Underwater glacial melting is occurring at higher rates than modeling predicts

University of Oregon-led team uses a new approach to directly measure submarine melt, uncovering data that could enhance projections of sea level changes

University of Oregon

VIDEO: Video made from time-lapse photos taken daily from March 31, 2016, to Aug. 8, 2016 from LeConte Glacier camp. The glacier moves from right to left, with the ice front,… view more

Credit: Video by Jason Amundson, University of Alaska Southeast

EUGENE, Ore. — July 25, 2019 — Researchers have developed a new method to allow for the first direct measurement of the submarine melt rate of a tidewater glacier, and, in doing so, they concluded that current theoretical models may be underestimating glacial melt by up to two orders of magnitude.

In a National Science Foundation-funded project, a team of scientists, led by University of Oregon oceanographer Dave Sutherland, studied the subsurface melting of the LeConte Glacier, which flows into LeConte Bay south of Juneau, Alaska.

The team’s findings, which could lead to improved forecasting of climate-driven sea level rise, were published in the July 26 issue of the journal Science.

Direct melting measurements previously have been made on ice shelves in Antarctica by boring through to the ice-ocean interface beneath. In the case of vertical-face glaciers terminating at the ocean, however, those techniques are not available.

“We don’t have that platform to be able to access the ice in this way,” said Sutherland, a professor in the UO’s Department of Earth Sciences. “Tidewater glaciers are always calving and moving very rapidly, and you don’t want to take a boat up there too closely.”

Most previous research on the underwater melting of glaciers relied on theoretical modeling, measuring conditions near the glaciers and then applying theory to predict melt rates. But this theory had never been directly tested.

“This theory is used widely in our field,” said study co-author Rebecca H. Jackson, an oceanographer at Rutgers University who was a postdoctoral researcher at Oregon State University during the project. “It’s used in glacier models to study questions like: how will the glacier respond if the ocean warms by one or two degrees?”

To test these models in the field, the research team of oceanographers and glaciologists deployed a multibeam sonar to scan the glacier’s ocean-ice interface from a fishing vessel six times in August 2016 and five times in May 2017.

The sonar allowed the team to image and profile large swaths of the underwater ice, where the glacier drains from the Stikine Icefield. Also gathered were data on the temperature, salinity and velocity of the water downstream from the glacier, which allowed the researchers to estimate the meltwater flow.

They then looked for changes in melt patterns that occurred between the August and May measurements.

“We measured both the ocean properties in front of the glacier and the melt rates, and we found that they are not related in the way we expected,” Jackson said. “These two sets of measurements show that melt rates are significantly, sometimes up to a factor of 100, higher than existing theory would predict.”

There are two main categories of glacial melt: discharge-driven and ambient melt. Subglacial discharge occurs when large volumes, or plumes, of buoyant meltwater are released below the glacier. The plume combines with surrounding water to pick up speed and volume as it rises up swiftly against the glacial face. The current steadily eats away from the glacier face, undercutting the glacier before eventually diffusing into the surrounding waters.

Most previous studies of ice-ocean interactions have focused on these discharge plumes. The plumes, however, typically affect only a narrow area of the glacier face, while ambient melt instead covers the rest of the glacier face.

Predictions have estimated ambient melt to be 10-100 times less than the discharge melt, and, as such, it is often disregarded as insignificant, said Sutherland, who heads the UO’s Oceans and Ice Lab.

The research team found that submarine melt rates were high across the glacier’s face over both of the seasons surveyed, and that the melt rate increases from spring to summer.

While the study focused on one marine-terminating glacier, Jackson said, the new approach should be useful to any researchers who are studying melt rates at other glaciers. That would help to improve projections of global sea level rise, she added.

“Future sea level rise is primarily determined by how much ice is stored in these ice sheets,” Sutherland said. “We are focusing on the ocean-ice interfaces, because that’s where the extra melt and ice is coming from that controls how fast ice is lost. To improve the modeling, we have to know more about where melting occurs and the feedbacks involved.”


