Download evolución química de la nube grande de magallanes.

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

P-nuclei wikipedia, lookup

Planetary nebula wikipedia, lookup

Hayashi track wikipedia, lookup

Cosmic distance ladder wikipedia, lookup

Main sequence wikipedia, lookup

Stellar evolution wikipedia, lookup

Nucleosynthesis wikipedia, lookup

High-velocity cloud wikipedia, lookup

Star formation wikipedia, lookup

Astronomical spectroscopy wikipedia, lookup

Transcript
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 [email protected]: 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 ([email protected]) y fotometría de Washington, usando un instrumento de campo
amplio ([email protected]), 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 Astronomy 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 Collapse 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 ([email protected])
and Washington photometry, using a wide field imager([email protected]), 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 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Studying the Large Magellanic Cloud . . . . . . . . . . . .
1.2.1 Star Clusters in the LMC and the Age Gap . . . .
1.2.2 Using High Resolution Spectroscopy . . . . . . . .
1.2.3 Using Washington Photometry . . . . . . . . . . .
1.3 Chemical Evolution . . . . . . . . . . . . . . . . . . . . . .
1.3.1 Basic Concepts of Chemical Evolution of Galaxies
1.3.2 Enrichment models . . . . . . . . . . . . . . . . . .
1.4 The Aim of this Thesis . . . . . . . . . . . . . . . . . . . .
2 Method
2.1 Observations: Data Acquisition . . . . . . . . . . . . . .
2.1.1 ESO-VLT Paranal Observatory . . . . . . . . . .
2.1.2 Las Campanas Observatory . . . . . . . . . . . .
2.1.3 Cerro Tololo Inter American Observatory (CTIO)
2.2 Data Reduction . . . . . . . . . . . . . . . . . . . . . . .
2.2.1 Spectra . . . . . . . . . . . . . . . . . . . . . . .
2.2.2 Images . . . . . . . . . . . . . . . . . . . . . . . .
2.3 Imaging Techniques . . . . . . . . . . . . . . . . . . . . .
2.3.1 Photometry . . . . . . . . . . . . . . . . . . . . .
2.3.2 Astrometry: From X and Y to RA and DEC . .
2.4 Spectral Analysis . . . . . . . . . . . . . . . . . . . . . .
2.4.1 Radial Velocities . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
2
2
2
5
6
6
7
8
.
.
.
.
.
.
.
.
.
.
.
.
9
9
9
10
11
11
11
14
15
15
16
16
16
viii
CONTENTS
2.4.2
Abundance Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3 Hodge 11
3.1 Introduction . . . . . . . . . . . . .
3.2 Results from Photometry . . . . .
3.3 Radial Velocities for Hodge 11 . . .
3.4 Abundances Results . . . . . . . .
3.4.1 Fe-peak elements . . . . . .
3.4.2 α-elements . . . . . . . . .
3.4.3 Neutron Capture Elements
3.4.4 Na and O . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4 SL869 and the Surrounding Field of H11
4.1 Introduction . . . . . . . . . . . . . . . . .
4.2 Star Cluster SL869 . . . . . . . . . . . . .
4.2.1 Photometry . . . . . . . . . . . . .
4.2.2 Abundance Analysis . . . . . . . .
4.2.3 Age determination . . . . . . . . .
4.3 Field Stars . . . . . . . . . . . . . . . . . .
4.3.1 Abundance Analysis . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
5 Washington Photometry of the LMC and V, I
5.1 Introduction . . . . . . . . . . . . . . . . . . . .
5.2 Photometry of CTIO MOSAIC data: Hodge 11
5.3 Washington Photometry of the LMC Field . . .
5.3.1 Main Results . . . . . . . . . . . . . . .
5.4 Testing the initial "SkZ" pipeline and obtaining
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
photometry method tests.
. . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
PSF Photometry . . . . . . . .
6 Preliminary Results on the Chemical Evolution of the
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . .
6.2 Previous work on the AMR . . . . . . . . . . . . . . . .
6.3 LMC Chemical Evolution Models . . . . . . . . . . . . .
6.4 Our Results on the Age Metallicity Relation . . . . . . .
LMC
. . . .
. . . .
. . . .
. . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
.
.
.
.
.
.
.
.
27
27
28
29
29
29
35
37
40
.
.
.
.
.
.
.
45
45
46
46
47
50
50
50
.
.
.
.
.
57
57
57
58
60
63
.
.
.
.
69
69
69
70
73
7 Future Work
7.1 Data processing, abundance determination
and further analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.1.1 Chemical Evolution Models: LMC . . . . . . . . . . . . . . . . . . . . . .
77
8 Conclusions
81
A Error Analysis
A.0.2 Error Analysis for H11 abundance determinations . . . . . . . . . . . . . .
83
83
Tesis de Doctorado
77
79
CONTENTS
ix
B MOOG: The LTE Stellar Line Analysis Program
B.1 What is MOOG? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
B.2 How is it work? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
85
85
85
C Nucleosynthesis
C.1 Nuclear Physics Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C.2 Nuclear Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
91
91
91
❄❇❅❆❈❋
R. Mateluna P. - 2012
List of Figures
1.1
1.2
1.3
LMC Star Clusters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
LMC Star Clusters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Washington standard giant branches. . . . . . . . . . . . . . . . . . . . . . . . . .
3
4
5
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
Fibers Map. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
MOSAIC image example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Washington Photometric System. . . . . . . . . . . . . . . . . . . . . . . . . . . .
MOSAIC observed fields preparation. . . . . . . . . . . . . . . . . . . . . . . . . .
MOSAIC CCD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Detailed view of our best UVES spectrum, TARG.6, in the range of 6220 − 6270.
Map example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
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. . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
13
13
14
15
21
22
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
Hodge 11 image . . . . . . . . . . . . . . . . . . . . . .
Color-magnitude diagram for H11. . . . . . . . . . . .
Swope Map. . . . . . . . . . . . . . . . . . . . . . . . .
Velocity histogram for GIRAFFE and UVES data. . .
Abundances for iron peak elements. . . . . . . . . . . .
Abundances for α-elements compared to the literature.
Mean [α/Fe . . . . . . . . . . . . . . . . . . . . . . . .
Neutron-capture elements abundances. . . . . . . . . .
Ba to Y ratio vs. metallicity . . . . . . . . . . . . . . .
Ba to Eu ratio vs. metallicity. . . . . . . . . . . . . . .
