UPCommons. Portal del coneixement obert de la UPC
Publication
Artículo científico (JournalArticle)
Thermo-hydro-mechanical modeling of root–soil interaction in unsaturated slopes
Vegetation effects in unsaturated soils are often simplified by adjusting strength parameters such as cohesion or friction angle, but this approach overlooks the progressive nature of root–soil interaction. To overcome this, an extended Barcelona Basic Model (BBM-VEG) is developed within the thermo-hydro-mechanical (THM) framework. The model introduces a strain-dependent reinforcement parameter (Rpveg), linked to the root mass fraction and activation strain, which dynamically modifies soil stiffness and strength. By coupling suction-dependent elastoplasticity with Bishop’s effective stress, the yield surface evolves with both suction and root effects, capturing strain hardening and softening in vegetated soils. Unlike earlier approaches, BBM-VEG explicitly represents the mobilization and degradation of root reinforcement with strain, enabling more realistic stress–strain and strength predictions. Validation against triaxial and tensile strength tests on rooted soils confirms its ability to reproduce both initial stiffness and peak strength across varying root densities. Sensitivity analyses showed that while the peak reinforcement controls the maximum strength, the activation strain governs when reinforcement develops, with lower values causing earlier stiffening. Finally, simulations of a representative slope subjected to hydraulic (0.001, 0.002, 0.003, and 0.0005 kg/s infiltration) and thermal (30, 45, 60, and 15 °C) cycles demonstrate that vegetation promotes drier conditions by limiting infiltration and enhancing evapotranspiration. While rainfall triggers displacement, total deformation in vegetated slopes is reduced by 65–70 % compared with bare slopes., The study is funded by the research project called “Slope mass wasting under climate change (SMuCPhy)”, granted by the Ministry of Economy and Competitiveness of Spain (BIA2015-67500-R), EROSLOP project funded by MCIN/AEI/10.13039/501100011033 and E-TRANSITION (TED2021-132559B-I00) funded by the Spanish Research Agency (AEI)., Peer Reviewed