Co-authors with Sutherland and Jackson were: Christian Kienholz, Jason M. Amundson and W. Pat Dryer of the University of Alaska Southeast, Juneau; Daniel Duncan of the University of Texas at Austin; Emily F. Eidam of the University of North Carolina; Roman J. Motyka of University of Alaska Southeast and Fairbanks; and Jonathan D. Nash of Oregon State University

From EurekAlert!.   (Includes Timelapse Video which I am unable to embed here).

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
July 27, 2019 2:30 am

That’s 3 years ago. What about today? I could well imagine that for many reasons there could have been an advance, retreat, thickening, thinning, etc. What’s the point of this post?

Kurt in Switzerland
July 27, 2019 2:31 am

Sounds like sensible research.

Rich Davis
Reply to  Kurt in Switzerland
July 27, 2019 6:13 am

Please. (EurekAlert!)

Accepting at face value, three years ago melt rates were up to 100x what models predicted. That kind of proves that what they are measuring is negligible in terms of sea level rise. It didn’t result in any acceleration of SLR.

July 27, 2019 2:37 am

This is make-work research. A vertical glacier face in water will lose ice by calving. Calving data is available for any number of glaciers. Calving will of course be quickened if the ice-front is undercut by melting, as it normally is. That is why calving icebergs will usually “fall over” outwards as the calve (and why you should never, ever take a boat close to a calving glacier front). The result is jagged, irregular icebergs.

The opposite process happens in very cold areas like Antarctica and northernmost Greenland where the glaciers instead tend to flow out to sea as a shelf ice or a glacier tongue. At least in Antarctica it can even happen that new ice forms below the shelf ice. In such an environment it is usually tidal action that breaks off large fragments of the ice and the result is the characteristic “plateau icebergs” of the southern ocean.

Kurt in Switzerland
Reply to  tty
July 27, 2019 3:32 am

But if the models being used are underestimating the melt rate, then it is only logical to obtain data to improve the models’ accuracy.

Unless you know how to guess the ratio of discharge melt to ambient melt at a particular glacier, e.g., as a function of latitude.

Kurt in Switzerland
Reply to  tty
July 27, 2019 3:36 am

But I doubt whether this will have significant impact on sea level rise rate projections.

Reply to  Kurt in Switzerland
July 27, 2019 5:56 am

Nope, the projections will remain ludicrously high.

Geoff Sherrington
Reply to  tty
July 27, 2019 3:40 am

Are you saying that this newish method of measurement of flux is redundant, because one can simply sum the volumes of the chunks of ice falling from the face?
In Australia, we have no glaciers or experts on local conditions, so please excuse the simplification and lack of usual terminology. Geoff S

Reply to  Geoff Sherrington
July 27, 2019 5:40 am

The flow rate and geometrical dimensions of the glacier outlet is sufficient. It has been used for many years. How large part of this discharge melts before as against after the calving moment is rather immaterial.

Reply to  tty
July 27, 2019 6:24 am

Thanks for this comment, as I was thinking the same and I don’t even have sociology degree but a science one. It didn’t make much sense to me.

Moderately Cross of East Anglia
Reply to  Geoff Sherrington
July 27, 2019 6:46 am

You mean all Australia’s glaciers have Gone! When did that happen – things must be far worse than we thought. Next we will discover a record high temperature in the U.K. based on a data tortured urban weather station…oh wait…
Do I come over as too cynical…?

Rich Davis
Reply to  Moderately Cross of East Anglia
July 27, 2019 3:52 pm

Too cynical? No, just moderately cross 🙂

spangled drongo
Reply to  Moderately Cross of East Anglia
July 27, 2019 10:29 pm

Yes, MC of EA, they’ve gone, but strangely a recent Oceanographer’s report on the Fort Denison tide gauge in Sydney Harbour is showing the current mean sea level as 6 cms lower than it was in 1914.

