STUDY ON LANDSLIDE CATEGORY BASE ON TEMPORAL- SPATIAL CHARACTERISTIC DISTRIBUTION IN NORTHERN VIETNAM USING SATELLITE IMAGES

Abstract

North Vietnam has a lot of landslide-prone regions due to heavy rainfall or tropical storms, steep
slopes on mountainous sides and human activities such as road or house constructions. Heavy rainfall events
are estimated by climate change in the future. Therefore, it is necessary to understand landslide characteristic
due to rainfall variation. This paper applied Landsat satellite images and calculated NDIs (Normalized
Difference Indexes) to evaluate the condition of vegetation, soil and water and detected 43 landslide points in
North Vietnam. Landslide points are then evaluated to compare digital geographic information data to divide
landslide type. As result, three typical types of landslides are classified as landslide caused by rainfall on
construction (road etc..) sites (Type 1), landslide caused by rainfall on natural slopes (Type 2), and landslide
caused by rainfall and water drawdown on banks of reservoirs or streams (Type 3). Considering the Landsat
result Type 2 and Type 3 consist larger proportions compared to those in historical data. This means that many
landslide would occur in mountainous areas that far from residential or road areas (Type1).