Expert knowledge for translating land cover/use maps to General Habitat Categories (GHC)

• DOI: 10.1007/s10980-014-0028-9
• Publication Date: 2014-04-15T10:07:01Z
• Authors: Maria Adamo; Cristina Tarantino; Valeria Tomaselli; Vasiliki Kosmidou; Zisis Petrou; Ioannis Manakos; Richard M. Lucas; Caspar A. Mücher; Giuseppe Veronico; Carmela Marangi; Vito De Pasquale; Palma Blonda
• Keywords: Land Cover

Monitoring biodiversity at the level of habitats and landscape is becoming widespread in Europe elsewhere as countries establish international national habitat conservation policies monitoring systems. Earth Observation (EO) data offers a potential solution to long-term through direct mapping or by integrating Land Cover/Use (LC/LU) maps with contextual spatial information situ data. Therefore, it appears necessary develop an automatic/semi-automatic translation framework LC/LU classes classes, but also challenging due discrepancies domain definitions. In context FP7 BIO_SOS ( www.biosos.eu ) project, authors demonstrated feasibility Food Agricultural Organization Cover Classification System (LCCS) taxonomy class translation. They developed automatically translate LCCS into recently proposed General Habitat Categories classification system, able provide exhaustive typology types, ranging from natural ecosystems urban areas around globe. However terminology, plant height criteria basic principles between two domains inducing number one-to-many many-to-many relations were identified, revealing need additional ecological expert knowledge resolve ambiguities. This paper illustrates how phenology, topological arrangement landscape, spectral signature multi-temporal Very High Resolution (VHR) satellite imagery measurements can be used such Concerning height, this compares results obtained using accurate values extracted LIght Detection And Ranging (LIDAR) exploiting EO texture features (i.e. entropy) proxy information, when LIDAR are not available. An application for Natura 2000 coastal sites Southern Italy discussed.