juliensimon/silso-sunspot-number

SILSO Daily Sunspot Number

Credit: NASA/SDO

Part of a dataset collection on Hugging Face.

Dataset description

Daily total sunspot numbers from the World Data Center SILSO at the Royal Observatory of Belgium. This is the longest continuous scientific observation in history, with systematic daily records since 1818 and international coordination since 1981.

The International Sunspot Number is the primary index of solar activity, tracking the number of sunspots visible on the solar disk each day. Sunspots are temporary phenomena on the Sun's photosphere caused by magnetic flux concentrations. Their number follows an approximately 11-year cycle (the Schwabe cycle) that profoundly affects space weather, satellite operations, radio communications, and Earth's upper atmosphere.

Sunspots are regions where intense magnetic flux tubes (typically 0.1-0.3 T) emerge through the photosphere, inhibiting convective energy transport and creating dark spots roughly 1000-1500 K cooler than the surrounding ~5800 K surface. The daily sunspot number is computed using the Wolf formula (R = k(10g + s), where g is the number of sunspot groups, s is the total number of individual spots, and k is a station-dependent scaling factor), then combined across the observer network into a single international index.

The approximately 11-year Schwabe cycle drives variations in total solar irradiance (order 0.1%), extreme ultraviolet flux (factor of 10 or more), solar flare and CME rates, and the overall heliospheric magnetic field strength. These variations have direct consequences for satellite drag, HF radio propagation, radiation exposure, and modulation of galactic cosmic ray flux at Earth.

The SILSO Version 2.0 series, released in 2015, recalibrated the entire historical record back to 1818 to correct for discontinuities introduced by changes in the reference observer. This makes it the most homogeneous long-baseline solar activity record available, spanning over 200 years of daily observations and encompassing roughly 19 complete solar cycles.

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

Schema

Column	Type	Description	Sample	Null %
date	datetime64[us]	Observation date (UTC); SILSO daily records begin 1818-01-01.	1818-01-01 00:00:00	0.0%
decimal_date	float64	Fractional year representation of the date (e.g., 2024.5 ~ July 2 2024); useful for continuous time-series computations.	1818.001	0.0%
sunspot_number	Int64	International Sunspot Number (ISN v2.0) -- daily count of sunspots on the solar disk using the Wolf formula (R = k(10g + s)); solar minimum: ~0-10; solar maximum: 150-300+; approximately 11-year cycle; null for days with no observation; v2.0 values are ~1.6x the previous v1.0 series.	65	4.3%
std_dev	float64	Standard deviation of individual station observations around the network mean for that day; larger values indicate disagreement between observers or complex sunspot groups.	-1.0	0.0%
n_observations	Int64	Number of SILSO network observing stations that contributed valid measurements that day; typical range 10-50; lower values (early historical records) reflect fewer contributing observers.	0	0.0%
is_provisional	bool	True if the value has not yet been finalized by SILSO (recent ~30 days); provisional values may be revised when additional observer data arrives; False for definitive historical values.	True	0.0%

Quick stats

• 76,091 daily records (1818-01-01 to 2026-04-30)
• All-time maximum: 528 on 1870-08-26
• 75,971 provisional values
• Solar Cycle 25 (current): peak so far 290, mean 86.5

Usage

from datasets import load_dataset

ds = load_dataset("juliensimon/silso-sunspot-number", split="train")
df = ds.to_pandas()

from datasets import load_dataset
import pandas as pd

ds = load_dataset("juliensimon/silso-sunspot-number", split="train")
df = ds.to_pandas()

# Plot solar cycles
import matplotlib.pyplot as plt
df["date"] = pd.to_datetime(df["date"])
monthly = df.set_index("date").resample("MS")["sunspot_number"].mean()
monthly.plot(figsize=(14, 4), title="Solar Cycles - Monthly Mean Sunspot Number")
plt.ylabel("Sunspot Number")
plt.show()

# Current solar cycle 25
sc25 = df[df["date"] >= "2019-12-01"]
print(f"Cycle 25 max so far: {sc25['sunspot_number'].max()}")

Data source

https://www.sidc.be/SILSO/

Related datasets

• juliensimon/solar-flare-events

• juliensimon/geomagnetic-kp-index

• juliensimon/dst-index

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{silso_sunspot_number,
  title = {SILSO Daily Sunspot Number},
  author = {juliensimon},
  year = {2026},
  url = {https://huggingface.co/datasets/juliensimon/silso-sunspot-number},
  publisher = {Hugging Face}
}

License

CC-BY-4.0