GISP2 Notebook

Number 2, Spring 1992

As of camp closing September 14, 1991, the GISP2 program had reached a depth of 1510 meters, recovering ice with an age of approximately 8000 years BP. During the 1991 season, discrete and continuous samples were taken to a depth of 719 meters (approximately 3000 BP) and from 1372 to 1510 meters. Between 719 meters and 1372 meters, the brittle ice zone was encountered and sampling over this depth range was postponed until the 1992 field season to assure sufficient time for core relaxation and thus facilitate sampling of this section.

Since the last issue of the Notebook, we have convened GISP2 science working meetings at Boulder, Lake Tahoe and Miami. These sessions have been focused primarily on preliminary reporting of data and interpretations plus planning for the next yearÍs field season. The figure on this page is a generalized example (50 year smoothing) of the type of multi-disciplinary data sets now evolving from GISP2. From this record it is clear that even relatively small (by comparison with hose to be encountered in the 200,000 year record) climatic events of the last millennium are sensitively recorded by several core properties. Detailed interpretations of the last 2000 years of the GISP2 paleo-environmental ice-core record calibrated with historical and instrumental records will provide the necessary analog for interpreting the record in pre-instrumental and pre-historic time.

In addition to our group meetings, GISP2 researchers have presented results of their work at several national and international meetings and have published several articles. Abstracts of published papers and titles of those in press and in review appear in this Notebook.

The Atmospheric Sampling Facility, 30 km upwind from GISP2 Camp, has continued to grow. There are now 13 sampling lines in operation powered by 6 kw of solar panels. Experiments by, or on the behalf of, 30 U.S. and European investigators are being carried out to develop our understanding of the atmospheric environment of the Summit region.

As of the 1991 field season, GISP2 has welcomed several new scientific efforts. These new projects are summarized in this issue of the notebook.

We continue to enjoy and appreciate the interaction with our colleagues to the east in the GRIP (Greenland Ice Core Project) camp. A joint GRIP-GISP2 scientific workshop is being planned for Spring, 1993.

Finally, as the 1992 field season rapidly approaches and we look to the future, we should not forget the hard work and dedication that has already gone into GISP2 and the individuals who have contributed to our success (see List of Participants, page 9).

We are proud of the achievements won by the efforts of all of the individuals who developed, constructed, and operated the drill, the science trench and the camp, those who provided support from Sondrestrom and the good hearted dedication of the 109th ANG who maintained the flow of equipment and personnel necessary to a successful field season.

R.D. Borys, Desert Research Institute, University of Nevada, Reno

Collections of snow crystal replicas and cloud water made at the GISP2 ice coring site in Greenland were analyzed to ascertain the mode of ice crystal growth and aerosol scavenging. Ice particle sizes and habits, cloud water chemistry and total snow chemistry were used to estimate the importance of ice particle riming on chemical wet deposition at summit. The field observations showed that riming was not an important ice particle growth mechanism and thus it was estimated that riming does not contribute significantly to the scavenging of aerosols by snow. Cloud droplet (fog) deposition (occult precipitation) was briefly investigated. Because of the frequency of fogs, the relatively high solute concentrations found in fog water and the large droplets present in the fogs it was concluded fog deposition may contribute a significant fraction of the total wet chemical deposition to the ice sheet at summit. Fogs may also contribute up to 10% of the water mass.

