Ice Cores and Climate Change

Contents
The Earth's Climatic History on Ice
Mass Spectrometry and Electroconductivity
The Dansgaard-Oeschger Discovery
Deuterium Excess Increases Precision
Carbon Dioxide and Other Indirect Indicators
A Race Against Time
References

The Earth's Climatic History on Ice


Paleoclimatologists are concerned with the cycle of glacials and interglacials that have occurred throughout Earth's history. The variety of factors contributing to the climatic system results in complexities that are difficult to unravel. Ice core geochemistry has been instrumental in the quest for an understanding of Earth's climatic past.

Researcher Mary Davis examines a thick layer of dust inside an ice core taken from Mt. Kilimanjaro in Tanzania, Africa. The dust layer signifies a major drought event that struck the region in the past. By Lonnie G. Thompson, Ohio State University. Used by permission.

Ice core analysis is a fairly new science; the first deep cores were drilled in the 1960s. However, it has already yielded a wealth of information. Scientists turned to ice cores in earnest in the early 1980s, primarily to determine the effect of anthropological activities on the Earth. Although not reaching back deep into geologic time, ice cores do represent times long before humans began to influence the environment, and are therefore very valuable. Scientists can gather from their constituents a record of temperature, precipitation, atmospheric composition, volcanic eruption, solar variability, sea-surface productivity. Ice core records are most applicable to the study of greenhouse gas concentrations. They are in fact the most detailed record available. Plotting the depth against age creates an ice core chronology.

Glaciers are the result of snow accumulation over long periods of time. When dry snow accumulates, it is preserved in layers. Water trapped in glaciers remains in a pristine condition, preserving for us today clues about the climatic conditions that prevailed when it was deposited. Trace elements and aerosols are deposited on the surface, and gases are trapped inside the snow. It is from the analysis of these impurities that data can be obtained. For example, carbon dioxide, methane, and oxygen remain in proportion to that in the atmosphere.

Not all ice cores are equal. The variations include the depth to which they are drilled and the latitude from which they are obtained. Depths are generally categorized as either shallow, intermediate, or deep. Glaciologists do not randomly plunge their drills into ice sheets. Careful consideration of the drilling site is essential to a successful endeavor. Using ice cores from several latitudes scientists can gain several perspectives on the world's climatic oscillations. It is therefore not surprising to know that ice cores have been retrieved from around the world. Cores drilled in Greenland include Crete, Milcent, Summit, the U.S. Greenland Ice Sheet Project (GISP), GISP2, and the Russian-US-France collaborative efforts Greenland Ice Core Project (GRIP) and GRIP2. Cores drilled in the Antarctic include Byrd, Vostok and Dome C. Unfortunately, finding a prime location is not the only problem with which paleoclimatologists must contend. Analysis of the geochemistry can require a great deal of finesse as well.

Because the atmosphere mixes throughly every few years, changes in long-lived consitutents are excellent time indcators. Gas chromatographs equipped with electron capture dectectors measure gas concentration of air entrapped in ice cores. The trapping mechanism that isolates the air from the atmosphere works over a period of time, so the air is representative of a time period, not a single age. Because resolution is limited by this factor, times must be expressed with a corresponding uncertainty. The Byrd ice core, for example, has a ~200 year uncertainty associated with its data. Vostok's uncertainty is even greater, reaching several thousand years. This age difference can be estimated from models, provided the snow accumulation rate and temperature are known.

The upper parts of ice cores are dated by counting annual rings, as with tree rings; by indentification of annual layering scientists can ascertain the age of ice at a particular band. Specific events leave distinctive tracers that can be used to calibrate the ice core against another for which dates have already been established. Such markers are called "reference horizons." For example, the eruption of a volcano creates a reference horizon, and knowing the precise date of this eruption enables scientists to establish dates for bands above and below the eruption. Oxygen-18 ratios not only indicate ocean isotopic composition but can also correlate with marine isotope record of sea water oxygen isotopes, a well-established paleochronologic system (Bender et al., 1994). Radioactive decay can also be used as a dating mechanism. Lower down the ages are reconstructed by modeling accumulation rate variations and ice flow.

Although most of the analysis must take place under laboratory conditions, borehole logging measurements are critical. Measurements at the site of the core extraction can reveal vertical straining, hole tilting, ice temperature and acoustic and electrical properties allow scientists to determine ice sheet dynamics, and thus changes that may effect the ice core's stratigraphy.

