PHYSICAL PROPERTIES RESEARCH ON THE GISP2 ICE CORE

Cold Regions Research And Engineering Laboratory (CRREL)
A. J. Gow and D.A. Meese

The CRREL team has been involved in the basic physical properties of the core (including detailed density and ultrasonic velocity measurements, thin section studies of crystal size and shape, c-axis orientations, and entrapped air bubble characteristics) as a function of depth in the ice sheet. A major part of effort has also centered around the age dating of the core. This has been accomplished by continuous stratigraphic analysis of the core based on annual layer counting and correlations with other data sets that also provide an annual signal. This project has been carried out with close collaboration with researchers from Penn State University (R.B. Alley) and the U.S. Geological Survey (J.J. Fitzpatrick).

Contents:Recent Accumulation Rates || Depth-Age Scale || Ice Density || Crystal Structure || Ultrasonic Velocities || Below 2811m

Recent Accumulation Rates

On 27 July 1990, 20 stakes were emplaced at intervals of 0.5 km along a line due south of the GISP2 drilling site in order to measure accumulation rates. Measurements over the last several years show remarkably uniform snow deposition of about 70 cm annually equivalent to 24 cm water. This uniformity of accumulation is consistent with the very low scale of surface relief observed around the GISP2 site. A number of snow pits were excavated beside snow stakes to evaluate stable isotope variations and depth hoar occurrence with respect to measured increments of snow accumulation at the stakes. Winter and summer accumulations appear similar in magnitude but the distribution characteristics of the winter snow fall is not yet known. Current measurements taken by the Automatic Weather Stations (AWS) may resolve seasonal accumulation rates. Accumulation measurements continue to be collected by researchers involved in ATM work at the GISP2 site. Back to Contents

Depth-Age Scale

A preliminary depth-age scale to 2811 m for the GISP2 core based on annual layer counting has been completed by D. Meese and the dating committee. Annual counts based on visual stratigraphy was initiated in the field and completed at the National Ice Core Laboratory (NICL) in Denver. This work was facilitated by use of a focused fiber optic light source that illuminated stratigraphic detail permitting essentially continuous layer counting to 2811 m, corresponding to ice at least 85,000 years old. Comparison of visible stratigraphy, electrical conductivity measurements (ecm), laser light scattering of dust and oxygen isotopes (only in the top 300 m) revealed an excellent correlation between these annual signals to a minimum depth of 2500 m. Correlation of the various methods provides the most accurate depth-age scale possible as questionable signals can be compared and the strength of the signals can be weighed. We are now in the process of completing the analysis on the ecm and laser light scattering records on the remainder of the core. These data will be merged to obtain the most continuous and accurate scale possible. As such the GISP2 depth-age scale constitutes the longest and oldest continuously dated stratigraphic sequence ever obtained and has provided the time scale against which the high-resolution GISP2 climatic record is being evaluated.

In collaboration with Ed Waddington, John Bolzan and Nadine Cutler the annual layer thickness data is being converted into accumulation. This will provide information as to how accumulation rates at Summit varied with time and with climatic fluctuations. Back to Contents

Ice Density

Ice density was measured every 20 m throughout the depth of the core and have been completed to 3040 m (Fig.1). Ice densities increase with depth in the core and start to level off once the firn-ice transition has been reached. This is attributed to hydrostatic compression of the entrapped air bubbles. However, in the deepest portion of the core the bubbles had begun to get very small and fewer in number an observation we attribute to pressure induced diffusion of gas into the ice, probably via a gas hydrate or clathrate mechanism similar to that originally postulated for the disappearance of air bubbles in cores from below 1100 m at Byrd Station, Antarctica. Temperature-corrected densities of deeper bubble-free ice have attained values close to 0.921 Mg/m3. A significant decrease in density following several months relaxation is clearly evident throughout the core. Ice densities and ultrasonic velocities are being remeasured on a regular basis to determine the amount and rate of relaxation in the core. Back to Contents