So where have they gone?

Must be the same place as all the others.

July 27, 2019 3:30 am

Two orders of magnitude error. The theory is falsified, again.

Reply to  Adrian Roman
July 27, 2019 4:14 am

new model estimating melt rate from measured parameters disagreed with old model. One or both is wrong.

Since the volume flow of the glacier is pretty clearly defined by section and linear speed, I don’t see why you need underwater sonar. There is nothing beyond the coast to restrain the downward flow so how can there be that much uncertainty. Does not make sense.

But “it’s worse than we thought” so it will certainly get grant money for next year.

July 27, 2019 3:31 am

The only logical conclusion of this study is that models do not reliably predict melt rate of glaciers. Inferring that a model, because it inaccurately understates melting from the only glacier whose melt was measured in the study, therefore understates melting of all glaciers is not rational. All you know is that, in your only test of the model’s accuracy, the model failed and it failed miserably.

Al Miller
Reply to  Kurt
July 27, 2019 4:16 pm

Ditto, the models are wrong, still and again. That is not news, but is something the MSM should be making the public aware of. Of course there’s no money in just being correct…

steve case
July 27, 2019 3:32 am

Underwater glacial melting is occurring at higher rates than modeling predicts

So the models were wrong. No news there.

I didn’t see where there was any claim that this is a new phenomenon. Wikipedia tells us that glacier retreat has been going on since at least 1850. I look forward to the “Worse than previously thought” head lines from our wonderful objective news media.

July 27, 2019 3:57 am

“EUGENE, Ore. — July 25, 2019 — Researchers have developed a new method to allow for the first direct measurement of the submarine melt rate of a tidewater glacier, and, in doing so, they concluded that current theoretical models may be underestimating glacial melt by up to two orders of magnitude.”

Our model, based upon one glacier, is better than their model(s)…

Obviously another “Give us more money” research announcement. Quelle Surprise!

Ron Long
July 27, 2019 4:00 am

My reading of the melting report suggests they are measuring ambient melting at the glacier-seawater interface. The ice of the glacier is already floating and in isostatic equilibrium as regards sea level, so who cares how it melts? If Glaciers melt backward from contact with the ocean and the melt water runs into the ocean you get a sea-level rise. The fact that the Leconte glacier is accumulating snow and converting it to ice in its head area and that it issues glaciers that flow to the sea, where the glacial ice melts, is everything normal as it should be. Lets be clear here, if enough snow is falling and accumulating as continental glaciers and glaciers run from them down to the sea, there is either not a net change in sea-level or some lowering. If continental glaciers are in dramatic retreat you are adding to seal level.

steve case
Reply to  Ron Long
July 27, 2019 5:41 am

Ron Long … Lets be clear here …
Greenland and Antarctica are well below freezing nearly everywhere nearly all the time. Their ice mass balance has everything to do with how much snow fell years, decades, centuries or perhaps millennia ago and nothing to do with global temperature. As such, headlines from our wonderful objective news media that say Greenland and Antarctica are melting faster than previously thought and sea levels blah… blah… blah… are always, without putting too fine a point on it, bullshit.

That’s not to say that sea level isn’t increasing, it is, and Greenland and Antarctica are probable sources of the increase, it’s just that global temperature isn’t a factor.

July 27, 2019 4:10 am

“We measured both the ocean properties in front of the glacier and the melt rates, and we found that they are not related in the way we expected,” Jackson said.

That is untrue, what they did was measure both the ocean properties in front of the glacier and ESTIMATE the melt rates from those measurements. They did NOT measure the melt rates. That means that somewhere they are using a model to estimate melt rates based on other measured parameters.

“sometimes up to a factor of 100, higher than existing theory would predict.”

That glacier is barely making it off the land limit. How can this be melting 100 times faster. You look at the section and measure the linear velocity and you have the volume melt rate. The linear speed of the glacier is easy to measure from the timelapse, so are they saying there are two orders of uncertainty in the cross-section.