Return to Top

Mass Spectrometry and Electroconductivity

Light stable isotopes, calculated as delta values, have been integral to climate research since the 1950s. It was at this time that researchers developed methods to make precise measurements of stable isotopes. The development of the oxygen isotope paleotemperature scale (Epstein et al., 1951; Urey et al, 1948; Urey et al., 1951) heralded the era of light stable isotope geochemistry. Isotopic differences are created by different numbers of neutrons in the nucleus of the atom. For example, one oxygen isotope has eight protons and eight neutrons. Another has ten neutrons instead. These different configurations are referred to as isotropomers.

Typically, ice cores are sealed in polyethylene bags and transported and stored below -15 ºC. After weighing, surface contaminants are removed by sublimating the outer edges of the core. Accelerator mass spectrometer measurements are taken on each section. Mass spectrometry is concerned with the separation of matter according to atomic and molecular masses. When an ion of mass m and charge z is acted upon by a potential difference of V, it acquires an energy E equal to zV. This energy represents the kinetic energy (KE) of 1/2 mv2. The equation thus becomes 1/2 mv2 = zV. The centrifugal force on the ion path is deflected by a magnetic field (B) such that mv/r=zB. Some algebraic manipulation yields m/z =B2 r2 /2V. If B and V are set in the mass spectrometer, m/z becomes solely dependent on r, and the heaviest isotopomer is the least deflected.

A researcher prepares a sample for mass spectrometry. Photo courtesy of Argonne National Laboratory. Used by permission.

Electroconductivity is used to quickly find regions of interest in the ice core. It detects the pH values across sections of the core. Significant events such as volcanic eruptions leave acidic tracers in the ice core.

Return to Top

The Dansgaard-Oeschger Discovery

Because oxygen-18 is heavier than oxygen-16, water vapor from the ocean tends to have a slightly higher ratio of 16O to 18O than that of the remaining ocean water. As water vapors traverses the Earth to the poles, the snow precipitated from this water vapor is becomes increasingly richer in 16O relative to 18O, as 18O precipitates first. If glaciers expand, the oxygen isotope ratio of ocean water becomes increasingly rich in 18O, as 16O is locked away in glacial ice. In addition, the process is dependent on temperature, so snow that falls in the winter is enriched is 16O. Thus, ice cores can provide a dating chronology, as well as a measure of climate cooling. Each annual layer begins 18O rich, becomes 18O poor, and ends 18O rich again. Furthermore, during cold periods, glaciers are relatively enhanced in 16O, while the oceans are relatively enriched in 18O. This imbalance is more marked for colder climates than for warmer climates.

For paleoclimatological purposes, a delta value is used to indicate the relative enrichment of 18O. The delta notation signifies { [(18O/16Osample) - (18O/(16O standard)]/(18O/(16O standard)} x 1000. The Vienna standard for mean ocean water is 12005.2 x 10-6.

Dansgaard-Oeschger (DO) events are characterized by rapid increases in 18O relative to 16O in ice cores of the last glacial period. These high values persist for several hundred to several thousand years. When these increases were discovered in 1972 it was not clear whether they reflected sudden arctic climatic changes, local effects, or stratigraphic disturbances in the core. Because of their widespread geographic extent they are now interpreted as warmer interstadial periods.

Return to Top

Deuterium Excess Increases Precision

Hydrogen also has isotopes. If only a single proton is present in the nucleus, this atom is called ordinary hydrogen. If a neutron is present as well, the atom is called heavy hydrogen, or deuterium. Deuterium is a stable atomic species that composes up to .014 to 0.015 percent of natural hydrogen compounds. It reacts more slowly than ordinary hydrogen, it can be used as a climatological tracer in the same manner as O-18. In fact is is also calculated as a delta value. In Antarctica, a cooling of 1°C results in a decrease of 9 per mil deuterium (Abysov et al. 1995).

Deuterium is calculated as a delta value, just as oxygen-18, and also reveals paleotemperature. It is not entirely redundant however, because of a quantity known as deuterium excess. The line comparing both deuterium and oxygen-18 at various locations is called the Global Meteoric Water Line (GMWL), which defines a unit called deuterium excess: d= dD – 8xd18O (Craig, 1961). Deuterium excess (d), or the deviation of a sample from the meteoric water line (GMWL) is an indicator of the complexity of the air circulation. Deuterium excess analysis supports complex histories of precipitation and re-evaporation from ice cores (Thompson et al., 1995). Delta d increases with elevation and can therefore represent conditions at the place where snow fell before it flowed as glacier ice. Deuterium signatures in snow and ice become lighter with elevation of distance inland.

