Root reinforcement effect of different grass species: A comparison between experimental and models results

Abstract

Root systems effects along the slopes are widely investigated with the aim to quantify the contribution to prevent soil erosion. Previous studies and methodologies have been carried out to estimate the additional cohesion provided by roots to the soil. In this research five grass species were tested, two of them ( Festuca pratensis and Lolium perenne) belonging to the Poaceae family, while Trifolium pratense, Lotus corniculatus and Medicago sativa are three species of the Fabaceae family. We realized in situ tests in order to quantify the contribution of the root system to the soil shear strength, and laboratory tests estimated the root tensile strength of the different species. Laboratory data and root area ratio (RAR) values, measured during the field tests, were used to implement the models of Wu et al. [Wu, T.H., Mckinnell, W.P., Swanston, D.N., 1979. Strength of tree roots and landslides on Prince Of Wales Island, Alaska. Can. Geotech. J. 16, 19–33], and of Pollen and Simon [Pollen, N., Simon, A., 2005. Estimating the mechanical effects of riparian vegetation on stream bank stability using a fiber bundle model. Water Resour. Res. 41]. These models are widely used in the literature to simulate the roots behavior in soil during a shallow landslide phenomenon. The in situ measured shear strength values were compared with the models outcomes, verifying if the models describe well the natural behavior of the five tested species. The experimental tests provided a measured value about the effective increase (in percentage) of soil shear strength at 10 cm depth, due to the presence of root systems (i.e. the percentage increase given by T. pratense is 515%). Some discrepancies between the experimental tests and the models results became visible. The applied models do not account for some complex factors involved in the soil reinforcement. The root–soil matrix should thus be characterized by mechanisms not fully captured by the W&W and FBM models.