Design of long-term monitoring solutions of the vegetation of flood plains by means of high-resolution satellite data (BMVBS)

Initial position
The structure and the inventory of plant species in the flood plain of the estuaries of the Elbe river are subject to constant change. The reasons for this are naturally or anthropogenically induced hydro-morphological changes. It is assumed that because of the growing frequency of extreme hydrological events, changes in the vegetation on the banks and in the flood plains of estuary water ways are induced to an increasing extent. Therefore quantitative and qualitative changes especially to the reed stock which is important for protecting the flood plains can be expected. On the one hand, the reed belts as a means of protection for the embankment may disappear and on the other hand, neophytes which have a less developed net of roots may grow at the so called disruptive spots. Consequently, the flood plains with their buildings are exposed to an increasing danger of erosion. Safety measures for banks and dykes will become necessary to a greater extent. Adaptation strategies in regard to maintenance measures are therefore essential.

Aims and objectives
The aim of the project is to check and quantify with high spatial resolution the changes in the tidal waterway, the Elbe, as described in the initial position. In order to achieve this, it is necessary to record the qualitative and quantitative changes by means of long-term monitoring. Through the multi-temporal analysis of remote control data, the aims of maintaining the embankment and the flood plains can be confirmed, adapted or set anew.
The aim of this part of the project consists of developing a plan for a robust monitoring concept supported by satellite images for long-term monitoring of the vegetation in the flood plains of the estuaries. In the course of this, the potential of the temporally and geometrically high-resolution RapidEye sensors will be evaluated. The results will be compared to existing analyses made using digital aerial photographs of the highest possible resolution and to hyper-spectral analyses which still have to be done (Prof. Schmidtlein, Bonn), and will be assessed with regard to their suitability for long-term monitoring.
Sensitivity and transferability of the method of evaluating the RapidEye data will be tested at the same time. The method used will be a combination of spectral and object-based classification. Indices of different spectral bands will be determined in order to optimize the differentiation of several varieties of vegetation. In addition to the well-known NDVI, new procedures with the RedEdge band of the RapidEye sensor will also be tested. The result is an analysis of the potential of high-resolution satellite data for use in long-term monitoring the vegetation of flood plains. The aim of the concept being developed is to provide decision support in regard to the long-term monitoring of the changes in estuaries.