Return to Top

Carbon Dioxide and Other Indirect Indicators

While oxygen and deuterium are used as direct paleoclimatic indicators, indirect methods are available was well. Greenhouse gases and atmospheric dust load indicate radiative forcing, and cosmogenic isotopes such a beryllium indicate solar forcing. Dry gas extraction enables scientists to find the isotopic composition of CO2 in ice core gas. Carbon dioxide can further be compared with deuterium values, both of which indicate global warming. When their rises and falls correlate, the carbon dioxide record can be taken as relatively accurate. Discovering which gas increased first can shed light on climate dynamics. For example, Monnin et al. found that the start of CO2 increases lagged the start of deuterium increases by 800 ± 600 years in ice cores from Dome Concordia, Antarctica (2001).

Monnin et al. also investigated the possibility of alteration of carbon dioxide by chemical reactions in glacier ice (2001). The most likely sources would be acid-carbonate reactions and the oxidation of organic compounds. However, they concluded that because the scatter of carbon dioxide values from neighboring neighboring samples was in agreement with the analytical uncertainty, the measurements were an accurate representation of atmospheric carbon dioxide.

From this photo a glacier in Banff, Canada, the striated nature of glaciers is apparent. Photo courtesy of Public Domain Photo (URL: www.pdphoto.org).

Nitrous oxide is an atmospheric trace greenhouse gas produced from ocean upwelling and soils in tropical and temperate regions. It photodissociates in the stratosphere. Due to human activities its concentration increases about 0.25% per year. Flückiger et al., using two ice cores from Summit in Central Greenland, discovered that nitrous oxide concentrations increased in concert with temperature variations in the Northern hemisphere (1999). However, they note that nitrous oxide measurement could be affected by a chemical reaction in the ice, a postcoring effect, or an analytical artefact. Retesting eliminated the possibility of contamination, but not chemical reaction or artefact sources.

Microparticles, soluble impurities, heavy metals, the ionic concentrations of trace elements and entrapped gases are also useful. Sodium and magnesium are used to trace sea salt. Aluminum, iron, and rare earth elements represent mineral dust. Zinc and lead indicate anthropogenic activities. When lakes are dry, salts are carried by the wind and deposited in snow.

Return to Top

A Race Against Time

In 2000, a team of scientists from Ohio State scaled Mount Kalamajaro to retrieved 214 meters of ices. Using a chlorine-36 marker, they were able to correlate a section of the ice with the 1951-1952 atomic bomb testing. Depletion of oxygen-18 in the cores 5,200 years ago has been interpreted differently: Barker et al. argue that it reflects anomalously heavy snowfall (2001), while Thompson et al. interpret it in terms of a substantial cooling (2002). Because warm and wet regions have similar delta O-18 averages, Thompson insists that temperature must be the deciding factor.

This is an illustration of how the confounding factors outlined in this Web site can prevent an accurate interpretation of the current climate. New methods of geochemical analysis may be helpful in resolving these issues. Reinhardt et al. experimented with laser ablation inductively coupled plasma mass spectrometry for trace element analysis (2001). The laser vaporizes a small inclusion of particles in the ice and analyzes them as well as dissolved minerals. They discovered that it presents a lower risk of contamination than conventional methods, and that a high spatial resolution is possible. In addition Maruo et al. has discovered that fluorometric detection can be used to detect ammonia and calcium in ice cores (2001). However, there is no time to spare in these developments. Ice core analysis is the greatest asset paleoclmatologists have, but it is also a disappearing one. The ice cores that are so critical for climate study are also vunerable to its effects, as illustrated by the map below.

This map shows the retreat of Mt. Kilimanjaro’s ice cap since 1912. During the period shown on the map, more than 80 percent of the mountain's glaciers were lost. All ice will probably be lost on the mountaintop within 15 years. By Lonnie G. Thompson, Ohio State University. Used by permission.

Isotope signals from the poles and Greenland have shown a weak response to small-scale climatic changes such as the Little Ice Age. However, cold mountain glaciers are an excellent indicator of these changes. Unfortunately, alpine drill sites are subject to many more restrictions than those of higher latitude. For example, complex ice flow patterns and wind erosion can complicate calculations (Wagenbach). Only six low latitude ice cores represent the last millennium, and tropical ice core chronologies are especially important because this region has considerable impact on the world as a whole (Thompson et al. 2002).

Alpine ice masses exist close to the melting point and are thus sensitive to small changes in temperature. Furthermore, Thompson et al. argue vehemently that the tropics are undergoing an anomalous warming (2002). The Kilimanjaro dust layer that indicated a drought severe enough to topple civilizations didn't cause the complete melting of Kilimanjaro's ice cap. The anticipated melting will be unprecedented in the Holocene.

