Study: 'glaciers don't need periods of warming to trigger retreat'

A new model yields insights into glaciers’ retreats and advances, shows cyclic nature of tidewater glaciers


A University of Alaska Fairbanks study looking at the physics of tidewater glaciers has yielded new insights into what drives their retreat-and-advance cycles and the role that climate plays in these cycles.

Lead author and UAF geophysics doctoral student Douglas Brinkerhoff said the study in Nature Communications reveals that shifting sediments drive the cycles among tidewater glaciers in temperate climates such as southern Alaska.

The study also reveals that these glaciers don’t need periods of warming within the temperate climates to trigger the glacier’s retreat, as previously thought.

“Tidewater glaciers can advance further into the ocean by riding on top of a pile of their own sediment, but this can get them into trouble,” said Brinkerhoff. “Eventually the glacier snout slows down, but it continues to push that sediment pile further out to sea, essentially pulling the rug out from under itself. When the glacier floats, there’s no resistance at its base holding it back and the whole floating part tends to disintegrate.”

The findings stem from a mathematical model that Brinkerhoff developed to better understand the periodic retreats and advances of glaciers that flow into the ocean. One-third of Alaska’s approximately 60 tidewater glaciers are advancing as part of this cycle, despite the otherwise widespread glacier loss. The advances usually last several centuries, while the retreats take only decades.

The terminus of the Taku Glacier, in the images at left, advanced more than three miles from 1933 to 2016, as reflected by the yellow and blue lines superimposed on the photographs. The location of the red dot on a mountain in the top image matches the location of the righthand red dot in the lower image. The images and lines at right show the Columbia Glacier’s retreat from 1985 to 2016. Image courtesy of Douglas Brinkerhoff, from photography by the US Navy, the National Snow and Ice Data Center, Martin Truffer and the Landsat 7 and 8 satellites via US Geological Survey

Co-author Martin Truffer, a glaciologist with UAF’s Geophysical Institute, said that glaciologist Austin Post first noted the tidewater glacier cycles in the 1970s. He saw that sediment piles developed in the ocean in front of advancing glaciers. These piles allowed the glaciers to extend their advance, followed by often catastrophic collapse and retreat.

“While the work of Austin Post and others clearly showed that the erosion, evacuation and deposition of glacial sediments play an important role, the work reported here managed, for the first time, to capture all the relevant processes in one single model,” said Truffer.

Brinkerhoff said that many scientists believed that warming periods within the temperate climate triggered the extended glacier to collapse. They thought that the extended part of the glacier was vulnerable to slight periods of warming because it was flatter and at sea level.

Illustration by Meghan Murphy This illustration shows a tidewater glacier slowly advancing on a sediment pile. The melting glacier terminus eventually begins to erode the sediment pile. That undercuts the ice’s support and triggers a rapid collapse, in a process explained by a new model developed by a University of Alaska Fairbanks researcher.

But the model shows that the extended portion of the glacier would collapse even without these warming periods because the glacier erodes the pile supporting it.

Co-author Andy Aschwanden, a modeler with UAF’s Geophysical Institute, said that tidewater glaciers in climates colder than Alaska, such as southeastern Greenland or the northern Antarctic Peninsula, generally do not exhibit a cycle yet.

“The model suggests if temperatures keep rising some of the tidewater glaciers in colder areas may start to advance, even though that seems counterintuitive,” Aschwanden said.

To develop and test the model, Brinkerhoff used observations from glaciers such as Prince William Sound’s Columbia Glacier, which started retreating in 1985. He said they also looked at advancing tidewater glaciers such as Hubbard Glacier, which has a submarine pile in front of it, and Taku Glacier near Juneau, the terminus of which now rests on an exposed pile of sediment.