Na and O abundances . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28
30
31
32
36
38
39
41
42
43
44
4.1
4.2
4.3
4.4
4.5
4.6
SL869 photometry calibration . . . . . . .
Color-magnitude diagram for SL869 . . .
Age estimation for SL869 . . . . . . . . .
Alpha element abundances for field stars.
Mean [α/Fe . . . . . . . . . . . . . . . . .
Iron-peak abundances for field stars. . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
47
48
51
54
55
56
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
26
xii
LIST OF FIGURES
5.1
5.2
5.3
5.4
5.5
Washington standard giant branches. . . .
Washington Color-magnitude diagrams for
Zoom in washington CMD. . . . . . . . .
Fit of SGB to H11 photometry. . . . . . .
Color-magnitude diagram for NGC1841. .
.
.
.
.
.
58
65
66
67
68
6.1
6.2
6.3
6.4
Composite AMR for the 21 studied LMC fields from Piatti et al.(2012). . .
AMR for the star clusters in the LMC from Hill et al.(2000). . . . . . . . .
SFR history for the LMC according to Pagel Tautvaisiene (1998). . . . . .
Composite AMR for the 21 studied LMC fields and the cluster results from
work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . .
. . .
. . .
this
. . .
71
72
72
74
7.1
Type II SNe models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
78
B.1
B.2
B.3
B.4
Example
Example
Example
Example
.
.
.
.
86
87
88
89
C.1 BigBang Nucleosynthesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C.2 Abundances in the Solar Neighborhood. . . . . . . . . . . . . . . . . . . . . . . .
C.3 Chart of nuclides. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
93
94
95
of
of
of
of
parameter file: synth. . .
parameter file: abfind. .
MOOG graphics: abfind
MOOG graphics: synth.
Tesis de Doctorado
.
.
.
.
.
.
.
.
.
.
.
.
. . . . . .
Hodge11.
. . . . . .
. . . . . .
. . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
List of Tables
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.8
2.8
Observing Time Proposals . . . . .
LMC target star clusters . . . . . .
Telescopes and Instruments set-up
Spectroscopic data set . . . . . . .
Imaging data set from LCO . . . .
LMC Cluster data set from CTIO .
LMC star fields from CTIO. . . . .
Line list used. . . . . . . . . . . . .
continued. . . . . . . . . . . . . . .
continued. . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
12
16
17
18
18
19
20
23
24
25
3.1
3.2
Important parameters for our target stars . . . . . . . . . . . . . . . . . . . . . .
[X/Fe] values for Hodge 11 stars. . . . . . . . . . . . . . . . . . . . . . . . . . . .
33
34
4.1
4.2
4.3
Important parameters for our target stars . . . . . . . . . . . . . . . . . . . . . .
Iron abundances for our target stars . . . . . . . . . . . . . . . . . . . . . . . . .
[X/Fe] values for field stars. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
49
50
53
5.1
5.2
Washington non-calibrated magnitudes of H11 target stars . . . . . . . . . . . . .
Estimated ages and dispersions (in Gyr) for the representative populations in LMC
fields. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Estimated metallicities and dispersions (in dex) for the representative populations
in LMC fields. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
59
61
A.1 Errors in [X/Fe] for stellar parameters. . . . . . . . . . . . . . . . . . . . . . . . .
84
5.3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
62
Chapter 1
Introduction
1.1
Motivation
Understanding how our Galaxy and other galaxies formed is one of the big questions in Astronomy 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 Accretion 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 concentrated 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 Grocholski 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 fundamental 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 isoabundance 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.
NF e
NF e
[F e/H] = log10
− log10
(1.1)
NH star
NH Sun
The metal enrichment of the ISM comes mainly from :
– Low and intermediate mass stars (0.8 ≤M/Msun ≤8.0) produce mainly 4 He,
plus some CNO isotopes and s-process (A > 90) elements.
12 C, 14 N
– 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: W R = −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:
P
i
. According to this definition it holds:
Xi = MMgas
i=1,n Xi = 1, where n represents the total
number of chemical species.
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 &
Tautvaišienė (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 Observatory (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
1
http://archive.eso.org/wdb/wdb/eso/sched_rep_arc/querytel=UT2&from_date=01Oct2008&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 Magellan 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 [email protected] 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 January 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
2.2.1
Data Reduction
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
3
http://www.lco.cl/telescopes-information/irenee-du-pont/instruments/website/direct-ccd-manuals/directccd-users-manual/ccd-manual-for-the-40-inch-100-inch-telescopes
4
http://www.eso.org/sci/software/pipelines/
5
IRAF 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
PI
S. Villanova
D. Geisler
R. Mateluna
D. Geisler
R.Mateluna
Observer
service-mode
Mateluna, Geisler
Mateluna
Mateluna, Geisler, Muñoz
Mateluna
Objects
H11, NGC2257, NGC1718
LMC star clusters
NGC1846, NGC1841
LMC and SMC star clusters
NGC1846
Instrument
FLAMES
MOSAIC
IMACS+MOE
Y4KCam
IMACS+MOE
Tesis de Doctorado
CHAPTER 2. METHOD
12
Table 2.1 Observing Time Proposals
ID Proposal
Semester-Year
082.B-0458
ESO82-B-2008
CNTAC-0912
2008-B
CNTAC09B_024
2009-B
CNTAC09B_010
2009-B
CNTAC10B_062
2010-B
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
2.3.1
Imaging Techniques
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)
h:m:s
d:m:s
Hodge11
06 14 22
−69 50 54
NGC2257
06 30 12
−64 19 34
NGC 1841
04 45 23
−83 59 49
NGC1718
04 52 25
−67 03 05
NGC1846
05 07 35
−67 27 39
SL869
06 14 41
−69 48 07
a
b
[Fe/H]a
dex
−1.84 ± 0.04
−1.59 ± 0.02
−2.02 ± 0.02
−0.80 ± 0.03
−0.49 ± 0.03
−0.40 ± 0.04
RVa
Gyr
245.1±1.0
301.6±0.8
210.3±0.9
278.4±2.2
235.2±0.9
258.4±2.1
Age b (51)
kms−1
15.2
11.9
12.5
9.31
9.17
9.15
From G06
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
2.4.1
Spectral Analysis
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
❄❇❅❆❈❋
Table 2.3 Telescopes and Instruments set-up
Telescope
Observatory
Instrument
UT2-8m
VLT-Paranal
FLAMES(G+U)
Blanco-4m
Yale-1m
Magellan-Baade-6m
Swope-1m
CTIO
CTIO
LCO
LCO
MOSAIC
Y4KCam
IMACS+MOE
SiTe#3 Cam
Instrument set up
H651.5A/HR14A(GIRAFFE)
580nm(UVES)
Imaging
Imaging
WB56-92+Mask
Imaging
Filters
...
