The recovery and dating of carbon dioxide in polar ice cores
Wilson, A. T. and Donahue, D. J.
Radiocarbon, Vol. 31, No. 3, p. 579-584, 1989

A new method is described for recovering trapped carbon dioxide from polar ice cores. The ice is sublimed under vacuum, and water vapor and carbon dioxide are collected at appropriate cold traps. The application of this method to obtain carbon dioxide from a specific ice core, the conversion of that carbon dioxide to graphite, and the measurement of radiocarbon in the carbon dioxide are described in detail. The potentialities and problems of the method are discussed.

AMS carbon-14 dating of ice: progress and future prospects
Wilson, A. T. and Donahue, D.J.
Nuclear Instruments and Methods in Physics Research , Vol. B52, p. 473-476, 1990

The "sublimation technique" for the recovery of carbon dioxide from ice samples and the conversion of the recovered carbon dioxide into graphite for AMS dating will be described, together with its use in some applications. The technique involves placing the ice sample in a carefully degassed glass vacuum system, "cleaning" the ice by removing the outer few millimeters by sublimation, then subliming the ice completely using infrared lamps as an energy source. The gases evolved from the subliming ice first pass through a trap at -80íC to remove water. The remaining gases pass through a trap immersed in liquid nitrogen where the carbon dioxide is recovered. The gases which pass the liquid nitrogen trap are trapped on molecular sieve held at liquid nitrogen temperatures. In our apparatus we can sublime a 3 kg sample of ice core in 18 hours. We measure the quantity of water sublimed, the amount of air and carbon dioxide recovered and the C14 content of the carbon dioxide. From these data we can calculate the altitude of the ice sheet and the CO2 concentration of the atmosphere at the time given by our C14 date. We have run ice samples as small as 1 kg in the "bomb pulse" zone where the activity is high and the spatial resolution in the core is of concern. The future prospects for the technique look promising and we ultimately hope to be able to date polar ice cores back in time as far as is achieved by AMS dating of more conventional methods.

A simple technique for converting CO2 to AMS target graphite
Wilson, A. T.
Radiocarbon, Vol. 34, No. 3, p. 318-320, 1992

I describe a simple, rapid and inexpensive method for converting CO2 samples into AMS target graphite. The technique is applicable for both small- and large-scale production facilities. With some modification, the method is suitable for use with very small samples.

AMS radiocarbon dating of ice: Validity of the technique and the problem of cosmogenic in-situ production in polar ice cores
Wilson, A. T., and D.J. Donahue
Radiocarbon, Vol. 34, No. 3, p. 431-435, 1992

In the "sublimation technique," carbon dioxide entrapped in the ice is recovered by sublimation, converted to graphite and ratio of l4C/l3C in the CO2 determined by AMS measurements. We describe here several experiments performed to check the validity of such measurements and to study the effect of cosmogenically produced in-situ 14C on the measurements.

Ice-core dating and chemistry by direct-current electrical conductivity
Taylor, K., R. Alley, J. Fiacco, P. Grootes, G. Lamorey, P. Mayewski, and M. J. Spencer
Journal of Glaciology, Vol. 38, No. 130, p. 325-332, 1992

Although quantitative interpretation of the low frequency electrical conductivity of ice cores from central Greenland is complicated by temperature variations of the measured core, annual layers can be recognized in sections of the core that are not impacted by non seasonal features. Ambiguities in counting of annual layers can be minimized by comparing the electrical conductivity measurements to measurements of dust concentration and visual stratigraphy. A non-linear relationship between applied voltage and the current measured across two electrodes complicates the quantitative comparison of measurements made with different equipment, but does not affect the overall shape of the observed features.

Determination of major ions in snow and ice cores by ion chromatography
Buck, C. F., P. A. Mayewski, M. J. Spencer, S. Whitlow, M. S. Twickler and D. Barrett
Journal of Chromatography, Vol. 594, p. 225-228, 1992

The determination of major anions (Cl-, NO-3, SO2-4) and cations (Na+, NH+4, K+, Mg2+, Ca2+) in snow and ice cores by ion chromatography at trace level concentrations (ng/g) is presented. Total acidity (H+) was determined using an acid titration method in order to complete the ionic balance. Unique sampling techniques and sample preparation methods were developed to avoid contamination of the snow and ice samples.

Toward using borehole temperatures to calibrate an isotopic paleothermometer in central Greenland
Cuffey, K. M., R. B. Alley, P. M. Grootes and S. Anandakrishnan
Palaeogeography, Palaeoclimatology, Palaeoecology, Global and Planetary Change Section), Vol. 98, p. 265-268, 1992

Analysis of borehole temperatures in ice sheets can improve the accuracy of isotopic paleothermometry by making possible independent calibrations of isotope-temperature relations. Here we present a preliminary calibration of the delta 18O paleothermometer for central Greenland. A numerical thermal model converts isotope-derived surface-temperature histories to temperature-depth profiles. Comparison with measured borehole temperatures allows calibration of the paleothermometer using formal inverse techniques.