The Paper:

Sediment transport drives tidewater glacier periodicity

Douglas Brinkerhoff, Martin Truffer & Andy Aschwanden


Most of Earth’s glaciers are retreating, but some tidewater glaciers are advancing despite increasing temperatures and contrary to their neighbors. This can be explained by the coupling of ice and sediment dynamics: a shoal forms at the glacier terminus, reducing ice discharge and causing advance towards an unstable configuration followed by abrupt retreat, in a process known as the tidewater glacier cycle. Here we use a numerical model calibrated with observations to show that interactions between ice flow, glacial erosion, and sediment transport drive these cycles, which occur independent of climate variations. Water availability controls cycle period and amplitude, and enhanced melt from future warming could trigger advance even in glaciers that are steady or retreating, complicating interpretations of glacier response to climate change. The resulting shifts in sediment and meltwater delivery from changes in glacier configuration may impact interpretations of marine sediments, fjord geochemistry, and marine ecosystems.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Roger Knights
July 26, 2017 10:37 am

Cycles. Whaddaya know!

July 26, 2017 10:52 am

It should be obvious to anyone paying attention that the alarmists take any, any natural phenomena and 1, Declare it catastrophic and 2. Attribute it to man’s use of fossil fuels. Man is but a tiny blip of existence on this planet and has done a good job of cleaning up after itself.

Th3o More
Reply to  markl
July 26, 2017 8:48 pm

The fact that “Glo-buncha-bull warming” is false is not that hard to find evidence for. The idea that mankind has done a good job of cleaning up after itself is much harder to make a case for. To the extent that trash is visible it often is cleaned up. Measuring pollutants in the depth of a stream bed can show us how poorly we are doing at cleaning up the part of the pollution that is not visible.

July 26, 2017 10:54 am

Reality is chipping away at the supposed evidence of CAGW. This work goes along with evidence that glaciers in Antarctica are melting because of geothermal heat. link
I can’t think of any alarmist claims that can’t be chipped away at because of inconvenient facts. The trouble is that people will remember the dramatic picture of a glacier calving and they won’t remember the fact that it’s a natural expected occurrence.

Rob Dawg
July 26, 2017 10:57 am

Must be an observation error. Cycles? None of the models incorporate cycles. Cardinal Wiggles! Break out the [da ta dah] “snowy coverlet.” We shall make the data CONFESS!

Reply to  Rob Dawg
July 26, 2017 1:08 pm

What no “pointy cushion”?

July 26, 2017 11:02 am

How dare modelers commit science!

Will Nelson
July 26, 2017 11:10 am

The evidence has been around for awhile: it is the time-traveling CO2…comment image

Don Gleason
July 26, 2017 11:13 am

I just spent an incredible week in SE Alaska, and visited several of the fjords carved by glaciers. When George Vancouver visited the Glacier Bay area in the 1790’s, the glacier extended to the mouth of the fjord. When John Muir visited in 1870’s (?), the glacier had retreated over 60 miles. Also of note, the water depth at the mouth of the fjord is in the 300-400 foot range. Further in, it’s over 1400 feet, which demonstrates the mechanism in the paper. This retreat of the glacier is locally attributed to the ending of the LIA. Note further that in a week surrounded by naturalists, I did not once hear the expression “global warming”

Jim Allison
Reply to  Don Gleason
July 26, 2017 2:08 pm

You beat me to it! I, too, just returned from a two-week Alaska trip. The naturalists aboard our ship pointed out the glacial growth and retreat of the several area glaciers did not follow any common pattern. Some of the mountain glaciers are currently growing, and the ocean-edge glaciers are calving because they are growing.

Ron Long
Reply to  Don Gleason
July 26, 2017 2:21 pm

Excellent Don and Jim. Nothing like being face-to-face with reality.

michael hart
July 26, 2017 11:25 am

Bizzare, really, that they are now in the process of rediscovering what I thought they already knew about glaciers several decades ago: Glacial movement, mass gain, or mass loss, is extremely complex above and beyond simple short term considerations of temperature and precipitation. Just what the hell have we been paying for in the last ~4 decades?

Reply to  michael hart
July 26, 2017 11:34 am

Socialist indoctrination ?

July 26, 2017 11:26 am

“but some tidewater glaciers are advancing ………. contrary to their neighbors.”
Temperature has nothing to do with it

Reply to  Latitude
July 26, 2017 1:14 pm

Global Warming both the bane and boon of the century! It causes glaciers to advance and retreat. Creates more violent weather events and fewer events. Causes sea level to both rise and fall.
“It’s a tasty dessert topping and a floor wax!”

