAA000 .43 AAAccc .88 AAA000 .01 AAA000 13.41 12.99 12.96 AAA000 14.82 13.78 13.06 AAA000 14.76 13.87 13.59 AAA000 14.48 13.67 13.43 AAA000 14.74 13.88 13.52 AAA000 12.43 11.63 11.30 AAA000 14.39 13.71 13.49 AAA000 14.34 13.41 12.95 AAA000 13.71 13.33 13.24 AAA000 14.74 13.86 13.65 AAA000 13.91 12.78 12.05 AAA000 063424+043012063424+041908 17.30 19.30 16.50 17.25 16.00 16.08 063423+042715 17.95 17.21 17.02 063422+042800 18.58 17.05 16.48 063421+041934 17.94 16.63 15.87 063418+041923 18.24 16.50 15.63 Optical/Infrared Photometry 06342629+0429394 063426+042939 14.46 13.53 13.24 AAAc00 06342584+043019006342578+0420324 063425+043019 063425+042032 14.03 16.18 13.40 14.30 13.17 13.06 AAAcpp AAA000 06342376+0424449 063423+042444 18.4306342481+0418323 16.4606342511+0426536 063424+041832 063425+042653 14.64 17.99 14.17 DBA000 16.01 13.26 14.99 12.98 UUB00c AAA000 06342156+0421285 063421+042128 15.73 14.22 13.38 AAA000 06342204+0430528 063422+04305206342263+0424454 063422+042445 16.48 14.25 15.01 13.14 14.36 12.74 BAA000 AAA000 06341994+0422234 063419+042223 16.78 15.18 14.42 UUA00c 06341944+0421325 063419+042132 15.93 14.31 13.55 AAAccc 06341745+0421549 063417+04215406341794+0426120 063417+042612 16.57 16.01 15.19 14.78 14.51 14.34 BAB000 AABccc 06341652+0429292 063416+042929 16.80 16.01 15.65 BCC000 ··· 18.83 06342203+0424293 063422+042429 15.94 14.71 14.27 AA Table 5.5—Continued 18.82 17.21 06342532+0424062 063425+042406 14.58 13.57 13 19.37 20.57 18.11 06342043+0424050 063420+042405 15.24 13.87 13 19.29 17.98 06342014+0427348 063420+042734 15.50 14.58 14 19.79 18.07 06341735+0425097 063417+042509 15.26 14.03 13 18.27 17.14 06341731+0428120 063417+042811 15.12 14.16 13 19.28 17.62 06341723+0425028 063417+042502 14.57 12.89 12 ··· ··· ··· ··· ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0096383 15.53 13.87 13.35 06342608+0430221 063426+043022 0095604 21.03 18.57 17.20 06342573+0428157 063425+042815 0095603 0096370 20.30 17.75 16.46 06342354+0430125 063423+043012 0095582 0095588 19.53 16.99 16.16 06342288+0429512 063422+042951 0095581 20.41 17.89 17.00 06342198+0426248 063421+042624 0095709 0095578 18.73 15.59 14.45 06342102+0424419 063421+042442 0095570 0095568 0095564 19.15 16.49 15.76 06341992+0427445 063419+042744 0095563 20.36 17.11 16.06 06341954+0427398 063419+042739 0095553 16.34 15.11 14.51 06341798+0424205 063417+042420 0095551 0095550 0095549 0096346 20.05 17.71 16.17 06341708+0431468 063417+043146 0095542 19.52 16.80 15.55 06341689+0424191 063416+042419 ··· ··· ··· ··· ······································· ··· ··· ··· ··· ··· ··· ································· ·············································································· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ········· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ······································· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· − − − − − − − − − − − − − − − − − − − X-ray Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 329 063426.29+042939.1 325326 063425.83+043018.6 327 063425.85+042031.6 328 063425.90+043241.7 063426.08+043021.5 0945 323324 063425.46+042731.3 063425.73+042815.3 0944 315316 063423.78+042444.4 317 063423.97+042116.0 318 063424.42+043011.1 319 063424.51+043155.4 320 063424.64+041909.3 321 063424.85+041832.8 322 063425.12+042653.1 063425.32+042406.1 0944 312313 063423.46+043542.7 314 063423.49+042715.1 063423.52+043012.3 0945 307 063422.05+042428.9 0944 308309 063422.05+043051.6 310 063422.57+042759.9 311 063422.72+042444.0 063422.88+042950.7 0944 305306 063421.72+042452.3 063421.99+042624.3 0944 302303 063421.26+041933.0 304 063421.34+042515.0 063421.57+042127.6 0943 301 063421.03+042441.6 0944 300 063420.43+042404.7 0944 298299 063420.04+042223.5 063420.14+042734.5 0944 297 063419.96+042744.0 0944 294295 063418.67+041920.6 296 063419.46+042131.8 063419.56+042739.2 0944 290291 063417.50+042153.9 292 063417.90+043438.0 293 063417.94+042611.6 063417.97+042419.5 0944 289 063417.37+042509.5 0944 288 063417.31+042811.7 0944 287 063417.24+042502.3 0944 286 063417.11+043146.6 0945 284285 063416.63+042938.9 063416.92+042419.1 0944 283 063416.49+042928.6 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 169 A000 ··· ··· ··· ··· ··· ··· ··· ··· .29 AAA000 .64 AAB000 .99 AAA000 .69 AAA000 .15 AAA000 .70 AAA000 .36 AAA000 18.00 ··· ··· 13.68 12.67 12.26 AAA000 13.99 13.16 12.83 AAA000 14.34 13.34 12.97 AAA000 14.99 14.03 13.69 AAA00s 15.17 14.52 14.39 AAA000 15.01 14.28 14.09 AAA000 14.24 13.37 13.08 AAA000 10.81 10.78 10.69 AAA000 14.25 13.44 13.17 AAA00c 13.34 12.50 12.06 AAA000 12.85 12.12 11.78 AAA000 15.56 14.61 14.30 AAA000 063436+042138 16.45 15.33 14.72 063430+042239 18.64 16.65 15.65 063427+043321 063427+042327 17.70 16.28 15.47 Optical/Infrared Photometry 06343595+0425332 063435+042533 15.92 14.76 14.33 AAA000 06343440+0419005 063434+041900 13.33 12.40 12.07 AAA000 06343333+0422119 063433+042211 15.36 14.11 13.48 AAA000 06343069+042140506343062+0419321 063430+042140 063430+041932 15.95 17.92 14.5706343157+0423172 16.07 063431+042317 13.8906343215+0419455 14.79 063432+041945 16.22 UUA00c UUB00c 14.87 14.70 13.17 14.02 12.29 AAA000 AAA000 06342912+0419273 063429+041927 15.87 14.30 13.56 AAAccc 06342855+0421263 063428+042126 15.30 13.99 13.39 AAA000 06342782+0420568 063427+042056 16.80 15.40 14.80 CBB000 06342662+0425081 063426+042508 15.36 14.11 13.43 AAA000 17.96 06343267+0423210 063432+042321 14.89 13.66 12.94 AA Table 5.5—Continued 19.45 18.40 06343704+0425355 063437+042535 15.87 14.68 14 19.56 18.20 06343440+0430594 063434+043059 15.94 15.03 14 19.20 17.45 06343327+0427522 063433+042752 15.30 14.38 13 20.26 18.68 06342938+0426232 063429+042623 16.19 15.11 14 19.13 18.11 06342889+0428105 063428+042810 15.56 14.68 14 18.99 17.89 06342743+0425180 063427+042518 15.28 14.12 13 19.14 17.58 06342727+0424350 063427+042435 14.88 13.79 13 ··· ··· ··· ··· ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0095865 20.36 17.84 15.90 06343808+0422105 063438+042210 0096464 20.29 17.98 16.09 06343769+0431481 063437+043148 0095655 0095817 20.01 18.35 16.54 06343522+0418481 063435+041848 0096447 0095647 20.59 18.56 17.08 06343360+0428065 063433+042806 0095644 19.14 17.20 16.40 06343348+0425427 063433+042542 0095790 0095639 0095765 19.80 17.76 16.32 06343008+0423434 063430+042343 0095630 20.31 17.22 15.95 06342945+0427244 063429+042724 0095629 0095621 0096405 11.60 11.55 11.54 06342882+0431039 063356+043321 0095617 19.78 16.80 15.73 06342824+0429541 063428+042954 0095738 20.27 17.34 15.65 06342781+0421277 063427+042127 0095612 0095610 0096397 18.12 15.62 14.42 06342724+0435164 063427+043516 0096394 20.01 19.26 17.41 06342686+0431151 063426+043115 ······ ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ································· ································· ···································· ··· ··· ··· ············································································································ ······ ···································· ······································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ·································································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· − − − − − − − − − − − − − − − − − − − − X-ray Source # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 376 063438.09+042209.9 0943 375 063437.67+043147.7 0945 372373 063436.02+042532.0 374 063436.55+042138.8 063437.08+042535.8 0944 369370 063434.41+042323.9 371 063434.49+041900.8 063435.21+041847.4 0943 367368 063434.27+042916.3 063434.39+043059.8 0945 365366 063433.50+042834.2 063433.60+042805.8 0944 362363 063433.33+042242.0 364 063433.33+042210.9 063433.47+042543.0 0944 348349 063430.65+042140.3 350 063430.66+041932.0 351 063430.75+042240.3 352 063431.45+042427.4 353 063431.47+041912.4 354 063431.58+042938.1 355 063431.59+042316.1 356 063431.81+042832.4 357 063432.10+041946.2 358 063432.17+041750.9 063432.69+042322.3 0943 359360 063432.85+041944.4 361 063433.09+042649.0 063433.26+042751.7 0944 347 063430.13+042342.9 0943 346 063429.47+042724.1 0944 344345 063429.17+041926.7 063429.39+042623.2 0944 343 063428.91+042810.9 0944 341342 063428.48+042125.2 063428.83+043103.5 0945 340 063428.22+042953.9 0944 337 063427.61+043321.0 338339 063427.70+042055.3 063427.78+042128.0 0943 334 063427.28+042327.5 335336 063427.41+042547.2 063427.48+042517.2 0944 333 063427.26+042435.2 0944 332 063427.22+043515.5 0945 331 063426.83+043114.8 0945 330 063426.62+042507.7 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 170 A000 ··· ··· .99 AAA000 .43 AAA00s .20 AAA000 Johnson, & Degioia- roege et al. 2006), [VGK85]=Voroshilov 12.18 11.89 11.84 AAA000 16.02 15.33 15.11 AAB000 14.75 13.66 13.26 AAA000 14.35 13.71 13.53 AAA000 063438+042144 16.58 15.18 14.41 Optical/Infrared Photometry m Table 5.1 and Table 5.2. For convenience, [MJD95]=Massey, 06343852+0425029 063438+042502 15.80 14.55 14.05 UAA0cc 16.51 06345161+0431096 063451+043109 15.51 14.80 14.62 AA : type A0V Table 5.5—Continued 19.60 18.33 06343866+0422420 063438+042242 15.54 14.45 13 18.77 17.75 06343909+0428074 063439+042807 15.67 14.77 14 19.16 17.43 06344557+0424223 063445+042422 14.92 13.81 13 ASS4]=TASS Mark IV Photometric Survey of the Northern Sky (D ··· ··· ··· ··· ··· (mag) (mag) (mag) ID ID (mag) (mag) (mag) 0095869 0095670 0095672 13.88 12.72 13.07 06343922+0427312 063439+042731 0096479 20.42 18.54 17.64 06343946+0433483 063439+043348 0095674 21.33 18.67 17.15 06343983+0424464 063439+042446 0095703 0096546 18.24 16.74 15.44 06344963+0431047 063449+043104 0096559 ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ··································································································· ·································································· ·································································· ································· − − − − − − − − reproduce the sequence number and source identification fro Columns 1–2 X-ray Source =[BC02] 136 =HD 46485=[BC02] 138=[TASS4] 663303; spectral type=[TASS4] O8 663323 =[BGD93] IRS 4 =[BGD93] IRS 2 =[BGD93] IRS 5 =[TASS4] 663391=[VGK85] NGC 2244 +04 226;=[TASS4] spectral 1472087=[VGK85] type NGC B9V 2244 +04 229; spectral type A0 =[MJD95] 289 =[MJD95] 278=[TASS4] 712279; V=12.6 =[MJD95] 265 =[MJD95] 237 =[MJD95] 230 =[MJD95] 259 =[MJD95] 256 =[MJD95] 260 =HD 259533=[TASS4] 663214; V=10.6, spectral type=[BC02] G0 135 =[TASS4] 712310=[VGK85] NGC 2244 +04 199; V=13.2; spectral # Note. — 24 25 27 32 42 48 54 57 65 72 95 164 185 207 209 210 342 381 119 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 377 063438.43+042146.2 378379 063438.55+042501.8 063438.76+042242.1 0943 380 063439.10+042807.3 0944 381 063439.22+042732.1 0944 382 063439.35+043348.0 0945 383384 063439.46+042317.9 063439.88+042444.6 0944 385386 063439.92+042527.0 387 063440.41+043344.8 388 063443.25+042351.3 063445.58+042423.2 0944 389390 063446.45+043101.8 391 063447.41+042857.8 063449.51+043105.9 0945 392393 063449.98+042739.4 394 063450.44+042702.2 063451.49+043110.4 0945 395 063452.40+043126.3 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg Eastwood (1995), & [BC02]=Berghöfer Christian (2002), [T et al. (1985), [BGD93]=Block, Geballe, & Dyson (1993). 171 north of PL5 ′ tions 0 keV) contained in . 2 MedE > –Number of IR sources contained in the IR,Ex N Estimated Notes & IR,Ex f –Number of hard X-ray sources ( IR,Ex X,h N N IR N X,h f X,h N –Number of IR sources contained in the region; X 5 4 80 3 1 33 ... MSX6CG206.82-01.97 4 1 25 3 0 0 ...... 6 2 33 3 0 0 ... IRAS06297+0453 –Fraction of IR-excess sources. 17 13 76 14 1 7 ... IRAS06317+0426,3 67 22 33 57 5 9 ... IRAS06314+0427,PL4 14 7 50 11 4 36 ... IRAS06306+0437,PL2 11 6 55 5 0 0 ... MSX6CG206.70-01.99 52 24 46 37 8 22 200 ... 29 5 1723 4 17 ...... 91 27 30 61 6 10 300 ... IR 160 66 41 131 9 7 800 ... N N ′ ′ Table 5.6. X-ray Sampled Stellar Clusters ′ ′ 5 IR,Ex 5 ′ ′ ′ ′ ′ . ′ ′ . 5 5 f . . 4 6 8 3 2 3 8 3 6 3 1 × × × × × × × × × × × ′ ′ ′ ′ ′ ′ ′ ′ ′ ′ ′ 5 5 . . –Fraction of hard X-ray sources; –Number of X-ray detected sources contained in the region; X X,h N f RMC R.A. Dec. Approx. Note. — ....C2 06:34:24 +04:22:00 4 ....C1 06:34:12 +04:28:15 5 C..... 06:34:12 +04:28:15 8 ....B3 06:33:15 +04:35:11 3 ....B2 06:33:14 +04:31:30 2 ....B1 06:32:54 +04:37:00 3 B..... 06:33:15 +04:35:11 4 ....A3 06:32:38 +04:46:18 6 ....A2 06:32:25 +04:39:00 3 ....A1 06:32:24 +04:50:30 1 Structure (J2000) (J2000) Extent (%) (%) Total Pop. Associa A..... 06:32:38 +04:46:18 9 the region; region that show significant K-band excess; 172

12 Fig. 5.1 A 1◦ 1◦ MSX 8.3 µm image overlaid with CO continuum emission contours × (Heyer et al. 2006), outlining the distribution of the ISM in the RMC. The squares with sequence numbers denote the locations of Phelps & Lada (1997) IR clusters (except cluster PL7, which is oﬀ the MSX image). The large boxes outlines the ACIS-I coverage. The embedded clusters are clearly associated with the CO clumps and the dust emission peaks. 173

Fig. 5.2 A Chandra mosaic of the RMC region (0.5-8 keV) overlaid with source extraction regions derived by AE from appropriate Chandra PSFs. Note that these PSFs grow larger in size with increasing off-axis angles. Sources covered by different pointings are represented by polygons of different color (red: ObsID 1875; green: ObsID 1876; blue: ObsID 1877). Note that some sources have two extraction regions as they lie in the overlapped region of two observations. 174

Fig. 5.3 The variable X-ray light curve of ACIS source #119. The K S test finds a − significant variation (P 0.005), but no flaring characterized by fast rise and slow KS ≤ decay as commonly seen in PMS stars is present. 175

Fig. 5.4 (a) The X-ray spectrum and the spectral fit for the O7 star (#164). The best fit adopts a two-temperature thermal plasma model, with soft kT and a low absorb- 21 2 ing column N 4.1 10 cm− . (b) The X-ray spectrum and the spectral fit for H ∼ × the brightest X-ray source in the fields (#119). The spectrum can be fit with a two- 21 temperature thermal plasma model (kT1 = 0.3 keV and kT2 = 2.1 keV, NH 3.2 10 −2 ∼ × cm ). 176

Fig. 5.5 Three-color Spitzer/IRAC+MIPS image of the neighborhood region of source #89. Red, green, and blue represents 24µm, 8µm, and 3.6µm emission, respectively. Overlaid symbols are the X-ray-selected stars that are Class I (magenta circle), Class II (green diamond), and Class III (cyan cross) based on their NIR colors. 177

2 Fig. 5.6 (a) The stellar surface density (log#stars arcmin− ) map for all RMC sources smoothed with a 3′ radius kernel. (b) The same as (a) but using a 2′ radius kernel. (c) The same as (a) but for the unobscured population. (d) The same as (b) but for the unobscured population. (e) The same as (a) but for the obscured population. (f) The same as (b) but for the obscured population. The density scaling is the same between maps of the unobscured population and obscured population for a fair comparison. 178

Fig. 5.7 NIR J H vs. H K color-color diagram for Chandra stars with high-quality − − JHK photometry (error in both J H and H K colors < 0.1 mag) in regions A, B, − − and C as deﬁned in Figure 5.6. The green circles and red triangles represent sources with and without signiﬁcant K-band excess (E(H K) > 2σ(H K)). The black solid − − and long-dash lines denote the loci of MS stars and giants, respectively, from Bessell & Brett (1988). The purple dash dotted line is the locus for classical T Tauri stars from Meyer et al. (1997), and the cyan solid line is the locus for HAeBe stars from Lada & Adams (1992). The blue dashed lines represent the standard reddening vector with crosses marking every AV = 5 mag. 179

Fig. 5.8 NIR J vs. J H color-magnitude diagram using the same sample and symbols − as Figure 5.7. The purple solid line and dashed line is the 2 Myr isochrone and the 1 Myr isochrone for PMS stars from Siess et al. (2000), respectively. The dash dotted line marks the location of Zero Age Main Sequence (ZAMS) stars. The blue dashed lines represent the standard reddening vector with asterisks marking every AV = 5 mag and the corresponding stellar masses are marked. 180

Fig. 5.9 Spatial distribution of all the IR counterparts to our X-ray sources. The dense 12 molecular cloud is outlined by the CO emission contours. Overlaid red polygons are the regions A, B, and C defined by X-ray source densities, and the symbols are the X-ray-selected stars classified as Class I (magenta circle), Class II (green diamond), and Class III (cyan cross) based on their NIR colors. The O7 star and the Class I protostar #89 are labeled. The overall distribution of the Class II/I sources are more confined to the CO molecular ridge than the Class III stars. 181

Fig. 5.10 Surface density contours based on 10th neighbor distances for the Chandra RMC sources. Yellow contours are densities for all candidates. The blue lines are contours for unobscured sources (medE 2.0keV), and the red lines are contours for obscured ≤ sources (medE > 2.0keV). All contours follow the levels of density for obscured sources, 2 2 starting at 0.057 stars arcmin− and steps of 0.45 stars arcmin− . 182

Chapter 6

A Chandra Study of the Triggered Cluster NGC 2237

6.1 Introduction

The formation of the lower mass stars in the vicinity of massive stars is an important but intriguing question (see Gorti & Hollenbach 2002 and Briceno et al. 2007 for recent reviews). In their immediate neighborhood, massive stars apparently play a destructive role: they quickly disrupt and disperse the molecular clouds after their births, surpressing further star formation in general (Herbig 1962). At larger radii from massive stars, they are more constructive to star formation activities; the shocks driven by powerful stellar winds or supernovae in the final stages of massive star evolution are crucial in triggering the collapse of molecular cores (Whitworth et al. 1994; Preibisch & Zinnecker 2001; Lee & Chen 2007; Urquhart et al. 2007). Likely triggered star formation events by massive stars have been observed at different spatial scales, such as W3/W4 (Oey et al. 2005), Cep OB2 (Sicilia-Aguilar et al. 2004), Cep B/OB3b (Getman et al. 2006), IC 1396N (Getman et al. 2007), NGC 6357 (Bohigas et al. 2004; Wang et al. 2007a), M 17 (Jiang et al. 2002; Broos et al. 2007), and 30 Doradus Nebula (Walborn et al. 1999; Townsley et al. 2006a). One particularly interesting example is the Rosette star forming complex (Phelps & Lada 1997; Román-Zúñiga et al. 2007a; Wang et al. 2007b,c). Powered by a number of OB stars, the massive young cluster NGC 2244 (d 1.4 kpc; t 2 Myr) has created a ∼ ∼ visually spectacular expanding HII region known as the Rosette Nebula. As the ionized nebula is clearly interacting with the adjacent Rosette Molecular Cloud (RMC), the well-known sequential formation of OB subgroups (as in M17, M42, and W3; reviewed in Elmegreen & Lada 1977) may have also occurred here. A few embedded clusters in the RMC have already been identified (Phelps & Lada 1997; Román-Zúñiga et al. 2007a; Wang et al. 2007c). To the west of the NGC 2244 cluster, a subcluster or satellite cluster was recently discovered in near-infrared (NIR) studies (Li 2005; Li & Smith 2005c; Román-Zúñiga et al. 2007b). This region was historically designated as NGC 2237, referring to the diffuse emission nebula. In the Digital Sky Survey optical plate (Figure 6.1), the NGC 2237 region is clearly the boundary between the HII region and the cold neutral molecular materials, as evidenced from the numerous elephant trunks pointing towards the O stars in NGC 2244. Its morphology closely resembles the peripheries of other HII regions. Almost no literature exists on the stellar content in this region. Although not noted as a subcluster, a handful of X-ray sources associated with this region were clearly seen in the ROSAT/PSPC X-ray image (0.5–2.0 keV) of the NGC 2244 cluster (Berghöfer & Christian 2002, see Figure 1 in ). Using the spatially complete Two Micron All Sky Survey (2MASS) data, Li (2005) statistically estimated a physical scale of 3.2 pc and a 183 stellar population of 230 stars for this small subcluster. To the south of the NGC 2244 ∼ core, another concentration of NIR excess sources in an arc-like structure were found (Li 2005). These substructures were suggested to be originated from inner or former swept-up layer of the HII region (Li 2005), and NGC 2237 may represent the earliest triggered stellar population near NGC 2244. Román-Zúñiga et al. (2007a) also identified 1 NGC 2237 as a new cluster based on the Nearest Neighbor analysis (Casertano & Hut 1985; Diggle 2003) of their deep FLAMINGOS NIR images. However, these NIR studies rely on color criteria to select YSOs, which is difficult to distinguish unrelated field stars and unreddened cluster members. Low mass young stars are emitting X-rays from violent magnetic reconnection flares, orders of magnitude more luminous than seen in older Galactic stars (Feigelson et al. 2007). Hence an X-ray study can effectively identify the young stars in this region, especially the diskless Class III sources missed in the color selection. X-ray studies of NGC 2244 and the embedded clusters have been reported in Wang et al. (2007b,c), which have shown new knowledge in the emerged from the clean X-ray-selected samples of the stellar populations. In this chapter, we focus on the X-ray sources identified by Chandra X-ray Ob- servatory in this region and study the young stellar population of NGC 2237, about which little is known previously. The goal of this observation is to identify the X-ray emitting young stars and investigate the NIR properties of the counterparts to our X-ray sources and their spatial distributions, seeking clues of the star formation activities here. If there is clear evidence of spatial gradient in the distribution of Class III, II, and I YSOs (from the HII region to the molecular pillars), the timescale and efficiency of the triggering process can be learned and compared to theories of triggered star formation (e.g., Froebrich et al. 2005; Getman et al. 2007).

