I arrived at Summit Station on July 13th, while traveling with the
Joint Science and Education Program (JSEP) for a short visit to the
camp. When we arrived, Summit Station had been experiencing above
freezing temperatures for multiple days prior to our arrival and a
melt layer formed on the near surface snow. I have been studying the
physical properties of the top layers of the ice, the firn, at Summit
and NEEM for my Ph.D. research. Recently, I have been focused on the
melt layers present in both firn cores because they occur very
infrequently. At Summit, there is only one other melt layer besides
the melt layer from this past week and this previous melt layer dates
to 1889.
The most interesting part of being at Summit Station just after a melt
event had occurred, is that the melt layer formation process could be
observed. When studying a firn core, there is only a small cross
section of the firn column that can be examined, which makes it hard
to understand how the melt layer formed and how evenly distributed it
is. Studying snow pits at Summit, including the recent melt layer,
presents a unique opportunity for us to understand how previous melt
events occurred. While at Summit density, stratigraphy, and
permeability measurements have been taken and samples will also be
brought back to the laboratory at Dartmouth, which will give us a clue
about melt layers in the past.
New Summit Melt Layer
July 21, 2012 by keegankm
WOW, this is an important and disturbing observation. I imagine the C-130 pilots are not happy with this event and the possible impact on the runway. Get some good data- go IGERT.
WOW! That is super neat! I’d love to hear more about this melt layer … and see how it compares with the melt experiment we did two years ago. Crazy-cool. Go IGERT indeed!
[…] such wide-spread melting is an event that only happens once every 150 years or so, e.g, PhD student Lora Koenig is quoted as “Ice cores from Summit show that melting events of this type occur about once […]
How far back in time do these ice cores you are looking at go? That is, does the bottom of the ice core correspond to 1889 or does go back another 100 or 1000 or 10000 years?
I study the firn column at Summit and NEEM, which is approximately 80m deep at both sites. These depths correspond to an age of about 1750. So, in the firn core we can see back about 250 years.
Thanks, I’ll correct my blog to reflect this. So, are there similar or even larger warming/melting events between 1750-1889 from what you see in the cores? I somehow feel much more comfortable with this type of data than remotely sensed products.
NASA has just today posted a press release about the severe melting on the Greenland Ice Sheet and refers to Kaitlin’s work to establish 1889 as the last date such pronounced melting occurred. http://www.nasa.gov/home/hqnews/2012/jul/HQ_12-249_Greenland_Ice_Sheet_Melt.html
Reblogged this on SERMERSUAQ.
I’m disturbed by the ” every 150 years” statement. Are there other cores going back further than 1750 that validate this? If there has only been one previous example how does one discern a trend?
Terry
Yes, in Greenland there have been many deep ice core drilling projects which drilled ice to the bedrock. At Summit, the deep ice core (GISP2) reaches a depth of about 3km with the oldest ice being about 123,000 years old. Alley and Anandakrishnan’s paper looks at the ice just below the firn, and into the last 10,000 years. In the past 10,000 years (the Holocene), there is on average a melt layer every 150 years.
Thanks for this clarification, or added context fascinating valuable research
Thanks for the clarification.
However, looking at the Alley and Anandarkrishnan paper, that “on average” covers a *very* large standard deviation, doesn’t it? There is nothing especially magical about a 150 year cycle (as some blogs and even the NYT seem to be crowing). Indeed, the opening line of that paper is “The rare melt features in the GISP2, central Greenland deep ice core have decreased in frequency over the most recent 7000 years.” And looking at the actual melt dates, there is one in 1889, but then none in the 700 or so years before that. To point out that the average of the whole period is 153 years is accurate, but I would suggest quite easily misunderstood.
Am I misrepresenting the data?
How does this year’s melt layer (so far, as there appears to be more above freezing weather coming on Saturday) compare with the melt layers of other years in which melt occurred? All of the data reporting I’ve seen about this talks about years in which either there “is” or “is not” a melt layer, but surely there analog numbers which could tell us something about the actual amount of melting which occurred.
Several other recent years have also seen temperatures at Summit going above freezing. Was there any melting at during these incidents, or did wetbulb temperatures remain below freezing? If there was a small amount of melting, is it possible to detect years like this in the snow record, even though these were years in which there “was not” a melt layer?
Thanks so much for the reply.
The 150 year figure, while accurate, gives the false impression that this is nothing too far from the ordinary. Event frequency from the last 1,000, 2,000 – even 5,000 years would paint a clearer picture.
Terry
[…] sheet that will, over time, become the next dense layers in the great frozen mass. Here’s her blog post on witnessing the first surface melt event since 1889 (with a photo of the resulting ice layer).Here’s my 2004 video report, the accompanying […]
[…] will, over time, turn a subsequent unenlightened layers in a good solidified mass. Here’s her blog post on witnessing a initial aspect warp eventuality given 1889 (with a print of a ensuing ice […]
Since the 150 year span has been previously observed has anyone ever hypothosized what the cause could be? They must have been predicting this to happen any time now so it shouldn’t be such a big surprise.
[…] sheet that will, over time, become the next dense layers in the great frozen mass. Here’s her blog post on witnessing the first surface melt event since 1889 (with a photo of the resulting ice […]
Was the 1889 melt event widespread? Or was it at only a few spots? Or is the data too sparse to tell?
Yes, this is my question too. Are these previous melt events local or is there evidence for analogous earlier events that cover ~97% of the ice sheet?
Thanks for your work!
