Annual Records in Ice Cores Go Back Tens of Thousands of Years!

Although seafloor sediments provide evidence further back in time, ice cores have the advantage of recording annual changes in both the precipitation amounts and the chemistry of the air. In fact, ice cores provide 8 different kinds of evidence of past conditions.

Ice cores provide 8 important types of evidence

Dr. Mary Davis is a senior research associate with the Ice Core Paleoclimatology Research Group (ICP) at BPRC. She has traveled with Dr. Lonnie Thompson as a member of ice coring expeditions to mountain (or alpine) glaciers around the world. In 2001, they drilled a core from a glacier in a col, or saddle, between Mt. Bona and Mt. Churchill in the Wrangell-St. Elias range in southeastern Alaska.

Ice core drilling in Southeastern Alaska

Ice core drilling team at work

Coring sites are selected on glaciers that have existed for centuries. Sites are selected where the ice flow is simplest. This is either on the tops of dome-shaped glaciers (where the ice moves out in all directions--a radial flow) or in a col (the lowest point between two adjoining peaks). Both represent a location where the ice would accumulate, rather than a slope where the ice flow would be fast and chaotic, thus jumbling the climate record.

Arctic Ice Cores

Although there are many peer-reviewed articles based on analyses of ice cores from the Arctic Region, only cores obtained since the mid-1990s can help to assess the Arctic's recent rate of response to climate change.

It's important to remember that the Arctic is a huge area of Earth, and that coring sites are chosen to answer specific questions. While they provide an annual record of the atmospheric conditions and give some idea of the relative amount of snowfall at that location, it is impossible to derive a full climate history of the entire region from such a small set of sample sites. This would be like choosing climate histories for 5 or 6 cities in the U.S. and then using them to describe what has happened for the "lower 48 states" based on those records.

Comparable Arctic Ice Cores

Greenland (ice sheet) Latitude/Longitude
GRIP 72°35’N, 37°38’W
GISP-2 72°36’N, 38°30’W
NEEM 77°27’N, 51°4’W
NGRIP 75°6’N, 42°18’W
Russian Arctic  
Windy Dome (alpine) 80°47’N, 63°32’E
Alaska & NW Canada  
Bona Churchill, AK 61°24’N, 141°42’W
Mt. Logan 60°35’N, 140°35’W

The BPRC Ice Core Paleoclimatology research group has helped to retrieve ice cores from these Arctic locations: Alexander Graham Bell Island (Franz Josef Land in Russia) in 1994; 30 sites in Greenland in 1997 and 1998 (PARCA); and from the Wrangell-St. Elias Range of SE Alaska in 2002.

Cores from Ice Sheets

Cores obtained from the great ice sheets (Greenland and Antarctica) are used to study changes that have happened over tens to hundreds of thousands of years. Scientists drill cores through these thick bodies of ice to learn more about climate changes that occurred across the ice ages (through both glacial and interglacial periods).

Cores from Greenland were obtained when Dr. Ellen Mosley-Thompson and her team from BPRC collaborated in the Program for Arctic Regional Climate Assessment (PARCA), during which the OSU team obtained 14 cores in 1997 and 16 cores in 1998. The PARCA project was set up in 1993 and data were collected between 1995-1999.



Ellen is also one of the collaborators on an international initiative funded by the National Science Foundation to study the abrupt environmental change in the Larsen Ice Shelf System on the Antarctic Peninsula. This project, whose acronym is LARISSA (LARsen Ice Shelf System, Antarctica) began in 2009. Ellen and Victor Zagarodnov from BPRC led the team that drilled cores from the Bruce Plateau, located at 66.037 °S; 64.003 °W, on the Antarctic Peninsula. They worked at the drilling site from January 4 through January 31, 2010. While they suffered some technological setbacks during that time, they obtained 445.65 meters of core, from the surface of the ice all the way to bedrock.

Getting to the Bruce Plateau, first from Chile and then from Rothera

The drill team on the last day of drilling the Bruce Plateau

Cores from Mountain Glaciers

For alpine cores, Dr. Thompson's team often selects sites that are in a col (pronounced like "kall"), which was described above as "the lowest point between two adjoining peaks". This represents a location where snow, and therefore ice, would accumulate, rather than slide downslope past the proposed drilling site. This helps to assure that an ice core obtained there would have a continuous record and that the ice at the bottom would include precipitation that fell at that location (and would not have moved into that spot with glacial flow).

Mt. Logan site (Note:  This site was not drilled by BPRC scientists.)

Mt. Logan, Canada’s highest mountain at 5959 meters above sea level, is located in the St. Elias Mountains near Alaska. It was successfully cored in the 2001-2002 field seasons through international collaboration between several institutions from Canada, the USA, Japan, and Denmark. This effort updated information gained from a 102.5 meter ice core obtained from Mt. Logan that was recovered by a research team from Canada's National Hydrology Research Institute, led by Dr. Gerald Holdsworth. The 2000-2002 effort was intended to extend the climate record back in time over the entire Holocene interglacial epoch and possibly into the last ice age (>12,000 years ago). The Mt. Logan cores contribute information about recent changes (up to the date they were obtained, of course). They are comparable to other Arctic cores at BPRC.

