Can. Geotech. J. 35: 579–590 (1998) © 1998 NRC Canada 579 In situ shear tests of soil blocks with roots Tien H. Wu and Alex Watson Abstract: In situ shear tests were performed on soil blocks that contained roots to study the contribution of roots to the shear strength in a case where the shear deformation is not constrained to a thin zone. The shearing resistance of the soil–root system, the tensile force in selected roots, and the deformation of the soil block were measured. The roots were exposed after the test and their positions were determined and used to estimate the initial positions. The root force and the shearing resistance of the soil–root system were estimated with known solutions and compared with measured root force and shearing resistance. None of the roots that passed through the shear zone failed in tension at the maximum displacement. As a consequence, the root resistance is much less than that found in a case where the failure surface is restricted to the boundary between a weak soil and a firm base and where roots are anchored in the firm base and fail in tension. Simplified procedures for estimating root forces are suggested for the case of a thick shear zone. Key words: in situ test, roots, shear strength, slope stability, soil reinforcement, soil–root interaction. Résumé : Des essais de cisaillement in situ ont été réalisés sur des blocs de sol contenant des racines pour étudier la contribution des racines à la résistance au cisaillement dans un cas où la déformation en cisaillement n’est pas confinée dans une zone mince. La résistance au cisaillement du système sol–racine, la force de traction dans des racines sélectionnées, et la déformation du bloc de sol ont été mesurées. Les racines ont été mises à découvert après l’essai, et leurs positions ont été déterminées et utilisées pour estimer leurs positions initiales. La force des racines et la résistance au cisaillement du système sol–racine ont été estimées au moyen desolutions connues et comparées avec la force des racines et la résistance au cisaillement mesurées. Aucune des racines passant à travers la zone de cisaillement n’a cassé en traction au déplacement maximum. En conséquence, la résistance due aux racines est beaucoup moindre que celle trouvée dans le cas où la surface de rupture est localisée à la frontière entre un sol mou et une fondation ferme et où les racines sont ancrées dans la fondation ferme et cassent en traction. Des procédures simplifiées pour estimer les forces des racines sont suggérées pour le cas d’une zone épaisse de cisaillement. Mots clés : essais in situ, racines, résistance au cisaillement, stabilité des talus, renforcement des sols, interaction sol–racine. [Traduit par la Rédaction] Introduction The role played by vegetation in improving slope stability is well recognized, and comprehensive reviews may be found in several publications (e.g., Greenway 1987; Coppin and Rich- ards 1990; Morgan and Rickson 1995; Gray and Sotir 1996). Many studies of roots as a soil reinforcement have been made. These include laboratory shear tests of soils with roots (Wal- dron 1977) or soils reinforced by fibers that simulate roots (Gray and Ohashi 1983; Jewell and Wroth 1987; Wu et al. 1988b; Shewbridge and Sitar 1989, 1996), in situ shear tests on soil blocks with roots (Endo and Tsuruta 1969; OLoughlin 1974; Ziemer 1981; Abe and Iwamoto 1988; Wu et al. 1988a; Nilaweera 1994), and evaluation of root forces from slope fail- ures (OLoughlin 1974; Wu et al. 1979; Riestenberg and Sovonick-Dunford 1983; Riestenberg 1994). These studies produced data that show increases in shear strength due to soil–root interaction. Analytical models of soil–root interaction have been derived and used to analyze results of laboratory shear tests (Waldron 1977; Waldron and Dakessian 1981; Gray and Ohashi 1983; Wu et al. 1988b) and slope failures (Wu et al. 1979; Riestenberg and Sovonick-Dunford 1983; Riestenburg 1994). These all considered a failure mode where shear defor- mation is constrained to a thin zone, as in the case of a direct shear test in the laboratory or a slope failure, in which a rela- tively weak soil layer slips over a firm base. Laboratory tor- sion tests by Shewbridge and Sitar (1996) provided observations of shear deformation in reinforced soil where shear deformation is not constrained to a thin zone. They showed that the thickness of the shear zone increased with the amount of reinforcement. Such conditions would apply to a slope failure in a thick soil layer that is being saturated from the top down. Then the critical slip surface may be at the sat- uration front, which may be above the firm base (Wu 1995). We know of no detailed observations of failure in a root- reinforced soil under similar conditions. To study soil–root interaction where the shear zone is not restricted to a thin zone, several in situ shear tests were per- formed on soil blocks that contained roots. The tests were conducted at a site where the roots grow in a thick layer of fairly uniform soil and shearing was made to take place through this soil. The boundary conditions in in situ shear tests differ from those in laboratory direct shear tests. The common practice is to excavate the soil around a test block. A force is applied to shear the soil at the base of the block. No rigid boundaries are imposed on the soil below the base. In the present case, the base of the block is parallel to the slope and simulates the potential failure surface. The roots that pen- Received October 29, 1997. Accepted April 16, 1998. T.H. Wu. The Ohio State University, 2070 Neil Avenue, Columbus, OH 43210, U.S.A. A. Watson. Landcare Research Ltd., P.O. Box 69, Lincoln, 8152, Canterbury, New Zealand.