Reply to  rocketscientist
July 26, 2017 1:17 pm

Manmade climate change means that whatever weather occurs, it’s our fault and always bad. Whether flood or drought, hot spell or cold snap, storm or calm, too much ice or too little, it’s all dangerous anthropogenic “climate change”.

James Francisco
Reply to  rocketscientist
July 27, 2017 5:26 am

It’s no Bass-O-Matic though.

July 26, 2017 11:29 am

Snow falls in the interior and maybe a century later it calves into the sea as an iceberg. Cause and effect separated by several decades. I doubt that warmer weather at the terminus has much to do with it.

Reply to  Steve Case
July 26, 2017 5:13 pm

Additionally, calving rates were not strongly related to forces that primarily act on the glacier’s surface, including rainfall, temperature, and wind speed.

July 26, 2017 11:31 am

From the article: “despite increasing temperatures”
There’s that unsupported claim again.
Depending on where you choose to start, the temperatures are not increasing, they have been decreasing.
The authors of this study have visions of Hockey Stick charts in their heads. That’s their proof that it is warming. Their proof is really a lie. I assume they are dupes and are unaware of this. Let’s hope so.

July 26, 2017 11:32 am

Could someone please explain to me why we care about GIANT CHUCKS of ICE that cannot support life ??

Reply to  Butch2
July 26, 2017 11:55 am

Glaciers do support life. Not that that is necessarily a reason to care about them.

July 26, 2017 11:33 am

So we knew this decades ago, and now we know it with more precision than before. Another small climb down from ‘its all AGW’. To be in Nature is an important climb down signal.

July 26, 2017 11:42 am

Talking about ice >>>>>>>
How is every ones prediction on the Arctic sea ice turning out so far.
Looks like it is still leaning towards a 2012 outcome …
Early days, however August is usually the decider.

Reply to  ozonebust
July 26, 2017 11:53 am

Is that you, Griffiepoo?
Why do you say that it’s headed for 2012, then that August is the decider, which latter statement is true?
Most years converge to about the same extent in late July and early August. What determines September extent is storminess in August. If there is a big cyclone, sea ice extent will be lower. If more than one, lower still.
Right now, Arctic sea ice is higher than in 2012 and 2011 but lower than in 2013 and 2014. Even with a storm, lower than 2012 in September seems unlikely. This year will probably be lower than 2009, 2013 and 2014, the highs of the past decade, but above 2012, the record low.

Reply to  Gloateus
July 26, 2017 12:10 pm

No not Griff. I dont know why you would assume that.
Although I do enjoy his comments, as it gives others a reason to comment.

Reply to  Gloateus
July 26, 2017 12:18 pm

Reduced chance of an Arctic cyclone this year as NH pressure not very high, SH even lower.
That is why there has not been a lot of hurricane / cyclone activity so far. When they start up there is not a lot to support them.
A forecast —
Global cyclone / huricane ACE this year will be below mean – no energy in system.
Antarctic Blozone hole area and duration will be about same as 2012, at best

Reply to  Gloateus
July 26, 2017 12:35 pm

Without Arctic cyclones, sea ice extent can’t get as low as in 2012. The other low years, eg 2007 and 2016, also saw at least one cyclone. The Super El Nino also affected 2015 and 2016 Arctic sea ice.
However, even lacking a cyclone next month, Arctic sea ice this year probably won’t return to its recent highs, ie 2009, 2013 and 2014. So, maybe somewhere between 4.1 and 4.6 million sq. km., but probably not the 3.3 million of 2012.

Roger Knights
Reply to  Gloateus
July 26, 2017 3:21 pm

I guess there won’t be a thread this year where we can cast our votes on the minimum extent number and compare it to the predictions of othere climate commenter sites and organizations.

