Author: Lucas M. Pulgar-Escobar
Institution: Departamento de Astronomía, Universidad de Concepción (Chile)
Degree: Master of Science in Astronomy
Thesis title: Characterizing NGC 6383: Membership, Pre-Main Sequence Stars, and Mass Segregation using Gaia DR3 and 2MASS
Supervisors: Dr. Ronald Mennickent, Dr. Pierluigi Cerulo
This repository contains the LaTeX source files, analysis scripts, and supporting material for the MSc thesis focused on the young open cluster NGC 6383, located in the Carina–Sagittarius arm within the Sh 2-012 star-forming region.
The work combines Bayesian inference, machine-learning clustering, and classical stellar population diagnostics to derive robust estimates of cluster parameters, identify pre-main-sequence (PMS) stars, and evaluate evidence for primordial mass segregation.
| Component | Description |
|---|---|
| Data sources | Gaia DR3, 2MASS, VPHAS+ |
| Membership analysis | HDBSCAN clustering with astrometric fidelity filtering |
| Bayesian modeling | PyMC 5 with the No-U-Turn Sampler (NUTS) for age, distance, and extinction inference |
| Isochrone fitting | ASteCA with MIST isochrones |
| PMS identification | Sagitta neural network (Gaia DR3 + 2MASS) |
| Statistical tools | Python 3.12 (Astropy, HDBSCAN, PyMC, Matplotlib, NumPy, Pandas) |
| Custom package | COSMIC — Characterization Of Star clusters using Machine-learning Inference and Clustering |
.
├── src/
│ ├── main.tex # Entry point for the thesis
│ ├── cites.bib # Bibliography database (apalike format)
│ ├── chapters/ # Chapter subfiles
│ ├── frontmatter/ # Title page, abstract, dedication, etc.
│ ├── preamble/ # Shared packages, metadata, and front-matter helpers
│ └── figures/ # Figures and graphics
├── build/ # LaTeX outputs (PDF, aux, log — gitignored)
├── Makefile # latexmk wrapper for reproducible builds
├── LICENSE # MIT License for code and analysis
└── README.md # Project overview
The legacy Spanish-named folders (Capítulos/, Otros/, Images/) and root-level LaTeX files were migrated into the src/ hierarchy to keep the project portable and fully English. Shared LaTeX configuration now lives under src/preamble/ so you can reuse the setup across chapters or derivative documents.
Install latexmk (TeX Live or MacTeX include it by default) and run:
make # Builds build/thesis.pdf
make watch # Continuous compilation (latexmk -pvc)
make clean # Remove auxiliary files under build/All intermediate files and the final PDF live under build/, which is ignored by git.
- Edit thesis metadata (title, advisor, dates, etc.) in
src/preamble/metadata.tex; changes propagate to the title and grading pages automatically. - Adjust packages, counters, or global layout via
src/preamble/thesis.sty. - Customise headers, hyperlink colours, and other page styles in
src/preamble/page_styles.tex. - Reorder or tweak the licence/dedication/acknowledgements flow inside
src/preamble/frontmatter_macros.tex.
NGC 6383 is a young open cluster (~3–4 Myr) embedded in the Sh 2-012 region.
This thesis refines its fundamental parameters through a unified Bayesian–machine-learning pipeline:
- Cluster membership: Robust determination via unsupervised clustering and astrometric fidelity weighting.
- Age and extinction: Joint posterior inference combining Gaia and 2MASS photometry.
- Pre-Main Sequence population: Neural-network classification of PMS stars, cross-validated with CMD and Sagitta outputs.
- Mass segregation: Quantitative assessment of stellar-mass stratification via cumulative radial distributions and K–S statistics.
If you use this repository or the COSMIC pipeline, please cite:
Pulgar-Escobar, L. M., Henríquez-Salgado, N. A., Mennickent, R. E., & Cerulo, P. (2025).
Characterizing NGC 6383: A study of pre-main-sequence stars, mass segregation, and age using Gaia DR3 and 2MASS.
Submitted to Astronomy & Astrophysics (A&A).
- Text and figures: © 2025 Lucas M. Pulgar-Escobar — All Rights Reserved.
The thesis text may not be redistributed or reproduced without explicit permission. - Code (COSMIC, scripts, and analysis): Released under the MIT License.
See LICENSE and LICENSE_thesis for details.
This work was supported by:
- ANID BASAL project FB210003
- SOCHIAS GEMINI project 32230014
and makes use of:
- ESA Gaia mission data (DPAC)
- Two Micron All-Sky Survey (2MASS)
- The Astropy community ecosystem
Email: lescobar2019@udec.cl
GitHub: https://github.com/notluquis