Earth’s largest water reservoirs: How much ice are they losing?

As recently as the late 1990s, Greenland and Antarctica were thought of as enormous but slow-to-change reservoirs of freshwater. We knew that the waxing and waning of ice sheets during glacial cycles had caused sea level to rise and fall by 400 feet, but these changes had happened over tens of thousands of years.

A big wake-up call came in from both ice sheets in the early 2000s. In Antarctica, the Larsen B Ice Shelf collapsed in 2002 in a matter of weeks. It was not so much the collapse of the ice shelf that caused alarm — that ice is already floating — but the fact that the ice behind the ice shelves, that sits on land, began to flow faster into the ocean. This is because ice shelves act like a buttress for the grounded ice; if you weaken that buttress, the ice it is holding back will flow faster.

Around the same time, huge outlet glaciers in southeast and west Greenland, such as Jakobshavn, Helheim and Kangerlussuaq glaciers lost their floating tongues and began to flow faster — sometimes doubling their flow rate.

Eventually, satellite records showed that similar changes were occurring in West Antarctica (in the Amundsen and Bellingshausen seas) and all around Greenland. Floating ice was thinning and/or breaking off and the ice behind it was moving faster into the ocean. As more ice flowed from land into the ocean the sea level contribution from both ice sheets was rapidly growing.

The scientific community got to work to understand what was happening. Consensus arose on attributing these changes to increased melting above the water line, by a warming atmosphere, and increased melting below the water line, by a warming ocean. Through a combination of mechanisms that include thinning, fracturing and, at times, collapse of ice shelves and ice tongues, lubrication of the bed over which the ice flows, the buttressing provided by floating ice was reduced and ice was speeding up. Warming of ocean waters reaching the ice sheets was not surprising — we knew that the ocean had absorbed about 90 percent of the heat accumulated by the planet due to increased greenhouse gases. We also knew that air temperatures were rising globally, and especially in the Arctic. What was new was the time scale: The ice sheets could no longer be thought of as sleeping giants that took centuries to respond to climate change.

The changing perception of dynamic ice sheet changes is reflected in the reports by the IPCC, the United Nations body that evaluates the state of Earth’s climate based on published research. In the 1990s, the IPCC’s first and second assessment reports, contained statements like: “No major dynamic response of the ice sheets was expected during the 21st century, and the main contributor to sea level rise is thermal expansion and melting of glaciers.” In other words, the scientific consensus summarized by the IPCC back then was that ice sheets changed too slowly for us to worry about them. These are the reports many of us who are now “mid-career” had on our desks when we were Ph.D. students just entering the field.

In response to the rapid changes witnessed in the 2000s, which had not been predicted by models, the IPCC challenged the polar science community in its fourth Assessment Report in 2008 (the one that earned it the Nobel Prize) to up the world’s understanding of ice sheets. They concluded that the “understanding of these effects (rapid dynamical changes in ice flow) is too limited to assess their likelihood or provide a best estimate or an upper bound for sea level rise.”

This motivated the ice community to change its thinking about ice sheets and sea level rise and quickly realign research priorities. Major campaigns were launched to combine computer simulations of ice flow (the models) with a flood of new data coming in from satellites and from teams working on the ice and at the ice/ocean margins.

The IPCC fifth report in 2014 contained some initial estimates of future mass loss, which included dynamic changes, based on four carbon emission scenarios from best-case to worst-case. The Paris Agreement came that same year.

By 2018, both our knowledge and the ice sheets were changing so quickly that the update couldn’t wait for the next IPCC, prompting an intermediate report for the cryosphere and the ocean — the “Special Report on the Ocean and Cryosphere in a Changing Climate.” In this report the projected rise in sea level in 2100 from Antarctica and Greenland ranges from 3.5 inches to 1.8 feet.

The upcoming sixth IPCC report in 2022 will contain updated projections of future sea level rise based on tens of different simulations provided by research groups around the world. These groups all worked together in a community-led effort, involving ice sheet, ocean and atmosphere modeling and observational teams.

We have come a long way, but even after all this we are still playing “catch-up,” and there are still gaps in our understanding. We do know, however, that the ocean is warming and that both Antarctica and Greenland are vulnerable to this warming. The same goes for the atmosphere. We worry that the biggest portion of Antarctica, East Antarctica — which we still think of as a sleepy giant since it is so thick and vast, making it harder for warming ocean waters and increasing air temperatures to reach it — is starting to show signs of change. We also worry that there may be mechanisms, that we have not been able to witness in the modern record and hence that are not in the models, that may amplify the ice loss. Scientists are using paleo-reconstructions to figure out whether these may be important.

Still, we can say with confidence that sea level will continue to rise (faster) in the future and that our projections are conservative estimates. Indeed, satellite observations that measure the changing height (altimetry) and changing mass (gravimetry) of ice sheets are tracking the worst-case predictions from IPCC’s fifth report.

As we gather more data, both on and around the ice sheets using all available tools, including satellites, our observational record gets longer and our understanding improves. As our understanding improves, our models get better. Long-term measurements, sometimes acquired by launching new satellites (such as NASA’s ICESat-2 and other follow-on missions), coordinated modeling and international collaboration are key to delivering more accurate predictions, so that coastal communities can make informed decisions to protect infrastructure and citizens and manage resources.

Helen Amanda Fricker and Fiamma Straneo are professors of polar science and co-directors of the Polar Center at Scripps Institution of Oceanography, UC San Diego.