Spatial and temporal characterization of hoar formation in central Greenland using SSM/I brightness temperatures
Shuman, C.A. and R. B. Alley
Geophysical Research Letters, Vol. 20, No. 23, p. 2643-2646, 1993

The summertime formation and burial of low-density, coarse-grained, hoar layers in the snow of central Greenland can be mapped using satellite passive-microwave data. Variations in a signal derived from Special Sensor Microwave/Imager (SSM/I) brightness temperature trends correlate temporally and spatially with hoar complex formation over 120,000 km2 of the Greenland ice sheet's dry firn facies. Observations at the Greenland Ice Sheet Project II (GISP2) site indicate that changes in surface conditions and microwave data correspond to four hoar events over two summers, as expected from theory. Following a snowfall, the smooth, high-density surface reflects some emitted, 37 GHz, horizontally- polarized (H) radiation but little vertically-polarized (V) radiation. Progressive surface roughness increase and density decrease during hoar formation causes a gradual decrease in H reflection. Formation and burial of a hoar layer thus causes a slow decrease followed by an abrupt increase in the V/H ratio. Hoar layers have been used to date the GISP2 ice core through the entire Holocene; archived microwave data now can be studied to assess the timing and frequency of the formation of these extensive stratigraphic markers in central Greenland.

Characterization of a hoar-development episode using SSM/I brightness temperatures in the vicinity of the GISP2 site, Greenland
Shuman, C. A., R. B. Alley and S. Anandakrishnan
Annals of Glaciology, Vol. 17, p. 183-188, 1993

Formation of a surface-hoar/depth-hoar complex at the GISP2 site in central Greenland was correlated with large changes in Special Sensor Microwave/Imager (SSM/I) brightness-temperature data. Pass- averaged SSM/I brightness-temperature data over a 1/2 degree latitude by 1 degree longitude cell for the 19 and 37 GHz, vertically (V) and horizontally (H) polarized bands were manipulated to yield differential (V-H) trends which clearly show a gradual decline as the hoar formation caused a progressively rougher surface with progressively lower density. The hoar episode ended as snowfall and high winds buried and destroyed the surface-hoar layer and caused rapid V-H increases in approximately 1 day. Comparison of the differential trends to changes in the field-monitored variables and theoretical values suggests that the V-H trends are sensitive primarily to changes in surface roughness, and secondarily to near-surface density changes. Consistent expression of trends in microwave brightness temperature over 35 adjacent study cells indicates that this technique may provide a remote-sensing signature capable of defining the timing and spatial extent of surface- and depth-hoar formation in central Greenland.

Deconvolution of continuous borehole temperature logs: Example from the Greenland GISP2 ice core hole
Saltus, R.W. and G.D. Clow
USGS Open File Report, No. 94-254, 1994

Measurements of borehole temperatures obtained from a continuously moving thermistor probe are a convolution of the actual borehole temperatures with the finite thermal response function of the temperature measurement system. To estimate actual temperatures from the measurements requires deconvolution, an operation that can magnify noise. We discuss the use of optimal (Wiener) filtering for noise reduction, followed by an exact deconvolution using serial division. We test the method using synthetic data and then apply it to measurements made in May 1992 at the GISP2 ice core hole in Greenland. After optimal filtering of the 1992 GISP2 measurements, the data power spectrum falls to 10% of maximum as frequencies drop below 0.004 Hz (periods of 250 seconds, equivalent to about 20 m in depth at a logging velocity of about 7.4 cm/s). The noise at higher frequencies (shorter wavelengths) is probably a result of noise caused by the mechanical slip-ring that provides electrical connection to the thermister cable.

137Cs gamma-rays from fallout detected at Summit, Greenland
Dunphy, P. and J. E. Dibb
Journal of Glaciology, Vol. 40, No. 134, p. 87-92, 1994

Global fallout from atmospheric testing of thermonuclear weapons produced horizon markers corresponding to the initiation of testing in 1953 and the maximum fallout in 1963. The radioactive isotope 137Cs associated with these events has a half-life of 30.2 years. Therefore, with the appropriate radiation detectors, this fallout can be used as a long-term temporal indicator in glaciers and snowpack. A prototype g-ray detector system was successfully tested and was used to make in-situ measurements of the 137Cs marker in a borehole at Summit, Greenland. The system consisted of a 7.6 cm by 7.6 cm NaI(Tl) scintillation crystal/photomultiplier detector, commercial preamplifier, amplifier, and power supplies, and a microcomputer-based pulse height analyzer. The measurements were made in boreholes of 25.4-cm and 12.7-cm diameter to depths of 22 m. Based on the results reported here, the g-ray detection technique promises to be a powerful way to quickly locate horizon markers in the field.

