On-site strategy for estimating range variances of terrestrial laser scanners based on (un-)scaled intensity values

Abstract Recent advancements in technology have established terrestrial laser scanners (TLS) as a powerful instrument in geodetic deformation analysis. As TLS becomes increasingly integrated into this field, it is essential to develop a comprehensive stochastic model that accurately captures the measurement uncertainties. A key component of this model is the construction of a complete and valid variance-covariance matrix (VCM) for TLS polar measurements, which requires the estimation of variances for range, vertical, and horizontal angles, as well as their correlations. While the need for correct quantification of correlations in the VCM might be case dependent, realistically quantifying the variances in the VCM is crucial for most applications demanding high accuracy. Given that the angular variances can be obtained from manufacturer specifications, the range variance varies with different intensity measurements. As a primary contribution, this study presents an effective methodology for measuring and estimating TLS range variances using both raw and scaled intensity values. A two-dimensional scanning approach is applied to both controlled targets and arbitrary objects using TLS instruments that provide raw intensity values (e.g., Z + F Imager 5016A) and those that output scaled intensities (e.g., Leica ScanStation P50). The methodology is further evaluated using field observations on a water dam surface. Overall, this work introduces a comprehensive workflow for modeling range uncertainties in high-end TLS systems.