6.2 Chandra Observations and Data Reduction

The NGC 2237 region was imaged with the Imaging Array of the Chandra Ad- vanced CCD Imaging Spectrometer (ACIS-I), which has a ﬁeld of view 17′ 17′. The ∼ × 20 ks ACIS-I observation was conducted on Feburary 9 2007, in the “Timed Event, Very Faint” mode with 5 pixel 5 pixel event islands (see Table 1 in Wang et al. 2007b for × details of pointing and roll angle). The pointings towards NGC 2244 are also outlined in Figure 6.1. The same customized data reduction and source extraction described in TFM03 and Wang et al. (2007a) is followed (see also Chapter 2). Only a number of bright point sources are visible, which are illustrated in the ACIS-I image (Figure 6.2). A smoothed X-ray image is created with the CIAO (version 3.4) tool csmooth (Ebeling et al. 2006) and shown in Figure 6.3, which gives the relative spectral hardness of the sources. Most of the sources appear dominated with soft band X-ray emission (0.5-2.0 keV). The source ﬁnding and photon extraction procedures are identical to that described in Wang et al.(2007b). The event extraction was made with our customized IDL script

1 2 The Nth nearest neighbor surface density for a star is calculated by ρ = (N 1)/πD , N − N where DN is the distance from this given star to its Nth neighbor star. See Casertano & Hut (1985); Román-Zúñiga et al. (2007a). 184

2 ACIS Extract (version 3.128; hereafter AE, Broos et al. 2002). We rejected sources with > 0.01 likelihood of being a background fluctuation after a careful review of the net counts distribution for all candidate sources. The trimmed source list includes 168 valid sources, which are further divided into a primary list of 130 highly reliable sources (Table 6.1) and a secondary list of 38 tentative sources with 0.001 likelihood of being spurious background fluctuations (reported as ≥ PB, column 13 in Table 6.1 and 6.2). Table 6.1 and 6.2 have formats that are identical to Tables 1 and 2 in Townsley et al. (2006a), Wang et al. (2007a), and Broos et al. (2007). A detailed description of the table columns is given in the table footnotes. The south-east corner of this field was also covered by ObsIDs 1874 and 3750. Six sources (# 160, 163, 164, 166, 167, and 168) are in the overlapped region, of which 163, 166, 167, and 168 have been detected in previous observations. The source event arrival times were compared to that of a uniform light curve model in AE, and five sources (#4, 39, 118, 149, and 152) were found significant variable (PKS < 0.005 in column 15 of Tables 6.1 and 6.2) during the short 20 ks observation. Only one of them (#149) have more than 100 net counts, which is also the brightest X-ray source in this observation. Its light curve shows a typical X-ray flare seen in PMS stars (Wolk et al. 2005), captured in the rising phase near the end of the observation. For relatively bright sources with photometric significance Signif> 2.0 (column 12 in Tables 6.1 and 6.2), the extracted spectra were fit using single temperature apec thermal plasmas (Smith et al. 2001) and power law models subjected to an absorbing 3 column (NH ) of interstellar material with the XSPEC package (version 12.2.1ap, Arnaud 1996). The best-fit model was achieved by the maximum likelihood method (Cash 1979). Abundances of 0.3 Z⊙ were assumed for the automated fitting performed by AE. The single temperature thermal plasma apec model is preferentially used. Because of the limited number of counts in our sources and the adequate fit for the bright sources using apec model, no two-temperature thermal plasma model was invoked. A power law model was only adopted if it represented the data more adequately than the thermal model or the thermal model required nonphysical parameters, provided that the source was not identified with a known stellar counterpart or exhibited X-ray flares as seen in the PMS stars. Spectral analysis results for the 65 sources with Signif & 2.0 are presented in Tables 6.3 (61 sources; thermal plasma fits) and 6.4 (4 sources; power-law fits). The table notes give detailed descriptions of the columns. Best-fit absorbing column densities range 2 from negligible to log N 23.2 cm− , equivalent to a visual absorption of A 100 mag H ∼ V ∼ (Vuong et al. 2003). Temperatures range from kT 0.1 keV to kT = 15 keV (truncated ∼ value to reflect the hardest photons detectable by the ACIS detector). The range of total band (0.5 8 keV) absorption corrected luminosities derived from spectral modeling is − −1 30.0 . log Lt,c . 32.1 ergs s .

2 http://www.astro.psu.edu/xray/docs/TARA/ae users guide.html 3 http://heasarc.gsfc.nasa.gov/docs/software/lheasoft/xanadu/xspec 185

4 To estimate the X-ray sensitivities, we use PIMMS tool , which gives an apparent 1 total band luminosity log L 29.4 ergs s− estimated for the faint 3 counts on-axis t ∼ detection in Table 6.2, assuming a 2 keV plasma temperature and an average AV = 2 1.5 mag visual extinction (corresponding to an absorbing column log N 21.4 cm− ), . H ∼ A conservative estimate for limiting sensitivity of the NGC 2237 observation is log Lt −1 ∼ 30.0 erg s ; the exact value depends on oﬀ-axis location and absorption. This implies the X-ray detections are nearly complete to the solar mass range when estimated from the empirical L M correlation (Preibisch et al. 2005; Telleschi et al. 2007). x − 6.3 Stellar Counterparts to Chandra Sources

Since the NGC 2237 region is not an extensively studied region, we searched optical and infrared catalogs from the literature and recent observations for counterparts identification. In addition to the available catalogs listed in Chapter 5 (USNO, 2MASS, and FLAMINGOS), other related catalogs are: UBV Photometry of NGC2244 (Ogura & Ishida 1981=OI81), UBV photometry of NGC 2244/Mon OB2 (Massey et al. 1995 =MJD95), BVIRHα photometry of NGC 2244 (Berghöfer & Christian 2002 =BC02). The reference frame offsets between the ACIS fields (astrometrically aligned to the Hip- parcos frame using 2MASS sources in the data reduction) and the catalogs are 0.3′′ to OI81, 0.4′′ to MJD95, 0.3′′ to BC02, 0.2′′ to USNO, and 0.2′′ to FLAMINGOS. These offsets were applied before matching sources. We associate ACIS X-ray sources with optical and near IR (ONIR) sources using positional coincidence criteria (Broos et al. 2007). Likely associations between ACIS sources and ONIR sources are reported in Table 6.5. Thanks to the low obscuration in this region and the sensitive FLAMINGOS images, 134 of the 168 ACIS sources (80%) have an ONIR counterpart identified. Although USNO photometry is reported for the optical counterparts, readers are cautioned that it is photographic system with 0.3 ∼ magnitude photometric accuracy (Monet et al. 2003). JHK magnitudes from FLAMIN- GOS photometry are reported for Chandra sources. The SIMBAD and VizieR online catalogs are searched for complementary information. Notes of the matched sources can be found in the table footnotes. The NIR J H vs. H K color-color diagram for 119 Chandra stars is shown − − in Figure 6.4. These are the NIR stars with high-quality JHK photometry (error in both J H and H K colors < 0.1 mag) listed in Table 6.5. Most Chandra sources − − occupy the color space associated with diskless young stars (Class III objects) that are reddened by interstellar extinction. A concentration of cluster members subjected to A 1.5 mag (assuming late type stars) is apparent, approximately centered at V ∼ J H=0.35, H K=0.8. Note that it is also the typical extinction seen towards NGC − − 5 2244. To the right of this reddening band are 7 K-band excess sources , defined as stars that have colors (J H) > 1.7(H K) + 2σ(H K). These are likely PMS stars with − − −

4 Portable, Interactive Multi-Mission Simulator is software for high-energy astrophysicists, written and maintained by Koji Mukai. See http://heasarc.gsfc.nasa.gov/docs/software/ tools/pimms.html 5 These sources are #44, 61, 80, 97, 111, 117, and 121. For the two sources (#111 and #121) that are close to the reddening band, the errors in both J H and H K colors are small, − − 186 circumstellar accretion disks (Class II objects). One outlier among the Class III sources, #86 appears highly obscured with a visual extinction of 20 mag. It only have 4 net ∼ counts, among the faintest detections. It is located in an optically dark region. The ideal way to identify the Class I and II protostars is using Spitzer mid-IR observations. However, no public Spitzer data is available for our region of interest; the observations by IRAC GTO team complete the coverage of the Rosette Nebula region have been scheduled. Figure 6.5 shows the NIR J vs. J H color-magnitude diagram for the same − stars shown in Figure 6.4. The 1 Myr and 2 Myr PMS isochrones (Siess et al. 2000) are shown for reference. Assuming that the NGC 2237 cluster is triggered to form by NGC 2244 thus must be younger than NGC 2244, it is still not possible to draw conclusion on the exact age of this cluster without obtaining spectra of the cluster members; the extinction and age of a star are coupled when inferred from its location in the color- magnitude diagram. Nevertheless most of the stars seem consistent with sharing the properties of the NGC 2244 cluster (an age of 2 Myr and an extinction of 1.5 2 mags). − A few stars have bright J magnitudes, consistent with being early type stars (B and A types) reddened from the ZAMS with A 1.5 2 mag. Their locations in the color-color V ∼ − diagram (Figure 6.4) also indicate early spectral types. These are bright in optical and have been identiﬁed in the limited number of optical surveys (e.g., OI81 and MJD95). In fact, some of these stars have early spectral types found in the literature. #53 is a B1 star (Philip & Egret 1980), the earliest spectral type known in the NGC 2237 region. The presence of a number of B stars are rather interesting. Sources #44 and 61 have masses > 2M⊙ estimated from their locations in the color-magnitude diagram. They are good candidates for Herbig Ae/Be stars and worth follow-up observations. The overall fraction of K-band excess sources is 13% among stars that have greater than one solar mass estimated from their NIR colors and magnitudes (this is also the approximate mass completeness limit for our X-ray-selected stars). This matches well with the K-excess disk fraction of 15 3% reported in FLAMINGOS NIR study of NGC 2237. ± There are 34 ACIS sources without any matched counterparts. These are typically a mixture of newly discovered embedded members, and distant background stars and extragalactic sources. Figure 6.6 shows the spatial distribution of X-ray sources that do not have matched IR counterparts and sources that have hard median photon energies (medE > 2.0 keV). Ten of them are located in the apparent HII region cavity seen in optical. The rest lie in the molecular pillars and the optically darker regions where residual molecular materials are. Such X-ray discovered stars are commonly found in the molecular clouds surrounding other clusters (Getman et al. 2005a; Wang et al. 2007a; Broos et al. 2007). The level of contamination by extragalactic X-ray sources and Galac- tic disk stars is evaluated following the simulations described in Wang et al. (2007a). A dozen sources may be extragalactic, and 10 could be foreground and background stars. ∼ These contaminants constitute 10% of the 168 ACIS sources. ∼

0.015 mag, therefore their K-excess are signiﬁcant. Also see Wang et al. (2007b) for a brief discussion∼ in deﬁning the K-excess. 187

6.4 Properties of the Stellar Population

6.4.1 Spatial Appearance of the NGC 2237 Cluster The central position and physical scale of the NGC 2237 cluster has been estimated in Li (2005) and Román-Zúñiga et al. (2007a), based on the surface densities of NIR sources. Note that the NIR samples are based on NIR color excess, which only traces stars with significant disk emission, while the X-rays mostly trace magnetic activities in PMS stars and do not require presence of disks. Hence our X-ray-sampled NGC 2237 stars represents the bulk of cluster members, which provides a clear view of their spatial distribution and hints of their star formation history. To identify the clustering of the stellar populations, smoothed stellar surface density map is created using a top-hat smoothing kernel with a 0.5′ radius (Li 2005; Wang et al. 2007a,b). Adopting different sampling radii does not have significant effect on the resulting features discussed here. Spatial concentration of the X-ray point sources is clearly centered at R.A.=06h31m, Dec.=04◦58′ (Figure 6.7), which is commensurate with its central position identified in NIR (R.A.=06h30m56s, Dec.=04◦58′30′′; Li 2005) and the approximate center in low resolution ROSAT X-ray images (Berghöfer & Christian 2002). The dimension of the cluster seen in X-rays is 5′( 2pc) in radius, slightly ∼ ∼ elongated towards the west. This is consistent with the reported extent in NIR. In addition, the NIR stellar density map (Li 2005) also shows an extension of substructure in the northwest. It is clearly seen as a second density enhancement in the X-ray map as well.

6.4.2 Distribution of the NIR Counterparts to the Chandra Stars To further investigate the distribution of Class II and Class III sources classified in the color-color diagram (Figure 6.4), we show their spatial distribution in Figure 6.8. The locations of 34 sources without matched counterparts are also shown. The seven Class II sources appear to be aligned along an arc perpendicular to the direction towards the O stars in NGC 2244. A number of the Class III sources are clustered, and the rest are distributed over the field of view. However noticeably the lower quarter and the rightmost part of this region seem lacking in X-ray sources. The CO emission contours suggest that there are some molecular materials in the rightmost region. Stars are probably forming inside the molecular clumps, but the sensitivity of our observation is unable to identify them because of heavy absorption. Figure 6.9 shows the spatial distribution of stars in different mass range. We consider stars with identified B and A spectral types and stars with NIR estimated masses & 2M⊙ as the high mass sample, while stars in mass range M < 2M⊙ are the low mass sample. The high mass stars are spatially concentrated at the cluster center. Note that sources #44, 53, 61 are along the same arc where the Class II sources are (#44 and #61 are K-excess sources). In contrast, while 30% of the low mass stars are clustered, the rest are dispersed along the peripheries of the HII region and inside the dark cloud. Again, the void of sources in the lower bottom and the rightmost portion of the field is worth noting. 188

To summarize, the spatial distribution of the Chandra stars (Figure 6.7) and their NIR counterparts (Figure 6.8) show the following features: a compact elongated cluster previously recognized as NGC 2237, which is 1.5 2.5 pc elongated along east- ∼ × west, parallel to the direction of the possibly-triggering NGC 2244 O stars; a broader distribution of stars over a 4 4 pc region to the north-west of the cluster (in the optical × dark pillar region and with little change at the photon-dominated region), a remarkable lack of stars 1-3 pc south of the cluster, and a lack of stars 2-4 pc west of the cluster which can be attributed to absorption from the residual GMC. Combining the spatial distribution and properties of the NIR counterparts to our Chandra stars do not provide a simple and coherent star formation picture of this region. Similar substructures in the massive clusters have been seen in the NGC 6357 (Wang et al. 2007a), M 17 (Broos et al. 2007), and NGC 2244 (Wang et al. 2007b), but the number of stars in those substructures are less significant. This satellite cluster contains nearly ten intermediate to high mass stars, and has a population on the order of 200. In terms of obscuration, population, and relative location to the NGC 2244 cluster, it is very similar to the X-ray discovered RMC A cluster near the rim of RMC, which has an estimated population of 300 stars ∼ (Wang et al. 2007c). It is plausible that pre-existing massive molecular clumps (inhomogeneities in the turbulent natal cloud that formed NGC 2244) also collapsed and formed stars in RMC A and NGC 2237, accompanying the formation of the NGC 2244 cluster. They represent the early star formation episode and probably are indistinguishable in age from NGC 2244. Further towards the molecular pillars protruding into the HII region, the passing over of ionization front from the OB stars in NGC 2244 may have triggered some star formation, although the evidence for triggering in photoevaporating pillars is tantalizing. The alignment of the Class II sources is possible hint that the ionization front is triggering the latest star formation along the boundary of HII region. However, this is based on small number statistics. Dynamical instability (Garcia-Segura & Franco 1996) could be also at work here but not effective; only a handful of new stars are forming at the vertices of the elephant trunks, in great contrast to the spectacular association of EGGs at the tip of molecular pillars in M16 (Hester et al. 1996). Note that the low counts of X-ray detected pillar object could be due to the decreased sensitivity of our observation in this region and intrinsic low X-ray luminosity (low mass) of the protostars in the EGGs. Linsky et al. (2007) report X-ray non-detection of EGGs in M16 and conclude that either the EGGs do not contain protostars or the protostars have not yet become X-ray active. More X-ray sources are seen distributed ahead of the ionization front that account for the density enhancement northwest of NGC 2237 seen in Figure 6.7. Most of these X-ray sources show hard X-ray emission, suggesting that they are embedded young stars. A previously unknown optical outflow feature is seen originated inside of the optically dark region, which may be a newly identified Herbig Haro (HH) object and supports the ongoing formation of stars in this region. These stars may have formed as a result of interaction between the molecular materials here and the shock front from the Rosette Nebula. Spectroscopy and Spitzer mid-IR data can determine their ages and evolutionary stages. 189

6.4.3 Notes on Interesting Sources The X-ray spectra of two stars with over 100 net counts are shown in (Figure 6.10). #149 is the X-ray source with the largest number of counts in this 20 ks observation. The single temperature thermal plasma fit provides an absorption column of log NH −2 ∼ 21.3 cm and a plasma temperature of kT = 4.1 keV. Because most of the photons arrive during the flaring stage and there are only a limited number of photons, we did not attempt separate fits for the quiescent and the flare phase. The high kT value is caused by the X-ray flare seen near the end of the observation. It is known from previous observations that spectral hardening is commonly seen during the X-ray flaring in PMS stars (e.g., Preibisch et al. 2005). The absorption corrected total band (0.5-8 keV) X-ray −1 luminosity log Lt,c 31.3 ergs . Berghöfer & Christian (2002) reported ROSAT/PSPC ∼ −1 detected X-ray luminosity (0.5-2 keV) log Lx = 31.01 erg s , which is in line with our observation when considering the different energy range and the X-ray flaring. #54 is another source that has over 100 counts to derive spectral information 2 reliably. The thermal plasma model suggests a low absorption of log N 20.0 cm− H ∼ and a plasma temperature of kT = 1.2 keV. The absorption corrected total band (0.5-8 1 keV) X-ray luminosity log L 31.0 erg s− . It was previously identified as a B2V star t,c ∼ (Philip & Egret 1980). The star with known earliest spectral type in the field is source #53 (B1V). It 1 only has 16 net counts and the spectral fitting gives log L 30.2 erg s− . The kT t,c ∼ and Lt,c of #53 and #54 are in the range of other early B-type stars seen in the Orion Nebula Cluster (Stelzer et al. 2005). The Hα image of a highly interesting region taken at KPNO (Li & Rector 2004) is shown in Figure 6.11. A number of X-ray stars are at or near the tip of elephant trunks (outlined by circles), which are probably very young. These are unlikely foreground stars projected onto the dark columns, since the morphology of molecular pillars around these stars are often affected by the ultraviolet irradiation from the stars. The most obscured source #86 is embedded in or behind the pillar object (see 6.3). No further information § is available about its X-ray luminosity because of its very limited counts (4 net counts). Li (2005) also noted one highly obscured 2MASS source with extinction A 17 mag V ∼ located behind the same pillar object, however no matching X-ray source was detected in our observation. Another interesting feature in this region is a bright visually identified outflow (outlined by a box), which was not reported previously in the literature. This may be a candidate HH object. There is a star 30′′northwest of the outflow, which is in the oppoiste direction of the outflow feature and possibly responsible for driving the outflow. No X-rays are detected from the candidate ionizing star or from the jet. This is additional evidence of active star forming activity further inside the pillars.

6.5 Summary

We present the ﬁrst high spatial resolution X-ray images of the NGC 2237 cluster in the Rosette Nebula obtained via a single Chandra observation. Our main ﬁndings are: 1. In this 20 ks observation, 168 X-ray point sources are detected with a limiting 30 1 X-ray sensitivity of L 10 ergs s− , nearly complete to the solar mass range. A total of t ∼ 190