A gamma-ray detector for in-situ measurement of 137Cs radioactivity in snowfields and glaciers
Dunphy, P., J.E. Dibb, and E.L. Chupp
Nuclear Instruments and Methods, Section A 353, p. 482-485, 1994

The rate of snow deposition at various cold regions on the earth is an important quantity for glaciological and climatological studies. Radioactive debris from above-ground tests of nuclear weapons (mainly 1954- 1970) and from the Chernobyl accident (1986) have been deposited on glaciers and snowfields, where they can be used as time and depth markers to determine the subsequent accumulation of snow. We discuss a technique to locate these markers that has been used just recently - in-situ measurement of g-rays from 137Cs. These g-rays, which are associated with radioactive fallout, have a distinctive depth profile and serve as markers of the historical nuclear events. The g-ray measurement involves lowering a scintillation detector down a borehole in the snow or ice and recording the response to the 137Cs g-rays as a function of depth. The in-situ measurement can be done relatively quickly and can replace sample retrieval, or it can be used to decide which ice or snow samples should be transported for later analysis in the laboratory. The feasibility of in-situ g-ray measurement has been demonstrated at sites in the French Alps and Greenland. We report on a portable detector system that is being developed for use in Antarctica. It is based, as much as possible, on inexpensive, commercially available detectors and electronics. The advantages and disadvantages of this approach are discussed. The problems involved with making these measurements in a harsh environment and the steps taken to deal with them are also presented.

Polar ice stratigraphy from laser-light scattering: Scattering from meltwater.
Ram, M., and M. Illing
J. Glaciology, Vol. 40, No. 136, p. 504-508, 1994.

We describe a new laser-light scattering instrument for measuring variations in dust concentration along polar ice cores. We have used this instrument with considerable success on the GISP2 ice core that is being retrieved from central Greenland. The distinct annual spring/summer dust peaks we observe are being used to date the core with tree-ring-like precision.

Calibration of the delta18O isotopic paleothermometer for Central Greenland, using borehole temperatures: Results and sensitivity
Cuffey, K., R. Alley, P. Grootes, J. Bolzan, and S. Anandakrishnan
Journal of Glaciology, Vol. 40, p. 341-349, 1994

We calibrate the delta 18-O paleothermometer for central Greenland using borehole temperatures, a thermal model forced by a measured delta 18-O record, and formal inverse techniques. The calibration is determined largely by temperature fluctuations of the last several centuries, including the little ice age. Results are generally insensitive to model variables. Results of this borehole calibration also seem to be in good agreement with modern spatial gradients of delta and temperature. We suggest that calibrations of isotopic paleothermometers using borehole temperatures are a useful paleoclimate tool because they are independent of spatial gradients and include the effects of prehistoric temperatures over ice sheets.

High-precision temperature logging at GISP2, Greenland, May 1992
Clow, G.D., R.W. Saltus and E.D. Waddington
USGS Open File Report, No. 95-490, 1995

We describe a logging system capable of making high-precision temperature measurements and present an analysis of the measurement errors. This system was used to acquire a continuous temperature log in the GISP2 ice core hole just before drilling operations began for the 1992 field season. The system's sensitivity was 0.14 mK (0.00014C) during this log. The precision of the processed data is 0.7 mK and the accuracy is 4.5 mK. Temperature gradients were found to slightly exceed the theoretical gradient required for the onset of natural convection of the borehole fluid (n-butyl acetate) in a narrow zone between 210 and 330 m. However, a monitoring experiment failed to detect evidence for convection at these depths. At the time of the 1992 log, temperatures in and around most of the borehole are predicted to have still been perturbed by 0.4-1.0 mK due to the previous year's drilling activities; within a couple hundred meters of the surface, the disturbance is predicted to have been much larger (~10 mK). If the near-surface drilling disturbance was in fact still this large during May 1992, it will take at least 7 additional years for the near-surface disturbance to dissipate to a level commensurate with the precision of the logging system, i.e. ~1 mK.

