FIELD EVIDENCE / PART I · SANCTSOUND OC01 · APRIL 2026
An anomaly.A vessel.
The model said vessel passage. The dataset said ambient. Ten audio chunks from a NOAA SanctSound recording led us from an apparent classification error to an independent investigation.
HOW IT UNFOLDED
Three steps from anomaly to proof.
Miss analysis surfaced a cluster
Running the full v6 validation set (1,145 samples) through the production CNN revealed 16 misses out of 1,112 clean samples — 98.6% accuracy. But 10 of those 16 came from a single source: SanctSound OC01. All in the same direction: model said ship, label said ambient. A random failure mode scatters. This clustered.
Every miss traced to one recording
All 10 flagged chunks came from a single 6-hour FLAC file recorded at Olympic Coast on 2019-03-09, in two tight windows at 12:14 UTC and 12:33–12:44 UTC. Measured against the 50 correctly-labelled ambient chunks from that same recording, they are markedly more tonal: spectral flatness 0.17 against 0.40 (p < 0.0001), with the peak-energy bin moving from 10 Hz up to 39 Hz, into the blade-rate region where propeller harmonics sit. Narrowband structure at low frequency is vessel acoustics, not background noise.
AIS named the ship
GFW vessel tracking was queried for OC01 (48.40°N, 124.70°W) on that day. 38 vessels were broadcasting AIS within 50 km. At ≤ 10 km during the suspect 12:00–13:00 UTC window: JOSCO HUIZHOU — a cargo vessel flying Hong Kong flag, present at 12:00 UTC, exactly when our CNN flagged the signature. OC01 sits at the western mouth of the Strait of Juan de Fuca, one of the world's busiest shipping corridors.
ACOUSTIC FEATURES
The signature that doesn't lie.
Per-chunk acoustic features, computed without any neural network and compared against the 50 correctly-labelled ambient chunks from the same OC01 recording. Same hydrophone, same deployment, same gain, so the only variable is what was in the water.
| Feature | Flagged chunks (n=10) | Ambient control (n=50) | Significance | Interpretation |
|---|---|---|---|---|
| Spectral flatness · 50–500 Hz | 0.17 | 0.40 | p < 0.0001 | Tonal, not broadband. Narrowband lines, not a raised noise floor |
| Peak frequency | 39.0 Hz | 10.2 Hz | p = 0.013 | Peak energy moves up into the blade-rate region, 5–50 Hz |
Mann-Whitney U, two-sided. Absolute band level did not separate the two groups (engine band −38.4 dB against −41.2 dB, p = 0.64), so the case rests on tonality and on the narrow-band PSD below, not on loudness. Both features are computed on a 256-bin mel spectrum spanning 0–8 kHz. Part II shows what happens when the same comparison is run across four different recordings instead of within one.
AIS CROSS-REFERENCE
The western mouth of the Strait of Juan de Fuca.
38 vessels were broadcasting AIS within 50 km of OC01 on 2019-03-09. JOSCO HUIZHOU was confirmed within 10 km at 12:00 UTC — the closest-point-of-approach physically consistent with the acoustic cluster at 12:33 UTC.
| Vessel | Class | Flag | Window (UTC) | Distance |
|---|---|---|---|---|
| JOSCO HUIZHOUCONFIRMED ≤ 10 KM | Cargo | HKG | 10:00 – 15:00 | ≤ 10 km at 12:00 UTC |
| ULTRA JAGUAR | Cargo | MHL | 12:00 – 16:00 | ≤ 50 km |
| HYUNDAI GRACE | Cargo | MHL | 10:00 – 14:00 | ≤ 50 km |
| WIND SONG | Fishing | USA | 05:00 – 17:00 | ≤ 50 km |
| VISHVA ANAND | Cargo | IND | 08:00 – 12:00 | ≤ 50 km |
| MONING | Cargo | PAN | 08:00 – 12:00 | ≤ 50 km |
| CAPE MCKAY / DENISE FOSS | Other | USA/CAN | All day | ≤ 50 km |
Verification
Six methods. One conclusion.
CNN classification
10 chunks flagged at 0.53-0.97 confidence, contiguous in the 12:14 and 12:33-12:44 UTC windows
Acoustic feature analysis
Spectral flatness 0.17 against 0.40 for ambient from the same recording (p < 0.0001): tonal, not broadband
AIS day-scale <= 50 km
38 vessels broadcasting, 7+ cargo active in suspect window
AIS tight <= 10 km
JOSCO HUIZHOU confirmed within 10 km at 12:00 UTC
Narrow-band PSD
+11.98 dB mean excess across 28-37 Hz, peaking at +13.17 dB at 31.25 Hz: cargo propeller blade-rate harmonics
Listening verification
Vessel-like acoustic rumble audible in suspect window
NARROW-BAND PSD
Propeller harmonics, measured.
Welch's PSD on the 12-minute suspect window vs a 5-minute quiet ambient control. The excess energy clusters tightly in 28–37 Hz — the cargo-ship propeller blade-rate harmonic band (80–120 RPM × 4–6 blades).
Random noise is broadband. Biological and geological events are broadband. This is narrow-band tonal at specific frequencies — a propeller.
Why it matters
Every passive acoustic archive can be improved.
Data quality at scale
SanctSound, NOAA NRS, and similar archives span thousands of hours of passive recordings. Each can be cross-referenced against GFW AIS automatically — surfacing vessel events that human labeling missed, without manual review.
Cross-modal verification loop
CNN flags acoustic anomaly → AIS confirms vessel presence → label is corrected. This loop runs entirely without human intervention, creating a self-improving dataset pipeline that gets more accurate over time.
Zero missed alerts
In production, these events route to Gemma for evaluation. A broadcasting vessel with no MPA proximity is graded LOW and logged. No false alarms reach operators; no real vessels are missed.
Publishable methodology
A tight AIS proximity heuristic (≤ 10 km, ± 30 min) combined with narrow-band PSD verification constitutes a reproducible, defensible method for AIS-corrected acoustic label generation.
VERDICT · AUDIT-GRADE
Six independent methods, one answer.
The dataset was wrong. Ocean Sentinel caught it.