INTRODUCTION

Any structure on the ground which is buried in snow cover sustains force due to snow load on it and snow settlement around it. We call this force settlement force of snow, and structures such as fences and guard-rails are sometimes destroyed by it when buried in snow cover.

In Japan, the settlement force on stakes or beams has been measured by many researchers (eg Amano 1943, Shidei and Takahashi 1949, Shoda 1953, Kobayashi and others 1976) mainly in Niigata prefecture of Hokuriku district, which is a rather warm area among the snowy regions of Japan (Figure 1). But in colder areas, as in Shinjo, few systematic measurements of settlement force have been carried out. Furthermore, settlement force is supposed to depend on the snow properties, and although the Japanese islands are small, snow properties vary from place to place. For these two main reasons, settlement force of snow cover onto horizontal beams has been measured since winter 1974/75 in the test field of the Shinjo Branch of the National Research Center for Disaster Prevention, and since winter 1978/79 at the Hijiori experimental site. The maximum snow depths on the ground at Shinjo and Hijiori experimental sites are, on average, 1.5 m and 4.5 m. respectively.

Fig. 1. Locations of two experimental sites in north Japan.

APPARATUS

Cross-sections of beams used in this research were rectangular, square, equilateral triangular and triangular with a 37_ vertical angle (Figure 2). The height of all beams is O.S m above the ground, and the horizontal width of all beams is 0.1 m. Each beam consists of three horizontal parts, each of length 1.0 m, (2.0 m at Hijiori), and settlement force onto the middle part only was measured by an electric force gauge fixed to a support, as shown in Figure 2. A penetrometer consisting of a steel corn and an aluminum plate was also inserted between the horizontal bar of the beam and another support, in case of failure of the electric gauge.

RESULTS AND DISCUSSION

Four examples of varying settlement forces, obtained from measurement of winter 1980/81 at both experimental sites, are shown in Figure 3. Force increases with increasing snow depth, and begins to decrease a few weeks later after the snow depth reached its maximum (as reported by Sh_da 1953). Sudden sharp increases of force were observed at the end of the winters 1980/81 and 1983/84 at Hijiori only. Reasons for these are not yet clear, but as shown in Figure 3, they appeared in the melting season only. Therefore they might be due to crust formation (see below).

Increment of settlement force onto a beam due to increase of the snow depth is shown in Figure 4, in comparison with the maximum force under maximum snow depth. Osada and others (1975), analyzing settlement force data measured at Niigata prefecture, proposed an equation to estimate maximum settlement force on a rectangular horizontal beam from maximum snow depth only; their relationship between force and maximum snow depth is shown in Figure 4 by a dashed line. The magnitude of maximum settlement force on rectangular and the square-section beams, measured each winter, is in good agreement with values calculated from Osada_s equation when snow lies thinner than about 3 m on the ground.

Settlement forces acting on the two kinds of beams of triangular section were smaller than on the rectangular beam; similar results were previously observed by Suto and Nakamata (1970). From more detailed analysis of our measurements we found that forces on the triangular beams were almost equal to those on rectangular beams when snow depth was lessr than about 1.3 m: in deeper snow forces were smaller on triangular than on rectangular beams.

We found also that the settlement forces increased on a cold night, (Figure 5). Although it was warm in the daytime on 11 March 1983, (maximum air temperature: 7.7 _C), the night became colder and air temperature dropped to -7.6 C in the early morning of 12 March. The snow cover surface turned to a hard crust as air temperature fell, the thickness of the crust reaching 15 cm at 7.35 am. Settlement force increased with the growth of the crust as shown in Figure 5. When the crust was cut off along the beam of triangular section, (Figure 6b), settlement force decreased suddenly by 200-300 N m¹ (Figure 5). So it is considered that the growth of the crust layer on the beam caused the increment of the settlement force.

Fig. 2. Horizontal beams with force measurement devices, and cross-sections of different types of beams

Fig. 3. Daily change of settlment force on two different types of beams in comparison with three parameters, snow depth, daily new snow depth and total weight of the daily new snow depth: in Shinjo (a) and Hijiori (b).

Fig. 4. Increase of settlement force on a beam 0.8 m above the ground, due to increase of snow depth, seen in comparison of maximum force with maximum snow depth.

Fig. 5. Changes in settlement force and air temperature at the Shinjo experimental site. Asterisks indicate the time when the hard crust layer adjacent to a triangular beam was cut, ie at 01:12 am and 08:28 am on 1) March 1983.

Fig. 6. Cross-section of snow cover adjacent to rectangular (a) and triangular (b) beams, and growth of the hard crust layer adjacent to the triangular beam. Downward arrows indicate the growth direction of the crust layer.

A similar process of bridging was derived through computer analysis of snow settlement, ie the increase of the settlement force from snow densification of a snow layer within the snowpack (Lang and others 1984).

CONCLUSIONS

Settlement forces acting upon beams 0.8 m above the ground were almost equal on beams of triangular or rectangular section in snow up to about 1.3 m deep. In snow deeper than this, forces were less on triangular than on rectangular beams. Settlement forces increased without further snowfall when the surface of wet snow cover turned into a crust.

From these observational results we conclude properties of the snow cover are an important factor in estimating settlement forces on beams, and need to be examined accurately and quantitatively.