E.A. Boyle, Massachusetts Institute of Technology

Large meteoritic impacts probably exert important modifications of global climates and ecosystems, but the size and frequency of these events is not well established. Polar ice cores preserve a record of the accretion of materials from the solar system. The element iridium, which is highly enriched in extraterrestrial debris relative to crustal materials, has been determined in Antarctica ice as a measure of the steady state cosmic debris influx, and a pulse of cosmic iridium has been reported to coincide with the 1908 Tunguska impact event. It is likely that many events of Tunguska magnitude should be preserved in a deep ice core. A complete record of Ir in a long polar ice core would allow us to determine the size -frequency history of previous cosmic impacts. It would also allow us to determine whether the more uniform background influx of small particles is truly constant. To observe these pulses, it is necessary to distinguish them from normal background crustal, volcanic, and cosmic dust Ir deposition. A study of Ir in the proximity of known volcanic events recorded in ice cores is needed to provides basis for this correction. Similarly, while the concentration of Ir in crustal dusts is low, the total concentration of crustal dusts in ice cores is high relative to the abundance of cosmic dusts. To allow for the possible influence of these high concentrations of crustal dusts on the cosmic Ir signal, it is necessary to examine the Ir concentration in relation to changes in the terrestrial dust record. Further confirmation of the extraterrestrial nature of Ir can be provided by measurement of the isotopic composition of osmium. Due to fractionation of radioactive parent ¹⁸⁷Re from radiogenic daughter ¹⁸⁷Os during magmatic processes, the ratio of ¹⁸⁷Os/¹⁸⁶Os is 400 in crustal rocks while it is only 3 in meteorites. Hence variation of the osmium isotope ratio can help to distinguish cosmic (or mantle- derived) Os from meteoritic Os. We propose to undertake these studies as an ancillary part of the ongoing GISP2 ice core program.

J-L Jaffrezo, C.I. Davidson, and M.J. Small, Carnegie Mellon University

GISP2 involves acquisition of a deep core to bedrock from the Summit region of Greenland. The core will provide a detailed record of some 200,000 years, yielding information of the earthÍs geologic history, climate change, and variations in chemical composition of atmospheric constituents. This project will investigate transport of chemical species in the atmosphere from source regions to the Greenland Ice Sheet at Summit, in order to better interpret data from the GISP2 core. There are three specific objectives. First, source regions and atmospheric pathways for the chemical constituents reaching Summit will be identified. Possible changes in characteristics of the air masses during transport will be explored. Second, incorporation of these chemical species into precipitation at Summit will be investigated, in order to determine the mechanisms and rates of deposition during snowstorms. Finally, changes in the chemical composition of the Ice Sheet after snowfall will be identified. These changes include additional input of chemical species by dry deposition onto surface snow, and also redistribution of the species within the snowpack. The objectives will be achieved by a program that incorporates field work, laboratory analyses, and computer modeling. The field work includes collection of aerosol, air, and precipitation samples; these will be analyzed in the laboratory for a wide variety of species such as anions and cations, trace elements, carbon compounds, and trace gases. Computer modeling work includes development of a climatology for Summit as well as mathematical modeling of the deposition processes. The results of this project will identify the most important source regions, atmospheric pathways, and deposition mechanisms influencing chemical constituents; in the deep core. This in turn will help separate the effects of variations in airborne concentration and variations in deposition rate in influencing these constituents in the core. The results will also improve our ability to derive relations between snow accumulation rate and concentrations throughout the core.

M. Ram and M. Illing, SUNY Buffalo

We have designed and built a 90Á laser-light-scattering (LLS) instrument for measuring the dust concentration along an ice core. A specially designed heater melts ice along the core and melt water is passed into an optical cell where the dust concentration is measured by scattering laser light off the melt water and observing the light that is scattered at 90Á. At present, the instrument is capable of a resolution of 2 mm which will allow us to observe sharp seasonal dust peaks even when the yearly accumulation is as small as 1 cm of ice core. Thus, we will be able to date the core down into the glacial stages with no difficulty. Indeed, we have found that the seasonal dust peaks become sharper and more distinct as the depth increases. At present, we are able to measure 12 m of ice in a 24 hr period. This is not sufficient to keep up with core processing and the ice that is not measured in the field has to be measured in our lab in Buffalo.

In addition to measuring seasonal dust peaks, we are also able to measure changes in background dust levels. We have already observed two very significant changes in dust background levels during the Holocene each lasting approximately fifteen years. We will look for more events of this kind and will analyze the mineralogy and size of the dust that is associated with them. This will help us understand the origin and nature of these unusual increases in dust levels.

We are developing an instrument for carrying our laser-light-scattering measurements directly on ice. We tested the technique in the summer of 1991 and it looked very promising. We will have an instrument ready for use in the summer of 1992. Direct measurements on ice are much faster (it is possible to process one meter in 10-15 minutes) and non-destructive. In addition, we believe that we will be able to measure other properties of the ice by this method and not just changes in dust concentration. Since many properties of ice change as it ages, these laser scattering measurements on ice will have to be done in the field.

