Earthflows: Measurement and Explanation: an Investigation into the Problems of Instrumentation and Measurement of Earthflow Properties and Their Application to the Explanation of mass movement

Abstract

A seasonally creeping earthflow about 200m long, 100m wide, and with a maximum depth of 11.75m has developed in a moderately sloping (21-35°), pasture-covered hillslope formed of Pliocene mudstone in east Wairarapa, New Zealand. Analysis of ground survey data and sequential aerial photograph series enabled assessment of the internal morphological components and their changes over a 40-year period. Movement vectors and rates over a 2-year period were determined at the surface by continuous recording and at depth by inclinometer readings. The flow was static for 5 months of the year but during the period of maximum movement attained surface rates of 1.1-4.4mlyr. Movement above the shear plane (which ranged in depth from 5.25 to 11.75m) varied within the feature from a form of 'true mudslide' below the crown scarp to a more fluid 'mudflow' at the toe. Soil water behaviour and local climatic factors were monitored continuously with specially developed instrumentation to provide a detailed record of temporal changes, while spatial variations were established from a less frequently monitored instrument network. This information was used to explain the relationships between various parameters, such as pore water pressure and precipitation, and to relate movement to soil hydrology and climate. Soil properties and stress-strain behaviour of the earthflow material were determined by a range of laboratory tests which quantified the moisture content-shear strength relationships for various failure conditions and established the factors determining the nature of these relationships. The rheological attributes of the material were also determined through a series of laboratory tests. Two distinct patterns of stress-strain behaviour emerged which appear to have played a critical role in the evolution of instability and the development and subsequent behaviour of the earthflow. In a densely packed, 'intact’ condition and under high normal stresses the material behaves as an elastic solid with a shearing resistance determined by the frictional characteristics of the particles. In contrast, when the particles are less tightly packed and under low normal stress, the behaviour of the material is controlled by its cohesive attributes and it can deform as a 'plastic' indefinitely under low shear stresses. The stress-strain behaviour of the material and the prevailing soil water and climatic conditions were used to determine the most appropriate form of stability analysis. Failure of the intact mudstone can be accurately modelled by limit equilibrium analysis using a discrete shear plane, and 'realistic' groundwater levels. After failure, particle arrangement is modified in such a way that individual grains can move under low stress levels within the relatively thick films of water bonded to the clays. Existing conditions within the earthflow suggest that this material will continue to undergo downslope deformation unless restrained in some manner. The compressive strength of the material at the toe varies greatly with seasonal moisture levels. Under summer conditions it provides sufficient resistance against the prevailing low shear stresses to restrain downslope movement. Movement during winter occurs in response to the decreased compressive strength of the material (resulting from increased soil moisture levels) and the higher levels of hydrostatic head at the toe, which are accentuated by the slope morphology. Progressive removal of material from the toe by active stream erosion reduces the lateral support of the slope. Increased stress levels and decreased strength lead to successive failure of intact material upslope. In this manner slope instability has been maintained and the earthflow has migrated retrogressively upslope in stages.