OCR Text |
Show Figure 3 gives settling curves for the lower layers of the snowcover. The upper curve, obtained during the winter of 1938- 39, is for a layer 40 cm in thickness with a density of 0.29. At the time the receptacle was set out, the snow had already settled considerably and consisted essentially of ice grains measuring 0.5- 1.5 mm. The middle curve records the settling of a layer 43 cm thick during the winter of 1939- 40. This layer had 20 cm of loose granular snow of density 0.32 immediately below it; over it was a layer of loose, windblown snow, with an average density of 0.13, that had fallen just three days before installation of the device. Mean density of the whole cover was 0.22. The lower curve shows settling of a 57 cm layer with a mean density of 0.18. Under it was a layer of loose granular snow 25 cm in thickness; above it was a layer of fresh snow. This last curve was obtained during the winter of 1940- 41. The zero point represents the day the receptacle was placed on the snowcover, without regard to calendar date. The data show that settling is pronounced the first 2- 3 weeks following a snowfall; although it is still marked at the end of a month, it is no more than a fraction of 1% after the first 24 hours. Until the beginning of thaw, settling is minimal, as little as a small fraction of a millimeter in 24 hours. Once thawing begins, heavy settling sets in. Parallel observations of changes in snow structure made at the same site showed that the heaviest settling takes place with transformation of the snow to a fine granular texture and gradually slows down with further transformation to granular snow. With later transformation to coarse granular snow, settling is minimal and, practically speaking, may be considered nil. Settling, it is thus seen, is largely due to recrystallization. The downward pressure of freshly fallen snow, at least in amounts of 10- 20 kg/ m2, does not appreciably increase settling of the underlying layers, even of layers of loose snow of 0.15 density. The insignificance of the weight of the overlying snow is confirmed by the gradual lessening of settling toward the end of winter, at the same time the overlying snow pressure increases to as much as 150- 200 kg/ m2. Under field conditions, even with weight pressures of 500- 600 kg/ m2, the bottom layer may maintain a density of 0.15- 0.20. Likewise, pressure produced by wind does not increase settling rate. Any results of wind pressure would be evidenced particularly in moist snow. Observations show, however, that loose, moist snow does not increase appreciably in density when subjected to windstorms. For example, on January 6, 1941, a gusty wind of 20 m/ sec velocity lasting 12 hours did not produce any settling as shown by the self- recorder measuring settling in fraction of a millimeter. Here, the receptacle for the self-recorder, covered by 3 cm of moist snow, lay on top of a 90 cm snowcover, the upper 10 cm of which consisted of partly saturated snow not exceeding 0.15 in density. The settling curve in this case is given in Fig. 3B. A strong wind arose on the eighth day of the curve. The amount of settling for this day, 15.3 mm, is nearly twice as much as on the sixth day, 8.4 mm. We can see that settling was not due to wind here, but to thawing to + 4.5° C; for on the seventh day, with thawing to + 1.5° C and a weak to moderate wind, settling was practically the same, 14.7 mm. Since pressure on snow does not produce friability, the theory that snow is compacted by wind action is inconsistent with the presence of loose material on the surface of windblown snow, especially where such windblown material originates from already bound snow. Also, the wind compaction theory does not explain the alternation of loose and dense layers deposited by the same windstorm. - 23- |