Investigation of 𝑀-𝑉 interaction domain for small-scale simplified root system architectures

Assessing the stability of trees against transversal loads requires the definition of a reliable limit condition. This latter is usually estimated for practical application from the interpretation of non-destructive loading tests on trees, along a single loading path. Toppling of root plates is however a complex phenomenon, whose activation depends (i) on the specific loading path (i.e. the combination of vertical, 𝑉𝑉, and horizontal, 𝐻𝐻, forces, and overturning moment, 𝑀𝑀) that can arise e.g. during a windstorm, and (ii) on additional second order geometrical effects, associated to possible large rotations of the tree. In this perspective, the limit condition should then be defined by a complete interaction domain in the generalized loading space, rather than by a single limit value for the toppling action, and the failure criterion must be formulated in a large rotation scheme. To this goal, some original small-scale experimental tests investigating the shape of the interaction domain for simplified root plate geometries in the 𝑀𝑀-𝑉𝑉 plane, also accounting for upward load components, have been conducted. An original load-controlled experimental setup is conceived, in order to allow for the development of large rotations in the model and to capture the onset of the unstable toppling phenomenon. Different simplified root system architectures are considered and critically compared. The results show that the interaction domain presents a noticeable coupling in the 𝑀-𝑉 plane, which can lead to largely inaccurate estimations of on-site toppling resistance if not correctly accounted for.