Chemical Evolution of the Large Magellanic Cloud)

EVOLUCIÓN QUÍMICA DE LA NUBE GRANDE DE MAGALLANES.

(CHEMICAL EVOLUTION OF THE LARGE MAGELLANIC CLOUD)

Profesor Guía: Dr. Douglas Geisler

Tesis para optar al grado académico de Doctor en Ciencias Físicas

Autor

RENEÉ CECILIA MATELUNA PÉREZ

CONCEPCIÓN - CHILE

NOVIEMBRE 2012 Director de Tesis : Dr. Douglas Geisler

Departamento de Astronomia, Universidad de Concepción, Chile.

Comisión Evaluadora : Dr. Giovanni Carraro.

European Southern Observatory, Santiago, Chile. Dipartimento di Astronomia, Universitá di Padova, Padova, Italia.

Dr. Sandro Villanova.

Departamento de Astronomia, Universidad de Concepción, Chile.

Dr. Tom Richtler. Departamento de Astronomia, Universidad de Concepción, Chile. Dedicado a Mi Padre

Agradecimientos

He llegado al final de un ciclo, y son muchas las personas que me han acompañado de alguna u otra forma en este proceso. Por esta razón, es que decidí hacer estos agradecimientos en un orden más o menos cronológico. Comenzaré por mis padres: Cecilia y René, ya que gracias a ellos estoy aquí. Mamá has sido un gran apoyo en este camino, te agradezco cada gesto de amor y cada sabio consejo que me has dado. Papá, aunque no estas físicamente presente para presenciar este momento, agradezco la oportunidad que me diste para ser fuerte y seguir adelante con mis sueños a pesar de las dificultades y se que estarías muy orgulloso de mi. Muchas gracias papá por el legado que me dejaste, mis hermanos: Alejandra, Gabriel, Mariela, José Luis y Alfredo, con ellos aprendo cada día de que en la diversidad esta la belleza y la armonía, muchas gracias, son un gran apoyo, los amo. A mis tios y primos: tia Quelita, tio Rene, Dany, Pauta y a mi comadre(Cecilia), gracias por entregarme su amor, sus consejos y esos momentos de celebración y risas. A los amig@s: Candy-Candy, Chany, Dany, Anis, Jessika, Nico, ustedes que creyeron en que este sueño se haría realidad mucho antes de que comenzara, me dieron su apoyo, su confianza y muchas risas compartidas hasta hoy, mil gracias. A la comunidad ’Physilandia’: Leo, Carlitos, Arturo G., Omar, Yaz, Marisol y Faio, les agradezco los grandes momentos vividos en la que fuera por un tiempo nuestra casa, el "Phys". Rodrigo F., gracias por la ayuda técnica y el apoyo durante el proceso final. A Roger, gracias por tus sabias palabras. A Nelson, quien fue un gran compañero durante parte de este camino, gracias por ser un apoyo y por la alegría que me entregaste en momentos importantes de mi vida. A Guille D. y Coté mis amiguitos ’PIA’...gracias por compartir este sueño. A mis nuevos ’compañeritos’, el equipo de divulgadores AstroUdec: Pamela, Paula, Gustavo, Matias y Roy, han sido un apoyo importante durante mi último año de este proceso, muchas gracias por aparecer en el momento preciso. Especialmente a Pamela, Gustavo y Paula gracias por la alegría y los momentos ñoños!!! Agradezco a mis referentes femeninos dentro de la astronomía: Stella, Karen, Amelia, Maja, Sonia y Barbara, las admiro mucho y son una fuente de motivación constante, gracias por el apoyo y los consejos. A mis compañeritos de la ESO: Mauricio, Mono, Florian, Joachim, Matias, Karla, Lucy, Daniela (compañera de oficina y amiga), muchas gracias por los almuerzos de sushi en ’patota’ y esos cafecitos de tarde conversados. A los postdocs que apoyaron mi trabajo de Tesis: Sandro y Karen, muchas gracias por el tiempo y la dedicación en resolver mis dudas. Christian, un gran apoyo en el primer ’run’ de observación en Las Campanas, muchas gracias. A mis tutores: Doug y Giovanni, muchas gracias por su paciencia, apoyo y visión. A Hugo Schwarz, el primero en depositar su confianza en mi ii y quien me guió al comienzo de este camino: aunque ya no estás para ver el resultado, muchas gracias por las enseñanzas, no olvidaré tus consejos. A quienes son parte de la Facultad y me apoyaron durante este proceso: Calu, muchas gracias por los necesarios ’breaks’ y los consejos. Jeanette, gracias por tu buena onda y por tus ricos ’coffees’. Paulina, gracias por el apoyo y el dinamismo y Marcelita, muchas gracias por tu paciencia y tu buena voluntad. Muchas gracias al comité evaluador por sus oportunos comentarios sobre esta Tesis. Agradezco a CONICYT, ESO y al proyecto BASAL por el financiamiento en la realización de esta tesis de doctorado. A todos aquellos que me han acompañado en algún momento y que no fueron nombrados, gracias por existir y haber sido parte de mi vida y de esta experiencia. Finalmente agradezco a la organización Condor Blanco: Suryavan Solar, Sol, LAma Norbu, Lama Dorje, Lexim de Gerand, Shirayam, Mankardo, Kin-Yasai, Lantui, Sekuyali, Manuvari, Samirati, Sayaru, Yamkaishi y a toda la tribu CB, gracias a todos por su apoyo en el mejor de los caminos que he conocido: el del autoconocimiento y la búsqueda de la realización.

Tesis de Doctorado Resumen

Una de las grandes preguntas en astronomía es cómo se formó nuestra Galaxia y otras galaxias, siendo ésta la motivación principal para esta tesis. Los dos escenarios más aceptados para formación de Galaxias, en particular el halo, para el propósito de este trabajo, son ’El Modelo de Colapso Monolítico’ y ’El Modelo Jerárquico de Acreción’. El último, es una versión temprana, independiente de lo que ha sido generalizado y ampliamente expandido para formar el modelo de acreción jerárquico ΛCDM, el cual predice una estructura de formación jerárquica en todas las escalas físicas, convirtiendo a las galaxias esferoidales enanas (dSph) y enanas irregulares (dIrr), en muy buenas candidatas para ser ’building block’ de nuestra Galaxia. Tomando en cuenta este último escenario, además de la intrigrante población de cúmulos estelares y lo poco que se conoce sobre evolución química de nuestra galaxia vecina, la Nube Mayor de Magallanes (LMC), se decidió llevar a cabo un estudio de esta galaxia basado en tres objetivos claros. Primero, limitar químicamente el modelo jerárquico para formación de Galaxia determinando abundancias en SCs, en la LMC. Segundo, estudiar la evolución química de la LMC al sumar la mayor cantidad de puntos como sea posible a la relación edad-metalicidad (AMR), incluyendo tanto los cúmulos como las estrellas del campo, y estudiando tantos elementos como sea posible para investigar una gran variedad de vías nucleosintéticas, y tercero, para comprobar si la técnica de CaT presenta posibles fallas en la determinación de la metalicidad en la LMC. Para este propósito, se recolectó un gran conjunto de datos. Se realizó espectroscopía de alta resolución (FLAMES@VLT) y fotometría de Washington, usando un instrumento de campo amplio (MOSAIC@4mCTIO), para cúmulos estelares y estrellas del campo de la LMC. El rango espectral me permitió medir una variedad de elementos, incluyendo Fe, Mg, Ca, Ti, Si, Na, O, Ni, Cr, Sc, Mn, Co, Zn, Ba, La, Eu e Y. Se analizaron dos cúmulos estelares (Hodge11 y SL869) y 21 campos de la LMC. Del análisis de la espectroscopía de alta resolución (HRS), se obtuvo una metalicidad de

[Fe/H]= −2.00 ± 0.04( σobs = 0.11 ± 0.03), para el cúmulo viejo Hodge 11(H11), en acuerdo con estudios anteriores. Y por primera vez para SL869, una metalicidad de [FeI/H]= −0.47

(σint = 0.04). Además, se estimó una edad de 1.45 Gyr, σ = 0.2Gyr, para SL869 usando un ajuste de isocronas. Uno de los resultados más importantes en este estudio es aquel que proviene del valor medio de [α/Fe] vs [Fe/H] (Fig. 3.7). Encontramos que H11 se encuentra en el rango de las dSph y bajo el valor observado en la Galaxia. Este resultado conﬁrma estudios previos y abre la posibilidad a que galaxias como la LMC, que se asumieron como ’building blocks’ de nuestra galaxia, puedan de hecho no satisfacer los requerimientos químicos, incluso en la metalicidad baja representada iv por H11. En los elementos iron-peak, también se ven abundancias similares a los resultados de dSph, tales como baja Cr, Mn y Ni. Los resultados de este trabajo, respecto de la evolución química de la LMC, están bien descritos por un modelo de ’bursting’ propuesto por Pagel & Tautvaisiene (1998), con la excepción del campo más antiguo, donde los errores para la edad son considerablemente mayores. Estos resultados muestran un incremento muy suave en la metalicidad sobre el período de edad de ∼4-11 Gyr, estando en excelente acuerdo con el modelo de ’bursting’. SL869 también está en muy buen acuerdo con este modelo, e incluso cúmulos más jóvenes muestran inequívocamente un incremento en el enriquecimiento químico predicho por este modelo en los últimos Gyrs. A pesar de que nuestros datos ajustan bien con el modelo, es necesario analizar muchos más datos, especialmente a bajas metalicidades, con el ﬁn de tener una idea más clara sobre la formación de la LMC y por tanto de nuestra propia galaxia.

Tesis de Doctorado Abstract

Understanding how our Galaxy and other galaxies formed is one of the big questions in As- tronomy and is the principal motivation for this thesis. The two most widely accepted scenarios for Galaxy formation, in particular the halo, for the purpose of this work, are ’The Monolithic Col- lapse Model’ and ’The Merger/Hierarchical Accretion Model’. The last is an early, independent version of what has now been generalized and greatly expanded to form the ΛCDM hierarchical accretion model, which predicts hierarchical structure formation on all physical scales, turning dwarf spheroidal (dSph) and dwarf irregular (dIrr) galaxies very good candidates to be building blocks of our Galaxy. Taking this last scenario into account, plus the intriguing star cluster population and what little is known of the chemical evolution of our galaxy neighbour, the Large Magellanic Cloud(LMC), it was decided to proceed with a study of this galaxy based on three clear aims. First, to chemically constrain the hierarchical model for Galaxy formation by determining abundances in SCs in the LMC. Second, to study the chemical evolution of the LMC by adding as many points as possible to the age-metallicity relation(AMR), including both cluster and field stars, and studying as many elements as possible to investigate a wide variety of nucleosynthetic pathways, and third, to check if the CaT technique presents possible flaws in metallicity determination in the LMC. For this purpose, a big set of data was collected. High resolution spectroscopy (FLAMES@VLT) and Washington photometry, using a wide field imager(MOSAIC@4mCTIO), were performed for star clusters and field stars in the LMC. The spectral range allowed me to measure a variety of elements, including Fe, Mg, Ca, Ti, Si, Na, O, Ni, Cr, Sc, Mn, Co, Zn, Ba, La, Eu and Y. Two star clusters (Hodge11 and SL869) and 21 fields of the LMC were analyzed. From the high resolution spectroscopy(HRS) analysis, it was obtained a metallicity of [Fe/H]=

−2.00 ± 0.04( σobs = 0.11 ± 0.03) for the old star cluster Hodge 11(H11), in agreement with previous studies. And for the first time for SL869, a metallicity of [FeI/H]= −0.47 (σint = 0.04). In addition, an age of 1.45 Gyr (σ = 0.2Gyr) was estimated for SL869 using isochrone fitting. One of the most important results in this study is that from the mean [α/Fe] vs [Fe/H] plot (Fig. 3.7). We find that H11 lies in the range of the dSph trend and below the Galactic one. This result confirms previous studies and opens the possibility that galaxies like the LMC, assumed to be building blocks of our galaxy, may not in fact satisfy the chemical requirements, even at the low metallicity represented by H11. In the iron-peak elements, abundance similarities to the dSph results are also seen, such as low Cr, Mn and Ni. For the field stars, 21 fields divided in 16 subfields were studied, determining age and metallicity from δT1 and SGB method, respectively. They are well described, from the chemical vi evolution, by a bursting model proposed by Pagel & Tautvaisiene (1998), with the exception of the oldest field, where the errors are considerably larger in age. These results show a very smooth increase in metallicity over the age period ∼4-11 Gyr, during the cluster age gap, in excellent agreement with the bursting model. SL869 is also in good agreement with this model, and even younger clusters unequivocally show the increase in chemical enrichment predicted by the bursting model over the last few Gyrs. Much more data, especially at low metallicities, are necessary to be analyzed, in order to have a more clear idea of the formation of the LMC and therefore our own galaxy.

Tesis de Doctorado Contents

Agradecimientos i

Resumen iii

Abstract v

List of Figures x

List of Tables xiii

1 Introduction 1 1.1 Motivation ...... 1 1.2 Studying the Large Magellanic Cloud ...... 2 1.2.1 Star Clusters in the LMC and the Age Gap ...... 2 1.2.2 Using High Resolution Spectroscopy ...... 2 1.2.3 Using Washington Photometry ...... 5 1.3 Chemical Evolution ...... 6 1.3.1 Basic Concepts of Chemical Evolution of Galaxies ...... 6 1.3.2 Enrichment models ...... 7 1.4 The Aim of this Thesis ...... 8

2 Method 9 2.1 Observations: Data Acquisition ...... 9 2.1.1 ESO-VLT Paranal Observatory ...... 9 2.1.2 Las Campanas Observatory ...... 10 2.1.3 Cerro Tololo Inter American Observatory (CTIO) ...... 11 2.2 Data Reduction ...... 11 2.2.1 Spectra ...... 11 2.2.2 Images ...... 14 2.3 Imaging Techniques ...... 15 2.3.1 Photometry ...... 15 2.3.2 Astrometry: From X and Y to RA and DEC ...... 16 2.4 Spectral Analysis ...... 16 2.4.1 Radial Velocities ...... 16 viii CONTENTS

2.4.2 Abundance Analysis ...... 21

3 Hodge 11 27 3.1 Introduction ...... 27 3.2 Results from Photometry ...... 28 3.3 Radial Velocities for Hodge 11 ...... 29 3.4 Abundances Results ...... 29 3.4.1 Fe-peak elements ...... 29 3.4.2 α-elements ...... 35 3.4.3 Neutron Capture Elements ...... 37 3.4.4 Na and O ...... 40

4 SL869 and the Surrounding Field of H11 45 4.1 Introduction ...... 45 4.2 Star Cluster SL869 ...... 46 4.2.1 Photometry ...... 46 4.2.2 Abundance Analysis ...... 47 4.2.3 Age determination ...... 50 4.3 Field Stars ...... 50 4.3.1 Abundance Analysis ...... 50

5 Washington Photometry of the LMC and V, I photometry method tests. 57 5.1 Introduction ...... 57 5.2 Photometry of CTIO MOSAIC data: Hodge 11 ...... 57 5.3 Washington Photometry of the LMC Field ...... 58 5.3.1 Main Results ...... 60 5.4 Testing the initial "SkZ" pipeline and obtaining PSF Photometry ...... 63

6 Preliminary Results on the Chemical Evolution of the LMC 69 6.1 Introduction ...... 69 6.2 Previous work on the AMR ...... 69 6.3 LMC Chemical Evolution Models ...... 70 6.4 Our Results on the Age Metallicity Relation ...... 73

7 Future Work 77 7.1 Data processing, abundance determination and further analysis ...... 77 7.1.1 Chemical Evolution Models: LMC ...... 79

8 Conclusions 81

A Error Analysis 83 A.0.2 Error Analysis for H11 abundance determinations ...... 83

Tesis de Doctorado CONTENTS ix

B MOOG: The LTE Stellar Line Analysis Program 85 B.1 What is MOOG? ...... 85 B.2 How is it work? ...... 85

C Nucleosynthesis 91 C.1 Nuclear Physics Concepts ...... 91 C.2 Nuclear Processes ...... 91

❄❇❅❆❈❋ R. Mateluna P. - 2012

List of Figures

1.1 LMC Star Clusters...... 3 1.2 LMC Star Clusters...... 4 1.3 Washington standard giant branches...... 5

2.1 Fibers Map...... 10 2.2 MOSAIC image example...... 13 2.3 Washington Photometric System...... 13 2.4 MOSAIC observed fields preparation...... 14 2.5 MOSAIC CCD...... 15 2.6 Detailed view of our best UVES spectrum, TARG.6, in the range of 6220 − 6270. 21 2.7 Map example...... 22 2.8 Example of spectral synthesis method. Five different spectra, with Fe abundance varying from 5.12 to 5.52, are shown fitted to the data in the measurement of a line of FeI (6335.3) in TARG.6...... 26

3.1 Hodge 11 image ...... 28 3.2 Color-magnitude diagram for H11...... 30 3.3 Swope Map...... 31 3.4 Velocity histogram for GIRAFFE and UVES data...... 32 3.5 Abundances for iron peak elements...... 36 3.6 Abundances for α-elements compared to the literature...... 38 3.7 Mean [α/Fe ...... 39 3.8 Neutron-capture elements abundances...... 41 3.9 Ba to Y ratio vs. metallicity ...... 42 3.10 Ba to Eu ratio vs. metallicity...... 43 3.11 Na and O abundances ...... 44

4.1 SL869 photometry calibration ...... 47 4.2 Color-magnitude diagram for SL869 ...... 48 4.3 Age estimation for SL869 ...... 51 4.4 Alpha element abundances for ﬁeld stars...... 54 4.5 Mean [α/Fe ...... 55 4.6 Iron-peak abundances for ﬁeld stars...... 56 xii LIST OF FIGURES

5.1 Washington standard giant branches...... 58 5.2 Washington Color-magnitude diagrams for Hodge11...... 65 5.3 Zoom in washington CMD...... 66 5.4 Fit of SGB to H11 photometry...... 67 5.5 Color-magnitude diagram for NGC1841...... 68

6.1 Composite AMR for the 21 studied LMC ﬁelds from Piatti et al.(2012)...... 71 6.2 AMR for the star clusters in the LMC from Hill et al.(2000)...... 72 6.3 SFR history for the LMC according to Pagel Tautvaisiene (1998)...... 72 6.4 Composite AMR for the 21 studied LMC ﬁelds and the cluster results from this work ...... 74

7.1 Type II SNe models...... 78

B.1 Example of parameter file: synth...... 86 B.2 Example of parameter file: abfind...... 87 B.3 Example of MOOG graphics: abfind ...... 88 B.4 Example of MOOG graphics: synth...... 89

C.1 BigBang Nucleosynthesis...... 93 C.2 Abundances in the Solar Neighborhood...... 94 C.3 Chart of nuclides...... 95

Tesis de Doctorado List of Tables

2.1 Observing Time Proposals ...... 12 2.2 LMC target star clusters ...... 16 2.3 Telescopes and Instruments set-up ...... 17 2.4 Spectroscopic data set ...... 18 2.5 Imaging data set from LCO ...... 18 2.6 LMC Cluster data set from CTIO ...... 19 2.7 LMC star ﬁelds from CTIO...... 20 2.8 Line list used...... 23 2.8 continued...... 24 2.8 continued...... 25

3.1 Important parameters for our target stars ...... 33 3.2 [X/Fe] values for Hodge 11 stars...... 34

4.1 Important parameters for our target stars ...... 49 4.2 Iron abundances for our target stars ...... 50 4.3 [X/Fe] values for ﬁeld stars...... 53

5.1 Washington non-calibrated magnitudes of H11 target stars ...... 59 5.2 Estimated ages and dispersions (in Gyr) for the representative populations in LMC ﬁelds...... 61 5.3 Estimated metallicities and dispersions (in dex) for the representative populations in LMC ﬁelds...... 62

A.1 Errors in [X/Fe] for stellar parameters...... 84

Chapter 1

Introduction

1.1 Motivation

Understanding how our Galaxy and other galaxies formed is one of the big questions in As- tronomy and is the principal motivation for this work. In Geisler et al.(2007)(32) there is a good summary of the state of art of this field from the point of view of the chemistry of the halo, globular clusters (GCs) and Local Group dwarf galaxies. The two most widely accepted scenarios for Galaxy formation, in particular the halo for the purpose of this work, are "The Monolithic Collapse Model" (Eggen et al.(1962)(25) -hereafter ELS62) and "The Merger/Hierarchical Ac- cretion Model" (Searle and Zinn (1978)(90) -hereafter SZ78), which is an early, independent version of what has now been generalized and greatly expanded to form the Λ CDM hierarchical accretion model. The first (ELS62) propose that a general monolithic gravitational collapse of matter brought together the baryons now observed as a spherical halo and continued on to form the disk. The infalling material was initially very metal-poor, but the collapse time was longer than the stellar evolution time, allowing some astration of material to be processed. This model predicts a gradient in metallicity and kinematics in the halo. The second model (SZ78) was proposed by the authors after they found no radial abundance gradient in the cluster system of the outer halo and a significant difference in HB morphology between inner and outer halo GCs. These results led them to suggest that the outer halo formed over a longer time via accretion of external systems, where they introduced the term "fragments" for these postulated Galactic "building blocks". This accretion model is in very good agreement with modern cosmological theories based on the current Λ CDM paradigm which predicts hierarchical structure formation on all physical scales (Navarro et al.(1997)(68)). Obvious candidates to be building blocks of our Galaxy, noted by SZ78 and after by Zinn(1980) (110), are dwarf spheroidal (dSph) and dwarf irregular (dIrr) galaxies, found in abundance con- centrated around the Milky Way and M31. Although such dwarfs are the most numerous galaxies in the universe, there are not enough such satellites remaining surrounding the Galaxy and M31to match the numbers predicted by Λ CDM theories, leading to the "missing satellite" problem. Nevertheless, an observational proof of accretion has been the discovery of the Sagittarius dSph (47) which is in the process of being accreted by the Milky Way. The above has motivated various groups, including the present thesis, to investigate these "fragments", especially from a 2 CHAPTER 1. INTRODUCTION chemical point of view, in order to compare them to the Milky Way halo.

1.2 Studying the Large Magellanic Cloud

The Large Magellanic Cloud (LMC) is one very good candidate for a prototypical halo building block , lying in the right mass regime predicted by the latest Λ CDM theories (11) which successfully explain the chemical evolution of the halo from massive fragments. It is also very near (about 50kpc) and thus an excellent target to test these theories. One of the most direct and powerful ways to learn about a galaxy’s chemical evolution and star formation history is through the study of its star clusters, which preserve a record of their galaxy’s chemical abundances at the time of their formation and are relatively easy to age date. Particularly important are the ancient globular clusters (GCs), which were witnesses to the construction of their parent galaxy long ago.

