Effect of root orientation on the strength characteristics of loess in drained and undrained triaxial tests

Ecological engineering measures are gaining popularity in the control of mountain hazards, particularly in shallow landslides, in which plant roots improve the shear strength of shallow soil. To measure such improvement, a series of drained and undrained triaxial tests were performed to investigate the role of Cunninghamia R. Br roots in loess soil mechanical behavior collected from Longchi Forest Center, southwest China. The specimens were prepared by distributing roots in four (4) distinct spatial orientations types, i.e., Horizontal, Vertical, Horizontal-Vertical (grid), and Randomly mixed in the soil. All specimens were first fully saturated and subsequently sheared in compression tests under drained and undrained conditions. The test results are evaluated in terms of monotonic stress-strain behavior, volume change, development of excess pore water pressure, and critical state behavior. The results showed that the addition of roots increased the soil shear strength, provides additional resistance to volumetric deformations and increased the positive excess pore water pressure as compared to plain loess soil. The results also revealed that the mechanical response of the soil-root composite depends on the type of root orientation. The shear strength improved highest with horizontal-vertical (grid) root orientation in both drained and undrained conditions among four root orientations. In contrast, the horizontal root orientation showed the least shear strength improvement. Additionally, for the same root orientation type, the shear strength improvement was more pronounced in undrained conditions compared to drained conditions. The plain and soil-root composite shear strength parameters showed that the addition of roots increased the cohesion with little effect on the mobilized angle of friction in both drained and undrained conditions. The findings of this research will provide a strong baseline for developing a predictive constitutive model for soil–root composite.