5:15 PM - 7:15 PM
[HDS07-P04] Preparation of a zoning map considering slope vulnerability during earthquakes
★Invited Papers
Keywords:slope failure, landslide, geologic map, terrain classification map, earthquake, engineering rock mass classification
Landslides are major ground hazard during earthquakes, and identifying vulnerable areas in advance is an urgent issue for disaster prevention and mitigation. It is known that there is a strong correlation between slope gradient and the occurrence of co-seismic landslides, but recent case studies have shown that the distribution of vulnerable geological features and the existence of tephra layers are also very important in predicting where landslides will occur. Slope vulnerability during earthquakes is mainly estimated by modeling the slope gradient, but if a zoning map that integrates both geomorphology and geology is created, it will be possible to construct a model for each zone, which may contribute to appropriate estimation according to the characteristics of earthquake ground motion.
In this study, we compiled topographic and geologic data from all over Japan at a resolution of the Quarter Grid Square (approximately 250 m) to create a zoning map for co-seismic landslides. The method is as follows.
(1) Creation of a map compiling the legend of the Seamless Geological Map V2
With reference to engineering rock mass classification and previous seismic landslides, the legend of Seamless Geological Map V2 (AIST) was grouped into seven types of map data: hard to medium-hard massive rock, soft massive rock, medium-hard layered rock, soft layered rock, man-made land, unconsolidated sediment, and volcanic ash layers.
(2) Creation of a map compiling the legend of the DEM terrain classification map
A terrain classification map was created by typifying similarly shaped slopes using a DEM based on the Digital Elevation Model of Fundamental Geospatial Data (GSI) interpolated to a 30-m grid. For each terrain category, the median values of depth to bedrock inferred from borehole data (NGIC) and slope gradient were obtained, grouped by hierarchical clustering, and map data were created for five categories: mountains, hilly mountains, hilly terrace and debris valley, terrace/alluvial fan/sand bar, and plain (plus high fills obtained separately by manual operation).
(3) Collection of supplementary information
Because the DEM resolution and classification process may have overlooked small cliffs in the plains, polygon data for the special landslide hazard warning areas (DNLI; MLIT) were used to supplement the data. Polygon data of the Soil Map (NARO) were used to supplement areas where the topsoil was relatively new volcanic ejecta and its alteration products.
(4) Construction of zoning maps for all of Japan
The terrain classification and geologic map data were superimposed and compared to the existing history of co-seismic landslides (the Quarter Square Grid Inventory; GSI) to make the final classification, taking into account the amount and type of landslides. Some zones were divided or reclassified with supplemental data. The final zoning map was divided into seven classifications, from Zone 0 to 6.
In this study, we compiled topographic and geologic data from all over Japan at a resolution of the Quarter Grid Square (approximately 250 m) to create a zoning map for co-seismic landslides. The method is as follows.
(1) Creation of a map compiling the legend of the Seamless Geological Map V2
With reference to engineering rock mass classification and previous seismic landslides, the legend of Seamless Geological Map V2 (AIST) was grouped into seven types of map data: hard to medium-hard massive rock, soft massive rock, medium-hard layered rock, soft layered rock, man-made land, unconsolidated sediment, and volcanic ash layers.
(2) Creation of a map compiling the legend of the DEM terrain classification map
A terrain classification map was created by typifying similarly shaped slopes using a DEM based on the Digital Elevation Model of Fundamental Geospatial Data (GSI) interpolated to a 30-m grid. For each terrain category, the median values of depth to bedrock inferred from borehole data (NGIC) and slope gradient were obtained, grouped by hierarchical clustering, and map data were created for five categories: mountains, hilly mountains, hilly terrace and debris valley, terrace/alluvial fan/sand bar, and plain (plus high fills obtained separately by manual operation).
(3) Collection of supplementary information
Because the DEM resolution and classification process may have overlooked small cliffs in the plains, polygon data for the special landslide hazard warning areas (DNLI; MLIT) were used to supplement the data. Polygon data of the Soil Map (NARO) were used to supplement areas where the topsoil was relatively new volcanic ejecta and its alteration products.
(4) Construction of zoning maps for all of Japan
The terrain classification and geologic map data were superimposed and compared to the existing history of co-seismic landslides (the Quarter Square Grid Inventory; GSI) to make the final classification, taking into account the amount and type of landslides. Some zones were divided or reclassified with supplemental data. The final zoning map was divided into seven classifications, from Zone 0 to 6.