134 ONIR counterparts are matched to the X-ray sources. We estimate at most 10% are extragalactic and galactic contamination. The rest of the X-ray sources without ONIR counterparts are likely young NGC 2237 stars deeply embedded in molecular materials. Our X-ray sample provides the first probe of the low mass population in this recently discovered satellite cluster to NGC 2244. 2. The locations of most ACIS sources in the color-magnitude diagram are consistent with a small population of PMS low mass stars (M . 2M⊙) in the NGC 2237 region with a visual extinction of 1 . AV . 2 at 1.4 kpc, assuming a similar age to NGC 2244 (2 Myr old). We derive an overall K-excess disk frequency of 13% for stars with mass M & 1M⊙ using the X-ray selected sample, consistent with the reported disk fraction from NIR study. 3. The central position and extent of the X-ray-sampled NGC 2237 cluster agree well with previous NIR studies. The distribution of Chandra stars do not provide a simple and coherent star formation picture of this region. Similar to RMC A, collapse of pre-existing massive molecular clumps accompanying the formation of the NGC 2244 cluster may have formed NGC 2237 as a satellite cluster. They represent the early star formation episode in the Rosette region. There are little evidence suggestive of a triggered formation process in the optical dark pillars by the NGC 2244 O stars, including discovery of young Class II sources aligned in an arc, a handful of X-ray stars at or near the tip of the elephant trunks, and embedded hard X-ray sources across the optical photodominated region. 4. We examine the X-ray properties of a number of B stars in the field. Several X- ray emitting stars are located at the peripheries of pillar objects. A previously unknown optical outflow feature is seen originated inside of the optically dark region, which may be a newly identified HH object and supports the ongoing formation of stars in this region. median E 8.8 0.7 7.6 1.3 11.8 1.4 10.5 3.4 11.4 1.1 13.9 2.4 14.1 1.7 13.7 1.1 12.9 1.7 13.9 1.5 16.4 1.0 14.4 1.7 16.4 3.4 16.1 1.7 16.4 2.3 17.4 0.9 16.0 1.1 17.5 1.5 15.0 1.5 (ks) (keV) (16) (17) 191 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom. Var. Eff. Exp. B -5 g... -5 g... -5 .... a 14.9 1.6 -5 g... -5 .... a 16.6 1.2 -5 .... a 16.3 2.9 -5 .... a 17.1 3.8 -5 .... a 17.5 4.1 -5 g... -5 g... -5 .... a 16.0 1.7 -5 g... -5 .... c 16.7 1.2 -5 g... -5 g... -5 .... a 17.4 1.6 -5 .... a 17.7 2.5 -5 .... a 16.6 4.8 -5 .... a 18.2 1.4 -5 g... -5 .... a 18.1 1.1 -5 .... a 14.7 2.4 -5 g... -5 g... -5 .... a 18.3 2.0 -5 g... -5 .... a 15.2 3.7 P < < < < < < < < < < < < < < < < < < < < < < < < < < < PSF Frac. Signif. log 7.82.47.3 0.90 0.90 0.89 2.0 1.9 1.8 -3.9 -3.3 -3.2 ...... a c a 15.4 14.8 15.4 4.3 1.7 2.6 h,net counts) C 2 30.1 0.91 4.8 8 2.2 0.90 2.7 9 6.5 0.77 1.9 -4.5 g... 2 7.1 0.90 2.2 6 3.2 0.90 1.7 -3.4 .... a 15.2 1.8 89 2.4 2.3 0.88 0.91 1.6 1.4 -4.7 -3.4 .... g... a 17.5 1.7 7 1.5 0.90 2.0 0 5.2 0.91 1.6 -4.0 g... 8754 0.0 4.5 0.9 3.0 0.91 0.91 0.91 0.90 1.5 1.5 1.8 1.5 -3.8 -3.9 -4.5 .... -3.2 ...... a a a a 17.0 16.7 16.5 16.2 1.1 4.0 1.3 2.3 4 2.7 0.90 2.1 3 2.7 0.89 1.9 3 6.8 0.90 2.1 5 7.6 0.90 1.9 2 0.8 0.89 1.2 -4.2 g... 02 2.2 1.9 0.67 0.89 1.4 2.0 -3.1 .... a 15.8 2.0 4 2.7 0.89 1.8 52 3.6 0.0 0.90 0.89 1.5 1.0 -4.6 -3.0 .... g... a 17.3 2.1 .4 1.1 0.90 2.0 -3.4 g... .5 4.1 0.85 2.4 .3 5.1 0.90 3.4 .4 2.1 0.90 4.7 .7 3.5 0.90 2.1 -4.5 g... .3 8.0 0.90 2.5 .9 13.4 0.91 2.8 .1 15.8 0.90 5.6 .9 0.4 0.90 3.0 .7 5.9 0.90 3.4 .5 6.3 0.89 2.0 -3.3 .... a 14.1 2.4 .7 11.5 0.91 7.8 .0 6.3 0.91 2.3 .4 2.7 0.89 3.6 .0 1.3 0.89 3.1 .0 8.5 0.85 2.6 .2 3.9 0.89 3.2 .1 3.9 0.89 3.6 .7 8.3 0.89 2.5 t B Extraction 2 0.3 10.8 0.90 9.8 t,net σ t,net Catalog: Basic Source Properties θ C Chandra Position (J2000.0) Error δ Table 6.1. Primary (deg) (deg) (arcsec) (arcmin) (counts) (counts) (counts) ( (J2000.0) α Source 2 063016.82+050452.2 97.570122 5.081179 1.4 10.0 10.6 4.7 6 34 063018.34+045953.95 063020.40+050351.76 063022.82+045538.2 97.576458 063023.75+050555.8 97.585038 97.595104 4.998324 97.598962 5.064382 4.927297 5.098853 1.0 1.2 1.0 0.8 8.0 8.7 7.8 9.4 9.8 9.2 8.2 36.8 4.3 4.3 4.0 7.1 4.2 4.8 3.8 6. 8 063025.39+050540.5 97.605833 5.094589 1.0 8.9 15.2 5.0 4. 10 063026.78+045341.5 97.611598 4.894866 0.9 8.1 8.1 3.7 1. 11 063027.33+045347.7 97.613897 4.896602 0.8 7.9 11.5 4.3 2 12 063028.13+050530.4 97.617210 5.091779 0.8 8.3 20.7 5.5 4 14 063028.66+050137.8 97.619421 5.027193 0.3 5.9 32.6 6.3 1 16 063028.95+050512.4 97.620662 5.086801 1.0 8.0 10.3 4.3 3 17 063029.28+050113.2 97.622006 5.020345 0.5 5.6 9.8 3.8 1. 1819 063029.46+050412.0 063029.51+050053.1 97.622768 97.622997 5.070019 5.014759 0.9 0.5 7.1 5.4 7.4 11.7 3.7 4.1 2. 1 20 063030.14+045949.3 97.625604 4.997038 0.4 5.0 14.1 4.4 0 2123 063030.17+045943.4 063030.88+050050.0 97.625713 97.628669 4.995407 5.013916 0.5 0.6 5.0 5.1 6.2 5.1 3.2 3.0 0. 0. 25 063032.56+050616.9 97.635685 5.104705 0.5 8.3 44.9 7.4 3 28 063032.97+050142.3 97.637400 5.028418 0.4 4.9 15.1 4.5 0 30 063033.26+050335.8 97.638616 5.059956 0.4 6.0 19.3 5.1 1 3132 063033.47+045052.1 063033.49+050142.0 97.639484 97.639552 4.847833 5.028351 1.2 0.5 9.4 4.8 10.5 8.3 4.7 3.5 6 0. 33 063033.87+050236.5 97.641126 5.043490 0.6 5.3 6.0 3.2 1. 3435 063034.06+045651.036 063034.15+045728.738 063034.26+050319.639 97.641920 063034.68+050337.2 97.642331 063034.78+045641.0 97.642787 4.947507 97.644504 4.957996 97.644945 5.055466 5.060339 0.6 4.944724 0.6 0.6 0.8 0.2 4.7 4.4 5.7 5.8 4.7 5.2 5.3 7.5 5.6 76.3 3.0 3.0 3.5 3.2 9.3 0. 0. 1. 1. 0 40 063034.93+045826.2 97.645553 4.973952 0.4 3.9 8.6 3.5 0. 41 063035.67+045535.8 97.648644 4.926631 0.5 5.2 10.0 3.8 1 43 063036.58+050043.9 97.652423 5.012199 0.3 3.7 7.7 3.4 0. 45 063037.33+050005.5 97.655565 5.001548 0.3 3.3 8.7 3.5 0. 46 063037.40+050228.1 97.655856 5.041158 0.4 4.5 7.5 3.4 0. 51 063039.44+045636.6 97.664370 4.943518 0.2 3.8 20.6 5.1 0 52 063039.60+045952.7 97.665031 4.997973 0.4 2.7 3.8 2.5 0. 53 063039.77+045439.6 97.665738 4.911022 0.4 5.3 16.0 4.6 1 54 063040.03+045759.4 97.666792 4.966507 0.1 2.9 114.7 11. 55 063040.51+050606.6 97.668796 5.101842 0.7 7.2 13.0 4.4 2 56 063040.98+045730.9 97.670769 4.958596 0.2 2.9 16.8 4.6 0 5759 063041.03+050605.4 063041.97+050059.6 97.670990 97.674908 5.101510 5.016560 0.9 0.3 7.2 2.7 5.0 7.8 3.0 3.4 1. 0. 60 063042.63+045620.5 97.677661 4.939030 0.3 3.5 6.6 3.2 0. 6164 063042.78+045533.0 063043.21+050227.4 97.678273 97.680052 4.925836 5.040955 0.5 0.6 4.2 3.6 5.5 2.8 3.0 2.3 0. 0. 65 063044.14+050221.2 97.683941 5.039226 0.2 3.4 19.9 5.0 0 66 063044.51+045131.1 97.685466 4.858657 0.8 7.9 13.3 4.7 3 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq. No. CXOU J median E 9.3 1.3 8.5 1.3 7.6 1.1 7.3 1.6 12.4 1.0 14.2 2.9 14.0 2.6 12.9 1.2 16.5 1.5 16.2 1.3 13.2 1.8 19.4 1.3 13.1 2.1 16.1 1.9 15.6 1.7 15.9 1.5 13.0 1.2 15.7 1.2 15.2 1.1 (ks) (keV) (16) (17) 192 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Characteristics Anom. Var. Eff. Exp. B -5 g... -5 .... a 18.9 4.3 -5 .... a 18.4 1.6 -5 g... -5 g... -5 .... a 18.7 1.6 -5 g... -5 .... a 14.4 1.6 -5 .... a 19.2 1.1 -5 g... -5 .... b 17.2 2.4 -5 g... -5 .... a 18.4 1.5 -5 g... -5 g... -5 .... b 18.7 2.9 -5 .... a 18.8 1.8 -5 g... -5 .... a 19.0 1.4 -5 .... a 18.8 1.0 -5 .... a 18.1 2.0 -5 .... a 17.9 2.9 -5 .... b 17.5 1.7 -5 .... a 14.5 1.5 -5 g... -5 .... a 17.5 1.3 -5 .... a 17.7 1.0 -5 .... a 17.6 1.5 -5 .... a 18.0 1.0 -5 g... -5 g... -5 g... -5 .... a 18.2 1.7 P < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < PSF Frac. Signif. log h,net counts) C 1 0.9 0.90 1.2 2 5.8 0.89 1.6 4 1.7 0.89 2.3 1 4.9 0.90 1.8 43 0.0 0.8 0.90 0.89 1.4 1.4 -4.5 -4.6 ...... a a 18.2 18.3 0.9 1.2 2 0.9 0.90 1.4 1 0.0 0.90 1.0 -3.4 g... 2 1.8 0.90 1.6 1 0.0 0.90 1.0 -4.1 g... 1 1.9 0.90 1.6 1 0.9 0.89 1.0 -4.2 g... 1 1.9 0.90 1.2 -4.9 .... a 19.0 1.7 1 2.9 0.90 2.2 2 2.9 0.89 1.6 2 2.9 0.90 1.6 2 0.0 0.90 1.0 -3.1 .... 72 1.8 0.8 0.90 0.90 1.8 1.4 -4.2 .... a 16.2 1.4 3 0.8 0.89 1.6 .1 10.9 0.90 3.2 .2 16.8 0.90 5.9 .1 5.0 0.90 3.9 .1 4.2 0.90 3.2 .7 13.5 0.90 3.9 .2 28.9 0.90 8.5 .4 1.7 0.89 3.2 .5 5.5 0.90 2.3 .2 8.9 0.90 3.1 .3 2.7 0.89 1.6 .3 1.8 0.89 1.4 -4.9 g... .2 0.9 0.90 1.6 .3 0.8 0.90 1.6 .2 0.8 0.89 1.4 .2 0.0 0.90 1.6 .2 3.8 0.90 2.0 .2 0.8 0.90 1.2 -3.9 g... .2 1.9 0.89 2.1 .2 2.8 0.90 2.0 .2.5 0.0 4.7 0.89 0.90 1.2 1.3 -4.1 -3.8 ...... a a 18.2 17.5 1.3 2.9 0.3 12.8 0.90 3.8 0.6 4.6 0.89 3.0 3.3 7.6 0.90 4.3 0.2 2.8 0.89 2.6 3.3 5.7 0.90 2.2 -5.0 .... a 14.2 2.2 t B Extraction t,net σ t,net θ C Table 6.1—Continued Position (J2000.0) Error δ (deg) (deg) (arcsec) (arcmin) (counts) (counts) (counts) ( (J2000.0) α Source 67 063044.98+045859.8 97.687428 4.983293 0.3 1.4 3.9 2.5 0. 68 063045.20+045726.8 97.688355 4.957452 0.3 2.2 5.8 3.0 0. 69 063045.43+050234.3 97.689312 5.042869 0.3 3.5 9.6 3.7 0. 70 063045.43+045823.2 97.689314 4.973135 0.2 1.5 6.9 3.2 0. 7172 063045.56+045556.273 063046.04+050235.3 063046.15+045838.8 97.689864 97.691850 97.692318 4.932305 5.043142 4.977449 0.4 0.4 0.1 3.5 3.5 1.2 4.6 4.7 16.9 2.8 2.8 4.6 0. 0. 0 74 063046.18+050151.3 97.692424 5.030929 0.1 2.8 46.8 7.4 0 75 063046.77+045904.2 97.694892 4.984502 0.1 0.9 22.9 5.3 0 76 063046.82+045033.0 97.695106 4.842502 0.9 8.8 19.9 5.8 7 77 063047.85+050000.2 97.699397 5.000058 0.2 0.9 4.8 2.8 0. 78 063048.22+045735.8 97.700918 4.959970 0.4 1.8 2.9 2.3 0. 79 063048.38+045638.4 97.701593 4.944013 0.3 2.7 5.8 3.0 0. 80 063049.17+045750.7 97.704880 4.964090 0.3 1.5 2.9 2.3 0. 81 063049.40+050423.1 97.705847 5.073090 0.3 5.1 23.3 5.4 0 82 063049.54+045743.8 97.706440 4.962190 0.2 1.6 5.9 3.0 0. 83 063049.72+045822.5 97.707201 4.972944 0.1 0.9 88.8 9.9 0 85 063050.08+045716.1 97.708670 4.954478 0.4 2.0 2.9 2.3 0. 8687 063050.33+050028.7 063050.35+045535.4 97.709720 97.709823 5.007985 4.926517 0.3 0.2 1.2 3.7 3.9 16.6 2.5 4.6 0. 0 88 063050.84+045902.3 97.711848 4.983999 0.2 0.3 8.9 3.5 0. 90 063051.52+050133.8 97.714680 5.026059 0.3 2.3 5.8 3.0 0. 91 063051.81+050047.7 97.715905 5.013258 0.2 1.6 5.8 3.0 0. 93 063051.93+045806.7 97.716385 4.968533 0.3 1.2 2.8 2.3 0. 94 063052.71+045144.3 97.719630 4.862310 0.9 7.6 11.5 4.4 3 9597 063052.93+050539.0 063053.84+045706.2 97.720567 97.724364 5.094187 4.951746 0.8 0.3 6.4 2.4 7.3 4.8 3.5 2.8 1. 0. 98 063054.20+045648.8 97.725873 4.946902 0.3 2.7 5.7 3.0 0. 99 063054.68+045904.1 97.727854 4.984497 0.1 1.1 15.8 4.5 0 100 063054.76+050242.1 97.728184 5.045038 0.2 3.6 21.7 5.2 101 063055.16+050327.1 97.729835 5.057531 0.3 4.3 15.4 4.5 102 063055.45+050637.6 97.731060 5.110463 0.5 7.5 28.7 6.2 103 063055.47+045605.6 97.731154 4.934902 0.4 3.4 5.7 3.0 0 104 063055.80+045621.1 97.732502 4.939202 0.4 3.2 4.7 2.8 0 105 063055.83+045818.6 97.732630 4.971836 0.2 1.7 5.8 3.0 0 106 063056.01+045715.7 97.733384 4.954377 0.3 2.5 5.7 3.0 0 107 063056.08+045705.6 97.733692 4.951575 0.3 2.6 4.8 2.8 0 108 063056.28+045832.0 97.734516 4.975560 0.2 1.7 5.8 3.0 0 109 063056.42+045715.8 97.735119 4.954391 0.3 2.5 7.8 3.4 0 110 063056.44+045619.8 97.735186 4.938850 0.3 3.3 11.8 4.0 112 063056.89+045645.0 97.737053 4.945835 0.4 3.0 3.8 2.5 0 113 063057.10+045721.7 97.737922 4.956040 0.2 2.6 8.8 3.5 0 114 063057.25+045832.0 97.738563 4.975560 0.2 1.9 7.8 3.4 0 115116 063057.70+050112.6117 063057.85+045143.0 063058.02+050254.4 97.740417 97.741082 97.741791 5.020182 4.861963 5.048472 0.4 1.0 0.5 2.7 7.8 4.1 3.8 10.7 4.5 2.5 4.3 2.8 0 0 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Seq. No. CXOU J ); KS e local Effective < P median E 05 . 9.1 1.4 16.5 1.5 15.5 1.9 12.8 2.0 12.9 1.5 10.8 1.9 (ks) (keV) (16) (17) 193 Column 16: Fraction of the PSF (at ector (FRACEXPO from eV); average of the upper ··· ··· ··· ··· ··· ··· Estimated random component of Photometric significance computed as Column 11: Characteristics Anom. Var. Eff. Exp. eV). Column 5: B -5 .... a 18.9 1.4 -5 .... a 18.0 1.5 -5 .... c 18.5 1.1 -5 .... a 16.6 1.7 -5 .... a 18.7 1.4 -5 .... a 18.7 1.5 -5 .... b 18.6 1.5 -5 .... a 15.2 1.6 -5 .... b 17.9 1.8 -5 .... a 16.7 4.0 -5 .... a 18.3 1.2 -5 g... -5 .... a 16.6 1.4 -5 .... a 17.2 1.1 -5 .... c 17.4 1.6 -5 .... a 17.7 1.1 -5 .... a 16.7 1.4 -5 .... a 16.9 2.0 -5 .... c 13.2 1.3 -5 g... -5 .... a 14.5 1.5 -5 g... -5 g... P < < < < < < < < < < < < < < < < < < < < < < < Column 12: l band): a = no evidence for variability (0 values above the 1% threshold that defines the catalog becaus s or for sources in chip gaps or on field edges. B P PSF Frac. Signif. log Background-corrected median photon energy (total band). h,net Estimated net counts extracted in total energy band (0.5–8 k counts) C Source anomalies: g = fractional time that source was on a det e that the source is in a crowded region. .2 0.0 0.89 1.4 .2 2.8 0.89 1.8 .4 0.7 0.89 1.4 -4.4 g... .2 1.8 0.89 2.1 .3 0.8 0.89 1.8 .3.3 0.8 2.8 0.90 0.89 1.2 1.2 -3.4 -3.4 ...... a a 18.1 18.2 1.1 4.7 .4.4 0.0 0.9 0.89 0.90 1.4 1.7 -4.3 -4.0 ...... a a 18.0 16.9 1.4 1.5 .6 0.6 0.89 1.3 -3.5 .... a 17.1 1.0 .9 5.4 0.90 1.8 .2.3.6 0.2 6.1 3.6 0.89 0.90 0.89 1.5 1.5 1.9 -3.6 -3.5 ...... a a 15.6 16.3 1.1 6.2 .0.6.1 5.4 0.4 7.4 0.89 0.90 0.90 1.9 2.0 2.0 -3.5 -4.0 .... -4.9 ...... a a a 14.8 13.9 14.8 2.3 1.3 3.4 .8 6.1 0.90 1.8 -3.8 .... a 14.2 4.4 Column 17: 0.2 2.9 0.89 4.1 0.7 12.6 0.89 5.2 0.2 6.8 0.89 3.7 3.0 2.9 0.89 2.3 0.5 3.7 0.89 2.4 0.3 6.8 0.89 6.3 1.5 4.9 0.90 3.1 0.8 1.4 0.91 2.4 0.5 2.6 0.89 3.2 0.8 2.5 0.90 2.7 1.0 17.4 0.89 4.8 8.2 1.0 0.90 2.6 1.0 1.3 0.90 2.7 3.04.8 1.9 5.0 0.90 0.90 2.1 3.8 -4.8 .... a 15.2 1.3 7.2 6.4 0.91 2.1 -3.7 g... 7.66.6 7.0 1.7 0.91 0.89 2.2 2.5 -3.9 .... a 12.7 2.8 4.7 5.5 0.90 2.3 t Right ascension and declination for epoch J2000.0. B Extraction Columns 7,8: Column 14: .1 0.7 47.5 0.89 10.6 Estimated net counts extracted in the hard energy band (2–8 k Variability characterization based on K-S statistic (tota t,net σ Columns 3,4: ly from background. Some sources have Off-axis angle. t,net Column 10: Column 15: Column 6: IAU designation. . θ C 005). No value is reported for sources with fewer than 4 count out streak. . 0 Table 6.1—Continued < KS Column 2: P bserved on axis to obtain the reported number of counts. a reduced PSF fraction (significantly below 90%) may indicat eration after sources are removed from the catalog. Position (J2000.0) Error δ Background counts extracted (total band). # of counts extracted √ Log probability that extracted counts (total band) are sole 05); c = definitely variable ( . (deg) (deg) (arcsec) (arcmin) (counts) (counts) (counts) ( 0 (J2000.0) α < Column 9: KS standard deviation of PSF inside extraction region < P Column 13: . 005 X-ray catalog sequence number, sorted by RA. . , computed as σ Source errors on column 7. 9 ; e = source on field edge; p = source piled up; s = source on read . 0 Column 1: σ < net counts 118 063058.19+050019.5 97.742490 5.005444 0.1 2.2 24.8 5.5 119 063058.43+045503.7 97.743469 4.917720 0.2 4.7 38.3 6.8 121 063058.59+045912.0 97.744147 4.986671 0.3 2.1 4.8 2.8 0 122 063058.68+045740.7 97.744530 4.961312 0.3 2.6 6.8 3.2 0 123 063059.47+045614.7 97.747821 4.937443 0.5 3.8 4.6 2.8 0 124 063059.88+045846.4 97.749515 4.979561 0.2 2.4 8.8 3.5 0 125 063100.05+045932.7 97.750225 4.992438 0.3 2.4 6.7 3.2 0 126 063100.55+045807.2 97.752315 4.968683 0.2 2.8 20.8 5.1 127 063100.71+050614.4 97.752976 5.104026 0.8 7.4 11.0 4.3 128 063101.33+045610.3 97.755571 4.936199 0.3 4.2 10.5 3.8 129 063101.54+045734.3 97.756436 4.959534 0.1 3.3 52.7 7.8 130133 063102.14+045711.5134 063102.81+045750.6 063102.91+050348.6 97.758933 97.761723 97.762143 4.953214 4.964079 5.063502 0.5 0.5 0.4 3.6 3.4 5.5 3.7 3.7 16.5 2.5 2.5 4.8 0 0 136137 063103.22+045728.2138 063103.39+045442.2 063104.57+045543.5 97.763455 97.764140 97.769059 4.957840 4.911739 4.928762 0.4 0.7 0.4 3.7 5.6 5.0 4.6 6.6 11.2 2.8 3.4 4.0 0 1 140141 063105.38+050130.3 063105.49+045749.3 97.772420 97.772913 5.025102 4.963709 0.6 0.3 4.4 4.1 4.4 16.5 2.8 4.6 0 150 063108.44+050126.5 97.785196 5.024051 0.4 5.0 13.2 4.3 142 063105.57+045525.4 97.773231 4.923727 0.3 5.4 33.0 6.3 151 063109.16+050109.1 97.788202 5.019214 0.6 5.1 7.1 3.4 0 145147 063107.14+045525.6148 063107.77+050248.7 063107.82+045740.8 97.779791 97.782416 97.782598 4.923780 5.046870 4.961343 0.7 0.7 0.5 5.7 5.6 4.7 5.8 5.7 7.4 3.2 3.2 3.4 1 1 0 152 063109.95+050745.9 97.791462 5.129425 1.1 9.8 15.8 5.5 154 063110.97+045546.5 97.795725 4.929585 0.6 6.2 13.0 4.3 149 063108.03+045930.3 97.783496 4.991772 0.1 4.4 134.3 12 156160 063112.02+050441.3161 063115.18+045439.2162 063115.47+050215.1163 97.800097 063117.02+045744.5 97.813270 063118.15+045513.6 97.814463 5.078145 97.820939 4.910909 97.825645 5.037544 4.962363 0.9 4.920467 0.9 0.8 0.8 0.6 7.6 7.7 6.9 6.8 8.0 9.0 9.4 10.0 8.9 25.2 4.1 4.1 4.1 3.8 6.0 4 3 2 164 063119.52+045310.1 97.831342 4.886157 1.2 9.5 11.8 4.9 165166 063120.08+045824.3167 063120.87+045326.1 063121.42+045406.9 97.833699 97.836976 97.839283 4.973431 4.890610 4.901928 1.0 1.2 1.1 7.5 9.6 9.3 8.2 12.4 14.4 3.9 5.1 5.1 2 168 063125.31+045640.0 97.855464 4.944456 1.2 9.1 12.3 4.7 ) is (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Note. — Seq. No. CXOU J upper error on net counts background estimates can rise during the final extraction it position error, 1 1.497 keV) enclosed within the extraction region. Note that mkarf exposure time: approximate time the source would have to be o b = possibly variable (0 and lower 1 ); KS e local Effective < P median E 05 . 8.0 2.8 12.9 1.6 17.2 3.4 19.0 2.1 17.7 1.0 15.7 1.4 18.0 1.7 15.0 1.8 13.9 1.0 15.5 1.7 17.2 1.0 16.8 2.5 15.5 1.0 17.6 1.5 15.4 4.0 14.5 1.3 (ks) (keV) (16) (17) 194 Column 16: Fraction of the PSF (at ector (FRACEXPO from eV); average of the upper ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· Estimated random component of Photometric significance computed as Column 11: Characteristics Anom. Var. Eff. Exp. eV). Column 5: B P Column 12: l band): a = no evidence for variability (0 values above the 1% threshold that defines the catalog becaus s or for sources in chip gaps or on field edges. B P PSF Frac. Signif. log Background-corrected median photon energy (total band). 0.00.5 0.90 0.89 1.2 1.1 -2.3 -1.7 ...... a a 16.7 16.2 1.1 1.3 h,net Estimated net counts extracted in total energy band (0.5–8 k counts) C Source anomalies: g = fractional time that source was on a det e that the source is in a crowded region. 34755 3.3 1.7 0.5 0.6 3.6 0.90 0.89 0.91 0.89 1.0 0.90 1.1 1.0 1.1 -1.5 1.1 -3.0 -2.2 .... -2.9 .... -2.8 ...... a .... a a a a 16.2 18.2 16.3 17.1 17.7 5.3 1.8 1.1 1.1 6.8 82 0.4 1.8 0.62 0.89 1.2 0.9 -2.8 -2.9 ...... a 15.9 1.4 8 3.8 0.90 1.4 -2.6 g... 2 0.0 0.90 0.9 -2.7 g... 2 1.1 0.90 0.9 -1.5 g... 9 0.0 0.89 1.1 -1.5 g... 5 0.9 0.90 1.2 -2.3 g... 9774 0.9 2.5 0.5 1.7 0.90 0.89 0.91 0.89 1.1 1.0 1.0 0.9 -1.9 -2.2 -2.2 .... -2.2 ...... g... a a a 15.4 16.8 17.0 1.2 4.2 1.1 7 0.0 0.91 0.8 -1.5 g... 265 2.3 3.8 2.9 0.89 0.90 0.91 1.3 1.2 1.2 -2.1 -2.2 -2.3 ...... g... a a 16.6 16.9 2.1 5.5 3 2.2 0.89 1.4 -2.6 g... .9 0.9 0.89 1.5 -2.5 g... .3 2.8 0.90 0.9 -2.5 .... .5 0.0 0.90 0.9 -1.8 .... .3 0.0 0.90 0.9 -2.3 .... .3 0.8 0.90 0.9 -2.5 .... .0.5 0.0 0.6 0.89 0.89 1.1 0.8 -1.8 -1.8 ...... a 15.5 1.4 .0 0.9 0.90 1.5 -1.8 .... a 10.2 1.2 .3.2.5.8.8 4.2.5 0.5.8 1.1 6.7 0.0 0.91 1.8 0.89 3.8 0.91 0.90 1.1 0.89 1.5 0.90 1.3 0.91 1.8 -2.0 1.5 -2.7 1.2 -2.4 .... 1.7 -2.7 .... -2.6 .... -2.3 .... -2.6 a .... a .... a .... a a 16.8 a 15.6 a 16.4 13.8 14.8 4.1 15.7 1.7 14.1 1.0 3.5 1.4 3.6 1.8 Column 17: t Right ascension and declination for epoch J2000.0. B Extraction Columns 7,8: Column 14: Estimated net counts extracted in the hard energy band (2–8 k Variability characterization based on K-S statistic (tota t,net σ Columns 3,4: ly from background. Some sources have Off-axis angle. t,net Column 10: Catalog: Basic Source Properties Column 15: Column 6: IAU designation. . θ C 005). No value is reported for sources with fewer than 4 count out streak. . 0 Chandra < KS Column 2: P bserved on axis to obtain the reported number of counts. a reduced PSF fraction (significantly below 90%) may indicat eration after sources are removed from the catalog. Position (J2000.0) Error δ Background counts extracted (total band). # of counts extracted Table 6.2. Tentative √ Log probability that extracted counts (total band) are sole 05); c = definitely variable ( . (deg) (deg) (arcsec) (arcmin) (counts) (counts) (counts) ( 0 (J2000.0) α < Column 9: KS standard deviation of PSF inside extraction region < P Column 13: . 005 X-ray catalog sequence number, sorted by RA. . , computed as σ Source errors on column 7. 9 ; e = source on field edge; p = source piled up; s = source on read . 0 σ Column 1: < 79 063025.18+045941.0 063026.53+050240.9 97.604953 97.610561 4.994746 5.044707 0.9 1.1 6.3 6.8 4.5 4.0 3.0 3.0 1.5 2.0 1 063007.57+050440.7 97.531553 5.077982 1.9 11.9 9.0 5.5 14 6384 063043.18+050540.289 063049.92+050248.292 063051.19+050410.096 97.679953 063051.82+050310.8 97.708005 063053.23+050319.8 97.713322 5.094515 97.715940 5.046730 97.721828 5.069451 5.053005 1.2 5.055511 0.5 0.8 0.5 0.6 6.6 3.5 4.9 3.9 4.1 3.7 3.6 3.3 3.5 3.5 3.0 2.5 2.5 2.5 2.5 2. 0. 0. 0. 0. 5862 063041.25+050600.1 063043.08+045743.3 97.671885 97.679506 5.100029 4.962030 0.9 0.4 7.1 2.4 4.2 2.8 2.8 2.3 0. 0. 50 063039.36+050630.9 97.664005 5.108604 1.4 7.7 5.2 3.2 1. 49 063038.80+045834.3 97.661694 4.976217 0.5 2.9 2.8 2.3 0. 48 063038.08+050408.0 97.658701 5.068899 1.1 5.7 2.8 2.5 1. 47 063037.66+050559.2 97.656929 5.099788 1.3 7.4 4.1 3.2 2. 44 063037.14+045403.0 97.654752 4.900840 0.9 6.2 4.5 3.0 1. 2729 063032.83+050407.337 063033.14+045704.642 063034.52+045702.9 97.636818 063036.03+050052.4 97.638123 97.643841 5.068696 97.650145 4.951299 4.950821 5.014576 1.0 0.8 0.7 0.6 6.5 4.8 4.5 3.9 4.1 3.3 3.3 2.6 3.0 2.5 2.5 2.3 1. 0. 0. 0. 26 063032.75+045834.5 97.636484 4.976272 0.8 4.4 2.3 2.3 0. 1522 063028.91+045516.224 063030.63+050223.4 063032.45+050311.8 97.620469 97.627642 97.635212 4.921175 5.039857 5.053278 1.0 0.8 0.9 6.7 5.8 5.9 4.8 4.4 4.5 3.2 3.0 3.0 2. 1. 1. 13 063028.23+045447.3 97.617645 4.913166 1.0 7.1 5.7 3.4 2. net counts 111 063056.62+050704.1 97.735946 5.117824 1.2 7.9 6.1 3.5 2 120 063058.53+045647.2 97.743912 4.946450 0.5 3.2 2.7 2.3 0 131 063102.24+045604.1 97.759343 4.934490 0.7 4.4 2.5 2.3 0 132 063102.50+045708.6 97.760442 4.952394 0.6 3.7 2.7 2.3 0 135 063103.04+045902.9 97.762685 4.984146 0.5 3.2 2.7 2.3 0 139143 063104.70+050507.6 063106.12+045753.8 97.769600 97.775533 5.085466 4.964966 1.1 0.7 6.9 4.2 4.0 2.5 3.0 2.3 2 0 144146 063106.27+045525.8153 063107.31+050422.9155 063110.51+050116.8157 97.776131 063111.04+050640.5158 97.780489 063113.08+050301.0159 97.793808 063114.11+045929.0 4.923855 97.796036 063114.61+050515.7 5.073039 97.804520 5.021354 97.808804 5.111260 97.810880 0.9 5.050292 0.9 4.991407 0.8 5.087715 1.2 0.9 5.5 0.9 6.6 1.3 5.4 9.0 6.8 3.7 5.9 5.8 8.5 4.5 9.2 6.2 2.8 4.5 3.4 8.2 3.0 4.6 3.5 1 3.0 2 4.3 1 6 2 1 5 ) is (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) Note. — Seq. No. CXOU J upper error on net counts 1.497 keV) enclosed within the extraction region. Note that exposure time: approximate time the source would have to be o and lower 1 b = possibly variable (0 background estimates can rise during the final extraction it position error, 1 mkarf · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes t,c L ··· ··· ··· ··· ··· ··· log t c L 195 log ) 1 − h,c L log h X-ray Luminosities L 28.25 29.76 29.63 30.68 31.06 30.68 30.17 30.45 30.21 log s L · · · · · · · · · log 06) 30.88 30.22 30.22 30.97 30.98 . 4) 29.88 30.34 30.36 30.47 30.61 2) 29.82 31.15 31.24 31.17 31.41 2) 29.73 30.62 30.66 30.67 30.82 3) 30.13 30.06 30.07 30.40 30.49 3) 30.05 29.78 29.80 30.24 30.45 2) 30.40 30.71 30.72 30.88 30.92 ...... ) 30.29 30.53 30.55 30.73 30.88 8) 29.53 30.12 30.14 30.22 30.30 0) 5) 29.69 29.41 29.51 29.88 30.61 2) 29.84 29.72 29.79 30.08 30.58 3) 1) 30.69 30.49 30.49 30.90 30.91 4) 30.36 30.21 30.25 30.59 30.92 0) 30.32 29.55 29.65 30.38 31.27 3) 29.97 29.11 29.24 30.02 31.25 5) 30.02 29.54 29.65 30.14 31.09 8) 30.34 30.26 30.28 30.60 30.80 2) 0) 30.06 28.85 28.88 30.08 30.55 ) (ergss +0 ...... 3 , +0 +0 +0 · · · +0 +0 +0 , , , , , , , − EM +0 +1 +0 +0 +1 +0 +0 +1 +0 +0 +0 +1 +1 3 2 08 3 2 3 3 2 ...... , , , , , , , , , , , , , 0 0 0 0 0 0 0 0 log · · · · · · − · · · · · · − − − · · · − · · · · · · · · · · · · · · · · · · − · · · − · · · − 4 29.41 30.19 30.25 30.25 30.52 1( 5( 3 29.46 30.25 30.29 30.32 30.49 5( 3 29.30 30.30 30.36 30.34 30.53 6( 2 29.89 30.10 30.10 30.31 30.32 6( 2( 0( 6( 0 29.95 30.01 30.12 30.28 30.91 3( 2 30.08 29.43 29.43 30.17 30.18 8( 9( 0( 3( 4( 1( 9( 8 29.59 30.32 30.44 30.40 30.83 5 29.88 30.53 30.55 30.62 30.72 5( 1( 5( 6( 8( 5 30.18 29.47 29.49 30.26 30.46 ...... 53 b 8) 53 1) 54 0) 53 9) 53 1) 54 7) 54 9) 54 2) 53 8) 53 2) 53 . d Sources: Thermal Plasma ...... ) 53 ) 53 ) 54 ) 53 2) 53 . ) 53 +3 +0 +1 · · · +6 · · · +1 +0 +5 +3 · · · +0 · · · +0 , , , , , , , , , , , , , , · · · +0 3 2 7 6 5 6 5 3 2 1 8 4 1 2 ...... , , 8 0 0 0 0 1 1 0 0 0 1 0 0 4 0 . kT 8 − − − − − − − − − − − − · · · − − − 9( 0 53 2( 1 62 3 53 3 53 2( 1( 9( 3( 8 53 3( 6( 6( 7( 2 53 5( 2 54 4 54 4( 5( 6( 4 54 0( 1 53 Spectral Fit ∗ ...... 9 53 7 53 8( 0 8 54 1 1 2 . . . . . Fits 4) 2 3) 2 5) 10 3) 1 3) 15 3) 1 4) 3 2) 0 ...... ) 1 2) 0 6) 1 5) 1 4) 0 6) 6 4) 1 6) 2 ...... ) (keV) (cm +0 +0 +0 +0 · · · +0 +0 +0 +0 H 2 , , , , , , , , , +0 +0 +0 +0 +0 +0 +0 6 5 0 0 3 4 7 3 3 − N ...... , , , , , , , 0 0 1 1 2 0 0 0 0 · · · · · · − − − − − − − · · · · · · − · · · · · · − · · · (cm ∗ ∗ ∗ 0 0 1 4 0( 5( 5( 3( 7( 7( 1( 0( 0 6( 5 14 0 0 0 0 7 0 5( 7 4 5( 0( 0( 1 10 4 5 9 0 9 1 3( 0( 2( 4 12 ...... Signif. log t,net C a Source Table 6.3. X-ray Spectroscopy for Photometrically Selecte 8 063025.39+050540.5 15.2 2.7 21 2 063016.82+050452.2 10.6 2.0 22 6 063023.75+050555.8 36.8 4.8 22 69 063045.43+050234.3 9.6 2.3 22 66 063044.51+045131.1 13.3 2.5 22 56 063040.98+045730.9 16.8 3.2 22 75 063046.77+045904.2 22.9 3.9 21 74 063046.18+050151.3 46.8 5.9 21 55 063040.51+050606.6 13.0 2.6 22 81 063049.40+050423.1 23.3 3.9 21 65 063044.14+050221.2 19.9 3.6 21 76 063046.82+045033.0 19.9 3.2 22 73 063046.15+045838.8 16.9 3.2 22 54 063040.03+045759.4 114.7 9.8 20 45 063037.33+050005.5 8.7 2.1 22 40 063034.93+045826.2 8.6 2.1 21 32 063033.49+050142.0 8.3 2.0 23 53 063039.77+045439.6 16.0 3.1 20 41 063035.67+045535.8 10.0 2.3 22 39 063034.78+045641.0 76.3 7.8 20 51 063039.44+045636.6 20.6 3.6 21 30 063033.26+050335.8 19.3 3.4 21 28 063032.97+050142.3 15.1 3.0 21 20 063030.14+045949.3 14.1 2.8 23 25 063032.56+050616.9 44.9 5.6 21 17 063029.28+050113.2 9.8 2.2 22 16 063028.95+050512.4 10.3 2.1 20 14 063028.66+050137.8 32.6 4.7 21 11 063027.33+045347.7 11.5 2.4 21 12 063028.13+050530.4 20.7 3.4 21 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) Seq.No. CXOUJ d · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Notes t,c L ··· ··· ··· ··· log

