USE OF 3D POINT CLOUD DATA FOR ROCK SLOPE ASSESSMENT

The use of detailed 3D models allows geologist to make better-informed decisions regarding geologic hazards. This talk will cover two case studies of the use of Structure from Motion (SfM) to collect structural information of rock slopes. The New Hampshire Department of Transportation (NHDOT) maintains a Rockfall Hazard Rating System (RHRS) that seeks to quantify the hazard posed by rock cuts over 20ft high adjacent to state roads. The current hazard rating is based on a number of field observations including geologic structure. These structural measurements are limited to the area accessible by foot because of limitations on rope access by NHDOT staff, thereby excluding a number of important structural measurements. In July 2017, NHDOT was able to work with the University of Vermont (UVM) to fly an Unmanned Aerial System (UAS) equipped with a global positioning system (GPS) and camera to acquire a series of images and video of a rock cut in Crawford Notch, New Hampshire. These pictures, in conjunction with ground control points collected by NHDOT using survey grade GPS during the flight, were used with commercial SfM software to create a 3D point cloud model of the rock cut. The resulting point cloud allowed for an exponential increase in the number of structural measurements and greatly improved the stability analysis of the cut, the highest rated in the RHRS. Additionally, the point cloud model allows for measurement of volumes, future change detection, and increased time and detail working on the cut and the identification of potential stabilization methods. Following the success of the UAS flight by UVM, a second cut in Warner, NH was visited and photographed with a GPS enabled smartphone, ground control points collected using a mapping grade GPS and the photographs processed using SfM freeware. This talk will share the process of collecting a 3D point cloud for a near vertical rock face as well as the additional analysis that is capable with such a rich data set. Some pros and cons of each method will be discussed as well as potential future applications of this technology within NHDOT.