Digital Elevation Model Construction Using Geostatistics and Geological Expert Knowledge -- A Case Study in Oitti Area in Southern Finland

Authors

• Rangsima Sunila Aalto University
• Kirsi Virrantaus Aalto University

Keywords:

DEM, Kriging interpolation, Variogram modelling, Geological expert knowledge

Abstract

Digital Elevation Models (DEMs) are an important topic in geographical information science as DEMs are commonly used in GIS analysis applications related to physical environment. There are various techniques used for modelling the elevation surface and in this study the focus is on interpolation methods based on geostatistical techniques. This study is further research extended from previous work in which the kriging method was applied in elevation modelling. In the previous research, it was shown that kriging is a suitable tool for constructing an elevation model in a study area which presented glacial and postglacial clays. Therefore, it was rather simple to build a DEM in that area by using the ordinary kriging method for interpolation. However, in many locations in Finland, it is not simple to build a DEM. The complex structure of the land formation may result in a complicated structure. Quaternary deposits consist of elongated moraine ridges that affect the geomorphology and, so, the elevation model. In Finland, the elevation model of moraine ridge areas is important because sources of fresh water are situated in these kinds of land formations. Mapping of the groundwater areas is necessary because of the EU directives. The aim of this research is to present a comparison of two kriging approaches in building elevation models. In the first one, elevation model is built by using the same variogram model in the whole study area. The second kriging approach uses geological expert knowledge in order to divide the study area into three sub-areas, a clay-dominated area in the west and east and a moraine ridge in the centre. It was shown that expert knowledge of Quaternary deposits can be applied in digital elevation modelling in order to produce a higher-quality result.