|| Institutes |
Dutch title: Radarbeelden voor kartering en monitoring van wetlandecosystemen in het Tsjaadbekken|
Parent project: Research action SPSD-I: Sustainable management of the North Sea, more
Reference no: T4/DD/35
Period: December 1998 till June 2001
|| Top |
- Ghent University; Faculty of Bioscience Engineering; Department of Forest and water management; Laboratory of Forest management and spatial information techniques (FORSIT), more
- Belgian Science Policy (BELSPO), more, sponsor
In the framework of their sustainable management, a better understanding of ecosystem processes in wetlands, and of the impacts of natural and human origin, require techniques to characterize these wetlands spatially, and to monitor the spatial and temporal dynamics.
Wetlands in Africa in general, and in the proposed test zone in the Chad Basin in particular, are of paramount ecological and economical importance, but are threatened by population pressure, with serious implications for the local population and their environment.
Management authorities need up-to-date spatial data that is able to provide an accurate picture of present land use en de extent of the flooded area. In view of the very dynamic nature of wetlands, research to characterise these spatio-temporal patterns is a considerable added value.
Research on the usefulness of imaging radar is being presented with the following objectives:
- Quantitative investigation of the radar backscatter signal in relation to wetland ecosystem components (aquatic, macrophytic, and terrestrial in the agricultural buffer zone) through statistical separability measures and testing against existing theoretical backscatter models.
- Optimisation of cartography of wetlands by development of procedures for image segmentation and classification: multitemporal classification and image fusion: visible/NIR and microwave data.
...with the pre-operational objective to test and supplement methodologies to contribute a qualitative (land use categories) and quantitative (statistics) input for wetland management, and for the pilot area in the Chad Basin in particular (thematic objective: sustainable management of tropical ecosystems).
On one hand the proposed project's activities are supported by the expertise acquired in a current SPOT VEGETATION project, while on the other hand the developed methodologies will envisage the use of ENVISAT ASAR data, that will become available 2 years from now. Hence the project features a development phase (using RADARSAT data), and a subsequent validation phase testing and applying the developed techniques and methods using ENVISAT ASAR. This explains while a project period of 2.5 years is suggested, in order to guarantee the completion of the validation phase.
It is expected that the project's outcome will contribute to the knowledge of radar backscatter patterns, mapping using space borne imagine radar with a reasonable accuracy, and methods (algorithms) for image processing and image synergy between visible/NIR and microwave data.
The results should be capable of extrapolation to wetland mapping elsewhere, but it is expected that certain research aspects should be useful for mapping and monitoring of flooding patterns in general.