juliensimon/pulsar-glitch-catalog

Jodrell Bank Pulsar Glitch Catalogue

Credit: NASA/JPL-Caltech

Part of a dataset collection on Hugging Face.

Dataset description

Comprehensive catalog of pulsar glitch events from the Jodrell Bank Centre for Astrophysics. Pulsar glitches are sudden spin-up events in neutron stars, thought to arise from angular momentum transfer between the superfluid interior and the solid crust.

During a glitch, the rotation frequency of the pulsar increases abruptly, typically by a fractional amount delta_nu/nu ranging from ~1e-11 to ~1e-5. The largest glitches (delta_nu/nu > 1e-6) are called "giant glitches" and are characteristic of young pulsars like the Vela pulsar, which glitches roughly every 2-3 years.

The physics of pulsar glitches provides a unique window into neutron star interior structure. The leading model invokes a reservoir of angular momentum stored in quantised vortices pinned to the inner crust lattice. As the star spins down electromagnetically, a rotational lag develops between the superfluid and the crust. When the Magnus force on pinned vortices exceeds the pinning force, vortices unpin catastrophically and transfer angular momentum outward, producing the observed spin-up. The cumulative glitch activity constrains the fractional moment of inertia of the superfluid component, which in turn constrains the neutron star equation of state.

This dataset is suitable for tabular classification tasks.

Schema

Column	Type	Description	Sample	Null %
bname	str	Pulsar B1950 designation (e.g. B0531+21 = Crab, B0833-45 = Vela); older naming convention, not all pulsars have a B-name	B1737-30	0.0%
jname	str	Pulsar J2000 designation (e.g. J0534+2200 = Crab, J0835-4510 = Vela); format encodes position in right ascension and declination	1740-3015	0.0%
n_glitches	int64	Total number of glitches recorded from this pulsar in the catalog; Vela-like pulsars glitch most frequently (~every 2-3 years)	38	0.0%
epoch_mjd	float64	Glitch epoch in Modified Julian Date (MJD = JD - 2400000.5); MJD 40000 ~ 1968, MJD 60000 ~ 2023	60980.0	0.0%
epoch_mjd_err	float64	1-sigma uncertainty on glitch epoch in days; typically days to weeks depending on timing cadence	5.0	6.5%
delta_nu_nu	float64	Fractional spin-up magnitude delta-nu/nu (dimensionless); typical range 1e-9 to 1e-5; large (Vela-class) glitches: ~1e-6; converted from catalog units of 1e-9	2.81e-06	0.3%
delta_nu_nu_err	float64	1-sigma uncertainty on delta_nu_nu (dimensionless)	3.0000000000000004e-09	2.7%
delta_nudot_nudot	float64	Fractional change in spin-down rate delta-nudot/nudot (dimensionless) after the glitch; typically negative (spin-down rate increases); converted from catalog units of 1e-3	0.00044	16.2%
delta_nudot_nudot_err	float64	1-sigma uncertainty on delta_nudot_nudot (dimensionless)	7.000000000000001e-05	16.9%
references	str	Bibliographic references for this glitch event; may cite multiple papers if the glitch was reported or reanalysed in multiple studies	Shaw+2025 [bak+25]; Also in Grover+25...	0.0%
epoch_datetime	datetime64[ns]	Glitch epoch as UTC datetime, derived from epoch_mjd; provided for convenience	2025-11-01 00:00:00	0.0%
is_large_glitch	object	True if delta_nu_nu > 1e-6 (giant/Vela-class glitch involving large angular momentum transfer from the superfluid interior); False for smaller glitches; null if delta_nu_nu is unmeasured	True	0.3%

Quick stats

• 728 glitch events
• 223 unique pulsars
• 194 large/giant glitches (delta_nu/nu > 1e-6)
• Largest glitch: 1647-4552 (delta_nu/nu = 6.50e-05)
• Most frequently glitching pulsar: 0537-6910 (65 glitches)

Usage

from datasets import load_dataset

ds = load_dataset("juliensimon/pulsar-glitch-catalog", split="train")
df = ds.to_pandas()

from datasets import load_dataset

ds = load_dataset("juliensimon/pulsar-glitch-catalog", split="train")
df = ds.to_pandas()

# Glitch size distribution
import matplotlib.pyplot as plt
sizes = df["delta_nu_nu"].dropna()
sizes[sizes > 0].hist(bins=50, log=True)
plt.xlabel("delta_nu/nu")
plt.ylabel("Count")
plt.title("Pulsar Glitch Size Distribution")
plt.show()

# Most glitching pulsars
top = df.groupby("jname").size().sort_values(ascending=False).head(10)
print(top)

# Giant glitches only
giants = df[df["is_large_glitch"] == True]
print(f"{len(giants):,} giant glitches")

Data source

https://www.jb.man.ac.uk/pulsar/glitches.html

Related datasets

• juliensimon/pulsar-catalog

• juliensimon/mcgill-magnetar-catalog

• juliensimon/gravitational-wave-events

• juliensimon/supernova-remnants

If you find this dataset useful, please consider giving it a like on Hugging Face. It helps others discover it.

About the author

Created by Julien Simon — AI Operating Partner at Fortino Capital. Part of the Space Datasets collection.

Citation

@dataset{pulsar_glitch_catalog,
  title = {Jodrell Bank Pulsar Glitch Catalogue},
  author = {juliensimon},
  year = {2026},
  url = {https://huggingface.co/datasets/juliensimon/pulsar-glitch-catalog},
  publisher = {Hugging Face}
}

License

CC-BY-4.0