Identification of Anomalous Morphological Landforms and Structures Based on Large Scale Discrete Wavelet Analysis

The identification and delineation of morphological patterns through a quantitative approach constitutes an important stage both for the geomorphological characteriza-tion of a region and for the interpretation of landforms for engineering geological purposes. This task is usually based on the expert judgment of terrestrial and aerial surveys or of thematic maps. This is potentially a cause of uncertainty, since personal and empirical opinions may bias the procedure. Here an automatic numerical approach, based on 2D discrete wavelet transform of the digital elevation model, is applied in order to map the anomalies of the topographic surface at large scale. The approach focuses on the digital elevation model of the Salento peninsula in Southern Italy de-noted by a widespread outcropping of Quaternary sedimentary units largely covering an intensively fractured Cretaceous limestones substratum. The data-mining used pro-cedure allowed for emphasizing details related to morphological structures.