t 196 c L nd upper uncertainties. log ) 1 − h,c L ols. (5) and (6) present the best-fit log ectively unconstrained. Fits lacking h X-ray Luminosities L intervals. More significant digits are used ived from the model spectrum, assuming a Table 6.1. 29.72 30.14 30.03 28.72 30.02 29.74 log ta to obtain rough estimates of luminosities; s L · · · · · · log 2) 30.60 30.82 30.83 31.02 31.14 3) 29.78 30.27 30.27 30.39 30.44 3) 30.23 30.53 30.55 30.70 30.88 1) 30.20 29.85 29.85 30.36 30.37 2) 29.70 29.65 29.66 29.98 29.99 3) 29.76 29.73 29.73 30.05 30.05 2) 30.06 29.80 29.80 30.25 30.26 1) 30.79 31.06 31.08 31.25 31.34 2) 30.22 29.29 29.29 30.26 30.28 ...... 8) 29.75 30.47 30.59 30.55 31.00 0) 30.04 29.34 29.53 30.11 31.68 4) 5) 29.75 30.22 30.28 30.35 30.65 2) 30.24 30.03 30.15 30.45 31.25 0) 29.77 29.73 29.83 30.05 30.69 3) 29.98 29.96 30.11 30.27 31.19 4) 29.60 28.91 28.99 29.68 30.46 0) 29.70 29.80 29.94 30.05 30.86 4) 29.99 29.47 29.54 30.11 30.73 3) 30.45 30.08 30.14 30.61 31.15 9) 30.01 30.63 30.72 30.72 31.11 ...... ) (ergss +0 +0 +0 +0 +0 +0 +0 +0 +0 3 , , , , , , , , , +0 +1 +0 +0 +0 +1 +0 +0 +1 +0 +0 +0 2 3 2 2 3 3 3 1 4 − ...... EM , , , , , , , , , , , , 0 0 0 0 0 0 0 0 0 · · · − · · · · · · · · · − · · · · · · · · · · · · − − · · · − · · · − · · · − · · · − − log 0( 1( 9( 2( 6( 2( 3( 7( 2( 5( 8( 4( 3 30.00 30.04 30.04 30.32 30.33 9( 9( 8( 0( 2( 7 29.83 29.72 29.81 30.08 30.66 3 30.03 29.11 29.12 30.08 30.23 9 29.84 29.41 29.52 29.97 30.89 3 29.80 29.46 29.49 29.97 30.21 3( 1( 7 30.16 29.83 29.87 30.33 30.65 2( 8 30.21 30.12 30.15 30.47 30.70 7 3( ...... 2) 54 9) 54 0) 54 4) 53 0) 53 0) 53 7) 53 5) 53 1) 54 8) 54 5) 53 ...... ) 52 ) 53 ) 53 b +4 +0 +1 +1 +1 · · · · · · +2 · · · +0 +1 +6 +4 +1 , , , , , , , , , , , , , , 2 1 3 6 5 5 3 6 1 7 8 5 3 6 ...... 1 0 0 0 0 1 0 0 2 0 0 1 1 0 kT − − − − − − − − − − − − − − , the same number of significant digits is used for both lower a 2( 4( 2 54 9( 1( 4 55 5 53 7( 9 54 9( 8( 6 53 0( 0( 0 53 4 53 9 54 1 53 3( 9( 3( 9 53 7 53 1( 0( 2 53 6 53 1 60 ...... Spectral Fit 5 53 1 2 3 1 1 . abundances (Imanishi et al. 2001; Feigelson et al. 2002a). C ⊙ Z 3) 1 2) 0 3) 2 3) 3 3) 2 3) 4 4) 1 3) 0 ...... ) 0 7) 14 2) 0 3) 0 5) 1 8) 0 5) 1 hem or when their values were so large that the parameter is eff ...... ) (keV) (cm +0 +0 +0 +0 · · · +0 +0 +0 +0 H 2 , , , , , , , , , ere frozen in the fit. Uncertainties represent 90% confidence +0 +0 +0 +0 +0 +0 8 3 7 7 4 4 6 2 2 ture parameters. Col. (7) presents the emission measure der − fication, net counts, and photometric significance data from N ...... , , , , , , 0 0 0 0 0 0 0 1 0 Table 6.3—Continued − − − · · · · · · · · · − − · · · · · · − − · · · − − (cm rozen parameters should be viewed merely as splines to the da ∗ ∗ ∗ ∗ ∗ 4 3 2 0 0( 6( 0( 3 2 2( 1 0 8( 9( 0 6( 0 2 0 0 2 1 8( 3 3 5( 1( 2( 3( 0 1 0 6( 0 0 6( 0 9( ...... Signif. log t,net C a Source 0.1 in order to avoid large rounding errors; for consistency < 83 063049.72+045822.5 88.8 8.5 21 87 063050.35+045535.4 16.6 3.2 22 88 063050.84+045902.3 8.9 2.2 22 94 063052.71+045144.3 11.5 2.3 22 99 063054.68+045904.1 15.8 3.1 22 100 063054.76+050242.1 21.7 3.8 22 101 063055.16+050327.1 15.4 3.0 21 102 063055.45+050637.6 28.7 4.3 22 110 063056.44+045619.8 11.8 2.6 21 116 063057.85+045143.0 10.7 2.2 21 118 063058.19+050019.5 24.8 4.1 20 119 063058.43+045503.7 38.3 5.2 21 124 063059.88+045846.4 8.8 2.1 20 126 063100.55+045807.2 20.8 3.7 22 127 063100.71+050614.4 11.0 2.3 20 128 063101.33+045610.3 10.5 2.4 22 129 063101.54+045734.3 52.7 6.3 21 134 063102.91+050348.6 16.5 3.1 20 138 063104.57+045543.5 11.2 2.4 21 141 063105.49+045749.3 16.5 3.2 21 142 063105.57+045525.4 33.0 4.8 22 149 063108.03+045930.3 134.3 10.6 21 150 063108.44+050126.5 13.2 2.7 22 152 063109.95+050745.9 15.8 2.6 20 154 063110.97+045546.5 13.0 2.7 21 161 063115.47+050215.1 10.0 2.1 22 163 063118.15+045513.6 25.2 3.8 21 167 063121.42+045406.9 14.4 2.5 20 164 063119.52+045310.1 11.8 2.1 22 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) For convenience, cols. (1)–(4) reproduce the source identi All fits used the “wabs(apec)” model in XSPEC and assumed 0.3 b a Seq.No. CXOUJ uncertainties, fits with large uncertainties, and fits with f Uncertainties are missing when XSPEC was unable to compute t values for the extinctiondistance column of density 1.4 andfor kpc. plasma uncertainties Quantities tempera marked with an asterisk (*) w 197 . d c · · · · · · · · · ensity = soft Notes ge that marked used for Table 6.1. t,c ed with a unreliable. L ··· ··· zen parameters log t c L log ) 1 − h,c L log h X-ray Luminosities L 30.50 30.72 30.51 log s d Sources: Power Law Fits L · · · log 3 . Γ +0 ··· N , the same number of significant digits is used for both lower 8 29.29 30.48 30.52 30.51 30.63 9 9 29.83 29.78 29.78 30.11 30.11 6 29.77 30.35 30.37 30.46 30.57 . . . . b 5 4 5 5 − − − − nd (6) present the best-fit values for the extinction column d 1 1 6 ng a distance of 1.4 kpc, are presented in cols. (8)–(12): (s) . . . Γ log –8 keV). Absorption-corrected luminosities are subscript power law normalization for the model spectrum. Quantities 1 2.1 2 estimates of luminosities; the listed parameter values are resent 90% confidence intervals. More significant digits are fication, net counts, and photometric significance data from 3 certainties, fits with large uncertainties, and fits with fro 5 3 PEC was unable to compute them or when their values were so lar ) (ergss Spectral Fit . . . H 2 0 since the soft band emission is essentially unmeasurable. − +0 ··· +0 − ∗ N 2 1 0 6 1 7 . . . . − (cm 5 cm . 22 > Signif. log H N t,net C a Source Table 6.4. X-ray Spectroscopy for Photometrically Selecte 0.1 in order to avoid large rounding errors; for consistency < 19 063029.51+050053.1 11.7 2.5 22 113 063057.10+045721.7 8.8 2.1 20 168 063125.31+045640.0 12.3 2.3 21 162 063117.02+045744.5 8.9 2.0 22 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) X-ray luminosities derived from the model spectrum, assumi For convenience, cols. (1)–(4) reproduce the source identi All fits used the “wabs(powerlaw)” model in XSPEC. Cols. (5) a a b c Seq.No. CXOUJ band (0.5–2 keV); (h) hard band (2–8 keV); (t) total band (0.5 Cols. (8) and (12) are omitted when log and upper uncertainties. Uncertainties are missing when XS the parameter isshould effectively be unconstrained. viewed merely Fits as splines lacking to un the data to obtain rough and power law photonwith index an parameters. asteriskuncertainties Col. (*) (7) were presents frozen the in the fit. Uncertainties rep 198 B000 A000 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· .37 AAA000 17.69 16.80 18.36 17.43 18.22 17.45 18.53 17.26 ··· ··· ··· ··· ··· ··· ··· 063022+045539 063026+045344 063028+045513 15.64 14.66 14.26 063032+050312 063034+050319063034+050337 14.87 14.05 13.69 063037+050006 063037+050557 16.73 16.14 15.85 680+0504518 063016+050452 14.52 13.74 13.42 AAA000 519+0459412550+0505402 063025+045941 063025+050540 14.80 13.89 14.20 13.15 13.86 12.81 AAA000 AAA000 2725+04534712814+0505302 063027+045347 063028+050530 15.78 14.502906+0505132 15.15 063029+050513 13.70 14.85 15.09 13.37 ABB000 14.28 AAA000 13.92 AAA000 3087+0500493 063030+050049 14.47 13.62 13.30 AAA000 3260+0506165 063032+0506163282+0504070 063032+050406 14.613323+0503362 14.48 13.74 063033+050336 14.03 13.40 14.39 13.80 AAA000 13.58 AAA000 13.17 AAA000 3478+0456409 063034+045640 13.89 13.23 13.08 AAA000 3567+0455366 063035+0455363720+0454026 14.93 063037+045402 13.68 12.99 13.43 11.78 AAA000 10.84 AAA000 Optical/Infrared Photometry 06300735+0504423 063007+050442 14.12 13.26 12.89 AAA000 06302871+0501380 063028+05013806302937+0501125 13.23 063029+050112 12.61 15.70 12.41 14.82 AAA000 14.45 AAA000 06303289+0458342 063032+04583406303292+0501427 063032+050142 16.05 13.89 15.23 13.27 14.7806303398+0456507 063033+045650 13.03 AAB000 06303453+0457043 15.91 EAAccc 063034+045704 14.97 16.20 14.57 15.19 AAA000 14.83 BBAccc 06303491+0458260 063034+04582606303586+050051406303649+0500443 063035+050051 15.24 063036+050044 16.54 14.32 15.62 15.85 13.94 14.79 15.55 AAA000 14.28 BCC000 AAA000 16.90 06302046+0503528 063020+050352 15.04 14.25 13.89 AA 17.68 06302940+0504123 063029+050412 15.78 15.13 14.84 AA ··· ··· 18.50 17.32 06302649+0502411 063026+050241 15.61 14.74 14 Table 6.5. Stellar Counterparts ··· ID (mag) (mag) (mag) ID ID (mag) (mag) (mag) ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ································· ······································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ······································································································································· ··· ··· ··· ··· ··· ··· ··· ··· ········································································ ··· ··· ··· ··· ··· ··· ··· ··· ······································· ····································································· ······ ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ······ ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ··· X-ray Source 5 063022.82+045538.2 1 063007.57+050440.7 23 063016.82+050452.24 063018.34+045953.9 063020.40+050351.7 0950-0096116 0950-0096157 20.03 20.41 17.53 16.05 06301 67 063023.75+050555.8 8 063025.18+045941.09 063025.39+050540.5 063026.53+050240.9 0949-0093548 0950-0096215 0950-0096223 20.41 18.79 18.04 16.75 16.33 15.39 06302 06302 10 063026.78+045341.5 1112 063027.33+045347.713 063028.13+050530.414 063028.23+045447.3 063028.66+050137.8 16 0948-009557617 0950-0096238 063028.95+050512.4 19.90 063029.28+050113.2 19.91 20.07 0950-0096249 17.50 16.77 21.13 16.12 0630 0630 19.07 17.49 0630 15 063028.91+045516.2 18 063029.46+050412.0 0950-0096256 20.42 1920 063029.51+050053.1 21 063030.14+045949.3 22 063030.17+045943.4 23 063030.63+050223.4 24 063030.88+050050.0 063032.45+050311.8 0950-0096271 20.37 17.61 16.25 0630 37 063034.52+045702.9 27 063032.83+050407.32930 063033.14+045704.6 063033.26+050335.8 0950-0096310 18.0935 0950-009631636 16.42 063034.15+045728.7 19.98 063034.26+050319.6 38 15.53 17.59 063034.68+050337.2 0630 16.18 0630 2526 063032.56+050616.9 063032.75+045834.5 28 063032.97+050142.3 0951-00956233132 20.42 063033.47+045052.1 33 063033.49+050142.0 34 17.81 063033.87+050236.5 063034.06+045651.0 16.39 0630 3940 063034.78+045641.0 063034.93+045826.2 0949-0093787 17.38 15.91 14.90 0630 41 063035.67+045535.84344 063036.58+050043.9 063037.14+045403.0 0949-0093800 18.89 0949-0093813 17.10 17.95 15.56 16.65 0630 15.24 0630 42 063036.03+050052.4 45 063037.33+050005.5 4647 063037.40+050228.1 063037.66+050559.2 # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 199 ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· .62 AAA0cc .50 ABA000 .03 AAA000 18.17 17.20 17.21 17.20 ··· ··· ··· ··· ··· ··· ··· ··· 063049+050423 17.41 16.67 16.31 063051+050134 063050+045715 063046+045838 063045+045822 063100+050759 11.23 10.47 9.75 ··· 5031+04553525082+0459025 063050+0455355116+0504106 063050+045902 063051+050410 14.48 14.64 15.07 13.64 13.83 14.21 13.29 13.54 13.93 AAA000 AAA000 AAA000 4956+04574394973+0458226 063049+0457444991+0502484 063049+045822 063049+050248 14.16 13.09 16.75 13.24 12.28 15.96 12.67 11.90 15.44 AAA000 AAA000 CCU000 4819+0457364 063048+045736 15.59 14.69 14.36 AAA000 4617+05015134676+0459041 063046+050151 063046+045904 13.58 13.44 12.82 12.73 12.41 12.45 AAA000 UUA00c 4200+0500596 063042+050059 12.77 12.22 12.01 AAA000 4414+0502209 063044+050220 14.11 13.31 12.98 AAA000 3943+04563673957+0459522 063039+0456363978+0454400 063039+0459524004+0457594 063056+050401 13.98 063040+045759 16.134100+0457307 10.79 13.14 063041+045730 12.21 15.10 10.36 12.83 14.02 11.90 14.61 10.22 AAA000 13.22 11.79 AAA000 AAA000 12.92 AAA000 AAA000 Optical/Infrared Photometry 06305192+0458055 063051+045805 15.71 14.79 14.42 AAA000 06305187+0503110 063051+050311 15.31 14.25 13.78 AAA000 06305185+0500480 063051+050047 16.68 15.86 15.44 BCC000 06305033+0500288 063050+050028 17.72 15.40 14.11 UBA000 06304692+0450325 063046+045032 15.5906304920+0457507 063049+045750 14.61 14.91 14.27 13.97 AAA000 13.31 AAA000 06304599+0502344 063045+050234 14.58 13.77 13.37 AAA000 06304563+0455569 063045+045556 12.36 12.06 11.97 AAA000 06304542+0502344 063045+050234 15.57 14.60 14.16 AAA000 06304498+0458597 063044+045859 15.43 14.58 14.22 AAAccc 06304118+0506006 063041+050600 15.38 14.53 14.15 AAA000 06304098+0506048 063040+050604 15.89 15.02 14.63 AAAc00 18.31 Table 6.5—Continued 18.35 18.03 06304785+0459596 063047+045959 15.78 15.03 14 19.45 17.15 06304838+0456381 063048+045638 15.61 14.81 14 13.93 16.38 06303885+0458343 063038+045834 14.49 13.52 13 ··· ··· ··· ··· ··· ID (mag) (mag) (mag) ID ID (mag) (mag) (mag) ································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ···································· ··· ··· ··· ···································· ··· ··· ··· ··· ······································································································································· ··· ··· ··· ··· ··· ··· ··· ································· ····································································· ··· ··· ··· X-ray Source 94 063052.71+045144.3 93 063051.93+045806.7 92 063051.82+050310.8 91 063051.81+050047.7 87 063050.35+045535.4 0949-0094042 20.19 17.22 15.92 0630 8889 063050.84+045902.390 063051.19+050410.0 063051.52+050133.8 0949-0094048 0950-0096642 18.68 20.34 16.36 17.75 16.16 16.70 0630 0630 86 063050.33+050028.7 77 063047.85+050000.281 0949-0094006 063049.40+050423.1 0950-0096603 8283 063049.54+045743.884 063049.72+045822.585 063049.92+050248.2 063050.08+045716.1 0949-0094030 0949-0094034 0950-0096610 19.73 18.07 20.21 17.05 15.68 17.66 15.67 14.59 0630 19.04 0630 0630 7879 063048.22+045735.8 063048.38+045638.4 0949-0094010 0949-0094013 20.96 19.51 17.50 0630 80 063049.17+045750.7 7475 063046.18+050151.376 063046.77+045904.2 063046.82+045033.0 0950-0096515 0949-0093993 19.45 18.20 16.93 16.23 15.33 14.74 0630 0630 73 063046.15+045838.8 72 063046.04+050235.3 71 063045.56+045556.2 70 063045.43+045823.2 6869 063045.20+045726.8 063045.43+050234.3 59 063041.97+050059.6 0950-0096433 16.48 14.03 13.88 0630 6061 063042.63+045620.5 063042.78+045533.0 6263 063043.08+045743.3 64 063043.18+050540.2 65 063043.21+050227.4 66 063044.14+050221.267 063044.51+045131.1 063044.98+045859.8 0950-0096475 18.90 16.45 15.78 0630 58 063041.25+050600.1 4849 063038.08+050408.0 063038.80+045834.3 0949-0093839 5051 063039.36+050630.9 52 063039.44+045636.653 063039.60+045952.754 063039.77+045439.655 063040.03+045759.456 0949-0093851 063040.51+050606.6 57 0949-0093856 063040.98+045730.9 0949-0093860 19.01 063041.03+050605.4 0949-0093863 20.09 13.17 16.37 0949-0093876 13.93 17.18 12.08 15.51 19.30 13.28 17.14 0630 11.61 0630 16.85 13.15 0630 0630 15.55 0630 # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 200 A000 ··· ··· ··· ··· ··· ··· .57 AAB000 .48 AAA000 .37 AAA000 .65 AAA000 17.25 16.33 ··· 063105+050127 063059+045846 12.74 11.84 11.52 063103+045902 17.84 16.96 16.54 5469+0459039 063054+045903 14.97 14.16 13.85 AAA000 5285+0505396 063052+050539 14.71 14.01 13.82 AAA000 Optical/Infrared Photometry 10549+0457496 063105+045749 12.39 11.93 11.76 AAA000 10008+045932610054+0458072 063100+045932 063100+04580710133+0456106 16.4310153+0457344 063101+045610 15.09 063101+045734 15.72 11.38 14.09 13.08 15.50 11.08 13.4810288+0503493 12.27 BCD000 063102+050349 10.9010321+0457285 AAA000 11.95 063103+045728 13.7910456+0455433 AAA000 AAA000 063104+045543 15.31 12.92 12.05 14.39 12.41 11.78 14.07 AAA000 11.66 AAA000 AAA000 05583+0458189 063055+045818 14.84 14.02 13.70 AAA000 05667+0507047 063056+050705 14.8305787+0451429 13.92 063057+04514205819+0500195 13.32 063058+050019 14.8005861+0459118 AAA000 13.81 14.17 063058+045911 13.23 13.92 14.08 12.90 AAA000 13.17 AAA000 12.55 AAA000 06305948+0456135 063059+045613 14.9706310071+0506142 063100+050614 14.0706310208+0457126 14.78 13.7306310223+0456045 063102+04571206310247+0457089 063102+045604 13.9806310283+0457506 AAAcc0 063102+045708 15.38 063102+045750 15.52 13.64 16.48 14.44 16.92 14.68 AAA000 15.65 14.20 16.21 14.29 15.25 AAAcc0 15.78 AAAc00 BCC000 CUD000 06305546+050637106305554+0456056 063055+050637 063055+045605 16.2106305601+0457153 14.97 063056+045715 15.38 14.02 14.96 15.03 13.54 14.03 ABB000 AAA000 13.67 AAA000 06305643+045715506305640+0456193 063056+045715 063056+04561906305687+0456444 15.3506305712+0457222 063056+045644 13.6406305727+0458325 063057+045722 14.4306305773+0501128 063057+045832 15.25 12.77 063057+050112 14.59 13.9106305806+0502546 15.36 14.39 12.46 063058+050254 15.88 13.9806305842+0455033 AAA000 14.51 14.01 AAAccc 063058+045503 15.28 15.05 13.65 14.13 AAA000 16.09 14.54 14.62 AAA000 AAA000 15.12 13.98 AAA000 14.52 AAA000 AAAccc 06305423+0456489 063054+045649 15.77 14.87 14.51 AAA000 06305383+0457057 063053+045705 15.73 14.28 13.20 AAA000 16.47 06305610+0457055 063056+045705 15.02 14.13 13.80 AA Table 6.5—Continued ··· 18.97 18.05 06310469+0505076 063104+050508 15.76 14.95 14 19.13 18.23 06305867+0457402 063058+045740 15.82 14.93 14 18.98 17.17 06305513+0503272 063055+050327 14.67 13.77 13 18.29 16.77 06305628+0458317 063056+045831 14.89 14.01 13 ··· ··· ··· ··· ID (mag) (mag) (mag) ID ID (mag) (mag) (mag) ··· ··· ··· ··· ··· ········· ··· ········· ······ ··· ········· ··· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ·········································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··············· ······ ······ ··· ··· ··· ······ ···································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ································· ··· ··· ··· ··· ···································· ··· ··· ··· X-ray Source 99 063054.68+045904.1 0949-0094092 20.30 17.97 16.57 0630 98 063054.20+045648.8 9596 063052.93+050539.097 063053.23+050319.8 063053.84+045706.2 0950-0096668 19.15 16.90 15.83 0630 # (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 140 063105.38+050130.3 141 063105.49+045749.3 0949-0094237 14.77 13.68 13.64 063 123124 063059.47+045614.7 125 063059.88+045846.4 126 063100.05+045932.7127 063100.55+045807.2128 063100.71+050614.4 129 063101.33+045610.3130 0949-0094162 063101.54+045734.3131 0949-0094168 063102.14+045711.5 132 19.40 063102.24+045604.1 133 0949-0094176 19.76 063102.50+045708.6 134 0949-0094180 17.30 063102.81+045750.6 135 13.64 17.09 063102.91+050348.6136 17.51 17.37 063103.04+045902.9 137 12.72 16.18 063103.22+045728.2 063 138 15.39 063103.39+045442.2 139 063 0950-0096798 12.72 063104.57+045543.5 14.70 063104.70+050507.6 063 0949-0094207 19.14 063 0949-0094224 20.88 16.34 0950-0096820 13.33 18.51 15.39 063 12.96 17.24 063 13.18 063 102103 063055.45+050637.6 104 063055.47+045605.6 105 063055.80+045621.1 106 063055.83+045818.6107 063056.01+045715.7 063056.08+045705.6 0949-0094103 0949-0094104 19.90 20.47 17.75 16.66 063 109110 063056.42+045715.8 111 063056.44+045619.8 112 063056.62+050704.1113 063056.89+045645.0 114 063057.10+045721.7 115 063057.25+045832.0 116 0951-0095780 063057.70+050112.6 117 063057.85+045143.0118 18.23 063058.02+050254.4 119 063058.19+050019.5120 16.67 063058.43+045503.7 121 0948-0096025 063058.53+045647.2 122 16.40 063058.59+045912.0 0950-0096749 20.74 063058.68+045740.7 063 19.61 19.31 0949-0094139 0949-0094141 17.02 16.48 19.42 063 15.44 16.40 063 15.69 063 100101 063054.76+050242.1 063055.16+050327.1 0950-0096702 108 063056.28+045832.0 0949-0094107 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg 201 hida (1981), ··· ··· ··· ··· ··· ··· .20 AAA000 .39 AAAc00 17.46 18.84 17.63 ··· ··· ··· 14.83 13.96 13.57 AAA000 15.26 14.34 13.85 AAA000 15.49 14.57 14.30 AAA000 ··· ··· ··· 063120+045823 063109+050107 17.14 16.40 16.14 063105+045525 m Table 6.1 and Table 6.2. For convenience, [OI81]=Ogura & Is Optical/Infrared Photometry 11948+0453096 11540+0502159 063115+050216 13.22 12.92 12.80 AAA000 12142+0454057 063121+045405 13.99 13.25 12.99 AAA000 10729+0504229 063107+05042310804+0459300 15.8510845+0501259 063108+045929 063108+050125 15.0110996+0507460 13.3911048+0501168 063109+050746 14.82 14.6811098+0455455 063110+050116 12.75 063110+045545 15.30 14.01 AAA000 15.75 12.52 13.19 14.92 13.69 14.88 AAA000 12.44 14.52 AAA000 14.54 12.20 AAA000 AAA000 AAA000 06311821+0455144 06312100+0453238 063120+045323 14.93 14.14 13.97 AAA000 06311189+050443106311320+0502596 063111+050443 063113+05025906311454+0505184 16.25 063114+050518 14.18 15.57 16.61 13.81 15.39 16.11 13.70 BBC000 15.89 AAA000 BDUcc0 06310620+0455257 063106+045525 16.80 16.18 15.79 BDU000 ; pmRA=0.6 mas/yr, pmDE=-12.6 mas/yr 1 − mDE=2.1 mas/yr ; pmRA=2.8 mas/yr, pmDE=0.4 mas/yr , pmDE=-2.8 mas/yr , pmDE=3.6 mas/yr , pmDE=-11.9 mas/yr Table 6.5—Continued 18.96 17.08 06311524+0454385 18.74 17.43 06310719+0455262 063107+045526 15.31 14.50 14 19.12 17.63 06310608+0457529 063106+045752 15.56 14.75 14 ]Brhoe &2]=Berghöfer Christian (2002) ··· ··· ··· (ROSAT/PSPC)=31.01 erg s Lx ID (mag) (mag) (mag) ID ID (mag) (mag) (mag) ···································· ··· ··· ··· ···································· ··· ··· ··· ··· ··· ··· ··· ··· ········································································································· ·········································· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· ··· reproduce the sequence number and source identification fro Columns 1–2 X-ray Source =[MJD95] 653; pmRA=-1.0 mas/yr, pmDE=-5.4 mas/yr =[MJD95] 498; pmRA=-3.0 mas/yr, pmDE=1.9 mas/yr =[BC02] 11; known X-ray source; log =[OI81] 35 =[MJD95] 471; spectral type A3:; pmRA=-0.8 mas/yr =[OI81] 36 =[MJD95] 497; spectral type B5; pmRA=6.5 mas/yr, p =[OI81] 14 =[MJD95] 104; spectral type=[OI81] B1V; 10 pmRA=11.0 =[MJD95] mas/yr 108; spectral type B2V; pmRA=-2.3 mas/yr =[OI81] 12 =[MJD95] 102; pmRA=6.8 mas/yr, pmDE=0.6 mas/yr =V539 Mon =[OI81] 13 =[MJD95] 110=MSX6C G206.1821-02.3456 # 161 141 149 Note. — 53 54 71 128 61 138 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) 164 063119.52+045310.1 0948-0096247 20.34 18.40 16.68 063 161162 063115.47+050215.1163 063117.02+045744.5 063118.15+045513.6 0950-0096973 15.00 14.48 14.24 063 166167 063120.87+045326.1 168 063121.42+045406.9 063125.31+045640.0 0949-0094425 18.85 16.72 15.58 063 165 063120.08+045824.3 146147 063107.31+050422.9148 063107.77+050248.7 149 063107.82+045740.8 150 063108.03+045930.3151 0950-0096860 063108.44+050126.5152 063109.16+050109.1 153 20.39 063109.95+050745.9154 0949-0094272 063110.51+050116.8155 0950-0096879 19.33 063110.97+045546.5156 16.67 063111.04+050640.5 157 0951-0095904 20.26 17.80 063112.02+050441.3 158 0950-0096913 15.16 063113.08+050301.0 159 063 0949-0094306 19.55 18.10 063114.11+045929.0 160 21.03 14.83 063114.61+050515.7 17.33 20.22 16.49 063115.18+045439.2 063 18.92 063 15.60 16.92 17.52 063 0949-0094356 14.45 063 063 142 063105.57+045525.4 144145 063106.27+045525.8 063107.14+045525.6 0949-0094260 143 063106.12+045753.8 0949-0094249 Seq CXOUJ USNOB1.0 B R I 2MASS FLAMINGOS J H K PhCcFlg [MJD95]=Massey, Johnson, & Degioia-Eastwood (1995), [BC0 202