1.2.1 Star Clusters in the LMC and the Age Gap

Herschel(1847) mentions for the first time the existence of star clusters (SCs) in the LMC and since then several authors have catalogued these objects (104). A very good discussion of LMC SCs can be found in Westerlund’s book, The Magellanic Clouds (108), and an up to date catalogue can be found in Bica et al.(2008)(8). The LMC has a particularly interesting population of SCs. For example, young massive clusters were found there for the first time, with no counterparts in the Milky Way, changing our preconceptions of such objects in general and how they were characterized (43). From Grochol- ski et al.(2006)(39)(hereafter G06) intermediate-age star clusters (1-3 Gyr) show a surprisingly narrow range of metallicities. There are some 13 known true GCs in the LMC; i.e. clusters with ages and masses comparable to their Milky Way counterparts (100), such as Hodge 11 (H11). Given the significance of GCs as witnesses to the construction of structures in an early time, a detailed study of any of this sample is of significant astrophysical importance. Another interesting phenomena was noticed from the earliest integrated color-magnitude diagrams (CMDs) of LMC clusters (see Fig. 1.2): the population of SCs is divided into two very distinct groups, blue and red(103). This led to the eventual discovery of an age gap in the cluster age distribution and age-metallicity relation (AMR- details in a subsequent chapter) of the LMC star clusters (23). This gap contains only one star cluster (ESO121-SC03: ∼ 9 Gyr) between the old GCs (about 13 Gyr) and the large population of massive, intermediate age (1−3 Gyr) clusters. The lack of star clusters in this huge age range, which covers most of the LMC’s history, prevents us from using them to study the chemical evolution during this time. To help fill this gap, we turn to the field stars, which are found to cover the whole age range.

1.2.2 Using High Resolution Spectroscopy

The most detailed knowledge of the chemistry of a star is given by high resolution spectroscopy (HRS), which provides accurate abundances for a wide variety of elements with a range of

Tesis de Doctorado 1.2. STUDYING THE LARGE MAGELLANIC CLOUD 3

Figure 1.1 Distribution of Star Clusters in the Large Magellanic Cloud studied by G06. The dashed curve roughly outlines the bar. nucleosynthetic histories, yielding a wealth of information on the various processes involved in the cluster’s chemical evolution. The ages, kinematics, metallicities and abundance ratios derived from HRS of Galactic globular clusters (GGCs) have yielded a vast array of fascinating insights into their formation and that of the Milky Way. With the advent of 8m class telescopes equipped with multi-object HR spectrographs, such observations in the LMC are now quite tractable. However, remarkably, despite the wealth of information that HRS of LMC star clusters could provide, only a handful of LMC GCs have been observed with HRS (Hill et al. 2000(42), Johnson et al.(2006)(50)- hereafter J06, Mucciarelli et al 2009 (66), Mucciarelli et al. 2010Mu10-hereafter M10). Such studies are desperately needed to tell us about the formation and early chemical evolution of our second nearest galactic neighbor, as well as that of our own Galaxy. Detailed abundances in dwarf spheroidal (dSph) galaxies (e.g. Schetrone et al (2001)(91), Geisler et al.(2005) (31)) show that the abundances of giants in dSphs and dIrrs are quite distinct from those in the halo of the Milky Way. In particular, the dwarfs have depleted abundances of the α elements at a given [Fe/H] compared to their halo counterparts. This potentially serious problem for hierarchical formation models can be overcome, as suggested by Robertson et al. (2005)(86), by assuming that the bulk of the halo was built by the accretion of only a very small number of very massive dwarfs (∼LMC size) very early in the galaxy’s history. Such dwarfs would have had high star formation rates and fast chemical evolution, with enrichment

❄❇❅❆❈❋ R. Mateluna P. - 2012 4 CHAPTER 1. INTRODUCTION

Figure 1.2 Integrated Color-magnitude diagram of the Large Magellanic Cloud star clusters from van den Bergh (104). Note the color gap between red and blue SCs. dominated by Type II SNe, leading to high α abundances. In contrast, the dSphs we see today are low mass survivors, with low star formation rates and slower evolution and thus relatively depleted α abundances due to enrichment from both Types Ia and II SNe. An important test of their prediction is that the lowest metallicity stars in the LMC should follow similar abundance patterns to those in the halo. In particular, they should have enhanced [α /Fe] ratios. Only a tiny sample of such very metal-poor LMC stars have been studied in detail (42, 50, 81), and results are inconclusive. Note that this problem may have been partly ameliorated by the recent discovery of two halo populations with distinct [α /Fe] ratios (69). The high-α population is that normally associated with the halo, with a relatively constant value of [α/Fe] ∼0.3, while the low-α population at the same [Fe/H] has [α/Fe] lower by an amount that increases with [Fe/H], from ∼ 0.05 at [Fe/H]∼-1.6 to ∼ 0.15 at [Fe/H]∼-0.8. However, in this same metallicity range, dSph stars have typical [α/Fe] values that are even lower, by ∼0.1 - 0.3 dex (32). This in fact leaves the bulk of the α problem intact. In addition, previous large scale metallicity determinations for LMC clusters (Olszewski et al.(1991)(71)-hereafter O91, Grocholski et al.(2006)(39)-hereafter G06) have relied on using the Ca IR triplet (CaT) as a proxy for measuring Fe abundances. This approach, however, has one potential ﬂaw: it assumes that [Ca/Fe] is the same for the LMC as for the Galactic clusters used to calibrate the relationship. The very limited HRS studies available so far ((42, 94, 82? ), J06) have not yet arrived at a consistent, deﬁnitive picture in this regard. Further work is required to clarify and quantify the use of Ca as a proxy for Fe, especially for extraGalactic studies. An independent HRS study of LMC star clusters, similar to that of J06 and M10, was therefore considered of great importance and led to the development of this work. The objective

Tesis de Doctorado 1.2. STUDYING THE LARGE MAGELLANIC CLOUD 5 is to trace the chemical evolution of the LMC from its earliest beginnings and thus we necessarily required several old GCs (like H11) and as many intermediate-age SCs, like SL869, as possible. Of equal importance are the field stars because in order to test different chemical evolution models, we need to fill the cluster age-gap in the AMR of the LMC (chapter 6). For this purpose, ages needed to be determine and the best way is using photometry.

1.2.3 Using Washington Photometry

According to the "Handbook of CCD Astronomy" (46), Photometry forms one of the fun- damental branches of astronomy. In addition to this, photometry of star clusters has been a powerful tool to understand and learn about stellar populations and stellar evolution processes. Using a variety of photometric systems, for example the Washington system, allow us to better describe and unravel diﬀerent parameters of these objects like age and metallicity, together with describing stellar populations in the LMC. The Washington photometric system was developed by Canterna (1976)(13) with the purpose of obtaining accurate temperatures, metal abundances and a CN index for G and K giants. Today, this system is mainly applied to derive metallicities and ages of SCs, with very good results Geisler et al.(1997)(28), Geisler & Sarajedini(1999)(29),Geisler et al.(2003)(30),Piatti et al. (2009)(76), Piatti(2012)(79), J06. The derivation of metallicity is usually done using the standard giant branches (SGB) method developed in Geisler & Sarajedini(1999)(29), where each giant branch corresponds to an isoabun- dance curve (see Fig.5.1). Age determination comes either from isochrone ﬁtting and/or from

Figure 1.3 Washington standard giant branches(SGB) in the [MT 1 −(C −T1)0] plane from Geisler & Sarajedini(1999), (29) (paper Fig.4). the δT1 index. This latter corresponds to the magnitude diﬀerence (δ) between the giant branch

❄❇❅❆❈❋ R. Mateluna P. - 2012 6 CHAPTER 1. INTRODUCTION clump(RC) in intermediate age clusters (the horizontal branch in old clusters) and the main sequence turnoff(TO - more details in Geisler et al.(1997)(28) The Washington system is a very efficient tool for determining the two key parameters for the study of galaxy evolution - age and metallicity. That is why this photometric system is used in this work - to help in the derivation of ages and metallicities for clusters and field stars in the LMC. More details about how this system was applied in this work is described in chapter 5.

1.3 Chemical Evolution

Chemical evolution of galaxies concerns the origin, distribution and evolution of nuclear species in stars and gas within a galaxy. First it is necessary to understand what are the basic concepts of chemical evolution in general.

1.3.1 Basic Concepts of Chemical Evolution of Galaxies

From Matteucci, F.(2008)(60), the basic ingredients to build a model of galactic chemical evolution (GCE) are:

• Initial conditions: The initial conditions for a model of galactic chemical evolution consist in establishing whether : a) the chemical composition of the initial gas is primordial or pre-enriched by a pre-galactic stellar generation; b) the studied system is a closed box or an open system (infall and/or outﬂow).

• Stellar birthrate function: The birthrate function can be deﬁned as: B(M,t) = ψ(t)ϕ(t), where ψ(t) is called the star formation rate (SFR - the rate at which the gas is turned into stars, and ϕ(t) is the initial mass function (IMF - the mass distribution of the stars at birth). More details are found in Matteucci, F. (2008)(60) and references there in.

• Stellar yields: Diﬀerent stars contribute with diﬀerent elements to the interstellar medium (ISM) depending on initial mass and chemical composition, in this context, the stellar yield is the amount of newly formed and pre-existing elements ejected by stars of all masses at their death and can be calculated by knowing stellar evolution and nucleosynthesis. At this point is important to describe a commonly used term in astronomy, the metallicity, which is symbolized by [Fe/H] and it tells us about the amount of metals (elements heavier than Hydrogen and Helium), referenced to the Sun (see eq.(1.1)) in a star.

NFe NFe [F e/H]= log10 − log10 (1.1) NH star NH Sun The metal enrichment of the ISM comes mainly from :

4 12 14 – Low and intermediate mass stars (0.8 ≤M/Msun ≤8.0) produce mainly He, C, N plus some CNO isotopes and s-process (A > 90) elements.

– Massive stars (8 ≤ M/Msun ≤ 40) end their life as Type II SNe and explode by core- collapse; they produce mainly α-elements (O, Ne, Mg, Si, S, Ca), some Fe-peak

Tesis de Doctorado 1.3. CHEMICAL EVOLUTION 7

elements, s-process elements (A < 90) and r-process elements. Stars more massive than 40 Msun can end up as Type Ib/c SNe: they are also core-collapse SNe and are linked to γ-ray bursts (GRB). – Type Ia SNe (white dwarfs in binary systems). They produce mainly Fe-peak elements.

– Very massive objects (M > 40Msun). They should produce mainly oxygen although many uncertainties in the models are still present.

• Gas flows: Various parametrizations have been suggested for gas flows and the most common is an exponential law for the gas infall rate: IR ∝ e−t/τ with the timescale τ being a free parameter, whereas for the galactic outflows the wind rate is generally assumed to be proportional to the SFR: WR = −λSF R where λ is again a free parameter. Both τ and λ should be fixed by reproducing the majority of observational constraints.

When all these ingredients are set, it is necessary to write a set of equations describing the evolution of the gas and its chemical abundances which include all of them. These equations will describe the temporal variation of the gas content and its abundances by mass (details in Matteucci, F.(2008) (60)). The chemical abundance of a generic chemical species i is deﬁned as: X = Mi . According to this deﬁnition it holds: X = 1, where n represents the total i Mgas i=1,n i number of chemical species. P

1.3.2 Enrichment models The three most considered models of chemical enrichment, according to ’Galactic Astronomy Book’(9), are:

• Closed-box model: Introduced by Talbot and Arnett (1971)(101) as ’simplest possible model’. It focused on a narrow annulus of galacto-centric radius and assumes that ’in the period under study no material either enters or leaves the region’. Initially: material entirely gaseous free of heavy elements. As time goes on stars are formed from the interstellar gas and massive stars return H, He and heavy elements to the ISM. For equations and details see (9).

• Leaky-box model: part of the gas is driven out by stars that form within the box. We suppose that SNe drive gas out of the box at a rate proportional to the SFR. This model was formulated by Hartwick, 1976 to reproduce the observed metallicity distribution within the Milky Way’s halo.

• Accreting-box model: the systems accretes gas and as a consequence of this, the system has the tendency to make metal-poor stars much less than in the absence of accretion.

In the case of the LMC a closed-box model has been often used to ﬁt the AMR data from the study of star clusters and ﬁeld stars (Carrera et al. 2011(15)). Other authors, like Pagel & Tautvaišien˙e(1998)(73) and recently Bekki et al. 2012 (5) have proposed new models to described the chemical evolution of the LMC, which will be tested in chapter 6 with our observational results.

❄❇❅❆❈❋ R. Mateluna P. - 2012 8 CHAPTER 1. INTRODUCTION 1.4 The Aim of this Thesis

The main purposes of this thesis are:

• To chemically constrain the hierarchical model for Galaxy formation by determining abundances in SCs in the LMC.

• To study the chemical evolution of the LMC by adding as many points as possible to the age-metallicity relation(AMR), including both cluster and ﬁeld stars, and studying as many elements as possible to investigate a wide variety of nucleosynthetic pathways.

• To check if the CaT technique presents possible ﬂaws in metallicity determination in the LMC.

Tesis de Doctorado Chapter 2

Method: Observations, Data Reduction and Techniques

2.1 Observations: Data Acquisition

All the observations were made in three Chilean observatories: ESO-VLT Paranal Obser- vatory (VLT), located near Antofagasta; Las Campanas Observatory (LCO) and Cerro Tololo InterAmerican Observatory (CTIO), both located near La Serena. We applied for observing time through the Chilean Time Allocating Comittee (CNTAC) and through the ESO proposal process. In Table 2.1 are listed all the accepted observing proposals done during the thesis period. In order to fulﬁll the main purpose of this work, ﬁve LMC star clusters were selected from the previous study of G06, including three very old GCs (H11, NGC2257, NGC 1841) and two intermediate-age SCs(NGC1718, NGC1846). In addition, a third intermediate-age cluster was added for free, because it is located in the proximity of H11. Position, metallicity, radial velocity and age for our selected SCs from the literature are given in Table2.2.

2.1.1 ESO-VLT Paranal Observatory

This Observatory has four 8m telescopes named UT1,2,3 and 4. The observations were performed with the FLAMES(UVES+GIRAFFE) 1, a high resolution spectrograph, mounted on UT2 (Kueyen) during January of 2009 (ID Proposal 082.B-0458) in service mode. The GIRAFFE dataset was obtained using the H651.5A/HR14A (wavelength range= 6308 − 6701,R= 17, 700) set-up (see Table2.3). The observations consist of 10 exposures of 45min each for the 108 fibers. In the case of UVES, five of the brighter RGB stars were selected and observed using the 580nm setting with 1.0” fibers, covering the wavelength range 4800 − 6800 with a mean resolution of R= 47, 000. The selection of the targets was based on the study of G06, which confirmed metallicity and radial velocity membership for a number of red giants stars in Hodge 11, NGC 2257 and NGC

1http://archive.eso.org/wdb/wdb/eso/sched_rep_arc/querytel=UT2&from_date=01Oct- 2008&progid=082.B0458%28A%29&period=82&remarks= 10 CHAPTER 2. METHOD

1718. In addition numerous surrounding ﬁeld stars were observed.

Figure 2.1 Position of the fibers from the instrument in the Hodge 11 field. Black dots are all the fibers, in green the possible target stars(cluster members), in red and blue the final H11 members from GIRAFFE and UVES respectively. Magenta dots are targets in SL869(see chapter 4)

2.1.2 Las Campanas Observatory LCO has various telescopes and for the acquisition of this thesis data, two diﬀerent instruments and telescopes were used. First, spectroscopic data were acquired with IMACS+MOE 2 mounted on one of the Magel- lan Telescopes (http://www.lco.cl/telescopes-information/magellan/), 6.5m Baade, during three separate runs listed in Table2.1. For this instrument it was necessary to prepare the observations in advance by making a MASK, to be installed in the instrument in order to observe the 2 http://www.lco.cl/telescopes-information/magellan/instruments/imacs/

Tesis de Doctorado 2.2. DATA REDUCTION 11 selected targets in NGC 1841 and NGC 1846, one mask per cluster. Target selection was based on the study of G06, the same as for the FLAMES@VLT observations and for the mask creation process. The instrument set up and dataset are listed in tables 2.3 and 2.4 respectively. For imaging the observations were made using the SITe#3 CCD camera 3 mounted on the Swope 1.0m telescope(http://www.lco.cl/telescopes-information/henrietta-swope/), during Jan- uary of 2011. Images in V and I ﬁlters were obtained for the clusters: Hodge 11, NGC 2257, NGC 1718, NGC 1841 and NGC 1846. Details of the observations appear in Table2.5.

2.1.3 Cerro Tololo Inter American Observatory (CTIO)

Two telescopes of CTIO were used by the author in the data acquisition, in both cases imaging data was obtained. The first set of data was obtained using MOSAIC (http://www.ctio.noao.edu/mosaic/), a wide field imager (36’ x 36’ field with 8Kx8K CCD detector) mounted on CTIO-4m Blanco telescope. An example of a typical MOSAIC image is in Fig.2.2. We covered 21 fields of the LMC main body for a total area of ∼ 7.6 square degrees. Only a single image was taken in each filter, as we judged that the dynamic range required to suit our science goals could be met most efficiently this way. Some fields have shorter exposure times simply due to time constraints. From this observation a set of LMC star clusters and field stars were observed using Washington filters T1, T2 and C (Fig.2.3). Details about the target are given in Table 2.6. Because of the wide field of the Instrument, fields were centered so that more than one star cluster fits in the field (see Fig. 2.4), and we set the center of the field on im6 (see MOSAIC CCD map, Fig. 2.5). The log of the observations is presented in Table2.1.3, where the main astrometric, photometric and observational information is summarized. Details abut the fields observed and results obtained with these observations appear in chapter 5. The second set of data was acquired using the imager, Y4kCam, mounted on the 1mYale telescope. In this observing run of 9 nights, several clusters of the LMC, as well as SMC and Galactic open clusters were observed in V, R and I filters. From the total 9 nights, 1 was lost for bad weather conditions. The rest of the data was processed but deemed useless because of strong fringing in the I filter during this bright time run.

2.2 Data Reduction

2.2.1 Spectra

GIRAFFE data were reduced using the ESO pipeline 4. Data reduction includes bias subtraction, flat-field correction, wavelength calibration, sky subtraction, and spectral rectification. UVES data were reduced using the UVES pipeline (Ballester et al. (2000)(2)), where raw data were bias-subtracted, flat-field corrected, extracted and wavelength calibrated. Using IRAF5 tasks rvcorrect we compute heliocentric corrections, with dopcor and continuum, each spectrum

3http://www.lco.cl/telescopes-information/irenee-du-pont/instruments/website/direct-ccd-manuals/direct- ccd-users-manual/ccd-manual-for-the-40-inch-100-inch-telescopes 4 http://www.eso.org/sci/software/pipelines/ 5IRAF is distributed by the National Optical Astronomy Observatory, which is operated by the Association of Universities for Research in Astronomy, Inc., under cooperative agreement with the National Science Foundation.

❄❇❅❆❈❋ R. Mateluna P. - 2012 12 CHAPTER 2. METHOD al . bevn ieProposals Time Observing 2.1 Table NA1B0221- .aeuaMtln G14 IMACS+MOE NGC1846 I NGC1841 NGC1846, Mateluna LMC Mateluna Muñoz Geisler, Mateluna, R.Mateluna Geisler D. Mateluna R. 2010-B 2009-B 2009-B CNTAC10B_062 CNTAC09B_010 CNTAC09B_024 NA-9220- .GilrMtln,GilrLCsa clu star LMC Geisler Mateluna, Geisler D. 2008-B CNTAC-0912 DPooa eetrYa IOsre bet Instrument Objects Observer PI Semester-Year Proposal ID 8.-48EO2B20 .Vlaoasriemd 1,NGC2 H11, service-mode Villanova S. ESO82-B-2008 082.B-0458 n M trcutr Y4KCam clusters star SMC and 5,NC78FLAMES NGC1718 257, tr MOSAIC sters MACS+MOE

Tesis de Doctorado 2.2. DATA REDUCTION 13

Figure 2.2 Example of MOSAIC image, a display of a random image to show the MOSAIC wide ﬁelds and how crowded the ﬁeld is, small star clusters are seen in the image.

Figure 2.3 Washington system of ﬁlters from Canterna (13). is shifted to rest-frame velocity and continuum-normalized, respectively. Finally, orders were

❄❇❅❆❈❋ R. Mateluna P. - 2012 14 CHAPTER 2. METHOD

Figure 2.4 Here is how the fields were prepare and cluster arrange in order to better used the wide field(represented by a rectangle) and avoid time lost. merged to obtain a 1D spectrum and for each star the 10x45 min. exposure sky-subtracted spectra were combined to obtain the final spectrum for the analysis (like Fig.2.6). The mean S/N is 25 for UVES and ∼ 50 for GIRAFFE, per resolution element, at 6650. A detailed view of a spectrum with identified lines is in Fig.2.6

2.2.2 Images

All the Swope images were pre-reduced using IRAF tasks for bias, linearity (based on the recipe discussed in Hamuy et al.(2006)(40)) and flat field corrections. The procedure is the following: first all the bias images (zero exposure time) are combined using zerocombine IRAF task, the resulting ZERO image is then subtracted from all the images using the IRAF task imarith. Then the flat field images (calibration images for each filter) are combined per filter,

Tesis de Doctorado 2.3. IMAGING TECHNIQUES 15

Figure 2.5 MOSAIC CCD distribution, field in this work were centered on chip im6. in our case just V and I, this combined FLAT image is normalized first and then all the images are divided by this FLAT image using the IRAF task imarith. This procedure is standard for all the instruments, as a preparation for the photometry. For MOSAIC 4m data, because it is a wide-field imager divided in 8 different detectors, each CCD was treated independently and the procedure described above applied. This means each chip was treated as a single image. In order to display MOSAIC images a different IRAF package is used called mscdisplay in the package mscred.

2.3 Imaging Techniques

2.3.1 Photometry

The photometry was performed using the IRAF DAOPHOT/ALLSTAR and PHOTCAL packages. Instrumental magnitudes were extracted following the point-spread function (PSF) method (Stetson (1987)(98)). A quadratic, spatially variable, master PSF (PENNY function) was adopted because of the large field of view of both detectors. Aperture corrections were then determined performing aperture photometry of a suitable number (typically 15 to 20) bright, isolated stars in the field. These corrections were found to vary from 0.160 to 0.290 mag, depending on the filter. The PSF photometry was finally aperture corrected, filter by filter.

❄❇❅❆❈❋ R. Mateluna P. - 2012 16 CHAPTER 2. METHOD

Table 2.2 LMC target star clusters. Cluster RA(J2000) DEC(J2000) [Fe/H]a RVa Age b(51) h:m:s d:m:s dex Gyr kms−1 Hodge11 06 14 22 −69 50 54 −1.84 ± 0.04 245.1±1.0 15.2 NGC2257 06 30 12 −64 19 34 −1.59 ± 0.02 301.6±0.8 11.9 NGC 1841 04 45 23 −83 59 49 −2.02 ± 0.02 210.3±0.9 12.5 NGC1718 04 52 25 −67 03 05 −0.80 ± 0.03 278.4±2.2 9.31 NGC1846 05 07 35 −67 27 39 −0.49 ± 0.03 235.2±0.9 9.17 SL869 06 14 41 −69 48 07 −0.40 ± 0.04 258.4±2.1 9.15

a From G06 b Piatti et al. (2009)(76), Geisler et al.(1997)(28), Milone et al.2009(63), Kerber et al.2007

The calibration was done using standard field SA98 (Landolt(1992)(54)) with approximately 20 standard stars in the field. The final rms of the fit to the standards was 0.032 and 0.033 for V and I filters, respectively. At the mean magnitude of the LMC targets, the internal error in V magnitude is 0.025 and in (V-I) color is 0.029 for the Swope data.

