Astronomical prediction
The tide table gives the expected level from regular tidal cycles, measured above the station’s MLLW reference.
Selected NOAA station · regional station context
Checking current water conditions…
Latest 24-hour average difference
—
Observed water level minus astronomical tide prediction.
Waiting for this station’s latest NOAA observations.
A what-if, not a forecast. Check NWS advisories.
Waiting for source timestamps.
NOAA station observations ↗Upcoming high tides
Loading this station’s next notable high tides…
Regional station heights are not Newport flood depths.
NOAA high-tide predictions ↗High-tide comparison
Waiting to compare with Oct 27
Selected high tide + latest extra water, compared with the same station’s observed high.
Scenario station: Los Angeles · NOAA 9410660
Checking three-day what-if bands…
Fixed historical residual-change bands require fresh current residuals and a valid selected tide, regardless of its date.
Feet above MLLWSame station and datum
MLLW is a fixed reference based on the average lower low tide, not ground level.
Observed referenceOct 27, 2026
Selected tideNov 25, 2026
Choose a high tide
November 25, 2026
Loading station predictions…
——
Slider shows a 91-day window. A date outside it opens a new window, within NOAA availability.
Jump to a high tide
Loading event dates…
Highest predicted days in the next 90 days; elevated tides can span several days.Retrieving NOAA high/low predictions…
Selected-day NOAA prediction source ↗
Why it’s happening
The tide table is only part of the water level.
Illustrative positive-residual example at one timestamp. Shapes are not to scale.
Combined extra water
The residual is the difference between observed and predicted water. It can also be negative.
Observed water level
At the same station and instant, the prediction plus the residual equals the measured water level.
MLLW is the reference zero, not an extra layer. No separate sea-level-rise amount is added on top.
Weather, seasonal patterns, broader ocean variability and longer-term changes can all contribute to the residual. These data do not isolate Kelvin waves or separately measure long-term sea-level rise.
Tidal residual trend
Daily mean7-day trailing average
Water above the predicted tide (inches)
Fetching the history from NOAA…
Move across the chart, tap a day, or use the date slider for details.
Custom dates and data detailsCalifornia calendar days · gaps are not interpolatedDownload daily CSV ↓
How this works
Southern California high tides and coastal flooding: regional context from two NOAA gauges, not local flood forecasts for Los Angeles, Orange or San Diego counties.
How this works
How to use this site
Choose a high tide to see where its station height might be given current tidal residuals: the astronomical tide plus the latest 24-hour average observed-minus-predicted difference. The gray bands add a historical reference for changes in that average over the next three days, even if the selected tide is months away. This is a what-if under current conditions, not a forecast for that date.
Tidal residuals fluctuate. Check back often in the days leading up to the high-tide event for an updated view of current conditions. Consult official forecasts and advisories for decisions. Station heights above MLLW are not local flood depths.
Data sources and calculations
Chart observations and astronomical predictions come from the selected NOAA station, using feet relative to that station’s MLLW and matching GMT timestamps. Long Beach history starts in April 2026; choose Los Angeles for November 2024 onward. The station series are never spliced. Nearby gauges can share datum controls, so their agreement is not fully independent confirmation. The chart, current residual and scenarios all use the selected station. Los Angeles is the default; an explicit station choice is remembered on this browser. Each residual is (observed − predicted) × 12 inches. Daily means, minima and maxima use America/Los_Angeles calendar days, including daylight-saving changes.
Days need at least 90% of expected six-minute matches to display a mean. Today uses elapsed intervals and is labeled partial. The trailing trend requires seven consecutive eligible completed days. Missing data breaks the line. Tolerance failures and verified inferred values are excluded; quality screening is applied before averages are shown. NOAA preliminary data can change. For the October event reference only, an expected-water-level-limit flag can be retained as an explicitly provisional reference if the peak window otherwise has complete readings without flat/rate faults. Missing readings or instrument-fault flags near the peak leave its comparison unavailable.
