Vintage film shows Thwaites Glacier ice shelf melting faster than previously observed

Stanford’s School of Earth, Energy & Environmental Sciences

Professor Dustin Schroeder (foreground) and art historian Jessica Daniel splice 50-year-old film containing radar measurements of Antarctica into a reel in preparation for digital scanning at the Scott Polar Research Institute in the UK. Credit Courtesy of Dustin Schroeder

Professor Dustin Schroeder (foreground) and art historian Jessica Daniel splice 50-year-old film containing radar measurements of Antarctica into a reel in preparation for digital scanning at the Scott Polar Research Institute in the UK. Credit Courtesy of Dustin Schroeder

Newly digitized vintage film has doubled how far back scientists can peer into the history of underground ice in Antarctica, and revealed that an ice shelf on Thwaites Glacier in West Antarctica is being thawed by a warming ocean more quickly than previously thought. This finding contributes to predictions for sea-level rise that would impact coastal communities around the world.

The researchers made their findings by comparing ice-penetrating radar records of Thwaites Glacier with modern data. The research appeared in Proceedings of the National Academy of Sciences Sept. 2.

“By having this record, we can now see these areas where the ice shelf is getting thinnest and could break through,” said lead author Dustin Schroeder, an assistant professor of geophysics at Stanford University’s School of Earth, Energy & Environmental Sciences (Stanford Earth) who led efforts to digitize the historical data from airborne surveys conducted in the 1970s. “This is a pretty hard-to-get-to area and we’re really lucky that they happened to fly across this ice shelf.”

Researchers digitized about 250,000 flight miles of Antarctic radar data originally captured on 35mm optical film between 1971 and 1979 as part of a collaboration between Stanford and the Scott Polar Research Institute (SPRI) at Cambridge University in the U.K. The data has been released to an online public archive through Stanford Libraries, enabling other scientists to compare it with modern radar data in order to understand long-term changes in ice thickness, features within glaciers and baseline conditions over 40 years.

Sea-level predictions

The information provided by historic records will help efforts like the Intergovernmental Panel on Climate Change (IPCC) in its goal of projecting climate and sea-level rise for the next 100 years. By being able to look back 40 to 50 years at subsurface conditions rather than just the 10 to 20 years provided by modern data, scientists can better understand what has happened in the past and make more accurate projections about the future, Schroeder said.

“You can really see the geometry over this long period of time, how these ocean currents have melted the ice shelf – not just in general, but exactly where and how,” said Schroeder, who is also a faculty affiliate at the Stanford Woods Institute for the Environment. “When we model ice sheet behavior and sea-level projections into the future, we need to understand the processes at the base of the ice sheet that made the changes we’re seeing.”

The film was originally recorded in an exploratory survey using ice-penetrating radar, a technique still used today to capture information from the surface through the bottom of the ice sheet. The radar shows mountains, volcanoes and lakes beneath the surface of Antarctica, as well as layers inside the ice sheet that reveal the history of climate and flow.

Newly uncovered features

The researchers identified several features beneath the ice sheet that had previously only been observed in modern data, including ash layers from past volcanic eruptions captured inside the ice and channels where water from beneath the ice sheet is eroding the bottom of ice shelves. They also found that one of these channels had a stable geometry for over 40 years, information that contrasts their findings about the Thwaites Glacier ice shelf, which has thinned from 10 to 33 percent between 1978 and 2009.

“The fact that we were able to have one ice shelf where we can say, ‘Look, it’s pretty much stable. And here, there’s significant change’ – that gives us more confidence in the results about Thwaites,” Schroeder said.

The scientists hope their findings demonstrate the value of comparing this historical information to modern data to analyze different aspects of Antarctica at a finer scale. In addition to the radar data, the Stanford Digital Repository includes photographs of the notebooks from the flight operators, an international consortium of American, British and Danish geoscientists. (Read more about the process of digitizing the archival film.)

“It was surprising how good the old data is,” Schroeder said. “They were very careful and thoughtful engineers and it’s much richer, more modern looking, than you would think.”


Schroeder also holds a courtesy appointment in electrical engineering. Stanford co-authors include postdoctoral researchers Winnie Chu and Matthew Siegfried, now with the Colorado School of Mines; graduate students Emma MacKie, Katherine Vega and John Emmons; and assistant professor of computer science Keith Winstein. Other co-authors are affiliated with the University of Cambridge, Imperial College London and the University of Edinburgh.

