Wind pumping in a snow pack related to atmospheric turbulence
Thesis DisciplineChemical Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
An extensive investigation has been carried out to evaluate the possible influence of wind pumping on snow metamorphism, and hence snow stability. Wind pumping refers to a pulsile forced flow of air in a snow pack due to surface pressure fluctuations. An extensive review of atmospheric turbulence from related disciplines was carried out in order to identify the characteristics of turbulence in mountain terrain. This was followed by an experimental programme to characterise and measure the pressure fluctuations on the snow surface in the mountain terrain of Arthur's Pass National Park. Frequencies between 0.0005 and 1 Hz were studied and power spectral/correlation function analysis was made. Gust exceedance statistics were also calculated. The turbulence intensities and rms pressure fluctuations were found to be significantly larger than those reported in the literature. The spectral analysis showed similar form to other documented spectra, but was shifted substantially toward lower frequencies. The same low frequency components, and a semi-periodic component was evident in the autocorrelation analysis. It was suggested that these features arise from turbulent structures shedding off large upstream structures and/or gravity wave activity. Following this, laboratory tests on beds packed with granular materials were carried out to study wind pumping airflow dynamics and the influence of wind pumping airflow on diffusive mass transfer. A mathematical basis for the dissipation of the surface pressure fluctuations in a permeable bed has been experimentally verified in the laboratory. Three one-dimensional models proposed were then applied to a wide variety of snow pack conditions. One model (finite-sealed) was found to be widely applicable to wind pumping modelling in the seasonal snow pack. A linear approximation to this model is applicable to most seasonal snow packs. Sublimating naphthalene was used as a tracer in a set of laboratory experiments to measure the enhanced diffusive mass transfer rates in a packed column due to wind pumping. Compared to stagnant air diffusion rate, an enhancement of one to two orders of magnitude was measured in mass loss rates. The mass loss rate enhancement depended strongly on both the period and magnitude of the applied wind pumping. The mass loss per cycle showed a large, linear dependence on cyclic displacement. This showed that the naphthalene mass loss process was 1-D planar convective/diffusion controlled. A slight dependence on peak velocity was also found. This indicated that the naphthalene mass loss process had a small dependence on boundary layer diffusion. Finally, cold laboratory experiments were carried out to directly study the effects of wind pumping in snow in a controlled environment. A number of strength tests were applied. The cone penetrometer is recommended as the most useful strength testing tool in this type of analysis. Pressure fluctuations associated with gale force winds (60 to 90 km/hr) were used to provide strong wind pumping conditions. Two types of experiments were run, both in high density snow (=350 kg/m3): Initially bonded grains had moderate to high temperature gradients imposed on them to evaluate any enhancements in the faceting process. Grains without any initial bonding had low temperature gradients imposed upon them to evaluate any enhancements in the rounding process. It was found that wind pumping enhanced the rate of the process in both cases. There was a significant wind pumping enhancement of strength loss at high temperature gradients. The influence of wind pumping on the growth rates of depth hoar appears to be almost as significant as the influence of the temperature gradient magnitude. This wind pumping enhancement is most likely to occur when there is a large temperature difference near the surface of the snow pack between the air and snow or between two layers of snow. There was a significant wind pumping enhancement of strength increase at low temperature gradients. The effect of wind pumping on the development of snow strength appears to be more important than the effect of elapsed time. The differences in enhancement with and without wind pumping appear to be larger at the beginning of the bonding process. It is conceivable that the wind pumping could play a major role in the initial bonding. It is therefore concluded that wind pumping is possibly a major contributing mechanism in the development of wind slabs. Hence it can be concluded that wind pumping significantly enhances dry snow metamorphism.