Physically-based digital geomorphological mapping: Case study of glacial and karst topography

Abstract

Digital geomorphological mapping is considered here as a semi-automated procedure of division the land surface into genetically meaningful objects. The basis is automatic segmentation into elementary forms, the smallest and indivisible elements of the land surface, should be geometrically clearly identifiable, with maximal internal homogeneity and clear discontinuities at their boundaries. These elements can then be combined into more complex genetically homogeneous composite forms. Given the crucial role of gravitational energy in landform formation, it should also be considered in elementary land surface segmentation . This research builds on the theoretical foundation of physical geomorphometry, which explores the relationship between gravitational energy and geomorphometric variables. Specifically, we apply the recently published algorithm for physically-based elementary land surface segmentation by Minár et al. (2024), which utilizes dynamic least squares (DLS) generalization within a GEOBIA framework. The algorithm was initially tested in structurally fluvial hilly terrain using nine physically interpretable gravity-specific point-based variables (elevation, slope aspect and gradient, three basic curvatures, and three changes in curvature). In this study, we extend the application of this algorithm to two different areas of the Western Carpathians: the glacial topography of its highest part and a karst plateau. By using a slightly simplified and specifically modified version of the physically-based algorithm, we achieved plausible and genetically interpretable results in both case studies, which confirms the value of physical geomorphometry in geomorphological mapping. Additionally, a novel concept of physical-geomorphometric signature was applied in both case studies as a support for the physical-geomorphometric analysis. The physical-geomorphometric signature is very helpful in the quantitative comparison between various genetic groups of landforms.