This sounds like typical “it’s worse than we thought” grant seeking pseudo-science.

July 27, 2019 4:28 am

This sounds like the “we’ve discovered it for the first time so it must be happening for the first time” logical fallacy.

Meanwhile there is record cold in eastern Russia. A big place. Arctic temperatures are a zero sum game.


Reply to  Phil Salmon
July 27, 2019 8:42 am

“Arctic temperatures are a zero sum game.”

What is the basis of that assertion? The rather sparse measurements indicate it is generally warming on average: ie not a “zero sum game.”

If you wish to contest that, you need something better than a snappy phrase.

Reply to  Phil Salmon
July 27, 2019 4:42 pm

If the measurements are too sparse then you can prove neither that it’s OMG warming nor that it’s a zero sum game. Both are equally empty snappy phrases.

Mark Broderick
July 27, 2019 4:30 am

Wow, that is a fast moving glacier. Think that is why it melts faster and why they chose this glacier…

Reply to  Mark Broderick
July 27, 2019 6:04 pm

Yes – after looking at that time lapse, I agree completely. That glacier is being pushed rapidly into the ocean from the weight of the glacier above it, and it seems to me the only thing the study did was say that the ratio of ambient melt to plume melt was much higher than the models had assumed. But there was nothing in the press release that supported any conclusion that the total melt was greater from this glacier than what was previously assumed.

If you can look at a time lapse of a glacier flowing into an ocean, you should be able to get a good estimate of the annual discharge rate of water into the ocean. What proportion of that total melt amount manifests from rising plumes across the glacier face and what manifests from ambient melt may be an interesting academic question, and may ultimately improve modeling, but in terms of how much water is being fed into the ocean, it doesn’t tell you much.

July 27, 2019 4:37 am

Well, it is not like they didn’t discover anything new. They were able to determine that “the melt rate increases from spring to summer.” Of course, I suspect I could have determined THAT from my Lazy Boy recliner, with one and a half seconds of thought.

July 27, 2019 4:44 am

august and May only?
so may is summer starting and aug is sort of end summer
couldnt get near em to do a winter data run?
hmm too much ice then?;-)

July 27, 2019 4:46 am

Seriously?…Glaciologists are just now gathering salinity and temperatures in calving fields?

Unbelievable. We’ve had remote subs for decades.

steve case
Reply to  DocSiders
July 27, 2019 5:49 am

DocSiders …
Seriously?…Glaciologists are just now gathering salinity and temperatures in calving fields?

Unbelievable. We’ve had remote subs for decades.

Kinda makes you wonder where their billions in funding is being spent.

Total Funding for USGCRP

July 27, 2019 4:54 am

I would have more faith in the purity of their intentions if they showed all twelve months of the ice front. Choosing the period from yearly maximum extent to yearly minimum is… predictable.

steve case
Reply to  MST
July 27, 2019 5:52 am

MST I would have more faith in the purity of their intentions if they showed all twelve months of the ice front. …

My common sense says that glaciers calve into the sea year round, but I’ve never seen the monthly data, and I have looked.

Kurt in Switzerland
Reply to  MST
July 27, 2019 6:27 am

MST: how many months of the year do you reckon Alaska glaciers have significant discharge?
I would think May-October ought to cover it.

More important would be year to year (or longer term) fluctuations, which would be heavily affected by local weather.

Robert W. Turner
Reply to  MST
July 27, 2019 6:41 am

Team measures glacial melt water during melt water season only, determines that rate of ice melt might be twice that predicted for a full year. No surprise there. In climastrology, when you want to know long term trends and rates of change in a dynamic system over decades of time, you look at six months of summer data and extrapolate from there.

Walter J Horsting
July 27, 2019 5:16 am

Let’s not forget the 93 volcanoes under the Antarctic on it rift zone!

July 27, 2019 5:36 am

I don’t need any mangled theories or dubious research or expensive computer models to predict sea levels, I’ve got a straight ruler.