Reply to  Gloateus
July 27, 2017 11:26 am

ozonebust July 26, 2017 at 12:18 pm
Reduced chance of an Arctic cyclone this year as NH pressure not very high, SH even lower.
That is why there has not been a lot of hurricane / cyclone activity so far. When they start up there is not a lot to support them.
A forecast —
Global cyclone / huricane ACE this year will be below mean – no energy in system.

And yet a week ago there were eight tropical cyclones simultaneously active in the North Pacific.

Reply to  Gloateus
July 27, 2017 3:19 pm

Yes you are correct but nothing of substance that progress to higher value hurricanes.
They rise and fall tropical storm s.
There is no real pressure and not a lot of available atmosphere for transport.
As soon as these ts were active nh tropospheric pressure drops.

Reply to  Gloateus
July 31, 2017 11:45 am

ozonebust July 27, 2017 at 3:19 pm
Yes you are correct but nothing of substance that progress to higher value hurricanes.
They rise and fall tropical storm s.

I guess I imagined super typhoon Noru this weekend!

Reply to  ozonebust
July 26, 2017 12:13 pm

Because Griffiepoo predicted a new record low for Arctic sea ice extent this year.
And every time he, she or it commented upon the progress of the melt season, sea ice melting promptly slowed. Uncanny.
We”ll see if you have the same magic touch. If so, this year’s extent will soon cross over yet more years at the same date.

Reply to  Gloateus
July 26, 2017 12:46 pm

I was not predicting as low as 2012, just saying that it is hugging the same line currently.
No majic touch here, but note the predictions on wind / energy related issues.

Reply to  Gloateus
July 26, 2017 12:47 pm


Reply to  Gloateus
July 26, 2017 12:54 pm

It is also hugging the lines of other years, just as this year has those of other years previously. The lines converge at this time of year, then diverge in early August, depending upon the weather.
Actually 2012 started diverging in late July.

Reply to  Gloateus
July 26, 2017 1:03 pm

This month, high winds and rain across the Bering Strait have already affected Arctic sea ice, but if they let up next month, record lows can be avoided.

Dave Fair
Reply to  Gloateus
July 26, 2017 1:28 pm

Get with the times, Gloateus: It is not “… he, she or it …”; it is S/H/It.

Reply to  Gloateus
July 26, 2017 1:31 pm

My bad. I’m not ready for the 21st century, it appears.

Joe - the non climate scientist
July 26, 2017 11:55 am

One observation from numerous “the Glaciers are Melting”
They almost always show a late 1800’s or early 1900’s photo along side a current photo to show all the retreat with the implication that the retreat all occurred in the very recent past.
Rarely do the show a series of photos (by decade or over 5-10 years). On the rare occasions when a time series of photos are provided, they almost show that most of the retreat was in the 1920/1930’s (at least the North american glaciers)
That comes across as cherry picking data to create a false impression

Reply to  Joe - the non climate scientist
July 26, 2017 12:01 pm

Everything goes back to the 1930’s.

Reply to  Joe - the non climate scientist
July 26, 2017 12:29 pm

“They almost always show a late 1800’s or early 1900’s photo along side a current photo to show all the retreat with the implication that the retreat all occurred in the very recent past.”
Indeed. This is particularly true for Jakobshavn/Ilulissat glacier in Greenland a favorite poster child for “Global Warming”. Usually you see an image like this:
However there exists a photograph taken by USAF in 1946 that shows that the glacier front was then behind the 2001 position. It’s not easy to find, but it is on p. 141 of this paper:

Reply to  tty
July 26, 2017 12:40 pm

Also doesn’t show the retreat between 1851 and 1902.

