Robotic Kayaks Discover High Rates of Underwater Glacier Melt

The LeConte Glacier in southeast Alaska is a 21-mile long fast-flowing tidewater glacier, which terminates abruptly in a fjord––spilling its contents into the ocean. The glacier sheds ice from its 200-meter face in calamitous calving events when large blocks of ice drop into LeConte Bay. Researchers seeking to collect data on the glacier’s submarine melt rate needed a way to get close enough to the terminus to collect the data they needed—the solution: a fleet of robotic kayaks sent into waters too dangerous for human researchers to enter.

The LeConte Glacier from above.
Source: Jeremy Keith/ Flickr

The November 2019 study, which was published in the journal Geophysical Research Letters, was led by Rebecca Jackson, an assistant professor of physical oceanography at Rutgers University, and a team of researchers from Oregon State University, University of Alaska Southeast, University of Oregon and University of Alaska Fairbanks. 

Tidewater glaciers are glaciers that reach all the way to the ocean. At their border with the sea, they melt either through calving or through submarine melting. 

Submarine melt matters because it is a significant contributor to glacier melt and is sensitive to rises ocean temperature and shifts in ocean circulation. It’s also more difficult to observe directly than surface melt because it occurs on the underside of glaciers. It can take place through two processes. The first is more easily detectable and comes from the drainage of freshwater discharge due to upstream melt on the glacier. It creates fast-moving plumes of water entering the ocean at the glacier’s terminus. The second type of submarine melting is the slower and harder to measure process of ambient melting where a glacier melts directly into the sea. 

Ambient submarine melting is hard to measure: as a result it is typically estimated using laboratory experiments and models. The idea that emerged from theory and lab experiments suggested that ambient melting was responsible for only a small amount of total terminus melting. 

One of the autonomous kayaks used in the study in front of the glacier terminus. 
Source: David Sutherland/ University of Oregon

Jackon’s study used autonomous kayaks to study plumes created by ambient melting at LeConte. The glacier has been recorded moving at velocities of up to 18 meters per day along its one kilometer-wide terminus. The kayaks were deployed along the length of LeConte over a seven-day period in September 2018. As they cruised close to the steep face of the glacier, they used complex instruments to observe water velocity, temperature, and salinity. The near-glacier data they gathered was supplemented with data collected downstream aboard a research vessel. 

Plumes created by ambient melt only exist within 400 meters of a glacier’s terminus and as such were difficult to access without technology given the risks created by the glacier. The autonomous kayaks were essential to this project because of the hazards of working directly next to a glacier’s terminus, where calving pieces of ice can crash into the water without warning, producing life-threatening waves––or in the case of LeConte, from below the surface, too. The glacier is known for “shooter” icebergs that calve beneath the water and launch up to the surface, propelled by their buoyancy. 

A shooter glacier emerging from below the surface at Dawes Glacier, another glacier in Alaska. Source: AdventureM/ Youtube

Glaciologist and founding director of The Dasht Foundation, Faezeh M. Nick, who was not involved with the study, told GlacierHub “The autonomous kayaks taking measurements in front of the calving glaciers sound very promising. It has been very risky and expensive to get this kind of data at glacier fronts. It would be very beneficial if these kayaks are not too expensive and are robust enough to be used at several locations.” 

The kayaks used in the study were developed at the University of Oregon and are called Robotic Ocean Surface Samplers (ROSS). They were designed to function in harsh conditions, to be resilient in the face of unforeseen challenges due to redundancies of critical systems, and to be inexpensive enough that they could be used in areas with high risk of being lost.  

The kayaks’ base component is a commercially available Mokai gas-powered kayak. The researchers then built upon it, adding in the necessary control electronics, communications systems, and scientific instruments needed for the tasks it would be sent to accomplish. 

Kayaks have historically been associated with Indigenous peoples of the Arctic, though in earlier times were only found in the areas further north. Monitoring climate impacts in Alaska has brought scientific and Indigenous technologies together as people strive to understand the changes taking place on the planet. 

