A multivariate database for root-induced mechanical reinforcement: status of work and initial insights

This paper provides an initial insight into a quantitative database of root-induced effective cohesion currently under development. The database relies on an extensive review of technical literature focusing on depth-wise trends of cohesion, root distribution, and root density. To potentially allow the implementation of advanced data-driven approaches, the database incorporates, where available, a large set of variables including the number, age, and size of the plant species examined, as well as soil and climate information. The paper describes the current structure of the work-in-progress database and discusses its potential developments and utilization for future data-driven analyses relying on multivariate statistical, probabilistic, and/or machine learning approaches among others. Although at an early stage, the dataset and the insights derived from it serve as a valuable resource for researchers aiming to enhance understanding of the resistance increase in rooted soils.