Solar Forcing of Terrestrial Geophysics — The Definitive Test

Author: Claude (TerraPulse Lab)

Status: Complete — V2 with 6-month DSCOVR data

Created: 2026-03-24

Updated: 2026-03-25

Verdict (V2): Signal is real but tiny — r=0.09 at hourly resolution, vanishes at daily

The Question

Do solar flares, CMEs, and the solar cycle directly drive earthquakes,

volcanic eruptions, and severe weather? We test this at three timescales

using the most comprehensive cross-domain dataset assembled in TerraPulse.

Data Sources

Results

Timescale 1: LONG-TERM (Solar Cycle, 1818-2026)

The solar cycle does not modulate geophysical event rates.

Zero significant correlations across 205 years of sunspot-eruption overlap,

74 years of sunspot-tornado overlap, and 6 years of sunspot-earthquake overlap.

Eruptions show an 8.2-year spectral peak, not the 11-year solar cycle.

Timescale 2: MEDIUM-TERM (Days, Mar 2026)

One significant result: Solar flux anti-correlates with earthquake count

(r=-0.41, p=0.012). Higher solar flux → fewer earthquakes. This is

counterintuitive — but may reflect that high solar flux (solar max)

corresponds to a different magnetospheric coupling state.

High Kp days average 6,366 earthquakes vs 3,619 on low Kp days —

suggestive (1.76x ratio) but not significant with only 17 days of data.

Timescale 3: SHORT-TERM (Hours, Mar 2026)

Two highly significant results at the hourly scale:

1. Solar wind speed → earthquake count: r=+0.54, p<0.00001.

Faster solar wind = more earthquakes. This is the strongest

solar-terrestrial coupling signal in our entire dataset.

1. Bz → earthquake count: r=-0.36, p=0.003.

Negative Bz (southward interplanetary magnetic field) = more earthquakes.

Southward Bz is exactly what drives geomagnetic storms — it opens Earth's

magnetopause and allows solar wind energy into the magnetosphere.

X-ray flux (the flare itself) is marginal at p=0.10 — the flare alone

doesn't trigger earthquakes, but the **solar wind and magnetic field

that arrive hours later** do.

Interpretation: The Timescale Paradox

V1 Results (74 hours, March 2026 only)

The initial analysis found strong correlations:

• Solar wind → earthquakes: r=0.54, p<0.00001
• Bz → earthquakes: r=-0.36, p=0.003

V2 Results (6 months, 1800+ hours from NCEI DSCOVR archive)

With 120 days and 1,800 hours of data, the picture changed dramatically:

Daily resolution (120 days):

Nothing survives at daily resolution. High-wind days average 712 eq/day vs 719 for low-wind days — essentially identical.

Hourly resolution (1,800 hours):

The signal is real but tiny. r=0.09 at hourly resolution (p=0.0001) — statistically significant with 1,800 data points, but explains less than 1% of variance. The signal decays with lag and vanishes by 24 hours, consistent with a transient electromagnetic coupling.

Bz shows no hourly signal (r=0.03, p=0.19) — only the solar wind speed component matters.

Interpretation: The V1-to-V2 Story

Why did r=0.54 shrink to r=0.09?

The V1 analysis (74 hours in March 2026) likely captured a single geomagnetic event where elevated solar wind coincided with above-average earthquake activity. With 1,800 hours of data spanning 6 months, this coincidence dilutes into the baseline noise.

The effect is real but negligible. r=0.09 means solar wind speed explains 0.8% of hourly earthquake variance. For comparison, tidal forces (Moon+Sun gravitational) explain ~1-2% of seismicity variance in the literature. Both are statistically detectable with enough data but practically insignificant for forecasting.

The honest conclusion: Solar wind has a statistically significant but physically negligible influence on earthquake rate. It is not a useful predictor. The independence assumption is almost correct for this pair — the coupling exists but is too weak to matter.

Key Finding for the Big Paper

Solar wind speed has a statistically significant (p=0.0001) but

physically negligible (r=0.09) association with hourly earthquake

count over 6 months. The V1 result (r=0.54) was a single-event

artifact. This underscores the importance of multi-month baselines

for solar-terrestrial coupling claims — a lesson the field has

learned repeatedly over 50 years of null results.

Next Steps

• Separate by earthquake depth — do shallow quakes show stronger coupling?
• Test during specific geomagnetic storms (Kp≥7 events) as natural experiments
• Compare with tidal forcing effect size as calibration
• This result argues AGAINST a solar forcing paper — the honest finding is the null

Visualizations

• Three-timescale dashboard

References

• Bakhmutov, V. & Sedova, F. (2020). Solar activity–seismicity correlations. Geofizicheskiy Zhurnal.
• Love, J.J. & Thomas, J.N. (2013). Insignificant solar-terrestrial triggering of earthquakes. GRL.
• Marchitelli, V. et al. (2020). On the correlation between solar activity and large earthquakes worldwide. Scientific Reports.
• Urata, N. et al. (2018). Geomagnetic Kp index and earthquakes. NHESS.