How we track high-tide flooding around the world
What the dashboard shows, where the numbers come from, how current they are, and what they can’t tell you.
The short answer
As sea level rises, ordinary high tides reach further. In many coastal towns, streets and parking lots now flood on sunny days at the highest tides, not only during storms. This dashboard counts, for tide gauges around the world, how often high tide reaches flood level now compared with 2005–2014. For US gauges it adds NOAA’s live readings, the high tides of the coming week, official flood counts and projections to 2050.
How we know
Tide gauges worldwide. The University of Hawaiʻi Sea Level Center gathers hourly sea level from national tide gauge networks and publishes it as “fast-delivery” data, about one to two months behind real time. We use every gauge that is still reporting and has enough data in 2005–2014 to set its flood level: about 130 gauges in more than 40 countries and territories, alongside NOAA’s 18 US gauges. Gauge coverage is uneven: the Pacific, Japan, Australia and the Americas are well covered; Europe, Africa and mainland Asia much less, because many national networks do not share hourly data through this archive.
Flood level. There is no worldwide official flood threshold, so we use NOAA’s method. NOAA sets the minor flood level of each US mainland gauge at the average daily high tide plus 0.50 m plus 4% of the tide range (moderate and major levels are set the same way, higher up). We apply the same formula to every gauge, using its own tides in 2005–2014. Minor flooding is disruptive: low roads and parking lots flood. A flood day is a day the highest hourly reading reaches that level. The formula was set from US gauges, most with tide ranges under 3 m; where tides are much larger (Broome in Australia, Brest in France) ordinary spring tides pass it, so we show those gauges but leave them out of the comparisons.
Pacific islands. On Pacific islands the formula fails the other way. NOAA found that the water at its island gauges had rarely reached the formula’s level although residents reported flooding. Since June 2023 it uses one level for its eleven Pacific island gauges, from Hawaiʻi to Guam, Kwajalein and American Samoa: 0.30 m above the average daily high tide, the average of the local flood levels the National Weather Service uses in Hawaiʻi. We apply that level to every gauge in Hawaiʻi and in the Pacific island nations and territories. Until November 2026 this page used the mainland formula there, and wrongly showed most Pacific islands as almost never flooding. NOAA adds two cautions that hold here too: the true number of flood days may be lower, and at Kwajalein, where tides are larger, the level may not describe flooding well. On the atolls with similar tides (Majuro, Tarawa, Funafuti, Nauru) the water passes it on roughly 40 to 60 days a year.
More often, or about the same. We compare each gauge’s average flood days a year in the last ten complete years with 2005–2014, the same ten years for every gauge. A gauge counts as “more often” or “less often” only if the two differ by at least a quarter and by at least one day a year; smaller differences are too close to chance and count as “about the same”. Under one day a year in both decades is “still rare”.
Checked against NOAA. Twenty-seven US gauges are in both networks, and NOAA publishes official flood counts for them. At the 16 where NOAA uses the formula, our flood level is within 3 mm of NOAA’s at 15. Our flood counts are a little over half of NOAA’s for the same gauge, because NOAA measures from the average high tide of 1983–2001, when the sea was lower, while the hourly records we use mostly start around 2000. The class of change is the same by both counts at 22 of the 27 gauges and one class apart at four. Guam is the exception: NOAA counts about 15 flood days a year there and we count almost none, because the sea at Guam stood as high in 2005–2014 as it has since. That is why the map and the ranking compare changes, not counts. The comparison can be repeated with scripts/check_against_noaa.py.
Records that jump. The fast-delivery data is not checked for changes in a gauge’s reference level. A jump of 20 cm would look like a flood of flood days. We therefore look at each gauge’s yearly average level and set a gauge aside when the years after some point all stand well above or below the years before. In November 2026 that was the case at seven gauges: Stockholm and Smögen in Sweden (from 2021), Juan Fernández and San Félix in Chile (2019), Legazpi in the Philippines (2018), Booby Island in Australia (2021) and Ofunato in Japan, where the land dropped in the 2011 earthquake. They are shown on the map but not compared.
