The tide is not a climate subject. It is set by celestial mechanics, it was here long before us, and warming barely alters it. That is exactly what makes it useful: because it does not change, it serves as a yardstick, and what does change can be read against it.
Why two tides a day
Lunar attraction stretches the ocean into two bulges: one facing the Moon, where the pull is strongest, the other opposite it, where the pull is weakest and the rotation of the Earth–Moon system wins out. Both bulges stay aligned with the Moon.
The common misunderstanding is to picture a wave travelling around the globe. That is not what happens: the bulges are roughly stationary relative to the Moon, and it is the Earth that turns beneath them. Each coast therefore passes through two bulges a day, hence two high waters — and the daily lag of about fifty minutes comes from the Moon having moved along its orbit in the meantime.
The Sun produces the same effect, twice as weak. When the two line up, at new and full moon, tides are strong: these are spring tides. In quadrature they work against each other: these are neap tides.
The 18.6-year cycle
The plane of the lunar orbit is tilted by about 5° from that of the Earth's orbit, and that plane swings slowly: it returns to the same position every 18.6 years. This is the nodal cycle, and it modulates the tidal range by a few percent.
A few percent on a range of one metre forty makes several centimetres — a signal that is perfectly legible in a long series, and one of the few astronomical quantities a tide gauge records directly.
NOAA Center for Operational Oceanographic Products and Services — Tides and Currents — The Battery, New York · accessed 15 November 2026
The New York series spans more than a century, that is five and a half cycles. The two curves tell two superimposed stories: the range oscillates without drifting, the mean level drifts without oscillating.
The threshold does not move, the sea rises
Here is where the tide stops being a matter of astronomy. A tide gauge counts the number of days a year on which high water exceeds a flooding threshold fixed once and for all, set from the height at which the water starts causing trouble — flooded streets, storm drains backing up.
The tidal range has not changed. Neither has the threshold. What changes is the mean level on which that tide is superimposed.
NOAA Center for Operational Oceanographic Products and Services — High Tide Flooding — Annual Flood Count · accessed 15 November 2026
The result is not proportional, and that is the whole interest of this indicator. High water heights are distributed around a mean; raising that mean by a few centimetres does not shift the distribution by one notch, it tips above the threshold a whole population of tides that used to pass just below it. A steady rise therefore produces growth that runs away.
High tide flooding days at New York, by decade
Bar heights are proportional to the gap from the period's minimum, 0.25 days for the 1920 decade. The 1920, 1950, 1970, 1990, 2000, 2020 decades have fewer than ten years on record. NOAA Center for Operational Oceanographic Products and Services — High Tide Flooding — Annual Flood Count · accessed 15 November 2026
The annual count is very noisy — one storm season or one El Niño episode is enough to double it. It is the decade average that carries the signal, and it is unambiguous.
What this changes for the notion of risk
Global mean sea level
81.90mm
+3.22 mm / yr
1993–2024 · 32 points
Global ocean heat content
32.5310²² J
+1.12 10²² J / yr
2005–2025 · 21 points
Coastal flooding is the textbook case of what specialists call a compound event: it results not from a single cause but from the coincidence of several — a spring tide, a depression lifting the water surface, an onshore wind, sometimes a river flood arriving at the same time from upstream A et al., 2018. Each of these factors taken alone remains ordinary; it is their meeting that floods.
The rise in mean level creates none of these factors. It raises the common starting point for all of them, which amounts to shortening the return period of every combination. A hundred-year event becomes a ten-year one without any storm having changed its nature Jiang et al., 2024.
The mean level, and how we know it
NOAA Laboratory for Satellite Altimetry — Global Mean Sea Level — Laboratory for Satellite Altimetry · accessed 15 November 2026
Two independent measurement systems coexist. Tide gauges, the older of the two, give a relative level: they are attached to ground that moves as well — subsiding deltas, rebounding regions formerly under ice. Satellite altimetry, since 1993, gives an absolute level, referenced to the centre of the Earth.
Their agreement over the common period is what grounds confidence in the reconstruction of the previous century, made from tide gauges alone Church et al., 2011. The global rate is not constant: it has more than doubled since the start of the altimetry series Hamlington et al., 2024.
Sources
- Tides and Currents — The Battery, New York — NOAA Center for Operational Oceanographic Products and Services
- High Tide Flooding — Annual Flood Count — NOAA Center for Operational Oceanographic Products and Services
- Permanent Service for Mean Sea Level — National Oceanography Centre
Studies cited in this brief
John Alexander Church et al. · Surveys in Geophysics · 2011
- Ocean and sea level
Benjamin Dillon Hamlington et al. · Communications Earth & Environment · 2024
- Ocean and sea level
Shijie Jiang et al. · Science Advances · 2024
- Water, droughts and floods
- Natural hazards
Future climate risk from compound events
Open accessJakob Zscheischler A et al. · Nature Climate Change · 2018
- Natural hazards
- Heat and heatwaves