Study reveals new insights into how fast-moving glaciers may contribute to sea level rise

Models of sea level rise based on our understanding of how Earth’s ice sheets respond to a warming atmosphere could be incorrect, a new study has found. This could have significant implications for future predictions of global sea level rise from the Greenland and Antarctic ice sheets.

Climate change is resulting in sea level rise as oceans expand and ice on land melts. How much and how fast sea level will rise in the current period of human-induced increased warming depends in part on how Earth’s ice sheets in Greenland and the Antarctic change their flow speeds in response to a warming atmosphere.

A new study, carried out in collaboration between Oxford University’s Earth Science department, the Oxford University Mathematical Institute, and Columbia University has shown that predictions of the impact of melting on Greenland tidewater glacier speeds could be incorrect, with implications for future predictions of global sea level rise from the Greenland and Antarctic ice sheets.

Laura Stevens, Associate Professor of Climate and Earth Surface Processes, and her colleagues used Global Positioning System (GPS) observations of the flow speed of Helheim Glacier—the largest single-glacier contributor to sea level rise in Greenland—and captured a near perfect natural experiment: high-temporal-resolution observations of the glacier’s flow response to a supraglacial lake drainage.

Greenland’s tidewater glaciers, including Helheim Glacier, contribute to sea level rise through calving events, where large chunks of ice detach from the glacier and fall into coastal fjords as icebergs. These icebergs calve from the glacier where the glacier meets the ocean, and the faster a tidewater glacier flows, the more ice enters the ocean and contributes to sea level rise.

The surface of glaciers often melts during warmer summer months and this meltwater can pool into supraglacial lakes that are kilometres across and rest on the top surface of the glaciers. Occasionally these lakes drain their water through the glacier and down into the glacier’s basal drainage system. For inland ice-sheet regions, we know from previous work by Laura and others that this drained lake water reduces friction between the ice and ground, causing the glacier to slide faster for a few days. However, for tidewater glaciers that end in the ocean, our understanding of the effects of such lake drainage events on glacier speeds have until now been limited due to a lack of in situ observations.

The results of this new study show that instead of the fast, downhill movement observed during lake drainages in inland ice-sheet regions, Helheim Glacier exhibited a relatively small ‘pulse’ of movement where the glacier speed up for a short amount of time and then moved slower than its speed prior to the lake drainage over the next few days. The team used a numerical model of the subglacial drainage system, originally developed by Ian Hewitt, Professor of Applied Mathematics and a co-author on the study, to interrogate these GPS observations. They discovered that tidewater glaciers like Helheim can have an efficient system of channels and cavities along their beds that allows flood waters to be quickly evacuated from the glacier bed without causing an increase in the total net movement of the glacier.

Although this sounds like positive news in terms of sea-level-rise implications, Laura says that a different effect should be expected for glaciers where surface melt is currently low, but will increase in future due to climate change. “When outputs from the subglacial drainage system model were consistent with our GPS data for the Helheim event, we were motivated to run additional model cases. What would happen if this lake had drained during the winter, when water inputs to the basal drainage system from summertime surface melting are turned off?”

They ran this winter model case, which is more akin to the year-round conditions of colder, Antarctic tidewaters, and the model outputs indicated that lake drainages under these conditions would produce a net increase in glacier movement, largely due to the less efficient winter-time subglacial drainage system not being able to evacuate flood waters quickly. This suggests that lake drainage events on currently colder, Antarctic tidewater glaciers should lead to only faster flow, though there are, as of yet, no in situ observations of tidewater-glacier response to lake drainage for any Antarctic tidewater glaciers.

The study also calls into question some common approaches for inferring subglacial drainage conditions from glacier velocities that are observed by satellite observations. “There’s a growing approach in observational glaciology of mapping tidewater glacier velocity patterns onto estimates of surface melt, with the thinking being that the amount of melt should explain the velocity pattern in some way,” Laura commented. “But what we’ve observed here at Helheim is that you can have a big input of meltwater into the drainage system during a lake drainage event, but that melt input doesn’t result in an appreciable change in glacier speed when you average over the week of the drainage event.” With the highest temporal resolution of satellite-derived glacier speeds currently available being roughly one week, lake drainage events like the one captured in the Helheim GPS data go unnoticed.

“These tidewater glaciers are tricky,” Laura says. “We have a lot more to learn about how meltwater drainage operates and modulates tidewater-glacier speeds before we can confidently model their future response to atmospheric and oceanic warming.”

Associate Professor Laura Stevens (right) and co-author Professor Meredith Nettles (left, Columbia University) approach a Greenland supraglacial lake via helicopter. Photo by Marianne Okal (UNAVCO, Inc.).

Notes to editors:

The study ‘Tidewater-glacier response to supraglacial lake drainage’ is published in Nature Communications and the paper is available at: https://doi.org/10.1038/s41467-022-33763-2.

For media inquiries, and to view a pre-embargo copy of the paper, contact Prof Laura Stevens, Earth Sciences Department, University of Oxford: laura.stevens@earth.ox.ac.uk.