The research was supported by NASA, NSF and the George Thompson Fellowship at Stanford University.

From EurekAlert!

81 thoughts on “Vintage film shows Thwaites Glacier ice shelf melting faster than previously observed

  1. ‘“It was surprising how good the old data is,” Schroeder said. “They were very careful and thoughtful engineers and it’s much richer, more modern looking, than you would think.”‘ am I alone in thinking this is somewhat patronising?
    When I saw vintage film I thought they’d obtained film from the early part of the 20th century, not from the post Apollo era.

    • John
      While technology has advanced, I’m of the opinion that the competence of many who use the technology has declined. I’m reminded of the old joke that every generation of teenagers thinks that it is they who first discovered sex.

      • If you can’t buy the technology off the shelf from Fisher, the modern scientist is at a dead end. And if it breaks they need somebody else to fix it. Research trends are now strongly controlled by readily available technology. Most modern graduate students are completely baffled by the concept that they might build their own instruments.

        • Quite a few years ago, I grumbled to an old physicist that the new PhDs coming up seemed to be lacking something. He agreed and put it down to the fact that they didn’t have to build their own equipment. Bingo!

          You can train someone ’til you’re blue in the face but they won’t have a clue until they’re forced to deal with physical reality.

      • In 2009 as I approached my 59th birthday I young acquaintance attending the University of Maryland informed me pedantically that I didn’t know anything about Karl Marx because he, the young acquaintance, had only just learned about him in school.

        • Daughter in late primary school noting how we watched classic black and white movies and the same with old photo albums asked seriously- When did the world turn to colour mum?

      • We’re standing on the shoulders of giants.

        Every generation to now was the avant-garde, they had to cope with never known challenges – and of course they had to regard themselves as the advanced, most modern generation ever.

        They “gave their lives” for “a better future”.

        Never forget our heroes.

      • “…every generation of teenagers thinks that it is they who first discovered sex.”

        You mean…we weren’t?

    • They have no clue.

      One of my hobbies is military technology. When I look at some of the things they did in WW2, without the aid of modern test equipment, I never cease to be amazed at how smart those guys were.

      I am amazed that modern writers assume that ancient people couldn’t think of the obvious. Usually, it takes a proper historian to set them right.

      People in historical times were every bit as smart as we are. We think we’re smart because we have a bunch of modern technology that makes things easy. We have the illusion of knowledge. It’s a serious handicap.

      • It boggles the mind to recall graduating in 4 years with a BS in Ch E w/o calculators, cell phones, or Google using just a Post slide rule. Thankfully hp calculators came along in time for grad school. When on the job, In the late ’70s it was really cool to use a Fortran program to modify and improve a process flowsheet for plutonium nitrate. The original process had been up and running for many years in a remotely-operated “concrete canyon” with 5-foot thick walls to shield the operators and engineers from radioactivity. All maintenance was done remotely with specially-designed tools.

        The WWII developments were truly amazing – the Lancaster, B-29, jet & rocket technology, and the list goes on.

    • “They were very careful and thoughtful engineers and it’s much richer, more modern looking, than you would think.”‘

      I’m thinking how unlikely it is that this sort of thing will ever be said about the ‘modern’ data.

      • or modern machinery!
        Ive just spent 2 days pulling a series of line trimmers to bits trying to find bits that arent bloody plastic and worn or splitting
        only ONE unit was made of decent products and its so good I cant get it dis assembled;-/

    • This is one of his many books, one of latest goes into why the engineering profession goes downhill corresponding to their professional career. He writes a column for American Scientist. Much of the rest of it is going the way of Scientific American. Brooklyn Bridge still standing.

      Petroski, Henry. 2010. The essential engineer : why science alone will not solve our global problems . New York : Alfred A. Knopf.

      While the movie industry has been preserving these all the way back, science has been remiss, maybe this is a wake-up that science didn’t start in this millennium.

    • Thank goodness they could analyse photographs because all the “climate experts” tell us that they were too dumb to measure temperature correctly. So bad in fact, that all their temperature data required homogenization.

    • yeah its a bit weird, things I still use are being sold as “vintage” or even “antique”
      makes me laugh and also feel very old

    • If you want to see some eye-popping “old data” taking, check out “Recollections and Reminiscences at the 25th Anniversary of THE FIRST WEIGHING OF PLUTONIUM.”

