Mapping of aggregated floodplain plant communities using image fusion of CASI and LiDAR data

摘要: Combined optical and laser altimeter data offer the potential to map monitor plant communities based on their spectral structural characteristics. A problem unresolved is, however, that narrowly defined communities, i.e. at a low hierarchical level of classification in Braun-Blanquet system, often cannot be linked directly remote sensing for vegetation mapping. We studied whether how floristic dataset can aggregated into few major discrete, mappable classes without substantial loss ecological meaning. Multi-source airborne (CASI LiDAR) field were collected floodplain along river Waal Netherlands. Mapping results similarity alone did not achieve highest levels accuracy. Ordination showed terrain elevation soil moisture main underlying environmental drivers shaping vegetation, but grouping position ordination space is always obvious. ordination-based with clustering led syntaxonomically relevant assemblages yielded mapping accuracies.