2.3.2 Astrometry: From X and Y to RA and DEC

After the photometry was done, the only coordinates to identify stars are X and Y. In order to check our members it is necessary to identify each star by general coordinates: RA and DEC. For that reason, a map like Fig. 2.7 is made and it is used to select a certain amount of stars (about 30). This list of stars has to be identified in an image of the same field where the stars have identified RA and DEC. The ESO-skycat tool is used to display an image from a known catalog, like the Sloan Digital Sky Survey (SDSS). After doing this correlation by eye, a table with X, Y from the photometry and RA, DEC from the catalogue image for the selected stars is made. That list is used as an input in the IRAF task ccmap, which creates a transformation plate, which will be used in the IRAF task cctran. This task performs the transformation of coordinates from X, Y of the total photometry list to RA, DEC using the information of the transformation plate from the task before. Now the stars can be identified by RA and DEC in the photometry list.

2.4 Spectral Analysis

2.4.1 Radial Velocities

Membership of the studied stars for each cluster was established by radial velocity measurement. We used the fxcor IRAF utility to measure radial velocities for both sets of data (GIRAFFE and UVES). This routine cross-correlates the observed spectrum with a template, in this case a synthetic spectrum with the mean atmospheric parameters of the targets (eﬀective temperature (Teﬀ )= 4600K, surface gravity (log g)= 1.5, microturbulence velocity (vt)= 1.6 km/s, metallicity ([m/H])= −2.0). For each studied cluster the radial velocity results are discussed in the next chapters.

Tesis de Doctorado 2.4. SPECTRAL ANALYSIS 17 Å s spectral range 580nm(UVES) ... 4769-5752+5818-6798 Å H651.5A/HR14A(GIRAFFE) ... 6293-6686.9 Å Yale-1m CTIO Y4KCam Imaging V,R,I ... UT2-8m VLT-Paranal FLAMES(G+U) Telescope Observatory Instrument Instrument set up Filter Swope-1m LCO SiTe#3 Cam Imaging V,I ... Blanco-4m CTIO MOSAIC Imaging C, R(T1), I(T2) ... Magellan-Baade-6m LCO IMACS+MOE WB56-92+Mask ... 5600-9200 Table 2.3 Telescopes and Instruments set-up

❄❇❅❆❈❋ R. Mateluna P. - 2012 18 CHAPTER 2. METHOD

Table 2.4 Spectroscopic data set Object Telescope/Observatory Instrument set up Exp. time (s) Exposures GIRAFFE H651.5A/HR14A 2700 10 Hodge11 UT2/VLT UVES 580nm/1” fibers 2700 10 GIRAFFE H651.5A/HR14A 2700 10 NGC1718 UT2/VLT UVES 580nm/1” fibers 2700 10 GIRAFFE H651.5A/HR14A 2700 5 NGC2257 UT2/VLT UVES 580nm/1” fibers 2700 5 NGC1841 Baade/LCO IMACS+MOE WB56-92+MASK 3600 4 3600 3 NGC1846 Baade/LCO IMACS+MOE WB56-92+MASK 2500 1 1800 1

Table 2.5 Imaging data set from LCO Object Telescope/Observatory Instrument ﬁlter Exp. time in (s) seeing V 45(1), 90(2), 300(1), 400(1) 1.45 Hodge 11 Swope/LCO Site#3 CCD I 45(1), 90(2), 300(1), 400(1) 1.45 V 90(2), 300(1), 400(1) 1.40 NGC 1718 Swope/LCO Site#3 CCD I 30(1), 90(2), 300(1), 400(1) 1.30 V 45(1), 90(2), 300(1), 400(1) 1.40 NGC 2257 Swope/LCO Site#3 CCD I 45(1), 90(2), 300(1), 400(1) 1.30 V 90(1),400(2) 1.50 NGC 1841 Swope/LCO Site#3 CCD I 90(1),150(1), 400(1), 480(1) 1.50 V 90(2), 300(1), 400(1) 1.50 NGC 1846 Swope/LCO Site#3 CCD I 90(1), 150(1), 400(1), 480(1) 1.40

Tesis de Doctorado 2.4. SPECTRAL ANALYSIS 19

Table 2.6 LMC Cluster data set from CTIO Object Telescope/Observatory Instrument ﬁlter Exp. time in (s) seeing C 1200(1) 1.0 NGC 1852 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0 NGC 1751 C 1200(1) 1.0 4m/CTIO MOSAIC T1 180(1) 1.0 +NGC 1795 T2 180(1) 1.0 C 1200(1) 1.0 NGC 1917 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0 C 1200(1) 1.0 NGC 1987 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0 C 1200(1) 1.0 IC 2146 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0 C 1200(1) 1.0 NGC 2108 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0 C 1200(1) 1.0 Hodge 3 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0 C 1200(1) 1.0 Hodge 11 4m/CTIO MOSAIC T1 180(1) 1.0 T2 180(1) 1.0

❄❇❅❆❈❋ R. Mateluna P. - 2012 20 CHAPTER 2. METHOD al . M trﬁlsfo CTIO. from ﬁelds star LMC 2.7 Table 42 84 6 93. 7.8-92 0.01 -39.20 278.68 32.7 29 -66 08.41 23 04 1 Field einto hms dms mg sc I R C I R C (sec) I R C (mag) s) m (d s) m (h designation 42 52 6 32. 7.8-89 0.01 -38.97 277.48 23.8 43 -65 45.27 28 04 2 43 13 6 33. 7.5-82 0.02 -38.25 278.95 33.0 03 -67 11.35 31 04 3 43 11 7 95. 8.0-48 0.09 -34.86 287.90 55.5 59 -74 31.11 32 04 4 43 09 6 62. 7.7-80 0.01 -38.03 277.77 26.4 16 -66 40.92 36 04 5 43 79 6 83. 7.9-73 0.04 -37.37 279.09 35.4 28 -67 17.96 39 04 6 44 10 7 95. 8.4-48 0.09 -34.84 286.54 56.6 59 -73 01.09 40 04 7 44 22 6 01. 7.4-71 0.02 -37.19 277.94 18.7 40 -66 12.27 44 04 8 44 08 6 63. 7.4-64 0.04 -36.43 278.64 32.3 26 -67 50.84 49 04 9 00 70.2-94 71211 3.00.11 -35.10 281.19 37.1 48 -69 01.32 57 04 10 10 75.9-75 38288 3.80.06 -35.58 278.87 43.8 51 -67 52.29 57 04 11 20 74.8-81 14289 3.90.06 -34.59 278.97 21.4 11 -68 49.38 07 05 12 30 92.7-74 03284 3.40.06 -34.54 278.45 40.3 46 -67 23.97 09 05 13 40 90.0-85 92296 3.20.08 -33.42 279.66 59.2 59 -68 02.70 19 05 14 50 71.4-04 82215 3.00.08 -32.40 281.54 08.2 44 -70 17.84 27 05 15 60 32.9-80 86284 3.60.06 -32.26 278.42 08.6 09 -68 21.19 33 05 16 70 74.6-44 99260 3.10.11 -30.91 286.05 59.9 46 -74 48.36 37 05 17 80 35.1-91 81295 3.00.06 -31.20 279.50 48.1 10 -69 56.31 43 05 18 90 71.7-21 21229 2.90.11 -29.09 282.93 32.1 16 -72 15.77 07 06 19 00 42.7-95 22201 2.10.10 -28.51 280.16 52.2 50 -69 28.07 14 06 20 10 00.3-24 66234 2.20.08 -28.12 283.46 16.6 44 -72 06.93 20 06 21 α 2000 δ 2000 E b l ( B ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± − .120 e 050101012112412812080.8 0.8 1.2 1.278 1.274 1.281 120 120 500 20 Dec 2008 0.01 .120 e 050101012512612012091.0 0.9 1.2 1.280 1.276 1.285 120 120 500 20 Dec 2008 0.01 .120 e 050101013613513011100.8 1.0 1.1 1.320 1.315 1.326 120 120 500 20 Dec 2008 0.01 .120 e 910 8 8 .2 .2 .2 . . 0.8 0.8 1.1 1.423 1.421 1.426 180 180 1200 19 Dec 2008 0.01 .120 e 050101013913313112100.9 1.0 1.2 1.381 1.373 1.389 120 120 500 20 Dec 2008 0.01 .120 e 050101014014314213100.9 1.0 1.3 1.432 1.423 1.440 120 120 500 20 Dec 2008 0.01 .120 e 910 8 8 .2 .1 .1 . . 0.7 0.9 1.0 1.417 1.414 1.421 180 180 1200 19 Dec 2008 0.01 .120 e 050101014514614511100.7 1.0 1.1 1.455 1.446 1.485 120 120 500 20 Dec 2008 0.01 .120 e 050101015214015113100.9 1.0 1.3 1.501 1.490 1.512 120 120 500 20 Dec 2008 0.01 .120 e 810 0 0 .0 .9 .0 . . 1.0 1.2 1.4 1.300 1.299 1.302 300 300 1500 18 Dec 2008 0.01 .120 e 050101015915615712110.9 1.1 1.2 1.537 1.526 1.549 120 120 500 20 Dec 2008 0.01 .120 e 050101015115715910090.9 0.9 1.0 1.569 1.557 1.581 120 120 500 20 Dec 2008 0.01 .120 e 810 8 8 .7 .7 .7 . . 1.0 1.2 1.4 1.276 1.273 1.279 180 180 1200 18 Dec 2008 0.01 .120 e 810 8 8 .0 .8 .9 . . 0.9 1.0 1.4 1.394 1.385 1.403 180 180 1200 18 Dec 2008 0.01 .120 e 810 8 8 .7 .5 .5 . . 1.0 1.0 1.4 1.456 1.457 1.475 180 180 1200 18 Dec 2008 0.01 .120 e 810 8 8 .9 .7 .8 . . 0.9 1.0 1.4 1.485 1.474 1.497 180 180 1200 18 Dec 2008 0.01 .120 e 910 8 8 .4 .3 .4 . . 0.7 0.8 1.0 1.443 1.439 1.447 180 180 1200 19 Dec 2008 0.01 .120 e 910 8 8 .6 .5 .6 . . 0.7 0.8 1.0 1.362 1.355 1.368 180 180 1200 19 Dec 2008 0.01 .120 e 910 8 8 .3 .2 .3 . . 0.8 0.8 1.1 1.430 1.424 1.437 180 180 1200 19 Dec 2008 0.01 .120 e 810 8 8 .1 .1 .1 . . 0.9 0.9 1.1 1.314 1.311 1.318 180 180 1200 18 Dec 2008 0.01 .120 e 910 8 8 .8 .6 .7 . . 0.8 0.9 1.0 1.474 1.466 1.482 180 180 1200 19 Dec 2008 0.01 V ) aeepsr ims seeing airmass exposure date

Tesis de Doctorado 2.4. SPECTRAL ANALYSIS 21

Figure 2.6 Detailed view of our best UVES spectrum, TARG.6, in the range of 6220 − 6270. Several important lines are identiﬁed.

2.4.2 Abundance Analysis

Eﬀective temperature for one of the studied clusters (H11) was derived from the (V-I) color using the relation by Alonso et al. (1):

5040 T = , where:θ = 0.5379 + 0.3981(V − I) + 0.04432(V − I)2 − 0.02693(V − I)3 (2.1) eﬀ θ o o o

and the reddening E(V-I)= 1.24∗E(B-V), where E(B-V)= 0.08 from W93(107). Surface gravities (log g) were obtained from the canonical equation:

g M T L log( ) = log( ) + 4 · log( eﬀ ) − log( ) (2.2) g⊙ M⊙ T⊙ L⊙

where the mass M/M⊙ was assumed to be 0.8 M⊙, and the luminosity L/L⊙ was obtained from the absolute magnitude MV assuming a true distance modulus of (m − M)0 = 18.5 from Gieren et al.(2005)(35). The bolometric correction (BC) was derived by adopting the relation

❄❇❅❆❈❋ R. Mateluna P. - 2012 22 CHAPTER 2. METHOD

Figure 2.7 Map from the Swope data, used in Astrometry to identify stars from an image with RA and DEC, X and Y coordinates in pixels.

BC-Teﬀ from Alonso et al. (1999)(1). Finally, microturbulent velocity (vt) was obtained from the relation vt-log g used in Marino et al.(2008)(57) for the same spectral type of stars. In the case of SL869 for the determination of the atmospheric parameters the spectroscopic method was used, since this cluster had more FeI and FeII lines than H11. First, for an idea of the parameters the method described above was used and then using the lines EQWs in

MOOG, adjusting iteratively each parameter: Teff by fulfilling the equilibrium of the excitation potential of FeI (see Fig.B.3, top panel), logg from the ionization balance of FeI and FeII (smallest difference) and microturbulence velocity vt by having a value of FeI abundance independent of the line strength (see Fig.B.3, middle panel). Because of the low metallicity and relatively low signal to noise for the H11 spectra, spectrum- synthesis method was used for all the elements. For this purpose five synthetic spectra were

Tesis de Doctorado 2.4. SPECTRAL ANALYSIS 23 calculated (Fig. 2.8) having different abundances for each element, and interpolated to derive the value that minimizes the r.m.s. of the fit. The Local Thermodynamic Equilibrium (LTE) program MOOG (Sneden (1973)(95)) was used for the abundance analysis (see Appendix B). Line list used, based on Villanova et al.(2009)(106), is shown in Table 2.8. In the case of Mn the hyperfine structure was used, for Ba solar isotopic ratios were used. In column two of Table 2.8 was adopted the MOOG notation (two-digit designation to the left of the decimal point, and a single digit to the right of the decimal point to represent the ionization stage, where zero denotes neutral and one denotes singly ionized) and in parentheses is denoted the different isotopes used in the case of Ba. Atmospheric models by Kurucz(1970)(52) were utilized and non-LTE corrections were made for Na lines based on Mashonkina et al.(2000)(58). The solar abundances used are from Villanova et al.(2009)(106).

Table 2.8: Line list used.

Wavelength (Å) A E.P. (eV) log gf 6300.300 08.0 0.000 −9.797 5688.204 11.0 2.104 −0.450 5889.951 11.0 0.000 0.157 5895.924 11.0 0.000 −0.234 5711.083 12.0 4.340 −1.670 6696.014 13.0 3.140 −1.562 6155.134 14.0 5.619 −0.900 6155.687 14.0 5.619 −2.390 6161.287 20.0 2.523 −1.293 6162.170 20.0 1.899 +0.457 6163.745 20.0 2.521 −1.303 6166.429 20.0 2.521 −1.136 6169.032 20.0 2.523 −0.644 6169.555 20.0 2.526 −0.227 6439.070 20.0 2.526 0.474 6493.776 20.0 2.521 +0.129 5526.808 21.1 1.768 0.104 5657.863 21.1 1.507 −0.403 5684.184 21.1 1.510 −1.000 6258.098 22.0 1.440 −0.340 6261.094 22.0 1.430 −0.440 6261.225 23.0 0.267 −2.300 5345.800 24.0 1.004 −0.930 5348.315 24.0 1.004 −1.140 5420.256 25.0 2.143 −3.018 5420.261 25.0 2.143 −2.988 5420.270 25.0 2.143 −2.733 5420.272 25.0 2.143 −3.766

❄❇❅❆❈❋ R. Mateluna P. - 2012 24 CHAPTER 2. METHOD

Table 2.8: continued.

Wavelength (Å) A E.P. (eV) log gf 5420.281 25.0 2.143 −2.812 5420.295 25.0 2.143 −2.511 5420.298 25.0 2.143 −3.687 5420.311 25.0 2.143 −2.745 5420.329 25.0 2.143 −2.327 5420.333 25.0 2.143 −3.812 5420.351 25.0 2.143 −2.771 5420.374 25.0 2.143 −2.169 5420.379 25.0 2.143 −4.164 5420.402 25.0 2.143 −2.947 5420.429 25.0 2.143 −2.029 6230.725 26.0 2.559 −1.231 6232.634 26.0 3.654 −1.083 6297.789 26.0 2.223 −2.620 6318.017 26.0 2.453 −1.928 6335.330 26.0 2.198 −2.177 6393.602 26.0 2.433 −1.522 6430.845 26.0 2.176 −2.006 6494.980 26.0 2.404 −1.403 5176.110 27.0 2.080 −2.120 5342.695 27.0 4.021 0.600 5688.605 27.0 2.080 −2.220 6586.306 28.0 1.950 −2.800 6643.625 28.0 1.676 −2.300 6767.769 28.0 1.830 −1.960 4810.527 30.0 4.080 −0.040 5087.416 39.1 1.080 −0.290 6496.910 56.1(134) 0.604 −0.230 6496.899 56.1(135) 0.604 −1.736 6496.902 56.1(135) 0.604 −1.036 6496.906 56.1(135) 0.604 −0.589 6496.916 56.1(135) 0.604 −1.433 6496.917 56.1(135) 0.604 −1.036 6496.920 56.1(135) 0.604 −1.036 6496.910 56.1(136) 0.604 −0.230 6496.898 56.1(137) 0.604 −1.736 6496.901 56.1(137) 0.604 −1.036 6496.906 56.1(137) 0.604 −0.589 6496.916 56.1(137) 0.604 −1.433 6496.918 56.1(137) 0.604 −1.036

Tesis de Doctorado 2.4. SPECTRAL ANALYSIS 25

Table 2.8: continued.

Wavelength (Å) A E.P. (eV) log gf 6496.922 56.1(137) 0.604 −1.036 6496.910 56.1(138) 0.604 −0.230 6165.891 59.1 0.923 -0.299 6645.095 63.1 1.378 0.130

❄❇❅❆❈❋ R. Mateluna P. - 2012 26 CHAPTER 2. METHOD

Figure 2.8 Example of spectral synthesis method. Five diﬀerent spectra, with Fe abundance varying from 5.12 to 5.52, are shown ﬁtted to the data in the measurement of a line of FeI (6335.3) in TARG.6.

Tesis de Doctorado Chapter 3

The Globular Cluster Hodge 11

3.1 Introduction

Hodge 11(H11) is a typical old, massive globular cluster(GC) in the Large Magellanic Cloud(LMC). As such, and given its relatively uncrowded location, it has been the subject of a number of studies. Since the earliest CMD by its discoverer Hodge(1960)(43), it was suspected that H11 was a bonafide old cluster. (Walker (1993) (107)) -hereafter W93- showed that its very blue and bifurcated horizontal branch(HB) was similar to that of M15, and that it was very metal-poor, with [Fe/H] = −2.0 ± 0.2. Curiously, however, he did not find any RR Lyraes. Johnson et al.(1999)(49) confirmed that H11 was as old as the oldest globulars in the Galaxy. The low metallicity found by W93 has been corroborated by all subsequent investigations, including low resolution blue (Cowley et al.(1982)(22)) and IR Ca triplet (Olszewski et al.(1991)(71)-O91, Gro- cholski et al.(2006)(39)-G06) spectra and finally HRS. H11 was one of the first LMC GCs to be studied with HRS. In a pioneering study by Johnson et al.(2006)(50)-J06, they not only observed H11 but also three other old LMC GCs. They used the MIKE spectrograph at the Magellan telescope to study two stars in H11. Despite the careful HRS study of J06, we felt a new investigation of H11 was warranted. First, J06 used the 6.5m Magellan telescope, while we targeted the 8m VLT. MIKE’s single slit allowed only one star to be observed at a time, and they only observed two stars, while we used the multiplexing capability of FLAMES to eventually observe eight members. Their resolution was 19,000, comparable to our GIRAFFE data, while our UVES data is a much higher 47,000. Their total exposure time was only 1 - 1.5h while we integrated for a total of 7.5 hrs. They obtained abundances for 12 elements but these did not include Na, O, Al, Si, or Zn while we observed these five elements as well as their original 12. Na and O are particularly important given their almost ubiquitous anti-correlation in both GGCs (e.g. Carretta et al.(2009)(16)) as well as the few old, massive LMC GCs in which these elements have so far been studied (Mucciarelli et al.(2010)(67)-M10). Silicon(Si), as one of the α elements, is also critical for examining such key issues as the Robertson et al. (2005)(86)’s prediction. J06 also found several results which sparked our interest in confirming or denying them. First, their metallicity of [Fe/H] = −2.21 ± 0.01 was lower than any other previous measurement, and 0.37 dex lower than the value we derived in our Ca triplet study (G06). Second, they found that, 28 CHAPTER 3. HODGE 11

Figure 3.1 Image from CTIO4m MOSAIC of H11. although the α element Mg was halo-like, both Ca and Ti were depleted with respect to the halo, at odds with M10. J06 suggested that the early chemical evolution of the LMC was significantly different from that of the halo, particularly with regards to the α elements, and more like that of dwarf spheroidals(dSphs), at odds with the Robertson et al. (86)’s prediction and M10 findings for other LMC GCs. Clearly, a more detailed investigation of H11 chemical abundances can address a number of outstanding issues. We present here our findings published in Mateluna et al. (59).

3.2 Results from Photometry

A previous set of images for H11 were obtained using FORS2@VLT during the course of the preimages obtained for the spectroscopic study by G06 and we reduced this data as well. The photometry procedure is described in chap.2. The results were calibrated using published

Tesis de Doctorado 3.3. RADIAL VELOCITIES FOR HODGE 11 29 photometry from W93. A color-magnitude diagram of H11 is shown in Fig. 3.2 using both photometric sets. The FORS2 data is much deeper than the Swope data but the brightest stars appear saturated, including some of our targets. This was the main reason to use the Swope photometry and not FORS2 in our analysis. The mean diﬀerence, without regard to sign, between the Swope and FORS2 photometry around the RGB is ∆V=+0.1 and ∆(V-I)=+0.025. We decided to use the Swope data for our stars except for TARG.2, because in the 1.0m photometry it is contaminated by another star.

3.3 Radial Velocities for Hodge 11

For the 101 stars observed with GIRAFFE we found a total of seven stars with radial velocities

(RV) comparable to H11 radial velocity value from G06 (RV= 245.1 ± 0.3 km/s and σRV = 1.0 km/s), but only four of them were finally selected as H11 members. The other three stars have a completely different metallicity (as revealed by the numerous strong FeI lines in their spectra). From the UVES data, four stars have a radial velocity and metallicities comparable to the values of H11 from G06. The final analysis thus gives a total of eight members for our cluster. A histogram of radial velocities is shown in Fig.3.4. From the measured radial velocities we obtained a mean heliocentric radial velocity < RV >H= 245.9 ± 0.9 km/s and a dispersion σ = 2.5 km/s. Table 3.1 lists the basic parameters of the selected stars:

3.4 Abundances Results

All results for the abundances of the diﬀerent chemical species for our eight members of H11 are presented in Table 3.2. In parentheses we give the number of lines used.

3.4.1 Fe-peak elements

The mean value of [Fe/H] from our eight cluster members is: <[Fe/H]>= −2.00 ± 0.04 and

σobs = 0.11 ± 0.03, with no evidence for any intrinsic variation. J06 give values of −2.21 ± 0.01 (FeI) and −2.05±0.06 (FeII) from their two stars. Their FeI value is significantly lower than ours (also obtained with FeI), while their FeII value is comparable to our FeI value. Two previous studies employed the Ca triplet technique (CaT) to derive the mean metallicity of H11. O91 observed two stars, finding [Fe/H]= −2.06 ± 0.2 while G06 derived a mean of −1.84 ± 0.04 (internal error only) from 12 giants. Our value agrees well with the former and is significantly lower than that of the latter. Photometrically, W93 found a value of [Fe/H]= −2.0 ± 0.2, the same as ours. H11 clearly remains one of the LMC’s most metal-poor constituents.