      It’s simply astounding how we went from Chadwick’s discovery of the neutron in 1932 to the atomic bomb in 1945, going through steps such as those recalled in this document. It is well worth reading. Anyone who thinks climate “science” is actually a science should take note.

  2. ““It was surprising how good the old data is,” Schroeder said. “They were very careful and thoughtful engineers and it’s much richer, more modern looking, than you would think.””

    No, more than some clueless millenial would think. They built the world you’re living in and gave you the privilege and ease of life no other people on earth have ever known.

    • It exposes the mindset that allows modern “thermometrists” the rationalization for adjusting 80-100 year old temperature data. Our great grandfathers just didn’t know how to make an accurate thermometer apparently in 1910-1940 or know then how to read it.

    • These clueless millennials also rename everything they “invent”, which shows clearly that they bothered to read up on previous technologies They think they are so clever.

  3. This ice sheet sits on top of a volcanically active rift and thus behaves differently to the rest of antarctica. The assumption that it is warming due to purported warmer ocean temperatures reveals that these scientists are very likely to reach the wrong conclusions.
    They should be looking at this volcanic activity rather than chasing their obsession with the CO2 Meme.

    • “They should be looking at this volcanic activity rather than chasing their obsession with the CO2 Meme.” It would be interesting to get scientist Dustin Schroeder’s reaction to that comment, since he’s the chap who discovered the greater-than-expected geothermal heat under the Thwaites Glacier:

      • “since he’s the chap who discovered the greater-than-expected geothermal heat “
        And he quantified it. It’s tiny; about 100 milliwatts/sq m. Enough to melt a few mm of ice per year.

    • And it’s always worthwhile to pull up a full map of the entire continent of Antarctica, realize just what an incredibly small part of it Thwaite’s glacier represents, and then wonder why they chose this particular location as the poster-child of alleged glacial retreat.

      • A few minutes on Google Earth is very informative, with a look at it the area around Mt Erebus a real eye-opener.

    • And they were able to show, if you check the article, that previously stable areas (i.e. not affected by volcanics) were now retreating…

      • The language is not clear what they mean. It sounds like they acknowledge that melt water from under the ice sheet (geothermal) is melting the ice shelf.

        The researchers identified several features beneath the ice sheet that had previously only been observed in modern data, including ash layers from past volcanic eruptions captured inside the ice and channels where water from beneath the ice sheet is eroding the bottom of ice shelves.

        • Adding to my comment of a few minutes ago (probably appearing below, because my comments now seem to take half an hour to appear at WUWT), I think, being generous, that Griff has wilfully misinterpreted the article presented.

          They certainly did not claim that previously stable areas were retreating, other than the unsupported statement about general melting. This happens normally all the time as glaciers reach the ocean. These glaciers necessarily will have sub-glacial channels where melting occurs. These channels will change over time, just as a river does on the land surface.

          What the researchers do claim, is to have found a channel where melting doesn’t occur, so that they can then use it as a calibration standard for other measurements.

  4. It cracks me up when these people report that sea level will rise as ice shelves melt. It is just the opposite – ice displaces more water. And is anyone surprised that over 50 years the underwater currents change? Anyone ever heard of fish migrating to new areas as their temperatures change, and then migrating BACK when temperatures change back? There is no news here, except that they are getting better at mapping the ice extent. Unfortunately, in the rush to report on global warming, they taint their own work with fanaticism.

    • SOP for climate numpties is to pretend that all ice melting activity is the same, and due to the same “cause”, because they know that the science and logic-challenged Believers won’t know any better. When called out on this, though, their lame excuse is “oh, but these ice shelves are “holding back” the land-based ice, and once the shelves go, the ice sheet will race to the sea because now, there’s nothing to stop it. And they say that with a straight face.

      • Because it is true. The ice shelves hold back land-based ice.

        Please do tell me how they are not/could not?

