Vegetation indices' spatial prediction based novel algorithm for determining tsunami risk areas and risk values.

Area of Science:

• Remote Sensing and Geospatial Analysis
• Natural Hazard Assessment
• Environmental Modeling

Background:

• Tsunami risk assessment requires accurate identification of vulnerable areas.
• Traditional methods may lack the spatial and temporal resolution needed for effective prediction.
• Vegetation indices offer potential indicators for tsunami impact assessment.

Purpose of the Study:

• To propose a new algorithm for detecting tsunami risk areas.
• To develop a prediction model for calculating tsunami risk values.
• To utilize spatial modeling of vegetation indices for enhanced tsunami risk detection.

Main Methods:

• Atmospheric correction using the DOS1 algorithm.
• Classification and prediction using the k-Nearest Neighbors (k-NN) algorithm.
• Spatial modeling incorporating vegetation indices (NDWI, NDVI, SAVI), slope, and distance.

Main Results:

• The proposed model demonstrated superior performance compared to other classification algorithms.
• Achieved minimal Mean Squared Errors (MSEs): 0.0002 for MNDWI and SAVI, 0.0003 for NDWI and NDBI, and 0.0006 for NDVI.
• The prediction model achieved an accuracy rate of approximately 93.62%.

Conclusions:

• The developed algorithm effectively detects tsunami risk areas using vegetation indices and spatial modeling.
• The model's high accuracy indicates its potential for practical application in tsunami hazard management.
• Integration of vegetation indices, slope, and distance provides a robust approach to tsunami risk assessment.