Fig. 6.1 A 30′ 30′ DSS2 R-band image of the NGC 2237 region. The box outlines the ′ ′ × 17 17 ACIS-I ﬁeld of view for the ObsID 8454 centered on NGC 2237. The polygons × outline the coverage of NGC 2244 region in the ObsIDs 1874 and 3750. 203

5:00 Dec. (J2000)

4:48 31:30 6:31:00 30:30 R.A. (J2000)

Fig. 6.2 ACIS-I image of the NGC 2237 region (0.5-8 keV) overlaid with source extraction regions. 204

Fig. 6.3 Smoothed ACIS-I image of the NGC 2237 region created using csmooth. Red represents soft X-ray emission and blue represents hard X-ray emission. 205

Fig. 6.4 NIR J H vs. H K color-color diagram for Chandra stars with high-quality − − JHK photometry (error in both J H and H K colors < 0.1 mag). The green − − circles and red triangles represent sources with and without signiﬁcant K-band excess (E(H K) > 2σ(H K)). The black solid and long-dash lines denote the loci of MS − − stars and giants, respectively, from Bessell & Brett (1988). The purple dash dotted line is the locus for classical T Tauri stars from Meyer et al. (1997), and the cyan solid line is the locus for HAeBe stars from Lada & Adams (1992). The blue dashed lines represent the standard reddening vector with crosses marking every AV = 5 mag. 206

Fig. 6.5 NIR J vs. J H color-magnitude diagram using the same sample and symbols − as Figure 6.4. The purple solid line and dashed line is the 2 Myr isochrone and the 1 Myr isochrone for PMS stars from Siess et al. (2000), respectively. The dash dotted line marks the location of Zero Age Main Sequence (ZAMS) stars. The blue dashed lines represent the standard reddening vector with asterisks marking every AV = 5 mag and the corresponding stellar masses are marked. 207

Fig. 6.6 The spatial distribution of X-ray sources that do not have matched IR counterparts (shown as blue crosses) and sources that have hard median photon energies (MedE > 2.0 keV; shown as red circles). About ten hard X-ray sources without matched NIR counterparts are located inside the HII region cavity, which are likely AGNs. The rest located inside the optical dark pillar region are probably embedded young stars. 208

2 Fig. 6.7 The stellar surface density (log#stars arcmin− ) map for all NGC 2237 sources smoothed with a 0.5′ radius sampling kernel. 209

To HD 46150 (O5V)

To HD 46223 (O4V)

Fig. 6.8 The spatial distribution of all the IR counterparts to our X-ray sources. The 12 molecular cloud is outlined by the (yellow) CO emission contours (Heyer et al. 2006). The symbols are the X-ray-selected stars classiﬁed as Class II (red diamond) and Class III (green circle) based on their NIR colors, and sources without counterparts (plus). 210

10:00.0

05:00.0

5:00:00.0 124 83 141 54 129

128 Dec. (J2000) 154 138 61 53 55:00.0 44

4:50:00.0

45:00.0 30.0 31:00.0 30.0 6:30:00.0 R.A. (J2000)

Fig. 6.9 The spatial distribution of stars in diﬀerent mass range. The extent of the compact cluster is 2pc. Stars with identiﬁed B and A spectral types and stars with ∼ NIR estimated masses & 2M⊙ as the high mass sample, shown as red boxes. Stars in mass range M < 2M⊙ are the low mass sample shown as blue crosses. 211

Fig. 6.10 (a) The X-ray spectrum and the spectral fit for the brightest X-ray star (#149) in the field. The best fit one-temperature thermal plasma model gives kT = 4.1 keV and 21 2 an absorbing column N 4 10 cm− . (b) The X-ray spectrum and the spectral fit H ∼ × for the brightest X-ray source in the fields (#54). The spectrum can be fit with a single 20 2 temperature thermal plasma model (kT = 1.2 keV, N 10 cm− ). H ∼ 212

1 arcmin

Fig. 6.11 The KPNO Hα image of a highly interesting region. Extraction regions of X-ray sources in this region are shown. Two optically dark pillar objects are circled. Source #86 (only 4 net counts) appears highly obscured with a visual extinction of 20 ∼ mag in the NIR diagrams, possibly a star embedded in or behind the dense pillar. A possible HH-ﬂow object is marked with a box. 213

Chapter 7

Conclusions and Future Prospects

7.1 Summary

The low mass stellar populations of many MSFRs were poorly studied in the optical and IR wavelengths because of their large distance, high extinction, and the heavy contamination from unrelated sources. Yet a thorough understanding of massive star and cluster formation cannot be achieved without studying them. X-rays open a new window to examine the stellar populations residing in the MSFRs, but the low angular resolution of previous generation X-ray telescopes limited the outcome from such studies. Driven by the high sensitivity high spatial resolution Chandra observations since the advent of Chandra X-ray Observatory, the understanding of the X-ray-emitting populations in the MSFRs is taken to a higher level in the last few years. In this thesis, we have analyzed Chandra observations of two massive star forming complexes and studied the X-ray sampled stellar populations in great details. Now we are able to revisit the outstanding issues in understanding the MSFRs introduced in Chapter 1. The unique power of X-ray selection of young stellar cluster members yielded the following main ﬁndings:

1. High spatial resolution Chandra images of MSFRs can effectively identify the lower- mass cluster populations accompanying the massive OB stars. In many cases, the massive clusters are much more populous than the populations sampled in the optical and IR. We have significantly improved the census of the young stellar populations in the MSFRs we studied. Depending on the sensitivity of the complementary optical and IR catalogs, many of the X-ray sources have stellar counterparts identified. A great number of these X-ray emitting stars are previously uncatalogued PMS stars. In the case of the poorly-studied cluster Pismis 24 in the NGC 6357 region (Chapter 3), the X-ray detected population provides the first deep probe of the rich population of this massive cluster, increasing the number of known members from optical study by a factor of 50. In the Rosette complex, ∼ we identified the low-mass populations in the NGC 2244 cluster (Chapter 4), the embedded populations in the Rosette Molecular Cloud (Chapter 5), and possibly a triggered population in the NGC 2237 region (Chapter 6), totaling 1500 stars ∼ revealed in X-rays. The typical fraction of contaminants from foreground stars, background stars, and extragalactic sources is on the order of 5 10%, most of − which can be identified based on counterpart information. The clean sample of the full PMS young stellar population and the massive stars is truly the unique power of the Chandra observations of the MSFRs. When complemented with long wavelength observations to determine the stellar masses and 214

ages, we can gain a better understanding of the past and present processes of star formation. A number of candidate O stars in obscured regions and a rich sample of intermediate-mass stars emerge from our observations, which are ideal for optical and IR followup.

2. Chandra studies of MSFRs offer a new way to investigate spatial structure and of young clusters. We determine the spatial distribution of cluster members to sub- arcsecond precision. Distinctive structures and substructures are identified based on the absorption-stratified stellar surface density maps. NGC 6357 and NGC 2244 are both large spherical clusters with substructures (Chapter 3 and 4). The spherical distribution of stars with relatively little substructure is somewhat expected if the formation process of rich star clusters is quasi-equilibrium, as explained in a model developed by Tan et al. (2006), in which the star cluster builds up gradually over many free-fall times. The overdensity in NGC 6357 consisting of a group of obscured stars (behind the optical dark column) separated from the main cluster could be the remnant core of the previous generation of stars that created the large scale morphology. The existence of a compact subcluster around HD 46150 and a secondary density peak, is direct evidence that the NGC 2244 cluster has not attained dynamical equilibrium. The radial density profile of NGC 2244 shows a larger relaxed structure around its central compact subcluster. Similar structure exists in NGC 6357. As star formation occurs within molecular clouds, could this be the consequence of the earliest structure of the protostellar cores and the fragmentation process within them? A numerical simulation by Bonnell et al. (2003) investigating the fragmentation a turbulent molecular cloud show formation of such subclusters with a massive star in the center via merging and competitive accretion (see also Bonnell et al. 2007, Beuther et al. 2007). We need to examine other young clusters to see if they have high stellar densities around central massive stars and relaxed outer cluster as well. Morphological evidence for triggered star formation at the peripheries of expanding HII regions is found. X-ray observations are capable of identifying the triggered populations with no selection bias against stars without color-excess in the IR surveys. This is best demonstrated in the Rosette region (Chapter 5 and 6), where around the main ionizing cluster NGC 2244 a small cluster RMC-A in the southeast and a satellite cluster NGC 2237 in the west were identified in the X-ray. Similar cases may have been found in Sh2-219 (Deharveng et al. 2005, 2006), SFO 78 (Urquhart et al. 2007), and RCW 120 (Zavagno et al. 2007). Compact elongated cluster inside the HII region, Class II sources aligned in an arc, broad distribution of stars across the PDR, and a remarkable lack of stars in some portion of the field are seen in NGC 2237, suggesting a complex star formation history. Stepping into the cloud to the southeast, the spatial distribution of RMC X-ray sources nicely traces the embedded clusters (Chapter 5), confirming the IR clusters identified in Román-Zúñiga et al. (2007a), preferentially in the molecular clumps along the RMC ridge, corresponding to the sequential triggered star formation picture (Elmegreen & Lada 1977). In NGC 6357, we may also witness triggering along the northern rim where the EGG is found. The triggered population does not appear significant, if 215

we take the ratio of triggered cluster to the central cluster as a first order estimate of the triggering efficiency ( 7%). Nevertheless, in both NGC 6357 and Rosette we ∼ need more than the morphological evidence to prove the triggered star formation scenario. The ages of the YSOs should be measured to firmly establish the causality of the events. Our X-ray sampled clusters shed light on the debate on the clustered and distributed star formation modes. The concentration of X-ray identified young stars implies that .35% of stars could be in a distributed population throughout the RMC region. This result is also confirmed by the clustering analysis in Román- Zúñiga et al. (2007a). We speculate that clustered star formation (Lada & Lada 2003) is the dominant mode in this molecular cloud, and perhaps all GMCs with molecular clumps ready to collapse.

3. Mass segregation of both primordial and dynamical origins are present in young stellar clusters, which can be investigated through Chandra observations of MSFRs as well. When combined with available stellar mass information, the spatial distribution of the X-ray stars can be used to study mass segregation in young clusters. Diversities have been seen among the clusters we observed: the massive stars in NGC 6357 are clearly concentrated in the center, while the massive stars in NGC 2244 are more dispersed with one O5 star accompanied by a small subcluster in the center and one isolated O4 star in the outer region. If they are coeval, any successful cluster formation model will need to account for the diversity seen in this cluster. Our results strongly constrain models of the cluster formation process.

4. Similar to the KLF frequently used in NIR studies, the XLF can be used to estimate the size of a stellar population and examine the IMF of a cluster empirically. By scaling the COUP ONC XLF to the observed XLFs, the total stellar population in NGC 6357, NGC 2244, and the RMC clusters is estimated to be 10000, 2000, ∼ ∼ and 1300 respectively. The XLFs suggest normal IMFs for NGC 6357 and NGC ∼ 2244; we do not confirm the top-heavy IMFs suspected from earlier optical studies. This may be attributed to the greatly improved census of the low-mass populations detected in the X-rays that were unidentified in the optical. However, cautions must be taken that the XLFs may not be identical in all regions. We compare the XLFs of the ONC, M17, Cep B, NGC 6357 and NGC 2244 (Chapter 4) and find diversities among them. In addition, new measurements of distance to clusters are constantly updated. This brings uncertainties to the luminosities (in X-ray band as well as ONIR bands). For example, a number of new distances to the ONC have been reported (414 7 pc, Menten et al. 2007; 389 24 ± ± pc, Sandstrom et al. 2007; 437 19 pc, Hirota et al. 2007; 434 14 pc, Kraus ± ± et al. 2007; 392 32 pc, Jeffries 2007). A new consensus value for the distance ± to ONC seems to be 400 pc. As the COUP studies adopted d = 450 pc, the ∼ X-ray luminosities based on the new distance can be 20% lower, and a recalibrated COUP XLF will shift to the fainter end by 0.1 dex in log Lx. Similarly, there ∼ are also uncertainties in the distance to NGC 6357 (1.7 kpc was reported although the latest measurement d = 2.56 0.10 kpc is considered more accurate) and the ± 216

Rosette Complex (1.6 kpc instead of 1.4 kpc is also used in literature). Given that the COUP XLF spanning 5 orders of magnitude in luminosity and that the range of XLF scaling to estimate the relative populations, this uncertainty will not affect our results. The uncertainties in the ages of cluster members do not affect the XLFs. However, the stellar masses estimated from PMS isochrones will change if a different age is used. For example, if we adopt a younger cluster age, the PMS isochrone will shift upwards and yield a lower dereddened mass and less amount of extinction. We attempted a younger age (0.3 Myr) for NGC 6357 (Chapter 3), and the inferred properties for stars with spectral types later than B3 are less dependent on the age assumption. The reported age uncertainty for NGC 2244 is much smaller (2-3 Myr), which does not affect our IMF results (Chapter 4). Note that we assume coevality of the stellar populations, which is reasonable for these young clusters but certainly more complex. Although star formation may occur rapidly, the process is not completed at the same instant (Palla & Stahler 2000; Hartmann 2001; Tan et al. 2006). For example, age dispersion is known to exist among the Orion stars (Hillenbrand 1997; Huff & Stahler 2006; Palla et al. 2007). Non-coeval populations may cause anomaly in the XLFs as well as in the IMFs (e.g., Cep B XLF and IMF, Getman et al. 2006). It also worths noting that the binary fraction among the massive stars and intermediate-mass stars remains a grey area. Surveys of intermediate-mass member stars in rich massive clusters to measure their multiplicity, such as the binarity study of Sco OB2 intermediate-mass stars by Kouwenhoven et al. (2007), are needed. If the binary fraction among stars in different mass ranges are similar, the number counts in the most massive bins will drop but the overall slope of the inferred IMF does not change significantly. Further studies of the IMFs through the XLFs are warranted, since the slopes of the high-mass end of IMFs place strong constraints on the massive star formation theory. Astrometric distance measurements and optical spectroscopy to accurately determine distances and to measure ages are needed for most of the clusters.

