Semiautomatic construction of isobase surfaces: A case study from the central Western Carpathians

• The analysis of isobase surfaces extends possibilities of morphostructural research.
• A semi-automated method was developed to replace manual construction of isobase surfaces.
• The analysis runs in a server environment, everyone can try it in web browser using own data.
• A comparison with the manually constructed isobase surfaces successfully verified the algorithm's results.

This article describes a method for semiautomated construction of isobase surfaces in the environment of a Geomorphological Information System (GmIS). The motivation to automate the process is that the manual construction of isobase surfaces involves several decision procedures, and also a lot of manual vectorization which can be time-consuming if the area of interest is large.A (semi-)automated process of isobase surface construction could accelerate the process of morphostructural analysis. Therefore, the contribution focuses on design, development and testing of such a method. The developed tools were compared to an expert driven construction of isobase surfaces. The automatically constructed data layers were compared to the layers that were constructed manually in the southwest part of the study area (Turčianska kotlina basin (Western Carpathians) and surrounding area in Slovakia). Three different statistical methods were used (correlation, linear regression and DTM volume difference checking). All three methods proved very high similarity of automatically created isobase surfaces to expert driven isobase surfaces creation. The results show that the method of semiautomated creation can be used in morphostructure analysis and save the time needed for manual isobase surface creation.The method's technological background is based on the ESRI platform. The original desktop solution has been developed for use on a client/server architecture that is able to present the capabilities of GmIS to a wider geomorphological audience.

Figure optionsDownload as PowerPoint slide