Measurement of DEM roughness using the local fractal dimension

The relationships between geological features and DEM surface roughness are studied using fractal geometry analysis. The estimate of the fractal dimension in a 3D space is performed locally on DEM surfaces using an adaptive “box-counting” technique. This procedure has been applied to two regions chosen to represent differences in lithological and tectonic conditions. The first one corresponds to a faulted, homoclinal sedimentary sequence and the second to a compound stratovolcano. In both cases, the results obtained show that the local fractal dimension detects distinct morphologic features. In the first case (Vittel, eastern France), the local fractal dimension detects the limits of the geological units as borders of homogeneous zones or transitions between different dissection depths. Fine structural lines underlying the Vittel fault can also be extracted. In the case of Mount Ararat (Eastern Turkey), different classes can be distinguished within the volcanic formations using a statistical analysis of the fractal dimension values. The treatment can also provide information about the local and regional geological structures. These first results show that measuring DEM surface roughness represents a helpful tool for extracting and mapping morphometric features.