...
C, R(T1), I(T2)
V,R,I
...
V,I
spectral range
6293-6686.9 Å
4769-5752+5818-6798 Å
...
...
5600-9200 Å
...
17
R. Mateluna P. - 2012
18
CHAPTER 2. METHOD
Table 2.4 Spectroscopic data set
Object
Telescope/Observatory
Hodge11
UT2/VLT
NGC1718
UT2/VLT
NGC2257
UT2/VLT
NGC1841
Baade/LCO
Instrument
GIRAFFE
UVES
GIRAFFE
UVES
GIRAFFE
UVES
IMACS+MOE
NGC1846
Baade/LCO
IMACS+MOE
Table 2.5 Imaging data set from LCO
Object
Telescope/Observatory
Instrument
Hodge 11
Swope/LCO
Site#3 CCD
NGC 1718
Swope/LCO
Site#3 CCD
NGC 2257
Swope/LCO
Site#3 CCD
NGC 1841
Swope/LCO
Site#3 CCD
NGC 1846
Swope/LCO
Site#3 CCD
Tesis de Doctorado
set up
H651.5A/HR14A
580nm/1” fibers
H651.5A/HR14A
580nm/1” fibers
H651.5A/HR14A
580nm/1” fibers
WB56-92+MASK
WB56-92+MASK
filter
V
I
V
I
V
I
V
I
V
I
Exp. time (s)
2700
2700
2700
2700
2700
2700
3600
3600
2500
1800
Exp. time in (s)
45(1), 90(2), 300(1), 400(1)
45(1), 90(2), 300(1), 400(1)
90(2), 300(1), 400(1)
30(1), 90(2), 300(1), 400(1)
45(1), 90(2), 300(1), 400(1)
45(1), 90(2), 300(1), 400(1)
90(1),400(2)
90(1),150(1), 400(1), 480(1)
90(2), 300(1), 400(1)
90(1), 150(1), 400(1), 480(1)
Exposures
10
10
10
10
5
5
4
3
1
1
seeing
1.45
1.45
1.40
1.30
1.40
1.30
1.50
1.50
1.50
1.40
2.4. SPECTRAL ANALYSIS
19
Table 2.6 LMC Cluster data set from CTIO
Object
Telescope/Observatory Instrument
NGC 1852
4m/CTIO
MOSAIC
4m/CTIO
MOSAIC
NGC 1917
4m/CTIO
MOSAIC
NGC 1987
4m/CTIO
MOSAIC
IC 2146
4m/CTIO
MOSAIC
NGC 2108
4m/CTIO
MOSAIC
Hodge 3
4m/CTIO
MOSAIC
Hodge 11
4m/CTIO
MOSAIC
NGC 1751
+NGC 1795
❄❇❅❆❈❋
filter
C
T1
T2
C
T1
T2
C
T1
T2
C
T1
T2
C
T1
T2
C
T1
T2
C
T1
T2
C
T1
T2
Exp. time in (s)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
1200(1)
180(1)
180(1)
seeing
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
R. Mateluna P. - 2012
l
b
278.68
277.48
278.95
287.90
277.77
279.09
286.54
277.94
278.64
281.19
278.87
278.97
278.45
279.66
281.54
278.42
286.05
279.50
282.93
280.16
283.46
-39.20
-38.97
-38.25
-34.86
-38.03
-37.37
-34.84
-37.19
-36.43
-35.10
-35.58
-34.59
-34.54
-33.42
-32.40
-32.26
-30.91
-31.20
-29.09
-28.51
-28.12
E(B − V )
(mag)
0.01±0.01
0.01±0.01
0.02±0.01
0.09±0.01
0.01±0.01
0.04±0.01
0.09±0.01
0.02±0.01
0.04±0.01
0.11±0.01
0.06±0.01
0.06±0.01
0.06±0.01
0.08±0.01
0.08±0.01
0.06±0.01
0.11±0.01
0.06±0.01
0.11±0.01
0.10±0.01
0.08±0.01
date
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
2008
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
Dec
20
20
20
19
20
20
19
20
20
18
20
20
18
18
18
18
19
19
19
18
19
exposure
C R I (sec)
500 120 120
500 120 120
500 120 120
1200 180 180
500 120 120
500 120 120
1200 180 180
500 120 120
500 120 120
1500 300 300
500 120 120
500 120 120
1200 180 180
1200 180 180
1200 180 180
1200 180 180
1200 180 180
1200 180 180
1200 180 180
1200 180 180
1200 180 180
airmass
CRI
1.281 1.274 1.278
1.285 1.276 1.280
1.326 1.315 1.320
1.426 1.421 1.423
1.389 1.373 1.381
1.440 1.423 1.432
1.421 1.414 1.417
1.485 1.446 1.455
1.512 1.490 1.501
1.302 1.299 1.300
1.549 1.526 1.537
1.581 1.557 1.569
1.279 1.273 1.276
1.403 1.385 1.394
1.475 1.457 1.456
1.497 1.474 1.485
1.447 1.439 1.443
1.368 1.355 1.362
1.437 1.424 1.430
1.318 1.311 1.314
1.482 1.466 1.474
seeing
CRI
1.2 0.8 0.8
1.2 0.9 1.0
1.1 1.0 0.8
1.1 0.8 0.8
1.2 1.0 0.9
1.3 1.0 0.9
1.0 0.9 0.7
1.1 1.0 0.7
1.3 1.0 0.9
1.4 1.2 1.0
1.2 1.1 0.9
1.0 0.9 0.9
1.4 1.2 1.0
1.4 1.0 0.9
1.4 1.0 1.0
1.4 1.0 0.9
1.0 0.8 0.7
1.0 0.8 0.7
1.1 0.8 0.8
1.1 0.9 0.9
1.0 0.9 0.8
Tesis de Doctorado
CHAPTER 2. METHOD
20
Table 2.7 LMC star fields from CTIO.