Reply to  tty
July 26, 2017 1:15 pm

If you are interested in more details about the waxing and waning of Greenland glaciers, which is quite complex, with one glacier advancing while another in the next valley retreats, I recommend this, written by real glaciologists:

Reply to  tty
July 26, 2017 1:27 pm

Awesome! Thanks. Real glaciologists are of no interest to “climate science”. Only retreating glaciers count!
Concluding Remarks:
Radiocarbon ages of marine material, transported from
the subsurface of the West Greenland ice sheet to its
margin, demonstrate a widespread recession of the ice
margin during the Holocene. Dated shells from the Qassimiut
lobe show that the ice margin was located behind
the present margin as late as 2.9 ka BP (Weidick et al.
2004) and at Kangilinnguata Sermia at 4.3 ka BP (this
study). At the southern part of Jakobshavn Isbræ, the
ice margin was behind the present margin at 2.2 ka BP
(Weidick & Bennike 2007) and at the northern part of
this ice stream at 0.4 ka BP (Briner et al. 2010). These
investigations focused on ice-free areas adjacent to the
largest calf-ice-producing outlets from the ice sheet, and
the behaviour of other parts of the ice-sheet margin may
have been different.
The fluctuations of the ice-sheet margin are still poorly
documented, especially before the 1900s. However,
from the sparse data available we can conclude that the
Inland Ice margin had a retracted position, not only at
the peak of the Holocene thermal maximum, but also in
the following millennia, at least in some lowland areas.
Local deviations may have occurred, such as the advance
during the Narssarssuaq stade (around 2000 years ago).
This advance may have been related to the first large cold
spell after the Holocene thermal maximum, as shown in
the temperature history inferred from the ice core from
Camp Century (Fig. 35; Dansgaard 2004). A cold period
is recognised at 2000 years BP, which could have initiated
the Narssarssuaq stade that only locally went beyond
the present extent of the ice margin. The subsequent cold
period of the Little Ice Age seems to have two minima at
about AD 1550 and AD 1850 (Dahl-Jensen et al. 1998).
Traces of the early advances of the ice margin have been
obliterated by later advances.
With respect to the response of the individual sectors
of the Inland Ice margin to climatic changes it must first
be stressed that the current information about amplitude
and exact dating of culmination is often uncertain, and
that the trend of the fluctuations, even for the relatively
well-documented last century shows a high degree of
variability (Fig. 36). It is clear that if the trimline zones
around outlets indicate the maximum Little Ice Age extent
of the glacier, then localities without trimline zones
must delineate areas where the present-day position coincides
with the Little Ice Age maximum. It is concluded
that the Little Ice Age maxima show a spread of ages,
although for minor lowland outlets, the majority are related
to the youngest Little Ice Age maximum from the
mid- or end of the 1800s.
With respect to outlets without a trimline zone, it
should be mentioned that some of these (Narsap Sermia,
Sermeq), at the end of the 1900s and the beginning of
this millennium, seem to show an initial recession. In
Fig. 36 the fluctuations of the outlets and ice margin are
shown. If this picture is compared with a map showing
the topography of the ice-free coastland and the subglacial
topography of the adjoining parts of the ice sheet, it
can be suggested that the large fluctuations of Kangiata
Nunaata Sermia may be related to the presence of a large
valley system that is evident under this ice stream (Fig.
37), and therefore to the mass balance and the dynamics
of this ice stream.
Otherwise outlets without trimline zones, such as
Saqqap Sermersua in the north and Isortuarsuup Sermia
in the south, seem to be related to nearby uplands and
highlands although this can scarcely explain the steady
advance of a local outlet. As regards the role of glaciers
and their fluctuations as a ‘climatoscope’ for climate
change in general, this seems to have been most successful
for minor local glaciers where a well-defined delineation
of the glacier form favours a more direct expression
of the ruling local climatic and mass-balance conditions
(Leclercq 2012; Leclercq et al. 2012). In contrast, the accumulation
(catchment) area and ablation area within
specific sectors of the Inland Ice may not be constant
since the boundaries of these areas may shift position as
a result of changing climatic conditions in combination
with a series of factors that influence the dynamics and
response time of the outlet. Such factors include surface
elevation, bed elevation, local and temporal variations in
snow fall, mass balance, basal ice temperature, depth of
the transition between Weichselian and Holocene age
ice, curvature of the surface contours and slope aspect of
the surface terrain (listed for modelling of the Inland Ice
dynamics by Ahlstrøm et al. 2008, p. 24). The specific
gaps in understanding and modelling of the ice-margin
response to climatic change are also dealt with by DahlJensen
et al. (2009); the observations and information
discussed here can serve as examples of the strong variability
in the response of the individual segments of the
ice-sheet margin. This does not, however, weaken the
credibility of estimates of the total mass balance of the
entire Inland Ice. It is more related to the problem of
prognoses of the response time of the individual sectors,
where detailed short-term predictions are required to
evaluate glacier hazards with respect to future exploration
for minerals or hydropower.

