1. A Status of Landslide Damage in Mountainous National Park of Korea by Using Temporal Spatial Images
This study analyzed damages from landslides in 14 mountainous national parks with the temporal spatial data between 1986 and 2013. First, landslide disaster areas were selected through visual identification. Then a DB was made by estimating data including time of occurrence, location and scale of landslides. It showed Mt. Seorak National Park recorded the largest damage area from landslide followed by Mt. Jiri and Mt. Naejang National Parks. The year 1995 showed the highest number of cases when landslides occurred in eight National Parks. Also, the largest area damaged by landslides reached 513.7 ha in 2006 while the second largest recorded 90 ha in 1998, which were occurred in Mt. Seorak and in Mt. Jiri respectively. Compared to the result of previous field survey, total damaged area calculated by this analysis amounted to 989 ha. It was about 13 times higher than the result from field survey 73 ha. When it comes to frequency, the result from this analysis 1,600 times outnumbered field survey 142 times by approximately 7 times.
2. GIS-based Analysis on Characteristics of the Influential Factors Causing Landslides in Mountainous National Parks
This study was carried out to analyze characteristics of the influential factors causing landslides in mountainous National Parks (Mt. Seorak and Mt. Jiri) and Non-National park areas (Bonghwa and Geochang). Through analytical methods landslide-damaged spots were extracted from aerial images and spatial DB of influential factors were made causing landslides with using GIS. Eleven factors were analyzed through spatial analysis technique and characteristics of landslide occurrence in target areas were analyzed by using likelihood ratio. Average drainage area in mountainous National Parks was two times greater than that of Non-National park areas. Also, in mountainous National Parks, slope and curvature was complex and concave. Major age class was seventh and the diameter was third or fourth. As a result of analyzing likelihood ratio to compare landslide occurrence rate in mountainous National Parks and Non-National park areas, areas with an drainage area between 2,000 to 10,000㎡ have greater drainage area than general mountainous areas. Also, in mountainous National Parks, slope and curvature showed more concave areas, was compared start contrast to general mountainous areas showing more convex areas. Age class of fourth and seventh, and DBH of four have the highest likelihood ratio in mountainous National Parks, which was different from general mountainous areas where age class of two and three, and diameter of two and three had the highest likelihood ratio.
3. Landslide Hazard Prediction Map Development Based on Logistic Regression Model for Applying in Mt. Seorak and Mt. Jiri
This study was carried out to develop landslide hazard prediction map for Mt. Seorak and Mt. Jiri by using the data of 733 landslides that occurred between 1986 and 2013. These data set was extracted front the temporal spatial data. Logistic regression model was developed by using the data mentioned above. The spatial data of database on landslide of the Mt. Seorak and Mt. Jiri was constructed for this study by using aerial the temporal spatial data because the model based on probability such as logistic regression analysis needs a lot of data concerned on landslide. Seven factors of Mt. Seorak closely related to landslide were selected through logistic regression analysis and the classification accuracy of model was 83%. Applied factors are as follows; slope, drainage length, TWI, aspect, forest stand, DBH class, age class. Four factors of Mt. Jiri closely related to landslide were selected through logistic regression analysis and the classification accuracy of model was 78%. Applied factors are as follows; slope, drainage length, TWI, DBH class. Using the developed model and GIS analysis, the new landslide hazard prediction map of Mt. Seorak and Mt. Jiri was made.