Field
α2000
δ2000
designation
(h m s)
(d m s)
1
04 23 08.41 -66 29 32.7
2
04 28 45.27 -65 43 23.8
3
04 31 11.35 -67 03 33.0
4
04 32 31.11 -74 59 55.5
5
04 36 40.92 -66 16 26.4
6
04 39 17.96 -67 28 35.4
7
04 40 01.09 -73 59 56.6
8
04 44 12.27 -66 40 18.7
9
04 49 50.84 -67 26 32.3
10
04 57 01.32 -69 48 37.1
11
04 57 52.29 -67 51 43.8
12
05 07 49.38 -68 11 21.4
13
05 09 23.97 -67 46 40.3
14
05 19 02.70 -68 59 59.2
15
05 27 17.84 -70 44 08.2
16
05 33 21.19 -68 09 08.6
17
05 37 48.36 -74 46 59.9
18
05 43 56.31 -69 10 48.1
19
06 07 15.77 -72 16 32.1
20
06 14 28.07 -69 50 52.2
21
06 20 06.93 -72 44 16.6
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):
Teff =
5040
, where:θ = 0.5379 + 0.3981(V − I)o + 0.04432(V − I)2o − 0.02693(V − I)3o
θ
(2.1)
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:
log(
M
Teff
L
g
) = log(
) + 4 · log(
) − log(
)
g⊙
M⊙
T⊙
L⊙
(2.2)
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: Tef f 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, spectrumsynthesis 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)
6300.300
5688.204
5889.951
5895.924
5711.083
6696.014
6155.134
6155.687
6161.287
6162.170
6163.745
6166.429
6169.032
6169.555
6439.070
6493.776
5526.808
5657.863
5684.184
6258.098
6261.094
6261.225
5345.800
5348.315
5420.256
5420.261
5420.270
5420.272
❄❇❅❆❈❋
A
08.0
11.0
11.0
11.0
12.0
13.0
14.0
14.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
21.1
21.1
21.1
22.0
22.0
23.0
24.0
24.0
25.0
25.0
25.0
25.0
E.P. (eV)
0.000
2.104
0.000
0.000
4.340
3.140
5.619
5.619
2.523
1.899
2.521
2.521
2.523
2.526
2.526
2.521
1.768
1.507
1.510
1.440
1.430
0.267
1.004
1.004
2.143
2.143
2.143
2.143
log gf
−9.797
−0.450
0.157
−0.234
−1.670
−1.562
−0.900
−2.390
−1.293
+0.457
−1.303
−1.136
−0.644
−0.227
0.474
+0.129
0.104
−0.403
−1.000
−0.340
−0.440
−2.300
−0.930
−1.140
−3.018
−2.988
−2.733
−3.766
R. Mateluna P. - 2012
24
CHAPTER 2. METHOD
Table 2.8: continued.
Wavelength (Å)
5420.281
5420.295
5420.298
5420.311
5420.329
5420.333
5420.351
5420.374
5420.379
5420.402
5420.429
6230.725
6232.634
6297.789
6318.017
6335.330
6393.602
6430.845
6494.980
5176.110
5342.695
5688.605
6586.306
6643.625
6767.769
4810.527
5087.416
6496.910
6496.899
6496.902
6496.906
6496.916
6496.917
6496.920
6496.910
6496.898
6496.901
6496.906
6496.916
6496.918
A
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
26.0
26.0
26.0
26.0
26.0
26.0
26.0
26.0
27.0
27.0
27.0
28.0
28.0
28.0
30.0
39.1
56.1(134)
56.1(135)
56.1(135)
56.1(135)
56.1(135)
56.1(135)
56.1(135)
56.1(136)
56.1(137)
56.1(137)
56.1(137)
56.1(137)
56.1(137)
Tesis de Doctorado
E.P. (eV)
2.143
2.143
2.143
2.143
2.143
2.143
2.143
2.143
2.143
2.143
2.143
2.559
3.654
2.223
2.453
2.198
2.433
2.176
2.404
2.080
4.021
2.080
1.950
1.676
1.830
4.080
1.080
0.604
0.604
0.604
0.604
0.604
0.604
0.604
0.604
0.604
0.604
0.604
0.604
0.604
log gf
−2.812
−2.511
−3.687
−2.745
−2.327
−3.812
−2.771
−2.169
−4.164
−2.947
−2.029
−1.231
−1.083
−2.620
−1.928
−2.177
−1.522
−2.006
−1.403
−2.120
0.600
−2.220
−2.800
−2.300
−1.960
−0.040
−0.290
−0.230
−1.736
−1.036
−0.589
−1.433
−1.036
−1.036
−0.230
−1.736
−1.036
−0.589
−1.433
−1.036
2.4. SPECTRAL ANALYSIS
25
Table 2.8: continued.
Wavelength (Å)
6496.922
6496.910
6165.891
6645.095
❄❇❅❆❈❋
A
56.1(137)
56.1(138)
59.1
63.1
E.P. (eV)
0.604
0.604
0.923
1.378
log gf
−1.036
−0.230
-0.299
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, Grocholski 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 [email protected] 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<RV> = 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
TARG.6
TARG.9
TARG.4
TARG.11
TARG.2
TARG.16
TARG.8
TARG.10
06:14:21.85
06:14:21.70
06:14:17.05
06:14:31.46
06:14:22.42
06:14:23.34
06:14:24.65
06:14:30.36
-69:50:32.2
-69:49:56.4
-69:50:42.6
-69:49:35.3
-69:51:17.6
-69:52:38.6
-69:50:15.4
-69:49:50.0
RGB
RGB
RGB
RGB
RGB
AGB
AGB
RGB
3.4. ABUNDANCES RESULTS
❄❇❅❆❈❋
Table 3.1 Important parameters for our target stars
Object
R.A.
Dec.
Type of Star
Instrument
UVES
UVES
UVES
UVES
GIRAFFE
GIRAFFE
GIRAFFE
GIRAFFE
RV
(km/s)
241.7
247.5
244.2
244.7
245.2
249.9
246.8
247.4
V
(mag)
15.86
16.88
16.75
17.42
17.81
17.98
18.04
18.21
V-I
1.50
1.18
1.21
1.13
1.12
0.97
0.99
1.07
Teff
(K)
4081
4590
4535
4706
4720
5072
5022
4837
vt
(km/s)
1.85
1.62
1.67
1.56
1.51
1.46
1.46
1.46
log g
0.31
1.20
1.03
1.47
1.64
1.86
1.86
1.85
S/N
@6650 ()
42
22
20
12
60
52
50
40
33
R. Mateluna P. - 2012
34
a
Where X corresponds to any chemical species.
TARG.4
−2.06(5)
0.49(1)
0.60(2)
0.24(2)
−0.08(1)
−0.23(1)
0.26(1)
...
