Identification and classification method of landslide pattern in the soil water index-based early warning system

Area of Science:

• Geotechnical Engineering
• Earthquake Engineering
• Natural Hazard Assessment

Background:

• Slope failures pose significant risks, necessitating advanced early warning systems (EWS).
• Existing methods often lack precise classification of landslide patterns and scales.
• Meteorological data and soil properties are crucial inputs for slope stability analysis.

Purpose of the Study:

• To develop and validate a Soil Water Index (SWI)-based EWS for identifying and classifying slope failure patterns.
• To evaluate the fuzzy scale of slope failures using meteorological data.
• To analyze the relationship between rainfall intensity, soil type, and landslide behavior.

Main Methods:

• Simulated 8,976 slope stability scenarios using homogeneous slope models (volcanic soil, Toyoura sand) under 22 rainfall conditions.
• Identified 374 slope failures with a factor of safety (FOS) < 1.0.
• Analyzed potential slip surface depths and correlated them with Soil Water Index (SWI) and water storage height (H2).

Main Results:

• The SWI-based EWS successfully identified and classified four distinct landslide patterns (Sliding, Buckling, Toppling, Crumbling).
• Landslide patterns correlated with rainfall intensity, shifting from long-term low-intensity (LL) to short-term high-intensity (SH) rainfall.
• A strong correlation was found between landslide pattern, slip depth, and water storage height (H2) in the second tank layer.

Conclusions:

• The Soil Water Index (SWI) is a viable metric for an early warning system (EWS) to classify slope failure patterns.
• Water storage height (H2) in the second tank layer shows potential for evaluating the scale of slope failures.
• This research provides a foundation for more accurate and nuanced slope failure risk assessment and management.