Return to Top

References

Abysov, S.S., Angelis, M., Barkov, N.I., Barnola, J. M., Bender, M., Chappellaz, J., Chistiakov, V. K., Duval, P., Genthon, C., Jouzel, J., Kotlyakov, V. M., Kudriashov, B. B., Lipenkov, V. Y., Legrand, M., Lorius, C., Malaize, B., Martinerie, P., Nikolayev, V. I., Petit, J. R., Raynaud, D., Raisbeck, G., Ritz, C., Salamatin, A. N., Saltzman, E., Sowers, T., Stievenard, M., Vostretsov, R. N., Wahlen, M., Waelbroeck, C., Yiou, F., Yiou P. 1995. Deciphering Mysteries of Past Climate From Antarctic Ice Cores Earth in Space 8(3): 9.

Barker, P.A., Street-Perrott, F.A., Leng, M.J., Greenwood, P.B., Swain, D.L., Perrott, R.A., Telford, R.J., Ficken, K.J. 2001. A 14,000-year oxygen isotope record from diatom silica in two alpine lakes on Mt. Kenya. Science 292:2307-2310.

Bender, M., T., Souers, M.L. Dinckson, J. orchardo, P. Grootes, P. Mayewski, and D. Meese. 1994. Climate connectison getween Greenalnd and Antarctica during the last 100,000 years. Nature 372: 663-666.

Craig, H. 1961. Isotopic variations in meteoric waters. Science 133: 1702-1703.

Flückinger, J., Dällenbach, A., Blunier, T., Stauffer, B., Stocker, T.F., Raynaud, D., and Barnola, J.-M. 1999. Variations in Atmospheric N2O Concentration During Abrupt Climate Changes. Science 285: 227-230.

Maruo, Masahiro, Nakavama, Eiichior, Obata, Hajime, Kamiyama, Kokichi, and Kimoto, Takashi 2001. Application of flow-through analyses of ammonia and calcium in ice core and fresh water by fluormetric detection. Field Analytical Chemistry and Technology 5(1-2):29-36.

Monnin, Eric, Indermühle, Andreas, Dällenback, Andre, Flückinger, Stauffer, Bernhard, Stocker, Thomas F., Raynaud, Dominique, and Barnola, Jean-Marc 2001. Atmospheric CO2 Concentrations over the Last Glacial Termination. Science 291: 112-114.

Reinhardt, H., Kriews, M., Miller, H., Schrems, O., Ludke, C., Hoffman, E., and Skole, J. 2001. Laster ablation inductively couple plasma mass spectrometry: a new tool for trace element analysis in ice cores. Fresenius J. Anal. Chem. 370:629-636.

Thompson, L.G., Mosley-Thompson, E., Davis, M.E., Lin, Ping-Nan, Henderson, Keith A., Cole-Dai, J., Bolzan, J.F. and Liu, K-b. 1995. Glacial Stage and Holocene tropical Ice core Records from Huascaran, Peru. Science 269:46.

Thompson, Lonnie G., Mosley-Thompson, Ellen, Davis, Mary E., Henderson, Keith A., Brecher, Henry H., Zagorodnov, Victor S., Mashiotta, Tracy A., Lin, Ping-Nan, Mikhalenko, Vladimir, Hardy, Douglas R., Beer, Jurg. 2002. Kilimanjaro Ice Core Records: Evidence of Holocene climate hange in tropical Africa. Science 298 (5593): 589-593.

Thompson, Lonnie G., Mosley-Thompson, Ellen, Davis, M. E., Lin, Ping-Nan, Henderson, K., and Mashiotta, T.A. 2003. Tropical Glacier and Ice Core Evidence of Climate Changes on Annual to Millennial Time Scales. Climatic Change 59: 137-155.

Urey, H.C. 1948. Oxygen isotopes in nature and in the laboratory. Science 108: 489-196.

Urey, H.C., Lowenstam, H.A., Epstein, S., and McKinney C. R. 1951. Measurement of paleotemperatures and temperatures of the Upper Cretaceous of England, Denmark, and Southeastern United Staets. Bull. Geol. Soc. Am. 62: 399-416.

Wagenbach, D. Alpine ice cores as climate proxies. URL: http://www.zamg.ac.at/ALP-IMP/downloads/session_wagenbach.pdf
Retrieved 11/30/2004.

Report for ES 767 Quaternary Geology
Cheryl Sedlacek, December 2004