Variations in multiscale curvature distribution and signatures of LiDAR DTM errors

Abstract

The development of high resolution LiDAR digital terrain models (DTMs) has enabled the exploration of the statistical signature of morphology on curvature distributions. This work analyzes Minimum Curvature distributions to identify the statistical signature of two types of LiDAR-DTM errors (outliers and striping artifacts) in the derived estimates, rather than morphology itself. The analysis shows the importance of modeling these errors correctly, in relation to the scale of analysis and DTM resolution, in order to have reliable curvature estimates. Nine DTMs of different morphological areas are considered, and grouped into a training dataset (without errors) and a test dataset (with errors). In the training dataset, the original DTMs are considered as true values; errors are then applied to these data. Minimum Curvature is computed at multiple scales from each DTM: changes in curvature distributions due only to morphology and scale are characterized from the original data; error effects are then identified from the datasets with simulated errors, and validated against the test dataset. The analysis shows that outliers and striping artifacts can be realistically simulated by heavily left tailed distributions. For DTMs without errors, the scale-dependent change in curvature distribution is primarily controlled by real morphology. When DTMs include errors, curvature distributions become controlled by these errors, whose propagation depends on error distribution, error spatial correlation, and the scale of analysis. This study shows that the curvature distributions are impacted upon differently by striping artifacts and outliers, and that these are clearly distinguishable from the signal of morphological features: a scale-dependent change in curvature distribution can therefore be interpreted as the signature of these specific errors, rather than morphology

Rainfall simulation and Structure-from-Motion photogrammetry for the analysis of soil water erosion in Mediterranean vineyards

“…From Table 2, emerges that all the DEMs PHO are comparable to DEMs NKN . Mean values are of the order of about 0.0001 m and SDE values of the order of about 0.001 m. Skewness and kurtosis confirm the fact that the elevation differences do not follow normal distributions (Höhle and Höhle, 2009;Sofia et al, 2013), and this supports the choice of considering more robust parameters too such as NMAD and median. However, also when considering these more robust approaches, DEMs PHO confirm to be comparable to DEMs NKN , showing NMAD and median values of the order of about 0.001 and 0.001 m, respectively.…”

Modeling Change of Topographic Spatial Structures with DEM Resolution Using Semi-Variogram Analysis and Filter Bank

“…The degree of smoothing in this case has been expressed in the S 0 term for the model. In other cases, gridded DEM data can also be derived from Radar (SAR) data, Lidar data and digital photogrammetry, the effects of processing and degree of implied smoothing as well as the existence of coherent noise has also been investigated [41][42][43]. (3).…”

A review of marine geomorphometry, the quantitative study of the seafloor

“…4, this has significant implications for marine geomorphometry, which shows as widely recognized in the terrestrial literature (Florinsky, 1998;Zhou and Liu, 2004;Oksa-nen and Sarjakoski, 2005) that errors and artefacts in a DTM propagate to and may be amplified in terrain attributes. As with DEMs (Harrison et al, 2009;Sofia et al, 2013), errors and artefacts in DBMs can be caused by the interpolation method (Erikstad et al, 2013), movement and positioning of the supporting platform (Hughes-Clarke et al, 1996), and a temporal (Lecours and Devillers, 2015) or spatial (Hughes-Clarke, 2003a, b) misalignment between the different elements of the surveying system. Data from radar altimetry are the least sensitive to platform motion .…”