Sea level trends since 1993
How the rates are spread across the ocean
Is the rise speeding up?
How acceleration is spread across the ocean
Key takeaways
- Sea level has risen almost everywhere since 1993, but not at the same rate.
- Parts of the western Pacific, the Gulf of Mexico and the western boundary currents have risen two to three times faster than the global average.
- A few areas, such as the eastern Pacific, have changed little, mostly because of shifts in winds and currents.
What are we looking at?
Each color shows how fast the sea surface has risen or fallen at that location over the satellite era, in millimeters per year. Warm colors mark faster rise and blue marks slower rise or a slight fall. Averaged over the whole ocean, the map gives about … mm per year, matching the global mean sea level record.
Why is the rise uneven?
Ocean currents and winds move heat and water around, piling water up in some regions and drawing it down in others. Over a few decades these patterns can add or subtract several millimeters per year. The strong rise in the western tropical Pacific, for example, reflects stronger trade winds in the 2000s and 2010s. Changes in the Gulf Stream and the Kuroshio show up as bands of fast rise along their paths. Over longer periods, the gravitational effects of melting ice sheets and the slow rebound of land from the last ice age also shape the pattern.
Why do we care?
The rise that matters for any coast is the local rise. This map shows the ocean part of that change. Combined with sinking or rising land, it sets how quickly high tide flooding is growing in a given place. Explore local projections with the AR6 Sea Level Projections tool or, for U.S. coasts, the Coastal Sea Level Explorer.
How the map is made
The map is computed on this site from NASA-SSH, NASA's standardized sea surface height record from the reference altimeter missions. NASA-SSH publishes a new global map every week on a half degree grid. At every location we fit a straight line through the full weekly record, while also fitting the yearly and half yearly seasonal cycles so they do not bias the trend. The map on this page keeps the full half degree resolution of the grids, and the smaller maps elsewhere on the site use a one degree average. Both are recomputed every week as new data arrive.
This map is the sea level trend map published by the NOAA Laboratory for Satellite Altimetry, averaged to a one degree grid for display. It is a stand in until the map computed on this site from NASA-SSH weekly grids is available.
Method: least squares trend at each location. Data: satellite altimetry.
How acceleration is measured
At every location we fit, all at once, a curve with a steady acceleration, the yearly and half yearly seasonal cycles, and two climate indices: the NASA-SSH El Niño index and the NASA-SSH Pacific Decadal Oscillation index. Both indices come from the same satellite record. Fitting them together with the curve means the large swings in sea level that come with El Niño, La Niña and the Pacific Decadal Oscillation are not mistaken for acceleration. This matters most in the Pacific, where a strong El Niño near the end of the record can otherwise look like the sea speeding up. Acceleration is twice the curvature of the fitted curve.
To judge whether an acceleration is real, we average what the fit leaves unexplained into yearly values, measure how large those leftovers are and how strongly one year carries into the next, and widen the uncertainty to match. An acceleration is called significant when it is larger than the 90 percent confidence limit. Faded areas are not evidence of no acceleration, only that this record alone cannot confirm it at that level. At 90 percent, about one in ten places with no real acceleration will still show as significant by chance.
Method: