Glacial burial and decomposition of ancient organic carbon: a scientific expedition to King George Island, Antarctica

View some trip photos     A neat article about the trip from "The Polar Times"    Random Science from the trip     Lindsay's blog from Antartica     Wilderness Research Foundation


Ning Zeng, Project Scientist, Associate Professor, University of Maryland, College Park                  

Jay Gregg, Junior Scientist, University of Maryland, College Park

Email contact:




An expedition to King George Island (KGI), Antarctica was conducted during January, 2010. The main goal was to search for ancient organic carbon buried under ice and to understand the role of such organic carbon in glacial-interglacial CO2 and climate changes. Three trips were taken to study the periglacial environment of the Collins Glacier (Bellingshausen Dome) on the southern edge of the KGI icecap. A glacial moraine was found to contain a large quantity of organic carbon. An outcrop was found to contain several clearly distinguishable layers: rubble, soil, moss, soil, shell, moss, muddy soil, and ice. The surrounding area and the glacial outwash downstream contain large amounts of organic material. CO2 fluxes were measured at two locations using a LICOR-8100 soil CO2 analyzer, with soil CO2 fluxes ranging 15-20ppmv/30min (0.15 mmol m-2 sec-1). Because there was no observable new vegetation growth on the site, and because the chamber where the flux was measured was dark (preventing photosynthesis), it appears that the CO2 was the result of the decomposition of the organic carbon that was once buried under ice. Our findings support the hypothesis that organic carbon, including both vegetation and soil carbon, can be buried under ice, and later released back into the atmosphere, thus contributing to climate change through the emission of CO2, a greenhouse gas. The age of the ancient carbon and the processes and circumstances under which they were buried are yet to be determined.


The expedition was part of a summer school organized in conjunction with the International Polar Year (IPY) 2007-2010.  The IPY is a international effort to foster collaborative research in the polar regions of the Earth.  The first IPY was from 1882-1883, the second from 1932-1933.  This research comes at the tail end of the third IPY, the goal of which was to increase our knowledge of the poles, how they are changing, and to better understand the influence of the poles on the global climate system and vice-versa.  The third IPY marks the largest international collaborative science effort since the International Geophysical Year 50 years prior.


King George Island (KGI)


Research for this project was conducted on King George Island, situated at the tip of the Antarctic Peninsula.  King George Island is the largest of the South Shetland Islands, close to the Antarctic circle at approximately 62 S.  At the north shore is the Drake Passage and to the South is Bransfield Strait. The island is approximately 95 km in length and 25 km across, and has an area of over 1100 square kilometers, of which over 90% is permanently glaciated.

King George Island is a dynamic place.  As the climate has warmed, glaciers are receding from Fildes peninsula on the south-west part of the island.  The newly exposed soil supports lichens, mosses, grasses and other vegetation, as well as Antarctic Terns and Skua.  Coastal areas support Chinstrap and Gentoo penguins, Elephant, Weddell and Leopard seals, Snow Petrels and Kelp Gulls.

Because of the wildlife and the dynamic geology, many countries have established research stations (most of them operating year-round) on King George Island, including Argentina, Brazil, Chile, China, Ecuador, South Korea, Peru, Poland, Russia, and Uruguay.  Research for this project was conducted from the Russian Bellingshausen station, though the field sites were near the edge of the Collins glacier near the glacial moraine.

Trip to/from and around KGI


Getting to Antarctica is not easy.  Some tourists manage the journey on expensive cruise ships, visiting much of the continent, but only coming ashore for short (hourly) stays. Other tourists charter flights, however, again the time spent in Antarctica is short.  Because Antarctica is reserved for scientific research, scientists may arrange longer stays with the invitation of a research station.  The Russian Bellingshausen station hosted more than a dozen researchers and scientists as part of the 2010 King George Island Summer Institute. We booked passage on a Uruguayan Air Force Hercules transport plane. Because of the short runway and treacherous conditions, the flights only occur when the weather on King George Island is fair, with no low clouds. This is by nature hard to predict, as King George Islands maritime climate can produce quickly changing and unpredictable weather conditions.  For instance, our inbound flight experienced a number of delays, and our return flight was canceled altogether.  Flexibility is a necessity when conducting field research in such an environment.


We spent the first three days on the island conducting reconnaissance hikes with experienced scientists and orienting ourselves with the environment.  We learned the safety protocols when doing field research, the hazards of the island (snow swamps,  slippery permafrost under a layer of mud, and its unpredictable and quickly changing weather. We also learned about the various wildlife on the island, safe distances to maintain, and sensitivity to the fragile flora in this environment.  On these hikes, we discovered areas of particular interest in recently exposed glacial moraines at the edge of the Collins glacier.  It is these areas that served as our field sites for this project.



Environment of Fildes Peninsula, the ice-free corner of King George Island


Climate and Ice


For three days, scouting hikes were conducted along the edge of Bellingshausen Dome (Collins Glacier), guided by Russian glaciologist Bulat Mavlyudov. The edge is marked with glacial moraines and nunataks (rock islands in ice). Over the last several decades, the edge of the ice cap has been retreating rapidly. A moraine about 50 meters away from the edge of the ice cap was under ice 20 years ago. This is consistent with the general warming in the region. The Antarctic Peninsula is one of the fastest warming regions in the world. The temperature measured at the Russian Bellingshausen Station shows an annual increase of 1.3 C increase in winter temperature over the last 50 years, while the winter temperature has increased by about 2.4 C. Further back in time, the entirety of  King George Island, including the Fildes Peninsula, was covered by ice during the last glacial maximum (LGM) 21,000 years ago when New York was under the Laurentide Ice Sheet.


Collins glacier and Bellingshausen dome in background. Moraine is in front of the glacier and a Nunatuk appears on the left side of dome.


Bellingshausen temperature record shows that Antarctic Peninsula is one of the fastest warming places in the world.




As the ice retreats, vegetation develops quickly on newly exposed land. Lichen, moss and grass grow on KGI. A transect perpendicular to the ice edge was made. Visual inspection showed no visible sign of vegetation in the few meters nearest to the ice edge. This is supported by a team from the University of Wisconsin, led by Professor Les Werner,

which measured the photosynthesis and respiration along the edge of the Collins glacier.


Moss vegetation near the coast of the King George Island


Sampling of organic carbon and CO2 measurement


At one location north of Uruguays Artigas Station (6211'04 S 58 54 W), a glacial moraine was found to contain large amounts of organic carbon. An outcrop that was cut out contained several clearly distinguishable layers: rubble, soil, moss, soil, shell, moss, muddy soil, and ice. Samples were taken from the moss layers. This area is called Site 1. In front of the outcrop is a muddy field of glacial outwash originating from the layered material. A stream winds around the moraine and flows down towards the sea, which was approximately 50m away. Slightly before the stream enters the sea, a mud field appeared to contain the same organic material that was washed out from the moraine and deposited there. Clumps of moss were found in the river-cut deposit and samples were collected at 1 meter and 2 meter depths (Site 2).

Jay Gregg set up the LICOR LI-8100 Automated Soil CO2 Flux System near Site 1.


A pile of moss layer (brown) exposed. Organic odor was clear.

Layered moraine outcrop: moss layers are brown, one layer above the spoon, one layer below. Shells (small white pieces) are above the lower moss layer. Exposed ice (permafrost) is white-blue at the lowest level, partly covered by loose fallen soil.


The moraine outcrop from distance; note the spoon as in the above picture.



Glacial outwash deposit, 50m downstream from the outcrop. Organic rich material including dead moss was present inside the deposit.


Results: CO2 measurement


CO2 flux measurements were conducted using the LI-8100 Automated Soil CO2 Flux System at Site 1. Two measurements, both immediately in front of the layered outcrop were conducted: one test spot had visible (dead) moss lumps on the surface, while the other had no visible moss. Tests showed discernable CO2 flux after the standard 3 minute collection time. The collection time was then increased to 30 min afterwards.


Data from the LI-8100 Automated Soil CO2 Flux System were analyzed with LICOR File Viewer software, and further analyzed in Microsoft Excel. To calibrate the instrument to the background level of atmospheric CO2 concentration (388.6 ppmv), data from the CO2 monitoring station on the King Seong Station (South Korea) were used to offset the data recorded by the LI-8100 Automated Soil CO2 Flux System.  This makes no difference on the flux calculations, but is done only to accurately represent the CO2 concentration in the chamber at any given point in the observation trial. The first three minutes of the 30-minute readings were not used in the analysis to allow for the chamber concentration to stabilize once the chamber was sealed.

Site 1, Spot 1, CO2 concentration in the chamber increased by 0.47 ppmv min-1, corresponding to a CO2 flux of 0.09 mmol m-2 sec-1.

Site 1, Spot 2, CO2 concentration in the chamber increased by 0.68 ppmv min-1, corresponding to a soil respiration CO2 flux of 0.15 mmol m-2 sec-1.



Large amounts of organic carbon in a glacial moraine outcrop and the downstream outwash were found at the front edge of Collins Glacier, King George Island. The organic carbon was deposited in the past as layers of moss, interlaced with shells and soil. This suggests that the area was once at sea level when shells were deposited. There were multiple periods of moss development and sedimentation.


The measured CO2 flux was significant, indicating the comparatively rapid decomposition once the old carbon is exposed. There was no evidence of new vegetation growth around the outcrop where measurement was done. This indicates a relatively fresh exposure, possibly within last few seasons.  These results are intriguing, suggesting that soil organisms (decomposers) are active within newly exposed soil from the glacial moraine. Whether they lie dormant under the ice for millennia, or they are newly transported to the area from somewhere else will be determined by carbon dating the soil.  Further work will be undertaken to date the sample to understand the age and developmental history of the organic carbon.




We are most grateful to the financial support from the Wilderness Research Foundation, New York, and  its president Sheldon Bart, without whose persistent effort, this work would not have been possible. We also gratefully acknowledge LICOR for supplying us the LICOR-8100 soil CO2 analyzer.




Hall, B. L., 2007: Late-Holocene advance of the Collins Ice Cap, King George Island, South Shetland Islands. The Holocene, 17, 1253, DOI: 10.1177/0959683607085132.

Zeng, N., 2003: Glacial-Interglacial Atmospheric CO2 Change--The Glacial Burial Hypothesis. Adv. Atmos. Sci., 20, 677-693.

Zeng, N., 2007: Quasi-100ky glacial-interglacial cycles triggered by subglacial burial carbon release. Climate of the Past, 135-153.

Random Science

Leaning trees

At Punta Arenas, Patagonia, I noticed that many plants, including conifer trees and small shrubs appear to be titled, or more precisely the leaves and branchs tend to be better developed on one side. This is in response to the prevailing winds. Over the growth of a tree, the strong winds in Patagoina forced the tree to grow asymmetrically. To put it another way, the flexible tree adapts itself to the strong wind.

This leads to an interesting possibility: the direction of the trees can be used as an indicator for local prevailing winds. In Patagonia, the strong winter winds mostly comes from the south, i.e., Antarctica. If the trees all tend to 'lean' in one direction, that will prove the above reasoning. A related conjecture is that the direction of tree leaning is not necessarily related to, e.g., annual average wind, but probably more to the strong wind which tends to be in the winter. This is not surprising as weak wind won't really influence the plants, but only really severe condition has impact. This is made clear later after I arrived at KGI, where we stayed in the guesthouse of the Russian Bellingshausen Station. The house is perched up a little hill where the wind is super strong, and we have to walk with our body bent against wind, sometimes at almost a 45 degree angle so that we don't get blown down. Of course, the direction of bending is the opposite of the trees leaning.

I plan to do some measurement with a compass on the way back from  Antarctica so as to define the direction of the prevailing strong wind . This could provide an independent direction measure if we are lost in the dark, or we are confused about the direction of the Sun in the southern Hemisphere :-).

Blue vs white iceberg

The day after our arrival, Hans, a German biologist who studies penguins, seals and birds took us for a hike along the west coast of KGI. We saw icebergs floating off the shore. Most icebergs were white, but one or two were blue. Hans said that the blue icebergs are old ice, could be as old as thousands of years. I put on my physicist's hat and offer the following explanation.

White icebergs are relatively fresh snow, and the snow flakes/particles scatter sun light and appears white. Cloud, a collection of many many water droplets appears white for the same reason. Instead, old iceberg is compacted solid ice chunk. Sunlight penetrates into the ice and is reflected back. Light rays of different wavelengths have different refractivity, and blue light is more preferentially reflected back so that old iceberg appears blue. This is the same reason why ocean is blue.