How much the sea has risen
Global ocean, 1808 to 2024
Global mean sea level and the extra volume of sea water it represents, in mm and cubic kilometres relative to the 1900-1909 average. These records are corrected for land movement, so they measure the ocean itself.
Sources:
- Jevrejeva et al. 2014: global sea level from 1,277 tide gauge records, 1807-2010, each corrected for datum changes and for land moving after the ice age. Before 1870 it is drawn dashed: until the 1850s it rests on about six northern European gauges, and the dip of about 80 mm in 1853-1866 lines up with San Francisco, New York, three fast-rising Finnish gauges and Liverpool joining the network in 1855-1858, not with any known loss of ocean water.
- Frederikse et al. 2020, "The causes of sea-level rise since 1900" (Nature): observed global sea level 1900-2018, and the separate contributions from warmer water expanding, glaciers, Greenland, Antarctica, and water stored on land (dams, groundwater pumping, lakes and soil).
- NOAA Laboratory for Satellite Altimetry: satellite radar altimetry 1993-2024 (TOPEX/Poseidon, Jason-1, 2, 3, Sentinel-6), measured from orbit so the land plays no part. Altimetry data are provided by NOAA Laboratory for Satellite Altimetry.
Volume is the rise times the ocean surface area of 361.8 million km², so 1 mm is about 362 km³. The satellite series has 0.3 mm/yr added for the ocean floor still sinking after the ice age, which makes the basins hold more water without the surface showing it; that is the standard correction for turning surface height into volume. The Jevrejeva and satellite series are shifted to match Frederikse over the years they overlap. Lough Neagh holds about 3.5 km³.
The coloured bands stack up the causes. Warmer water expanding adds volume without adding water; everything else is water that was on land and is now in the sea. The causes add up to a little less than the observed rise, which is the unexplained part of the budget in the paper.
Around Ireland, 1918 to 2024
Annual mean sea level from every PSMSL tide gauge on the island, in mm relative to the Dublin 1960-1990 mean. These gauges measure the sea against the land they stand on.
Data: PSMSL Revised Local Reference annual means. Each gauge has its own datum, so each one is offset to match the gauges it overlaps with (median difference over shared years, using unflagged readings only). Belfast and Malin Head are tied to Dublin over 1938-1997. Portrush is tied to Dublin and Malin Head over 1996-2001, which is the weakest link in the chain (about 11 mm of uncertainty, from 6 shared years). The Marine Institute gauges that start in 2008 or later are tied to Portrush and Bangor, and Howth to those.
Corrections to the raw PSMSL records:
- Dublin is split in two at 2002. PSMSL notes that up to 2001 the port supplied mean tide level, and from 2002 the harbour master assumed the datum was chart datum after the hydrographic department closed. The 2002-2009 series (dashed) is aligned on its own and left out of the running mean, because its datum is unconfirmed.
- Malin Head 1998-2001 is excluded. It falls 130 mm while Portrush, 40 km away, stays flat. The gauge had a documented calibration fault in 1993.
- Bangor 2014 is excluded. PSMSL notes the gauge read 20-40 mm high and had blocked channels that year.
- Howth 2018-2019 is excluded. Against the other Irish gauges its monthly values sit about 120 mm high until May 2019, then drop in a single month, which is a datum change rather than the sea.
- Ballyglass is split at its 2020 gap. It comes back about 55 mm lower relative to its neighbours, so the two parts are aligned separately.
- Any year where a gauge departs more than 60 mm from its usual difference to the other gauges is excluded. This removes Castletownbere 2011 and 2016, and Killybegs 2016.
- Years PSMSL flags for attention are shown hollow and are not used for alignment or the running mean. Excluded years are shown as grey crosses.
Trend of the combined record: about 0.1 mm/yr for 1918-1990, 2.3 mm/yr for 1990-2024. Castletownbere rises faster than its neighbours through 2008-2024, which may be local land subsidence or gauge drift; it is kept in. Much of the north of Ireland is still rising slowly after the last ice age, so these gauges understate the rise of the sea itself.
The time axis is logarithmic in years before a reference year; the slider moves that reference year, so older years get squeezed harder the further right it sits. Click a gauge to hide or show it. The running mean only uses visible gauges.