11.3 Landslide and slope stability analysis

Slope Stability Factors and Landslide Triggers

Landslides pose significant risks to lives and infrastructure. Slope stability depends on the balance between forces trying to move material downhill and forces resisting that movement. Understanding what controls that balance, and what disrupts it, is the foundation of landslide hazard assessment.

Geophysical methods and remote sensing give us tools to peer beneath the surface and monitor slopes over time. Seismic refraction, electrical resistivity surveys, InSAR, and GIS-based modeling all contribute to identifying vulnerable areas before failure occurs.

Geological and Geomorphological Factors

Slope stability is governed by the interplay of geological, geomorphological, and hydrological conditions. The key material property is shear strength, which depends on two components: cohesion (the "glue" holding particles together) and internal friction angle (resistance to sliding between grains). A slope remains stable as long as shear strength exceeds the shear stress imposed by gravity.

Several geological factors increase landslide susceptibility:

• Weak layers such as clay seams, weathered horizons, or bedding planes oriented parallel to the slope can act as potential slip surfaces
• Rock and soil properties like grain size, mineralogy, and degree of weathering control how strong or weak the slope material is
• Slope geometry matters directly: steeper angles, greater heights, and concave profiles all increase the ratio of driving forces to resisting forces
• Groundwater conditions affect pore water pressure, which reduces effective stress and therefore shear strength

Triggering Mechanisms and External Forces

A slope can sit near the edge of stability for years until a trigger pushes it past the threshold. Common triggers include:

• Intense or prolonged rainfall, which raises pore water pressure and saturates slope materials
• Earthquakes, where seismic ground motion adds transient shear stress and can liquefy saturated soils
• Volcanic activity, including eruptions, lahars, and hydrothermal alteration of rock
• Human-induced changes such as excavation at the toe of a slope, loading at the crest, or altered drainage patterns
• Reservoir filling, which raises the water table and increases pore pressure in adjacent slopes

The factor of safety (FoS) quantifies how close a slope is to failure:

$FoS = \frac{\text{Resisting forces (shear strength)}}{\text{Driving forces (shear stress)}}$

• $FoS > 1$: the slope is stable (resisting forces exceed driving forces)
• $FoS = 1$: the slope is at the point of failure
• $FoS < 1$: the slope is unstable and failure is expected

In engineering practice, a FoS of 1.5 or higher is generally required for long-term slope stability to account for uncertainties in material properties and loading conditions.

Geophysical Methods for Landslide Investigation

Seismic Refraction Surveys

Seismic refraction maps the subsurface velocity structure by measuring how fast seismic waves travel through different layers. Because landslide debris is typically looser and more fractured than intact bedrock, there's a velocity contrast that helps delineate the geometry and thickness of the slide mass.

How the method works:

1. An array of geophones is laid out along a line on the slope surface
2. A seismic source (sledgehammer strike or small explosive charge) generates waves at one end
3. The geophones record the arrival times of waves refracted along layer boundaries
4. Travel-time data are inverted to produce a velocity model of the subsurface

What the results reveal:

• Bedrock depth and the thickness of overlying unconsolidated material
• Low-velocity zones that may indicate higher fracture density, water saturation, or weak materials prone to failure
• Groundwater table position, since saturated materials have higher P-wave velocities (close to 1500 m/s for water-saturated sediments)

Electrical Resistivity Surveys

Electrical resistivity imaging maps how easily current flows through the subsurface. Resistivity is sensitive to lithology, water content, and clay content, making it well suited for identifying conditions associated with landslide-prone zones.

Data collection follows a similar surface-array approach:

1. An array of electrodes is placed along a profile on the slope
2. Current is injected through selected electrode pairs while voltage is measured at others
3. Measurements are repeated for many electrode combinations to build up spatial coverage
4. The apparent resistivity data are inverted to produce a 2D or 3D resistivity model

Key interpretive guidelines:

• Low resistivity zones often correspond to water-saturated or clay-rich materials, both of which reduce shear strength
• Time-lapse monitoring (repeating surveys over weeks or months) can track changes in water content or detect the progressive development of slip surfaces
• Sharp resistivity contrasts may mark the boundary between slide material and intact bedrock

Integrating seismic refraction with electrical resistivity provides complementary information: seismic data constrain mechanical properties and layer geometry, while resistivity data highlight fluid content and clay distribution. Both datasets should be interpreted alongside geological mapping and geotechnical borehole data for reliable landslide characterization.