The Bona-Churchill Ice Core

Field Logistics

A deep drilling program was conducted on the col between Mt. Bona and Mt. Churchill in southeastern Alaska from April 30 to June 10, 2002. This was a time when weather conditions were likely to be the best for aircraft operations and when temperatures were still sufficiently cold at the col to drill the ice cores and store them on site. The drilling team was supported by an AS350 helicopter (photo) that was piloted by Lambert DeGavere, under contract from ERA Aviation, and an Otter piloted by Paul Klaus of Ultimate Thule Outfitters. VECO, under contract to NSF, provided the logistical support including air support to and from the location.

Drills and Cores

A lightweight, portable drilling system designed for coring to a depth of up to 700 meters was developed and tested at BPRC by Dr. Victor Zagorodnov for this project. The drill system included a complete setup with a 700-m cable capacity and a controller unit. Due to the thickness of the Bona-Churchill ice field it was necessary to use multiple drilling technologies. The OSU system is designed to be quickly switched from a dry-hole, electro-mechanical drill (used to 180 meters) to a thermal-alcohol electric drill that collected core from 180 to 460 meters. The electro-mechanical drill physically “chews” its way through the ice, whereas the thermal drill melts its way downward.

(L) Mechanical Drill Bit; (R) Thermal Drill Bit

Six cores, totaling 623 meters, were obtained from the Bona-Churchill site. Core 1, the longest, was drilled 460 meters to bedrock. This is the deepest ice core to be recovered from an alpine ice field. Core 2 was 114 meters long. Four shorter cores, from 1-12 meters in length, were also recovered, to determine the impact of drifting on the various chemical and physical signals that are preserved in the ice layers and to assess the reproducibility of the records from the col. These shallow cores, along with the two deep cores, are used to help characterize the distribution of annual snow accumulation across the col. A single pulse radar-sounding unit was used to measure bedrock depths at 34 sites across the col. These data were used to prepare an ice thickness map, and the surface of the ice was mapped using GPS and an electronic altimeter.

(L) Core from 458 m below the surface; (R) Lonnie Thompson examines another ice core

The quality of the ice cores that were obtained at Bona-Churchill ranged from good to excellent. The drilling system is powered with a lightweight, highly fuel-efficient diesel generator that minimized environmental impact on the col. Use of this lightweight system also made the project more cost efficient by reducing airlift requirements. Both drills (electro-mechanical and thermal) produce core sections that are 100 mm in diameter and up to 2.1 meters long. A newly developed, quick-assembly geodesic dome, designed and built at OSU, housed all drilling and core processing activities.

Polar v. Tropical (or Mid-Latitude) Ice Cores

As you would expect, mountain glaciers away from the polar regions often have ice-free areas with bare soil and rock below the "snow line" (within a few miles of the ice). As a result, they are more likely to have layers that contain dust that is blown upslope during the dry part of the year. Dust from distant sources is also found on glaciers, often carried by far-traveling air masses and even by jet streams. Also, storm systems that travel across the lower- and mid-latitude regions move over forests, croplands, and grasslands, so they can deliver pollen, seeds, insects and may even carry small birds up to the glaciers from ecosystems below. Finally, many areas of the tropics are under the influence of monsoons and are therefore more likely to experience distinct wet and dry seasons. This influences the time of year when the greatest snowfall is expected to occur at high altitudes, which is important in the distinction of annual layers in the ice cores. This is in contrast to middle latitude regions, in which winters and summers more often show wide differences in seasonal temperature, rather than in precipitation. Instead of having distinct wet or dry seasons within the year, middle latitudes often have distinct warm and cold seasonal differences.

Among the organic material found deep within one Sajama core was this insect, (family Hemiptera), that was found 105 meters down, a depth representing about 6,000 years ago.

Ice cores from high-altitude tropical glaciers in monsoon regions are also more likely to contain dark-colored, easily-visible, dry-season dust layers. In contrast, the dust layers in cores from the large continental polar ice sheets (Antarctica and Greenland) are much lighter in color because they contain considerably less dust, most of which has traveled from distant sources. While there are summer and winter seasons in the polar regions, the sites where the cores are drilled may be separated from areas with bare rocks and soil by vast expanses of snow and ice and/or by the oceans. So there is less seasonal delivery of dust onto the surface of the great ice sheets. Also, because they are usually located far from ecosystems such as forests and grasslands, ice cores from the continent-sized ice sheets are less likely than tropical cores to contain biological material such as plant and insect fragments.

Three things determine how much time is included in an ice core record: (1) the thickness of the ice sheet or glacier, (2) the temperature of the ice at the bottom, and (3) the amount of snow that falls each year. Polar ice cores are much longer, both in length and in time, because: (1) the ice covering Greenland and Antarctica is much thicker (maximum ~10,000 feet) and (2) the temperatures are far below freezing at the base. (3) There is usually less snowfall on the polar ice sheets (especially in the interiors) than in the tropical mountains. As a result of pressure from the weight of the ice above, the ice near the bottom of the core is compressed. Thus, with increasing depth time is also compressed.