Landform Classification in R
Jasiewicz and Stepinkski (2013) provided a novel method for classification and mapping of landform elements from a digital elevation model (DEM) based on the principle of pattern recognition rather than differential geometry. At the core of the method is the concept of a geomorphon (geomorphologic phonotypes) — a simple ternary pattern that serves as an archetype of a particular terrain morphology. From a total of 498 possible combinations a set of ten distinct patterns is formed (see figure 1).
The basis for this case study is the DEM model derived from NASA’s Shuttle Radar Topography Mission (SRTM) flown back in the year 2000. The original data set is scaled-up to an uniform grid size of 0.001 °. R package raster provides a focal function which is applied to the DEM data set. Both, the raster resolution and the size of the focal window determine the magnitude order of landforms which are to be detected. A window of 7 x 7 grid cells is used in this study. It allows for the detection of small-scale landforms in the complex topography in Switzerland. The R script plus supporting information can be found here.