juliensimon/space-weather-indices

Space Weather Indices (Kp, Ap, F10.7)

Credit: NASA

Part of a dataset collection on Hugging Face.

Dataset description

Daily geomagnetic and solar activity indices since 1957 from NOAA SWPC via CelesTrak. Includes Kp/Ap geomagnetic indices, F10.7 solar radio flux, and international sunspot numbers.

These indices together form the essential parameter set for characterizing the state of the heliosphere and its coupling to the terrestrial environment. The Kp index (quasi-logarithmic, 0-9 scale, 3-hourly) captures planetary-scale geomagnetic disturbances driven by solar wind-magnetosphere interactions, while the Ap index (its linearized daily equivalent in nanotesla) serves as the standard geomagnetic input to atmospheric density models. The F10.7 solar radio flux (measured daily at 2800 MHz in Penticton, Canada) is the primary proxy for solar extreme ultraviolet (EUV) radiation that heats the thermosphere -- the atmospheric layer where most satellites experience drag.

For operational space weather applications, this dataset provides the complete set of inputs required by the major atmospheric density models: NRLMSISE-00 (F10.7, F10.7bar, Ap), JB2008 (F10.7 plus supplementary indices), and DTM (F10.7, Kp). The storm classification (G1-G5) derived from Kp thresholds is the same scale used in NOAA space weather alerts.

This dataset is suitable for tabular regression, time-series forecasting tasks.

Schema

Column	Type	Description	Sample	Null %
date	datetime64[us]	Observation date in UTC	1957-10-01 00:00:00	0.0%
bartels_rotation	int64	Bartels Solar Rotation Number -- a 27-day cycle count since 1832, used to align solar data with the Sun's synodic rotation period as seen from Earth	1700	0.0%
bartels_day	int64	Day within the current Bartels rotation cycle (1-27)	19	0.0%
kp_0000	float64	3-hourly Kp geomagnetic index for the 00:00-03:00 UT interval (quasi-logarithmic, 0-9 scale); measures planetary-scale magnetic field disturbances	43.0	0.7%
kp_0300	float64	3-hourly Kp geomagnetic index for the 03:00-06:00 UT interval	40.0	0.7%
kp_0600	float64	3-hourly Kp geomagnetic index for the 06:00-09:00 UT interval	30.0	0.7%
kp_0900	float64	3-hourly Kp geomagnetic index for the 09:00-12:00 UT interval	20.0	0.7%
kp_1200	float64	3-hourly Kp geomagnetic index for the 12:00-15:00 UT interval	37.0	0.7%
kp_1500	float64	3-hourly Kp geomagnetic index for the 15:00-18:00 UT interval	23.0	0.7%
kp_1800	float64	3-hourly Kp geomagnetic index for the 18:00-21:00 UT interval	43.0	0.7%
kp_2100	float64	3-hourly Kp geomagnetic index for the 21:00-24:00 UT interval	37.0	0.7%
kp_sum	float64	Sum of the eight daily 3-hourly Kp values (0-72); daily aggregate measure of geomagnetic activity	273.0	0.7%
ap_0000	float64	3-hourly Ap index for the 00:00-03:00 UT interval -- linearized equivalent of Kp in nanotesla units; used as input to atmospheric density models (NRLMSISE-00, JB2008)	32.0	0.7%
ap_0300	float64	3-hourly Ap index for the 03:00-06:00 UT interval	27.0	0.7%
ap_0600	float64	3-hourly Ap index for the 06:00-09:00 UT interval	15.0	0.7%
ap_0900	float64	3-hourly Ap index for the 09:00-12:00 UT interval	7.0	0.7%
ap_1200	float64	3-hourly Ap index for the 12:00-15:00 UT interval	22.0	0.7%
ap_1500	float64	3-hourly Ap index for the 15:00-18:00 UT interval	9.0	0.7%
ap_1800	float64	3-hourly Ap index for the 18:00-21:00 UT interval	32.0	0.7%
ap_2100	float64	3-hourly Ap index for the 21:00-24:00 UT interval	22.0	0.7%
ap_avg	float64	Daily average Ap index; geomagnetic storm threshold at Ap >= 50; key input for satellite drag models	21.0	0.7%
cp	float64	Daily Character Figure Cp (0.0-2.5) -- a qualitative measure of the overall level of geomagnetic disturbance for the day	1.1	0.7%
c9	float64	Converted Cp on a 0-9 integer scale; derived from Cp for easier comparison with Kp	5.0	0.7%
sunspot_number	int64	International Sunspot Number (ISN) -- daily count of sunspot groups and individual spots; primary indicator of the ~11-year solar activity cycle	334	0.0%
f107_obs	float64	Observed 10.7 cm (2800 MHz) solar radio flux in solar flux units (SFU, 1 SFU = 10^-22 W/m2/Hz); measured at Penticton, Canada; primary proxy for solar EUV radiation that heats the thermosphere	269.3	0.0%
f107_adj	float64	F10.7 solar radio flux adjusted to 1 AU distance; removes the effect of Earth's orbital eccentricity for physical comparisons	269.8	0.0%
f107_data_type	str	Data source flag: OBS (observed), INT (interpolated from observations), PRD (predicted), PRM (predicted monthly mean)	OBS	0.0%
f107_obs_center81	float64	81-day centered running average of observed F10.7; smooths the 27-day solar rotation modulation to represent background EUV irradiance level	266.6	0.0%
f107_obs_last81	float64	81-day trailing (last 81 days) running average of observed F10.7; available in near-real-time unlike the centered average	230.9	0.0%
f107_adj_center81	float64	81-day centered running average of 1-AU-adjusted F10.7	266.8	0.0%
f107_adj_last81	float64	81-day trailing running average of 1-AU-adjusted F10.7	235.5	0.0%
is_storm	bool	True when daily average Ap >= 50, indicating a geomagnetic storm day; storms cause elevated satellite drag, GPS errors, and power grid disturbances	False	0.0%
storm_level	str	NOAA G-scale classification based on maximum 3-hourly Kp: G1 (Kp=5, minor), G2 (Kp=6, moderate), G3 (Kp=7, strong), G4 (Kp=8, severe), G5 (Kp=9, extreme); null for non-storm days	G5	1.4%
data_type	str	Simplified data type: 'observed' (OBS/INT -- based on measurements) or 'predicted' (PRD/PRM -- forecast values)	observed	0.0%

