Comparison of ArcticDEM with airbome laser scanning data

The article is devoted to the assessment of the accuracy of the ArcticDEM digital elevation model in comparison with airborne laser scanning (ALS) data on the Kola Peninsula. ArcticDEM, created on the basis of stereoscopic processing of satellite images, has a high level of detail (resolution up to 2 m), but its accuracy requires verification, especially in conditions of complex relief and vegetation cover. As a standard, the authors used airborne laser scanning data with a density of more than 6 points per m2, covering an area of 260 km2. The main results of the study show the presence of systematic deviations of several decimeters, which may require correction to improve the accuracy of the data. ArcticDEM data have good filtering of vegetation cover, without which the final values may have an order of magnitude greater deviations. Errors increase with increasing relief dissection. ArcticDEM meets the requirements for creating topographic plans at a scale of 1:10 000 with a relief section of 5 m, but is not suitable for a scale of 1:5 000 due to insufficient accuracy. Significant deviations (up to 1 m) are observed on the water surface on water bodies, which is due to the limitations of stereophotogrammetry for homogeneous textures. Thus, ArcticDEM demonstrates high accuracy, including in tundra and forest-tundra conditions, but its use requires taking into account systematic errors and assessing the ruggedness of the relief. The model is suitable for regional studies, but in mountainous areas its accuracy can drop significantly. The research results confirm the errors declared by the manufacturer if their values are interpreted as the value of one standard deviation. The study emphasizes the importance of validating global DEMs with local high-precision data, such as ALS, to ensure the reliability of results in scientific and applied problems.