Publication Date:
2019-07-16
Description:
Analysis of coastal marine algae communities enables to adequately estimate the state of
coastal marine environment and provides evidence for environmental changes.
Hyperspectral remote sensing provides a tool for mapping macroalgal habitats if the algal
communities are spectrally resolvable. We compared the performance of three classification
approaches to determine the distribution of macroalgae communities in the rocky intertidal
zone of Heligoland (Germany) using airborne hyperspectral (AISAeagle) data. The classification
results of two supervised approaches (maximum likelihood classifier and spectral angle
mapping) are compared with an approach combining k-Means classification of derivative
measures. We identified regions of different slopes between main pigment absorption
features of macroalgae and classified the resulting slope bands. The maximum likelihood
classifier gained best results (Cohan’s kappa = 0.81), but the new approach turned out as
time effective possibility to identify the dominating macroalgae species with sufficient
accuracy (Cohan’s kappa = 0.77), even in the heterogeneous and patchy coverage of the
study area.
Repository Name:
EPIC Alfred Wegener Institut
Type:
Article
,
isiRev
Permalink