Yes, the 1889 melt event is believed to be widespread because it shows up in many shallow ice cores across Greenland (Clausen et al. 1988, Glaciological investigations in the Crete area, Central Greenland: a search for a new deep-drilling site, Annals of Glaciology, Vol 10, 10-15). With satellites, we are now able to determine the extent of new melt events much more easily. In order to study melt events that occurred before the dawn of satellite technology we must drill ice cores.
Thanks for clarifying. That is a very helpful piece of information that has not been widely circulated in news stories or blogs discussing this event, leaving many people wondering.
[…] ice sheet that will, over time, become the next dense layers in the great frozen mass. Here’s her blog post on witnessing the first surface melt event since 1889 (with a photo of the resulting ice […]
The Alley and Anandakrishnan paper (Variations in melt-layer frequency in the GISP2 ice core: Implications for Holocene summer temperatures central Greenland, ANNALS OF GLACIOLOGY, VOL 21, 1995 pg:64-70) says that the average melt frequency over the last 10kyrs is 1 every 153yrs. However, it points out that these events have become less frequent with time, decreasing from an average of 1 every 82yrs (between 5500-8500yrs BP) to 1 every 250yrs (between 1000-4000yrs BP). They attribute it to orbitally induced decreasing local summer insolation values, although they can’t rule out reduced summer temperature variability.
I’m curious if this has been observed in any other Greenland ice cores with sufficient length? And how safely can you infer widespread surface melting from a summit site?
[…] m d’altitude, portait de l’eau liquide, ce qui n’était pas arrivé depuis 1889 d’après Kaitlin Keegan. Glaciologue du Darmouth College et membre du programme Darmouth Igert, elle se trouve actuellement […]
[…] was available to describe the exact findings, but in a blog posting detailing her work, Keegan noted that several cores dating back millennia have also reflected the 150-year […]
[…] was available to describe the exact findings, but in a blog posting detailing her work, Keegan noted that several cores dating back millennia have also reflected the 150-year […]
[…] was available to describe the exact findings, but in a blog posting detailing her work, Keegan noted that several cores dating back millennia have also reflected the 150-year […]
[…] was available to describe the exact findings, but in a blog posting detailing her work, Keegan noted that several cores dating back millennia have also reflected the 150-year […]
[…] was available to describe the exact findings, but in a blog posting detailing her work, Keegan noted that several cores dating back millennia have also reflected the 150-year […]
[…] was available to describe the exact findings, but in a blog posting detailing her work, Keegan noted that several cores dating back millennia have also reflected the 150-year […]
[…] New Summit melt layer […]
Remember that each time there is a surface melt, it removes material that would otherwise have ended up in ice core (or, in the case of local original depressions, it can add material to later core). In sedimentary geology, the topics of “unconformity” and related terms are part of the fundamental science. In ice cores, there is less mention, especially of “disconformity” because it would be difficult to recognise. This has several consequences. It makes it more difficult to equate layers in cores to years, especially when there as a low intrinsic accumulation rate, like Vostok & South Pole. Therefore, the time axis in cores can become wrong. Second, if there is a disconformity, there is no easy way to deduce the number of years of accumulation that were wiped off the surface, again leading to potential or actual errors in core time lines. There does not seem to be a reported disconformity in the Antarctic ice mass, so we can say, if this is correct, that the Antarctic has not warmed to melting in the last 800,000 years or so. However, Greenland has many interruptions to parameters such as oxygen isotopes that could be ascribed to disconformities. There has been a tendency, instead, to relate the breaks to changes in direction of incoming winds in the snow season. There are many papers on the latter topic.
The appearance of widespread Greenland melt reported by NASA 28 June is not likely due to hot air, because the heat capacity/transfer physics is dubious. Keep in mind some large, recent coronal ejections and the UV spectrum of incoming light; and the surface albedo change in Greenland that might have been cause by volcanos and forest fires. I have not seen the measurements/calculations mentioned in this last paragraph, so it is speculative.
[…] planes to land on the once-frozen runways. The last time Summit melted like this was in 1889, ice cores show. On average, these island-wide melts happen about once every 150 […]
I’m curious. What was the source of the warm air that led to the melt all over
Greenland. Was it a body of air from the Atlantic ocean or from the Arctic perhaps. To rise that high and still remain warm, it must have contained a lot of water vapor which condensed and release latent heat. Could we be seeing the harbinger of a Walker cell between rising air over the ocean and katabatic winds over the Ice sheet.
[…] will, over time, turn a subsequent unenlightened layers in a good solidified mass. Here’s her blog post on witnessing a initial aspect warp eventuality given 1889 (with a print of a ensuing ice […]
I actually found this posting , “New Summit Melt Layer «
Dartmouth IGERT – Polar Environmental Changeâ€, exceptionally enjoyable and it was in fact a superb read.
Thanks for your time-Pam
[…] Last summer, several IGERT fellows had the serendipitous and rare opportunity to witness a warming climate’s effect on Greenland first-hand. Julia Bradley-Cook was stationed in Kangerlussuaq collecting data on carbon cycling in soil when the bridge over the Watson river collapsed from anomalously high flows of meltwater (see https://dartmouthigert.wordpress.com/2012/07/11/glacial-melt-threatens-town-water-supply and https://dartmouthigert.wordpress.com/2012/07/11/update-the-river-powers-on). Days later, the 3rd cohort of Dartmouth IGERT students flew up to Summit Camp, Greenland’s highest point, and observed features of the ice sheet-wide surface melt. Fellow Kaitlin Keegan, Thayer Professor Mary Albert, and their collaborators study the frequency of such melt events; their work at the North Greenland Eemian Ice Drilling (NEEM) sight has suggested that such an event last transpired in 1889 and, therefore, is unprecedented in the satellite record. (See https://dartmouthigert.wordpress.com/2012/07/21/new-summit-melt-layer). […]