Abstract:
A hypothesis used to explain the relationship between timber harvesting
and landslides is that tree roots add mechanical support to soil, thus increasing
soil strength. Upon harvest, the tree roots decay which reduces soil strength and
increases the risk of management -induced landslides. The technical literature
does not adequately support this hypothesis. Soil strength values attributed to
root reinforcement that are in the technical literature are such that forested sites
can't fail and all high risk, harvested sites must fail. Both unstable forested sites
and stable harvested sites exist, in abundance, in the real world thus, the
literature does not adequately describe the real world. An analytical model was developed to calculate soil strength increase due
to root reinforcement. Conceptually, the model is composed of a reinforcing
element with high tensile strength, i.e. a conifer root, embedded in a material
with little tensile strength, i.e. a soil. As the soil fails and deforms, the reinforcing
element also deforms and stretches. The lateral deformation of the reinforcing
element is treated analytically as a laterally loaded pile in a flexible foundation
and the axial deformation is treated as an axially loaded pile. The governing
differential equations are solved using finite-difference approximation
techniques. The root reinforcement model was tested by comparing the final shape of
steel and aluminum rods, parachute cord, wooden dowels, and pine roots in
direct shear with predicted shapes from the output of the root reinforcement
model. The comparisons were generally satisfactory, were best for parachute
cord and wooden dowels, and were poorest for steel and aluminum rods. A parameter study was performed on the root reinforcement model which
showed reinforced soil strength increased with increasing root diameter and soil
depth. Output from the root reinforcement model showed a strain
incompatibility between large and small diameter roots. The peak increase in soil
strength attributed to roots was controlled by the small (<4mm) diameter root
fraction. These results were used to calculate the effect of timber harvesting on a
small, approximately 7.6 m^3 (10 yd^3), hypothetical landslide in a shallow,
cohesionless, forest soil. The root reinforcement model predicted a post-harvest
reduction in soil strength of 14 and 19 percent for a soil with and without 5 kPa
(105 lbs/ft^2) of cohesion, respectively.
