Catches
These are the earthquakes the station has recorded and that the USGS catalog confirms — magnitude, location, depth and origin time are the catalog’s; the waveform, the peak and the timing are ours. A single geophone detects; a network confirms, and every event here was also recorded by professional stations, which is what makes it an earthquake and not a truck.
Each image is the raw 1–15 Hz trace with its smoothed envelope, and a spectrogram below it: an earthquake arrives as a burst that starts at low frequency and stays there, because the ground filters out the high frequencies on the way; a local disturbance is broadband and impulsive. That difference is how the detector tells them apart.
How far can a 4.5 Hz geophone on a garage floor hear? The rings are the predicted reach by magnitude, inverted from the same amplitude model the harvester uses and corrected by what the confirmed catches actually measured. Three rings per magnitude: the inner is a noisy afternoon, the outer a quiet night. Dots are the confirmed catches — 28 of them, out to 89 km; the dashed line is the furthest one; the X is an M3.4 at 348 km that was looked for and not seen, the one hard constraint on the far field. Corrected for what the catches measured, the station reads about 1.8× quieter than the textbook California attenuation predicts.
Everything past the dashed line is extrapolation. Beyond ~150 km what survives is low-frequency Lg energy, exactly where this geophone is deaf, so the outer rings are upper bounds. The map is regenerated whenever the catalog comparison is re-run.
00:22:31 UTC · 38.821°N 122.843°W · depth 1.7 km · 45 km NNW
- Peak 79 µV at 17:22 on a weekday afternoon, through the daytime traffic floor. The P onset landed on the predicted +9.0 s to within a sample.
- Recorded with the station’s original Python acquisition loop, a few hours before it was replaced by the C reader that now owns the digitizer.
15:30:04 UTC · 37.755°N 122.150°W · depth 5.5 km · 88 km SSE, on the Hayward fault (first reported as M4.1)
- The furthest confirmed catch, nearly twice the previous record. Peak 503 µV at +30 s, with the energy arriving as a long train of S and surface waves rather than a sharp onset — what 88 km of crust does to a signal.
- The Hayward and Rodgers Creek faults are one connected system; this is the neighbour to the south announcing itself.
- A second event of M2.8 followed 37 minutes later and was recorded too. Both were used to calibrate this station against a USGS strong-motion instrument 1.6 km away.
10:28:21 UTC · 43 km NNW · the geothermal field
- Peak 227 µV. The Geysers is the world’s largest geothermal field and produces several small earthquakes a day, driven by the wastewater injected to recharge the steam reservoir — some of it Santa Rosa’s own.
- That makes it a permanent calibration source 45 km away: same path every time, a new event every few days.
21:35:14 UTC · 45 km NNW
- Peak 159 µV. The workhorse event: its waveform was scaled and superposed onto recorded cultural noise to measure how much sensitivity the station loses while someone is dragging trash cans past it (about two thirds of a magnitude unit, as it turns out).
- Compared against the strong-motion station 1.6 km away, it says this geophone reads about 3× quieter than its nameplate — the calibration now in use.
02:40:06 UTC · 38.777°N 122.936°W · depth 5.9 km · 46 km NW, on the Maacama fault
- The biggest event yet. Peak 1,406 µV in 1–15 Hz against a ~1.5 µV floor, an ~80 s coda, and a detector ratio of 8,535 — the previous record was 645.
- Not remotely clipped: the whole event used ~3.5 % of the digitizer’s range.
- First arrival at +9.06 s. That was 1.4 s later than the textbook P-velocity predicted, and with four other events the same day it pinned the station’s own crustal velocity at 5.19 km/s — the number every prediction on this site now uses.
- Four more confirmed events the same day: three aftershocks at 46 km and an M1.5 near Angwin at 29 km.
06:29:25 UTC · 38.798°N 122.781°W · depth 3.5 km · 41 km NNW
- The second confirmed event, two days after the first, and the first from The Geysers. Onset at +8.15 s — one of the five points that fixed the station’s P velocity.
11:31:41 UTC · 38.507°N 122.435°W · depth 6.2 km · 18 km ENE, the Rodgers Creek / Maacama step-over
- The first confirmed earthquake, five days after the station moved to the garage. A clean, impulsive local event: sharp P at +3.97 s, peak 117 µV, signal-to-noise ~85×.
- Close enough that P and S are only ~2.4 s apart and merge into one burst — there is no independent S–P distance from one trace; the 18 km is the catalog’s.
- Also recorded by BK.CMB (broadband, 185 km) and CE.68327 (Santa Rosa strong-motion). Their picks are what located it.
not detected">~1,140 km E
- The other edge of the envelope. At the predicted regional arrival times the trace is at 1× its own noise; the two bright bursts on the image are local — broadband to 25 Hz, which nothing 1,140 km away can deliver.
- Together with an M3.4 at 348 km (Toms Place, 2026-07-30) that was also looked for and not seen, this is what puts the far edge on the map above.