• The Journal of Engineering Geology
• /
• v.25 no.1
• /
• pp.93-102
• /
• 2015

The characteristics of frozen soils are one of most important factors for foundation design in cold region. The objective of this study is to evaluate the shear strength and stiffness of frozen soils according to the confining conditions during the freezing and shearing phase. A direct shear box is constructed for the frozen specimens and bender elements are mounted on the wall of the shear box to measure shear wave velocities. Specimens are prepared by mixing sand and silt with a silt fraction of 30% in weight and the degree of saturation of 10%, giving a relative density of 60% for all tests. The temperature of the specimens in the freezer is allowed to fall below -5℃, and then direct shear tests are performed. A series of vertical stresses are applied during the freezing and shearing phase. Shear stress, vertical displacement, and shear wave along the horizontal displacement are measured. Experimental results show that in all the tests, shear strength increases with increasing vertical stress applied during the freezing and shearing phases. The magnitude of the increase in shear strength with increasing vertical stress during shearing under fixed vertical stress in the frozen state is smaller than the magnitude of the increase in vertical stress during freezing and shearing. In addition, the change in shear wave velocities varies with the position of the bender elements. In the case of shear waves passing through the shear plane, the shear wave velocities decrease with increasing horizontal displacement. This study provides an evaluation of the properties of shear strength and stiffness of frozen soils under varied confining condition.

• Korean Journal of Agricultural Science
• /
• v.7 no.2
• /
• pp.131-144
• /
• 1980

The Linear Regression Model to extend the monthly runoff data in the short-recorded river was proposed by the author in 1979. Here in this study generalization precedure is made to apply that model to any given river basin and to any given station. Lengthier monthly runoff data generated by this generalized model would be useful for water resources assessment and waterworks planning. The results are as follows. 1. This Linear Regression Model which is a transformed water-balance equation attempts to represent the physical properties of the parameters and the time and space varient system in catchment response lumpedly, qualitatively and deductively through the regression coefficients as component grey box, whereas deterministic model deals the foregoings distributedly, quantitatively and inductively through all the integrated processes in the catchment response. This Linear Regression Model would be termed "Statistically deterministic model". 2. Linear regression equations are obtained at four hydrostation in Geum-river basin. Significance test of equations is carried out according to the statistical criterion and shows "Highly" It is recognized th at the regression coefficients of each parameter vary regularly with catchment area increase. Those are: The larger the catchment area, the bigger the loss of precipitation due to interception and detention storage in crease. The larger the catchment area, the bigger the release of baseflow due to catchment slope decrease and storage capacity increase. The larger the catchment area, the bigger the loss of evapotranspiration due to more naked coverage and soil properties. These facts coincide well with hydrological commonsenses. 3. Generalized diagram of regression coefficients is made to follow those commonsenses. By this diagram, Linear Regression Model would be set up for a given river basin and for a given station (Fig.10).