Geomechanical feature extraction and analysis using LiDAR data: IOC mine

CIM Bulletin, Vol. 4, No. 8, 2009

M. Lato, D.J. Hutchinson and M. Diederichs

The ability to remotely collect valuable information that can be used for geomechanical evaluations is now possible due to the development of rapid, highly accurate scanning devices, which are invaluable for mining operations. This is particularly the case where rockfalls or pit wall failures may occur, endangering personnel who are mapping the face using conventional methods. LiDAR (light detection and ranging), is a range-based imaging technique that measures the time a light source (typically a laser) reflection from an object is received, which is in turn used to determine a referenced location. Static LiDAR data collection equipment can sample up to 500,000 points per second at ranges up to 1,500 m, depending on the equipment employed. The advantages of remotely sensed digital data, in comparison to traditionally collected geomechanical data, include safety, logistics and the ability to re-check analyses at any time. This paper shows the accuracy and validity of remotely sensed digital data, compared with traditional methodologies. To this end, a collaborative practical research project was established between Rio Tinto, the Iron Ore Company of Canada (IOC) and the Department of Geological Engineering at Queen’s University. The goal of the project was to test the limitations of the I-Site 4400 in various working environments, establish workflows and limitations, and perform comparative analyses between traditionally collected data and LiDAR derived data. The project was specific to the IOC mines; thus, only the software and hardware currently in use by the mine was utilized in the geomechanical evaluations. The LiDAR data collected for this project was used for the extraction of planar features, such as joint or fault surfaces visibly present in the pit walls. The features were extracted using the Vulcan geotechnical module through a manual process. Although automated processes do exist, they are currently not in place at the IOC mine, nor is the software available on site. After the collection, processing and extraction of features, the results were compared to traditionally mapped analyses. Due to the magnetic influence of the iron present in the ore, structures were evaluated using an inclinometer and clino rule, with respect to a traverse line. Comparisons of the resultant data from both digital feature extraction using Vulcan and traditionally mapped data were completed using Dips, a graphical and statistical data orientation program.   The comparative analysis was completed by independently analyzing the datasets for integrity and then using common joint-set windows to perform statistical tests on the subsequently segmented data. The orientation of the individual sets, as well as the confidence in the mean and the variability with the dataset, were evaluated. The results demonstrated significant correlation between the methods, as well as limitations for each method. Joint surfaces parallel to the bench face pose the greatest challenge for accurate mapping using traditional methodologies. This challenge is due to the joint surfaces’ distant intersection with the traverse line. Structures perpendicular to the bench face proved to be the most challenging to accurately map using the LiDAR data, due to their minimal representation in the resultant point cloud. The results of this project included specific guidelines and operating workflows to enable the use of LiDAR equipment as a geomechanical mapping tool. Analysis proved the validity of the I-Site 4400 as a geomechanical mapping tool and the procedure has been adopted in the regular work of the IOC pit operations.