A new method for measuring deformation and closure of an ice tunnel has been devised. This technique requires the installation of grid lines on a smooth section of the tunnel wall. Although this method necessitates the work of smoothing the wall, it has the significant advantage over other ones of making the movement readily visible.

The technique was used on sections of tunnel wall that had been left relatively smooth by the excavation process. The larger irregularities were removed with an ice axe, and a smooth face was obtained by some slight melting with a blow torch. One to two centimeters deep vertical and horizontal slots were cut with a portable rotary electric saw. The cuts were made at 20 cm. intervals to form a one meter square grid. Pieces of dyed, wet string were forced into the slots and frozen into place by spraying with cold water.

The tunnel where the grids were installed has been described elsewhere.Reference Butkovich ^{1 ,} Reference Landauer ^{2 ,} Reference Rausch ³ It extended 366 m. into a cold ice cap, with its long axis in the direction of the glacier flow, and had several large rooms branching off. The ice temperatures varied between −5° and −10° C. Several grids were installed in 1956 and 13 more in the summer of 1957. The 1957 grids were remeasured and photographed after 10 months, in June 1958. Line separations were measured to the nearest millimeter. Angle measurements were taken on the upper and lower horizontal and left and right vertical lines. Table I gives the pertinent information. The height of the ice above the grids and the surface slope were determined from a level survey. The hydrostatic stress values are assumed equal to the weight of the overlying ice, and the shear stress is taken to be equal to the hydrostatic stress multiplied by the sine of the surface angle.

Table I.

Deformation of 1957 Grids After 10 Months

Figs. 1 a–c (opposite) show photographs of some of the grids. The vertical compression of the grids is obvious. The average reduction varied up to 16 per cent, generally increasing with increasing hydrostatic stress. The horizontal reductions are small, but indicate a tendency towards “extending flow”.

Fig. 1 (a)

Grid No. 7

Fig. 1 (b)

Grid No. 10

Fig. 1 (c)

Grid No. 11 Photographs of some grids showing change from originally square sections

The vertical compression can be explained by tunnel closure. The results (Table I) indicate some dependence of the amount of reduction on stress, but there is a great deal of scatter. The stress situation is evidently complicated.

Shear, as measured with a Brunton compass, was generally 1 to 3 degrees from the vertical in the direction of flow towards the portal. Tilting of the horizontal lines of the grids was less than I degree in most cases, but in one case as much as 4 degrees. However, the initial angular setting of the grids may have been in error by a few degrees. An attempt to relate shear stress to shear strain rate was quantitatively unsuccessful because of scatter in the data.

Two important visual observations were made. The first was that no shearing along the dirt bands was found. Apparently, then, strong shearing is not associated with dirt banding. A second observation was that heavy dirt bands that occurred throughout the tunnel had extruded as much as several centimeters into the tunnel opening. The amount of extrusion appeared to be greater with an increased concentration of dirt in the band. Fig. 1 c is an oblique view of grid 11, which contained a heavy dirt band and shows this effect. A probable explanation is that dirty ice is more plastic than clean ice. This unexpected result needs further study.