Monitoring the evolution of the tidal estuary of the Scheldt river and its upstream river network based on LiDAR and airborne imaging spectroscopy

Abstract submitted to "4th EARSeL Workshop on Remote Sensing of the Coastal Zone"
Monitoring the evolution of the tidal estuary of the Scheldt river and its upstream river network based on LiDAR and airborne imaging spectroscopy
Dries Raymaekers
Flemish Institute for Technological Research (VITO)
Belgium
Luc Bertels
Flemish Institute for Technological Research (VITO)
Belgium
Els Knaeps
Flemish Institute for Technological Research (VITO)
Belgium
Daphne van der Wal
Netherlands Institute of Ecology (NIOO-KNAW), Centre for Estuarine and Marine Ecology
Netherlands
Keywords: hyperspectral remote sensing, LiDAR, tidal mud flats, vegetation classification, microphytobentos, cartography of river banks
Presentation preference: oral

The port of Antwerp, being the second largest seaport in Europe, is uniquely situated at the upper end of the tidal estuary of the Scheldt river. Both the tidal estuary, the Westerscheldt, and the upstream part of the river need to be monitored to preserve the access towards the port and the unique environment of the estuary. The following two studies illustrate the technique to combine LiDAR and hyperspectral remote sensing for detailed and continuous monitoring of a river network, including it’s bank structures and the tidal estuary.
(1) From 2005 onwards, hyperspectral airborne data is collected on a yearly basis for the Westerscheldt. Supported with an extensive field campaign, the grain size, water content and microphytobentos biomass of the tidal flats was retrieved in a consistent way over the years. This allows to present an overview of the evolution of the estuary subject to alternative dumping techniques.
(2) For the upper part of the Scheldt river, a combination of LiDAR and hyperspectral airborne data was used to produce detailed maps of the river network and their mud flats, salt marches and bank vegetation. The hyperspectral imagery was classified with a multiple binary algorithm based on Fisher’s linear discriminant analysis and trained with field information, collected around the time of the flight campaign. In this manner, vegetation types, mud and soil types were mapped for a river network having a total length of more than 220 km. As situated within the European Union Water Framework Directive (WFD), this approach will be executed every 3 years.

No fulltext available