Terrain modelling of the complex volcanic terrain of Ischia Island, Italy

Abstract

A high-resolution digital elevation model (DEM) (2 m spacing) is used in an attempt to model the volcanic terrain of Ischia Island (~150 ka ≥ age ≥ AD 1302). Ischia Island consists of a high-relief, resurgent central-western sector, a western sector affected by landslides related to the resurgence, and an eastern, lowland sector in which a volcanic field was emplaced. The interpretation of elevation, slope, and aspect derived from a DEM revealed two volcanic cones near the northeast coastline and a huge semicircular ridge in the central portion of the island, corresponding to the Mount Epomeo resurgent block. The Mount Epomeo area shows an extensive drainage network developed on the southern flank, which represents the scar of a huge, prehistoric debris avalanche. The elevation and slope statistics and hypsometric curve allowed terrain classification according to well-known standards (present mean earth elevation, hypsometric integral ranking). Peaks and critical points were discovered in elevation and slope frequency distributions, and asymmetries in the directional development of island relief in relation to area extent, slope, and elevation were revealed from aspect rose diagrams. The results allow us to discriminate zones with different volcanic structures and deposits (e.g., zones affected in the past by resurgence phenomena, Mount Epomeo relief, central sector of the island), caused by the emplacement of a volcanic field with lava flows and domes (eastern sector), landslides (western sector), or controlled by tectonic processes (western flank of Mount Epomeo). Aspect regions provided the terrain partition framework (aggregation of adjacent pixels with aspect direction within a certain range) into objects. The objects were parametrically represented by a set of attribute-value pairs on the basis of elevation and slope and classified in eight clusters using cluster analysis (centroid method). The interpretation of cluster centroids and the spatial distribution of the eight clusters revealed a terrain organization controlled by both endogenic (volcanic and tectonic) and exogenic (erosion) processes. In particular, clusters allowed the identification of topographic features related to gravity flows (lava flows, domes, landslides, debris flows), tectonics (faulting), and volcanic structures (resurgent blocks, volcanic fields).