Grain Coarsening of Water-Saturated Snow

• Published in: Journal of Glaciology
• Main Authors: Raymond, Charles F., Tusima, Katsutosi
• Format: Article in Journal/Newspaper
• Language: English
• Published: Cambridge University Press (CUP) 1979
• Subjects: Earth-Surface Processes; Journal of Glaciology
• Online Access: http://dx.doi.org/10.1017/s0022143000014076; https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0022143000014076

Abstract Experimental measurements were made of changes in grain-size distribution with time in snow saturated with solutions of various impurity contents. Qualitatively, the changes in grain-size distribution occur by shrinkage and eventual disappearance of the relatively small particles and growth of the relatively large particles by a solid mass-exchange process which conserves the total solid mass. The distribution of relative grain size is found to be essentially time independent except for transient effects lasting only several to several tens of hours after the time of initial saturation. Mean grain volume increases at a constant rate, which for solutions of impurity concentration less than about 0.01 mol 1 –1 is (5 to 6) X 10 –3 mm 3 h –1 . In pure solutions the smallest particles shrink at a characteristic rate of about 1 x 10 –2 mm 3 h –1 . Once the steady relative-size distribution is established, the rate of volume change of typical grains varies linearly with grain volume from the characteristic negative rate for the smallest particles through zero for particles of mean volume to positive values for particles of larger volume. The basic features of the changes that take place are explained in terms of heat-flow controlled melting and freezing determined by temperature differences associated with the effect of particle surface curvature on melting temperature. The constant rate of increase of mean grain volume is a consequence of conservation of total ice volume. The expectation that particles of intermediate size would be neither shrinking or growing leads to the conclusion that the actual rate of increase in mean grain volume is about one-half the characteristic melting rate of the smallest particles, which fits the observations. The process of grain growth is slowed by impurities in a way which can be predicted from the melting-temperature depression caused by the impurity and its diffusion coefficient. The transport of heat between grain surfaces is largely through the liquid-filled gaps between ...