Abstract:

This study aimed toelucidate the reinforcement effect of grass roots on purple soil bunds. Three grasses commonly found growing on purple soil bunds （Echinochloa crusgalli， Digitaria sanguinalis and Eleusine indica） were studied. Soil is a material that cannot withstand tension but can withstand pressure， while the root system within the soil can withstand tension to some extent. When external force is applied， the contact interface between the root system causes roots and soil to dislocate each other and generate frictional resistance and aspects of this root-soil interaction were measured for roots artificially packed into specially constructed test apparatuses. Root systems were harvested from the field and manually packed into the test apparatuses. Parameters measured included： resistance to shear force under vertical load （5 pressures ranging from 25 to 150 kPa）， maximum force （N） to pull packed roots out of the test apparatuses， and tensile strength （MPa） of the roots in the test conditions. Tests were conducted under five soil water contents （5%， 10%， 15%， 20% and 25%） and the root-soil friction characteristics of the three grasses were examined by direct-shear force tests and ‘pull-out’ （extraction） force tests. In addition， the mechanisms of interaction between grass roots and soil during the friction test process were analyzed. It was found that： 1） There was a significant difference between the three grasses in extraction friction force （P<0.01）. The maximum pulling force of E. indica was on average 1.18 times greater than that of D. sanguinalis which in turn was 1.14 times higher than that of E. crusgalli. Similarly， the tensile strength of E. indica was on average 1.30 times higher than that of D. sanguinalis which in turn was 1.10 times higher than that of E. crusgalli. 2） The root-soil friction characteristics of the three tested grasses showed significant differences across the five different soil water contents tested （P<0.05）. At a soil water content of 15%-20%， the cohesion of the tested grass root-soil interface was relatively lower than at the other soil water contents， while the friction coefficient tended to be a maximised. In addition， the maximum extraction force and tensile strength of root-soil interface occurred at a soil water content of 15%. 3） The constitutive relationship between the shear strength at the root-soil interface and vertical load can be described by a hyperbolic model that satisfies the Mohr-Coulomb criterion. The shear strength of the root-soil interface reached a maximum at a soil water content of 20%-25%. 4） Under the same vertical load and soil water content， the shear strength of the goosegrassroot-soil interface was significantly higher than that of D. sanguinalis and E. crusgalli （P<0.05）. It can thus be concluded that among the three tested grasses herbs， E. indica had the most significant effect on enhancing soil shear strength， followed by D. sanguinalis and E. crusgalli. These results provide scientific data to assist with selection of grass species for soil-fixation to protect purple soil bunds.

Key words: purple soil bunds, root-soil interface, herb roots, root-soil friction, soil water content