reader comments 91
Some people have eyeballed satellite measurements of sea level rise and claimed that there is no sign of acceleration—just a linear increase. Then, ignoring the physics of melting glacial ice and the expansion of warming water, they declare that future sea level rise won’t be a big deal. Many studies have demonstrated accelerating rates of sea level rise over the past millennia, as well as the tide gauge record spanning the 20th century. But the short satellite record—which only started in 1993—is a slightly different question.
While the global satellite record is in many ways cleaner than coastal measurements that can be affected by processes that raise or lower the ground that the tide gauge sits on, there are still complications to account for. Since the record is still short, a small wiggle of natural variability can have a significant impact on seeing the subtle acceleration. The back and forth between El Niño and La Niña, for example, causes sea level to vary from year to year by changing the amount of precipitation that temporarily shifts water onto continents.
Accounting for all of this is complicated, but that hasn’t stopped researchers from trying.
Two recent studies helped to clear up some of the issues. One study found an error in the calibration system on the TOPEX/Poseidon satellite, which was throwing off its sea level measurements by a few millimeters. Another study pointed out that the satellite record happened to start just after the 1991 eruption of Pinatubo, which affected global climate for several years. The beginning of the satellite record captures a sea level “bounce-back” from that event, which makes sea level rise in the 1990s look faster than the real trend.
Accounting for this slows the rate of sea level rise at the start of the record, making it easier to note the faster sea level rise in recent years.
Greenland and Antarctica, front and center
A study published last year calculated that the measured rate of sea level rise in the second half of the satellite record was greater than the first half—meaning it had accelerated. But a new study led by the University of Colorado’s R. Steven Nerem does some work to get to more straightforward math—simply calculating the rate of acceleration that has now become apparent.
Using the compiled data from the TOPEX/Poseidon satellite and Jason-1 through Jason-3 orbiters, the researchers apply a couple of corrections to smooth out the natural variability a bit. Accounting for the effect of the Pinatubo eruption increases the calculated acceleration, while accounting for El Niño and La Niña reduces it. They also compare the satellite data to tide gauge records around the world to work out the error bars on their calculations.
The average rate of sea level rise for the 25-year period of 1993 to 2017 is 2.9 millimeters per year, but the researchers calculate that it has been accelerating by 0.084 ± 0.025 millimeters per year each year. The bulk of that acceleration is due to increasing ice loss from Greenland and Antarctica.
For context, the researchers extrapolate that acceleration out to 2100, finding that it would mean about 65 centimeters of sea level rise (more than double what we’d get with no acceleration). That’s well within the projected range of 52 to 98 centimeters given by the last IPCC report for the high-emissions scenario.
Both sets of numbers, however, are acknowledged as conservative. The IPCC reports have struggled with how to represent fast-changing sea level research, choosing to include only what could be confidently quantified and noting that additional factors could change extrapolation. Last year’s US National Climate Assessment instead summarized the research by including plausible sea level rise scenarios as high as 250 centimeters.
But even if the loss of ice from Greenland and Antarctica simply continued on the trajectory from recent years, this study suggests we would expect to see around 65 centimeters by the end of the century. That’s because sea level rise really is accelerating—as expected—enough that it’s now apparent even in the 25-year-long satellite record. And that’s a big deal.
PNAS, 2018. DOI: 10.1073/pnas.1717312115 (About DOIs).