5. X-ray observations naturally deliver a nearly disk-unbiased sample of young stars, and the observed disk fraction in the MSFRs as indicated by K-band excess appears lower than the NIR-excess disk fractions found in the nearby LMSFRs of similar age. We derive an overall K-excess disk frequency of 6% for X-ray selected stars ∼ in the intermediate- to high-mass range in the NGC 6357 region (Chapter 3), and 10% for stars with mass M & 2M⊙ in NGC 2244 (Chapter 4). The K-excess ∼ disk frequency for X-ray selected stars with mass 0.5M⊙ . M . 2M⊙ in NGC 2244 is 5% (Chapter 4). Similarly the FLAMINGOS JHK imaging study by ∼ Román-Zúñiga et al. (2007a) derive a low fraction ( 10%) of K-excess stars with ∼ mass M & 0.1M⊙. The low fraction of K-excess stars among the massive stars in clusters that are < 2 Myrs old is not surprising. The disk frequencies are in line with previous studies of other young clusters: the inner disks around massive stars evolve very rapidly (Hillenbrand et al. 1993; Hernández et al. 2005). The X-ray stars in these regions 217

provide an important new sample for studies of intermediate-mass PMS stars that are not accreting, in addition to the accreting HAeBe stars. Studies of disk fraction as a function of the stellar age (e.g., Haisch et al. 2001; Hernández et al. 2007) conclude that the “standard” fraction of stars in the solar mass range that have IR disk emission is expected to be 50 80% in clusters ∼ − younger than 3 Myrs (e.g., NGC 2024, NGC 2264, Trapezium, Taurus, NGC 2362, NGC 7129, Cha I, IC 348, see Figure 1 in Haisch, Lada, & Lada 2001). Comparing to these disk fractions, the low K-excess disk frequency ( 6%) for X-ray selected T ∼ Tauri stars in the 2 Myr old NGC 2244 cluster seems intriguing and contradicting their results. However, the disk fractions based on L-band excess (or Spitzer/IRAC 8µm excess) trace thermal emission from disk materials of cooler temperatures than the hot inner disks that show K-band excess. Haisch et al. (2001) show that JHK observations alone are not sensitive enough to detect circumstellar disks in a complete and unambiguous manner. Many of the disks may have already become the so-called transition disks (Forrest et al. 2004), where the inner disk has opened a gap. Such disks are likely missed in the K-excess selected samples. Therefore, the inferred disk fraction of a young cluster generally increases when measured in longer IR wavelengths (e.g., Lada et al. 2004). Indeed, Balog et al. (2007) report a 45% 8µm excess disk fraction for NGC 2244. To interpret our results in the context of Haisch et al. (2001), disk evolution from Haisch et al. (2001) including K-band excess fractions compiled from literature (see references in Haisch et al. 2001) are illustrated in Figure 7.1. It appears that the fraction of K-band excess sources in X-ray sampled NGC 2244 stars is slightly lower than the K-excess fraction reported in the deep FLAMINGOS IR study (Román-Zúñiga et al. 2007a). Both are 10 30% lower than the expected ∼ − K-excess fraction for a 2 Myr cluster (e.g., IC 348 or Taurus) estimated from Figure 7.1. Similarly, the 45% 8µm excess disk fraction in NGC 2244 is smaller than but still in agreement to the results for 2-3 Myr old clusters in Haisch et al. (2001) and Hernández et al. (2007). Close to the O stars, they report a drop in the disk fraction. The lower disk fraction in NGC 2244 compared to the disk frequencies in clusters of similar age but without presence of O stars may indeed reflects faster disk dissipation due to photoevaporation (Hollenbach et al. 2000). This implies the radiation and wind-dominated environment does have an impact to the evolution of the protoplanetary disks. On the other hand, IR surveys alone do not include unbiased representation of the diskless cluster members, which X- ray surveys effectively identify. This may also lead to a lower disk fraction in the X-ray-selected sample of young stars. Further systematic studies are needed to explore the disk-unbiased nature of the X-ray selected sample. This will be further elaborated in 7.3. § 6. X-ray observations reveal a large sample of massive stars, including well known O stars in the optical and obscured O stars newly-identified in the X-rays and IR (Chapter 3, Wang et al. 2007a; Wolk et al. 2006; Broos et al. 2007; Tsujimoto et al. 2007). The detection efficiency of O and early B stars is high, 100%. We detect ∼ 218

all 10 known OB stars with spectral type earlier than B1 in the Pismis 24 cluster and all 9 OB stars with spectral types B0.5 or earlier in NGC 2244. The NGC 2244 stars show soft X-ray spectra that are consistent with X-rays being generated in microshocks in fast stellar winds, whereas the NGC 6357 O stars show hard X- ray emission component in addition to the soft emission. Very recently one of the O3 stars, Pismis 24-1 has been resolved as a binary (Ma´ız Apellániz et al. 2007). Anomalously hard X-ray emission has been detected in early O stars in other regions (e.g., M17, W3; Townsley 2006), which may be an indicator of binarity. We confirm the long-standing log(L /L ) 7 relation for the O stars using x bol ∼ − samples from NGC 6357, NGC 2244, and M 17 (Chapter 4). Large deviation from this correlation was found for the B stars. This result suggests that the observed X-ray emission from non-peculiar O-type stars (i.e., not colliding wind binary) is directly coupled to their wind properties. The detected emission from B stars are probably associated with a physical PMS companion. We also report the first detection of X-ray emission from an EGG object at the tip of a molecular pillar; this source is likely a B0-B2 protostar.

7. X-ray flaring is ubiquitous among the PMS stars. Several flaring sources exhibit 1 luminosities above log L 32.0 ergs s− , which is exceptionally high for PMS t,c ≥ stars, comparable to the strongest X-ray flares known in PMS stars (Chapter 3). Magnetic loops connecting the stellar photosphere and the circumstellar disk have been postulated and evidence for large flaring structure in ONC stars have been found (Favata et al. 2005). Such powerful X-ray flares may efficiently irradiate protoplanetary disks and have impact on the chemistry and mineralogy of protoplanetary disks where the planet formation occurs. It is interesting to note that very recently detection of 12.81 µm [Ne II] emission in four dust disks have been reported (Pascucci et al. 2007), matching theoretical predictions that stellar X-rays can partially ionize the gas in the disk surface and produce detectable [Ne II] lines (Glassgold et al. 2007). The X-ray flares may be responsible for the flash-melted chondrules found in meteorites; the MeV particles associated with the X-ray flares may produce the short-lived radioactive nuclides in the calcium-aluminum-rich in- clusions (CAIs), the origin of which is highly debated (Feigelson et al. 2002b).

8. Diffuse X-ray emission is present in the NGC 6357 region and in the NGC 2244 cluster. However, the derived luminosity of diffuse emission in NGC 6357 is consistent with the estimated level of integrated emission from the unresolved PMS stars. However, the extent of diffuse emission complements the cavity seen in optical and infrared nicely, which may suggest the presence of diffuse emission is real. The NGC 2244 diffuse emission cannot be fully attributed to unresolved fainter PMS stars, hence it is truly diffuse in nature. It is likely originated from the wind termination shocks when the fast OB stellar winds collide with each other or hit the ambient medium. The presence of diffuse X-ray emission will probably require certain kinematic energy input (i.e., enough number of early-type stars, see Townsley et al. 2003) and favorable geometric configurations (confinement in the HII region cavity). Note that the diffuse emission has very low luminosity, 0.1L⊙. ∼ 219

7.2 Implications for Star Formation

Our results, in particular the large scale distribution of clusters and the smooth- ness/clumpiness within the clusters shed some light on the formation of stars and clusters. Clustered and distributed star formation both exist in the giant molecular clouds, as indicated by the three large-scale structures seen as overdensities and a distributed population in the RMC. The cluster mode of star formation is dominant here in Rosette, similar to what we learned from nearby clouds (Lada et al. 1991). More embedded clusters with higher IR excess fractions were found in the south-east part of the RMC (beyond our FOV) (Román-Zúñiga et al. 2007a), suggesting a temporal sequence of star formation across the complex. There have been extensive discussions about the possible formation mechanism for these star clusters in the RMC (see Chapter 5). Our coverage on NGC 2244, RMC-A, RMC-B, RMC-C, and NGC 2237 is summarized in Figure 7.2, which provides a global view of the locations of X-ray selected clusters relative to the molecular materials traced by CO emission (Heyer et al. 2006). For reference, an X-ray image using ROSAT/PSPC archival data is also shown since we do not have full Chandra coverage of the large field. The central NGC 2244 cluster must have formed and created an expanding HII region. It is worth noting that RMC-A and NGC 2237 are very similar in terms of obscuration and IR-excess fraction, and comparable with NGC 2244. Both are located in the evacuated cavity around NGC 2244. They may have formed simultaneously with NGC 2244 as substructures in the primordial molecular cloud. The CO emission clearly outlines a large shell of dense molecular materials, which may have been swept-up by the expanding HII region and stellar winds from NGC 2244 OB stars. This large shell became dense enough and fragmented to form RMC-B. The substructures seen in RMC- B are closely associated with the dense molecular clumps. In the lower arc located in the south of NGC 2244 (possibly another fragment of the shell), another young embedded cluster has been discovered by Phelps & Lada (1997), which is also consistent with this large scale triggering process known as “collect-and-collapse”. The clumps inside the RMC are under influence of the shock front from the nebula; clusters are formed in the most massive cores and appear extended with significant gas removed (Román-Zúñiga et al. 2007a). However we cannot rule out the possibility that they formed as a result of collapse from the large scale turbulence in the primordial cloud. Measurements of the cluster ages and comparison with models and numerical simulations are needed to test this hypothesis. Nevertheless, the process of star formation here seems different from the original sequentially triggered formation scenario proposed by Elmegreen & Lada (1977), which requires formation of massive stars in the first triggered subgroup to trigger the formation of another OB subgroup. From the ROSAT images in Figure 7.2 and Figure 3.1, there are notable similari- ties between the Rosette complex and the NGC 6357 complex. Although their molecular clouds may not be the same configuration because of spatial projection, both have a ring like structure in the optical and have clusters in the peripheries as indicated by the X-ray emission. This seems to be a rather common setting for HII region complexes and can be used to learn about the extreme giant HII region complexes such as 30 Dor. In the Rosette Complex, the clusters further from the Rosette Nebula have increased IR excess 220 fraction and likely younger; the expanding HII region possibly triggered the formation of the clusters. However, the HII region G353.2+0.9 in NGC 6357 is the youngest and the other HII regions G353.1+0.6 and G353.2+0.7 are more evolved with no indication of current star formation (Felli et al. 1990; Massi et al. 1997). It is not clear whether there is triggering between the HII regions. But the temporal divergence among the clusters allows such interaction and might imply the large scale turbulence in the parental molecular cloud probably does not play an exclusive role in determining cluster and star formation in the clouds.

7.3 Future Work

7.3.1 A Synergy Study of Young Stellar Clusters with Two Great Observa- tories One of our findings, the low NIR-excess disk fractions among X-ray selected stars in MSFRs compared to the observed disk frequencies in LMSFRs (e.g., Haisch et al. 2001), is intriguing. Further investigation is warranted. Using NIR-excess has limited sensitivity in detecting inner protoplanetary disks; a disk fraction assessed with Spitzer mid-IR observations will be more sensitive. While many photometric surveys of protoplanetary disks are being pursued with Spitzer, most of the targeted samples suffer from a few critical limitations: they lack a complete, unbiased representation including the diskless cluster members that cannot be effectively identified by Spitzer, and spatially they focus on relatively isolated environments like the nearby Taurus-Auriga molecular cloud. Thus the inferences about disk fractions and evolution are consequently affected. As we have shown, X-ray sampling of young stars are particularly effective. A flux-limited X-ray survey of a star forming region approximately gives a complete sample of PMS stars down to the corresponding mass limit (e.g., 0.1M⊙ for COUP, 0.5M⊙ for NGC 2244, 1M⊙ for NGC 2237). Chandra images can reveal heavily obscured (up to A 500 mag) low-mass cloud populations at all PMS phases considerably deeper than V ≃ NIR surveys (Getman et al. 2005). Because they trace magnetic activity rather than photospheric or disk blackbody emission, X-ray surveys are not biased towards PMS stars with IR-luminous disks (Preibisch et al. 2005). X-ray selection also effectively discriminates all PMS phases from older contaminating Galactic field stars. To fully understand the prevalence and evolution of protoplanetary disks in different star forming environments, we need: (1) Spitzer surveys of large populations of young stars, which already exist. Stun- ning details of Galactic MSFRs have been revealed by Spitzer, and millions of stars are available to study protoplanetary disks. The Spitzer GLIMPSE and GLIMPSE II surveys now offer complete coverage of the inner Galaxy. Numerous HII regions, open clusters, previously uncatalogued star clusters, massive stars, and supernova remnants have been seen in the survey. GLIMPSE 3D is on the way to enlarge the sample for investigating off-plane star formation. (2) A census of the cluster members identified in a disk-unbiased fashion, which can be best achieved with high spatial resolution, high sensitivity X-ray observations. 221

The complementary nature of the Chandra and Spitzer data will provide the best census to date for the young stellar populations. We can now study characteristics of protoplanetary disks around X-ray selected young stars using both Spitzer and Chandra data, to improve our understanding in the early evolutionary stages of star formation, particularly the environmental effects on the prevalence and evolution of protoplanetary disks. We can determine how the ionizing UV radiation and stellar winds from the massive stars affect the formation and evolution of protoplanetary disks in their nearby intermediate- and low-mass siblings. This will be addressed by examining the disk fractions among massive clusters compared to those in nearby embedded low mass star clusters of similar age. We can also constrain the timescale of disk dissipation in intermediate- and low-mass stars in the vicinity of massive stars. This can be done by comparing the disk frequencies in massive star clusters of different ages. Last but not least, although the Spitzer mission is ending, the ground-based telescopes such as VLT and the upcoming James Webb Space Telescope (JWST) offer sensitive high resolution IR imaging capabilities, which ensures a promising future for combined IR and X-ray studies.

7.3.2 X-ray Studies of Structures of Young Stellar Clusters Because X-ray sampled young clusters represent a good census of the cluster members identified in a disk-unbiased manner and with low fraction of contamination, X- ray observations of YSCs offer a great method to study the spatial structure of embedded clusters and distribution of the young stars. This unique power has been demonstrated in individual chapters and certainly deserves further attention. There are ongoing Spitzer surveys of YSCs aimed to study the range of cluster morphologies and assess information imprinted by their birth process (see review by Allen et al. 2007). The identification of cluster members is limited to YSOs with disks and envelopes. Many clusters show elongated structure, suggesting that they were born from elongated or filamentary molecular clouds (Gutermuth et al. 2005). Young clusters with subclusters or multiple density peaks have been seen, which may imply clumpy structures in their birth clouds. With rich archival data available, we can systematically survey the X-ray morphologies of a large number of YSCs, not limited to the IR-excess sources. The inclusion of Class III sources should greatly improve the number statistics and help to enhance any substructures. The smoothing technique can be significantly improved by adopting sophisticated smoothing algorithms and statistical methods, to detect features in a robust and quantitative fashion.

7.3.3 X-ray Observations of Precursors to Star Clusters Looking forward, another research area that X-ray observations can play a significant role is the earliest stages of massive star and cluster formation. With the latest arguments that the infrared dark clouds (IRDCs) represent the earliest phases of the formation of star clusters and massive stars (Rathborne et al. 2006, 2007), it is crucial to 222 understand their behavior and properties to have a complete picture of the star formation process (see Menten et al. 2005 for the latest review). The only way to evaluate their importance is through measurements of their embedded stellar populations. To date the primary techniques used to study embedded stars in IRDCs have been IR imaging and maser emission, both of which are largely restricted to massive OB stars – only X-ray emission consistently traces the intermediate- and low-mass population. While we appreciate that, due to large distance and high obscuration, the resulting X-ray images are not likely to be visually spectacular; the detection of even a handful of X-ray-faint embedded members may critically test the importance of IRDCs in Galactic star formation. Complementary to IR and radio studies, which can only reveal massive embedded protostars, X-ray emission provides a unique window into the earliest phases of clustered and massive star formation in IRDCs by allowing us to study the low mass protostellar population that is expected to precede the appearance of the first massive stars. This will allow us to characterize the environment of cluster formation and massive star formation before it is altered by feedback from massive stars, determining basic parameters such as the ratio of gas mass to embedded stellar mass and the demographics and spatial distribution of the embedded protostellar population. These conditions are critical inputs to, and predictions of, theories of massive star and cluster formation for which X-ray observations can provide a first observational determination. For example, the collisional model of massive star formation (Bonnell et al. 1998; 6 8 Bonnell & Bate 2005) requires that embedded clusters collapse to densities of 10 10 −3 − low mass stars pc , and explains the lack of revealed clusters at such high densities by appealing to disruption and dispersal of mass by massive stellar feedback. X-ray observations of protoclusters that have not yet formed massive stars are the ideal way 1 to search for this phase and determine if it exists In contrast, Tan et al. (2006) and Krumholz & Tan (2007) argue that cluster formation is a quasi-equilibrium process that requires several crossing times to complete, and where only a few percent of the mass forms stars per crossing time. This model predicts a correlation between the star-to-gas ratio and the amount of substructure: systems with < 5% of the mass in stars are expected to be less than a crossing time old, and therefore still retain a lot of substructure; systems with > 10% of the mass in stars are older, and should be smoother. X-ray observations make it possible to test this prediction. If ambipolar diffusion is the key process in regulating star formation (e.g. Shu et al. 2004), the spatial distribution of embedded protostars may be sheet-like, and this too would be visible to X-ray observations. For future work, our goal is to investigate, for the first time with X-rays, the characteristics of the embedded stellar population in an IRDC. The XLF and star formation efficiency in IRDCs will be evaluated. With the joint-force of Spitzer data, the X-ray

1 Even for the nearest IRDCs (d 1 2 kpc), the very high stellar density of million stars −3 5 ∼ − 2 pc implies surface density 10 arcmin . Because of the high absorption column, only the most luminous low mass stars∼ (a few percent of the total population) may be detected. Crowding may not be a problem for Chandra but should be further evaluated. 223 observation can unambiguously address the roles of IRDCs in star formation and test current theories of the process of clustered and massive star formation. 224

Fig. 7.1 IR excess disk fraction as a function of cluster age. Solid triangles are L- band excess disk fractions from Haisch et al. (2001). Open diamonds are K-excess disk fractions compiled from literature data in Haisch et al. (2001). For NGC 2244, the triangle represents Spitzer IRAC 8µm excess disk fraction in Balog et al. (2007), the diamond represents K-excess disk fraction in Román-Zúñiga et al. (2007a), and the star represents K-excess disk fraction in X-ray selected sample in Wang et al. (2007b). 225

Fig. 7.2 A global view of the locations of X-ray selected clusters. The top panel shows the distribution of molecular materials traced by CO emission (Heyer et al. 2006). The lower panel shows large scale ROSAT/PSPC X-ray image revealing the concentration of X-ray emitting stars. Some clusters (e.g., RMC-B, RMC-C) are associated with the CO emission peaks, and others (e.g., NGC 2244, NGC 2237, and RMC-A) are mostly unobscured. ACIS coverage is outlined with the polygons. The ROSAT archive data set ID shown here is RP900555N00 (see Bergh¨ofer & Christian 2002). 226 Bibliography

Abbott, D. C., Torres, A. V., Bieging, J. H., & Churchwell, E. 1986, ApJ, 303, 239

Adams, F. C., Lada, C. J., & Shu, F. H. 1987, ApJ, 312, 788

Adams, J. D., Stauﬀer, J. R., Monet, D. G., Skrutskie, M. F., & Beichman, C. A. 2001, AJ, 121, 2053

Albacete Colombo, J. F., Flaccomio, E., Micela, G., Sciortino, S., & Damiani, F. 2007, A&A, 464, 211

Albacete Colombo, J. F., M´endez, M., & Morrell, N. I. 2003, MNRAS, 346, 704

Alexander, D. M., Bauer, F. E., Brandt, W. N., Schneider, D. P., Hornschemeier, A. E., Vignali, C., Barger, A. J., Broos, P. S., Cowie, L. L., Garmire, G. P., Townsley, L. K., Bautz, M. W., Chartas, G., & Sargent, W. L. W. 2003, AJ, 126, 539

Allen, L. 1996, PhD thesis, , University of Massachusetts, Amherst, MA, USA, (1996)

Allen, L., Megeath, S. T., Gutermuth, R., Myers, P. C., Wolk, S., Adams, F. C., Muze- rolle, J., Young, E., & Pipher, J. L. 2007, Protostars and Planets V, 361

Allen, L. E., Calvet, N., D’Alessio, P., Merin, B., Hartmann, L., Megeath, S. T., Guter- muth, R. A., Muzerolle, J., Pipher, J. L., Myers, P. C., & Fazio, G. G. 2004, ApJS, 154, 363

Alter, G., Balazs, B., Ruprecht, J., & Vanysek, J. 1970, Catalogue of star clusters and associations (Budapest: Akademiai Kiado, 1970, 2nd ed., edited by G. Alter, B. Balazs, J. Ruprecht)

Andre, P. & Montmerle, T. 1994, ApJ, 420, 837

Arias, J. I., Barb´a, R. H., Ma´ız Apell´aniz, J., Morrell, N. I., & Rubio, M. 2006, MNRAS, 366, 739

Arnaud, K. A. 1996, in Astronomical Society of the Paciﬁc Conference Series, Vol. 101, Astronomical Data Analysis Software and Systems V, ed. G. H. Jacoby & J. Barnes, 17

Babel, J. & Montmerle, T. 1997a, ApJL, 485, L29

—. 1997b, A&A, 323, 121

Bagnulo, S., Hensberge, H., Landstreet, J. D., Szeifert, T., & Wade, G. A. 2004, A&A, 416, 1149

Ballesteros-Paredes, J., Klessen, R. S., Mac Low, M.-M., & Vazquez-Semadeni, E. 2007, Protostars and Planets V, 63 227

Bally, J., Feigelson, E., & Reipurth, B. 2003, ApJ, 584, 843

Bally, J. & Zinnecker, H. 2005, AJ, 129, 2281

Balog, Z., Muzerolle, J., Rieke, G. H., Su, K. Y. L., Young, E. T., & Megeath, S. T. 2007, ApJ, 660, 1532

Balog, Z., Rieke, G. H., Su, K. Y. L., Muzerolle, J., & Young, E. T. 2006, ApJL, 650, L83

Baraﬀe, I., Chabrier, G., Allard, F., & Hauschildt, P. H. 1998, A&A, 337, 403

Bate, M. R., Bonnell, I. A., & Bromm, V. 2003, MNRAS, 339, 577

Becklin, E. E. & Neugebauer, G. 1967, ApJ, 147, 799

Benaglia, P., Cappa, C. E., & Koribalski, B. S. 2001, A&A, 372, 952

Benjamin, R. A., Churchwell, E., Babler, B. L., Bania, T. M., Clemens, D. P., Cohen, M., Dickey, J. M., Indebetouw, R., Jackson, J. M., Kobulnicky, H. A., Lazarian, A., Marston, A. P., Mathis, J. S., Meade, M. R., Seager, S., Stolovy, S. R., Watson, C., Whitney, B. A., Wolﬀ, M. J., & Wolﬁre, M. G. 2003, PASP, 115, 953

Berghoefer, T. W., Schmitt, J. H. M. M., & Cassinelli, J. P. 1996, A&AS, 118, 481

Berghoefer, T. W., Schmitt, J. H. M. M., Danner, R., & Cassinelli, J. P. 1997, A&A, 322, 167

Bergh¨ofer, T. W. & Christian, D. J. 2002, A&A, 384, 890

Berghofer, T. W. & Schmitt, J. H. M. M. 1994, Science, 265, 1689

Bertoldi, F. & McKee, C. F. 1996, in Amazing Light: A Volume Dedicated to C.H. Townes on his 80th Birthday, ed. R.Y.Chiao (New York: Springer), 41

Bessell, M. S. & Brett, J. M. 1988, PASP, 100, 1134

Beuther, H., Churchwell, E. B., McKee, C. F., & Tan, J. C. 2007, Protostars and Planets V, 165

Bica, E., Dutra, C. M., Soares, J., & Barbuy, B. 2003, A&A, 404, 223

Binney, J. & Tremaine, S. 1987, Galactic dynamics (Princeton, NJ, Princeton University Press, 1987, 747 p.)

