A method for prediction of creeping slopes by viscohypoplasticity

Abstract

This study presents a comprehensive framework for predicting and reproducing the rate of slope movements in creeping slow-moving landslides. By employing an advanced viscohypoplastic constitutive model, the framework is capable of accounting for the mechanical behaviour of soil and changes in groundwater levels, and is able to effectively detect spatial and rate variability in landslide processes. The model's predictive performance is first validated through a detailed parametric analysis using finite element models of an idealized slope, incorporating site-relevant material parameters and groundwater level fluctuations. The integration of groundwater level oscillations demonstrated its impact on displacement rates, particularly during extreme events. The framework is then applied, as a proof of concept, to a real-world case of a creeping slope in the Prackovice landslide area, where a 2D finite element model of the area identified critical zones with excessive deformations. Calibration of viscous parameters and OCR was based on back-analysis of inclinometer measurements, while other soil parameters were determined experimentally. This framework offers a valuable tool for land management, infrastructure safety and mitigating the socio-economic impacts of landslides by determining the spatio-temporal evolution of slow-moving landslides.