Study of Ice on the Tom River Before the Ice Drift of 1913

Student E. Bessonov.

Annual studies of ice on the Tom River before the ice drift, organized by B. P. Weinberg, began in 1913 on March 26; it was necessary to confine ourselves to the same methods of investigation as in previous years: measurement of density and testing of the breaking strength of various horizontal layers of the ice cover.

Density and resistance to fracture, like many other physical properties of ice, must depend to a considerable degree on those external conditions that accompany both the very moment of its formation and its subsequent increase in mass: the thickness of the water layer, the speed of its current, the temperature gradient of the stream of cold, the state of the atmosphere, and many others. Differences in these conditions must cause fluctuations, within broad limits, in the numerical characteristics of ice. The latter is intensified still further if the conditions of the investigation themselves were not entirely favorable: if, for example, a piece of ice for twenty-four hours, or even longer, floats in the ice hole after being cut out; if it is transported slowly and is not kept sufficiently well in warm weather. Finally, even the very method of treatment may perhaps have an effect. If, as far as possible, all these conditions of investigation are carefully watched, then the comparability of the measurement results should increase thereby.

The tests were carried out in the physical laboratory of the Technological Institute on six ice specimens brought from the river.

The first was cut out not far from the upper ferry, and the other five opposite Gorokhov’s warehouse, near an island lying in the middle of the river, about 20 meters from its lower (downstream) end. The depth of the water at this place was about 3 m. The layer of snow on top of the ice was not thick:

about 30–40 cm. The specimens were cut out as large parallelepipeds through the entire thickness of the ice cover from a common opening. When the first of them were removed, the water already rose from beneath the cover, and its level was almost the same as the upper surface of the latter, so that the sun and wind could not act directly on the walls of the opening. As for the spring water that had accumulated in rather large puddles on top of the cover, it could have had the strongest effect on the last specimens.

The ice cover was, on average, more than 1 m thick. On top, over 30 cm, there was a layer of firn, completely opaque and dirty-yellow in color, separated by a sharp surface from the rest of the mass of ice. From this boundary, over a distance of 15–30 cm, a layer of opaque white color, owing to the large number of air bubbles, forms a gradual transition from firn to pure ice. The latter, throughout the remaining extent, is completely transparent and bluish in color.

The lower surface, in contact with the water, is sharply bounded by a plane.

Processing the tested specimens on a melting plate revealed that, perpendicular to the plane of freezing, the pieces of ice are penetrated, instead of cleavages of its crystals, by thin interlayers, which are nothing other than an accumulation of the tiniest air bubbles. These surfaces separate vertical prism-shaped bodies from one another, the perimeters of whose cross sections form very intricate figures. At the same time, they showed a certain tendency toward polygons, chiefly with 6 sides, so that a section in the horizontal plane of the whole piece represents a cobweb-like network. In examining several layers of the ice cover, it turned out that the cells of this network differ in size for different layers. Their areas vary within very wide limits. Thus, for example, in specimen No. 1, for layer No. 9 (see the second table, numbering of layers), the areas are 5–15 mm²; for layer No. 13, already 1–3 cm²; and for layer No. 21 they reach as much as 15 cm². This shows that the entire mass of ice consists of long and narrow needles of pyramidal form, directed with their sharp ends upward. How they are arranged relative to one another could not be traced, owing to the haste of the work. In this respect, photographic images of both horizontal and vertical sections of the entire mass would yield much. Moreover, the geometric axes of these needles do not stro-

are perpendicular to the freezing surface, but are situated at a certain angle to it. It would be interesting to trace whether this latter fact is in any way dependent on the flow of the water or on any other conditions, or not.

In view of the fact that such external signs of the condition of the ice may serve as a very substantial supplement to the results of experimental investigations, I give a brief outline of their changes over the entire period immediately preceding the ice drift.

March 27.—Total thickness, without the crust, about 85–90 cm. The upper opaque part is 15 cm. In the transparent part, chiefly in the middle layers between the crystals, mostly along their edges, there are small accumulations of air bubbles, as if strung on threads. In general, a strong, homogeneous mass.

