090°Open data
Open datasets, fully documented — searchable here, and readable by any LLM.
1 datasets
NOAA Tidal Water-Level Anomaly Intelligence
Every NOAA coastal water-level station's observed-minus-predicted tidal residuals over the trailing 24 complete UTC hours, transformed into a ranked surge/drawdown intelligence digest. Each row is one station: hourly residual statistics (mean, max, min, std in meters; observed minus astronomical prediction, so the ordinary tide is removed), counts of surge/drawdown hours at +/-0.3 m, sustained-event flags (3-hour running mean at +/-0.5 m), the residual least-squares trend (m/hour) flagging building surge situations, and a documented 0-100 anomaly_score ranked as anomaly_rank with tiers extreme/high/elevated/normal. Stations need >= 18 of 24 hours with both observed and predicted values; stations with no hourly predictions (mostly Great Lakes gauges) are excluded by design. Primary key: station_id. Cadence: daily; as_of is the window end (last complete UTC day), so identical input produces an identical content hash. Caveats: observations are preliminary and may be revised by NOAA; anomalies are non-astronomical water-level deviations, not weather forecasts. NOAA tide data is U.S. federal public domain — commercial_use = yes. Sample use: filter anomaly_tier in ('extreme','high') for stations seeing real surge right now, or track one station's residual_trend across snapshots for a building surge.
Once today's free allowance is used up, AI features can run on your own provider account.
Kept in this browser tab only (cleared when you close it) and sent with each AI request. Our servers use it for that request and never store or log it.