Scenarios use the mean of matched six-minute tidal residuals in the 24 hours ending at the latest available NOAA observation: the start is excluded and the end included. This is always 240 expected intervals, including across daylight-saving changes. At least 90% coverage and no missing run longer than one hour are required. Observations must be no more than two hours old and retrieval must have no errors. The rolling 24-hour average is held constant as an assumption; it is not a prediction of the residual at a future high tide.
Opening the page shows a prepared station snapshot immediately. The server checks both stations in the background about every five minutes; Refresh data requests a new check, with simultaneous requests combined. The page checks for updated snapshots every 15 minutes and briefly polls while a check is running. Saved snapshots retain their original Checked and source-fetched times. Recent observations reconcile the last three UTC days, including month boundaries, with a full current-month reconciliation at least daily when NOAA is available. Earlier history remains explicitly cached. Astronomical predictions are cached for up to seven days only while their dates cover the required instants. Failed checks remain visible, and scenarios pause on retrieval errors, checks older than 20 minutes, or observations older than two hours. No cached value is labeled newly checked merely because the page was opened.
NOAA API documentation ↗ · Quality flags ↗ · Station observations ↗
Station history and chart controls
Long BeachData begins April 1, 2026April 1 has partial usable coverage. Complete-day means begin April 2.
Los AngelesData begins November 1, 2024Default for two years of history. The chart and scenarios use the selected station.
Change over 30 days
—
Daily mean compared with 30 days earlier
Selected station
Los Angeles
NOAA 9410660 · Regional context for Newport Beach
Los Angeles has two years of history, from November 1, 2024. The chart and high-tide scenarios use this station.
Los Angeles · November 2024 onward · separate station series
Selected chart day: coverage and quality
Choose a point on the chart to inspect its coverage.
What the tide comparison assumes
Slide through the dates to compare each day’s highest predicted high-water event with October 27, 2026. The possible height shown holds the latest 24-hour tidal residual constant; it is a what-if scenario, not a forecast.
Latest 24-hour average differenceUnavailableRequires sufficient matched observations over the past 24 hours and fresh data.
The future-tide comparisons use the latest 24-hour average residual, held constant as a what-if. That is different from the same-instant relationship illustrated here, and it is not a forecast.
These are regional station heights, not Newport flood depths. Conditions can change; check regularly and consult official advisories. Being closer to an event does not guarantee accuracy.
October reference and November / December comparison details
Three-day residual-change reference and its limits
3-day residual-change historical what-if
Loading the reviewed three-day residual reference…
Dark gray is fixed at ±1.3 inches and light gray at ±2.9 inches around the selected astronomical tide plus the current trailing 24-hour mean residual. They are rounded, symmetric references close to the six-month historical 50th and 90th percentiles of maximum absolute three-day residual changes. They are not probabilities for the selected date. Inches are divided by 12 to place the bands on the feet-MLLW gauge.
Target, method and limitations
Target: the largest absolute change in the trailing 24-hour mean residual anywhere in the following 72 hours, checked every six minutes from overlapping hourly origins. This is not a distribution of observed high-tide errors, instantaneous residuals, or waves. An observed peak can differ further, so the gray endpoints do not bound the actual high tide. No separate peak-error or phase adjustment is added.
Display preferences
Residuals fluctuate. Check back often.
Return in the days leading up to the selected high tide to see how current residuals have changed. The page refreshes while open, but does not predict how residuals will evolve before that date.
The blue ripple uses a fixed visual amplitude, unrelated to the gray bands or real waves. It is decorative, not a wave-height or uncertainty estimate. Reduce motion or your device’s reduced-motion setting keeps it still.
Created by Tom Stewart
Tom is not an oceanographer, a data scientist, a meteorologist or a software developer of any kind. Naturally, he built a tide dashboard.
Tom is a marginally skilled surfer, sailor, fisherman and free diver. He does float up and down with the rising and falling tide.
He vibe-coded this whole thing with just an idea and the help of ChatGPT. The published site runs on Control Plane.
Powered by Control Plane
Control Plane brings your clouds together, with automatic failover, built-in security, and scaling.
Hosted simultaneously on AWS and Google Cloud in Oregon. Two clouds, one tide obsession.
Try the Control Plane plugin for ChatGPT (opens in a new tab)