Quick stats

• 25,032 observed days (1957-10-01 to 2026-04-13)
• 613 geomagnetic storm days (Ap >= 50)
• 24,913 severe storms (G3+)
• Strongest storm: Ap=280 on 1960-11-13

Usage

from datasets import load_dataset

ds = load_dataset("juliensimon/space-weather-indices", split="train")
df = ds.to_pandas()

from datasets import load_dataset

ds = load_dataset("juliensimon/space-weather-indices", split="train")
df = ds.to_pandas()

# Only observed data (exclude predictions)
observed = df[df["data_type"] == "observed"]

# Geomagnetic storms
storms = df[df["is_storm"] == True].sort_values("ap_avg", ascending=False)

# Solar cycle visualization
import matplotlib.pyplot as plt
df["year"] = df["date"].dt.year
yearly_ssn = df.groupby("year")["sunspot_number"].mean()
yearly_ssn.plot(figsize=(12, 4))
plt.ylabel("Mean Sunspot Number")
plt.title("Solar Cycle from Daily Sunspot Numbers")
plt.show()

# F10.7 flux trend (drives atmospheric drag)
df.set_index("date")[["f107_adj"]].rolling(81).mean().plot()
plt.ylabel("F10.7 (SFU)")
plt.title("81-day Running Mean F10.7 Solar Flux")
plt.show()

Data source

https://celestrak.org/SpaceData/

Update schedule

Daily at 11:00 UTC via GitHub Actions

Related datasets

• juliensimon/solar-flare-events

• juliensimon/space-track-satcat

• juliensimon/space-track-tle-history

• juliensimon/neo-close-approaches

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{space_weather_indices,
  title = {Space Weather Indices (Kp, Ap, F10.7)},
  author = {juliensimon},
  year = {2026},
  url = {https://huggingface.co/datasets/juliensimon/space-weather-indices},
  publisher = {Hugging Face}
}

License

CC-BY-4.0