A new approach to glaciochemical time series analysis
Meeker, L.D., P. A. Mayewski, and P. Bloomfield
Ice Core Studies of Global Biogeochemical Cycles, Annecy, France, R. Delmas (editor). NATO Advanced Sciences Institutes Series 1, Vol. 30, p. 383-400, 1995

Time series obtained from glacier ice cores present special difficulties to the scientist and time series analyst. These series generally represent multivariate non-stationary and non-linear processes sampled non-uniformly in time. As a consequence, few of the traditional methods of time series analysis are immediately applicable in their usual formulation. Here we discuss the analytical problems presented by glaciochemical time series and review some of the procedures available or under development to explore the paleoclimatological information they contain.

Polar ice stratigraphy from laser-light scattering: Scattering from ice
Ram, M., M. Illing, P. Weber, G. Koenig and M. Kaplan
Geophysical Research Letters, Vol. 22(24), p. 3525-3527, 1995

We demonstrate that laser-light scattering (LLS) off glacial (pre-Holocene) polar ice can be used to measure variations in dust concentration along sections of bubble-free ice core. We have used this technique with considerable success on the GISP2 ice core that was retrieved from central Greenland down to bedrock. The technique is fast, non-destructive and has a very high resolution (1mm or less).

An empirical technique for estimating near-surface air temperatures in central Greenland from SSM/I brightness temperatures
Shuman, C.A., R.B. Alley, S. Anandakrishnan and C.R. Stearns
Remote Sensing of Environment, Vol. 51, p. 245-252, 1995

Near-surface air temperatures in central Greenland can be estimated from satellite passive microwave brightness temperatures supported by limited air-temperature data from automatic weather stations. In this region, brightness temperature depends on snow emissivity, which varies slowly over time, and on snow temperature, which varies more rapidly and is controlled by air temperature. The air temperature and brightness temperature data define an emissivity trend which can be modeled as an annual sinusoid. Estimated air temperatures represent an integrated near-surface value that defines the overall temperature trend at the Greenland summit. The modeled emissivity trend allows daily-average air temperatures to be estimated across significant gaps in weather station records, as well as quality control of their temperature data. The technique also generates annual trends of emissivity which can be used to calibrate or test radiative transfer models of microwave emissivity from dry firn.

Temperature and accumulation at the Greenland Summit: Comparison of high-resolution isotope profiles and passive microwave brightness temperature trends
Shuman, C.A., R.B. Alley, S. Anandakrishnan, J.W.C. White, P.M. Grootes, and C.R. Stearns
Journal of Geophysical Research, Vol. 100, No. D5, p. 9165-9177, 1995

Long-term passive microwave brightness temperature trends, supported by short-term automatic weather station (AWS) temperature data, show that the Greenland Summit area experiences secondary warm periods in the late fall and/or winter as well as primary midsummer warmth. High-resolution isotope profiles from snow pits dug in 1989, 1990, and 1991 near the Greenland Ice Sheet Project II (GISP2) site reveal that stable isotope ratios (d18O and dD) preserve this distinctive temperature cycle. This indicates that snow accumulation occurs frequently through the year at the Summit and that the isotope record initially contains temperature information from many times of the year. Comparisons of the records have allowed amounts, rates, and seasonal distribution of accumulation to be estimated as well as the weighting function for isotope thermometry and isotopic diffusion to be evaluated. Through an empirically-derived emissivity model using AWS air temperature data and brightness temperatures, our technique allows isotope values preserved in the snow to be related to estimated near-surface air temperatures. Density- corrected, water-equivalent profiles allow the amounts and timing of accumulation to be determined. Our results indicate that stable isotope ratios from the near-surface snow at the Greenland Summit are a reliable, high-resolution temperature proxy. This gives confidence to the paleoclimatic interpretation of isotope signal variations in the GISP2 core.

Seasonal precipitation timing and ice core records
Steig, E.J., P.M. Grootes and M. Stuiver
Science, Vol. 266, p. 1885-1886, 1995

Gravitational enrichment of 84Kr/36Ar ratios in Polar Ice Caps: A measure of Firn Thickness and Accumulation Temperature
Craig, H. and R.C. Wiens
Science, Vol. 271, p. 1708-1710, 1996

Measurements of 84Kr/36Ar ratios in Greenland ice show that gravitational separation in the firn layer is responsible for the enrichments relative to atmospheric ratios. The 84Kr/36Ar ratio is enriched by 12.8 per mil and is 24 times the 18O/16O enrichment in trapped O2, as predicted for gravitational fractionation. Because gravitational enrichment depends on firn thickness, which in turn depends on annual mean temperature, noble gas ratios provide a method for determining paleotemperatures and ancient firn thicknesses in polar ice caps. The gravitational effects are modulated by about 10 to 15 percent by atmospheric concentration changes caused by temperature effects on oceanic gas solubilities. The availability of five noble gases should make it possible to deconvolute the solubility and gravitational enrichments for calibration of 18O paleotemperatures throughout the polar ice sheets.