The data gathered by the autonomous kayaks show that ambient melting is a significant contributor to total melting at a glacier’s terminus and represents a large part of the total submarine melt flux. It revealed that ambient melt has been underestimated by a factor of up to 100. 

“We need these types of measurements being performed in front of several other glaciers in different regions before making a new statement about the general pattern or magnitude of submarine melt and its effect on sea-level rise,” Nick said. This finding increases scientists’ understanding of submarine glacier melt and opens the door for further research to establish a generalizable melt parameter for modeling ocean‐glacier interactions. As scientists’ understanding of glacier melt dynamics improves through studies like this one, they are one step closer to being able to generate predictive models on critical issues like sea-level rise with greater accuracy. 

Alaskan Glaciers Are Melting Twice as Fast as Models Predicted

Scientists from the University of Oregon recently found that the underwater section of a glacier in southeast Alaska is melting at rates up to two orders of magnitude greater than those predicted by theory.  The results, published in the journal Science, challenge the current models used to predict the melting of tidewater glaciers worldwide.

Tidewater glaciers play an important role in maintaining glacier stability, and their melting is accelerating overall ice loss in Greenland and in Antarctica. According to the study, no one has yet directly measured the melting of the underwater portion of a tidewater glacier. Instead, scientists have relied on untested theoretical models.

This diagram of a typical marine-terminating glacier in Greenland is from Oceans Melting Greenland (OMG), a project to investigate the extent to which the ocean is melting Greenland’s glaciers from below. (Source: NASA Jet Propulsion Laboratory)

Glaciers that terminate in the ocean come in two forms: Ice shelves, which are horizontal slabs of ice that extend into the ocean, and tidewater glaciers, which end in relatively vertical ice faces, Rebecca Jackson, one of the study’s authors, told GlacierHub.

Jackson explained that observing tidewater glaciers from below is difficult. Ice shelves retreat horizontally, and this can be seen by satellites and other remote sensing techniques. But the horizontal retreat of a tidewater glacier’s vertical face is too slight to be detected with remote sensing.

With funding from the National Science Foundation, the University of Oregon scientists used a new method to directly observe tidewater glacier melt: creating and comparing sonar images of the glacier over time. They studied the melting of Alaska’s LeConte Glacier, mainly because of its accessibility, Dave Sutherland, the lead author of the study, told GlacierHub. The physics of the interactions between glacier and ocean at LeConte are the same as in other systems around the world, including tidewater glaciers in Greenland, Patagonia, and the west Antarctic, he said.

They made observations at the glacier six times from 2016 to 2017, Sutherland told GlacierHub.

The glacier-ocean boundary at LeConte. (Source: Dave Sutherland, University of Oregon)

The results were striking. “We have direct observations that show melt rates are much higher than we we expected,” Jackson told GlacierHub.

This doesn’t change the amount of glacier ice currently being lost to the ocean, she explained. “Right now we know, to a pretty good degree, how much glaciers are losing ice and raising sea level,” Jackson said. “That’s a pretty well-documented quantity and our results don’t change that.”

Instead, the study illuminates what portion of the glacier ice being lost to the ocean is the result of underwater ice melting as opposed to calving—the process by which chunks of the glacier break off and float into the ocean as icebergs. “The sub-marine melt rates are higher than we expect, which means that the amount calving off is slightly less,” said Jackson.

That means that the ocean is playing a larger role than expected in the loss of water-terminating glacial ice, Sutherland told GlacierHub.

With a warming ocean, this news suggests that tidewater glaciers could disappear quicker in response to climate change than previously thought, Jackson explained. “There’s a hypothesis that ocean warming can increase submarine melting and then that triggers a glacier acceleration that deposits more ice overall into the ocean,” she said.

In other words, if the portion of glacier submerged in ocean water melts quicker, then the rate at which the glacier flows toward the ocean will increase, and the rate of calving will increase as well.

“You could imagine that if sub-marine melting was depth-dependent, you could undercut the glacier and destabilize it, leading to increased calving,” Sutherland told GlacierHub. And indeed their study found that the melt rates of the glacier were depth-dependent.

