Duncan-Chang Constitutive Model of Vetiver Root and Soil Matrix at Different Growth Stages

[Objective] To investigate the constitutive relationship governing the entire failure process of root-soil matrixs. [Methods] By employing the Duncan-Chang model and conducting triaxial tests, a stress-strain constitutive relationship model was developed to analyze the entire failure process of the undisturbed root-soil matrixs. Vetiver samples at different growth stages (20, 40, 60 days) were utilized in conjunction with laterite to form root-soil matrixs. Subsequently, the response of model parameters to variations in root biomass was analyzed. [Results] (1) The Duncan-Chang constitutive relationship demonstrated an excellent capability in simulating the mechanical behavior of root-soil matrixs. Despite the inability to capture the strain softening phenomenon of individual soil, the stress-strain curve, predicted shear strength indicators, and measured values exhibit a strong correlation (R2>0.91). They effectively predicted the variation characteristics of the curve in the elastic, hardening and plastic stages. (2) The Duncan-Chang constitutive model parameters of root-soil matrixs exhibited a significant correlation with root characteristic parameters. With the growth of vetiver, the parameters cc、φc、K、Rf and initial shear strength Ei all increased. The variation pattern of parameter n was slightly unique, as it was larger in the bare soil group than in the root-soil group. However, within the root-soil group, n gradually increased with plant growth. (3) The vetiver root system effectively reinforced the soil. Throughout the growth peroid, the longer the growth time, the more developed the root system and the greater the root biomass, the greater improvements in soil shear strength and shear strength indicators (ct、φt) would be. [Conclusion] A constitutive relationship of root-soil matrixs based on the Duncan-Chang model has been established, and this can enhance the understanding of the mechanical behavior of root-soil matrixs in shallow surface landslides, and provide a theoretical basis for effective prevention and control of shallow surface landslide disasters.