DEVELOPMENT OF GEOMORPHIC PROTOCOLS FOR LIDAR-ENHANCED RECOGNITION OF LANDSLIDE HAZARDS IN THE BUFFALO NATIONAL RIVER OF ARKANSAS

The Buffalo National River in northern Arkansas traverses three physiographic sub regions: the Salem Plateau, Springfield Plateau, and Boston Mountains. It flows easterly from the Boston Mountains to the White River, carving steep slopes with near-vertical escarpments on the outer margins of bends.

The focus of this study was to prepare 7.5 minute landslide inventory maps along the river, where the bedrock geology was recently mapped by the U. S. Geological and Arkansas Geologic Surveys. The landslide features were tentatively identified by reviewing historic records, topographic expression of slopes underlain by shale, aerial photographs, shaded relief maps, and 1-meter LiDAR digital elevation models (DEM). ESRI’s ArcMap software was employed for analysis and interpretation of all the spatial data and digitization of suspected landside features. Review and reexamination of characteristic landslide features were then undertaken using 3D software such as ArcScence.

Initial mapping of the Ponca Quadrangle suggests that this area spawns various types of deep-seated landslide features, extending into the underlying bedrock formations. These features vary along northern and southerly-facing bluffs of the channel. The use of the high-resolution airborne LiDAR played a key role in identifying prehistoric bedrock landslide features not previously recognized. These features were field checked and investigated to verify spatial dimensions, where possible, and to confirm their existence with some level of confidence. The ultimate goal of this project is to ascertain whether a workable methodology of automated landslide recognition and extraction can be developed using algorithms for topographic and morphometric recognition of various types of landslides