Rock-slope Activity Index (RAI): a lidar-derived process-based rock-slope assessment system

Abstract

Rock-slope Activity Index (RAI): a lidar-derived process-based rock-slope assessment system Lisa Ann Dunham A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Engineering University of Washington 2015 Committee: Dr. Joseph Wartman Dr. Pedro Arduino Program authorized to offer degree: Civil and Environmental Engineering ©Copyright 2015 Lisa Ann Dunham University of Washington Abstract Rockslope Activity Index (RAI) Lisa Ann Dunham Chair of the Supervisory Committee: Dr. Joseph Wartman H.R. Berg Associate Professor Civil And Enviromental Engineering Inventory of unstable highway slopes is an immense challenge for Departments of Transportation (DOTs) due to the geographic dispersion of problematic slopes as well as the variable nature and speed of erosional processes. Due to advancements in lidar technology, acquisition of high resolution spatial data to map and monitor these slopes is becoming simpler, less expensive, and more widely available. Further, the collected data can be used for wide array of applications in addition to the slope inventory, enabling new discoveries for a variety of applications. However, several challenges remain in using lidar for slope assessment. One key problem is the amount of data collected requires significant data processing, a steep learning curve, and can be labor and computationally intensive. To reduce this bottleneck an automated classification system for characterizing rock slopes and calculating their likelihood of failure from lidar data has been developed. This algorithm quickly extracts morphological indices and evaluates them to determine the likelihood of failure throughout the entire face of each rock outcrop. To test this algorithm, a series of terrestrial lidar scans have been completed for several road cuts located adjacent to the Glenn and Parks Highways in Alaska over a three year period. Areas screened as highly unstable are being compared to erosion estimates obtained from the time series lidar data for validation. DOTs can then use this method directly with traffic information for risk assessment, improving safety and enabling them to efficiently determine how to allocate limited resources for road and slope improvements.