Tide Manifold @ Erlands Point

For about a week the tides were boring. The water slid up and back without committing to anything, none of the big lows that pull the water off the flats and leave the inlet looking half-drained. The tides app agreed: nothing dramatic coming. The reason was up in the sky. The moon was about half-lit, midway between new and full, which is the one point in the cycle where the sun and moon pull at right angles instead of together. Right angles, small tides. Neaps.

That got me curious about the whole thing, so I plotted every day at once. Each dot is one day at NOAA's Bremerton gauge, the nearest real-time sensor to Erlands Point. The axes never move: x is the sun–earth–moon angle (0° at new moon, 180° at full, spring tides at both ends), y is the earth–moon distance (perigee to apogee), and z is the day's tidal range, high minus low. Range is the clean signal: it strips out time of day, geography, and weather and leaves only the astronomy. The biggest ranges pile into the corner where the angle is extreme and the moon is close, the king-tide corner. That dull half-moon week sits in the low middle, right where it belonged. The whole surface is the moon, behaving exactly as advertised.

Then I pulled the range apart into the two numbers it's made of: the actual high and the actual low, each on its own surface. Mostly they behave. The highs climb toward the spring corners, the lows fall away from them, and the gap between the two clouds is the range from before. But one point refuses to sit where it should. December 18, 2024 is the highest high in the whole record, 14.65 ft, and it floats well outside the king-tide corner, out in the flat middle where the moon makes nothing special. The moon had already left full that morning and was drifting back toward its far point. By the astronomy alone it was an ordinary day. So something else put the water there. It was a storm surge. A low-pressure system sat over Puget Sound that morning. NOAA's barometer at the gauge bottomed out near 1006 mb with gusts to 30 knots out of the northwest, and forecasters had already flagged coastal-flooding risk for the Sound that week. The storm stacked about 1.6 ft of extra water onto an already-high tide.

You can pull the weather out on its own. NOAA also publishes the astronomical prediction: the tide the moon and sun alone would make. Subtract it from the observed high, matched to the same minute, and what's left is the surge residual. The part of the water the sky doesn't explain. Plot that on the same axes and the astronomy goes silent. Tide-driven days flatten toward zero. Only the weather stands up. The winter months bristle with low-pressure surges, a few summer high-pressure days dip slightly negative, and December 18, 2024 rises right out of the calm.

One last wrinkle, and it's the whole point. The day that started this, December 18, 2024, is the highest high water in the entire record (#1 of 549), yet its surge was only the third-largest, +1.6 ft. It topped out because the moon already had the tide near its peak: the astronomical prediction alone ranked 15th, and the storm just pushed it over. December 17, 2025 is the mirror image. Its surge is the biggest in the record: +2.4 ft, off a 990 mb low. But the moon was near apogee and well off the spring angles, so that day's predicted tide ranked 414th. A record surge only lifted it to a 13.6 ft high, beaten by thirteen ordinary days. For scale, consider the storm people here actually remember: the November 2024 bomb cyclone. It bottomed near a record-tying 942 mb and cut power to some 600,000 homes, almost 50 mb deeper than either, and it isn't even on this chart: it struck three weeks before the record begins, and it's remembered for wind, not water. That's the point. A big surge isn't a high tide, and the most violent storm isn't the highest water. The flooding danger needs the storm and a tide the moon has already stacked to arrive together. 2024 had the alignment; 2025 had the bigger surge and nowhere to put it; the bomb cyclone had the wind.

A few honest notes. The vertical numbers are measured: NOAA's verified daily highs and lows, with the trailing ~2 weeks stitched from 6-minute readings until NOAA verifies them. The splice is visible if you look. The angle and distance are computed: Skyfield with the JPL DE421 ephemeris at noon UTC. And it isn't a backyard sensor, it's NOAA's. There's a station right on Dyes Inlet at Tracyton, closer to me, but it's prediction-only with no instrument, so the real measurements come from the Bremerton gauge down the water.

Sources

• Tide observations + predictions — NOAA CO-OPS station 9445958, Bremerton ↗, via the CO-OPS Data API ↗ (high_low + water_level + predictions, datum MLLW).
• Dec 2024 storm context — KING 5: significant coastal flooding, Puget Sound ↗; Seattle Times: how king tides amp up flood risk ↗.
• Lunar geometry — computed with Skyfield ↗ and JPL's DE421 ephemeris ↗.
• The closer NOAA station, prediction-only (no instrument): Tracyton, Dyes Inlet (9445901) ↗.