A picture of LeConte Glacier taken during the study. (Source: Dave Sutherland, University of Oregon)

“Ultimately what we want to be able to do is start answering questions like, if the ocean warms by one or two degrees, how will that affect the glacier?” said Jackson. And in order to do that, the faulty model must be replaced by a more accurate theory. The same team of scientists is currently working on that new model, Jackson said.

Although the Science study does not address why the previous theory might be incorrect, the scientists involved have a hypothesis, Jackson said. The velocity of the water touching a tidewater glacier face affects its melting rate: A higher velocity means a higher melt rate, just as pouring hot coffee over an ice cube melts the ice cube faster than placing it into the hot coffee. The current theory takes into account how freshwater that seeps out from the bottom of the glacier increases the water velocity as it rises to the top of ocean along the surface of the ice. However, it doesn’t take into account other drivers of ocean currents near the glacier, including wind, tide, and waves. “Those can also enhance the velocity along the ocean ice boundary, and that can also enhance melt rate,” Jackson told GlacierHub.

Collecting data from glaciers with different fjord conditions and glacier characteristics will provide the scientists with the information needed to model tidewater glacier melt as a function of the physical properties of the glacier and adjoining ocean. Since ocean conditions vary from season to season, the team is continuing to collect data at LeConte Glacier throughout the year, with the same goal of discovering how oceanic properties affect glacier melt. “One thing we’re excited about is what we present is the new method for directly measuring sub-marine melting that hopefully can be used at many other glaciers around the globe,” said Jackson.

In the mean time, estimates of sea level rise might need adjustment.

Footage of LeConte Glacier taken during the study. (Source: Dave Sutherland, University of Oregon)

Roundup: Kayaks, Snow Machines and Drones

Roundup: Kayaks, Regrowing Glaciers, and the Bowdoin

 

Research Using Remote-Controlled Kayaks

From Alaska Public Media: “LeConte Glacier near Petersburg… [is] the southern-most tide water glacier in the northern hemisphere and scientists have been studying it to give them a better idea of glacial retreat and sea level rise around the world… to get close to the glacier, which is constantly calving, a team of scientists is relying on unmanned, remote controlled kayaks… these kayaks have been completely tweaked by Marion and an ocean robotics team from Oregon State University… The boats are customized with a keel, antennas, lights and boxes of computer chips and wires.”

Find out more about the kayaks and research here.

LeConte Glacier’s calving front (Source: Gomada / Creative Commons)

 

Regrowing Morteratsch Glacier with Artificial Snow

From New Scientist: “The idea is to create artificial snow and blow it over the Morteratsch glacier in Switzerland each summer, hoping it will protect the ice and eventually cause the glacier to regrow… The locals had been inspired by stories that white fleece coverings on a smaller glacier called Diavolezzafirn had helped it to grow by up to 8 metres in 10 years… Oerlemans says it would take 4000 snow machines to do the job, producing snow by mixing air blasts with water, which cools down through expansion to create ice crystals. The hope is that the water can be “recycled” from small lakes of meltwater alongside the glacier… But the costs… are immense.”

Find out more about how this works here.

Snow cannons like this could help regrow Morteratsch Glacier (Source: Calyponte / Creative Commons)

 

Drones Capture a Major Calving Event

From The Cryosphere: “A high-resolution displacement field is inferred from UAV orthoimages (geometrically corrected for uniform scale) taken immediately before and after the initiation of a large fracture, which induced a major calving event… Modelling results reveal (i) that the crack was more than half-thickness deep, filled with water and getting irreversibly deeper when it was captured by the UAV and (ii) that the crack initiated in an area of high horizontal shear caused by a local basal bump immediately behind the current calving front… Our study demonstrates that the combination of UAV photogrammetry and ice flow modelling is a promising tool to horizontally and vertically track the propagation of fractures responsible for large calving events.”

Find out more about the study here.

Drones are increasingly being used to study glaciers (Source: Creative Commons)