The mean values for the other iron-peak elements are <[Sc/Fe]>= 0.07±0.09, σobs = 0.16±06, <[Cr/Fe]>= −0.49 ± 0.01, σobs = 0.02 ± 0.08, <[Ni/Fe]>= −0.16 ± 0.05, σobs = 0.13 ± 0.04. We present our results for these iron-peak elements in comparison with the literature data in Fig. 3.5. In addition, we derive [Mn/Fe]=-0.49 and [Co/Fe]=0.26 from a single (albeit our best UVES) star. Our results are in excellent agreement with those of J06 for Sc and Ni, while the accord is reasonable for both Mn and Co. However, our results show a very low abundance in [Cr/Fe]

❄❇❅❆❈❋ R. Mateluna P. - 2012 30 CHAPTER 3. HODGE 11

Figure 3.2 Color-magnitude diagram for H11. In black dots, photometry from FORS2(VLT) pre-images from G06. In blue triangles our data from the Swope Telescope at Las Campanas Observatory. H11 stars observed spectroscopically in this study are shown in red triangles (Swope data) and red dot (FORS2 data).

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 31

Figure 3.3 Swope Map.

❄❇❅❆❈❋ R. Mateluna P. - 2012 32 CHAPTER 3. HODGE 11

Figure 3.4 Velocity histogram for GIRAFFE and UVES data. In red are the Hodge 11 members. There were three additional stars with radial velocities very close to that of the cluster, but the spectra showed many more lines, therefore their metallicities are much higher, and they were considered ﬁeld stars.

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 33 log g S/N t v 8190 1.8535 1.62 0.31 1.67 1.20 1.03 42 22 20 eﬀ 706 1.56 1.47 12 2 47209 1.51 5022 1.64 1.46 1.86 60 50 97 507207 1.46 4837 1.86 1.46 1.85 52 40 (km/s) (mag) (K) (km/s) @6650 () Object R.A. Dec. Type of Star Instrument RV V V-I T TARG.8 06:14:24.65 -69:50:15.4 AGB GIRAFFE 246.8 18.04 0.9 TARG.9 06:14:21.70TARG.4 -69:49:56.4 06:14:17.05 -69:50:42.6TARG.2 06:14:22.42 RGB -69:51:17.6 RGB RGB UVES UVES 247.5 GIRAFFE 244.2 16.88 245.2 16.75 1.18 17.81 45 1.21 1.1 45 TARG.6 06:14:21.85 -69:50:32.2 RGB UVES 241.7 15.86 1.50 40 TARG.11 06:14:31.46 -69:49:35.3TARG.16 06:14:23.34 -69:52:38.6 RGBTARG.10 06:14:30.36 -69:49:50.0 AGB UVES RGB GIRAFFE 244.7 249.9 GIRAFFE 17.42 17.98 1.13 247.4 0. 4 18.21 1. Table 3.1 Important parameters for our target stars

❄❇❅❆❈❋ R. Mateluna P. - 2012 34 CHAPTER 3. HODGE 11 al . XF]vle o og 1stars. 11 Hodge for values [X/Fe] 3.2 Table a uI05()...... 06()...... 06 0.05 0.62 ...... 0.66(1) ...... 0.59(1) EuII LaII a 58)0.03(1) (5688) NaI YII a . .42 .02 .62 . . . . .30.25 0.05 0.43 0.54 ...... 0.56(2) ... 0.60(2) 0.14(2) 0.49(1) ... 0.57(1) 0.57(1) NaD NaD OI MnI n 0.16(1) ZnI BaII [ [ o .63 ...... 0.26(3) NiI CoI a (5688) NaI i .92 .82 .11 ...... 60.06 0.16 ...... 0.21(1) 0.18(2) 0.09(2) CrI TiI cI0.15(3) ScII l .91 ...... 0.11 0.39 ...... 0.12 ... 0.25 ...... 0.33(1) 0.22(1) ... 0.33(1) 0.05(7) 0.26(1) 0.52(1) 0.11(1) 0.39(1) 0.41(1) CaI SiI AlI MgI hr orsod oayceia species. chemical any to corresponds X Where Fe/H X/Fe L E NLT ] ] a L E NLT − − − − − − − − AG6TR. AG4TR.1TR. AG1 AG8TARG.10 TARG.8 TARG.16 TARG.2 TARG.11 TARG.4 TARG.9 TARG.6 0 0 0 0 2 0 0 0 ...... 09 08 49 16 11 08 22 51 1 ...... (1) 1 .22 .91 ...... 40.14 0.04 ...... 0.19(1) 0.02(2) (1) 1 ...... (1) 1 .31 0.17(1) 0.23(1) (1) (7) (3) (1) (2) ... − − − − − − − − − 0 0 0 2 0 0 0 0 0 ...... 21 37 15 03 37 52 47 12 04 2 0.24(2) (2) (1) 1 .61 ...... 60.18 0.06 ...... 0.16(1) (1) (2) (2) (1) (1) 2 .73 ...... 70.16 0.07 ...... 0.17(3) (2) 5 .43 .31 .01 .41 .71 .51 .30. 0.13 0.05(1) 0.27(1) 0.14(1) 0.20(1) 0.23(1) 0.14(3) (5) − − − − − 0 2 0 0 0 . . . . . 08 06 10 23 49 (1) (5) 2 ... (2) (1) 2 ...... (2) − − − − − 0 2 0 0 0 . . . . . 16 02 17 31 12 1 ...... (1) (5) 2 . . . . .20.32 0.02 ...... (2) 1 ...... (1) (1) − − − 1 0 0 . . . 81 24 31 (2) 1 0.03(1) (1) (1) − − 2 0 . . 08 10 (2) 1 0.34(1) (1) − − 0 2 . . 11 06 1 ... (1) (2) − − 1 0 . . 86 13 (3) (1) < [X/Fe] − − − − − − 2 0 0 0 0 0 ...... 00 49 16 14 32 01 σ > 0.11 0.02 0.13 0.17 0.14 0.23 obs 10

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 35 compared to J06 and to halo stars. Venn et al. (105) also found a low value for Cr in Carina metal poor stars. This could be explained as a lack of high energy SNeII in the environment of the protocluster. Possibly these stars did not form or their gas was removed by SNe-driven winds (Umeda and Nomoto(2002)(102)). J06 found an even lower value, −0.67, for Mn. They also found a low abundance for [V/Fe] in their other clusters, reminiscent of our Cr and both Mn values for H11. We have no explanation for the Cr diﬀerence with J06, but note that our three stars all give very similar values. We generally conﬁrm what J06 found: Fe-peak elements in LMC clusters are relatively depleted or even strongly depleted in the case of Cr and Mn, and comparable to the values of dSph stars, although the latter data are sparse.

3.4.2 α-elements

The mean abundances for the main alpha-elements are <[Mg/Fe]>= 0.25 ± 0.06 and σobs = 0.12 ± 0.04, <[Si/Fe]>= 0.39 ± 0.06 and σobs = 0.11 ± 0.04, <[Ca/Fe]>= 0.13 ± 0.04 and σobs = 0.10 ± 0.03 and <[Ti/Fe]>= 0.16 ± 0.03 and σobs = 0.06 ± 0.02. The respective values from J06 are <[Mg/Fe]>= 0.46 ± 0.02, <[Ca/Fe]>= 0.30 ± 0.03 and <[Ti/Fe]>= −0.04 ± 0.05 (they did not measure Si). Fig. 3.6 shows our results for the α element ratios in comparison with the literature, in particular J06, M10 and halo stars. The agreement with J06 is fair. In the case of Mg we get a lower value than J06, M10 and halo stars. J06 got a value in agreement with the abundance of halo stars for H11, but the other GCs in their study show lower Mg values. M10 found values for their GCs comparable to the halo with some slightly lower values. We observe in our results a dispersion in Mg. Venn et al. (2012)(105) suggest that this dispersion could possibly be due to inhomogeneous mixing of the interstellar gas. For Si J06 has no data, but their other GCs lie on the halo trend, as do the mean of M10 clusters. Our results are slightly low in comparison to halo stars. We show a particularly low abundance in [Ca/Fe], confirming the J06 results for other GCs but not for H11, for which they find an abundance 0.17 dex higher than we do. M10, on the other hand, find halo-like Ca abundances in their sample. If our low Ca abundance is correct, this could portend a serious problem for the CaT technique, which assumes that [Ca/Fe] is the same for the LMC as for the Galactic clusters used to calibrate the Ca to Fe relationship. Our [Ca/Fe] value is about 0.3 dex less than that of halo stars used to calibrate the CaT technique. Thus, CaT, applied to H11 but using Galactic calibrators, should derive a relatively metal-poor metallicity compared to its actual [Fe/H]. This difference, however, is in the opposite sense to that required to explain the offset between our [Fe/H] value and that derived by G06 from CaT. However, it is not clear that the Ca triplet metallicities should scale in an obvious way with [Ca/Fe]. Also, to properly address this issue would require analysing a sample of Galactic GCs using the same techniques and solar abundances as adopted in our analysis of H11. Pompéia et al.(2008)(82) found that there was no real correlation between the Ca triplet metallicities, FLAMES high-resolution [Fe/H], and [Ca/Fe] for a large LMC field star sample. This may have to do with the continuum level around the Ca triplet lines, which is set by H- ions, whose abundance is controlled much more by α-elements and sodium than it is by

❄❇❅❆❈❋ R. Mateluna P. - 2012 36 CHAPTER 3. HODGE 11

Figure 3.5 Abundances for iron peak elements. Magenta dots are our results for Hodge 11, in red the LMC data: triangles correspond to J06, open circles correspond to Pompéia et al.(2008)(82), stars correspond to Mucciarelli et al.(2008)(65) and M10. In black dots Galaxy data from Fulbright(2000)(26), Lee et al.(2002)(55), Cayrel et al.(2004)(17) and Reddy et al. (2006)(84), in blue squares data of dSph galaxies from Shetrone et al. (2001)(91) and Sbordone et al. (2007)(89). Error bar stands for σtot from Table A.1.

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 37 the Fe abundance. There is also a complication that the Ca triplet abundances are determined using a V magnitude, and if the α elements are very different, then the stellar colors will shift, changing V for a given bolometric luminosity. In any case, any shift empirically seems to be small. Our Ti abundance is 0.2 dex higher than that of J06, who found their value, which was slightly subsolar, substantially lower than halo stars at the same metallicity. They found a similar result for their other GCs. Our value is only slightly lower than the halo mean and in good agreement with M10 values. Given the limitations of only a few lines in a small number of stars, and the similar nucleosynthetic genesis of these elements, the best metric is to derive the mean α abundance for all four elements. Unfortunately, we have all four elements in only three stars. The mean [α/Fe] ratio vs. [Fe/H] is plotted in Fig. 3.7, along with the equivalent data for J06 and M10 GC, LMC field stars from Pompéia et al.(2008)(82), Galactic data from Fulbright(2000)(26), Cayrel et al.(2004)(17), Lee et al.(2002)(55) and Nissen and Schuster(2010)(69), and dSphs data from Shetrone et al. (2001)(91), Sbordone et al. (2007)(89) and Monaco et al.(2005)(64). This diagram is an excellent diagnostic for separating out halo from dSph stars, especially at the more metal-rich end (e.g. Geisler et al. (2007)(32)), as is clearly seen. Note that the Nissen and Schuster(2010)(69) low-α stars are indeed a lower envelope to the halo distribution and that dSph stars are normally at significantly lower [α/Fe] for the same metallicity. According to the Robertson et al. (2005)(86) scenario, the halo was mainly built up by the merger of only a few very massive dwarfs with LMC-like masses, with fast evolutionary timescales and thus enhanced [α/Fe] ratios from SNeII. The dSphs we see today are much lower mass survivors, who experienced much slower star formation rates which allowed SNIa to eventually contribute their ejecta, leading to depleted [α/Fe] ratios at higher metallicities. Where does the LMC fit into this diagram? Looking at the M10 data yields the clear impression that, at least at metallicities up to about -1.6, the LMC does indeed overlap well with the halo. However, all other LMC data, including J06 GCs and the Pompéia et al.(2008)(82) field stars, as well as our three H11 stars, indicate that, even at the low metallicity of H11, the LMC was already showing SNe Ia depletion effects, mimicking the dSphs. This depletion compared to the halo continues to grow with metallicity, becoming quite severe by [Fe/H]= −1. The overall impression, then, is that the LMC is more dSph-like than halo-like in this diagram, confirming what J06 pointed out and in contradiction to M10’s findings and Robertson et al. (2005)(86)’s prediction.

3.4.3 Neutron Capture Elements

[Ba/Fe], [Y/Fe] and [Eu/Fe] abundances for H11 are shown in Fig. 3.8 in comparison with literature data for the Galaxy halo, dSphs and LMC GCs and ﬁeld stars. The average abundance for each neutron capture element is <[Y/Fe]>= 0.04 ± 0.08 and σobs = 0.14 ± 0.06, <[Ba/Fe]>= −0.01 ± 0.08 and σobs = 0.23 ± 0.06 and <[Eu/Fe]>= 0.62 ± 0.04 and σobs = 0.05 ± 0.02. Our results are in good agreement with J06 for both Y and Ba. However, it is diﬃcult to make a distinction between the Galactic halo and the dSph locus at this metallicity. In [Ba/Fe] we have particularly good agreement with M10 results. On the other hand, our [Eu/Fe] abundance is lower than J06 for H11 and again is comparable to

❄❇❅❆❈❋ R. Mateluna P. - 2012 38 CHAPTER 3. HODGE 11

Figure 3.6 Abundances for α-elements compared to the literature. Magenta dots are our results for Hodge 11, in red the LMC data: triangles correspond to J06, open circles correspond to Pompéia et al.(2008)(82), stars correspond to Mucciarelli et al.(2008)(65) and M10. In black dots Galaxy data from Fulbright(2000)(26) and Lee et al.(2002)(55), in blue squares data of dSph galaxies from Shetrone et al. (2001)(91), Sbordone et al. (2007)(89) and Monaco et al.(2005)(64). Error bar stands for σtot from Table A.1.

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 39

Figure 3.7 a bundance ratio vs. [Fe/H].]Mean [α/Fe] abundance ratio vs. [Fe/H]. Same symbols and colors as in Fig. 3.6 with the addition of Nissen and Schuster(2010)(69) high- and low-α halo stars (black open circle).

❄❇❅❆❈❋ R. Mateluna P. - 2012 40 CHAPTER 3. HODGE 11

M10 results, a little higher than Galactic halo stars. We obtained a value of log ε (La/Eu) of 0.07, which places the cluster in the r-process only region, according to Sneden et al. (2008)(97) (Figure 12.a). This means that the cluster was formed from material polluted by SNe only, with no contribution from AGB stars, similar to the most metal poor stars of our Galaxy ([Fe/H]< −3.0, Sneden et al. (2008)(97)). In Fig. 3.9 the [Ba/Y] ratio vs. [Fe/H] shows H11 falls in the same region as both Galactic and dSph stars. One clearly sees that both Galactic and dSph stars share the same low metallicity trend, with the [Ba/Y] ratio increasing as metallicity increases until approximately [Fe/H]= −2.0, the metallicity of H11. At higher metallicity, the LMC and dSph stars separate from the Galaxy. This can be observed in Fig. 3.9 where LMC data from other authors, in red, are included. Colucci et al.(2012)(21) claim that the LMC has undergone much slower star formation than the Galaxy, based on the fact that for the LMC the [Ba/Y] ratio increases with decreasing age and with increasing metallicity above −2. In addition, we show in Fig. 3.10 the [Ba/Eu] ratio vs. [Fe/H], our H11 data in magenta, LMC data in red, Galactic data in black and dSphs in blue. H11 clearly falls in the r-process regime, given that [Ba/Eu] has been used as a test for the s- or r- origin of these elements (Sneden et al. (1997)(96)).

3.4.4 Na and O In order to study the Na and O content in H11 compared to Galactic (Carretta et al.(2009)(16)) and LMC GCs (Mucciarelli et al.(2008)(65)), we present the abundances of [Na/Fe] and [O/Fe] in Fig. 3.11. Unfortunately, our data is limited to only three stars. The mean value in H11 for [Na/Fe]= −0.32 ± 0.07, σobs = 0.14 ± 0.05 and for [O/Fe]= 0.54 ± 0.03, σobs = 0.05 ± 0.02. H11 stars present a low [Na/Fe] abundance, conﬁrming J06’s suggestion that LMC stars were probably born with low [Na/Fe]. A very high [O/Fe] abundance is shown in Fig.3.11 for our stars. They stand at the extreme end of the trend of the anti-correlation displayed by Galactic

GCs. From the comparison of σobs and our total internal error, we ﬁnd a hint of a spread in the Na abundance, but more data is needed to conﬁrm this. No O spread is visible, but this is expected at the position of our stars in the Na:O relation.

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 41

Figure 3.8 Neutron-capture elements abundances. Magenta dots are our results for Hodge 11, in red the LMC data: triangles correspond to J06, open circles correspond to Pompéia et al.(2008)(82), stars correspond to Mucciarelli et al.(2008)(65) and M10. In black dots Galaxy data from Fulbright(2000)(26), Lee et al.(2002)(55), Reddy et al.(2003)(83), Reddy et al. (2006)(84), in blue squares data of dSph galaxies from Shetrone et al. (2001)(91) and Sbordone et al. (2007)(89). Error bars stands for σtot from Table A.1.

❄❇❅❆❈❋ R. Mateluna P. - 2012 42 CHAPTER 3. HODGE 11

Figure 3.9 [Ba/Y] ratio vs. [Fe/H]. H11 in magenta dots, Galactic data in black: dots from Fulbright(2000)(26) and open circles from Nissen and Schuster(2011)(70), blue squares are dSph data from Shetrone et al. (2001)(91) and Shetrone et al. (2003)(92), LMC data in red: triangles are J06, stars are old GCs from M10 and intermediate age clusters from Mucciarelli et al.(2008)(65), open circles from Pompéia et al.(2008)(82). Segmented line showing the Galactic trend.

Tesis de Doctorado 3.4. ABUNDANCES RESULTS 43

Figure 3.10 [Ba/Eu] ratio vs. [Fe/H]. H11 in magenta dots, Galactic data in black dots from McWilliam et al.(1995)(61), Reddy et al.(2003)(83), Reddy et al. (2006)(84), Burris et al.(2000)(12), Fulbright(2000)(26), blue squares are dSph data (Shetrone et al. (2001)(91), Shetrone et al. (2003)(92), Geisler et al.(2005)(31)), LMC data in red: triangles are J06, stars are old GCs from M10 and intermediate age clusters from Mucciarelli et al.(2008)(65).

❄❇❅❆❈❋ R. Mateluna P. - 2012 44 CHAPTER 3. HODGE 11

Figure 3.11 Left: [Na/Fe] and [O/Fe] vs. [Fe/H], right: [Na/Fe] vs [O/Fe] showing the Na:O anti-correlation in GCs. Magenta dots are our results for H11, in red other LMC data: triangles are results from J06, open circles are results from Pompéia et al.(2008)(82), stars are results from Mucciarelli et al.(2008)(65) and M10; in black ﬁlled circles are Galactic GC data from Carretta et al.(2009)(16) and Lee et al.(2002)(55), in blue squares dSph galaxies from Shetrone et al. (2001)(91), Sbordone et al. (2007)(89) and Monaco et al.(2005)(64). Error bars stands for σtot from Table A.1.

Tesis de Doctorado Chapter 4

SL869 and the Surrounding Field of H11

4.1 Introduction

From the H11 spectroscopic (FLAMES) and photometric (Swope and FORS2) data, we intentionally obtained simultanously data of the nearby star cluster SL869 as well as surrounding field stars. What little is known of this cluster suggests it is a typical intermediate-age cluster. This allows us to obtain a second point on the AMR, on the opposite end of the age, and presumably also metallicity spectrum, from H11, for free, taking full advantage of the wide field of the instruments. The only prior metallicity measurement of SL869 was performed by Grocholski et al.(2006)(G06) (39) spectroscopically, obtaining a value of −0.40 dex σavg = 0.04 from three stars using the Cal- cium triplet technique. They also obtained a mean radial velocity of RV= 258.4 , σavg = 2.1 kms−1. This velocity value from G06 was used by us to discriminate stars from the whole set of data (UVES+GIRAFFE) to select targets of this cluster (see radial velocities in Table4.1). Walker (1993)(107) in his study of Hodge11 photometrically (taken with the 0.9m telescope at CTIO) derived an isochrone age of 1.5 Gyr, [Fe/H]= −0.46 for SL869 and this is the only measurement of age for this object from isochrone fitting. Piatti (2011)(77) estimate an age for this star cluster from Washington photometry, using the δ(T1) technique(Geisler et al.(1997) (28)), of 1.70±0.15 Gyr. In this work dataset, LMC field stars are also included and spectroscopically analyzed, in order to determine abundances of different elements such as Fe, Ca, Ti, Cr and Ni. This allows the study of different abundance trends in the LMC and to compare abundances from the LMC fields to Galactic halo stars, especially at low metallicities if such stars are available. Such stars can tell us something about the possibility of the LMC to be a prototypical ’building block’ of our Galaxy. In addition, the field population does not show the pronounced age gap exhibited by the star clusters in the LMC. Therefore, if we can estimate age for these stars, they can help to fill in the famous age gap in the AMR and help us to try to understand better the chemical evolution of our neighbor galaxy. Pompéia et al.(2008) (82) studied a sample of LMC field stars with high resolution spec- 46 CHAPTER 4. SL869 AND THE SURROUNDING FIELD OF H11 troscopy finding that the α−elements (Ca, Si and Ti) show lower [X/Fe] ratios than Galactic stars at the same [Fe/H]. They also found an offset for the [iron-peak/Fe] ratios of Ni, Cr and Co, with a depleted pattern and sub-solar values. In general, they found that: "The overall chemical distributions of this sample indicates a slower star formation history relative to that of the solar neighborhood, with a higher contribution from type Ia supernovae relative to type II supernovae". Our analysis will attempt to shed additional light on these and other issues.

4.2 Star Cluster SL869

In the present work we analyze, for the ﬁrst time, HRS (GIRAFFE+UVES from VLT) data of SL869 (2 stars, in Table4.1). In addition, a new age estimation is determined with the isochrone ﬁtting method on color-magnitude diagram (CMD) from FORS2 photometry, using the Padova Isochrones (http://stev.oapd.inaf.it/cgi-bin/cmd) for a the metallicity value obtained from our HRS study.

