Analytical techniques for mapping multi-hazard with geo-environmental modeling approaches and UAV images

Area of Science:

• Geosciences
• Environmental Science
• Spatial Analysis

Background:

• Quantitative spatial analysis is crucial for understanding natural hazards and their interactions.
• Geographic Information Systems (GIS) and R software are increasingly utilized for spatial analysis.
• Multi-hazard assessments are essential for effective environmental management.

Purpose of the Study:

• To compare multi-hazard susceptibility maps generated in 2020 and 2021.
• To identify key morphometric parameters influencing collapsed pipes, gully heads, and landslides.
• To evaluate the effectiveness of various data mining techniques in multi-hazard assessment.

Main Methods:

• Utilized Unmanned Aerial Vehicles (UAVs), GIS tools, and data mining techniques.
• Employed seven classifiers: Boosted Regression Tree (BRT), Flexible Discriminant Analysis (FDA), Multivariate Adaptive Regression Spline (MARS), Mixture Discriminant Analysis (MDA), Random Forest (RF), Generalized Linear Model (GLM), and Support Vector Machine (SVM).
• Applied linear regression to determine influential morphometric parameters.

Main Results:

• The majority of the study area exhibited low susceptibility to collapsed pipes, landslides, and gully heads.
• In 2020 and 2021, 52.22% and 48.18% of the region were not susceptible to any hazards, respectively.
• 6.19% (2020) and 7.39% (2021) of the region faced risks from compound events. Model validation (Area Under the Curve) exceeded 0.70.

Conclusions:

• The study successfully mapped multi-hazard susceptibility and identified key influencing factors.
• Understanding the coexistence and interaction of multiple hazards is vital for sustainable environmental management.
• Future research should focus on determining the combined effects of multi-hazards to support policy development.