E.D. Waddington, D.R. MacAyeal, and J. Firestone, University of Washington

One of the major goals of glaciological research is the inference of past climate from ice cores. One way to make this inference is through the measurement and analysis of the temperature-depth profile at the drill site. Temperature analysis is independent of isotopic techniques, and thus can provide an independent check on interpretations of isotopic signals (e.g., was there a Younger Dryas signal in the Dye-3 oxygen-isotope profile, or does the oxygen-isotope profile merely reflect the abundance of glacial meltwater in the North Atlantic?)

Despite sophisticated understanding of the physics of heat transfer in ice sheets, analysis of temperature-depth measurements has been rather crude. Highly developed temperature models are used in a trial-and -error mode to select environmental histories which yield the best fit to todayÍs ice-sheet temperature observations. We have developed the rudiments of an analytical technique which, if coupled with the sophisticated heat transfer models, could yield a major improvement over the trial-and-error method (it is more efficient and less subjective). This method comes from a branch of applied mathematics called control theory ( the precise name of our technique is the adjoint trajectory method). We have tested our method by repeating the Dye-3 temperature analysis.

We need to develop the adjoint- trajectory method for the inference of past climate to a greater degree before it can be applied to the GISP2 data. We anticipate that application of our method to the GISP2 data will yield a surface-temperature history for the Greenland Ice Sheet (over the Holocene at least) that is independent of all isotopic techniques.

B.W. Mosher, Complex Systems Research Center, University of New Hampshire

Trace element records preserved in high latitude glaciers such as the Greenland Ice Sheet have the potential to tell us much concerning regional and global scale atmospheric and paleoatmospheric chemistry. Historical information concerning the timing and magnitude of anthropogenic emissions, volcanic eruptions, and climatic change is contained in this record. Before this record can be fully and accurately interpreted we must understand the link between atmospheric composition and snow chemistry. The establishment of an atmospheric sampling program as part of GISP2 provides an excellent opportunity to conduct studies of wet depositional processes active in the high Arctic. This investigation will provide measurements of aerosol and concurrent snow elemental composition with which possible fractionation processes may be examined. In addition this work, when combined with meteorological air mass trajectory analysis, will help to identify the types, locations and transport pathways of contaminants found in the ice sheet. Trace metal data from summer snow samples, when compared with annual trace metal accumulation data, will also allow us to examine the seasonal timing of net chemical deposition to the ice sheet. Additionally, this information will be of value in interpreting the trace metal record contained in the GISP2 deep ice core.

K. Nishiizumi, J.R. Arnold, and R.C. Finkel, University of California at San Diego and Lawrence Livermore National Laboratory

We will use accelerator mass spectrometry to measure ¹⁰Be, ²⁶Al, and ³⁶Cl at selected depths in the GISP2 Summit ice core. We will study cosmogenic nuclide concentrations in both Pleistocene and Holocene sections of the Summit deep core. The resulting time series of nuclide concentrations will be applied to three main problem areas: dating of ice cores, deducing the history of solar activity and of variations in he geomagnetic field, and studying climatic history through effects of atmospheric circulation and of atmospheric chemistry on nuclide deposition.

T. Sowers, Penn State University

I propose to construct a 110,000 year record of the concentration of N2O in the atmosphere from the occluded air parcels in the GISP II ice core. This record will provide information on nitrogen biogeochemical cycles over glacial/interglacial timescales. The concentration of N2O in the atmosphere today (1997) is close to 313 ppbv. Due to its ability to absorb long wave radiation, N2O is considered to be a "greenhouse" gas.

Variations in the paleoatmospheric levels will have impacted the latitudinal distribution of outgoing radiation, in turn impacting global climate. The concentration of N2O in air is determined by the sources and sinks of N2O on a global scale. The only sources of N2O are terrestrial and marine biospheres and the major sink of N20 is photodissociation in the stratosphere. Estimates of the magnitude of the sources and sinks are not well constrained but it appears as though the terrestrial biosphere contributes about 80% of the global N2O emissions with the other 20% being made up from an oceanic contribution.

Comparing an atmospheric N2O record with other bioactive gases, such as CO2 and CH4 from trapped gases in ice, will help in understanding the nature of changes in global carbon cycling throughout the past 200,000 years. In addition, N2O will be a useful stratigraphic tool for correlating Greenland and Antarctic ice cores since the N2O records versus time must be the same for each ice core when the records are compared on a common timescale.