        • The land based ice, upon losing the restraint of shelf ice, becomes the new shelf ice, unless its depth/thickness is low enough to float. This is how any glacier terminating in water works. The driver is the accumulating snow/ice in the highland.
          The nonsensical notion that the snapshot of current conditions represents some sort of unchanging steady state is echoed in other fallacies in the climate arena, and no doubt, in many others. For example, the contention that a little bit of (further) warming is suddenly going to result in permafrost melt and release of methane, ie a tipping point. Complete nonsense. The whole frontier of permafrost is continually melting/refreezing on various time intervals. That frontier is defined by a combination of latitude, proximity to maritime influence, aspect with relation to cardinal direction, elevation, etc. Obviously the frontier was somewhere across the middle of the continental USA in early Holocene times, receding northward in fits and starts through to the present day location. Tim Ball shares a photo of a white spruce stump dated to some 5000 years ago still approximately 100 km north of the current tree line in the Northwest Territory of Canada. Where was the tipping point due to methane release during the Holocene?

        • Griff, don’t you know how to think? How can ice that is floating hold back ice that is moving downhill?

    • some of the ice shelves are grounded… they are not floating -so if they melt, they DO add to sea level. and these also hold back ice on the mainland…

      Fish moving back? In UK waters we see only a northward migration of fish, with cold preferring species moving north and all sorts of exotic warm water species appearing for the first time

      • That is conclusive proof of catastrophic man-caused global warming for sure! Nothing anecdotal about that at all.

      • Griff, get thee to a historian. These fish aren’t appearing “for the first time”, they’re appearing for the first time since the Medieval Warm Period.

        That’s the problem with many warmers, they think the world is stable and therefore any change is “new”, “unusual” or “unprecedented”. Reality is quite different. The Isle of Wight only separated from England in 70 something AD. Harlech Castle, when under the famous siege was on the coast. Beaumaris Castle, built in the 1600s could sail a 40 ton to it’s south dock when new. The Cinque Ports were actually on the coast. The peninsula where William the Conqueror landed in 1066 is now several miles inland.

        The Earth is constantly in flux. Accept the reality.

      • That’s not really true, sand eel populations have moved northward and southward in cycles, these have caused problems for puffin populations with declines in Shetland recorded at various times-the early 80s (possibly earlier) I remember this being big news at that time it was thought to be pollution related, but that didn’t wash as there wasn’t significant pollution, later it was linked to ocean temperature increases, but also population increases further south at other times recently have occurred skokholm,skomer in Pembrokeshire for example. Ocean cycles seem to be the more likely culprit. Jim Steele’s book highlights other such cycles. The truth is all of these are poorly understood at the moment, hence pollution, breeding problems, then temperatures all being used as reasons. Ocean cycles would seem most likely.not co2..

      • Do me a favor, Griff. Do an experiment. Take a half full glass of water and put a couple of ice cubes in it, and mark the water level on the side of the glass. According to you when the ice melts the water level will rise. You will find out you are wrong. Floating ice NEVER causes water level to rise when it melts. You could get water level rise when you get more ice move from land to float in the water, but considering how much ocean there is compared to how much ice could move from land to ocean, not anything measurable.

    • John: 1) the ice shelf in question isn’t floating. It is grounded so all the ice volume is not fully displaced in the sea.
      2) It isn’t “just the opposite” for melting of floating ice. For floating ice, melting it will result in no sea level change. Ice is ~10 greater volume than an equal mass of water, but it floats 10% above the water, displacing the same volume as the original water that formed the ice.

    • John,
      A grounded ice sheet has significant above-water mass supported by the sea bed it is resting upon. Melting that grounded shelf until it floats can/will raise SL. The question is how fast/how much.

  5. Another EurekAlert!
    But somebody missed a Big one!
    Here it is: A EurekAlert! where the author frets that Scientific Conferences produce more carbon dioxide than strictly necessary.
    Feast your eyes.

    Several times a year, researchers from all over the world travel long distances in order to share their latest findings and establish contacts at conferences.

    Well, yes they do.

    According to Jäckle’s calculations, the average CO2 balance of a conference visitor ranges from 0.5 to 1.5 tons of CO2 equivalents per three-day ECPR meeting.


    “If researchers would then accept somewhat longer travel times by bus or train compared to air travel, up to 85 percent of a conference’s emissions could be saved,” says Jäckle.

    Eureka! We are saved!

    • “If researchers would then accept somewhat longer travel times by bus or train…”

      Why the half-measures, why not use video conferencing and save even more on ’emissions’? And we could power the video conferencing with Solar Panels and Wind Turbines and be even more eco-friendly!

