juliensimon/wolf-rayet-stars

Galactic Wolf-Rayet Stars

Credit: NASA/ESA/Hubble

Part of a dataset collection on Hugging Face.

Dataset description

Catalog of Galactic Wolf-Rayet stars — massive evolved stars with powerful stellar winds and broad emission lines. Wolf-Rayet stars represent a brief but spectacular late stage in the lives of the most massive stars (>25 solar masses), just before they explode as supernovae.

Wolf-Rayet (WR) stars are among the hottest and most luminous stars known, with surface temperatures of 30,000-200,000 K and luminosities up to a million times the Sun. Their spectra are dominated by broad emission lines from helium, nitrogen (WN subtype), carbon (WC subtype), or oxygen (WO subtype), produced by their extreme stellar winds losing mass at rates of 10^-5 solar masses per year.

This catalog is based on Rate & Crowther (2020), which combined the most complete census of Galactic WR stars with Gaia DR2 parallaxes to derive distances, luminosities, and spatial distribution. The dataset includes astrometric positions, spectral classifications, Gaia and infrared photometry, and distance estimates.

Wolf-Rayet stars represent a fleeting but critical phase in massive star evolution. Stars born with initial masses above roughly 25 solar masses shed their hydrogen envelopes through powerful radiation-driven winds and episodic mass loss, exposing first the products of CNO-cycle hydrogen burning (nitrogen-rich WN phase) and then the products of helium burning (carbon- and oxygen-rich WC/WO phases). Their powerful winds inject enormous mechanical energy and chemically enriched material into the surrounding interstellar medium, sculpting ring nebulae and contributing to Galactic chemical evolution.

Schema

Column	Type	Description	Sample	Null %
wr_number	object	Standard WR catalog number (e.g. 'WR 1', 'WR 140') from the VIIth Catalog of Galactic Wolf-Rayet Stars; primary identifier used throughout the literature	WR1	0.0%
spectral_type	object	Full WR spectral classification (e.g. 'WN4b', 'WC7+O5-8', 'WO2'); WN = nitrogen-sequence, WC = carbon-sequence, WO = oxygen-sequence; '+' indicates a spectroscopic binary companion	WN4b	0.0%
name	object	Alternative designation (usually HD number or other catalog ID); null for stars without a common alternative name	HD 4004	2.6%
ra_deg	float64	Right ascension, ICRS at Gaia DR2 reference epoch Ep=2015.5, in decimal degrees (0-360)	10.868291666666666	0.0%
dec_deg	float64	Declination, ICRS at Ep=2015.5, in decimal degrees (-90 to +90)	64.75983333333332	0.0%
parallax_mas	float64	Zero-point corrected Gaia DR2 parallax in milliarcseconds; many WR stars have negative or zero parallax due to faintness/crowding — distances are derived via a Bayesian method	0.314	0.0%
parallax_error_mas	float64	1-sigma uncertainty on Gaia DR2 parallax (mas)	0.04	0.0%
distance_kpc	float64	Distance from the Sun in kpc, derived from Gaia DR2 parallax using a Bayesian prior; null for stars where Gaia astrometry is too poor; most Galactic WR stars lie within 10 kpc	3.15	0.0%
distance_upper_error_kpc	float64	Asymmetric upper 1-sigma uncertainty on distance (kpc); distances are often asymmetrically uncertain because negative parallaxes give poorly constrained distances	0.47	0.0%
distance_lower_error_kpc	float64	Asymmetric lower 1-sigma uncertainty on distance (kpc)	0.36	0.0%
galactic_height_pc	int64	Perpendicular distance from the Galactic mid-plane in pc, calculated from distance and Galactic latitude; WR stars trace the thin disk, typically	z	< 200 pc
galactic_height_upper_error_pc	int64	Upper 1-sigma uncertainty on Galactic height (pc)	15	0.0%
galactic_height_lower_error_pc	int64	Lower 1-sigma uncertainty on Galactic height (pc)	12	0.0%
gaia_g_mag	float64	Gaia DR2 G-band (330-1050 nm) apparent magnitude; WR stars typically G = 8-18 mag; heavily reddened WR stars may be absent from Gaia	9.79	0.0%
gaia_bp_rp	float64	Gaia DR2 BP-RP colour index; WR stars are hot (blue) but often appear red due to interstellar extinction and emission lines; null where BP/RP photometry is unavailable	1.05	4.5%
astrometric_excess_noise	float64	Gaia DR2 astrometric excess noise in mas; elevated values indicate binarity, source confusion, or poor astrometric solution	0.0	0.3%
log_luminosity	float64	Log bolometric luminosity in solar units log(L/L_sun); WR stars: 10^5-10^6 L_sun (log L ~ 5.0-6.0); null for stars without a reliable distance or spectroscopic analysis	5.78	74.7%
error_flag	object	Quality or error flag from Rate & Crowther (2020) indicating issues with the spectral classification, photometry, or distance; null for clean entries	g	0.3%
ks_mag	float64	2MASS Ks-band (2.17 um) apparent magnitude; near-infrared photometry less affected by interstellar extinction than optical	7.48	5.3%
j_ks_color	float64	2MASS J-Ks colour index; traces near-infrared excess from free-free emission in WR winds; WC stars show strong IR excess	0.73	5.3%
h_ks_color	float64	2MASS H-Ks colour index; additional near-infrared wind emission diagnostic	0.38	5.3%
ks_extinction	float64	Ks-band extinction A_Ks in magnitudes, derived from comparison of observed and intrinsic colours; used to compute absolute magnitudes	0.3	5.3%
ks_abs_mag	float64	Absolute Ks-band magnitude of the WR star, corrected for extinction; useful for luminosity comparisons independent of optical reddening; null where distance or extinction is unknown	-5.4	5.3%
wr_subtype	object	Broad WR sequence derived from spectral_type: WN (exposing CNO-cycle products, WN2-WN11), WC (exposing He-burning products, WC4-WC9), WO (most evolved, WO1-WO4); null if spectral type is ambiguous	WN	2.9%
is_binary	bool	True if the spectral type contains '+', indicating a detected companion; WR+O binaries are important progenitors of compact object mergers	False	0.0%