0.33(1)
0.14(3)
0.17(3)
0.21(1)
−0.49(2)
...
...
−0.10(2)
0.16(1)
0.19(1)
0.17(1)
...
...
TARG.11
−2.02(5)
...
0.56(2)
−0.17(2)
−0.16(1)
−0.31(1)
0.22(1)
...
...
0.23(1)
...
...
...
...
...
...
...
...
−0.12(1)
...
...
TARG.2
−1.81(2)
...
...
...
...
...
...
...
...
0.20(1)
...
...
...
...
...
−0.24(1)
...
...
−0.31(1)
...
0.66(1)
TARG.16
−2.08(2)
...
...
...
...
...
...
...
...
0.14(1)
...
...
...
...
...
0.03(1)
...
...
−0.10(1)
...
...
TARG.8
−2.06(2)
...
...
...
...
...
...
...
...
0.27(1)
...
...
...
...
...
−0.11(1)
...
...
0.34(1)
...
...
TARG.10
−1.86(3)
...
...
...
...
...
...
...
...
0.05(1)
...
...
...
...
...
...
...
...
−0.13(1)
...
...
<[X/Fe]>
−2.00
0.54
0.43
0.02
−0.14
−0.32
0.25
...
0.39
0.13
0.07
0.16
−0.49
...
...
−0.16
0.06
0.04
−0.01
...
0.62
σobs
0.11
0.05
0.25
0.32
0.17
0.14
0.12
...
0.11
0.10
0.16
0.06
0.02
...
...
0.13
0.18
0.14
0.23
...
0.05
Tesis de Doctorado
CHAPTER 3. HODGE 11
Table 3.2 [X/Fe] values for Hodge 11 stars.
[X/Fe]a
TARG.6 TARG.9
[Fe/H]
−2.11(7) −2.03(2)
OI
0.57(1)
0.57(1)
NaD
...
0.14(2)
NaDN LT E
... −0.21(2)
NaI (5688)
0.03(1) −0.37(1)
NaI (5688)N LT E −0.22(1) −0.52(1)
MgI
0.41(1)
0.11(1)
AlI
0.39(1)
...
SiI
0.52(1)
0.33(1)
CaI
0.05(7) −0.04(5)
ScII
0.15(3) −0.12(2)
TiI
0.09(2)
0.18(2)
CrI
−0.51(2) −0.47(1)
MnI
−0.49(1)
...
CoI
0.26(3)
...
NiI
−0.08(3) −0.37(2)
ZnI
0.16(1) −0.15(1)
YII
−0.08(1)
0.02(2)
BaII
−0.16(1)
0.23(1)
LaII
−0.09(1)
...
EuII
0.59(1)
...
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 Calcium 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 preimages 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: VF S = CT(v-i)F + ∆ V + vF , and IF S = CT(v-i)F + ∆ I
+ iF , where VF S and IF S 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: Tef f 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
❄❇❅❆❈❋
Table 4.1 Important parameters for our target stars
Object
R.A.
Dec.
Type Instrument
of Star
TARG.13 06:14:40.79 -69:47:59.9 RGB
GIRAFFE
TARG.14 06:14:39.17 -69:47:31.0 RGB
UVES
RV
(km/s)
263.84
263.06
V(Swope)
(mag)
17.63
17.42
V-I(Swope)
1.21
1.28
Teff
(K)
4593
4578
vt
(km/s)
1.88
2.22
log g
1.39
1.91
S/N
@6620 (Å)
48
12
49
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
TARG.13 GIRAFFE 7.00
20
6.99
TARG.14
UVES
7.06
75
7.04
nr. lines
1
4
[FeI/H]
-0.50
-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
Table 4.3 [X/Fe] values for field stars.
ID
[Fe/H] [Ca/Fe]
BK41
−0.59
−0.42
BK58
−0.74
0.05
BK102
−0.65
−0.29
BK93
−0.79
−0.02
BK80
−0.66
0.23
BK14
−1.05
0.22
BK97
−0.95
−0.09
BK50
−0.97
−0.12
BK109
−0.71
−0.33
BK65
−0.83
0.02
BK87
−0.52
0.21
TARG.22 −0.88
0.21
BK55
−0.37
−0.10
BK37
−0.23
−0.33
BK44
−1.11
0.22
BK19
−0.86
−0.37
BK9
−0.68
−0.73
BK113
−0.79
0.41
BK105
−0.95
0.08
BK60
−0.62
−0.39
BK95
−0.78
0.00
BK98
−0.4
−0.39
BK31
−0.86
−0.08
BK2
−0.86
0.42
BK82
−0.99
−0.43
BK71
−1.24
0.64
BK81
−0.97
−0.01
BK108
−0.55
−0.02
BK99
−0.49
−0.05
BK8
−0.61
0.34
❄❇❅❆❈❋
53
[Ti/Fe]
...
...
...
−0.20
0.03
...
...
...
...
...
...
0.63
−0.22
...
...
...
0.11
...
−0.26
...
...
...
0.50
0.43
...
...
0.04
...
−0.34
0.27
[Cr/Fe]
−0.90
−0.70
...
...
−0.73
−0.21
−0.25
...
−0.46
−0.14
...
−0.55
−0.44
...
−0.13
−0.87
...
−0.55
−0.09
−0.92
−0.26
0.45
−0.32
−0.65
−0.47
...
−0.25
0.01
−0.81
−0.88
[Ni/Fe]
−0.75
−0.76
...
−0.28
...
−0.32
−0.40
...
−0.30
−0.21
...
−0.27
−0.47
−0.55
−0.35
−0.76
...
−0.07
−0.25
−0.57
−0.60
...
−0.65
−0.79
−0.29
0.30
−0.97
0.24
−0.45
−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 Fulbright(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 CT1 T2 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
Table 5.1 Washington non-calibrated magnitudes of H11 target stars
ID
RA
DEC
r(t1) i(t2)
c
TARG.11 06:14:31.460 −69 : 49 : 35.29 10.96 11.06 11.53
TARG.9 06:14:21.700 −69 : 49 : 56.40 10.37 10.47 11.08
TARG.8 06:14:24.650 −69 : 50 : 15.40 11.54 11.78 11.90
TARG.16 06:14:23.340 −69 : 52 : 38.60 11.54 11.78 11.69
TARG.2 06:14:22.423 −69 : 51 : 17.63 11.18 11.39 11.83
TARG.10 06:14:30.359 −69 : 49 : 49.98 11.75 11.89 12.19
a
59
c-t1
0.57
0.71
0.36
0.15
0.65
0.44
[Fe/H]a
−2.02
−2.03
−2.06
−2.08
−1.81
−1.86
Data 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 CT1 T2 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 V I technique (Geisler & Sarajedini(1999)(29)). Thus, for a given
photometric accuracy, metallicities can be determined 3 times more precisely with the Washington 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.