  6. Curious. Why do all the measurements/observations stop 10 years ago, 2009? At least in this paper. An inquiring mind would like to know. Anyone?

  7. “It’s Worse Than We Thought” alert No. 5,235. For sure this means we need more funding to study how much worse it is, and send it quickly because there are no years left on the Doomsday Clock. I think I will go play golf, that puts me in good disaster-dealing company. Just saying.

  8. From the article: “information that contrasts their findings about the Thwaites Glacier ice shelf, which has thinned from 10 to 33 percent between 1978 and 2009.”

    Well, 1978 was the coolest period since the early 1900’s, so it would be perfectly normal for the ice to be melting in the warming world that came after 1978.

    The ice also melted in a similar manner during the very similar warming that took place during the 1930’s, and during the hot El Nino year of 1998.

    After the 1930’s, the globe started cooling off and cooled into the 1970’s to the point that climate scientists were sounding the alarm about human-caused Global Cooling.

    Currently, the globe has been cooling for over three years, since the hot El Nino year of 2016, just like similar cooling occurred after the 1930’s and after the hot El Nino year of 1998. We are currently about 0.5C cooler than 2016, the “hottest year evah!”, according to NOAA and NASA.

    Nothing to see here. As Trump says: “The temperatures go up and then they do down.” You can infer from that, that Trump doesn’t think the temperatures go, Up, Up, Up, as NOAA and NASA claim. And currently temperatures are NOT going Up, Up, Up.

  9. If one cheats to get into some of the “big name” universities, like Stanford, Harvard, Yale, etc then why would they cheat on an issue they are irrationally passionate about? Move a decimal place, exclude contradictory data, not report contrary results, etc all could easily be defended in the minds of the “true believers.” They believe, as all evil people have in the past, that whatever advances their cause no matter how fallacious is OK, not in their minds required. Now if I must follow the AGW agenda to keep my incoming grants, my pay and benefits, what difference does it make if I misreport my research to please the high priests and priestesses of the Anthropogenic Global Warming religion.

    • I think you have little understanding of how little scientists get paid and why they pursue careers in science and not accountancy.

      • Simple enough Mr Griff…
        Why do Scientists pursue careers in Climate Science instead of Accountancy?
        Because Climate Science has no Accountancy.
        Because in Climate Science 1+1 doesn’t have to = 2.

      • Apparently they get paid vast sums directly from their grant providers and university funders much like Dr Willie Soon (NOT)being paid massive amounts from fossil fuel interests /sarc

      • My guess is because many want the easy gravy train where they don’t have to be accountable for what they produce. In the business world, if you don’t perform you get canned (unless you’re good at the heiny-lick maneuver). In academia, you can produce all kinds of crap and thrive.

    • You’re right of course. Since many of their parents paid the equivalent of historic church Indulgences (mentioned on the news this AM) to get ’em into Stanford, Harvard, Yale, etc on ‘ athletics scholarships’ why would they worry about a little ‘end justifies the means’?

  10. As of the 15th Aug (the latest NSIDCC plot shown here) the antarctic sea ice extent is (marginally) higher than on the corresponding date last year. Does that seem alarming?

  11. Kudos for digitizing the photographically recorded radar data from the 1970’s. More data and a longer temporal baseline is always valuable. Interpretation without preconceived ideation would be refreshing!

    • J Mac
      If properly cared for, the film has the potential to be usable for a longer period of time than the digitized data. Where would you go to get data stripped off an 8″ floppy disk? How easy would it be to translate spreadsheet data on a 3 1/2″ Amiga floppy? Digital technology is advancing so rapidly that the future problem of reading imperceptible bits encoded differently on different media will be more challenging than looking at a picture.

        • I was lucky none of the computer I repaired had no card readers, most were mini and to small for that, but the first one I was trained on had paper tape MTRs. Yep it was a beauty, sixteen bit with 128 K of memory. Yet with the right disk drive it could support up to forty users on terminals and if you had to wait more than ten seconds between screens something was wrong! It also had an eight inch floppy drive, 1 meg at that.

  12. To put it more factually, a very small part of Antarctica was examined back as far as the 1970s to draw conclusions about climate that fluctuates over decades regardless of direction of warmth to focus in on only a singular possible cause of thawing and project grave destruction for the whole world.

