This repository creates a series of plots of stellar spectra in multiple languages.
The command to run code from this repository is:
python3 spectral_types_plot.py
The following command line options are available:
--lang LANG add language code
--plot_dir PLOT_DIR add directory for output plots. Default is plots directory in this repository.
--translations_file TRANSLATIONS_FILE
add your own JSON file containing translations. Default is translations.json in this repository.
--output_format OUTPUT_FORMAT
add the output format for the plots. options: eps, jpg, jpeg, pdf, pgf, png, ps, raw, rgba,
svg, svgz, tif, tiff. Default is png.
--translate_filenames TRANSLATE_FILENAMES
If True output filenames will be in requested language. If False output filenames will be in
English. Default is False
Example usage, for English, using untranslated filenames, the default translation file, pdf output format and output directory "/home/user/plots/", one would use the command:
python3 spectral_types_plot.py --lang=en --output_format=pdf --plot_dir=/home/user/plots/
The code creates one plot for each of the seven spectral types (showing both a line of wavelength vs flux and a band plot showing light and dark patches on the spectrum) and two comparison plots showing all seven spectra (one with a line plot, one with a band plot).
The code released is available under an MIT license and should be credited to IAU OAE/Niall Deacon. The plots in the plots directory and the translations in the translations directory are published under a CC-BY-4.0 license. The data in the data folder are from the Sloan Digital Sky Survey and are public domain.
Please credit all plots created by this code to IAU OAE/SDSS/Niall Deacon adding credits to additional translators where appropriate. Some of the characteristics of the plots were inspired by the plots created by the MaStar Stellar Library team. These plots were generated using Sloan Digital Sky Survey (SDSS) spectroscopic data from the MaStar Stellar Library. More information on the MaStar Stellar Library and SDSS can be found in Yan et al. (in prep), Yan et al. (2019) and the Abdurro’uf et al. (2021) and Blanton et al. (2017).
Ali Al-Edhari
Samantha Brown-Sevilla
Rulx Narcisse
Giuliana Giobbi, Francesco Salvestrini
Lin Shijie
Lin Shijie
You can add your own translations by downloading this repository and editing the translations.json file. Each language starts with a language code followed by translations of each text element. We have included a placeholder zz language code that you can edit (or copy and edit if you want to add more than one language). For example if you were to translate the terms into Brazilian Portuguese you would edit:
"zz":{
"translation_approval_level":"N",
"Translation Credit":null,
"matplotlib_cairo": false,
"possible_fonts": [
"Noto Sans",
"Arial"
],
"xaxis_text":"Wavelength (nanometers)",
"yaxis_text":"Flux",
"spectral_features_title":"Important spectral features",
"balmer_text":"Hydrogen atoms",
"helium_atoms_text":"Helium atoms",
"helium_ions_text":"Helium ions",
"calcium_atoms_text":"Calcium atoms",
"calcium_ions_text":"Calcium ions",
"sodium_atoms_text":"Sodium atoms",
"titanium_oxide_text":"Titanium Oxide",
"title0":"Spectrum of an O-type star",
"title1":"Spectrum of a B-type star",
"title2":"Spectrum of an A-type star",
"title3":"Spectrum of an F-type star",
"title4":"Spectrum of a G-type star",
"title5":"Spectrum of a K-type star",
"title6":"Spectrum of an M-type star",
"spectrum_title0":"O-type star",
"spectrum_title1":"B-type star",
"spectrum_title2":"A-type star",
"spectrum_title3":"F-type star",
"spectrum_title4":"G-type star",
"spectrum_title5":"K-type star",
"spectrum_title6":"M-type star",
"bands_title":"Stellar Spectra",
"lambda_lower":"365 nm",
"lambda_upper":"900 nm",
"bands_filename":"spectra bands",
"lines_filename":"spectra lines"
}
And instead have:
"pt-br": {
"translation_approval_level": "N",
"Translation Credit":"Translator: Eduardo Penteado",
"matplotlib_cairo": false,
"possible_fonts": [
"Noto Sans",
"Arial"
],
"xaxis_text": "comprimento de Onda (nanômetros)",
"yaxis_text": "fluxo",
"spectral_features_title": "características espectrais importantes",
"balmer_text": "átomos de Hidrogênio",
"helium_atoms_text": "átomos de Hélio",
"helium_ions_text": "íons de Hélio",
"calcium_atoms_text": "Átomos de Cálcio",
"calcium_ions_text": "íons de Cálcio",
"sodium_atoms_text": "átomos de Sódio",
"titanium_oxide_text": "óxido de Titânio",
"title0": "Espectro de uma estrela do tipo O",
"title1": "Espectro de uma estrela do tipo B",
"title2": "Espectro de uma estrela do tipo A",
"title3": "Espectro de uma estrela do tipo F",
"title4": "Espectro de uma estrela do tipo G",
"title5": "Espectro de uma estrela do tipo K",
"title6": "Espectro de uma estrela do tipo M",
"spectrum_title0": "estrela do tipo O",
"spectrum_title1": "estrela do tipo B",
"spectrum_title2": "estrela do tipo A",
"spectrum_title3": "estrela do tipo F",
"spectrum_title4": "estrela do tipo G",
"spectrum_title5": "estrela do tipo K",
"spectrum_title6": "estrela do tipo M",
"bands_title": "Espectros Estelares",
"lambda_lower": "365 nm",
"lambda_upper": "900 nm",
"bands_filename": "bandas_espectrais",
"lines_filename": "linhas_espectrais" },
Then just run:
python3 spectral_types_plot.py --lang=zz
With zz replaced by your language code.
Below are links to the diagrams produced by this code. You can also find the diagram captions and any translations of these captions in the links.
- Spectrum of an O-type star
- Spectrum of a B-type star
- Spectrum of an A-type star
- Spectrum of an F-type star
- Spectrum of a G-type star
- Spectrum of a K-type star
- Spectrum of an M-type star
- Stellar spectral types
- Stellar spectral types - bands
The built-in fonts for matplotlib often struggle with non-Latin characters. The code is set up to try to load commonly used fonts for the writing system it is producing the plots for. If you want to load a font that is already installed on your system then you can tell the code to use that font by adding it to the start of the list in the possible_fonts list in translations.json. If you are struggling to get a particular writing system to work with this code then you can download the font you want to use and copy the .ttf file to the fonts folder of this repository. The code will then automatically load that font. The Google Noto Fonts project provides fonts in a wide range of writing systems. For some writing systems (mostly scripts used in South Asia such as Devanagari or Bengali) we recommend you use the mplcairo matplotlib backend. Once you have installed mplcairo, change "matplotlib_cairo" from false to true (lowercase, no quotemarks). We do not include mplcairo in the requirements.txt file as it is a little complex to install and only required by some users.
All of the spectra plotted have their fluxes normalised so that they all have a maximum of the same value. The line plots with multiple stars plotted together have the normalised spectra of each star offset by differing amounts for ease of comparison. The total flux emitted by a star depends on its temperature and size so while an O-type star appears faint in red light on these plots it will emit much more red light than a faint red star (red dwarf).
The color representation is a linear colour spectrum from 450-650nm. Bluer than 450nm is coloured blue, even though the human eye sees very little bluer than 400nm. Redder than 650nm is coloured red even thought the human eye has very little redder than 750nm.
For languages other than English, please check the translation approval level in the translations.json file. If approval level is marked as 'N' then the translation has not been reviewed, translations marked 'Y' have been approved by a reviewer in our review system.