Human-caused sea level rise drives 21st-century worldwide water level extremes

A paper in Science Advances (June 2026) led by Climate Central, with Daniel M. Gilford of Climate Central as an author and Robert E. Kopp of Rutgers University as a co-author. Its question is what share of the growth in extreme water level days, place by place and region by region, can be put down to human-caused climate change. The press release frames this as a gap: flood risk at the local and regional scale had not been studied this way before, leaving communities short of information.

The authors used two largely independent methods at 519 tide gauge sites. The first adds up the sea level rise that can be attributed to four contributors: the ocean expanding as it warms, mountain glaciers melting, and the Greenland and Antarctic ice sheets losing ice. The second compares observed sea levels from 1900 to 2018 with a modelled version of the same period without climate-warming pollution.

The findings, as the press release states them: a human-caused signal in sea level is found at 97% of the sites; 58% of the extreme water level days observed between 2000 and 2018 are attributed to it; and, averaged over all sites, the count of days above the extreme threshold has nearly tripled since the 1970s. Both methods reach similar conclusions. A companion paper in Nature Climate Change from Tulane University, released at the same time, is titled "Human-driven sea-level rise has quadrupled the frequency of coastal sea-level extremes since 1900". Both studies were co-supported by the Megalopolitan Coastal Transformation Hub (MACH), a National Science Foundation-funded consortium led by Rutgers.

What a reader can take from it: a site-by-site attribution of extreme water level days at hundreds of tide gauges, and a worked example of two attribution methods, one from physical contributions and one from a counterfactual record, applied to the same question. In this wiki's stack it belongs with the data and the models that feed a coastal flood twin.