TL;DR
This study investigates how cracks, road loads, and rainfall interact to cause shallow landslides, focusing on the Baiyansizu landslide. Using numerical simulations, the research reveals that cracks significantly amplify rainfall's impact, while road loads accelerate deformation, leading to a complex, non-linear system that requires a multi-faceted approach for effective monitoring and early warning.
What Problem Does This Paper Address?
Shallow landslides are a major geological hazard, particularly in mountainous regions like southwest China, posing significant risks to human lives and infrastructure. While previous research has often focused on single factors (like rainfall) or dual-factor interactions (e.g., rainfall and cracks), there's a critical gap in understanding the combined effects of multiple triggers. Specifically, how do pre-existing cracks, the dynamic loads from roads (like traffic), and rainfall infiltration synergistically influence landslide deformation and stability? This paper aims to bridge this gap by investigating the complex interplay of these three factors, which are often present simultaneously in real-world landslide scenarios.
Methods
The core of this research involves numerical simulations using GeoStudio software. The authors constructed a hydro-mechanical coupling model of the Baiyansizu landslide, incorporating its specific geological layers (sliding mass I, sliding mass II, sliding zone, and bedrock). A key aspect of their methodology was the creation of comparative models: one with simulated fissures (approximately 1m deep) at the trailing edge and another without. They also incorporated periodic dynamic loads to represent road traffic.
The study then simulated four different "working conditions" to isolate and then combine the effects of:
- Condition I: Effective rainfall only.
- Condition II: Effective rainfall plus periodic dynamic load.
- Condition III: Effective rainfall plus crack conditions.
- Condition VI: Effective rainfall, crack conditions, and periodic dynamic load (the most comprehensive scenario).
By analyzing the resulting pore water pressure and displacement fields, and tracking the stability coefficient over time for each condition, the researchers could quantify the synergistic effects of these coupled factors. They also performed a comparative analysis of five different landslides (Baiyansizu, Tanjiawan, Bazimen, Tudiling, and Chengnan) to understand how variations in landslide characteristics (like thickness and crack development) influence their response to these triggers.
Datasets
The study utilized field data from the Baiyansizu landslide, including:
- Ground deformation monitoring data: Collected from automatic GNSS monitoring points (GP01, GP02, GP03) from April 2022 to July 2024.
- Daily rainfall data: Recorded for the same period.
- Geological survey data: Including information on soil properties, stratigraphy, and crack characteristics.
Additionally, the paper references data from other landslides (Tanjiawan, Bazimen, Tudiling, and Chengnan) for comparative analysis, citing previous studies for their monitoring data and characteristics.
Key Insights and Ideas
- Cracks are the dominant amplifier: The research strongly suggests that fissures at the trailing edge are the primary determinant of deformation. They act as conduits, dramatically increasing rainfall infiltration efficiency and leading to localized saturation and instability.
- Rainfall's dual personality: Effective rainfall doesn't just cause landslides; it does so in two distinct ways. Long-term, moderate rainfall leads to gradual, progressive creep by weakening shear strength. Short-term, intense rainstorms cause abrupt, step-like deformation due to sudden pore water pressure surges.
- Road loads accelerate the inevitable: While not the primary trigger, road loading (and associated cyclic vibrations) significantly accelerates landslide deformation. It helps to connect plastic zones within the sliding mass, amplifying displacement.
- Synergy is key: The most severe deformation and instability occur when all three factors (cracks, rainfall, and road load) are present. This multi-factor coupling creates a non-linear system where the whole is greater than the sum of its parts.
- Landslide thickness matters: The study highlights that shallow landslides (like Baiyansizu) are much more sensitive to these coupled factors than thicker, more stable landslides (like Tanjiawan), which can better dissipate stresses.
Key Results
- Maximum Displacement: Under the combined effects of rainfall, fissures, and cyclic dynamic loading, the maximum displacement of the Baiyansizu landslide reached 1.45 meters. Without fissures, heavy rainfall alone caused 0.82m displacement, while adding road and cyclic loading increased it to 1.14m. With fissures, heavy rainfall alone caused 1.425m displacement.
- Stability Coefficient: The stability coefficient dropped significantly below the critical threshold (Fs < 1.0) to 0.995 in the fully coupled scenario (Condition VI), indicating a heightened risk of instability. In contrast, under natural conditions, the coefficient was around 1.1.
- Deformation Patterns: The rear edge of the Baiyansizu landslide showed significantly greater displacement (up to 2677 mm) compared to the front edge (20 mm), confirming spatial concentration of deformation.
- Comparative Analysis: The study demonstrated that variations in landslide thickness, crack development, road loading, and rainfall infiltration characteristics lead to distinct deformation patterns across different landslide types.
Limitations and Future Work
- Model Simplifications: While GeoStudio is a powerful tool, the numerical model is a simplification of complex reality. The paper notes the fissures were modeled at a depth of approximately 1m, which might not capture the full complexity of natural crack networks.
- Generalizability: While five landslides were compared, the findings are primarily derived from the detailed analysis of the Baiyansizu landslide. Further studies on a wider range of landslide types and geological settings would enhance generalizability.
- Future Work: The authors suggest that their findings provide a theoretical foundation for developing hierarchical early warning systems for landslides, specifically tailored to rainfall thresholds and considering the coupled effects of multiple triggers. Future work could focus on refining these models and integrating them into practical monitoring and warning systems.
Summary updated