Reply to  MrGrimNasty
July 27, 2019 8:45 am

You forgot no “stinkin” mangled theories….

Gunga Din
July 27, 2019 6:05 am

So …. we hear about the model being wrong because it didn’t match actual observations.
(The melting is “worse than we thought!”)
When are we going to hear about the warming in the models not matching actual observations?
(The warming is not as high as we thought!)

steve case
July 27, 2019 6:14 am

MrGrimNasty … I don’t need any mangled theories or dubious research or expensive computer models to predict sea levels, I’ve got a straight ruler.
The Permanent Service for Mean Sea Level PSMSL lists about 1500 tide gauges, and they all measure sea level in millimeters.

July 27, 2019 6:25 am

So an error in estimation of glacial melting, up to 100 times too low, now explains mm/yr sea level rise? STOP THE PRESSES!!!

July 27, 2019 7:35 am

The higher you can get that melt rate to appear, the more headlines you get and the higher is your Grant Amount and Research Payout likely to be.

July 27, 2019 7:43 am

What theory are they saying was out by 100 times? …Certainly not Conservation of Energy….Certainly not Heat Transfer theory……so likely out on some “parameterizations”, euphemism for “best guess fudge factors”….

steve case
Reply to  DMacKenzie
July 27, 2019 8:47 am

What theory are they saying was out by 100 times?

Just remember that 100 times nearly nothing is still nearly nothing (-:

July 27, 2019 8:23 am

They have it backwards. In climate science models are used to determine data, not the other way around. It is imperative that they find a clever way to adjust the data to match the models.
Previously, wooden boats were used in exploration and they have since been replaced by steel boats. This must have had a profound effect on the data which hasn’t been adjusted for the change.

Keith S.
July 27, 2019 10:08 am

So I ask, what is Univ of Oregon’s dept of Earth Science? What exactly is that and what is it supposed to do? oceanography is at Oregon’s other state university, Oregon State University.

Loren Wilson
July 27, 2019 10:55 am

I am not sure about their mechanism. Where is the heat coming from to melt the glacier? The melt-water that flows out from the base or near the base of the glacier, especially the underwater portion, is going to be very close to 0°C, since it was at one time ice itself, and has been in contact with ice during its path from higher up the glacier to the face. Water at 0°C does not have any energy to melt ice which is also at 0°C or cooler. I agree that this water is less dense than the cold sea water it meets, and that it will tend to rise as it slowly mixes with the ocean water. However, I don’t think that it will melt much of the glacier. It may draw up some sea water that is above 0°C to the face of the glacier and promote some melting. It could also serve as a cold buffer between the warmer ocean water and the glacier. It depends on the flow rates, tides, geometry of the interface, etc. I would like to see some high-quality temperature measurements of the water as it passes through the glacier, and then at the face. Small robotic submersibles could be safely used at the face to collect data.

michael hart
July 27, 2019 1:22 pm

And because of ice density being so low as compared to liquid water, the first-order effect of underwater melting is to lower sea level, not raise it.

Of course that’s not the whole story, but it is not something one is likely to hear from the mouths of alarmists or the general media.

Peter Wilson
July 27, 2019 5:55 pm

The implication seem to be that the ice is melting 100 times faster than “predicted”, so sea levels will rise more.

But surely one doesn’t have to have any sophisticated measure of melt rate, when clearly the whole glacier disappears into the sea. How is simply measuring the volume of the disappeared glacier not sufficient to quantify the amount of ice melted into the sea?

July 29, 2019 7:52 am

Seaport officials should take note of this post by Charles The Moderator …”two orders of magnitude” is robust (Seaports With Sea Level Change – 3).

Johann Wundersamer
July 31, 2019 4:14 am

A certain volume of ice from the ocean interior rises and the ocean basin stabilises at the respective volume of salt saturated water at sea surface.

What’s the point.

%d bloggers like this:
Verified by MonsterInsights