Loading...
Development of semi-automated lidar processing algorithms to correlate climate variables to rockfall patterns for a slope near Glenwood Springs, Colorado
Schovanec, Heather E.
Schovanec, Heather E.
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2020
Date Submitted
Keywords
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Rockfall in mountainous terrain is a pervasive hazard that negatively impacts roads and other infrastructure every year. The Colorado Department of Transportation (CDOT) has implemented monitoring on slopes which they deem pose a risk to travelers and their roads. One such slope is located near Glenwood Springs, Colorado, where rockfalls have been observed to occur with greater frequency in the winter and spring. This made it necessary to understand how climate variables (e.g. precipitation and freeze-thaw cycles) may be influencing rockfall patterns throughout the year. Previous studies have shown the advantages of using remote sensing, in the form of terrestrial laser scanning, for rock slope monitoring. However, high resolution databases with scans collected multiple times per month are rare. The logistics of data collection, the complexity of point cloud processing, and the computation time required, make high resolution data difficult to acquire and difficult to process. As a result, although studies have been performed making basic correlations between rockfall and climate, a thorough statistical analysis was not possible. Using high resolution lidar data acquired biweekly between October 2018 and November 2019 at the Glenwood Springs study site, a comprehensive point cloud processing workflow was developed to create a rockfall database, which could be used to complete a statistical correlation analysis between climate variables and rockfall The development of automation and optimization algorithms, the construction of a point cloud processing workflow, and the use of machine learning reduces the amount of manual work required to complete this study and similar future research. The results of the analysis correlating precipitation and freeze-thaw cycles to rockfall show that it is possible to identify the variables that has primary influence on rockfall over different times of the year and identify relevant time periods over which these variables are acting. Precipitation was identified as the variable influencing rockfall over a between December 2019 and March 2019, while freeze-thaw was the primary influencing factor between March 2019 and late-April 2019. During the summer of 2019, precipitation was the primary factor, although it was found to be influencing rockfall over longer triggering timescales (~30 days).
Associated Publications
Rights
Copyright of the original work is retained by the author.