Analysis of dolines using multiple methods applied to airborne laser scanning data

Delineating dolines is not a straightforward process especially in densely vegetated areas. This paper deals quantitatively with the surface karst morphology of a Miocene limestone occurrence in the Styrian Basin, Austria. The study area is an isolated karst mountain with a smooth morphology (former planation surface of Pliocene age), densely vegetated (mixed forest) and with a surface area of 1.3km2. The study area is located near the city of Wildon and is named “Wildoner Buchkogel”. The aim of this study was to test three different approaches in order to automatically delineate dolines. The data basis for this was a high resolution digital terrain model (DTM) derived from airborne laser scanning (ALS) and with a raster resolution of 1×1m. The three different methods for doline boundary delineation are: (a) the “traditional” method based on the outermost closed contour line; (b) boundary extraction based on a drainage correction algorithm (filling up pits), and (c) boundary extraction based on hydrologic modelling (watershed). Extracted features are integrated in a GIS environment and analysed statistically regarding spatial distribution, shape geometry, elongation direction and volume. The three methods lead to different doline boundaries and therefore investigated parameters show significant variations. The applied methods have been compared with respect to their application purpose. Depending on delineation process, between 118 and 189 dolines could be defined. The high density of surface karst features demonstrates that solutional processes are major factors in the landscape development of the Wildoner Buchkogel. Furthermore the correlation to the landscape evolution of the Grazer Bergland is discussed. •Airborne laser scan data facilitate accurate doline delineation.•Depending on the delineation process, the number of dolines varies significantly.•The Leitha limestone area of Wildon hosts a significant cluster of dolines.