Storms, “Supermoons,” and a persistent ~0.2 Hz ground hum near New Madrid (paper attached)

[David Thomson]

Hi PSN folks,

I’m sharing a short paper (PDF attached) about a simple observation that turned into a multi-year rabbit hole: a persistent, narrowband “hum” around 0.2 Hz (one wiggle every ~5 seconds) that shows up on my Illinois home station and on nearby regional stations. The paper looks at how that band behaves during big storms and around perigee full moons (“supermoons”), and what—if anything—that might mean for the New Madrid Seismic Zone (NMSZ).

What is the 0.2 Hz band?

It sits in the ocean microseism range. Even far inland, ocean waves banging into each other and the coast produce low-frequency vibrations that travel through the crust. Strong storms boost this background “hum.” Tides can also modulate the loads on the oceans and the solid Earth a little bit.

What we did (plain-English version)

• Summarized years of seismic data by hour and by frequency, then focused on the 0.18–0.24 Hz slice.

• Built a simple hourly tide index for the station location.

• Marked time windows for tropical cyclones, nor’easters, and supermoons (with ±72 h padding to catch swell and lingering effects).

• Asked three basic questions:

  1. Timing: Do the biggest 0.2 Hz bumps line up with big storms and supermoon windows?

  2. Tide phase: During those windows, do high 0.2 Hz hours prefer certain tide phases? (We show this with circular “rose” plots.)

  3. Lock-in: As the amplitude grows, does the peak frequency converge (“lock in”) toward 0.200 Hz, as you’d expect from a resonant mode?

What we found

• Storm link: The 0.2 Hz band generally strengthens from late summer through winter, and some of the largest excursions line up with major storms that would be expected to drive strong microseisms.

• Tidal fingerprint (weak but present): On average, the tides only have a small effect, but it’s detectable—we see a modest coherence at the main diurnal/semidiurnal lines and small lags (the ground responds a little after the tide).

• Phase preference during events: When we only look at hours that are both high in 0.2 Hz and inside event windows, the tide-phase distributions are not uniform. For example, the supermoon and nor’easter subsets favor certain phases, while the non-event (“quiet”) background looks much flatter.

• Resonant lock-in: At higher amplitudes, the apparent peak frequency within 0.18–0.24 Hz clusters tightly near 0.200 Hz. That’s the kind of behavior you see when a system’s natural mode gets strongly excited.

What this might mean (and what it does not claim)

• A working hypothesis (not a prediction engine): in a water-rich, fractured/karstic crust like parts of the NMSZ, slow cyclic loads from storms and tides could slightly modulate stress and pore pressure. When the system is already close to failure, that added nudge might influence timing. This could also help explain reports of distant rumbling/booms before strong shaking in 1811–1812—i.e., shallow layers ringing while deeper structures are being nudged.

• What we are not saying: storms or “supermoons” cause earthquakes. The coupling is weak. Many other factors dominate. This is about conditional loading in a near-critical system, not deterministic forecasting.

Why share with PSN?

This group is uniquely positioned to test or falsify this. A few concrete ways to help:

• Replicate the 0.18–0.24 Hz envelope on long records from quiet backyard stations or small arrays—especially across the central U.S.

• If you have (or can add) infrasound or outdoor microphones, note whether “mystery booms” line up with high 0.2 Hz hours.

• Compare against local well levels / river stage and barometric pressure where available.

• If you want to try this, I can share a tiny Python script that ingests a folder of miniSEED/SAC and spits out the hourly 0.2 Hz envelope.

What’s next

The paper outlines an immediate next step: a simple “wetness/loading index” that blends river stage near the Mississippi/Ohio confluence and coastal wave power (Gulf + Atlantic) to see how much of the 0.2 Hz variance we can explain. We’re also planning to keep an eye on the 2025–2026 perigee-syzygy cluster while tracking storm seasons—again, for monitoring, not prediction.

How to read the figures (no jargon)

• Overlays: show the 0.2 Hz strength and the tide index on the same timeline; year-by-year panels make the seasonal pattern obvious.

• Rose plots: circular histograms of tide phase at the times when the 0.2 Hz band is strong. A lopsided rose means “more of the strong hours happen around that phase.”

• Lock-in plots: scatter/histograms where the x-axis is frequency and you can see high-amplitude points huddling around 0.200 Hz.

The attached PDF has the plain-English methods, figure captions, and the exact command lines used to build the plots from simple CSVs. If you’re curious or want the small helper scripts, reply here and I’ll share them.

Thanks for taking a look—and for any replication attempts, critiques, or ideas to strengthen (or disprove!) the hypothesis.

—
David W. Thomson
Private researcher, Illinois
(PSN member; happy to follow up off-list if preferred)