March 30.—Total thickness with the crust, 120 cm. The crust is 30 cm, and the opaque part 20 cm. The remainder is entirely analogous to the preceding.

April 2.—Great brittleness, which is so considerable that the preparation of bars for determining the resistance to fracture proved extremely difficult. As one approaches the lower layers, the surfaces of the crystals are sharply bounded, and their edges have the appearance of fairly thick threads. Along the latter, thawing takes place, as a result of which the bars very easily fall apart into pieces.

April 5.—Entirely analogous to the condition on April 2, except for the somewhat smaller thickness: without crust, 85 cm.

April 8.—In general, the ice has become somewhat thinner: without crust, only 80 cm. The opaque part has increased to 25 cm. The lower half of the layers represents a strongly pitted mass, owing to considerable thawing at the contact surfaces of the crystals.

April 11.—Thickness without crust about 80 cm. The opaque part more than 30 cm. The edges of the crystals are interesting: from strong thawing they have become like channels from 1 mm to 5 mm in cross-section, passing through the entire thickness of the cover, so that the bars turned out to be as if drilled through. The comparative ease of working and the lesser degree of pitting indicate great strength.

It was impossible to obtain further specimens, since the ice on the river became covered with water.

The experimental setup of the investigation remained the same as in previous years; only the number of observations was increased. Their results are expressed in the following two tables.

Table 1.

The density, as is evident, changes very little, with the exception of the very uppermost layers. For layer 1 of the ice sheet of sample No. 2 the density was not determined because of the great difference between it and the density of the liquid during hydrostatic weighing.

The resistance to fracture varies within rather wide limits. For sample No. 1 two series of observations were made at a three-day interval. The resulting difference, especially for the upper layers, indicates the influence of insufficient preservation of the samples. Several new tests for sample No. 2 were made in a plane perpendicular to the plane of the preceding cut. As is evident, there is no special difference. The excessively large, comparatively, values for layers 18 and 20 of sample No. 5 were obtained, evidently, for the following reason. The fracture of bars always occurs along the surfaces of contact of the crystals, whereas here they, with their small length, turned out to be almost of a single crystal, so that the fracture pro-

Table 2.

passed into the stronger mass of the latter. The difference of the last sample, No. 6, from the nearest preceding ones can most likely be explained by fluctuations of the weather affecting them during transport and storage.

If one compares the purely external features cited above of the condition of the ice cover and their change over time with the numerical data obtained experimentally, one may arrive at the following conclusions:

The upper layers, immediately beneath the snow crust, constitute an opaque mass because of the large number of air bubbles. The density and flexural strength of these layers are less than the density and strength of the nearest middle layers—the transparent ones. With time, the thickness of this part gradually increases. Thus on 27.3 it was 15 cm, on 30.3 already 20 cm, and on 11.4 it reached 30 cm. Its density and strength, however, decrease, and the decrease in density passes correspondingly into the lower layers, as is evident from the tables of results (Table 1, layers III and IV, and Table 2, layers 5 and 6).

The lower layers constitute a transparent mass of crystalline structure. Toward the ice drift, gradual destruction sets in within this mass, expressed ultimately in the form of disintegration into crystals—curving, most strongly in the very lowest layers. If, at the same time, no noticeable change in density occurs, since the empty spaces between the crystals are filled with water, then the changes in flexural strength in the direction of decrease are very sharp. (Table 2, all the lower layers and part of the middle layers).

The middle part of the cover, in the region of layers 8, 9, and 10, throughout the entire period of observations before the ice drift, is distinguished by comparative constancy. The flexural strength, having reached a certain maximum, remains unchanged in these layers.

Thus, as the ice drift approaches, the ice cover is destroyed simultaneously from two sides, in the direction toward the inner layers. The latter, forming in some place as it were a neutral region, are distinguished by the greatest strength and invariability of physical properties than all the others.

What the connection is between these facts and the strength of the ice cover as a whole, and with the general condition of the ice drift; what their causal foundations are—only further, broader and more carefully organized investigations can show.

Tomsk.

June 1913.