Rapid landslide detection from free optical satellite imagery using a robust change detection technique

Area of Science:

• Geosciences
• Remote Sensing
• Geohazards

Background:

• Remote sensing advancements have improved landslide detection capabilities for geo-hazard management.
• Rapid landslide detection is crucial for risk assessment and timely evacuation planning.
• The complexity of landslides necessitates efficient and accurate detection methods.

Purpose of the Study:

• To propose a simple and rapid method for landslide detection using remote sensing data.
• To evaluate the effectiveness of the Tasseled Cap Transformation (TCT) for identifying land-change features associated with landslides.
• To demonstrate the applicability of the method for defining landslide footprints and alarm zones.

Main Methods:

• Utilized the Tasseled Cap Transformation (TCT) on pre- and post-event Sentinel-2 satellite images.
• Created an RGB color composite image by stacking delta TCT Brightness, Greenness, and Wetness indices.
• Applied the method to a landslide event in the Pomarico site, Basilicata, Southern Italy.

Main Results:

• The TCT-based method successfully identified the footprint of a 2019 landslide with high accuracy (ACC = 0.95).
• The RGB composite visually represented landscape changes linked to physical characteristics.
• The procedure effectively detected landslides involving spectral changes without requiring in-situ data.

Conclusions:

• The proposed Tasseled Cap Transformation method offers a rapid, accurate, and cost-effective approach for landslide detection.
• The free availability and frequent revisit times of Sentinel-2 data ensure global applicability of this technique.
• This method supports immediate landslide risk management by enabling rapid delimitation of affected areas.