Aurélie Dehouck

Hyperspectral field database in support to coastal wetland mapping

This paper presents the hyperspectral field measurements acquired at Arcachon (France) over sediments and vegetation species characteristic of the macrotidal lagoon. The reflectance spectra are analyzed to define the variables of interest in order to propose a classification scheme useful for management purposes. The main results show that coupling winter to summer optical multispectral imagery allows discriminating invasive (Spartina anglica) from endemic (Spartina maritima) lower salt-marsh vegetation species. Also, the bottom color and henceforth sediment mineralogy largely influences the determination of seagrass meadows cover. In addition, preliminary results show the potential of identifying biofilms over sandy to muddy sand. Finally, the characterization of macro-algae deposits on the intertidal flats may involve the exploitation of finer and more numerous spectral bands than those typical of classical multispectral imagery.

Use of optical and radar data in synergy for mapping intertidal flats and coastal salt-marshes (Arcachon lagoon, France)

This contribution explores the potential of high-resolution satellite SAR imagery (TerraSAR-X, ALOS-PALSAR) for mapping coastal habitats in complement of optical data (SPOT-5). It addresses X- and L-band SAR signatures over intertidal flats and coastal salt-marshes by investigating the mean backscattering coefficient σ0 over the major environmental units and its variation associated with physical parameters (soil roughness, soil moisture, tidal inundation) and instrumental configurations. The major findings outline a great potential of TerraSAR-X data to detect oyster beds and discriminate Zostera noltii seabed from salt-marsh vegetation species. Based on these results, a multi-sensor multi-temporal mapping strategy was tested running simple supervised classification algorithms. These tests show that mapping performance is greatly enhanced when running Mahalanobis classifier on a 6-band concatenated image. Further work is needed to account with green macro-algae deposits and microphytobenthos in order to demonstrate the full capabilities of spaceborne sensors to provide an exhaustive mapping of intertidal environments.

Wavelet based texture modeling for panchromatic very high resolution image classification: Application to oyster racks detection

This study evaluates the potential of wavelet-based texture multivariate modeling for the detection of cultivated oyster fields and their differentiation from abandoned fields in Very High Resolution panchromatic PLEIADES data. The proposed models are tested in a supervised classification context using a training database composed of extracted image patches representative of the land covers of interest. The obtained classification results exhibit high detection rate for cultivated fields. Classification errors appear however in the detection of abandoned fields. These results demonstrate the interest of such model for the characterization of inter-tidal ecosystems and opens up perspectives for their use in the management of oyster farming activities.