ORIGINAL PAPER Mediterranean forest mapping using hyper-spectral satellite imagery Selma Etteieb & Mounir Louhaichi & Chariton Kalaitzidis & Ioannis Z. Gitas Received: 20 January 2012 / Accepted: 31 October 2012 # Saudi Society for Geosciences 2012 Abstract Mediterranean forests are characterized by spatio- temporal heterogeneity that is associated with Mediterra- nean climate, floristic biodiversity and topographic variability. Satellite remote sensing can be an effective tool for characterizing and monitoring forest vegetation distribu- tion within these fragmented Mediterranean landscapes. The heterogeneity of Mediterranean vegetation, however, often exceeds the resolution typical of most satellite sensors. Hyper-spectral remote sensing technology demonstrates the capacity for accurate vegetation identification. The ob- jective of this research is to determine to what extent forest types can be discriminated using different image analysis techniques and spectral band combinations of Hyperion satellite imagery. This research mapped forest types using a pixel-based Spectral Angle Mapper (SAM), nearest neigh- bour and membership function classifiers of the object- oriented classification. Hyperion classification was done after reducing Hyperion data using nine selected band com- binations. Results indicate that the selection of band combi- nation while reducing the Hyperion dataset improves classification results for both the overall and the individual forest type accuracy, in particular for the selected optimum Hyperion band combination. One shortcoming is that the performance of the best selected band combination was superior in terms of both overall and individual forest type accuracy when applying the membership classifier of the object-oriented method compared to SAM and nearest neighbour classifiers. However, all techniques seemed to suffer from a number of problems, such as spectral similarity among forest types, overall low energy response of the Hyper- ion sensor, Hyperion medium spatial resolution and spatio- temporal and spectral heterogeneity of the Mediterranean ecosystem at multiple scales. Keywords Mediterranean forests mapping . Hyperion satellite imagery . Hyperion data reduction . Pixel-based Spectral Angle Mapper (SAM) . Object-oriented classification Introduction Forests are considered one of the most valuable renewable natural resources because of their economic, environmental, aesthetic and recreational values as well as their critical role in influencing global atmospheric cycles Vanhala et al. 2005; Gondard et al. 2006; Lawrence et al. 2006). Forest land cover mapping has become a widely used and an important resource for characterizing forest structure and function. Tremendous Mediterranean forest heterogeneity makes it difficult to accurately delineate and map forest land cover types. This spatiotemporal heterogeneity—typical of Mediterranean forest vegetation communities—is caused by high floristic richness, significant topographical diversity and variable climatic conditions. These forests are shaped and fragmented over time by natural ecological processes forming distinct patches that can be intermixed with other land cover types (Alados et al. 2004; Hernandez-Stefanoni, 2005; Mallinis et al. 2008; Torras et al. 2008). S. Etteieb (*) : C. Kalaitzidis Department of Environmental Management, Mediterranean Agronomic Institute of Chania, Alsyllio Agrokepiou, P.O. Box 85, Chania 73100 Crete, Greece e-mail: selmaetteieb@yahoo.fr M. Louhaichi International Center for Agricultural Research in the Dry Areas (ICARDA), P.O. Box 950764, Amman 11195, Jordan I. Z. Gitas Laboratory of Forest Management and Remote Sensing, Faculty of Forestry and Natural Environment, Aristotle University of Thessaloniki, P.O. Box 248, 541 24, Thessaloniki, Greece Arab J Geosci DOI 10.1007/s12517-012-0748-6