4.2.1 Photometry

A previous set of images for SL869 was obtained using FORS2 during the course of the pre- images obtained for the spectroscopic study by G06 by centering one chip on H11 and the other on SL869. We reduced this data as well. The photometry procedure is described in chap.2. The results were calibrated using the photometry from the Swope telescope from this work. The procedure consists in first finding the common stars in both photometries. We found 380 stars. We then limit our sample to just the stars with magnitudes brighter than 19.5 (134 stars) in order to maximize the photometric quality but also include the full color range in the CMD. We the plotted the difference between the two system magnitudes, with VS as the magnitude V(calibated) from Swope stars and vF , the magnitude in V(instrumental) from FORS2 (the same notation is applied to I filter), vs. (v-i)F , the color from the FORS2 data (color range is -0.2-2.5). To these points we fit a line using least square fitting, finding an rms= 0.054 for V and rms= 0.061 for I, and calculate a slope (CT) and intersection point (∆V). This procedure is done for the v and i filters independently. Finally, the parameters obtained are applied to the FORS2 photometry in the foillowing way: VFS = CT(v-i)F +∆ V+vF , and IFS = CT(v-i)F +∆ I + iF , where VFS and IFS are the V and I magnitudes in FORS2 calibrated with Swope. The coefficients are: CTV =0.052, ∆V= −0.449 , CTI = −0.00089 and ∆I = −0.482. After this procedure a color-magnitude diagram is made for SL869, see Fig. 4.2, using both photometric sets and a radial cut of 200 pixels for both sets. This CMD presents a clear turnoff(TO), stars in the red giant branch and a clear clump at around magnitude V∼ 19. Also, the CMD shows about 4 magnitudes of main sequence(MS), allowing us to derive a good isochrone age. Since the FORS2 data is much deeper than the Swope data, it is used for the age estimation. But in the spectroscopic analysis the Swope data was used because it was calibrated with standard fields observed during the same night as the cluster data, and to be consistent with the H11analysis.

Tesis de Doctorado 4.2. STAR CLUSTER SL869 47

Figure 4.1 Photometry calibration for FORS2 data of SL869 using Swope calibrated photometry.

4.2.2 Abundance Analysis

Before starting any spectroscopic analysis, first it is important to appropriately select the objects of study. The first selection criteria was radial velocity. We know the mean value for SL869 from G06. After comparing our results of radial velocities from both GIRAFFE and UVES datasets, 3 possible targets are found. Next the position criteria is used, where we discarded one star for being far removed from the cluster center, and finally we are left with our cluster sample of 2 stars, listed in Table4.1. With the information from the photometry, is possible to determine stellar atmospheric parameters (Table 4.1), to first order in the way described in section 2.4.2., for each star. Since this star cluster is metal rich we expect to find more FeI and FeII lines than in H11. Therefore, is possible to determine stellar atmospheric parameters based on the spectroscopic method, using the EQWs of iron lines (see Table4.2) in MOOG, adjusting each parameter: Teff by fulfilling the equilibrium of the excitation potential of FeI, basically minimizing the slope of the green line from Fig.B.3, top panel; logg from the ionization balance of FeI and FeII , by minimizing the difference between those two values; and microturbulent velocity vt by having a value of FeI abundance independent of the line strength, in other words a minimal value for the slope (green line) in Fig.B.3, middle panel. The abundance determination is divided in two parts: equivalent width determination and synthetic spectra. FeI and FeII abundances were determined with EQWs using abfind in MOOG. All other elemental abundances were determined with synthetic spectra using synth and the same lines used in H11. The results from this analysis unfortuntately has large uncertainties for one of the targets because of its low signal-to-noise ∼ 12 (TARG.14). However, the other target has a quite decent value (S/N∼ 50). From this work, the average iron abundance for the two stars in SL869 is <[FeI/H]>= −0.47,

❄❇❅❆❈❋ R. Mateluna P. - 2012 48 CHAPTER 4. SL869 AND THE SURROUNDING FIELD OF H11

Figure 4.2 Color-magnitude diagram for SL869. In black dots, photometry from FORS2(VLT) pre-images from G06. In blue triangles our data from the Swope Telescope at Las Campanas Observatory. SL869 stars observed spectroscopically in this study are shown in red (Swope data).

Tesis de Doctorado 4.2. STAR CLUSTER SL869 49 log g S/N t v eﬀ 4578 2.22 1.91 12 .21 4593 1.88 1.39 48 of Star (km/s) (mag) (K) (km/s) @6620 (Å) Object R.A. Dec. Type Instrument RV V(Swope) V-I(Swope) T TARG.14 06:14:39.17 -69:47:31.0 RGB UVES 263.06 17.42 1.28 TARG.13 06:14:40.79 -69:47:59.9 RGB GIRAFFE 263.84 17.63 1 Table 4.1 Important parameters for our target stars

❄❇❅❆❈❋ R. Mateluna P. - 2012 50 CHAPTER 4. SL869 AND THE SURROUNDING FIELD OF H11

Table 4.2 Iron abundances for our target stars Object Instrument Fe I nr. lines FeII nr. lines [FeI/H] TARG.13 GIRAFFE 7.00 20 6.99 1 -0.50 TARG.14 UVES 7.06 75 7.04 4 -0.44

σint = 0.04. This result is in reasonable agreement with what G06 obtained for this cluster: - 0.40. According to the G06 ﬁndings, intermediate-age clusters in the LMC show a very tight metallicity distribution (with mean metallicity −0.48 dex) and from the results present in this work, SL869 falls perfectly into this category. The abundances of α-elements, iron-peak elements and n-capture elements for this star cluster will be derived in the future.

4.2.3 Age determination

With the calibrated FORS2 photometry, we can fit isochrones to the CMD of SL869. The age estimation was done by fitting Padova isochrones, available online. Using the iron abundance from the spectroscopic analysis, an isochrone of Z= 0.007 ( [Fe/H]= −0.43) is chosen for different ages (1−2.8 Gyr). In Fig. 4.3 is the CMD for SL869 from FORS2 photometry with four different isochrones with ages 1Gyr in red, 1.3 Gyr in green, 1.6 Gyr in blue and 2 Gyr in magenta. The value for reddening used is the same as H11 (E(B-V)= 0.08 from Walker (1993)(107)) and the absolute distance modulus value ((m-M)o = 18.5 from Gieren et al.(2005)(35)). The two isochrones that best fit the TO, subgiant branch and clump are the 1.3 Gyr and 1.6 Gyr isochrones. The mean of these two ages becomes our best age estimation for SL869: 1.45 Gyr, σ = 0.2Gyr. This result is in good agreement with the value obtained by Walker (1993)Walker (107) of 1.5 Gyr, and in reasonable agreement with the Piatti (2011) (77) value of 1.70±0.15Gyr.

4.3 Field Stars

In the field of H11, a large number (30) of field stars were observed spectroscopically with FLAMES (see chap.2) with the purpose of studying these stars to determine different elements like Fe, Ca, Ti, Cr and Ni. Here we present preliminary results and they are compared mainly with the work of Pompéia et al.(2008)(82) (and references there in) for the LMC field stars. As mentioned in the introduction of this chapter, Pompéia et al.(2008)(82) studied a sample of LMC field stars with high resolution spectroscopy finding very interesting results.

4.3.1 Abundance Analysis

The ﬁnal data set consist of 30 stars from GIRAFFE observations. Color and magnitude for each star was necessary in order to estimate initial atmospheric parameters. Because at the time when this work was done, no photometric data was available from our own observations, it was decided to use the 2MASS catalogue, to obtain J, H, K magnitudes and use Alonso et al. (1999)(1) procedure to obtain atmosphere parameters (similar to what is described in section 2.4.2). The S/N ratio for these spectra is unknown, because the data were lost due to technical problems, and thus we cannot calculate our errors properly, but we do have data

Tesis de Doctorado 4.3. FIELD STARS 51

Z= 0.007 1 Gyr 1.3 Gyr 1.6 Gyr 2 Gyr

Figure 4.3 Age estimation for SL869 using Padova Isochrones with FORS2 photometry.

❄❇❅❆❈❋ R. Mateluna P. - 2012 52 CHAPTER 4. SL869 AND THE SURROUNDING FIELD OF H11 from new observations for which we will determine accurate abundances in the near future. The results presented in this section are preliminary and the large spread of the results could be due to the lack of radial velocity selection. Normally, stars with radial velocities between 235-310 kms−1(Fig.??) are selected as LMC members, but in this case this cut was not done. This is a further reason why a second revision must be done to this data in the near future. To determine the abundances, equivalent width (EQW) were measured for several lines in Fe, Si, Ca, Ti, Sc, Cr and Ni and then processed in MOOG using the driver abﬁnd and model atmospheres from Kurucz(1970)(52) with the atmospheric parameters found from the 2MASS colors. All preliminary abundances for this set of data are listed in Table 4.3 and plotted, in magenta triangles, in comparison with literature in Figs.4.4, 4.5 and 4.6. In red LMC data from Pompéia et al.(2008)(82) (open circles) and Mucciarelli et al.(2008)(65)(intermediate-age clusters), in black data of the Galaxy (Fulbright(2000)-(26), Lee et al.(2002)-(55)) and in blue data of dSphs from Shetrone et al. (2001)(91), Sbordone et al. (2007)(89) and Monaco et al.(2005)(64). The data shows a big scatter in comparison with Pompéia et al.(2008)Pompéia et al. (82) for all elements and metallicities. This could be due to either the preliminary nature of the abundances and /or the lack of a radial velocity cut, both of which will be improved later.

• α-element abundances: This work results are in agreement with Pompéia et al.(2008)(82)(red open circles), confirming the low [Ca/Fe] and [Ti/Fe](see Fig.4.4). They found sub-solar values for these α-elements: compared to halo abundances, LMC stars are deficient by a factor of 3. In this result is also seen a decreasing trend of Ti abundances for higher metallicities. But the data are disperse and no solid conclusions can be drawn at this time. In Fig.4.5 is plotted the average value for alpha elements (Ca and Ti). We add Nissen and Schuster (69) results for comparison. They found a low-α population in the Galactic halo which is slightly enhanced with respect to the abundances in this work, confirming what Pompéia et al.(2008)(82) found for the LMC field population. This low-α population found by Nissen and Schuster (69) may play an important role in the possible merging history of our Galaxy, because they indicate that they evolved separately from the main halo population and were probably captured by the halo. Low [α/Fe] ratios suggest that SNe Ia has contributed more to the interstellar medium content in the past than SNe II (Pompéia et al.(2008)(82)).

• iron-peak element abundances: This work results are very low for some stars and a large dispersion in the abundances of Cr and Ni is observed. Again, the errors need to be calculated and the results will be reanalyzed. Little can be said about these results, but in general they are in agreement with Pompéia et al.(2008)(82) except for a group of stars with very low Cr and Ni, more distinct in Cr.

Tesis de Doctorado 4.3. FIELD STARS 53

Table 4.3 [X/Fe] values for ﬁeld stars. ID [Fe/H] [Ca/Fe] [Ti/Fe] [Cr/Fe] [Ni/Fe] BK41 −0.59 −0.42 ... −0.90 −0.75 BK58 −0.74 0.05 ... −0.70 −0.76 BK102 −0.65 −0.29 ...... BK93 −0.79 −0.02 −0.20 ... −0.28 BK80 −0.66 0.23 0.03 −0.73 ... BK14 −1.05 0.22 ... −0.21 −0.32 BK97 −0.95 −0.09 ... −0.25 −0.40 BK50 −0.97 −0.12 ...... BK109 −0.71 −0.33 ... −0.46 −0.30 BK65 −0.83 0.02 ... −0.14 −0.21 BK87 −0.52 0.21 ...... TARG.22 −0.88 0.21 0.63 −0.55 −0.27 BK55 −0.37 −0.10 −0.22 −0.44 −0.47 BK37 −0.23 −0.33 ...... −0.55 BK44 −1.11 0.22 ... −0.13 −0.35 BK19 −0.86 −0.37 ... −0.87 −0.76 BK9 −0.68 −0.73 0.11 ...... BK113 −0.79 0.41 ... −0.55 −0.07 BK105 −0.95 0.08 −0.26 −0.09 −0.25 BK60 −0.62 −0.39 ... −0.92 −0.57 BK95 −0.78 0.00 ... −0.26 −0.60 BK98 −0.4 −0.39 ... 0.45 ... BK31 −0.86 −0.08 0.50 −0.32 −0.65 BK2 −0.86 0.42 0.43 −0.65 −0.79 BK82 −0.99 −0.43 ... −0.47 −0.29 BK71 −1.24 0.64 ...... 0.30 BK81 −0.97 −0.01 0.04 −0.25 −0.97 BK108 −0.55 −0.02 ... 0.01 0.24 BK99 −0.49 −0.05 −0.34 −0.81 −0.45 BK8 −0.61 0.34 0.27 −0.88 −0.48

❄❇❅❆❈❋ R. Mateluna P. - 2012 54 CHAPTER 4. SL869 AND THE SURROUNDING FIELD OF H11

Figure 4.4 Abundances of Ca and Ti compared to Fe for LMC ﬁeld stars. Magenta triangles: our data, in red values for LMC from Pompéia et al.(2008)(82) and Mucciarelli et al.(2008)(65), in blue dSph data from Shetrone et al. (2001)(91), Sbordone et al. (2007)(89) and Monaco et al.(2005)(64) and in black data from the Galaxy from Fulbright(2000)(26) and Lee et al.(2002)(55).

Tesis de Doctorado 4.3. FIELD STARS 55

Figure 4.5 a bundance ratio vs. [Fe/H].]Mean [α/Fe] abundance ratio vs. [Fe/H]. Same symbols and colors as in Fig. 3.6 with the addition of Nissen and Schuster(2010)(69) high- and low-α halo stars (black open circle).

❄❇❅❆❈❋ R. Mateluna P. - 2012 56 CHAPTER 4. SL869 AND THE SURROUNDING FIELD OF H11

Figure 4.6 Abundances of Ni and Cr compared to Fe for LMC ﬁeld stars in magenta triangles. In red values for LMC from Pompéia et al.(2008)Pompéia et al. (82) and Mucciarelli et al.(2008)Mucciarelli et al. (65), in blue dSph data from Shetrone et al. (2001)(91), Sbordone et al. (2007)(89) and Monaco et al.(2005)(64) and in black data from the Galaxy from Ful- bright(2000)Fulbright (26) and Lee et al.(2002)(55).

Tesis de Doctorado Chapter 5

Washington Photometry of the LMC and V, I photometry method tests.

5.1 Introduction

The Washington photometric system was developed by Canterna (1976)(13) with the purpose of obtaining accurate temperatures, metal abundances and a CN strength index for G and K giants. Today, this system is mainly applied to derive metallicities and ages of SCs, with very good results (e.g. Geisler et al.(1997)(28), Geisler & Sarajedini(1999)(29), Geisler et al.(2003)(30), Piatti et al. (1999)(74), Piatti et al. (2009)(76),Piatti (2011)(? ) and Piatti et al.(2012)(79)). The derivation of metallicity can be done using the standard giant branches (SGB) method developed in Geisler & Sarajedini(1999)(29) (see Fig.5.1) and the age determined from the δT1 index(Geisler et al.(1997)(28)). In addition, both of these quantities can be derived form ischrone fitting given sufficiently deep and well calibrated data. The Washington system is a very efficient tool for determining these two important parameters for the study of galaxy evolution as it is a broad-band system, with FWHM typically of 1000Å. That is why this photometric system is used in this work, to help in the derivation of ages and metallicities for clusters and field stars in the LMC, and with the advantage of 4m Mosaic data, we can go deep enough to determine the age from MS fitting. In this chapter we present results from Washington photometry of Hodge 11 and LMC field stars. In addition, we briefly describe PSF photometry and its application to a set of data, as part of testing an automatization method (SkZ pipeline) for photometry. Finally, we present a summary of the results of Washington Photometry from MOSAIC (4mCTIO) data obtained during the period of this thesis (see Table 2.1) and published in Piatti et al.(2012)(79).

5.2 Photometry of CTIO MOSAIC data: Hodge 11

After pre-reducing the MOSAIC data acquired during the thesis with the purpose of studying LMC star clusters and the ﬁeld population (see chap.2), the DAOPHOT package was applied to the images and a radial cut of 500pixels to obtain the H11 CMD (see Fig. 5.2). CHAPTER 5. WASHINGTON PHOTOMETRY OF THE LMC AND V, I PHOTOMETRY 58 METHOD TESTS.

Figure 5.1 Washington standard giant branches(SGB) in the [MT 1 − (C − T1)0] plane from Geisler & Sarajedini(1999)(29) (paper Fig.4). Each curve represents the ﬁducial distribution of stars along the giant branch in 12 globular or old open clusters with a range of well-known metallicity. The mean color of the RGB is a strong function of metallicity, ranging from very red for the most metal-rich cluster (NGC 6791) to very blue for the most metal-poor cluster (NGC 7078).

In Table 5.1 the values of Washington magnitudes (non-calibrated) for H11 targets and their metallicity from HRS from this thesis. H11 spectroscopic targets were identiﬁed in Piatti et al.(2012)’s(79) calibrated photometry, which was kindly supplied by Andres Piatti. Interestingly, one of the metal rich stars from HRS, TARG.10, was found to lie redward of the bulk of the RGB (see Fig.5.3 left panel) . The standard giant branches (SGB) (Geisler & Sarajedini(1999)(29)) method was applied to the calibrated photometry of H11 (see Fig.5.4) to determine metallicity. For H11 we used three metal-poor comparison clusters: NGC7078 (-2.15 in metallicity on the Zinn and West scale), NGC6397(-1.91) and NGC6752 (-1.54). We assumed the same reddening and distance as above. Based on the method described in Geisler & Sarajedini(1999)(29), a metallicity for H11 of −2.03 ± 0.17 was obtained. This is in excellent agreement with the mean value obtaining from our HRS study. The possibility that TARG. 10 is indeed more metal-rich than the bulk of the stars bears further investigation.

5.3 Washington Photometry of the LMC Field

Piatti et al.(2012) (79) present, for the ﬁrst time, CCD Washington CT1T2 photometry of some 5.5 million stars in twenty-one 36×36 ﬁelds, distributed throughout the entire LMC main body, covering a total area of 7.6 square degrees. The data was observed as part of

Tesis de Doctorado 5.3. WASHINGTON PHOTOMETRY OF THE LMC FIELD 59

Table 5.1 Washington non-calibrated magnitudes of H11 target stars ID RA DEC r(t1) i(t2) c c-t1 [Fe/H]a TARG.11 06:14:31.460 −69 : 49 : 35.29 10.96 11.06 11.53 0.57 −2.02 TARG.9 06:14:21.700 −69 : 49 : 56.40 10.37 10.47 11.08 0.71 −2.03 TARG.8 06:14:24.650 −69 : 50 : 15.40 11.54 11.78 11.90 0.36 −2.06 TARG.16 06:14:23.340 −69 : 52 : 38.60 11.54 11.78 11.69 0.15 −2.08 TARG.2 06:14:22.423 −69 : 51 : 17.63 11.18 11.39 11.83 0.65 −1.81 TARG.10 06:14:30.359 −69 : 49 : 49.98 11.75 11.89 12.19 0.44 −1.86

aData obtained in this work spectroscopically, see chap.3

this thesis (see Chap.2) using the MOSAIC camera mounted on the 4m-Blanco telescope at CTIO. Each of the fields was subdivided into 16 smaller fields and each subfield is treated independently(336 fields in total), with the purpose of finding the dominant age and metallicity in each subfield. This data has been analyzed in order to improve our knowledge of the structure, extent, star formation history (SFH), and age-metallicity relationship (AMR) of our galactic neighbor. Extensive artificial star tests over the whole mosaic image data set and the observed behavior of the photometric errors with magnitude demonstrate the accuracy of the morphology and clearly delineate the position of the main features in the CMDs.

This work aimed at presenting new Washington CT1T2 photometry of the LMC main body which goes much deeper than the Magellanic Cloud Photometric Survey(MCPS - (41)) and covers ≈ 1.7 times the currently available VISTA near-infrared Y JKs survey of the Magellanic system (VMC) survey area (87). Saha et al.(2010)(88) used the same telescope, instrument setup and filters as for the present data set, but they explored the very outer region of the LMC, so that its main body was not surveyed. Finally, Rubele et al.(2012)’s (87) results are based on a photometric dataset whose limiting Ks mag for a 100% completeness level barely reaches two magnitudes below the Red Clump (RC). In our case, the T1 mag for 100% completeness level reaches between ∼ 3.5 and 4.5 mags below the RC. Other advantages of the present dataset are that the Washington CT 1 system SGB technique is found to have 3 times the metallicity sensitivity of the analogous VI technique (Geisler & Sarajedini(1999)(29)). Thus, for a given photometric accuracy, metallicities can be determined 3 times more precisely with the Wash- ington technique. In addition, the ability of the Washington system to estimate ages of star clusters has long been proven (? )-and references therein. From the δ(T1) index, calculated by determining the difference in the T1 magnitude of the RC and the main sequence turnoff (MSTO) (28), ages older than ∼ 1 Gyr can be estimated with typical errors of 10% (Piatti(2012)(78)). This yields a unique and powerful tool in which ages and metallicities for both clusters and field stars are determined homogeneously.

Finally, in this work we set out to obtain a large, deep and homogeneous database of LMC ﬁeld and cluster star photometry in order to investigate the SFH, AMR, metallicity distribution, metallicity gradient, constrain times of starbursts, etc. Here I summarize the main results published in Piatti et al.(2012)(79). In the future we expect to analyze the clusters.

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 5. WASHINGTON PHOTOMETRY OF THE LMC AND V, I PHOTOMETRY 60 METHOD TESTS. 5.3.1 Main Results 1. Global CMD properties: After extensive artiﬁcial star tests over the whole mosaic image data set, it was shown that the 50% completeness level is reached at C ∼ 23.5-25.0 and T1 ∼ 23.0-24.5, depending on the crowding and exposure time (see ﬁg.2 in Piatti et al. (2012)(79)), and that the behavior of the photometric errors with magnitude for the observed stars guarantees the accuracy of the morphology and position of the main features in the CMDs investigated.

T1(MSTO) magnitudes for the so-called representative stellar population of each field was determined, namely, the TO with the largest number of stars. The resultant representative T1(MSTO) mags are on average ∼ 0.5 mag brighter than the T1 mags for the faintest 100% completeness level of the respective field, so the TO of the representative population of each field was reached with negligible loss of stars. The prevailing TOs are typically ∼ 25%-50% more frequent than the following less dominant population, represented by a secondary peak - sometimes there also exists a third peak - in the differential luminosity functions. The RCs of the studied LMC fields were also investigated, assuming that the peak of

T1(RC) mag distribution corresponds to the most populous T1(MSTO) in the respective field. T1 histograms for these RC stars were built and Gaussian fits were performed to derive the mean RC mag values and the FWHMs of the T1(RC) distributions (fig.4 in Piatti et al.(2012)(79)).

2. Representative LMC field ages and metallicities: δT1 indices, calculated by determining the difference in the T1 magnitude of the RC and the MSTO, were computed using the representative T1(MSTO) and T1(RC) magnitudes. From these values the ages of the prevailing population in the studied LMC field were estimated (see Table 5.1 from Piatti et al. (79)), using a well-proven δT1 index-age calibration. The dispersions associated with the mean values represent in general a satisfactory estimate of the age spread around the prevailing population ages, although a few individual subfields have slightly larger age spreads. These larger age spreads do not affect the subsequent results. Representative metallicities were also estimated following the standard SGB procedure of entering absolute

MT1 magnitudes and intrinsic (C − T1)o colors for each subﬁeld RGB into Fig.5.1 after applying the appropriate reddening and distance corrections. The measured metallicity values are presented in Table 5.2 from Piatti et al.(2012)(79).