Remote Sensing and GIS in Landslide Analysis

Remote Sensing Techniques for Landslide Mapping

Remote sensing provides high-resolution spatial data for identifying, mapping, and monitoring landslides, especially over large or inaccessible terrain.

• Stereoscopic aerial photography allows 3D visualization of the ground surface, making it possible to identify landslide features like head scarps, tension cracks, and displaced masses
• High-resolution satellite imagery (e.g., WorldView, Pléiades) can detect surface deformation, changes in vegetation cover, and fresh exposure of bare soil associated with recent slides
• LiDAR (Light Detection and Ranging) generates detailed digital elevation models (DEMs) with sub-meter resolution. Because LiDAR can penetrate vegetation canopy, it reveals subtle topographic signatures of old or slow-moving landslides that are invisible in optical imagery
• InSAR (Interferometric Synthetic Aperture Radar) measures ground deformation by comparing the phase of radar signals from repeat satellite passes. It can detect displacement rates as small as a few millimeters per year, making it a powerful tool for monitoring creeping slopes before catastrophic failure

GIS-Based Landslide Susceptibility Assessment

Geographic Information Systems (GIS) serve as the platform for combining all of these data sources into a spatial analysis framework.

The typical workflow for susceptibility mapping:

1. Build a landslide inventory by compiling historical records, field observations, aerial photo interpretation, and satellite-derived detections
2. Assemble thematic layers representing conditioning factors: slope angle, aspect, lithology, land use/land cover, distance to faults, drainage density, and soil type
3. Apply a statistical or machine learning model (logistic regression, random forest, or neural networks) to relate the spatial distribution of past landslides to the conditioning factors
4. Generate a susceptibility map that classifies the landscape into zones of low, moderate, high, and very high landslide probability
5. Validate the model using a portion of the landslide inventory withheld from training

GIS-based approaches combined with remote sensing enable rapid, cost-effective hazard assessment over regional scales, which is particularly valuable in developing countries or mountainous regions where field access is limited.

Case Studies of Significant Landslides

2014 Oso Landslide, Washington, USA

The Oso landslide on March 22, 2014, killed 43 people and destroyed dozens of homes in Snohomish County. The slide occurred in a region with a well-documented history of slope failures, but the scale and mobility of this event were far greater than anticipated.

• The slope consisted of weak, glacially-derived sediments (glaciolacustrine clays and outwash sands) deposited during Pleistocene glaciation
• Weeks of heavy rainfall saturated the slope, raising pore water pressures and reducing effective stress along a pre-existing weak layer
• Post-failure geophysical investigations using seismic refraction and electrical resistivity identified a low-velocity, high-conductivity layer at the base of the slide mass, consistent with a saturated clay-rich slip surface
• The slide mass traveled over 1 km across the valley floor, a runout distance that highlighted the danger of flow-like behavior in saturated, fine-grained debris

2017 Xinmo Landslide, Sichuan, China

On June 24, 2017, a massive rock avalanche buried Xinmo village in Maoxian County, killing over 100 people. This event demonstrated how seismic preconditioning and topographic effects combine to produce catastrophic failures.

• The slope had been weakened by the 2008 Wenchuan earthquake ($M_w$ 7.9) and subsequent aftershocks, which fractured the rock mass extensively
• A smaller, more recent earthquake triggered the final collapse of the heavily fractured ridge
• Seismic monitoring data suggested that topographic amplification of seismic waves along the narrow ridge concentrated shaking at the crest, accelerating the failure
• The resulting rock avalanche had a volume of approximately 13 million cubic meters and buried the village under tens of meters of debris

1963 Vajont Landslide, Italy

The Vajont disaster on October 9, 1963, is one of the most studied landslides in geotechnical history. Approximately 270 million cubic meters of rock slid into the Vajont reservoir, generating a wave that overtopped the dam by over 200 meters and killed nearly 2,000 people downstream.

• The slope above the reservoir contained a pre-existing slow-moving landslide along a clay-rich layer that acted as the basal sliding surface
• As the reservoir was filled, rising water levels increased pore water pressure along this clay layer, reducing effective stress and progressively destabilizing the slope
• Monitoring instruments, including inclinometers in boreholes, recorded accelerating creep in the weeks before failure, but warnings were not acted upon in time
• The dam itself survived intact, but the displaced water devastated the town of Longarone in the valley below