Ground Movement Prediction Models

“Soil slope stability in distributed hydrologic applications” Sudden and often catastrophic mass wasting events represent a growing threat in today’s shifting hydroclimatic world. Such hazards whether caused by soil slips or debris flows can and unfortunately do result in the costly destruction of properties while also claiming a significant death toll each year. Hydrologists, geomorphologists, and geotechnical engineers depend on a growing array of real-time monitoring and forecast systems typically relying on a variety of estimated rainfall thresholds for predicting landslides. Acknowledging the complex interactions between groundwater, runoff and geotechnical soil properties in determining where a soil slope failure will occur; hillslope hydrology plays a crucial role in the overall process.  To this end, distributed physically based hydrological models, operating either in steady state or in dynamic conditions, coupled with a soil stability model triggered either by shallow subsurface flow or by a wetting front advancement, have been used to map landslide sources in a watershed.   Over time, various methods have been used to model catchment soil stability (e.g., Infinite Slope approach, Transient Rainfall Infiltration Grid-based Regional Slope Stability model or the Shallow Landslides Instability Prediction model, to name but a few) with the Infinite Slope approach continuing in popularity. As the authors note, however, any effective forecast system needs something more precise and a physically based hydrological model coupled with a stability model, working with rainfall maps provided either by spatially distributed weather forecast models or by radar maps could, in principle, be the appropriate solution. Accordingly, to provide a solution to these limitations, a recent study proposed an analytical improvement based on a modification of the well-known 2D Janbu's method as a reasonable compromise between the simplicity of the IS model and the more rigorous complete limit equilibrium methods. The Mettman Ridge study site located about 15 km north of Coos Bay, Oregon consisting of steep, highly dissected soil-mantled hillslopes with narrow ridges and steep channels was selected to test this new method. The figure below illustrates vividly the predictive improvements using this new analytical method.   I enjoyed reviewing the paper by Bonomelli, Pilotti and Piciullo (2025) in Earth Surface Processes and Landforms, “A novel approach for the computation of soil slope stability in distributed hydrologic applications”

✅ Hardening Soil Small Strain Model In geotechnical engineering, accurately modeling soil behavior under various loading conditions is one of the biggest challenges. One model that has significantly improved prediction accuracy is the Hardening Soil Small Strain (HS-Small) Model. It is an extension of the conventional Hardening Soil model, specifically tailored to reflect the real stiffness behavior of soils at very small strain levels. ✅ Historical background The Hardening Soil Small Strain (HS-Small) model was developed by Prof. Ronald B.J. Brinkgreve and his colleagues at Delft University of Technology (TU Delft) in the Netherlands. It was later implemented and widely disseminated through the PLAXIS finite element software, which was also initially developed as a research project at TU Delft. ✅ What is the HS-Small Model? HS-Small is an advanced elasto-plastic constitutive soil model used primarily in: • Seismic response analyses. • Dynamic soil-structure interaction studies. • Cases where small deformation behavior (settlements or vibrations) plays a critical role. The key feature that differentiates this model is its incorporation of very small strain stiffness (G₀). Soils exhibit much higher stiffness at small strains (below 0.001%), and this model captures that behavior accurately, unlike simpler models that assume constant stiffness. ✅ Why Model Small Strain Behavior? In projects involving foundations of sensitive structures, tunnels, or nearby infrastructure, the expected deformations are usually minimal but critical. Standard models often underestimate stiffness at these small strains, leading to overly conservative or inaccurate results. The HS-Small model bridges this gap by including the strain-dependent stiffness behavior of soil—improving predictions of settlements, ground movements, and dynamic responses. ✅ Key Features of the Model • Based on non-associated plasticity theory. • Accounts for stiffness in compression (E₅₀), oedometer loading (Eₒₑₒᵤₙ), and unloading/reloading (Eᵤᵣ). • Introduces G₀ and γ₀.₇ to describe stiffness degradation with strain. • Data input can be derived from lab tests such as bender element tests or resonant column tests. ✅ Practical Applications • Tunnel-induced ground movements. • High-precision foundation settlement analyses. • Dynamic response modeling under seismic loading. #GeotechnicalEngineering #SoilMechanics #HSMS #NumericalModeling #CivilEngineering #Plaxis #FoundationDesign

A case study on the monitoring and analysis of ground settlement caused by tunnelling for the construction of stacked twin tunnels for an underground metro line using a slurry pressure-balanced Tunnel Boring Machine (TBM). 1. TBM Advantages: The use of a closed-type shield TBM offers several advantages in controlling ground settlement during tunnel construction. These advantages include minimizing ground settlement through measures such as face pressure application, radial ground support, and backfill grouting. 2. Three Ground Loss Categories: The study identifies three categories of ground loss that contribute to ground settlement during TBM tunnelling: face loss (due to ground deformation at the tunnel face), shield loss (induced by radial ground contraction around the shield skin plate), and tail loss (occurring in the annular void between the ground and concrete segmental lining). Each of these losses can be estimated based on various parameters, including ground conditions and TBM operation data. 3. Ground Loss Prediction: The gap model is introduced for estimating the total ground loss, which is the sum of face, shield, and tail losses. This model considers factors such as the friction between ground and shield skin plate, face pressure, undrained shear strength, and ground deformation. The face loss, shield loss, and tail loss are calculated separately. 4. Settlement Monitoring: The study involved monitoring ground settlement at different stages of TBM operation, including face excavation, shield passage, and ring build with tail grouting. Settlement profiles are characterized as Gaussian distributions. 5. Volume Loss Estimation: The study used Gaussian curve fitting to estimate total volume loss and trough width factor based on the observed settlement troughs. The results indicated that the estimated total volume loss and trough width factor were consistent with values reported in similar weathered granite ground. 6. Settlement Stages: The analysis revealed that, on average, most ground settlement occurred during the shield passage and ring build period, with minor settlement during face excavation. 7. Adjustment Factors: The gap model was compared with the measured volume loss for the down-track tunnel, and adjusting coefficients for shield and tail losses were determined to improve prediction accuracy. 8. Up-Track Tunnel Analysis: The study applied the adjusted factors for shield and tail losses to estimate ground settlement in the up-track tunnel. The predictions for ground settlement were mostly within the measured settlement range, with medium factors providing reliable estimations on average. 9. The study demonstrates the importance of monitoring and analyzing ground settlement during TBM tunnelling for metro line construction. It emphasizes the value of the gap model for predicting volume loss, and how adjusting coefficients based on field measurements can improve prediction accuracy. #Tunneling #Groundsettlement