Field
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
1
9.0
2.8
9.5
2.9
8.6
2.7
11.1
3.3
9.5
2.9
9.0
2.8
11.1
3.3
8.6
2.7
6.5
2.1
2.7
0.8
2.7
0.8
2.8
0.8
3.0
0.9
2.0
0.4
3.9
1.3
5.5
1.8
6.9
2.2
4.4
1.4
9.5
2.9
10.0
3.1
8.6
2.7
9.0
2.8
9.5
2.9
11.1
3.3
9.5
2.9
11.7
3.5
9.0
2.8
10.0
3.1
11.1
3.3
6.5
2.1
2.1
0.5
4.7
1.5
3.7
1.2
2.8
0.8
1.6
0.3
4.9
1.6
4.7
1.5
6.9
2.2
6.2
2.0
11.7
3.5
12.9
3.8
12.3
3.6
11.7
3.5
10.5
3.2
8.6
2.7
9.5
2.9
11.7
3.5
9.0
2.8
4.7
1.5
8.6
2.7
4.9
1.6
3.5
1.1
3.5
1.1
3.5
1.1
2.8
0.8
1.6
0.3
3.7
1.2
4.7
1.5
11.7
3.5
8.1
2.6
6.9
2.2
10.0
3.1
9.5
2.9
9.0
2.8
11.1
3.3
11.7
3.5
9.5
2.9
11.7
3.5
9.5
2.9
10.5
3.2
8.6
2.7
5.2
1.7
2.7
0.8
4.7
1.5
2.7
0.8
2.8
0.8
2.0
0.4
3.7
1.2
4.9
1.6
11.7
3.5
8.1
2.6
11.7
3.5
10.0
3.1
9.5
2.9
9.0
2.8
8.1
2.6
9.0
2.8
9.5
2.9
11.7
3.5
11.7
3.5
10.5
3.2
9.0
2.8
8.6
2.7
3.5
1.1
3.5
1.1
4.9
1.6
3.0
0.9
1.5
0.2
4.9
1.6
6.5
2.1
12.3
3.6
4.4
1.4
9.0
2.8
10.0
3.1
12.3
3.6
11.1
3.3
10.5
3.2
11.1
3.3
7.7
2.5
11.7
3.5
9.0
2.8
10.5
3.2
11.1
3.3
4.9
1.6
2.1
0.5
2.8
0.8
4.9
1.6
3.9
1.3
1.4
0.2
3.9
1.3
4.7
2.1
9.5
2.9
8.1
2.6
11.7
3.5
10.0
3.1
12.3
3.6
11.1
3.3
10.5
3.2
11.1
3.3
10.0
3.1
9.0
2.8
11.1
3.3
8.1
2.6
8.6
2.7
3.7
1.2
2.1
0.5
4.9
1.6
3.7
1.2
3.7
1.2
2.0
0.4
3.9
1.3
6.2
2.0
9.5
2.9
8.6
2.7
9.0
2.8
12.9
3.8
12.3
3.6
11.7
3.5
8.6
2.7
9.0
2.8
10.5
3.2
11.7
3.5
9.0
2.8
8.1
2.6
9.0
2.8
4.9
1.6
2.1
0.5
4.9
1.6
4.7
1.5
3.7
1.2
2.3
0.6
3.9
1.3
6.2
2.0
6.5
2.1
10.0
3.1
9.0
2.8
10.0
3.1
12.3
3.6
9.0
2.8
10.0
3.1
11.7
3.5
9.5
2.9
11.7
3.5
9.0
2.8
10.5
3.2
9.0
2.8
6.5
2.1
2.1
0.5
4.7
1.5
3.7
1.2
3.9
1.3
1.6
0.3
3.7
1.2
4.7
1.5
10.0
3.1
4.4
1.4
9.0
2.8
10.0
3.1
9.5
2.9
11.1
3.3
8.1
2.6
11.1
3.3
10.0
3.1
11.7
3.5
9.0
2.8
10.5
3.2
8.6
2.7
8.6
2.7
2.1
0.5
6.5
2.1
3.7
1.2
5.2
1.7
1.5
0.2
3.9
1.3
6.2
2.0
11.7
3.5
4.7
1.5
9.0
2.8
12.9
3.8
12.3
3.6
11.1
3.3
10.5
3.2
8.6
2.7
10.0
3.1
9.0
2.8
11.7
3.5
13.5
3.9
8.6
2.7
6.5
2.1
2.7
0.8
3.7
1.2
4.9
1.6
3.7
1.2
2.4
0.7
3.9
1.3
6.2
2.0
11.7
3.5
5.8
1.9
9.0
2.8
7.7
2.5
9.5
2.9
11.7
3.5
11.1
3.3
11.7
3.5
10.5
3.2
6.5
2.1
9.0
2.8
10.5
3.2
8.6
2.7
4.9
1.6
2.1
0.5
4.9
1.6
2.8
0.8
3.7
1.2
4.2
1.4
3.2
1.0
3.5
1.1
9.0
2.8
10.0
3.1
9.0
2.8
12.9
3.8
12.3
3.6
8.6
2.7
10.5
3.2
8.6
2.7
12.9
3.8
11.7
3.5
9.0
2.8
12.9
3.8
7.3
2.3
9.0
2.8
2.7
0.8
3.7
1.2
3.7
1.2
2.3
0.6
3.7
1.2
3.7
1.2
5.2
1.7
10.0
3.1
3.5
1.1
9.0
2.8
10.0
3.1
9.5
2.9
11.1
3.3
10.5
3.2
9.0
2.8
8.1
2.6
12.3
3.6
11.7
3.5
8.1
2.6
6.9
2.2
6.9
2.2
2.7
0.8
4.9
1.6
3.9
1.3
3.9
1.3
3.2
1.0
3.2
1.0
6.5
2.1
9.5
2.9
6.2
2.0
6.9
2.2
10.0
3.1
9.5
2.9
11.1
3.3
8.1
2.6
11.7
3.5
8.1
2.6
7.3
2.3
9.0
2.8
8.1
2.6
9.0
2.8
5.2
1.7
1.7
0.3
3.7
1.2
3.9
1.3
3.0
0.9
3.7
1.2
3.0
0.9
4.9
1.6
9.5
2.9
5.8
1.9
12.3
3.6
7.7
2.5
12.3
3.6
11.7
3.5
8.1
2.6
9.0
2.8
8.6
2.7
9.5
2.9
9.5
2.9
11.1
3.3
9.0
2.8
6.5
2.1
2.1
0.5
5.2
1.7
5.2
1.7
3.7
1.2
3.7
1.2
2.4
0.7
3.7
1.2
9.0
2.8
4.4
1.4
9.5
2.9
8.1
2.6
10.0
3.1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
❄❇❅❆❈❋
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.