3. The VS feature: Finally, we studied the so-called vertical structure (VS) phenomenon -a striking feature composed of stars that lie below the RC and extend from the lower blue end of the RC to ∼ 0.45 mag fainter- taking advantage of the present database. The VS phenomenon is not clearly seen in most of the studied ﬁelds, suggesting its occurrence is linked to some other condition(s) in addition to the appropriate age, metallicity, and the necessary red giant star density.

Tesis de Doctorado 5.3. WASHINGTON PHOTOMETRY OF THE LMC FIELD 61

Table 5.2. Estimated ages and dispersions (in Gyr) for the representative populations in LMC ﬁelds.

FieldA B C D E F G H I J K L M N O P

1 9.0 9.0 11.7 9.0 9.0 11.1 11.1 11.7 9.0 11.1 11.1 11.7 8.6 11.1 11.1 11.7 2.8 2.8 3.5 2.8 2.8 3.3 3.3 3.5 2.8 3.3 3.3 3.5 2.7 3.3 3.3 3.5 2 9.5 9.5 10.5 11.1 8.1 10.5 10.5 8.6 10.0 8.1 10.5 11.1 10.5 10.5 8.1 8.1 2.9 2.9 3.2 3.3 2.6 3.2 3.2 2.7 3.1 2.6 3.2 3.3 3.2 3.2 2.6 2.6 3 8.6 11.1 8.6 11.7 9.0 11.1 11.1 9.0 11.7 11.1 8.6 11.7 8.6 9.0 11.7 9.0 2.7 3.3 2.7 3.5 2.8 3.3 3.3 2.8 3.5 3.3 2.7 3.5 2.7 2.8 3.5 2.8 4 11.1 9.5 9.5 9.5 9.5 7.7 10.0 10.5 9.5 10.0 10.0 10.5 12.9 8.1 8.1 8.6 3.3 2.9 2.9 2.9 2.9 2.5 3.1 3.2 2.9 3.1 3.1 3.2 3.8 2.6 2.6 2.7 5 9.5 11.7 11.7 11.7 11.7 11.7 9.0 11.7 11.7 11.7 9.0 6.5 11.7 12.3 7.3 9.5 2.9 3.5 3.5 3.5 3.5 3.5 2.8 3.5 3.5 3.5 2.8 2.1 3.5 3.6 2.3 2.9 6 9.0 9.0 9.0 9.5 11.7 9.0 11.1 9.0 9.0 9.0 11.7 9.0 9.0 11.7 9.0 9.5 2.8 2.8 2.8 2.9 3.5 2.8 3.3 2.8 2.8 2.8 3.5 2.8 2.8 3.5 2.8 2.9 7 11.1 10.0 4.7 10.5 10.5 10.5 8.1 8.1 10.5 10.5 13.5 10.5 12.9 8.1 8.1 11.1 3.3 3.1 1.5 3.2 3.2 3.2 2.6 2.6 3.2 3.2 3.9 3.2 3.8 2.6 2.6 3.3 8 8.6 11.1 8.6 8.6 9.0 11.1 8.6 9.0 9.0 8.6 8.6 8.6 7.3 6.9 9.0 9.0 2.7 3.3 2.7 2.7 2.8 3.3 2.7 2.8 2.8 2.7 2.7 2.7 2.3 2.2 2.8 2.8 9 6.5 6.5 4.9 5.2 8.6 4.9 3.7 4.9 6.5 8.6 6.5 4.9 9.0 6.9 5.2 6.5 2.1 2.1 1.6 1.7 2.7 1.6 1.2 1.6 2.1 2.7 2.1 1.6 2.8 2.2 1.7 2.1 10 2.7 2.1 3.5 2.7 3.5 2.1 2.1 2.1 2.1 2.1 2.7 2.1 2.7 2.7 1.7 2.1 0.8 0.5 1.1 0.8 1.1 0.5 0.5 0.5 0.5 0.5 0.8 0.5 0.8 0.8 0.3 0.5 11 2.7 4.7 3.5 4.7 3.5 2.8 4.9 4.9 4.7 6.5 3.7 4.9 3.7 4.9 3.7 5.2 0.8 1.5 1.1 1.5 1.1 0.8 1.6 1.6 1.5 2.1 1.2 1.6 1.2 1.6 1.2 1.7 12 2.8 3.7 3.5 2.7 4.9 4.9 3.7 4.7 3.7 3.7 4.9 2.8 3.7 3.9 3.9 5.2 0.8 1.2 1.1 0.8 1.6 1.6 1.2 1.5 1.2 1.2 1.6 0.8 1.2 1.3 1.3 1.7 13 3.0 2.8 2.8 2.8 3.0 3.9 3.7 3.7 3.9 5.2 3.7 3.7 2.3 3.9 3.0 3.7 0.9 0.8 0.8 0.8 0.9 1.3 1.2 1.2 1.3 1.7 1.2 1.2 0.6 1.3 0.9 1.2 14 2.0 1.6 1.6 2.0 1.5 1.4 2.0 2.3 1.6 1.5 2.4 4.2 3.7 3.2 3.7 3.7 0.4 0.3 0.3 0.4 0.2 0.2 0.4 0.6 0.3 0.2 0.7 1.4 1.2 1.0 1.2 1.2 15 3.9 4.9 3.7 3.7 4.9 3.9 3.9 3.9 3.7 3.9 3.9 3.2 3.7 3.2 3.0 2.4 1.3 1.6 1.2 1.2 1.6 1.3 1.3 1.3 1.2 1.3 1.3 1.0 1.2 1.0 0.9 0.7 16 5.5 4.7 4.7 4.9 6.5 4.7 6.2 6.2 4.7 6.2 6.2 3.5 5.2 6.5 4.9 3.7 1.8 1.5 1.5 1.6 2.1 2.1 2.0 2.0 1.5 2.0 2.0 1.1 1.7 2.1 1.6 1.2 17 6.9 6.9 11.7 11.7 12.3 9.5 9.5 6.5 10.0 11.7 11.7 9.0 10.0 9.5 9.5 9.0 2.2 2.2 3.5 3.5 3.6 2.9 2.9 2.1 3.1 3.5 3.5 2.8 3.1 2.9 2.9 2.8 18 4.4 6.2 8.1 8.1 4.4 8.1 8.6 10.0 4.4 4.7 5.8 10.0 3.5 6.2 5.8 4.4 1.4 2.0 2.6 2.6 1.4 2.6 2.7 3.1 1.4 1.5 1.9 3.1 1.1 2.0 1.9 1.4 19 9.5 11.7 6.9 11.7 9.0 11.7 9.0 9.0 9.0 9.0 9.0 9.0 9.0 6.9 12.3 9.5 2.9 3.5 2.2 3.5 2.8 3.5 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.2 3.6 2.9 20 10.0 12.9 10.0 10.0 10.0 10.0 12.9 10.0 10.0 12.9 7.7 12.9 10.0 10.0 7.7 8.1 3.1 3.8 3.1 3.1 3.1 3.1 3.8 3.1 3.1 3.8 2.5 3.8 3.1 3.1 2.5 2.6 21 8.6 12.3 9.5 9.5 12.3 12.3 12.3 12.3 9.5 12.3 9.5 12.3 9.5 9.5 12.3 10.0 2.7 3.6 2.9 2.9 3.6 3.6 3.6 3.6 2.9 3.6 2.9 3.6 2.9 2.9 3.6 3.1

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 5. WASHINGTON PHOTOMETRY OF THE LMC AND V, I PHOTOMETRY 62 METHOD TESTS.

Table 5.3. Estimated metallicities and dispersions (in dex) for the representative populations in LMC ﬁelds.

FieldA B C D E F G H I J K L M N O P

1 -0.96 -0.86 -0.90 -0.76 -0.91 -1.00 -0.95 -1.00 -0.96 -0.95 -0.95 -0.95 -0.85 -0.95 -0.95 -0.90 0.31 0.31 0.20 0.31 0.31 0.20 0.20 0.20 0.31 0.20 0.20 0.20 0.28 0.20 0.20 0.20 2 -0.88 -0.88 -0.90 -0.90 -0.74 -0.90 -0.90 -0.85 -0.80 -0.84 -0.90 -0.85 — -0.90 -0.84 -0.89 0.35 0.35 0.20 0.20 0.26 0.20 0.20 0.28 0.20 0.26 0.20 0.20 0.20 0.20 0.26 0.26 3 -0.85 -0.95 -0.90 -0.90 -0.91 -0.95 -0.95 -0.86 -0.95 -0.95 -0.90 -0.90 -0.85 -0.86 -0.90 -0.86 0.28 0.20 0.28 0.20 0.31 0.20 0.20 0.31 0.20 0.20 0.28 0.20 0.28 0.31 0.20 0.31 4 -0.90 -0.88 -0.88 -0.93 -0.83 -0.93 -1.00 -1.00 -0.88 -1.00 -0.95 -0.95 -0.90 -0.94 -0.89 -0.90 0.20 0.35 0.35 0.35 0.35 0.25 0.20 0.20 0.35 0.20 0.20 0.20 0.20 0.26 0.26 0.28 5 -0.78 -0.90 -0.90 -0.90 -0.90 -0.90 -0.86 -0.90 -0.90 -0.90 -0.86 -0.86 -0.85 -0.85 -0.77 -0.88 0.35 0.20 0.20 0.20 0.20 0.20 0.31 0.20 0.20 0.20 0.31 0.25 0.20 0.20 0.24 0.35 6 -0.86 -0.91 -0.86 -0.93 -0.85 -0.96 -0.95 -0.91 -0.91 -0.96 -1.00 -0.91 -0.86 -1.00 -0.91 -1.08 0.31 0.31 0.31 0.35 0.20 0.31 0.20 0.31 0.31 0.31 0.20 0.31 0.31 0.20 0.31 0.35 7 -0.85 -0.95 -0.78 -0.95 -0.90 -0.95 -0.89 -0.89 -0.90 -0.95 -0.90 -1.00 -0.80 -0.89 -0.89 -1.00 0.20 0.20 0.31 0.20 0.20 0.20 0.26 0.26 0.20 0.20 0.20 0.20 0.20 0.26 0.26 0.20 8 -0.90 -1.00 -0.95 -1.00 -0.96 -1.10 -1.05 -1.06 -1.01 -1.10 -1.05 -1.05 -1.02 -1.06 -1.11 -1.26 0.28 0.20 0.28 0.28 0.31 0.20 0.28 0.31 0.31 0.28 0.28 0.28 0.24 0.24 0.31 0.31 9 -0.81 -0.81 -0.80 -0.86 -0.85 -0.85 -0.70 -0.80 -0.81 -0.95 -0.86 -0.80 -0.81 -0.81 -0.76 -0.91 0.25 0.25 0.30 0.29 0.28 0.30 0.33 0.30 0.25 0.28 0.25 0.30 0.31 0.24 0.29 0.25 10 -0.61 -0.66 -0.88 -0.71 -0.83 -0.61 -0.61 -0.66 -0.66 -0.61 -0.71 -0.61 -0.56 -0.66 -0.48 -0.56 0.33 0.31 0.34 0.33 0.34 0.34 0.31 0.31 0.31 0.31 0.33 0.31 0.33 0.33 0.28 0.31 11 -0.56 -0.78 -0.68 -0.78 -0.68 -0.58 -0.85 -0.85 -0.83 -0.91 -0.70 -0.80 -0.65 -0.80 -0.70 -0.81 0.33 0.31 0.34 0.31 0.34 0.34 0.30 0.30 0.31 0.25 0.33 0.30 0.33 0.30 0.33 0.29 12 -0.53 -0.70 -0.68 -0.51 -0.80 -0.85 -0.75 -0.78 -0.70 -0.70 -0.80 -0.53 -0.60 -0.72 -0.72 -0.86 0.34 0.33 0.34 0.33 0.30 0.30 0.33 0.31 0.33 0.33 0.30 0.34 0.33 0.33 0.33 0.29 13 -0.61 -0.58 -0.58 -0.48 -0.66 -0.77 -0.75 -0.70 -0.77 -0.91 -0.75 -0.70 -0.50 -0.72 -0.66 -0.70 0.34 0.34 0.34 0.34 0.34 0.33 0.33 0.33 0.33 0.29 0.33 0.33 0.32 0.33 0.34 0.33 14 -0.52 -0.40 -0.45 -0.52 -0.36 -0.40 -0.42 -0.45 -0.35 -0.31 -0.52 -0.69 -0.65 -0.58 -0.65 -0.65 0.30 0.27 0.27 0.30 0.26 0.25 0.30 0.32 0.27 0.26 0.33 0.32 0.33 0.34 0.33 0.33 15 -0.67 -0.80 -0.70 -0.65 -0.80 -0.77 -0.77 -0.77 -0.70 -0.77 -0.77 -0.73 -0.65 -0.68 -0.71 -0.67 0.33 0.30 0.33 0.33 0.30 0.33 0.33 0.33 0.33 0.33 0.33 0.34 0.33 0.34 0.34 0.33 16 -0.67 -0.73 -0.68 -0.75 -0.81 -0.73 -0.80 -0.80 -0.73 -0.75 -0.80 -0.63 -0.66 -0.76 -0.75 -0.60 0.28 0.31 0.31 0.30 0.25 0.31 0.26 0.26 0.31 0.26 0.26 0.34 0.29 0.25 0.30 0.33 17 -0.91 -0.91 -1.15 -1.05 -1.20 -1.18 -1.13 -0.96 -1.25 -1.15 -1.15 -1.06 -1.15 -1.13 -1.13 -1.11 0.24 0.24 0.20 0.20 0.20 0.35 0.35 0.25 0.20 0.20 0.20 0.31 0.20 0.35 0.35 0.31 18 -0.71 -0.70 -0.74 -0.74 -0.61 -0.74 -0.75 -0.80 -0.71 -0.63 -0.69 -0.80 -0.63 -0.70 -0.69 -0.61 0.31 0.26 0.26 0.26 0.31 0.26 0.28 0.20 0.31 0.31 0.27 0.20 0.34 0.26 0.27 0.31 19 -1.13 -1.10 -0.96 -0.95 -1.01 -1.15 -1.11 -1.11 -1.06 -1.11 -1.11 -1.11 -0.96 -1.06 -1.15 -1.28 0.35 0.20 0.24 0.20 0.31 0.20 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.24 0.20 0.35 20 -1.15 -1.30 -1.30 -1.10 -1.25 -1.30 -1.25 -1.20 -1.20 -1.20 -1.13 -1.15 -1.10 -1.20 -1.13 -1.24 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.20 0.25 0.20 0.20 0.20 0.25 0.26 21 -0.85 -0.95 -0.93 -0.98 -0.95 -1.00 -1.00 -1.00 -0.93 -1.05 -1.03 -1.00 -0.88 -1.03 -1.10 -1.15 0.28 0.20 0.35 0.35 0.20 0.20 0.20 0.20 0.35 0.20 0.35 0.20 0.35 0.35 0.20 0.20

Tesis de Doctorado 5.4. TESTING THE INITIAL "SKZ" PIPELINE AND OBTAINING PSF PHOTOMETRY63 5.4 Testing the initial "SkZ" pipeline and obtaining PSF Pho- tometry

PSF stands for Point Spread Function and this is the best or maybe the only method capable of producing scientifically valid results in the case of crowded fields like star clusters (46). This method consists in adjusting a mathematical function (most commonly Gaussian, Lorentzian or Moffat) to the flux of a star as measured by a digital detector like a CCD. IRAF can deal with this kind of photometry with daophot and in this part of the work Peter Stetson’s DAOPHOTII program (Stetson (1992)(99)) was decided to be used because this program was included inside an automatization performed by Francesco Mauro, which we tested as part of this work. The idea was to test the automatization of the photometry process based on DAOPHOTII and also learn about PSF photometry and how it is performed. This automatization was created by Francesco Mauro with the purpose to be applied in the future to big imaging surveys. The test was made for a set of images obtained at Las Campanas Observatory for the LMC globular cluster NGC1841. The results from this test allowed me to prepare the spectroscopic observing run at LCO for NGC1841, see Fig. 5.5. Before running the scripts from this automatization, it is necessary to create an INPUT file using some parameters from the images, like GAIN, READ NOISE, and HIGH VALUE (chip saturation level) and FWHM for each of the images, determined using IRAF: imexam, and information from the headers of the fits images including exposure time, filter and airmass. Finally one assigns the path of the image and the number of frames (if it is not MOSAIC data, just ’1’). With all these ingredients the INPUT file is created and now the scripts are ready to run. Before running ANTE, it is important to have all the images in pixel type ’real’, not ’ushort’. The different scripts for the process are:

1. ANTE: Access to DAOPHOT and runs FIND, PHOTOMETRY, PICK, PSF and ALL- STAR. All of these tasks within DAOPHOT are run automatically and non-interactively.

2. Check.cl: CL script written by Peter Stetson. Looks at each PSF star selected by DAOPHOT in order to ’check’ the quality of the subtraction and the shape of the PSF star candidate to decide if it is a good candidate.

3. INTRA: Reﬁnes the PSF function, can be run before Check.cl

4. INTRAFALS: Runs DAOPHOT: FIND to ﬁnd stars not subtracted from the subtracted image due to crowding.

5. POST: Performs the ﬁnal PSF solution reﬁnement.

After the pipeline is run, one continues with the usual photometry steps: DAOMATCH (gives a rough solution for the positional transformation from image to image given a best image based on fwhm), DAOMASTER (gets a better positional transformation and creates a master star list), ALLFRAME(performs better photometry) and MONTAGE (creates a montage image to check if the stars were subtracted correctly).

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 5. WASHINGTON PHOTOMETRY OF THE LMC AND V, I PHOTOMETRY 64 METHOD TESTS. All of these procedures were run several times to ﬁnally obtain the cluster CMD, see ﬁg.5.5. Since then, the SkZ pipeline has been changed several times and is far from what it was when originally tested in this work. Currently it is being applied to VVV data (Mauro et al.2012, in preparation). The results from the photometry were used to identify targets in the LMC star cluster NGC1841, based on the work of G06 and their spectroscopic targets (marked in red in Fig.5.5). This analysis was used in the preparation of a MASK to be used in IMACS+MOE@LCO during the acquisition of spectroscopic data of NGC1841, details in chap.2.

Tesis de Doctorado 5.4. TESTING THE INITIAL "SKZ" PIPELINE AND OBTAINING PSF PHOTOMETRY65

16 10

18 12

20 14

22 16

24 -2 -1 0 1 2 -1 0 1 2 3 c-t1 C-T1

Figure 5.2 Color-magnitude diagrams for Hodge11 from Washington photometry based on CTIO 4m MOSAIC data. In red H11 stars studied spectroscopically. Left panel is non-calibrated photometry from this work and the right panel is calibrated photometry from Piatti et al.(2012)(79), both from the same data set.

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 5. WASHINGTON PHOTOMETRY OF THE LMC AND V, I PHOTOMETRY 66 METHOD TESTS.

16 10

17 11

18 12

19 13 -2 -1 0 1 2 -1 0 1 2 3 c-t1 C-T1

Figure 5.3 Zoom in the CMD from Figs.5.2 in the area of the RGB.

Tesis de Doctorado 5.4. TESTING THE INITIAL "SKZ" PIPELINE AND OBTAINING PSF PHOTOMETRY67

Figure 5.4 Fit of SGB to H11 photometry. The most metal poor SGBs were selected in this case, from left to right: in green NGC7078([Fe/H] =-2.15), in blue NGC6397(-1.91) and in red NGC6752(-1.54).

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 5. WASHINGTON PHOTOMETRY OF THE LMC AND V, I PHOTOMETRY 68 METHOD TESTS.

Figure 5.5 Color-magnitude (instrumental not calibrated) diagram for NGC1841. In red, targets from G06, used for the preparation of a MASK to be used in IMACS+MOE(see chap.2). Data from Warsaw@LCO (courtesy of Matias Gómez).

Tesis de Doctorado Chapter 6

Preliminary Results on the Chemical Evolution of the LMC

6.1 Introduction

Despite much work on this subject over the past decade, using both star clusters and ﬁeld stars (Richtler et al 1989(85), Olszewski et al.(1991)(71), Geisler et al.(1997)(28), Bica et al.1998(7), Dirsch et al. 2000(24), Hill et al. 2000(? ), Harris & Zaritsky 2009(41), Piatti et al. 2009(74), Carrera et al. 2011(15), among others), a number of very interesting questions remain to be fully addressed and understood, including (from Piatti & Geilser (2012)(80)):

• What caused the general lull in SFH between ∼ 3-12 Gyr ago?

• Are the cluster and ﬁeld AMRs really tightly coupled?

• Can the LMC AMR best be described by a closed-box, bursting or other chemical evolution model?

• What, if any, are the radial dependences of the chemical evolution?

One of the main purposes of this work is to study the chemical evolution of the LMC through the age-metallicity relation (AMR), as well as through an investigation of detailed abundances for a variety of elements with a range of nucleosynthetic histories. For this, we ﬁrst determined age and metallicity using both Washington photometry and HRS of ﬁeld stars and several star clusters in the LMC. In this chapter we show our results in the AMR together with the models to see how these can be interpreted. We start with a brief historical review of the subject.

6.2 Previous work on the AMR

The most recent works on the AMR of the LMC, using both photometry and spectroscopy of field stars, are (Carrera et al. (2011) (15)), (Harris & Zaritsky 2009(41)) and (Piatti & Geilser (2012)(80)). The first one used CaT to determine metallicities of LMC field stars (red giants) and obtained BVRI photometry of 10 fields (34’x33’) to derive star formation histories. From their CHAPTER 6. PRELIMINARY RESULTS ON THE CHEMICAL EVOLUTION OF THE 70 LMC results the most important conclusions are: (i)The disk field star AMR ∼ AMR of SCs and is well reproduced by a closed-box model or models with a small degree of outflow, (ii)The lack of clusters between 3-10 Gyr is not observed in the field population, (iii)The rapid chemical enrichment observed in the last few Gyrs is only observed in fields with R 7 kpc and (iv)much better evidence is found for an outside-in than inside-out formation scenario, in contradiction to generic ΛCDM models. The second paper is the first- ever global, spatially resolved reconstruction of the SFH of the LMC. They found that there existed a long relatively quiescent epoch (∼12-5 Gyr) during which the star formation was suppressed throughout the LMC; the metallicity also remained stagnant during this period. They concluded that the field and cluster star formation modes have been tightly coupled throughout the LMC’s history, in contradiction to Carrera et al 2011 results where no age gap was observed in the field population, and that the LMC AMR is better described by a ’bursting’ enrichment model (Pagel & Tautvaišien˙e(1998)(73)). The last paper is the work from Piatti et al.(2012)(80) using the results for field stars of ages and metallicities from Piatti et al.(2012)(79). They concluded from the composite AMR (from the 21 observed fields), showed in Fig.6.1, that the LMC has not chemically evolved as a closed-box nor in complete agreement with the bursting model, exclusively, but more as a combination of both models, and that the cluster age gap is not observed in the field population. On the other hand, are the star clusters, where the pioneering work using HRS is Hill et al. 2000. Other work includes Richtler et al 1989. Moreover, using photometry of star clusters are Geisler et al 1997, 2003, Dirsch et al 2000 and Piatti et al. 2009, 2012. In the work of Hill et al. 2000, they derived an AMR combining their HRS results with ages from Geisler et al.(1997)(28) and models from Pagel & Tautvaišien˙e(1998)(73) and concluded that the LMC is well defined by a bursting model (see Fig.6.2), although their sample is very small - only 4 SCs.

