A woman stands on top of a rock holding a fish off Bikeman islet, located off South Tarawa in the central Pacific island nation of Kiribati, May 25, 2013. Kiribati consists of a chain of 33 atolls and islands that stand just meters above sea level, spread over a huge expanse of otherwise empty ocean. With surrounding sea levels rising, Kiribati President Anote Tong has predicted his country will likely become uninhabitable in 30-60 years because of inundation and contamination of its freshwater supplies. Reuters

Scientists overestimated the rate of sea level rise between 1901 and 1990, meaning Earth’s oceans have expanded more dramatically in the past 25 years and major coastal cities and island communities could be closer to going under than previous research has shown. That’s the crux of a new study from scientists at Harvard University published Wednesday in the journal Nature.

Based on their data, annual sea level rise from 1901 to 1990 was about 1.2 millimeters (.05 inches), roughly .3 to .6 millimeters less than the “standing wisdom” for that time period, said Robert Kopp, an associate professor of earth sciences at Rutgers University in New Brunswick, New Jersey, and a senior scientist on the study. “As you go back through the 20th century, the tide gauge record becomes sparser,” he said. “One of the things we’re interested in is global mean sea level – the volume of water in the ocean.” Previous research did not account for “gaps” in tide gauge data, according to Kopp. That was where the discrepancy between the Harvard data and previous data came from.

Researchers said their study could change scientists’ understanding of sea level rise today. "If we've been overestimating the sea-level change during [the 20th century,] it means that these models are not calibrated appropriately, and that calls into question the accuracy of projections” over the next 85 years, Eric Morrow, a recent Ph.D. graduate of Harvard’s department of earth and planetary sciences and co-author of the study, said in a statement. He described sea level acceleration as a “larger problem” than was previously believed.

Sea level rise is one of the most familiar effects of climate change, and the one with perhaps the biggest consequences. The consensus among climatologists is that Earth’s oceans are certainly expanding, a process linked to manmade global warming caused by the burning of fossil fuels. Over the past 100 years, the global mean sea level has risen by 10 to 20 centimeters (4 to 8 inches,) a process that has only accelerated in the 21st century.

From 1901 to 1990, the rate of sea level rise was estimated to be between 1.5 and 1.8 millimeters annually, according to the U.N.’s Intergovernmental Panel on Climate Change, or IPCC, widely considered the authority on changing global sea levels. That rate has spiked over the past 20 years to an estimated 3.2 millimeters (.13 inches) a year – twice as fast as during the last century.

There’s no indication that the acceleration will slow down any time soon. Most experts predict somewhere around one meter of sea level rise by 2100, enough to inundate many of the world’s major cities. “Sea level is going to go up. The IPCC is right on that,” said Tim Barnett, a marine physicist with the Scripps Institution of Oceanography at the University of California, San Diego. “We can argue on how much, but the boundaries of that argument are within ‘bad’ or ‘worse.’”

The reason sea levels have risen more rapidly in recent decades is connected to several things happening simultaneously. For one, about 80 percent of the heat added to Earth over the years due to global warming has gotten stored in the oceans. When water warms, it expands. Then there’s the unprecedented glacial and polar melt, coupled with huge ice loss in Greenland and West Antarctica, dumping more water into the global bathtub.

Greenland, for instance, is “melting faster than anybody thought it would. Ditto Antarctica,” Barnett said. “If you melted just a tenth of Greenland, you’d have .7 meter rise. That finishes off Miami, that finishes off Wall Street. It doesn’t take a lot to wreak havoc.”

Scientists have pinpointed 20 cities with the most to lose from rising sea levels where flooding will come at a high cost. By 2050, rising sea levels could cost New York around $2 billion; Miami, about $2.5 billion; Mumbai, roughly $6.4 billion and Guangzhou, China, a whopping $13.2 billion, according to a 2013 study published in the journal Nature Climate Change.

The effects of rising sea levels are perhaps most apparent in the Pacific island nation of Kiribati. The country is a chain of 33 low-lying atolls and islands, mere flecks on the vast ocean’s surface. Its highest point is just a few meters above sea level. By some estimates, Kiribati’s 103,000 residents will hardly have a stone to stand on by the end of the century. The government of Kiribati two years ago reportedly spent $9.6 million on 6,000 acres of land in Fiji for when that day comes.

Sea level rise is tied up in all of the processes contributing to ocean expansion. “A lot of our understanding of how much heat the oceans are absorbing, how much ice is melting, is all wrapped up in sea level rise,” said Jonathan Bamber, professor of glaciology at the University of Bristol in England. That’s where the Harvard study could have implications for future projections, he said. “If these results are correct, then we have to reinterpret some of our understanding of 20th century glacial melt and heat uptake.” He described the Harvard team’s approach as “reasonably compelling.”

Sea level rise during the 20th century was measured almost entirely from tide gauge data. U.S. scientists began measuring coastline changes in the early 1800s by placing tall wooden rulers directly into the surf. Many decades later, in 1851, the U.S. put is first self-recording tide gauge off the coast of San Francisco.

The problem with tide gauges is that they were never equally distributed and favored the Northern Hemisphere. Because they’re often located just off shore, tide gauges can be affected by forces like sedimentation and tectonic plate movement that are unrelated to climate change.

Today, climate scientists’ collective toolbox for measuring changing sea levels includes Arctic core samples, tide gauge readings and, in more recently, GPS and satellite measurements. It’s not a perfect science, but overall, scientists are much more confident in their estimates today than 100, 50 or even 20 years ago.

The Harvard study is sure to raise some eyebrows in the scientific community. “I have no doubt there will be lots of keen eyes pouring over these results,” Bamber said.