Field
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
1
-0.96
0.31
-0.88
0.35
-0.85
0.28
-0.90
0.20
-0.78
0.35
-0.86
0.31
-0.85
0.20
-0.90
0.28
-0.81
0.25
-0.61
0.33
-0.56
0.33
-0.53
0.34
-0.61
0.34
-0.52
0.30
-0.67
0.33
-0.67
0.28
-0.91
0.24
-0.71
0.31
-1.13
0.35
-1.15
0.20
-0.85
0.28
-0.86
0.31
-0.88
0.35
-0.95
0.20
-0.88
0.35
-0.90
0.20
-0.91
0.31
-0.95
0.20
-1.00
0.20
-0.81
0.25
-0.66
0.31
-0.78
0.31
-0.70
0.33
-0.58
0.34
-0.40
0.27
-0.80
0.30
-0.73
0.31
-0.91
0.24
-0.70
0.26
-1.10
0.20
-1.30
0.20
-0.95
0.20
-0.90
0.20
-0.90
0.20
-0.90
0.28
-0.88
0.35
-0.90
0.20
-0.86
0.31
-0.78
0.31
-0.95
0.28
-0.80
0.30
-0.88
0.34
-0.68
0.34
-0.68
0.34
-0.58
0.34
-0.45
0.27
-0.70
0.33
-0.68
0.31
-1.15
0.20
-0.74
0.26
-0.96
0.24
-1.30
0.20
-0.93
0.35
-0.76
0.31
-0.90
0.20
-0.90
0.20
-0.93
0.35
-0.90
0.20
-0.93
0.35
-0.95
0.20
-1.00
0.28
-0.86
0.29
-0.71
0.33
-0.78
0.31
-0.51
0.33
-0.48
0.34
-0.52
0.30
-0.65
0.33
-0.75
0.30
-1.05
0.20
-0.74
0.26
-0.95
0.20
-1.10
0.20
-0.98
0.35
-0.91
0.31
-0.74
0.26
-0.91
0.31
-0.83
0.35
-0.90
0.20
-0.85
0.20
-0.90
0.20
-0.96
0.31
-0.85
0.28
-0.83
0.34
-0.68
0.34
-0.80
0.30
-0.66
0.34
-0.36
0.26
-0.80
0.30
-0.81
0.25
-1.20
0.20
-0.61
0.31
-1.01
0.31
-1.25
0.20
-0.95
0.20
-1.00
0.20
-0.90
0.20
-0.95
0.20
-0.93
0.25
-0.90
0.20
-0.96
0.31
-0.95
0.20
-1.10
0.20
-0.85
0.30
-0.61
0.34
-0.58
0.34
-0.85
0.30
-0.77
0.33
-0.40
0.25
-0.77
0.33
-0.73
0.31
-1.18
0.35
-0.74
0.26
-1.15
0.20
-1.30
0.20
-1.00
0.20
-0.95
0.20
-0.90
0.20
-0.95
0.20
-1.00
0.20
-0.86
0.31
-0.95
0.20
-0.89
0.26
-1.05
0.28
-0.70
0.33
-0.61
0.31
-0.85
0.30
-0.75
0.33
-0.75
0.33
-0.42
0.30
-0.77
0.33
-0.80
0.26
-1.13
0.35
-0.75
0.28
-1.11
0.31
-1.25
0.20
-1.00
0.20
-1.00
0.20
-0.85
0.28
-0.86
0.31
-1.00
0.20
-0.90
0.20
-0.91
0.31
-0.89
0.26
-1.06
0.31
-0.80
0.30
-0.66
0.31
-0.85
0.30
-0.78
0.31
-0.70
0.33
-0.45
0.32
-0.77
0.33
-0.80
0.26
-0.96
0.25
-0.80
0.20
-1.11
0.31
-1.20
0.20
-1.00
0.20
-0.96
0.31
-0.80
0.20
-0.95
0.20
-0.88
0.35
-0.90
0.20
-0.91
0.31
-0.90
0.20
-1.01
0.31
-0.81
0.25
-0.66
0.31
-0.83
0.31
-0.70
0.33
-0.77
0.33
-0.35
0.27
-0.70
0.33
-0.73
0.31
-1.25
0.20
-0.71
0.31
-1.06
0.31
-1.20
0.20
-0.93
0.35
-0.95
0.20
-0.84
0.26
-0.95
0.20
-1.00
0.20
-0.90
0.20
-0.96
0.31
-0.95
0.20
-1.10
0.28
-0.95
0.28
-0.61
0.31
-0.91
0.25
-0.70
0.33
-0.91
0.29
-0.31
0.26
-0.77
0.33
-0.75
0.26
-1.15
0.20
-0.63
0.31
-1.11
0.31
-1.20
0.20
-1.05
0.20
-0.95
0.20
-0.90
0.20
-0.90
0.28
-0.95
0.20
-0.86
0.31
-1.00
0.20
-0.90
0.20
-1.05
0.28
-0.86
0.25
-0.71
0.33
-0.70
0.33
-0.80
0.30
-0.75
0.33
-0.52
0.33
-0.77
0.33
-0.80
0.26
-1.15
0.20
-0.69
0.27
-1.11
0.31
-1.13
0.25
-1.03
0.35
-0.95
0.20
-0.85
0.20
-0.90
0.20
-0.95
0.20
-0.86
0.25
-0.91
0.31
-1.00
0.20
-1.05
0.28
-0.80
0.30
-0.61
0.31
-0.80
0.30
-0.53
0.34
-0.70
0.33
-0.69
0.32
-0.73
0.34
-0.63
0.34
-1.06
0.31
-0.80
0.20
-1.11
0.31
-1.15
0.20
-1.00
0.20
-0.85
0.28
—
0.20
-0.85
0.28
-0.90
0.20
-0.85
0.20
-0.86
0.31
-0.80
0.20
-1.02
0.24
-0.81
0.31
-0.56
0.33
-0.65
0.33
-0.60
0.33
-0.50
0.32
-0.65
0.33
-0.65
0.33
-0.66
0.29
-1.15
0.20
-0.63
0.34
-0.96
0.31
-1.10
0.20
-0.88
0.35
-0.95
0.20
-0.90
0.20
-0.86
0.31
-0.94
0.26
-0.85
0.20
-1.00
0.20
-0.89
0.26
-1.06
0.24
-0.81
0.24
-0.66
0.33
-0.80
0.30
-0.72
0.33
-0.72
0.33
-0.58
0.34
-0.68
0.34
-0.76
0.25
-1.13
0.35
-0.70
0.26
-1.06
0.24
-1.20
0.20
-1.03
0.35
-0.95
0.20
-0.84
0.26
-0.90
0.20
-0.89
0.26
-0.77
0.24
-0.91
0.31
-0.89
0.26
-1.11
0.31
-0.76
0.29
-0.48
0.28
-0.70
0.33
-0.72
0.33
-0.66
0.34
-0.65
0.33
-0.71
0.34
-0.75
0.30
-1.13
0.35
-0.69
0.27
-1.15
0.20
-1.13
0.25
-1.10
0.20
-0.90
0.20
-0.89
0.26
-0.86
0.31
-0.90
0.28
-0.88
0.35
-1.08
0.35
-1.00
0.20
-1.26
0.31
-0.91
0.25
-0.56
0.31
-0.81
0.29
-0.86
0.29
-0.70
0.33
-0.65
0.33
-0.67
0.33
-0.60
0.33
-1.11
0.31
-0.61
0.31
-1.28
0.35
-1.24
0.26
-1.15
0.20
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Tesis de Doctorado
5.4. TESTING THE INITIAL "SKZ" PIPELINE AND OBTAINING PSF PHOTOMETRY63
5.4
Testing the initial "SkZ" pipeline and obtaining PSF Photometry
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 ALLSTAR. 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 [email protected] 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
-2
-1
0
c-t1
1
2
24
-1
0
1
2
3
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.