6.3 LMC Chemical Evolution Models

From the above section, it is clear that the LMC has suffered from some kind of memory loss and forgot how to form star clusters during a very long period, the famous age gap (3-12 Gyrs). Also, according to observations(Harris & Zaritsky 2009(41), Rubele et al. 2012(87), Piatti et al.(2012)(79)), field stars formed at a lower rate during that period, but suddenly about 3 Gyrs ago the LMC increased its star and cluster formation rates dramatically, subsequently showing more than one episode of a peak in the SFR. These interesting events led me to become curious about the evolution and the history of this galaxy and in a more specific way its chemical evolution. The most widely used models for describing the chemical enrichment of the LMC are Pagel & Tautvaisiene (1998)(73)(hereafter PT98), which assumes 2 models: a continuous SFR and a bursting SFR, and the closed-box model used by Geha et al 1998, among others. (73) assumed the LMC to have been built up by gradual infall of unprocessed material, this helps to alleviate the ’G-dwarf’ problem (deficit of metal-poor stars in the solar neighborhood relative to the one-zone model predictions of Galactic chemical evolution). They assumed linear laws of star formation and investigate both smooth and bursting models. Also, they assumed stellar yields and time delays identical to those which apply to the solar neighbourhood and appeal to galactic winds to

Tesis de Doctorado 6.3. LMC CHEMICAL EVOLUTION MODELS 71

Figure 6.1 From Piatti et al.(2012)(80), composite AMR for the 21 studied LMC fields as compared with different field AMRs: Harris Zaritski 2009 (yellow line), Rubele et al. (2011) (black line), Pagel Tautvaisiene (1998, hereafter PT98)) (blue line), and Geha et al. (1998) (red line). The red line AMR is based on a closed-box model, while the blue line relies on a bursting model. We also included with red and blue filled circles the AMRs derived by Carrera et al. 2011 for the LMC bar and disk, respectively. explain the low metallicities of the LMC in relation to their current gas fractions. However, they ignore selective winds and assume just a non-selective wind proportional to the SFR. In Fig.6.3 is shown the SFR history for both models. The full curve corresponds to the bursting model and the dashed-line curve corresponds to the smooth one. They assumed a quasi-linear star formation law, with an inverse time-scale for star formation(ω) constant for the ’smooth’ model. For the ’burting’ model, this quantity (ω) is assumed constant over certain periods, between which it changes discontinuously. According to this assumptions, two burst occur at 12Gyr and 3 Gyr, as it is shown in Fig.6.3 and are very drastic.

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 6. PRELIMINARY RESULTS ON THE CHEMICAL EVOLUTION OF THE 72 LMC

Figure 6.2 From Hill et al.(2000)(42): AMR for the star clusters in the LMC including Geisler et al.(1997)(28) and models from Pagel & Tautvaišien˙e(1998)(73), continuous (solid line) and bursting (dashed line) star formation rate (SFR).

Figure 6.3 SFR history for the LMC according to Pagel Tautvaisiene (1998)(73) model. The full curve describes the bursting model and the dashed-line curve shows the smoth model.

Tesis de Doctorado 6.4. OUR RESULTS ON THE AGE METALLICITY RELATION 73

In the future it is expected to get involved in running the models to describe chemical evolution of the LMC ( see chap.7). Recently, Bekki et al 2012(5) proposed a new model, that incorporates the delay time distribution (DTD) of type Ia supernova (SNe Ia). This means that the progenitor star can explode as early as 108 yr after their formation. They also incorporate the metallicity dependent chemical yields of AGB stars, in order to investigate the chemical evolution of s-process elements and to compare them with the observations. Their results were: (i)The present gas mass fraction and stellar metallicity as well as the higher [Ba/Fe] in metal- poor stars at [Fe/H]−1.5 can be more self-consistently explained by models with steeper initial mass functions, (ii)the observed higher [Mg/Fe] (≥ 0.3) at [Fe/H]∼ −0.6 and higher [Ba/Fe] ( 0.5) at [Fe/H]∼ −0.3 can be due to signiﬁcantly enhanced star formation about 2 Gyr ago. (iii)the observed overall [Ca/Fe]Ð[Fe/H] relation and remarkably low [Ca/Fe] (−0.2) at [Fe/H] −0.6 are consistent with models with short-delay supernova Ia and with the more eﬃcient loss of Ca possibly caused by an explosion mechanism of type II supernovae. (iv)the metallicity distribution functions do not show double peaks in the models with a starburst about 2 Gyr ago, but they show characteristic double peaks in the models with double starbursts at ∼ 200 Myr and ∼ 2 Gyr ago. The observed apparent dip of [Fe/H] around ∼ 1.5 Gyr ago in the ageÐmetal- 9 licity relation can be reproduced by models in which a large amount (∼ 10 Msun) of metal-poor ([Fe/H]−1) gas can be accreted onto the LMC. For the purpose of this work and due to limited time, the results from this thesis will be compared with this new model in the future and in the next section we only compare our results to the closed box model and the models from Pagel & Tautvaisiene (1998) (73).

6.4 Our Results on the Age Metallicity Relation

In the past chapters, metallicities for 2 star clusters (H11 and SL869) were determined both spectroscopically and photometrically. In addition, the age of one of these star clusters was derived from isochrone fitting. In addition, from Washington photometry we determined mean ages and metallicities for 21 fields in the LMC, which included some 5.5 million stars. We also include data for a large number of clusters studied by Piatti et al. 2009, where star clusters with ages greater than 13.75 Gyr (latest estimate for the age of the Universe from Jarosik et al. (2011)(48)), are assumed to have an age value of 12.3 Gyr, similar to GCs in our Galaxy. All this data is put together in the AMR and compared with the Geha et al 1998 closed-box model and Pagel & Tautvaišien˙e(1998)(73) models (continuous and bursting SFR) in fig6.4. From the figure (6.4) we see that our 2 clusters fall near the two extremes of the plot. SL869 falls on top of the model proposed by Pagel & Tautvaišien˙e(1998)(73) with a bursting SFH. This study concluded, from their comparison with the observations at the time it was published, when almost no HRS data were available, that neither a steepened IMF nor selective galactic winds are required to explain the abundances in the LMC. From these comparisons, they also suggested that the relatively high ratio of SNIa to core-collapse(Type II) SN observed in the LMC (Barbuy et al. (1994)(? )) is related to their star formation history rather than the IMF. The bursting model fits very well the metallicities for the old clusters, thought the ages for this clusters are not very accurate, is clear that the old GCs show a very big spread in metallicity,

❄❇❅❆❈❋ R. Mateluna P. - 2012 CHAPTER 6. PRELIMINARY RESULTS ON THE CHEMICAL EVOLUTION OF THE 74 LMC

Figure 6.4 Composite AMR for the 21 studied LMC fields (Piatti et al.(2012)(79)) and the cluster results from this work (red dots). Also included as black triangles are data of LMC star clusters from different authors collected in Table 9 from Piatti et al. 2009 (see Table??). For comparison we show models by Pagel Tautvaisiene (1998) (73) (blue line: bursting and green line: smooth ) and Geha et al. (1998) (27) (black line: closed-box). which is well described by the model. From the abundance ratio plots (chap.3), it is seen that the α-elements in the LMC start to diminish around lower metallicities(∼ −2.0) than in halo stars(∼ −1.0), which could be associated to this high ratio of SNIa to SNII mentioned by Pagel & Tautvaišien˙e (1998)(73), because the first are the ones responsible for both the iron and α enrichment of the ISM. Also the low α-elements abundance suggest a different (slower) star formation history than in the Galactic halo, related to the enrichment by SNe type II. It appears that the LMC formed less typeII SNe progenitors or some how lost these SNeII yields, supporting the idea of galactic winds

Tesis de Doctorado 6.4. OUR RESULTS ON THE AGE METALLICITY RELATION 75 considered in the PT98(73) model. More can be added to the final discussion about the LMC AMR when the results of field stars from Piatti et al. (2012)(79) are taken into account. These results are well described by the PT98(73) bursting model, with the exception of the oldest field, where the errors are considerably larger in age. These results show a very smooth increase in metallicity over the age period ∼4-11 Gyr, during the cluster age gap, in excellent agreement with the PT98(73) bursting model. SL869 is also in perfect agreement with this model, and even younger clusters unequivocally show the increase in chemical enrichment predicted by the bursting model over the last few Gyrs. The closed-box model in comparison to the results (Fig.6.4) looks fine for a low limit of the AMR , but a closed-box model with no inflows or outflows could not be consistent with the guess that the observed low α-element abundances could be due to galactic winds, for example. The "smooth" model does not fit well to the results because does not describe the abrupt rise of metallicity at age of ∼ 3Gyr. It seems that the AMR in general is very well described by the bursting model proposed by PT98(73), taking into account both clusters and field stars of the LMC. But much more data, especially at low metallicities, are necessary in order to have a more clear idea of the formation of the LMC and therefore our own galaxy. Also, new models become important to describe the physics involved in the formation and evolution of galaxies.

❄❇❅❆❈❋ R. Mateluna P. - 2012

Chapter 7

Future Work

7.1 Data processing, abundance determination and further analysis

Despite the effort put into this thesis, there is still a large amount of data to be processed from all the observations performed during the period of this thesis. The data collected at VLT, of two more star clusters (NGC1718 and NGC2257), will be analyzed using the same procedure used in this thesis. One of the two clusters, NGC1718, is already in the process of being analyzed in collaboration with Alessio Mucciarelli (Bologna, Italy). In this star cluster peculiar abundances have been recently found from integrated light spectroscopy (Colucci et al.(2012)(21)), including high iron-peak abundances and very low α abundances. We will perform a detailed abundance analysis of several stars in NGC1718 observed with HRS and verify if these findings are real or not. For the case of SL869, we will measure, for the two members, many elemental abundances (α- elements, iron-peak elements and n-capture elements) using the same spectral synthesis method as employed for H11. The two clusters observed at LCO, NGC1841 and NGC1846, need a different procedure for the reduction process which will require some initial investment in testing different methods. We anticipate starting with these studies as soon as the thesis is finished. In the data set of every star cluster, field stars were alsoobserved in order to have a broad range of ages and metallicities. These stars will undergo the same procedure of spectral analysis described in this thesis. This data will increase the amount of field stars so far studied and help in the understanding of the AMR. Also, field stars surrounding NGC1718 could be potentially interesting objects, considering the recent findings of iron-rich and α-poor stars (Colucci et al.(2012)(21)), in this LMC star cluster. Another interesting analysis that can be done with HRS abundances is related to observational constraints to SNe yields, especially with light elements like oxygen and magnesium. As an example of this future analysis we show a figure from Gibson et al.(2006)(34), Fig.7.1 shows nucleosynthetic abundance ratio ([O/Mg]) patterns predicted by the solar-metallicity Type II supernovae models from three authors, plus the sub-solar [O/Mg] value in the Galactic bulge (shaded region) (See Gibson et al.(2006)(34) for references). From the Washington data we can analyze the LMC cluster photometry, constructing a CMD for each cluster, determine metallicity from SGB fitting and ages using same techniques as (79). 78 CHAPTER 7. FUTURE WORK

Figure 7.1 Figure from Gibson et al.(2006)(34). Nucleosynthetic abundance ratio (oxygen-to- magnesium: [O/Mg]) patterns predicted by the solar-metallicity Type II supernovae models from three authors, plus the sub-solar [O/Mg] value in the Galactic bulge, shaded region (See Gibson et al.(2006)(34) for references).

Tesis de Doctorado 7.1. DATA PROCESSING, ABUNDANCE DETERMINATION AND FURTHER ANALYSIS 79 There are a large number of clusters available from our MOSAIC data for this analysis. Using both sets of data from spectroscopy and photometry, it will be possible to perform an improved analysis related to both star clusters and ﬁeld stars, allowing to continue the study of the ’Chemical Evolution of the LMC’.

7.1.1 Chemical Evolution Models: LMC During the period of this thesis work, I had the opportunity to participate in the Vatican Observatory Summer School 2012 (VOSS10), whose subject was ’The Chemistry of the Universe’. The faculty members were: Susan Trammell1, Sofia Cora2, Sue Lederer3 and Brad Gibson4. There I learned the importance of chemistry in astronomy, that can be studied in many different objects, from planets, comets and stars to the big bang, galaxies and galaxy clusters. Different projects were developed during the curse of the month at VOSS10 supervised by each faculty member. The most challenging one was related to chemical evolution models, supervised by Brad Gibson. It was a project developed in groups of four students during the last weeks of the school. My group had to determine ’Metallicity Distribution Functions (MDF)5 and Abundance Ratio Distributions’ in a dwarf spiral and a massive spiral galaxy, using two Hydrodynamical Simulations: DG1(Governato et al 2010(Nature,463,203)) and Massive Spirals: g15784(Stinson et al 2010 (astro-ph/1004.0675)). Both simulations were generated using the SPH code ’GASOLINE’, described in Wadsley et al (2004,New Astron,9,137). The principal motivation to study the MDF is that it is a potentially powerful clue to the evolutionary history of stellar populations and a good constraint for Galactic chemical evolution models. Therefore, we tested these simulations taking into account the G-dwarf problem (deficit of metal-poor stars in the solar neighborhood relative to the one-zone model predictions of Galactic chemical evolution). In the future the idea is to work in developing a model for the LMC, in collaboration with Brad Gibson and use the results from this thesis as constraints for this model. Also, study in more detail the recently proposed model from Bekki et al 2012 , which incorporates very interesting features on the chemical evolution, using the last data available from both photometric surveys and HRS data.

1Univ. of North Carolina at Charlotte, US. 2 FCAG/UNLP Observatorio Astronomico, La Plata, Argentina. 3 California State University SB, NASA Johnson Space Center, US. 4Jeremiah Horrocks InstituteUniversity of Central Lancashire, UK. 5 MDF is basically a histogram of the distribution of metals in a galaxy or in star cluster

❄❇❅❆❈❋ R. Mateluna P. - 2012

Chapter 8

Conclusions

For the purpose of this Thesis work, star clusters and ﬁeld stars were analyzed using two very diﬀerent techniques: high resolution spectroscopy and photometry, in particular Washington photometry.

In the case of H11a metallicity of [Fe/H]= −2.00±0.04 and σobs = 0.11±0.03 was determined, confirming it as one of the most metal poor clusters in the LMC (Olszewski et al.(1991)(71), Walker (1993)(107),Johnson et al.(2006)(50)-J06). Grocholski et al.(2006)(39) found a higher value using the CaT technique. In this work has been found that [Ca/Fe] is significantly lower than Galactic halo calibrators, so that CaT might be expected to give a relatively low, not high, metallicity. J06 also obtained low values of Ca in their study. However, M10 found values comparable to the Galaxy in their sample of three other old LMC GCs. Clearly, more investigation is need to clarify the appropriateness of using Ca as a proxy for Fe. One of the most important results in this study is that from the mean [α/Fe] vs [Fe/H] plot (Fig. 3.7). We find that H11 lies in the range of the dSph trend and below the Galactic one. This result confirms J06 and opens the possibility that galaxies like the LMC, assumed to be building blocks of our galaxy (from ΛCDM hierarchical formation models), may not in fact satisfy the chemical requirements, even at the low metallicity represented by H11. In the iron-peak elements, abundance similarities to the dSph results are also seen, such as low Cr, Mn and Ni. Another interesting result from H11 abundances has to do with observational constraints on the masses of SNII progenitors from relative abundances, especially of the alpha elements. Has been found low values for Mg, Ca and Ti, as did Venn et al. (2012)(105) in Carina metal- poor stars. The yields of these elements (Ca and Mg in particular) depend on the progenitor SNe mass. Assuming that Woosley and Weaver(1995)(109) reflects reality to first order, a high [O/Mg] abundance could indicate preferential pollution from 15-25 M/M⊙ SNeII (at least for Solar metallicities, Gibson(1997)(33)). In addition, Eu is an indicator of the r-process (main source: 8-10M/M⊙ stars) and in these results the Eu abundance is high, indicating lower mass progenitors (Woosley and Weaver(1995)(109)). Finally, the result for [La/Eu] shows that these neutron capture elements were formed totally by the r-process, implying SNe-only pollution, without the influence of AGB stars.

A hint of a Na spread is suggested, for H11, by comparing the σobs value with the internal errors (see Apendix A, Table A.1). This spread is normal at the position of the targets in their 82 CHAPTER 8. CONCLUSIONS location in the Galactic GC Na:O trend. H11 presents a behavior similar to that of intermediate- age LMC clusters from Mucciarelli et al.(2008)(65) shown in Fig. 3.11, with the hint of a Na but no O spread. The difference in the O abundance between Mucciarelli et al.(2008)(65) sample and this work data resides in the fact that the environment where H11 was formed was α-enhanced. The data fall in the extreme high O, low Na end of the Na:O anti-correlation trend (see Fig. 3.11). H11 is only slightly less massive (Mackey & Gilmore(2003)(56)) than the three old LMC GCs found by M10 to follow the same Na:O anticorrelation as Galactic GCs, and more massive than the intermediate-age LMC clusters from Mucciarelli et al.(2008)(65) which show a hint of anticorrelation. Now, from the study of SL869, it was obtained an average iron abundance of <[FeI/H]>= −0.47, σint = 0.04. This result is in reasonable agreement with what G06 obtained for this cluster: −0.40 dex. According to the G06 findings, intermediate-age clusters in the LMC show a very tight metallicity distribution (with mean metallicity −0.48 dex) and from the results present in this work, SL869 falls perfectly into this category. An age estimation of 1.45 Gyr, σ = 0.2Gyr, in good agreement with the value obtained by Walker (1993)(107) of 1.5 Gyr, and in reasonable agreement with the Piatti (2011)(77) value of 1.70 ± 0.15Gyr. From the abundance analysis of field stars, little can be said about these results, the data shows a big scatter in comparison with Pompéia et al.(2008)(82) for all elements and metallicities. This could be due to either the preliminary nature of the abundances and /or the lack of a radial velocity cut, both of which will be improved later. But in general they are in agreement with Pompéia et al.(2008)(82). In the case of the Washington photometry, very interesting results were obtained for both star clusters and field stars. A metallicity for H11 of −2.03 ± 0.17 was obtained based on the method of SGB described in Geisler & Sarajedini(1999)(29). This value is in excellent agreement with the mean value obtained from our HRS study. One of the metal rich stars from HRS, TARG.10, was found to lie redward of the bulk of the RGB, however the possibility that TARG. 10 is indeed more metal-rich than the bulk of the stars bears further investigation. For the field stars, 21 fields divided in 16 subfields were studied, determining age and metallicity from δT1 and SGB method, respectively. These results were plotted in the AMR, plus the results from HRS of H11 and SL869, in comparison with chemical evolution models proposed for the LMC, closed-box model and (Pagel & Tautvaisiene (1998) (73)) (PT98) models (smooth and bursting SFR). This thesis results are well described by the (PT98(73)) bursting model, with the exception of the oldest field, where the errors are considerably larger in age. These results show a very smooth increase in metallicity over the age period ∼4-11 Gyr, during the cluster age gap, in excellent agreement with the (PT98(73)) bursting model. SL869 is also in good agreement with this model, and even younger clusters unequivocally show the increase in chemical enrichment predicted by the bursting model over the last few Gyrs. Much more data, especially at low metallicities, are necessary in order to have a more clear idea of the formation of the LMC and therefore our own galaxy. Also, new models become important to describe the physics involved in the formation and evolution of galaxies.

Tesis de Doctorado Appendix A

Error Analysis

A.0.2 Error Analysis for H11 abundance determinations The calculation of the effect of internal errors on the determination of the abundances was made varying the atmospheric parameters in the following way: ∆Teff = +50 K (based on the error in (V-I)), ∆ logg= +0.10 (using the variation of +50K in Teff and the error in magnitude V in the canonical equation), ∆vt = +0.05 (from the variation in log g) and ∆[m/H]= +0.1 (from our dispersion(σobs) in [Fe/H]), and recalculating the abundances for TARG.4, assumed to be representative of our sample. These values can be easily rescaled to different errors in each atmospheric parameter if necessary. The value of σS/N for one line is the mean of the rms of the Fe abundance measurements per star, which is 0.06 dex. To obtain the error in S/N for a given element and a given star, we divided this value by the square root of the number of lines used for that element. The resulting errors for each [X/Fe] 1 ratio due to uncertainties in each atmospheric parameter are listed in Table A.1. The value of σtot was given by:

2 2 2 2 2 σtot = σTeﬀ + σlog g + σvt + σ[m/H] + σS/N (A.1) q The total internal error for each element is also compared to the observed error (standard deviation of the sample) in Table A.1.

1 Where X corresponds to any chemical specie. 84 APPENDIX A. ERROR ANALYSIS

Table A.1 Errors in [X/Fe] for stellar parameters.

[X/Fe] ∆Teﬀ ∆ log g ∆ vt ∆[m/H] σS/N σtot σobs +50K +0.10 +0.05 +0.1 [Fe/H] 0.08 −0.03 −0.09 −0.05 0.03 0.14 0.11 [O/Fe] −0.09 0.015 0.00 0.01 0.06 0.11 0.05 [Na/Fe] (D) −0.03 −0.01 0.08 0.04 0.04 0.10 0.25 [Na/Fe] −0.04 0.02 0.09 0.01 0.06 0.12 0.17 [Mg/Fe] 0.055 −0.005 0.09 0.06 0.06 0.13 0.12 [Al/Fe] −0.035 0.02 0.09 0.04 0.06 0.12 ... [Si/Fe] −0.12 −0.04 −0.05 0.01 0.06 0.15 0.11 [Ca/Fe] −0.045 −0.01 −0.01 −0.01 0.03 0.06 0.10 [Sc/Fe] −0.08 0.04 0.06 −0.02 0.03 0.11 0.16 [TiI/Fe] −0.005 0.01 0.08 0.04 0.06 0.11 0.06 [Cr/Fe] 0.01 0.005 0.06 0.04 0.04 0.08 0.02 [Mn/Fe] 0.015 0.05 0.15 0.13 0.06 0.21 ... [Co/Fe] −0.135 0.07 0.07 0.16 0.03 0.23 ... [Ni/Fe] −0.02 0.00 0.08 0.05 0.04 0.10 0.13 [Zn/Fe] −0.075 0.02 −0.02 −0.04 0.06 0.11 0.18 [Y/Fe] −0.06 0.06 0.07 0.04 0.06 0.13 0.14 [Ba/Fe] −0.025 0.065 0.05 0.04 0.06 0.11 0.23 [La/Fe] 0.020 0.05 0.10 0.10 0.06 0.16 ... [Eu/Fe] −0.085 0.055 0.04 0.02 0.06 0.13 0.05

Tesis de Doctorado Appendix B

MOOG: The LTE Stellar Line Analysis Program

B.1 What is MOOG?

MOOG is a FORTRAN code, developed by Chris Sneden (1973), that performs a variety of LTE line analysis and spectrum synthesis tasks. It is typically used to assist in the determination of chemical composition of a star. MOOG use the basic equations of LTE stellar line analysis, in particular using the formulation of F. N. Edmonds, Jr.(1969, JQSRT, 9, 1427).

B.2 How is it work?