Blaauw, A. 1961, Bulletin of the Astronomical Institutes of the Netherlands, 15, 265

—. 1964, ARAA, 2, 213

Blitz, L. 1993, in Protostars and Planets III, ed. E. H. Levy & J. I. Lunine, 125–161

Blitz, L. & Thaddeus, P. 1980, ApJ, 241, 676 228

Block, D. L. 1990, Nature, 347, 452

Block, D. L., Geballe, T. R., & Dyson, J. E. 1993, A&A, 273, L41

Bohigas, J., Tapia, M., Roth, M., & Ruiz, M. T. 2004, AJ, 127, 2826

Bonatto, C., Santos, Jr., J. F. C., & Bica, E. 2006, A&A, 445, 567

Bonnell, I. A. & Bate, M. R. 2005, MNRAS, 362, 915

Bonnell, I. A., Bate, M. R., Clarke, C. J., & Pringle, J. E. 1997, MNRAS, 285, 201

—. 2001, MNRAS, 323, 785

Bonnell, I. A., Bate, M. R., & Zinnecker, H. 1998, MNRAS, 298, 93

Bonnell, I. A. & Davies, M. B. 1998, MNRAS, 295, 691

Bonnell, I. A., Larson, R. B., & Zinnecker, H. 2007, Protostars and Planets V, 149

Bradt, H. V. & Kelley, R. L. 1979, ApJL, 228, L33

Brandt, W. N., Alexander, D. M., Hornschemeier, A. E., Garmire, G. P., Schneider, D. P., Barger, A. J., Bauer, F. E., Broos, P. S., Cowie, L. L., Townsley, L. K., Burrows, D. N., Chartas, G., Feigelson, E. D., Griﬃths, R. E., Nousek, J. A., & Sargent, W. L. W. 2001, AJ, 122, 2810

Brice˜no, C., Preibisch, T., Sherry, W. H., Mamajek, E. A., Mathieu, R. D., Walter, F. M., & Zinnecker, H. 2007, Protostars and Planets V, 345

Brinkman, A. C., Behar, E., G¨udel, M., Audard, M., den Boggende, A. J. F., Branduardi- Raymont, G., Cottam, J., Erd, C., den Herder, J. W., Jansen, F., Kaastra, J. S., Kahn, S. M., Mewe, R., Paerels, F. B. S., Peterson, J. R., Rasmussen, A. P., Sakelliou, I., & de Vries, C. 2001, A&A, 365, L324

Broos, P., Townsley, L., Bauer, F., & Getman, K. 2002, ACIS Extract, An ACIS Point Source Extraction Package (University Park: The Pennsylvania State Univ.) http: //www.astro.psu.edu/xray/docs/TARA/ae users guide.html

Broos, P. S., Feigelson, E. D., Townsley, L. K., Getman, K. V., Wang, J., Garmire, G. P., Jiang, Z., & Tsuboi, Y. 2007, ApJS, 169, 353

Burningham, B., Naylor, T., Littlefair, S. P., & Jeﬀries, R. D. 2005, MNRAS, 363, 1389

Carpenter, J. M. 2000, AJ, 120, 3139

—. 2001, AJ, 121, 2851

Carpenter, J. M., Meyer, M. R., Dougados, C., Strom, S. E., & Hillenbrand, L. A. 1997, AJ, 114, 198

Casertano, S. & Hut, P. 1985, ApJ, 298, 80 229

Cash, W. 1979, ApJ, 228, 939

Celnik, W. E. 1985, A&A, 144, 171

—. 1986, A&A, 160, 287

Chen, W. P., Chiang, P. S., & Li, J. Z. 2004, Chinese Journal of Astronomy and Astro- physics, 4, 153

Chlebowski, T., Harnden, Jr., F. R., & Sciortino, S. 1989, ApJ, 341, 427

Churchwell, E. 2002, ARAA, 40, 27

Churchwell, E., Whitney, B. A., Babler, B. L., Indebetouw, R., Meade, M. R., Watson, C., Wolff, M. J., Wolfire, M. G., Bania, T. M., Benjamin, R. A., Clemens, D. P., Cohen, M., Devine, K. E., Dickey, J. M., Heitsch, F., Jackson, J. M., Kobulnicky, H. A., Marston, A. P., Mathis, J. S., Mercer, E. P., Stauffer, J. R., & Stolovy, S. R. 2004, ApJS, 154, 322

Clarke, C. J., Bonnell, I. A., & Hillenbrand, L. A. 2000, Protostars and Planets IV, 151

Comer´on, F., Schneider, N., & Russeil, D. 2005, A&A, 433, 955

Corcoran, M. F. 1999, in Revista Mexicana de Astronomia y Astrofisica Conference Series, Vol. 8, Revista Mexicana de Astronomia y Astrofisica Conference Series, ed. N. I. Morrell, V. S. Niemela, & R. H. Barbá, 131–136

Corcoran, M. F., Waldron, W. L., Macfarlane, J. J., Chen, W., Pollock, A. M. T., Torii, K., Kitamoto, S., Miura, N., Egoshi, M., & Ohno, Y. 1994, ApJL, 436, L95

Cox, A. N. 2000, Allen’s astrophysical quantities (Allen’s astrophysical quantities, 4th ed. Publisher: New York: AIP Press; Springer, 2000. Editedy by Arthur N. Cox. ISBN: 0387987460)

Cox, P., Deharveng, L., & Leene, A. 1990, A&A, 230, 181

Cutri, R. M., Skrutskie, M. F., van Dyk, S., Beichman, C. A., Carpenter, J. M., Chester, T., Cambresy, L., Evans, T., Fowler, J., Gizis, J., Howard, E., Huchra, J., Jarrett, T., Kopan, E. L., Kirkpatrick, J. D., Light, R. M., Marsh, K. A., McCallon, H., Schneider, S., Stiening, R., Sykes, M., Weinberg, M., Wheaton, W. A., Wheelock, S., & Zacarias, N. 2003, 2MASS All Sky Catalog of point sources. (The IRSA 2MASS All-Sky Point Source Catalog, NASA/IPAC Infrared Science Archive. http://irsa.ipac.caltech.edu/applications/Gator/)

Dale, J. E., Bonnell, I. A., & Whitworth, A. P. 2007, MNRAS, 375, 1291

Damiani, F., Micela, G., Sciortino, S., & Harnden, Jr., F. R. 1994, ApJ, 436, 807

Damiani, F., Micela, G., Sciortino, S., Hu´elamo, N., Moitinho, A., Harnden, Jr., F. R., & Murray, S. S. 2006, A&A, 460, 133 230

Davies, M. B., Amaro-Seoane, P., Bassa, C., Dale, J., de Angeli, F., Freitag, M., Kroupa, P., Mackey, D., Miller, M. C., & Portegies Zwart, S. 2006, New Astronomy, 12, 201 de Jager, C. & Nieuwenhuijzen, H. 1987, A&A, 177, 217

De Marco, O., O’Dell, C. R., Gelfond, P., Rubin, R. H., & Glover, S. C. O. 2006, AJ, 131, 2580

Deharveng, L., Leﬂoch, B., Massi, F., Brand, J., Kurtz, S., Zavagno, A., & Caplan, J. 2006, A&A, 458, 191

Deharveng, L., Zavagno, A., & Caplan, J. 2005, A&A, 433, 565 den Boggende, A. J. F., Mewe, R., Gronenschild, E. H. B. M., Heise, J., & Grindlay, J. E. 1978, A&A, 62, 1

Dias, W. S., Assaﬁn, M., Fl´orio, V., Alessi, B. S., & L´ıbero, V. 2006, A&A, 446, 949

Dickey, J. M. & Lockman, F. J. 1990, ARAA, 28, 215

Diggle, P. 2003, Statistical Analysis of Spatial Point Patterns (2nd Edition, Hodder Arnold Publication, 2003, 159 p.)

Dorland, H., Montmerle, T., & Doom, C. 1986, A&A, 160, 1

Dougherty, S. M., Beasley, A. J., Claussen, M. J., Zauderer, B. A., & Bolingbroke, N. J. 2005, ApJ, 623, 447

Drake, J. J. 2002, in Astronomical Society of the Paciﬁc Conference Series, Vol. 277, Stellar Coronae in the Chandra and XMM-NEWTON Era, ed. F. Favata & J. J. Drake, 75

Droege, T. F., Richmond, M. W., Sallman, M. P., & Creager, R. P. 2006, PASP, 118, 1666

Dutra, C. M., Bica, E., Soares, J., & Barbuy, B. 2003, A&A, 400, 533

Ebeling, H., White, D. A., & Rangarajan, F. V. N. 2006, MNRAS, 368, 65

Egan, M. P., Shipman, R. F., Price, S. D., Carey, S. J., Clark, F. O., & Cohen, M. 1998, ApJL, 494, L199

Elmegreen, B. G. 1998, in Astronomical Society of the Paciﬁc Conference Series, Vol. 148, Origins, ed. C. E. Woodward, J. M. Shull, & H. A. Thronson, Jr., 150

Elmegreen, B. G. 2000, ApJ, 530, 277

Elmegreen, B. G. 2005, in Astrophysics and Space Science Library, Vol. 329, Astrophysics and Space Science Library, ed. R. de Grijs & R. M. Gonz´alez Delgado, 57

Elmegreen, B. G. & Lada, C. J. 1977, ApJ, 214, 725 231

Evans, II, N. J., Allen, L. E., Blake, G. A., Boogert, A. C. A., Bourke, T., Harvey, P. M., Kessler, J. E., Koerner, D. W., Lee, C. W., Mundy, L. G., Myers, P. C., Padgett, D. L., Pontoppidan, K., Sargent, A. I., Stapelfeldt, K. R., van Dishoeck, E. F., Young, C. H., & Young, K. E. 2003a, PASP, 115, 965

Evans, N. R., Schlegel, E. M., Waldron, W. L., Seward, F. D., Krauss, M. I., Nichols, J., & Wolk, S. J. 2004, ApJ, 612, 1065

Evans, N. R., Seward, F. D., Krauss, M. I., Isobe, T., Nichols, J., Schlegel, E. M., & Wolk, S. J. 2003b, ApJ, 589, 509

Ezoe, Y., Kokubun, M., Makishima, K., Sekimoto, Y., & Matsuzaki, K. 2006a, ApJ, 638, 860

—. 2006b, ApJL, 649, L123

Favata, F., Flaccomio, E., Reale, F., Micela, G., Sciortino, S., Shang, H., Stassun, K. G., & Feigelson, E. D. 2005, ApJS, 160, 469

Favata, F., Giardino, G., Micela, G., Sciortino, S., & Damiani, F. 2003, A&A, 403, 187

Favata, F. & Micela, G. 2003, Space Science Reviews, 108, 577

Feigelson, E., Townsley, L., G¨udel, M., & Stassun, K. 2007, in Protostars and Planets V, ed. B. Reipurth, D. Jewitt, & K. Keil, 313–328

Feigelson, E. D., Broos, P., Gaﬀney, III, J. A., Garmire, G., Hillenbrand, L. A., Pravdo, S. H., Townsley, L., & Tsuboi, Y. 2002a, ApJ, 574, 258

Feigelson, E. D. & Decampli, W. M. 1981, ApJL, 243, L89

Feigelson, E. D., Garmire, G. P., & Pravdo, S. H. 2002b, ApJ, 572, 335

Feigelson, E. D., Getman, K., Townsley, L., Garmire, G., Preibisch, T., Grosso, N., Montmerle, T., Muench, A., & McCaughrean, M. 2005, ApJS, 160, 379

Feigelson, E. D., Jackson, J. M., Mathieu, R. D., Myers, P. C., & Walter, F. M. 1987, AJ, 94, 1251

Feigelson, E. D. & Lawson, W. A. 2004, ApJ, 614, 267

Feigelson, E. D. & Montmerle, T. 1999, ARAA, 37, 363

Feldman, U. 1992, Physica Scripta Volume T, 46, 202

Felli, M., Persi, P., Roth, M., Tapia, M., Ferrari-Toniolo, M., & Cervelli, A. 1990, A&A, 232, 477

Forrest, W. J., Sargent, B., Furlan, E., D’Alessio, P., Calvet, N., Hartmann, L., Uchida, K. I., Green, J. D., Watson, D. M., Chen, C. H., Kemper, F., Keller, L. D., Sloan, G. C., Herter, T. L., Brandl, B. R., Houck, J. R., Barry, D. J., Hall, P., Morris, P. W., Najita, J., & Myers, P. C. 2004, ApJS, 154, 443 232

Freeman, P. E., Kashyap, V., Rosner, R., & Lamb, D. Q. 2002, ApJS, 138, 185

Froebrich, D., Scholz, A., Eisl¨oﬀel, J., & Murphy, G. C. 2005, A&A, 432, 575

Frogel, J. A. & Persson, S. E. 1974, ApJ, 192, 351

Gagn´e, M., Oksala, M. E., Cohen, D. H., Tonnesen, S. K., ud-Doula, A., Owocki, S. P., Townsend, R. H. D., & MacFarlane, J. J. 2005, ApJ, 628, 986

Garcia-Segura, G. & Franco, J. 1996, ApJ, 469, 171

Garmire, G. P., Bautz, M. W., Ford, P. G., Nousek, J. A., & Ricker, Jr., G. R. 2003, in Presented at the Society of Photo-Optical Instrumentation Engineers (SPIE) Confer- ence, Vol. 4851, X-Ray and Gamma-Ray Telescopes and Instruments for Astronomy. Edited by Joachim E. Truemper, Harvey D. Tananbaum. Proceedings of the SPIE, Volume 4851, pp. 28-44 (2003)., ed. J. E. Truemper & H. D. Tananbaum, 28–44

Gaustad, J. E., Rosing, W., McCullough, P. R., & van Buren, D. 1999, in IAU Sym- posium, Vol. 190, New Views of the Magellanic Clouds, ed. Y.-H. Chu, N. Suntzeﬀ, J. Hesser, & D. Bohlender, 99

Getman, K. V., Feigelson, E. D., Garmire, G., Broos, P., & Wang, J. 2007, ApJ, 654, 316

Getman, K. V., Feigelson, E. D., Grosso, N., McCaughrean, M. J., Micela, G., Broos, P., Garmire, G., & Townsley, L. 2005a, ApJS, 160, 353

Getman, K. V., Feigelson, E. D., Townsley, L., Broos, P., Garmire, G., & Tsujimoto, M. 2006, ApJS, 163, 306

Getman, K. V., Flaccomio, E., Broos, P. S., Grosso, N., Tsujimoto, M., Townsley, L., Garmire, G. P., Kastner, J., Li, J., Harnden, Jr., F. R., Wolk, S., Murray, S. S., Lada, C. J., Muench, A. A., McCaughrean, M. J., Meeus, G., Damiani, F., Micela, G., Sciortino, S., Bally, J., Hillenbrand, L. A., Herbst, W., Preibisch, T., & Feigelson, E. D. 2005b, ApJS, 160, 319

Giacconi, R., Murray, S., Gursky, H., Kellogg, E., Schreier, E., Matilsky, T., Koch, D., & Tananbaum, H. 1974, ApJS, 27, 37

Giacconi, R., Murray, S., Gursky, H., Kellogg, E., Schreier, E., & Tananbaum, H. 1972, ApJ, 178, 281

Gies, D. R. 1987, ApJS, 64, 545

Glassgold, A. E., Najita, J. R., & Igea, J. 2007, ApJ, 656, 515

Greene, T. P., Wilking, B. A., Andre, P., Young, E. T., & Lada, C. J. 1994, ApJ, 434, 614

Gregorio-Hetem, J., Montmerle, T., Casanova, S., & Feigelson, E. D. 1998, A&A, 331, 193 233

Grosso, N., Feigelson, E. D., Getman, K. V., Townsley, L., Broos, P., Flaccomio, E., McCaughrean, M. J., Micela, G., Sciortino, S., Bally, J., Smith, N., Muench, A. A., Garmire, G. P., & Palla, F. 2005, ApJS, 160, 530

Grosso, N., Montmerle, T., Feigelson, E. D., & Forbes, T. G. 2004, A&A, 419, 653

G¨udel, M. 2004, A&ARv, 12, 71

G¨udel, M., Guinan, E. F., & Skinner, S. L. 1998, in Astronomical Society of the Pa- ciﬁc Conference Series, Vol. 154, Cool Stars, Stellar Systems, and the Sun, ed. R. A. Donahue & J. A. Bookbinder, 1041

G¨udel, M., Padgett, D. L., & Dougados, C. 2007, in Protostars and Planets V, ed. B. Reipurth, D. Jewitt, & K. Keil, 329–344

Gutermuth, R. A., Megeath, S. T., Pipher, J. L., Williams, J. P., Allen, L. E., Myers, P. C., & Raines, S. N. 2005, ApJ, 632, 397

Haisch, Jr., K. E., Lada, E. A., & Lada, C. J. 2001, AJ, 121, 2065

Hamaguchi, K., Yamauchi, S., & Koyama, K. 2005, ApJ, 618, 360

Hanson, M. M., Geballe, T. R., Conti, P. S., & Block, D. L. 1993, A&A, 273, L44

Hanson, M. M., Howarth, I. D., & Conti, P. S. 1997, ApJ, 489, 698

Harnden, Jr., F. R., Branduardi, G., Gorenstein, P., Grindlay, J., Rosner, R., Topka, K., Elvis, M., Pye, J. P., & Vaiana, G. S. 1979, ApJL, 234, L51

Harries, T. J. & Hilditch, R. W. 1998, in Astronomical Society of the Paciﬁc Conference Series, Vol. 131, Properties of Hot Luminous Stars, ed. I. Howarth, 401

Hartmann, L. 2001, AJ, 121, 1030

Hartmann, L. & Burkert, A. 2007, ApJ, 654, 988

Hartmann, L., Megeath, S. T., Allen, L., Luhman, K., Calvet, N., D’Alessio, P., Franco- Hernandez, R., & Fazio, G. 2005, ApJ, 629, 881

Haynes, R. F., Caswell, J. L., & Simons, L. W. J. 1978, Australian Journal of Physics Astrophysical Supplement, 45, 1

Healy, K. R., Hester, J. J., & Claussen, M. J. 2004a, ApJ, 610, 835

Healy, K. R., Hester, J. J., Scowen, P. A., & Snider, K. D. 2004b, in Bulletin of the American Astronomical Society, Vol. 36, Bulletin of the American Astronomical Soci- ety, 1517

Heap, S. R., Lanz, T., & Hubeny, I. 2006, ApJ, 638, 409

Hensberge, H., Pavlovski, K., & Verschueren, W. 2000, A&A, 358, 553 234

Herbig, G. H. 1960, ApJS, 4, 337

—. 1962, ApJ, 135, 736

Herbig, G. H. & Bell, K. R. 1988, Catalog of emission line stars of the orion population : 3 : 1988 (Lick Observatory Bulletin, Santa Cruz: Lick Observatory, —c1988)

Hern´andez, J., Calvet, N., Hartmann, L., Brice˜no, C., Sicilia-Aguilar, A., & Berlind, P. 2005, AJ, 129, 856

Hernández, J., Hartmann, L., Megeath, T., Gutermuth, R., Muzerolle, J., Calvet, N., Vivas, A. K., Briceño, C., Allen, L., Stauffer, J., Young, E., & Fazio, G. 2007, ApJ, 662, 1067

Hester, J. J. & Desch, S. J. 2005a, in Astronomical Society of the Paciﬁc Conference Series, Vol. 341, Chondrites and the Protoplanetary Disk, ed. A. N. Krot, E. R. D. Scott, & B. Reipurth, 107

Hester, J. J. & Desch, S. J. 2005b, in Astronomical Society of the Paciﬁc Conference Series, Vol. 341, Chondrites and the Protoplanetary Disk, ed. A. N. Krot, E. R. D. Scott, & B. Reipurth, 107

Hester, J. J., Desch, S. J., Healy, K. R., & Leshin, L. A. 2004, Science, 304, 1116

Hester, J. J., Scowen, P. A., Sankrit, R., Lauer, T. R., Ajhar, E. A., Baum, W. A., Code, A., Currie, D. G., Danielson, G. E., Ewald, S. P., Faber, S. M., Grillmair, C. J., Groth, E. J., Holtzman, J. A., Hunter, D. A., Kristian, J., Light, R. M., Lynds, C. R., Monet, D. G., O’Neil, Jr., E. J., Shaya, E. J., Seidelmann, K. P., & Westphal, J. A. 1996, AJ, 111, 2349

Heyer, M. H., Williams, J. P., & Brunt, C. M. 2006, ApJ, 643, 956

Hillenbrand, L. A. 1997, AJ, 113, 1733

Hillenbrand, L. A. & Hartmann, L. W. 1998, ApJ, 492, 540

Hillenbrand, L. A., Massey, P., Strom, S. E., & Merrill, K. M. 1993, AJ, 106, 1906

Hillenbrand, L. A. & White, R. J. 2004, ApJ, 604, 741

Hirota, T., Bushimata, T., Choi, Y. K., Honma, M., Imai, H., Iwadate, K., Jike, T., Kameno, S., Kameya, O., Kamohara, R., Kan-ya, Y., Kawaguchi, N., Kijima, M., Kim, M. K., Kobayashi, H., Kuji, S., Kurayama, T., Manabe, S., Maruyama, K., Matsui, M., Matsumoto, N., Miyaji, T., Nagayama, T., Nakagawa, A., Nakamura, K., Oh, C. S., Omodaka, T., Oyama, T., Sakai, S., Sasao, T., Sato, K., Sato, M., Shibata, K. M., Shintani, M., Tamura, Y., Tsushima, M., & Yamashita, K. 2007, astro-ph/0705.3792

Hoare, M. G., Kurtz, S. E., Lizano, S., Keto, E., & Hofner, P. 2007, Protostars and Planets V, 181 235

Hollenbach, D. J., Yorke, H. W., & Johnstone, D. 2000, Protostars and Planets IV, 401

Huenemoerder, D. P., Canizares, C. R., & Schulz, N. S. 2001, ApJ, 559, 1135

Huensch, M., Schmitt, J. H. M. M., Schroeder, K.-P., & Reimers, D. 1996, A&A, 310, 801

Huensch, M., Schmitt, J. H. M. M., & Voges, W. 1998, A&AS, 132, 155

Huﬀ, E. M. & Stahler, S. W. 2006, ApJ, 644, 355

H¨unsch, M., Schmitt, J. H. M. M., Sterzik, M. F., & Voges, W. 1999, A&AS, 135, 319

Imanishi, K., Koyama, K., & Tsuboi, Y. 2001, ApJ, 557, 747

Isobe, T., Feigelson, E. D., & Nelson, P. I. 1986, ApJ, 306, 490

Jeﬀries, R. D. 2007, MNRAS, 376, 1109

Jessop, N. E. & Ward-Thompson, D. 2001, MNRAS, 323, 1025

Jiang, Z., Yao, Y., Yang, J., Ando, M., Kato, D., Kawai, T., Kurita, M., Nagata, T., Nagayama, T., Nakajima, Y., Nagashima, C., Sato, S., Tamura, M., Nakaya, H., & Sugitani, K. 2002, ApJ, 577, 245

Johnstone, D., Hollenbach, D., & Bally, J. 1998, ApJ, 499, 758

Joye, W. A. & Mandel, E. 2003, in Astronomical Society of the Paciﬁc Conference Series, Vol. 295, Astronomical Data Analysis Software and Systems XII, ed. H. E. Payne, R. I. Jedrzejewski, & R. N. Hook, 489

Kant, I. 1755, Allgemeine Naturgeschichte und Theorie des Himmels (Zeitz, Bei W. Webel, 1798. Neue auﬂ.)

Kastner, J. H., Huenemoerder, D. P., Schulz, N. S., Canizares, C. R., & Weintraub, D. A. 2002, ApJ, 567, 434

Kenyon, S. J., Calvet, N., & Hartmann, L. 1993, ApJ, 414, 676

Kenyon, S. J. & Hartmann, L. 1995, ApJS, 101, 117

Kharchenko, N. V., Piskunov, A. E., R¨oser, S., Schilbach, E., & Scholz, R.-D. 2005, A&A, 438, 1163

Kitamoto, S. & Mukai, K. 1996, PASJ, 48, 813

Kouwenhoven, M. B. N., Brown, A. G. A., Portegies Zwart, S. F., & Kaper, L. 2007, A&A, 474, 77

Kraus, S., Balega, Y. Y., Berger, J.-P., Hofmann, K.-H., Millan-Gabet, R., Monnier, J. D., Ohnaka, K., Pedretti, E., Preibisch, T., Schertl, D., Schloerb, F. P., Traub, W. A., & Weigelt, G. 2007, A&A, 466, 649 236

Kroupa, P., Aarseth, S., & Hurley, J. 2001, MNRAS, 321, 699

Kroupa, P. & Boily, C. M. 2002, MNRAS, 336, 1188

Krumholz, M. R. & Tan, J. C. 2007, ApJ, 654, 304

Ku, W. H.-M. & Chanan, G. A. 1979, ApJL, 234, L59

Kurtz, S., Cesaroni, R., Churchwell, E., Hofner, P., & Walmsley, C. M. 2000, Protostars and Planets IV, 299

Kurtz, S. & Hofner, P. 2005, AJ, 130, 711

Lada, C. J. 1987, in IAU Symposium, Vol. 115, Star Forming Regions, ed. M. Peimbert & J. Jugaku, 1–17

Lada, C. J. & Adams, F. C. 1992, ApJ, 393, 278

Lada, C. J. & Lada, E. A. 2003, ARAA, 41, 57

Lada, C. J., Muench, A. A., Haisch, Jr., K. E., Lada, E. A., Alves, J. F., Tollestrup, E. V., & Willner, S. P. 2000, AJ, 120, 3162

Lada, C. J., Muench, A. A., Lada, E. A., & Alves, J. F. 2004, AJ, 128, 1254

Lada, C. J., Muench, A. A., Luhman, K. L., Allen, L., Hartmann, L., Megeath, T., Myers, P., Fazio, G., Wood, K., Muzerolle, J., Rieke, G., Siegler, N., & Young, E. 2006, AJ, 131, 1574

Lada, E. A., Evans, II, N. J., Depoy, D. L., & Gatley, I. 1991, ApJ, 371, 171

Lada, E. A. & Lada, C. J. 1995, AJ, 109, 1682

Lada, E. A., Strom, K. M., & Myers, P. C. 1993, in Protostars and Planets III, ed. E. H. Levy & J. I. Lunine, 245–277

Leahy, D. A. 1985, MNRAS, 217, 69

Lee, H.-T. & Chen, W. P. 2007, ApJ, 657, 884

Li, J.-Z. 2003, Chinese Journal of Astronomy and Astrophysics, 3, 495

Li, J. Z. 2005, ApJ, 625, 242

Li, J. Z., Chu, Y. ., & Gruendl, R. A. 2006, astro-ph/0602456

Li, J. Z. & Rector, T. A. 2004, ApJL, 600, L67

Li, J. Z. & Smith, M. 2005a, AJ, 130, 721

Li, J. Z. & Smith, M. D. 2005b, ApJ, 620, 816

—. 2005c, A&A, 431, 925 237

Li, J. Z., Wu, C. H., Chen, W. P., Rector, T., Chu, Y. H., & Ip, W. H. 2002, AJ, 123, 2590

Li, W., Evans, II, N. J., & Lada, E. A. 1997, ApJ, 488, 277

Linsky, J. L., Gagne, M., & Mytyk, A. 2003, in IAU Symposium, Vol. 221, IAU Sympo- sium, 207P

Linsky, J. L., Gagn´e, M., Mytyk, A., McCaughrean, M., & Andersen, M. 2007, ApJ, 654, 347

