JENNIFER STEWART, DAVID A. BRAATEN, and CAROLE BENNETT, Department of Physics and Astronomy, University of Kansas, Lawrence, Kansas 66045
A snow surface with mobile snow grains is essentially a sediment bed that is influenced by a turbulent air flow. The formation and evolution of snow surface features (e.g., ripples) caused by wind-driven snow grains have been previously examined and classified by Kobayashi and Ishida (1979). The impact of snow grains on the surface by saltation results in the formation of ripples (Kosugi, Nishimura, and Maeno 1992) as well as the liberation of small snow grains, which may be transported by suspension (Anderson and Hallet 1986). The dynamic processes of surface-feature formation display self similarity but are essentially nonlinear and chaotic processes (Tufillaro, Abbott, and Reilly 1992) in which the redistribution of snow grains forming snow-surface features distorts the turbulent flow, which in turn distorts the features.
Although cold-temperature wind-tunnel studies suggest that eolian snow ripples are comparable to corresponding ripples formed in other sediments such as sand (Kosugi et al. 1992), there are some differences in the observed morphology. An important difference between snow grains and sand grains is that snow grains are subject to interparticle cohesive forces and to sublimation during transport unlike sand grains (Schmidt 1986; Pomeroy and Gray 1990). Snow surface features are also hypothesized to play a role in near-surface ice-sheet ventilation processes known as wind pumping (Waddington, Cunningham, and Harder 1996) by the production of high-frequency, micropressure fluctuations caused by the turbulent air passing over the surface features.
A detailed characterization of snow surface features on the Ross Ice Shelf was carried out during the 1996-1997 field season to characterize the morphology of naturally occurring snow ripples, a morphology that could be compared to ripple features in other sediments and would provide a data set against which numerical snow-surface feature simulation models could be validated. Snow-surface features were characterized using a new technique that involved capturing ripple cross-section shapes in digital images in the field for later analysis.
Snow-surface feature measurements were made adjacent to the Willie Field automatic weather station (AWS) (77.85°S 167.08°E) between 3 and 5 December 1996. The field team members responsible for these measurements were J. Stewart and C. Bennett. The features were characterized soon after a precipitation period that was associated with high wind speeds. The snow surface features observed were primarily transverse features (aligned perpendicular to the prevailing winds) such as snow ripples that were produced by winds 24 to 48 hours prior to the field measurements in the range of 8 and 14 meters per second. The basic sampling technique used in this investigation was initially devised by Werner et al. (1986) to characterize sand ripples.
This technique requires an apparatus (figure) consisting of a metal straight edge (on which is mounted a bubble level), a ruler, and a short post on the end of the apparatus. The straight edge is suspended above the snow surface by two stakes, which keep the apparatus from disturbing the surface feature. For transverse features, it is possible to use the shadow cast by the straight edge (which appears as an inverse ripple with the ridge of the surface feature casting the smallest shadow and the trough of the feature casting the largest shadow) to make accurate measurements of the feature. The shadow is cast by sunlight directed by a mirror perpendicular to the snow surface feature and apparatus. The length of the shadow cast by the small post on the straight edge is used to calculate the angle of the incident light (about 45 degrees). The angle is required to correct the surface-feature measurements using simple trigonometric relationships.
Images of the shadows were recorded digitally for later analysis using a Canon RC-250 still videocamera. The table summarizes the digital images obtained and the features characterized adjacent to the Willie Field AWS. Each digital image includes two internal calibrations: a scale, which enables the calibration of image pixels to centimeters, and the shadow length from the small post of known height, which allows the angle of the incident light to be calculated.
Images were later downloaded from the camera floppy disk to a Macintosh computer in the Crary Lab and analyzed using the public domain software package "NIH Image." NIH Image is a public domain image analysis program written by Wayne Rasband at the U.S. National Institutes of Health and is available over the Internet by anonymous file transfer protocol from codon.nih.gov/pub/nih-image . Image analysis techniques and corrections for the angle of the incident light using simple trigonometric relationships allowed surface-feature characteristics such as amplitude, wavelength, and length and angle of upwind and downwind slopes to be quantified with precision.
Two types of indices were determined for the surface features: a symmetry ratio, which is determined by dividing the windward portion of the wavelength by the leeward portion, and ripple ratio, which is determined by dividing the total wavelength by the ripple height. Because the partial upwind side and lee side wavelengths were measured independently of the total wavelength, a comparison of their sum to total wavelength was used to check the effect of image quality and operator skill on the data. Neither factor had a significant effect, suggesting the technique is both reliable and requires little practice. The projection ratio was found to be roughly the same for the two sampling days with mean values of 14.7 and 14.8. The symmetry ratio was found to have a larger difference in the mean values for the two sampling days with values of 0.353 and 0.451. The presence of the heightened ridge of the surface features did not significantly influence either index.
This research was supported by National Science Foundation grant OPP 94-17255
Anderson, R.S., and B. Hallet. 1986. Sediment transport by wind: Toward a general model. Geological Society of America Bulletin , 97, 523-535.
Kobayashi, S., and T. Ishida. 1979. Interaction between wind and snow surface. Boundary-Layer Meteorology , 16, 35-47.
Kosugi, K., K. Nishimura, and N. Maeno. 1992. Snow ripples and their contribution to the mass transport in drifting snow. Boundary-Layer Meteorology , 59, 59-66.
Pomeroy, J.W., and D.M. Gray. 1990. Saltation of snow. Water Resources Research , 26(7), 1583-1594.
Schmidt, R.A. 1986. Transport rate of drifting snow and the mean wind speed profile. Boundary-Layer Meteorology , 34, 213-241.
Tufillaro, N.B., T. Abbott, and J. Reilly. 1992. An experimental approach to nonlinear dynamics and chaos . Redwood City, California: Addison-Wesley.
Waddington, E.D., J. Cunningham, and S.L. Harder. 1996. The effects of snow ventilation on chemical concentrations. In E.W. Wolff and R.C. Bales (Eds.), Processes of chemical exchange between the atmosphere and polar snow (NATO ASI Series I). Berlin: Springer-Verlag.
Werner, B.T., P.K. Haff, R.P. Livi, and R.S. Anderson. 1986. Measurement of eolian sand ripple cross-sectional shapes. Geology , 14, 743-745.