I will not go into the details of coding, but lets say that use various subroutines that are called from driver routines and during the process of abundance determination gives automatic plots supported by sm(SuperMONGO), a graphic package chosen by the creator of MOOG because of its ease of use and ability to do color graphics, an example of the plotted output of MOOG is in fig.B.1. This code has certain internally-stored atomic and molecular data that have been culled from various literature sources. To run it, after installation, one simply writes MOOG on the computer console, and instantly the program will ask for the name of the "parameter file" in which you have specified which driver to use. In the case of this work, abfind and synth were used. Also, it is necessary to input a "model atmosphere file" and a "line list file". The "parameter file" is an important component of a MOOG run. This file tells MOOG which driver to use, how to process the data and how to output the results, examples of "parameter file" are found in fig. B.1 and B.2. Here is a description of the most used drivers in this work:

• abﬁnd: force-ﬁts abundances of species to yield computed equivalent widths that agree with observed ones previously measured with other software packages. Here is a sample graphical output from this mode. 86 APPENDIX B. MOOG: THE LTE STELLAR LINE ANALYSIS PROGRAM

• synth: computes a set of trial synthetic spectra and matches these to an observed spectrum. Abundances can be deduced either by visual inspection of the plot or by mathematical minimization of the observed-computed spectrum diﬀerence.

More information can be found in "MOOG: An LTE Stellar Line Analysis Program" 1

Figure B.1 Example of "parameter ﬁle: synth"

1 http //www.as.utexas.edu/˜chris/moog.html

Tesis de Doctorado Figure B.2 Example of "parameter ﬁle: abﬁnd" 88 APPENDIX B. MOOG: THE LTE STELLAR LINE ANALYSIS PROGRAM

Figure B.3 Example of abundance output from equivalent width matching (abﬁnd). These are Fe I abundances from individual lines plotted as functions of excitation potential (top panel), reduced equivalent width (middle panel), and wavelength (bottom panel). The dashed yellow lines represent the mean Fe I abundance, and the dashed blue lines represent (linear) trends of abundance with the three variables. The middle plot also contains information about the stellar model atmosphere used in this computation, and the bottom plot has information on the stellar equivalent widths. The vertical axis abundance units are logarithmic number densities on a standard scale in which log (H) = 12. The user can alter some of the computations (such as assumed microturbulent velocity) while the code is running.

Tesis de Doctorado B.2. HOW IS IT WORK? 89

Figure B.4 Example of synthetic spectrum computations and their comparison to an observed spectrum: This spectrum correspond to a CaI line. The colored lines represent the 5 synthetic spectrum computations, and the white dots represent the observed spectrum. The bottom panel shows the spectra plotted together, and the top panel shows the "o-c" comparisons of synthetic and observed spectra. The abundance units are logarithmic number densities on a standard scale in which log (H) = 12.

❄❇❅❆❈❋ R. Mateluna P. - 2012 90 APPENDIX B. MOOG: THE LTE STELLAR LINE ANALYSIS PROGRAM

Tesis de Doctorado Appendix C

Nucleosynthesis

First few minutes after the big-bang nucleosynthesis at a temperature of the order of 109 K created all the hydrogen, deuterium, some 3He, major part of 4He and some 7Li. (Primordial mass fraction X∼ 0.76, Y∼ 0.24, Z∼ 0.) From hydrogen and helium the ﬁrst stars (Population III) were formed. They were Very massive ( 100M sun ? ) ) and had short life time.

C.1 Nuclear Physics Concepts

Here are summarized basic concepts of nuclear physics, necessary to understand a part of this work. Basics Concepts

• Atomic nucleus: Z protons and N neutrons. Z is atomic number and A(=Z+N) is mass number.

• Nucleons or nuclides: protons and neutrons.

• Isotopes: group of nuclides conformed by the same element (Z=constant), varying N.

• Stability valley: occupied by stable nucleus.

• α -particle: helium nucleus , α-nuclei: 4He

• Iron-peak nuclide: those with 40 A 65: Sc, Ti, V,Cr,Mn,Fe,Co,Ni and Cu.

C.2 Nuclear Processes

• β-decay: emission of an electron or positron.

• α-decay: emission of an α -particle.

• P-P chain: Production of helium from hydrogen. Three diﬀerent branches. First step (decay of a proton into a neutron) involves the weak nuclear force. Characteristic in low mass stars (temperatures: ∼ 15x106 K ) 92 APPENDIX C. NUCLEOSYNTHESIS

• CNO-cycle: Production of helium from hydrogen. Production of helium using carbon, nitrogen and oxygen as catalysts, being consumed and regenerated in the process. Char- acteristic in high mass stars (temperatures: ∼ 4x107 K)

• Triple-α : synthesis of helium into carbon. First: two α-particles forms an unstable beryllium(8Be) nucleus, before 8Be decays, reacts with another α-particle forming carbon(12C). Is possible for carbon nuclei to capture α-particles producing oxygen and also 16O can capture α-particles to produce Neon. (typical temperature for triple- αprocess : T ∼ 108 K).

• s-process: slow neutron capture processes. Reactions are caused by low ﬂux of neutrons as in red giants during shell-burning phase (AGB). Form stable, neutron-rich isotopes of certain heavy elements. The s is for slow: Involves capture of a neutron suﬃciently slowly, to have time to decay before another neutron is captured.

• r-process: rapid neutron capture processes. Reactions are caused by dense ﬂux of neutrons during a supernova. Form unstable nuclei by absorption of neutrons. The r is for rapid: the formed unstable nuclei do not have time to decay before another neutron is absorbed because of the dense ﬂux.

All these processes occur in stars, depending principally on temperature and are responsible for the elements production.

Tesis de Doctorado Figure C.1 BigBang Nucleosynthesis. 94 APPENDIX C. NUCLEOSYNTHESIS

Figure C.2 The Õlocal galacticÕ abundance distribution of nuclear species, as a function of mass number A. The abundances are given relative to the Si abundance which is set to 106. Peaks due to the r- and s-process are indicated. Figure from Langer, N. (53)

Tesis de Doctorado C.2. NUCLEAR PROCESSES 95

Figure C.3 Chart of the nuclides, showing proton number Z vs neutron number N. Stable nuclei are in blue, and long-lived (>105 years) radioactive isotopes are in black. Other (lighter) colours show isotopes with shorter decay times. The arrows show the directions of some simple nuclear transformations.

❄❇❅❆❈❋ R. Mateluna P. - 2012 96 APPENDIX C. NUCLEOSYNTHESIS

Tesis de Doctorado Bibliography

[1] Alonso, A., Arribas, S. &Martínez-Roger, C. 1999, A&AS, 140, 261.

[2] Ballester P., Modigliani A., Boitquin O., Cristiani S., Hanuschik R., Kaufer A., & Wolf S. 2000, Msngr, 101, 31.

[3] [Ba94] Barbuy, B., de Freitas Pacheco, J.A.,Castro,S.,1994,AA,283,32.

[4] Bekki K., Couch W.J., Beasley M.A., Forbes D.A., Chiba M., Da Costa G.S. 2004, ApJ, 610, L93

[5] Bekki K., Tsujimoto T., Draft Version 2012

[6] Bica, E.L.D., Schmitt, H.R., Dutra, C.M. & Oliveira, H.L. 1999, AJ, 117,238-246.

[7] Bica E., Geisler D., Dottori H., Piatti A.E., Clariá J.J., Santos Jr. J.F.C. 1998, AJ, 116, 723

[8] Bica E., Bonatto C., Dutra C.M., Santos Jr. J.F.C., 2008, MNRAS, 389, 678

[9] Binney, J. and Merriﬁeld, M. 1998, Book: Galactic Astronomy. Princton series in astrophysics.

[10] Bonatto C., Bica E. 2010, MNRAS, 403, 996 [Bromm et al.(1999)]Br99 Bromm, Volker; Coppi, Paolo S.; Larson, Richard B. 1999, ApJ, 527, Issue 1, L5-L8.

[11] Bullock, J. S., Johnston, K. V. 2005, AJ, 635, 931-949.

[12] Burris, D., Pilachowski C. A., Armandroﬀ T.E., Sneden C., Cowan, J.J., Roe, H. 2000, ApJ, 544, 302.

[13] Canterna R. 1976, AJ, 81, 228

[14] Carrera R., Gallart C., Hardy E., Aparicio A., Zinn R., 2008, AJ, 135, 836

[15] Carrera R., Gallart C., Aparicio A., Hardy E., 2011, AJ, 142, 61

[16] Carretta, E., Bragaglia, A., Gratton, R.G., Lucatello, S., Catanzaro, G., Leone, F., Bellazzini, M., Claudi, R., D’Orazi, V., & Momany, Y. 2009, A&A, 505, 117. 98 BIBLIOGRAPHY

[17] R. Cayrel, E. Depagne, M. Spite, V. Hill, F. Spite, P. Francois, B. Plez, T. Beers, F. Primas, J. Andersen, B. Barbuy, P. Bonifacio, P. Molaro, and B. Nordstrm 2004, A&A 416, 1117.

[18] D.D. Clayton, Principles of Stellar Evolution and Nucleosynthesis, 1968, University of Chicago Press, ISBN 0 226 10953 4.

[19] Cohen J., 1981, ApJ, 258, 143

[20] Cole A., Tolstoy E., Gallagher J.S. III, Smecker-Hane T.A. 2005, AJ, 129, 1465

[21] Janet E. Colucci, Rebecca A. Bernstein, Scott A. Cameron, and Andrew McWilliam 2012,ApJ,746,29.

[22] Cowley, A. P.; Hartwick, F. D. A. 1982, ApJ, 259, 89.

[23] DaCosta, G. 1991, IAUS, 148, 183

[24] Dirsch B., Richtler T., Gieren W.P., and Hilker M., 2000, A&A,360,133.

[25] Eggen,O. J., Lynden-Bell, D., & Sandage, A. R. 1962, ApJ, 136, 748

[26] Fulbright, J. 2000, A.J.,120,1841.

[27] Geha M.C. et al. 1998, AJ, 115, 1045

[28] Geisler D., Bica E., Dottori H., Clariá J.J., Piatti A.E., Santos Jr., J.F.C. 1997, AJ, 114, 1920

[29] Geisler D., Sarajedini A. 1999, AJ, 117, 308

[30] Geisler D., Piatti A.E., Bica E., Clariá J.J. 2003, MNRAS, 341, 771

[31] Geisler, D., Smith, V.V., Wallerstein, G., Gonzalez, G., Charbonnel, C. 2005, AJ, 129, 1428.

[32] Geisler, D., Wallerstein, G., Smith, V. V., Casetti-Dinescu, D. I. 2007, PASP, 119, 939.

[33] Gibson, B. K. 1997, MNRAS, 290, 471.

[34] Gibson, B. K., MacDonald, A., Snchez-Blzquez, P., Carigi, L. 2006. ...

[35] Gieren, W., Storm, J., Barnes, T.G., III, Fouqu, P., Pietrzynski, G.; Kienzle, F. 2005, The ApJ, 627, 224.

[36] Gilmore G, Wyse R., 1991, ApJ, 367, L55.

[37] Girardi L. 1999, MNRAS, 308, 818

[38] Grevesse, N. & Sauval, A. J. 1998, SSRv, 85, 161.

Tesis de Doctorado BIBLIOGRAPHY 99

[39] Grocholski, A. J., Cole, A. A., Sarajedini, A., Geisler, D.,& Smith, V. V. 2006, AJ, 132, 1630.

[40] Hamuy, M., Folatelli, G., Morrell, N. I., et al. 2006, PASP, 118, 2.

[41] Harris J., Zaritsky D., 2009, AJ, 138, 1243

[42] Hill, V., Franois, P., Spite, M., Primas, F., Spite, F. 2000,A&A, 364, L19.

[43] Hodge, P.W. 1960, ApJ,131, 351.

[44] Holtzman J.A. et al. 1997, AJ, 113, 656

[45] Holtzman J.A. et al. 1999, AJ, 118, 2262

[46] Howell, Steve 2006, HANDBOOK OF CCD ASTRONOMY, 2nd ed., Cambridge Uni- versity Press.

[47] Ibata, R. A., Gilmore, G., & Irwin, M. J. 1994, Nature 370, 194.

[48] Jarosik et al. 2011, ApJS,192,14

[49] Johnson, J. A., Bolte, M., Stetson, P. B., Hesser, J.E., & Somerville, R.S. 1999, ApJ, 527, 199.

[50] Johnson, J. A., Ivans, I. I., & Stetson, P. B. 2006, ApJ, 640,801.

[51] Kerber, L. O.; Santiago, B. X.; Brocato, E.,2007,AA,462,139

[52] Kurucz, R.L. 1970, SAO, 309.

[53] Langer, N. 2012. Notes on an advanced astrophysics course on Nucleosynthesis at Bonn University.

[54] Landolt, A.U. 1992, AJ, 104, 340-371, 436-491.

[55] Lee, Jae-Woo; Carney, Bruce W. 2002, A.J., 124, 1511.

[56] Mackey, A.D., & Gilmore, G.F. 2003, MNRAS, 338, 85.

[57] Marino, A.F., Villanova, S., Piotto, G., Milone, A.P., Momany, Y., Bedin, L.R. & Medling, A.M. 2008, A&A, 490, 625.

[58] Mashonkina, L. I., Shimanski, V. V., Sakhibullin, N. A. 2000, Astronomy Reports, Volume 44, 790.

[59] Mateluna, R., Geisler, D., Villanova, S., Carraro, G., Grocholski, A., A. Sarajedini, A. Cole, V. Smith. 2012, A&A accepted.

[60] Matteucci, F. 2012, eprint arXiv 0804.1492

[61] McWilliam, A., Preston, G.W., Sneden, C., Shectman, S. 1995, AJ, 109,2736.

❄❇❅❆❈❋ R. Mateluna P. - 2012 100 BIBLIOGRAPHY

[62] Mighell K.J, Rich M., Shara M., Fall S.M., 1996, AJ, 111, 2314.

[63] Milone A.P., Bedin L.R., Piotto G., Anderson J. 2009, A&A, 497, 755

[64] Monaco, L., Bellazzini, M., Bonifacio, P., Ferraro, F. R., Marconi, G., Pancino, E., Sbordone, L., Zaggia, S. 2005, A&A, 441, 141.

[65] Mucciarelli, A., Carretta, E., Origlia, L., Ferraro, F. R. 2008, AJ, 136, 375.

[66] Mucciarelli, A., Origlia, L., Ferraro, F.R., Pancino, E. 2009, ApJ Letters, 695, L134.

[67] Mucciarelli, A., Origlia, L., Ferraro, F.R. 2010, ApJ, 717, 277.

[68] Navarro, J.F., Frenk, C.S, & White, S. D. M., 1997, ApJ, 490, 493.

[69] Nissen, P.E., & Schuster, W.J. 2010, A&A, 511, L10.

[70] Nissen, P.E., & Schuster, W.J. 2011, A&A, 530A, 15

[71] Olszewski, E. W., Schommer, R. A., Suntzeﬀ, N. B., Harris, H. C. 1991, AJ, 101, 515.

[72] Bernard Pagel, Book Review: Nucleosynthesis and chemical evolution of galaxies / Cambridge U Press, 1997 1998

[73] Pagel B.E.J., Tautvaišien˙eG. 1998, MNRAS, 299, 535 (PT98)

[74] Piatti A.E., Geisler D., Bica E., Clariá J.J., Santos Jr. J.F.C., Sarajedini A., Dottori H. 1999, AJ, 118, 2865 (P99)

[75] Piatti A.E., Sarajedini A., Geisler D., Bica E., Clariá J.J. 2002, MNRAS, 329, 556

[76] Piatti A.E., Geisler D., Sarajedini A., Gallart C. 2009, A&A, 501, 585

[77] Piatti A.E. 2011a, MNRAS Letters, 418, L40.

[78] Piatti A.E. 2012, MNRAS, 422, 1109

[79] Piatti, A., Geisler, D., Mateluna, R. 2012, AJ, 144, 4, article id. 100.

[80] Piatti, A. & Geisler, D. 2012, eprint arXiv:1208.3899, AJ accepted.

[81] Pompéia, L., Barbuy, B., Grenon, M., Gustafsson, B. 2007,IAUS, 241, 78P.

[82] Pompéia, L., Hill, V., Spite, M., Cole, A., Primas, F., Romaniello, M., Pasquini, L., Cioni, M.R., & Smecker Hane, T. 2008, A&A, 480, 379.

[83] Reddy, B. E., Tomkin, J., Lambert, D. L., Allende Prieto, C. 2003, MNRAS, 340, 304.

[84] Reddy, B. E., Lambert, D. L., Allende Prieto, C. 2006, MNRAS, 367, 1329.

[85] Richtler T., Spite M., Spite V., 1989, A&A, 225, 351.

Tesis de Doctorado BIBLIOGRAPHY 101

[86] Robertson, B., Bullock, J. S., Font, A. S., Johnston, K. V., Hernquist, L. 2005, ApJ, 632, 872.

[87] Rubele S., Kerber L., Girardi L., et al. 2012, A&A, 537, 106

[88] Saha A., Olszewski E.W., Brondel B., et al., 2010, AJ, 140, 1719

[89] Sbordone, L., Bonifacio, P., Buonanno, R., Marconi, G., Monaco, L., Zaggia, S. 2007, A&A, 465, 815.

[90] Searle, L., & Zinn, R. 1978, ApJ, 225, 357.

[91] Shetrone, M.D., Co˙té, P., Sargent, W. L. W. 2001, ApJ, 548, 592.

[92] Shetrone, M., Venn, K.A., Tolstoy, E., Primas, F., Hill, V., Kaufer, A. 2003, AJ,125,684.

[93] Smecker-Hane T.A., Cole A.A., Gallagher J.S. III, Stetson P.B. 2002, ApJ, 566,239

[94] Smith, V.V., Hinkle, K. H., Cunha, K., Plez, B., Lambert, D.L., Pilachowski, C.A., Barbuy, B., Melndez, J., Balachandran, S., Bessell, M.S., Geisler, D.P., Hesser, J.E., Winge, C. 2002, AJ, 124, 3241.

[95] Sneden, C. 1973, ApJ, 184, 839.

[96] Sneden, C., Kraft, R.P., Shetrone, M. D., Smith, G. H., Langer, G. E., Prosser, C. F. 1997. AJ, 114, 1964.

[97] Sneden, C., Cowan, J. J., Gallino, R. 2008, ARA&A, 46, 241S.

[98] Stetson, P. B. 1987, PASP, 99, 191.

[99] Stetson, P. B. 1992, Astronomical Data Analysis Software and Systems I, A.S.P. Con- ference Series, Vol. 25, Diana M. Worrall, Chris Biemesderfer, and Jeannette Barnes, eds., p. 297.

[100] Suntzeﬀ, N.B., Schommer, Robert A., Olszewski, E.W., Walker, A.R. 1992, AJ, 104, 1743-1764.

[101] Talbot, R. J., Jr., Arnett, W. D. 1971, AJ, vol. 170, p.409

[102] Umeda, H., & Nomoto, K. 2002, ApJ, 565, 385

[103] Van den Bergh,S. 1981b, Astron. Astrophys. Suppl., 46, 79.

[104] Van den Bergh,S. 2000, Book: The Galaxies of the Local Group

[105] Venn, K. A., Shetrone, M.D., Irwin, M. J., Hill, V., Jablonka, P., Tolstoy, E., Lemasle, B., Divell, M., Starkenburg, E., Letarte, B., Baldner, C., Battaglia, G., Helmi, A., Kaufer, A., Primas, F. 2012, ApJ, 751:102.

❄❇❅❆❈❋ R. Mateluna P. - 2012 102 BIBLIOGRAPHY

[106] Villanova, S., Carraro, G., & Saviane, I. 2009, A&A, 504, 845.

[107] Walker, A. 1993, AJ, 106, 999.

[108] Westerlund

[109] Woosley, S.E. & Weaver, T. A., 1995, ApJ S, 101, 181.

[110] Zinn, R., 1980, ApJS, 42, 19

Tesis de Doctorado List of Publications.

Articles.

• A Washington Photometric Survey of the Large Magellanic Cloud Field Star Population A. Piatti, D. Geisler, R. Mateluna, 2012. Astronomical Journal, Volume 144, page 100.

• Chemical Abundances in the Old LMC Globular Cluster Hodge 11 R. Mateluna, D. Geisler, S. Villanova, G. Carraro, A. Grocholski, A. Sarajedini, A. Cole, and V. Smith. Accepted August 13 of 2012 to be published in Astronomy and Atrophysics.

Proceedings.

• ’Detailed Abundances for Field Stars Surrounding the LMC Cluster Hodge 11’ Mateluna, Reneé; Geisler, Douglas; Villanova, Sandro. Stellar Populations Ð Planning for the Next Decade, Proceedings of the International As- tronomical Union, IAU Symposium, Volume 262, p. 385-387.(2009)

• ’Chemical abundances of the LMC cluster Hodge 11 and its surrounding ﬁeld’ Mateluna, Reneé; Geisler, Douglas; Villanova, Sandro. Star clusters: basic galactic building blocks throughout time and space, Proceedings of the International Astronomical Union, IAU Symposium, Volume 266, p. 474-476.(2009)

• ’A High Galactic Latitude Dust Template for CMB Polarization Studies’ Fraisse, Aurélien A.; Magalhães, A. M.; Schwarz, H. E.; Spergel, D. N.; Majewski, S. R.; Patterson, R. J.; Mateluna, R. C.; Semler, D. R.; Richards, J. W. American Astronomical Society, AAS Meeting 214, 413.03; Bulletin of the American As- tronomical Society, Vol. 41, p.680.(2009) 104 BIBLIOGRAPHY Meetings Presentations.

• Reunion Anual SOCHIAS 2012 10-12 Octubre 2012, Viña del Mar. Presentation oral: Chemical Abundances of the LMC Globular Cluster Hodge11

• Joint observatory meeting-ESO, 2010

• Reunion Anual SOCHIAS 2010 18-20 Enero 2010, Universidad de Concepcion, Concepcion. Poster: ’First Results from FLAMES VLT data for four LMC Clusters’.

• IAU General Assembly 2009, 3-14 Agosto 2009, Rio de Janeiro, Brasil. IAU Symposium 262: Stellar Populations Ð Planning for the Next Decade Poster:’Detailed Abundances for Field Stars Surrounding the LMC Cluster Hodge 11’

• IAU General Assembly 2009, 3-14 Agosto 2009, Rio de Janeiro, Brasil. IAU Symposium 266: Star clusters: basic galactic building blocks throughout time and space Poster: ’Chemical abundances of the LMC cluster Hodge 11 and its surrounding ﬁeld’

• Reunin Anual SOCHIAS 2009. 14-16 de Enero 2009, CEPAL, Stgo. Poster: "Chemical abundances of the LMC Cluster H11 and its surrounding ﬁeld".

• Reunion del Centro de Astrofsica y Tecnologas Aﬁnes (CATA) 17-18 de Junio 2009, Universidad de Chile, Stgo. Oral: "Chemical abundances of the LMC Cluster H11 and its surrounding ﬁeld".

Attended Schools, Workshops and Internship.

• ESO Chile Student from Nov 2009- Nov 2011.

• Vatican Observatory Summer School(VOSS) 2010, Castel Gandolfo, May 31-June 30. Italy

Tesis de Doctorado