9
16
10
17
11
18
12
19
13
-2
-1
0
1
2
c-t1
-1
0
1
C-T1
Figure 5.3 Zoom in the CMD from Figs.5.2 in the area of the RGB.
Tesis de Doctorado
2
3
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 [email protected] (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
70
CHAPTER 6. PRELIMINARY RESULTS ON THE CHEMICAL EVOLUTION OF THE
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 & Tautvaišienė (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 & Tautvaišienė (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
72
CHAPTER 6. PRELIMINARY RESULTS ON THE CHEMICAL EVOLUTION OF THE
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 & Tautvaišienė (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 metalpoor 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 ageÐmetallicity relation can be reproduced by models in which a large amount (∼ 109 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 & Tautvaišienė (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 & Tautvaišienė (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
74
CHAPTER 6. PRELIMINARY RESULTS ON THE CHEMICAL EVOLUTION OF THE
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 & Tautvaišienė (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 (αelements, 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-tomagnesium: [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.
1
Univ. of North Carolina at Charlotte, US.
FCAG/UNLP Observatorio Astronomico, La Plata, Argentina.
3
California State University SB, NASA Johnson Space Center, US.
4
Jeremiah Horrocks InstituteUniversity of Central Lancashire, UK.
5
MDF is basically a histogram of the distribution of metals in a galaxy or in star cluster
2
❄❇❅❆❈❋
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 metalpoor 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 intermediateage 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:
σtot =
q
2
2
2
2
σT2 eff + σlog
g + σvt + σ[m/H] + σS/N
(A.1)
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
+50K
+0.10 +0.05
[Fe/H]
0.08
−0.03 −0.09
[O/Fe]
−0.09
0.015
0.00
[Na/Fe] (D)
−0.03
−0.01
0.08
[Na/Fe]
−0.04
0.02
0.09
[Mg/Fe]
0.055 −0.005
0.09
[Al/Fe]
−0.035
0.02
0.09
[Si/Fe]
−0.12
−0.04 −0.05
[Ca/Fe]
−0.045
−0.01 −0.01
[Sc/Fe]
−0.08
0.04
0.06
[TiI/Fe]
−0.005
0.01
0.08
[Cr/Fe]
0.01
0.005
0.06
[Mn/Fe]
0.015
0.05
0.15
[Co/Fe]
−0.135
0.07
0.07
[Ni/Fe]
−0.02
0.00
0.08
[Zn/Fe]
−0.075
0.02 −0.02
[Y/Fe]
−0.06
0.06
0.07
[Ba/Fe]
−0.025
0.065
0.05
[La/Fe]
0.020
0.05
0.10
[Eu/Fe]
−0.085
0.055
0.04
Tesis de Doctorado
∆[m/H]
+0.1
−0.05
0.01
0.04
0.01
0.06
0.04
0.01
−0.01
−0.02
0.04
0.04
0.13
0.16
0.05
−0.04
0.04
0.04
0.10
0.02
σS/N
σtot
σobs
0.03
0.06
0.04
0.06
0.06
0.06
0.06
0.03
0.03
0.06
0.04
0.06
0.03
0.04
0.06
0.06
0.06
0.06
0.06
0.14
0.11
0.10
0.12
0.13
0.12
0.15
0.06
0.11
0.11
0.08
0.21
0.23
0.10
0.11
0.13
0.11
0.16
0.13
0.11
0.05
0.25
0.17
0.12
...
0.11
0.10
0.16
0.06
0.02
...
...
0.13
0.18
0.14
0.23
...
0.05
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(Super MONGO), 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 driver s 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"
Figure B.1 Example of "parameter file: synth"
1
http //www.as.utexas.edu/c̃hris/moog.html
Tesis de Doctorado
1
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 3 He, major part of 4 He and some 7 Li. (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: 4 He
• 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. Characteristic in high mass stars (temperatures: ∼ 4x107 K)
• Triple-α : synthesis of helium into carbon. First: two α-particles forms an unstable
beryllium(8 Be) nucleus, before 8Be decays, reacts with another α-particle forming carbon(12 C).
Is possible for carbon nuclei to capture α-particles producing oxygen and also 16 O 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 University 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., Tautvaišienė G. 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., Cȯté, 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. Conference 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 Astronomical 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 Astronomical 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