Land that moves. A gauge measures the sea against the land it stands on. In southern Alaska the land is rising faster than the sea: NOAA measures sea level falling by 2.5 mm a year at Sitka and 8.7 mm a year at Kodiak, and flood days there are falling. In American Samoa the land has been sinking since the 2009 earthquake, and NOAA measures sea level rising 15 mm a year at Pago Pago. Such gauges stay in the comparison, because that is the change people there live with.
United States. For 18 US gauges we use NOAA’s own data: live readings every six minutes, tide predictions, official thresholds and flood counts, NOAA’s outlook for the current year, and the 2022 interagency sea level rise scenarios. Along the US coast, sea level is projected to rise 10–12 inches (0.25–0.30 m) on average between 2020 and 2050. The difference between the measured level and the predicted tide is the weather effect: onshore winds and low air pressure push water up (storm surge).
NOAA’s outlook and projections. NOAA’s outlook year runs from May to April, so that one El Niño or La Niña winter falls in one year. The range it gives for each gauge is its forecast plus and minus one standard deviation. In the projections, “2050” is the average of the ten years ending in 2050, not the 2050s.
How current each part is
- Worldwide gauges: fast-delivery data arrives in monthly batches, one to two months behind; we check every night.
- US water level now: preliminary 6-minute data, read every night. A gauge that has not reported for six hours shows no reading.
- US high tides next week: NOAA tide predictions, refreshed every night.
- Flood days: complete calendar years with at least 80% of data. The comparison moves on by a year each January.
- US outlook and projections: the outlook is issued each summer for the year from May to April; the scenarios come from the 2022 report.
What the dashboard can’t tell you
It is not a flood warning. For a coming storm or king tide, follow your national weather service and local officials.
A gauge describes its own harbour. Streets a few kilometres away can flood at lower or higher levels, depending on their height, drainage and sea walls. Rain and river flooding are not included. Outside the US the flood level is our application of NOAA’s rules, not a threshold set by the local authority.
The comparison starts in 2005–2014 because most of the hourly records do. Ten years is short. At Guam and Palau the sea was no higher in 2016–2025 than in 2005–2014 (at Palau after rising almost 9 cm in the ten years before), so flood days there have not increased by this measure. Where there is almost no tide, as in the Baltic, the water reaches the level only in storms.
Further reading
- University of Hawaiʻi Sea Level Center: the worldwide tide gauge data used here. uhslc.soest.hawaii.edu
- NOAA, High Tide Flooding: outlooks, annual reports and the flood-day data used here. tidesandcurrents.noaa.gov/high-tide-flooding
- Sweet et al. (2022), Global and Regional Sea Level Rise Scenarios for the United States (NOAA Technical Report NOS 01; record at USGS). pubs.usgs.gov
- Kavanaugh et al. (2023), Notice of Methodology Update: NOAA High Tide Flooding Outlooks (NOAA Technical Service Publication NOS CO-OPS 001): the flood level for Pacific islands, and how the projections are averaged. repository.library.noaa.gov
- NOAA, Annual High Tide Flooding Outlook: the forecast for the current year, by region. tidesandcurrents.noaa.gov
- Sweet et al. (2018), Patterns and Projections of High Tide Flooding Along the U.S. Coastline Using a Common Impact Threshold (NOAA Technical Report NOS CO-OPS 086): how the flood thresholds are set. tidesandcurrents.noaa.gov
- NOAA, Sea Level Trends: measured trends at every US gauge. tidesandcurrents.noaa.gov/trends-and-extremes
Credits
Worldwide data comes from the University of Hawaiʻi Sea Level Center and the national agencies that run the gauges and share their records. US data comes from NOAA’s National Ocean Service and its Center for Operational Oceanographic Products and Services. We only download and present their public data.
← Back to the dashboard