Search for a command to run...
A series of flowslide impact tests was developed in a beam centrifuge with natural silty-sandy volcanic soils sampled on mountain slopes near Vesuvius volcano in Italy. Following the release of the material, the flow propagates up to the impact loading stage, exhibiting different flow–barrier interaction mechanisms. High-frequency image acquisition and processing are used to quantify the flow-structure interaction, which is then interpreted through an advanced large deformation numerical approach based on the Material Point Method (MPM), capable of including the hydromechanical behavior of non-associative elastoplastic materials such as those composing the flow and the barrier. The MPM model is calibrated on barrier displacement dynamics, which range from negligible values to 0.18 m depending on barrier position and released mass, and validated against flow arrival time, thickness and velocity of flow before the impact and load cell time measurements of base vertical total stress. Back-analysis reveals that saturated hydraulic conductivity ranges from 9.81e-4 m/s to 9.81e-3 m/s across the four selected tests, with lower conductivity values correlating with greater energy transfer to the barrier and a transition in interaction mechanisms from “bore” to “standing jump” and “airborne jet”. The most frequently observed mechanism is the standing jump, characterized by flow run-up, partial overtopping, and material accumulation behind the barrier. As the main novelty, the saturated conductivity of the flow under impact conditions is quantified within a physically consistent range, which is outlined as a key factor in the whole process, including propagation, interaction, and dissipation. These results provide essential insights for engineering practice in debris flow hazard mitigation, particularly for the design and assessment of protective barrier systems. While the centrifuge tests provide controlled conditions enabling detailed observation of flow-barrier interactions, the study considers a single soil type and simplified barrier geometry. Future research should investigate the influence of varying fine content, partial saturation conditions, and flexible barrier designs to extend the applicability of these findings to diverse landslide mitigation scenarios. • Centrifuge tests at 40 g simulate debris flow impact on a free-sliding barrier. • Numerical MPM model calibrated using the barrier's experimental data. • Calibration was performed on soil permeability and Young modulus. • Model validated by comparing numerical and experimental total vertical stresses. • Results compared with existing literature through Faug's diagram.