Crystal Structure

Analysis of the GISP2 core at Summit has included detailed examination of its crystalline structure, an important property in determining the rheological behavior of the ice and its potential impact on the preservation of paleoclimate records in the deeper parts of the ice sheet. Cursory crystal structure investigations including crystal size measurements and c-axis fabric analysis of horizontal thin sections have been substantially completed to 3040 m. Mean crystal diameter increased approximately linearly from 2-3 mm at the firn-ice transition (75-80 m depth) to 7-8 mm at 1000 m (Figure 2). Below 1000 m the crystal size remained fairly constant until the onset of the Younger Dryas event at 1678 m at which depth the crystals underwent a significant 2-3 fold decrease in size (Figure 2). A progressive clustering of crystallographic c-axes towards the vertical accompanied the crystal size changes, including a small but measurable increase in the degree of clustering of c-axes downward into the Younger Dryas (Figure 3). Increased dust levels that characterized Wisconsin ice has contributed to the maintenance of the generally fine-grained crystal texture which, with its strong vertical c-axis fabric, persisted to 2990 m. Beginning at about 2800 m, in regions of dust-poor ice, layers of coarse-grained ice begin to appear. Below 2990 m the ice becomes entirely coarse-grained, consisting of crystals 50-60 mm in diameters and larger (Figure 2). The growth of very large crystals, attributed to dynamic annealing recrystallization at elevated temperatures in the basal ice was also accompanied by a degrading of the c-axis fabric from a strong single vertical pole pattern to one exhibiting a ring-like or girdle-like distribution of c-axes about the vertical. A transition to much finer-grained ice, 13 m above the bed, occurred coincidentally with the appearance of silt-rich ice and a return to a strong vertical c-axis fabric. Back to Contents

Ultrasonic Velocities

The ultrasonic velocity profile through the ice in both the longitudinal and horizontal directions was measured on 10 cm of ice every 10 m throughout the length of the core. Ultrasonic p-wave velocities measured in the vertical direction of the cores increased progressively with depth to values exceeding 4000m/sec. This, and a progressive increase in the velocity difference between the vertical and horizontal directions of the core both signify the onset of a clustering of c-axes in the vertical, consistent with the optical characteristics of the crystals as observed in thin sections. Back to Contents

Below 2811 m

Below 2811 m visible stratigraphy proved difficult to impossible to decipher. Layering was frequently distorted into s-folds (Fig. 3) and structures resemblent of boudinage and was accompanied by intermittent inclined layering, with inclinations often exceeding 20. At this time it would appear that irregular ice flow below 2811 m, leading to the observed distortion of layers, could have disturbed the environmental record that may have existed in the basal 250 m of ice at GISP2. The extent of this disturbance is currently being evaluated in terms of the distorted layer/crystal structure relationships of the bottom 10-15% of the core. Generally clear glacial ice continued to a depth of 3041 m before transitioning abruptly into brown silty ice (Fig. 4). This silty ice, which contained a number of clear ice bands, continued for an additional 13 m before rock was encountered. The top 10 cm of rock showed signs of weathering (Fig. 5a), the top portion of rock is probably a boulder. At approximately 37 cm, 5 cm of till was encountered (Fig. 5b). This was underlain by bedrock which is composed of gray gneissic granite (Fig. 5c). A total of 1.5 m of rock was collected before drilling was terminated. The contrasted petrographic characteristics of the silty ice and the underlying bedrock poses major questions as to the origin of the silty ice and the mechanism of entrapment of the contained debris. Apart from a few pebbles (up to 2 cm in diameter) and the widespread occurrence of sand-sized particles the silty ice consisted primarily of fine-grained material whose color darkened appreciably as the bed was approached. Despite the dark appearance of the silty ice it's debris content generally represented less than 0.5% by weight of the core. A temperature of about -9C was measured at the ice/bedrock interface which is about 6C below the pressure melting point. The question of just when and where the silty ice was accreted must await more detailed examination (ongoing) of the precise nature of the debris, including its petrographic and textural characteristics, the nature of its disposition in the enclosing ice and the physical properties of the ice itself. Back to Contents


Figure Captions

Figure 1. Density profile from 0-3050 m from the GISP2 ice core. Small scale fluctuations in densities reflect relaxation effects of the core prior to measurement.

Figure 2. Photographs of horizontal thin sections showing crystal texture variations as a function of depth in the GISP2 ice core. Thin sections photographed between crossed polarizers to delineate individual crystal outlines. Color changes reflect changes in the orientations of crystals as they respond to englacial deformation. The enlarged size of crystals at 3001 m is attributed to annealing recrystallization in warmer basal ice. Smallest scale subdivision in all six sections measures 1 mm.

Figure 3. C-axis fabric profile of the GISP2 ice core showing changes with depth. All fabric diagrams (point scatter plots of individual c-axes) based on measurements on horizontal thin sections.

Figure 4. S-folds evident in the stratigraphic layers in the GISP2 core at 2545.77 m. The distance across the core is 12.3 cm.

Figure 5. The silty ice core. Notice the sharp transition between the clean ice and silty ice.

Figure 6. a) Top of rock core. This is probably the weathered surface of a boulder; b) Section of till between the boulder and bedrock; c) gneissic granite bedrock.