Evaluating the accuracy of Airborne Laser Scanning for Mountain Mapping: A Case Study at the Slumgullion Landslide Site in Colorado, U.S.

Airborne laser scanning (ALS) has become one of the most effective and reliable means for large-scale topographic mapping. While a general understanding of the accuracy of ALS surveys is known, too few empirical studies exist for assessing the accuracy of ALS point clouds, particularly in mountain areas. This study used three repeated ALS datasets collected for landslide investigations within a steep and heavily vegetated mountain area to evaluate the accuracy of ALS mapping. The National Center for Airborne Laser Mapping (NCALM) at the University of Houston conducted three repeated airborne LiDAR surveys at the Slumgullion landslide site (6.5 km by 2 km) in Colorado on July 3, 7, and 10, 2015. A terrestrial laser scanning (TLS) survey was also performed during the same period at the middle portion of the landslide area. The TLS survey provided valid ground truth references to evaluate the positional accuracy of ALS data. These repeated airborne ALS datasets also provided valuable datasets for evaluating the repeatability (precision) of ALS surveys. According to this investigation, the bare-earth ALS points used for the case study are able to retain approximately one foot ( 30 cm) accuracy (RMSE: Root-mean-square-error) and one decimeter ( 10 cm) precision in the vertical direction; digital elevation models (0.5-m by 0.5-m DEM) derived from bare-earth ALS points are able to retain half decimeter ( 5 cm) precision and one decimeter ( 10 cm) accuracy. While the vertical precision is not sensitive to topography, the vertical accuracy of ALS points can be considerably affected by the complexity of terrain features and vegetation coverage. The precision and accuracy estimations obtained from the case study could be used as a rule of thumb to estimate the performance ALS mapping in mountain areas with a similar topography and vegetation condition. Conventionally, ALS datasets are assessed according to isolated ground control targets for high-accuracy topographic surveys. This study introduces a method for evaluating and correcting ALS measurements according to millions of TLS points. The elevation accuracy of ALS points can be improved to a half-foot ( 15 cm) level after the correction. The method introduced in this article pave the way for large ALS datasets to be evaluated and corrected rapidly and confidently.