Reply to  tty
July 26, 2017 1:30 pm

Not a lot of aerial photographs taken in the arctic before 1904 (none), actually precious few aerial photographs of any where before then (1st in1858). And, even those were made with kites.

Reply to  tty
July 26, 2017 1:36 pm
Reply to  tty
July 27, 2017 1:32 am

And if yor read this:
You will see that historical data show that in c. 1750 the Illulissat glacier front was much in the same place as today, but that it then advanced to a maximum in c. 1850 (which is where the point the map always starts from)

July 26, 2017 12:21 pm

comment image
“Taking a more synoptic view, Fig. 4a (above) shows the terminus position over 3 kyr. The glacier undergoes a sawtooth oscillation with an amplitude of 6.5 km and a period of 326 years.”
Off hand I don’t know of natural cycle of that length.
Note of caution however: Maunder Minimum ended in 1705,
1705 + 326 = 2031
Blue line in this graph hits minimum in 2034. No, I’m not saying we are heading for a new Maunder minimum, that would be ‘cyclomania’ would it not? It’s written in the ice, is it not? :)

Clyde Spencer
July 26, 2017 1:18 pm

Taken from the paper: “When the glacier floats, there’s no resistance at its base holding it back and the whole floating part tends to disintegrate.” Yet, one routinely sees the claim that it is the floating ice shelves in Antarctica that are holding the glaciers back, and calving will remove the buttressing and allow the glaciers to surge forward. I guess there are different schools of thought on this! How can that be when the “science is settled?”

July 26, 2017 2:12 pm

There’s also noise about big chunks of Antarctic glaciers calving off. When you have trillions of tons of ice cantilevered off the edge of a continent and continuing to grow from the continent, floating on an ocean that rises and falls with tides and storms, wouldn’t you expect the portion over the water to inevitably break off and separate in a periodic manner?

Don Easterbrook
July 26, 2017 2:47 pm

Glaciers are equilibrium systems governed by glacier geophysics. Glacier models are governed by inputs and assumptions by computer modelers. I’ve found that real glaciers give far better data than models.
Tidewater glaciers are special critters, because of the wide variety of conditions at the termini. Long glaciers close to sea level with floating termini have to be studied on an individual basis–one rule doesn’t serve all. So I would challenge modelers to account for all real glacier behavior with any single model.

Gunga Din
July 26, 2017 8:12 pm

(Forgive me for modifying an older quip.)
“When glaciers recede, alarmist have a cow.
That explains all the bellowing!”

Gunga Din
Reply to  Gunga Din
July 26, 2017 8:15 pm

PS The original was “calve” instead of “recede”.
But the outcome is the same. 😎

July 27, 2017 3:34 am

It’s a model folks. That means it is dependant on inputs.

July 27, 2017 6:36 am

good article

July 27, 2017 6:54 am

“Winds from Siberia reduce Arctic sea ice cover, Norwegian researchers find”

July 28, 2017 12:21 am

Disappearing glaciers? Not in Antarctica
According to the new analysis of satellite data, the Antarctic ice sheet showed a net gain of 112 billion tons of ice a year from 1992 to 2001. That net gain slowed to 82 billion tons of ice per year between 2003 and 2008.
Not in the Northern Hemisphere
Winter snow and ice cover trend for the Northern Hemisphere has in fact been trending upwards since statistics started.

July 28, 2017 2:17 am

Anyone who has ever had ice cubes in a frost free freezer for very long knows that ice doesn’t just melt – it also “evaporates” – it’s called sublimation. It’s what really rock solid ice does on Iowa’s roads in the winter. Day by day it shrinks – but not from melting and not from wear. When it’s bitter cold and dry, it sublimates. Happens to glaciers, too.

Verified by MonsterInsights