Quick stats

• 380 Galactic Wolf-Rayet stars
• 220 WN (nitrogen sequence), 145 WC (carbon sequence), 4 WO (oxygen sequence)
• 64 spectroscopic binaries
• 380 with Gaia-based distance estimates (median 2.9 kpc)
• 96 with luminosity measurements

Usage

from datasets import load_dataset

ds = load_dataset("juliensimon/wolf-rayet-stars", split="train")
df = ds.to_pandas()

# WN vs WC distribution
print(df["wr_subtype"].value_counts())

# Nearest WR stars
nearest = df.nsmallest(10, "distance_kpc")[["wr_number", "spectral_type", "distance_kpc", "name"]]
print(nearest)

# Luminosity distribution by subtype
import matplotlib.pyplot as plt
for st in ["WN", "WC"]:
    sub = df[df["wr_subtype"] == st].dropna(subset=["log_luminosity"])
    plt.hist(sub["log_luminosity"], bins=15, alpha=0.6, label=st)
plt.xlabel("log(L/L_sun)")
plt.ylabel("Count")
plt.legend()
plt.title("Wolf-Rayet Luminosity Distribution")
plt.show()

Data source

https://vizier.cds.unistra.fr/viz-bin/VizieR-3?-source=J/MNRAS/493/1512

Related datasets

• juliensimon/bright-star-catalog

• juliensimon/gcvs-variable-stars

• juliensimon/pulsar-catalog

Citation

@dataset{wolf_rayet_stars,
  title = {Galactic Wolf-Rayet Stars},
  author = {juliensimon},
  year = {2026},
  url = {https://huggingface.co/datasets/juliensimon/wolf-rayet-stars},
  publisher = {Hugging Face}
}

License

CC-BY-4.0