  13. Climate science procedure:

    “Correct” answer: old data is surprisingly good.
    “Incorrect” answer: old data needs “adjustments.”

  14. Is this the same film that has Mickey Mouse at the helm of his tugboat steering to avoid the icebergs?

  15. I cannot see a link to the actual paper, but I see no support of the claim “Thwaites Glacier ice shelf melting faster than previously observed.”

    The only way adding data from 1971 to 1979 can add to our understanding is to measure changes between 1971 and whenever modern data starts. That say “10 to 20 years provided by modern data,” so call the start of modern data 2000. They can now measure changes between 1971 and 2000, if that shows faster melting than the modern data, then that indicates melting is slowing.

    They have established that some parts of the ice are stable and others are shrinking. A global effect would melt every area at a similar rate, so localised melting implies a localised cause, i.e. not CO2.

    Then why do that say “the Thwaites Glacier ice shelf, which has thinned from 10 to 33 percent between 1978 and 2009”? They have data going back to 1971 and presumably data since 2009, could it be that these data sets do not support their dogma.

  16. The paper is open-access available here:

    For Figure 7, the authors write in the Figure Legend:
    Expanded coverage of the Ross Ice Shelf and the Siple Coast. (A) Previously unscanned profiles (light red) dramatically expand the available data for the Siple Coast including (B) a 1974 60-MHz profile of the grounding line of Mercer ice stream showing grounded and floating basal topography as well as surface and crevassing, enabling assessment of ocean access (10, 41). Dashed vertical lines indicate where this survey crosses the modern grounding line as measured by satellite observations (4) suggesting that the location of the grounding line in this area has not significantly changed. Most of the radar-sounding profiles of ice shelves, like the Ross Ice Shelf (C), have been scanned and provide a baseline ice thickness and basal geometry from 1974 against which more recent surveys (24) can be compared to observe ice-shelf processes and change.”

    (My bold for emphasis.)
    Ooopppps… how inconvenient that the older 1970’s data now shows the massive Ross Ice shelf was thinning faster in first half of the ~40 year period than of late.

    Another inconvenient point they found that they had to rationalize away because of “expectation bias” is here where they discuss and compare the Thwaites Eastern Ice Sheet (TEIS) behavior of the past to the more recent measurements that used altimetry (and not the 60 MHz ice penetrating radar scans as was done in the 1970’s that they digitized in this project).

    “Through this comparison of archival and modern data, we estimate a thinning rate of ∼40 m per decade, higher than the thinning rate of 25 m per decade based on ice-shelf surface altimetry acquired from 1994 to 2012 (3, 32, 33). One possible explanation for the differences in these estimates is that submarine melting slowed from the 1978–1994 period to the 1994–2009 period. “


    That obvious conclusion they then had to hand-wave away with this statement, “However, given the cooler ocean temperatures in the earlier period (34), it is more likely that the ice-shelf-scale altimetry, which estimates a single value for both TEIS and Thwaites Ice Tongue, does not capture the finer-scale local thinning apparent in radar-sounding data. “

    So they had to assume a “given” on ocean temperature under the Thwaites ice shelf in the 1970’s (that it was colder) and then malign and “diss” the modern altimetry studies of TEIS as “does not capture the finer-scale local thinning apparent in [older] radar-sounding data.”

    This what religions do. that is obvious conclusions have to be rationalized away to the arrive at desired conclusions to remain within the dogma and belief system.

    • wow!!! they published that!!

      I guess the skeptics theory that group think and gatekeepers rule the world of peer review is flat busted

      • But the theory of maximum media spin and science through headlines lives on! One might say that it remains err… “robust”.

      • No Steven.
        They genuflected to the Consensus, a sort of it’s worse than we thought. They had to bend their conclusions from the simpler interpretation of their data to get there though. It (TEIS recession) began even earlier and has since slowed, we just didn’t know how bad it was (then) because be only used altimetry recently not ice penetrating radar.

  17. There are 52 cards in a deck, just because 1 falls in favor doesn’t mean that the remainder of the deck isn’t stacked

  18. The title of this article should include be Vintage 1970’s Film …

    “Vintage” used without a date usually implies something a bit older than forty-some years.

    I was thinking 1920″s or silent-film era, maybe some old piano music playing, to accompany fast, quirky moving figures tripped about the ice.

Comments are closed.