Evaluating the Use of LIDAR for Landslide Monitoring on Oklahoma Highway

Landslides cause huge human loss and property damage when they occur near infrastructure such as highways. The current approach for dealing with landslides by the Oklahoma Department of Transportation (ODOT) is primarily reactive because there is no effective monitoring mechanism to assess the risk of landslide properly. When the damage is already done, expensive repairs follows because the repair process is time driven and the use of resources may not be the most cost-effective. Traffic lane closures during the repair increases travel time and road users' cost. This gives an opportunity to look for alternative practices. Several studies have proved that the LIDAR technology can be used to detect the slope changes in mountains, but there is no readily available generalized framework to apply this technology to monitor or assess the risk of landslides. The objectives of this study are 1) to develop a comprehensive workflow to apply this technology, 2) to evaluate registration and vegetation algorithms on the collected data, 3) to assess the displacement change over various seasons, and 4) to assess the impact of vegetation removal and downsampling algorithms on displacement change. For this study, the data was collected from four different sites that include both rock type and soil type slopes on Oklahoma highways, collected in four different seasons (summer, dry, winter and warm seasons) of the year. Then, M3C2 displacement analysis was performed on different seasons' data to identify the displacement change over different seasons. Throughout the entire research process, various technical challenges associated with the application of the LIDAR technology were reported along with recommendations to overcome these challenges. Through M3C2 analysis, it was observed that the largest change was observed during June and September. By considering the current level of registration, no significant change was observed in the majority of the areas. It was also observed from the study that vegetation removal and downsampling have impacts on the result of statistical displacement and significant change analyses. The comprehensive workflow developed in this study can help ODOT to implement the LIDAR technology to monitor and assess the risk of landslides on highways in a cost effective manner.