Lortet, M. C., Testor, G., & Niemela, V. 1984, A&A, 140, 24

Lucy, L. B. 1974, AJ, 79, 745

Lucy, L. B. & White, R. L. 1980, ApJ, 241, 300

Luhman, K. L. 1999, ApJ, 525, 466

Luhman, K. L., Joergens, V., Lada, C., Muzerolle, J., Pascucci, I., & White, R. 2007, Protostars and Planets V, 443

Macfarlane, J. J., Waldron, W. L., Corcoran, M. F., Wolﬀ, M. J., Wang, P., & Cassinelli, J. P. 1993, ApJ, 419, 813

Maggio, A., Flaccomio, E., Favata, F., Micela, G., Sciortino, S., Feigelson, E. D., & Getman, K. V. 2007, ApJ, 660, 1462

Ma´ız-Apell´aniz, J., Walborn, N. R., Galu´e, H. A.,´ & Wei, L. H. 2004, ApJS, 151, 103

Ma´ız Apell´aniz, J., Walborn, N. R., Morrell, N. I., Niemela, V. S., & Nelan, E. P. 2007, ApJ, 660, 1480

Marschall, L. A., van Altena, W. F., & Chiu, L.-T. G. 1982, AJ, 87, 1497

Martins, F., Schaerer, D., & Hillier, D. J. 2005, in SF2A-2005: Semaine de l’Astrophysique Francaise, ed. F. Casoli, T. Contini, J. M. Hameury, & L. Pagani, 633

Massey, P. 1998, in Astronomical Society of the Paciﬁc Conference Series, Vol. 142, The Stellar Initial Mass Function (38th Herstmonceux Conference), ed. G. Gilmore & D. Howell, 17

Massey, P., DeGioia-Eastwood, K., & Waterhouse, E. 2001, AJ, 121, 1050

Massey, P. & Johnson, J. 1993, AJ, 105, 980

Massey, P., Johnson, K. E., & Degioia-Eastwood, K. 1995, ApJ, 454, 151

Massi, F., Brand, J., & Felli, M. 1997, A&A, 320, 972

McBreen, B., Jaﬀe, D. T., & Fazio, G. G. 1983, AJ, 88, 835 238

McCullough, P. R. 2000, PASP, 112, 1542

McKee, C. F. 1989, ApJ, 345, 782

McKee, C. F. & Ostriker, E. C. 2007, ARAA, 45, 565

McKee, C. F. & Tan, J. C. 2002, Nature, 416, 59

McMillan, S. L. W., Vesperini, E., & Portegies Zwart, S. F. 2007, ApJL, 655, L45

Meaburn, J., L´opez, J. A., Richer, M. G., Riesgo, H., & Dyson, J. E. 2005, AJ, 130, 730

Meaburn, J. & Walsh, J. R. 1986, MNRAS, 220, 745

Menten, K. M., Reid, M. J., Forbrich, J., & Brunthaler, A. 2007, A&A, 474, 515

Meyer, M. R., Calvet, N., & Hillenbrand, L. A. 1997, AJ, 114, 288

Mezger, P. G., Altenhoﬀ, W., Schraml, J., Burke, B. F., Reifenstein, III, E. C., & Wilson, T. L. 1967, ApJL, 150, L157

Moﬀat, A. F. J. & Vogt, N. 1973, A&AS, 10, 135

Monet, D. G., Levine, S. E., Canzian, B., Ables, H. D., Bird, A. R., Dahn, C. C., Guetter, H. H., Harris, H. C., Henden, A. A., Leggett, S. K., Levison, H. F., Luginbuhl, C. B., Martini, J., Monet, A. K. B., Munn, J. A., Pier, J. R., Rhodes, A. R., Riepe, B., Sell, S., Stone, R. C., Vrba, F. J., Walker, R. L., Westerhout, G., Brucato, R. J., Reid, I. N., Schoening, W., Hartley, M., Read, M. A., & Tritton, S. B. 2003, AJ, 125, 984

Moretti, A., Campana, S., Lazzati, D., & Tagliaferri, G. 2003, ApJ, 588, 696

Mori, K., Tsunemi, H., Miyata, E., Baluta, C. J., Burrows, D. N., Garmire, G. P., & Chartas, G. 2001, in Astronomical Society of the Paciﬁc Conference Series, Vol. 251, New Century of X-ray Astronomy, ed. H. Inoue & H. Kunieda, 576

Muench, A. A., Lada, E. A., Lada, C. J., & Alves, J. 2002, ApJ, 573, 366

Muno, M. P., Bauer, F. E., Bandyopadhyay, R. M., & Wang, Q. D. 2006, ApJS, 165, 173

Muzerolle, J., Calvet, N., & Hartmann, L. 1998, ApJ, 492, 743

—. 2001, ApJ, 550, 944

Nagayama, T., Nagashima, C., Nakajima, Y., Nagata, T., Sato, S., Nakaya, H., Yama- muro, T., Sugitani, K., & Tamura, M. 2003, in Proceedings of the Society of Photo- Optical Instrumentation Engineers (SPIE) Conference, Vol. 4841, Instrument Design and Performance for Optical/Infrared Ground-based Telescopes., ed. M. Iye & A. F. M. Moorwood, 459–464

Natta, A., Grinin, V., & Mannings, V. 2000, Protostars and Planets IV, 559 239

Neckel, T. 1978, A&A, 69, 51

—. 1984, A&A, 137, 58

Neuhaeuser, R., Sterzik, M. F., Schmitt, J. H. M. M., Wichmann, R., & Krautter, J. 1995, A&A, 297, 391

Nugis, T., Crowther, P. A., & Willis, A. J. 1998, A&A, 333, 956

O’dell, C. R., Wen, Z., & Hu, X. 1993, ApJ, 410, 696

O’dell, C. R. & Wong, K. 1996, AJ, 111, 846

Oey, M. S., Watson, A. M., Kern, K., & Walth, G. L. 2005, AJ, 129, 393

Ogura, K. & Ishida, K. 1981, PASJ, 33, 149

Oskinova, L. M., Ignace, R., Hamann, W.-R., Pollock, A. M. T., & Brown, J. C. 2003, A&A, 402, 755

Owocki, S. P., Castor, J. I., & Rybicki, G. B. 1988, ApJ, 335, 914

Owocki, S. P. & Cohen, D. H. 1999, ApJ, 520, 833

Padgett, D. L., Brandner, W., Stapelfeldt, K. R., Strom, S. E., Terebey, S., & Koerner, D. 1999, AJ, 117, 1490

Palla, F., Randich, S., Pavlenko, Y. V., Flaccomio, E., & Pallavicini, R. 2007, ApJL, 659, L41

Palla, F. & Stahler, S. W. 2000, ApJ, 540, 255

Pallavicini, R., Golub, L., Rosner, R., Vaiana, G. S., Ayres, T., & Linsky, J. L. 1981, ApJ, 248, 279

Park, B.-G. & Sung, H. 2002, AJ, 123, 892

Pascucci, I., Hollenbach, D., Najita, J., Muzerolle, J., Gorti, U., Herczeg, G. J., Hil- lenbrand, L. A., Kim, J. S., Carpenter, J. M., Meyer, M. R., Mamajek, E. E., & Bouwman, J. 2007, ApJ, 663, 383

Perault, M., Omont, A., Simon, G., Seguin, P., Ojha, D., Blommaert, J., Felli, M., Gilmore, G., Guglielmo, F., Habing, H., Price, S., Robin, A., de Batz, B., Cesarsky, C., Elbaz, D., Epchtein, N., Fouque, P., Guest, S., Levine, D., Pollock, A., Prusti, T., Siebenmorgen, R., Testi, L., & Tiphene, D. 1996, A&A, 315, L165

Perez, M. R. 1991, Revista Mexicana de Astronomia y Astroﬁsica, 22, 99

Persi, P., Ferrari-Toniolo, M., Roth, M., & Tapia, M. 1986, A&A, 170, 97

Persi, P. & Rosa Marenzi, A. 2006, Chinese Journal of Astronomy and Astrophysics Supplement, 6, 125 240

Pﬂamm-Altenburg, J. & Kroupa, P. 2006, MNRAS, 373, 295

Phelps, R. L. & Lada, E. A. 1997, ApJ, 477, 176

Philip, A. D. & Egret, D. 1980, A&AS, 40, 199

Pismis, P. 1959, Boletin de los Observatorios Tonantzintla y Tacubaya, 2, 37

Pizzolato, N., Maggio, A., & Sciortino, S. 2000, A&A, 361, 614

Pollock, A. M. T., Corcoran, M. F., Stevens, I. R., & Williams, P. M. 2005, ApJ, 629, 482

Pollock, A. M. T., Haberl, F., & Corcoran, M. F. 1995, in IAU Symposium, Vol. 163, Wolf-Rayet Stars: Binaries; Colliding Winds; Evolution, ed. K. A. van der Hucht & P. M. Williams, 512

Portegies Zwart, S. F., Pooley, D., & Lewin, W. H. G. 2002, ApJ, 574, 762

Preibisch, T. & Feigelson, E. D. 2005, ApJS, 160, 390

Preibisch, T., Kim, Y.-C., Favata, F., Feigelson, E. D., Flaccomio, E., Getman, K., Micela, G., Sciortino, S., Stassun, K., Stelzer, B., & Zinnecker, H. 2005, ApJS, 160, 401

Preibisch, T. & Zinnecker, H. 2001, in Astronomical Society of the Paciﬁc Conference Series, Vol. 243, From Darkness to Light: Origin and Evolution of Young Stellar Clusters, ed. T. Montmerle & P. Andr´e, 791

Prinja, R. K., Barlow, M. J., & Howarth, I. D. 1990, ApJ, 361, 607

Prisinzano, L., Damiani, F., Micela, G., & Sciortino, S. 2005, A&A, 430, 941

Rathborne, J. M., Jackson, J. M., & Simon, R. 2006, ApJ, 641, 389

Rathborne, J. M., Simon, R., & Jackson, J. M. 2007, ApJ, 662, 1082

Reipurth, B. & Bally, J. 2001, ARAA, 39, 403

Rho, J., Ram´ırez, S. V., Corcoran, M. F., Hamaguchi, K., & Leﬂoch, B. 2004, ApJ, 607, 904

Rieke, G. H. & Lebofsky, M. J. 1985, ApJ, 288, 618

Robin, A. C., Reyl´e, C., Derri`ere, S., & Picaud, S. 2003, A&A, 409, 523

Román-Zúñiga, C., Elston, R., Ferreira, B., & Lada, E. 2007a, astro-ph/0709.3004

Román-Zúñiga, C. G., Lada, E. A., & Elston, R. 2007b, Star Formation Handbook, in press 241

Román-Zúñiga, C. G., Lada, E. A., & Williams, J. P. 2005, in Revista Mexicana de Astronomia y Astrofisica Conference Series, Vol. 24, Revista Mexicana de Astrono- mia y Astrofisica Conference Series, ed. A. M. Hidalgo-Gámez, J. J. González, J. M. Rodr´ıguez Espinosa, & S. Torres-Peimbert, 66–69

Sakellis, S., Taylor, M. I., Taylor, K. N. R., Vaile, R. A., & Han, T. D. 1984, PASP, 96, 543

Sana, H., Rauw, G., Naz´e, Y., Gosset, E., & Vreux, J.-M. 2006, MNRAS, 372, 661

Sandstrom, K. M., Peek, J. E. G., Bower, G. C., Bolatto, A. D., & Plambeck, R. L. 2007, ApJ, 667, 1161

Scalo, J. 1998, in Astronomical Society of the Paciﬁc Conference Series, Vol. 142, The Stellar Initial Mass Function (38th Herstmonceux Conference), ed. G. Gilmore & D. Howell, 201

Schilbach, E., Kharchenko, N. V., Piskunov, A. E., R¨oser, S., & Scholz, R.-D. 2006, A&A, 456, 523

Schmitt, J. H. M. M. 1997, A&A, 318, 215

Schmitt, J. H. M. M. & Favata, F. 1999, Nature, 401, 44

Schmitt, J. H. M. M., Fleming, T. A., & Giampapa, M. S. 1995, ApJ, 450, 392

Schmitt, J. H. M. M., Golub, L., Harnden, Jr., F. R., Maxson, C. W., Rosner, R., & Vaiana, G. S. 1985, ApJ, 290, 307

Schneps, M. H., Ho, P. T. P., & Barrett, A. H. 1980, ApJ, 240, 84

Schulz, N. S., Testa, P., Huenemoerder, D. P., Ishibashi, K., & Canizares, C. R. 2006, ApJ, 653, 636

Seward, F. D., Forman, W. R., Giacconi, R., Griﬃths, R. E., Harnden, Jr., F. R., Jones, C., & Pye, J. P. 1979, ApJL, 234, L55

Shu, F. H., Li, Z.-Y., & Allen, A. 2004, ApJ, 601, 930

Sicilia-Aguilar, A., Hartmann, L., Calvet, N., Megeath, S. T., Muzerolle, J., Allen, L., D’Alessio, P., Mer´ın, B., Stauﬀer, J., Young, E., & Lada, C. 2006, ApJ, 638, 897

Sicilia-Aguilar, A., Hartmann, L. W., Brice˜no, C., Muzerolle, J., & Calvet, N. 2004, AJ, 128, 805

Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593

Skinner, S., G¨udel, M., Schmutz, W., & Zhekov, S. 2006, Ap&SS, 304, 97

Smith, M. D. 2004, The origin of stars (The origin of stars / Michael D. Smith. London (UK): Imperial College Press, ISBN 1-86094-501-5, 2004, 248 pp.) 242

Smith, R. K., Brickhouse, N. S., Liedahl, D. A., & Raymond, J. C. 2001, ApJL, 556, L91

Stahler, S. W., Palla, F., & Ho, P. T. P. 2000, Protostars and Planets IV, 327

Stassun, K. G., van den Berg, M., Feigelson, E., & Flaccomio, E. 2006, ApJ, 649, 914

Stelzer, B., Flaccomio, E., Montmerle, T., Micela, G., Sciortino, S., Favata, F., Preibisch, T., & Feigelson, E. D. 2005, ApJS, 160, 557

Stelzer, B., Hu´elamo, N., Hubrig, S., Zinnecker, H., & Micela, G. 2003, A&A, 407, 1067

Stelzer, B., Hu´elamo, N., Micela, G., & Hubrig, S. 2006a, A&A, 452, 1001

Stelzer, B., Micela, G., Hamaguchi, K., & Schmitt, J. H. M. M. 2006b, A&A, 457, 223

Stetson, P. B. 1987, PASP, 99, 191

Stevens, I. R. & Hartwell, J. M. 2003, MNRAS, 339, 280

Strom, K. M., Kepner, J., & Strom, S. E. 1995, ApJ, 438, 813

Strom, K. M., Strom, S. E., & Merrill, K. M. 1993, ApJ, 412, 233

Sung, H. & Bessell, M. S. 2004, AJ, 127, 1014

Tachibana, S., Huss, G. R., Kita, N. T., Shimoda, G., & Morishita, Y. 2006, ApJL, 639, L87

Tan, J. C., Krumholz, M. R., & McKee, C. F. 2006, ApJL, 641, L121

Telleschi, A., G¨udel, M., Briggs, K. R., Audard, M., & Palla, F. 2007, A&A, 468, 425

The, P. S., de Winter, D., & Perez, M. R. 1994, A&AS, 104, 315

Throop, H. B. & Bally, J. 2005, ApJL, 623, L149

Townsley, L., Feigelson, E., Montmerle, T., Broos, P., Chu, Y.-H., Garmire, G., & Getman, K. 2005, in X-Ray and Radio Connections (eds. L.O. Sjouwerman and K.K Dyer) Published electronically by NRAO, http://www.aoc.nrao.edu/events/xraydio Held 3-6 February 2004 in Santa Fe, New Mexico, USA, (E3.04) 10 pages

Townsley, L. K. 2006, astro-ph/0608173

Townsley, L. K., Broos, P. S., Chartas, G., Moskalenko, E., Nousek, J. A., & Pavlov, G. G. 2002, Nuclear Instruments and Methods in Physics Research A, 486, 716

Townsley, L. K., Broos, P. S., Feigelson, E. D., Brandl, B. R., Chu, Y.-H., Garmire, G. P., & Pavlov, G. G. 2006a, AJ, 131, 2140

Townsley, L. K., Broos, P. S., Feigelson, E. D., Garmire, G. P., & Getman, K. V. 2006b, AJ, 131, 2164 243

Townsley, L. K., Broos, P. S., Garmire, G. P., & Nousek, J. A. 2000, ApJL, 534, L139

Townsley, L. K., Feigelson, E. D., Montmerle, T., Broos, P. S., Chu, Y.-H., & Garmire, G. P. 2003, ApJ, 593, 874

Tsujimoto, M., Feigelson, E. D., Townsley, L. K., Broos, P. S., Getman, K. V., Wang, J., Garmire, G. P., Baba, D., Nagayama, T., Tamura, M., & Churchwell, E. B. 2007, ApJ, 665, 719 ud-Doula, A. & Owocki, S. P. 2002, ApJ, 576, 413

Urquhart, J. S., Thompson, M. A., Morgan, L. K., Pestalozzi, M. R., White, G. J., & Muna, D. N. 2007, A&A, 467, 1125

Vacca, W. D., Garmany, C. D., & Shull, J. M. 1996, ApJ, 460, 914 van der Hucht, K. A. 2001, New Astronomy Review, 45, 135

Verschueren, W. 1991, PhD thesis, , Vrije Universiteit Brussel (VUB), Brussels, Belgium, (1991)

Vuong, M. H., Montmerle, T., Grosso, N., Feigelson, E. D., Verstraete, L., & Ozawa, H. 2003, A&A, 408, 581

Walborn, N. R. 1994, in Astronomical Society of the Paciﬁc Conference Series, Vol. 60, The MK Process at 50 Years: A Powerful Tool for Astrophysical Insight, ed. C. J. Corbally, R. O. Gray, & R. F. Garrison, 84

Walborn, N. R., Barb´a, R. H., Brandner, W., Rubio, M., Grebel, E. K., & Probst, R. G. 1999, AJ, 117, 225

Walborn, N. R., Howarth, I. D., Lennon, D. J., Massey, P., Oey, M. S., Moﬀat, A. F. J., Skalkowski, G., Morrell, N. I., Drissen, L., & Parker, J. W. 2002, AJ, 123, 2754

Waldron, W. L., Cassinelli, J. P., Miller, N. A., MacFarlane, J. J., & Reiter, J. C. 2004, ApJ, 616, 542

Walter, F. M., Alcala, J. M., Neuhauser, R., Sterzik, M., & Wolk, S. J. 2000, Protostars and Planets IV, 273

Walter, F. M., Brown, A., Mathieu, R. D., Myers, P. C., & Vrba, F. J. 1988, AJ, 96, 297

Wang, J., Townsley, L. K., Feigelson, E. D., Getman, K. V., Broos, P. S., Garmire, G. P., & Tsujimoto, M. 2007a, ApJS, 168, 100

Wang, J., Townsley, L. K., Feigelson, E. D., Getman, K. V., Broos, P. S., Roman-Zuniga, C. G., & Lada, E. A. 2007b, ApJ in press

—. 2007c, ApJ in preparation

Wang, Q. D., Dong, H., & Lang, C. 2006, MNRAS, 371, 38 244

Weaver, R., McCray, R., Castor, J., Shapiro, P., & Moore, R. 1977, ApJ, 218, 377

Weidner, C., Kroupa, P., N¨urnberger, D. E. A., & Sterzik, M. F. 2007, MNRAS, 376, 1879

Weisskopf, M. C., Brinkman, B., Canizares, C., Garmire, G., Murray, S., & Van Spey- broeck, L. P. 2002, PASP, 114, 1

Weisskopf, M. C., Tananbaum, H. D., Van Speybroeck, L. P., & O’Dell, S. L. 2000, in Presented at the Society of Photo-Optical Instrumentation Engineers (SPIE) Con- ference, Vol. 4012, Proc. SPIE Vol. 4012, p. 2-16, X-Ray Optics, Instruments, and Missions III, Joachim E. Truemper; Bernd Aschenbach; Eds., ed. J. E. Truemper & B. Aschenbach, 2–16

Whitworth, A., Bate, M. R., Nordlund, A.,˚ Reipurth, B., & Zinnecker, H. 2007, Proto- stars and Planets V, 459

Whitworth, A. P., Bhattal, A. S., Chapman, S. J., Disney, M. J., & Turner, J. A. 1994, MNRAS, 268, 291

Williams, J. P. & Blitz, L. 1998, ApJ, 494, 657

Williams, J. P., Blitz, L., & McKee, C. F. 2000, Protostars and Planets IV, 97

Williams, J. P., Blitz, L., & Stark, A. A. 1995, ApJ, 451, 252

Williams, J. P. & Gaidos, E. 2007, ApJL, 663, L33

Wilson, T. L., Mezger, P. G., Gardner, F. F., & Milne, D. K. 1970, A&A, 6, 364

Winston, E., Megeath, S. T., Wolk, S. J., Muzerolle, J., Gutermuth, R., Hora, J. L., Allen, L. E., Spitzbart, B., Myers, P., & Fazio, G. G. 2007, astro-ph/0707.2537

Wolﬁre, M. G. & Cassinelli, J. P. 1987, ApJ, 319, 850

Wolk, S. J., Bourke, T. L., Smith, R. K., Spitzbart, B., & Alves, J. 2002, ApJL, 580, L161

Wolk, S. J., Harnden, Jr., F. R., Flaccomio, E., Micela, G., Favata, F., Shang, H., & Feigelson, E. D. 2005, ApJS, 160, 423

Wolk, S. J., Spitzbart, B. D., Bourke, T. L., & Alves, J. 2006, AJ, 132, 1100

Wolk, S. J. & Walter, F. M. 1996, AJ, 111, 2066

Wood, D. O. S. & Churchwell, E. 1989, ApJS, 69, 831

Yanagida, T., Ezoe, Y., Kawaharada, M., Kokubun, M., & Makishima, K. 2007, in Astronomical Society of the Paciﬁc Conference Series, Vol. 361, Active OB-Stars: Laboratories for Stellare and Circumstellar Physics, ed. A. T. Okazaki, S. P. Owocki, & S. Steﬂ, 530 245

Ybarra, J. E. & Phelps, R. L. 2004, AJ, 127, 3444

Yorke, H. W. 2004, in IAU Symposium, Vol. 221, Star Formation at High Angular Resolution, ed. M. Burton, R. Jayawardhana, & T. Bourke, 141

Yorke, H. W. & Sonnhalter, C. 2002, ApJ, 569, 846

Young, K. E., Harvey, P. M., Brooke, T. Y., Chapman, N., Kauﬀmann, J., Bertoldi, F., Lai, S.-P., Alcal´a, J., Bourke, T. L., Spiesman, W., Allen, L. E., Blake, G. A., Evans, II, N. J., Koerner, D. W., Mundy, L. G., Myers, P. C., Padgett, D. L., Salinas, A., Sargent, A. I., Stapelfeldt, K. R., Teuben, P., van Dishoeck, E. F., & Wahhaj, Z. 2005, ApJ, 628, 283

Zacharias, N., Urban, S. E., Zacharias, M. I., Wycoﬀ, G. L., Hall, D. M., Monet, D. G., & Raﬀerty, T. J. 2004, AJ, 127, 3043

Zavagno, A., Pomar`es, M., Deharveng, L., Hosokawa, T., Russeil, D., & Caplan, J. 2007, A&A, 472, 835

Zinnecker, H., McCaughrean, M. J., & Wilking, B. A. 1993, in Protostars and Planets III, ed. E. H. Levy & J. I. Lunine, 429–495

Zinnecker, H. & Preibisch, T. 1994, A&A, 292, 152

Zinnecker, H. & Yorke, H. W. 2007, ARAA, 45, 481 Vita Junfeng Wang Education The Pennsylvania State University State College, Pennsylvania 2001–Present Ph.D. in Astronomy & Astrophysics, expected in Dec. 2007 Area of Specialization: Chandra Observations of Massive Star Forming Regions Nanjing University Nanjing, Jiangsu, China 1997–2001 B.S. in Astronomy Awards and Honors Zaccheus Daniel Foundation for Astronomical Science Grant 2002–2006 Sigma Xi Grants-In-Aid Research Fellowship 2002 Braddock/Roberts Fellowship, Eberly College of Science 2001–2002

Observing Time and Travel Grants XMM-Newton Cycle 6 GO program (503740101) 60 ks (PI) Spitzer Cycle 3 GO small program (30361) 5 hours (PI) Spitzer Cycle 3 GO medium program (30726) 28 hours (joint Chandra time 50 ks; Co-I) Chandra Cycle 8 GTO program (200459) 20 ks Hobby-Eberly Telescope HRS 17 hours, LRS 5 hours (PI) International Travel Grant, American Astronomical Society 2007 Cool Stars 14 Travel Award, Spitzer Science Center 2006 XXVIth IAU General Assembly Travel Support, International Astronomical Union 2006 Protostars and Planets V Student Scholarship, University of Hawaii 2005

Research Experience Doctoral Research The Pennsylvania State University 2005–Present Thesis Advisors: Prof. Eric Feigelson and Dr. Leisa Townsley Analyzed a sample of Chandra/ACIS observations of the Galactic massive star forming regions to improve the census of young stellar populations inside. Graduate Research The Pennsylvania State University 2002–2005 Research Advisor: Prof. Jian Ge Silicon immersion grating technology for high resolution infrared spectroscopy. Dust extinction features towards high redshift quasars Undergraduate Research Nanjing University 2000–2001 Research Advisors: Prof. Zongyun Li Optical CCD photometry of Catalysmic Variable stars Teaching Experience Teaching Assistant The Pennsylvania State University 2001–2002 Taught independent astronomy lab sections, assisted activities in advanced-level undergraduate classes: Lab instructor for Astro 11; Grader for Astro 440; Telescope operator for Astro 1 and Astro 11