By Michel Huizenga
Abstract:
The Natura 2000 network is an important step towards nature conservation in the European Union. The EU member states have to report every six years an assessment about the preservation of their Natura 2000 habitat types. Remote sensing techniques could be an important tool for habitat monitoring. In this thesis the actual role of remote sensing in operational habitat monitoring is investigated by a literature research and a questionnaire. Yet, remote sensing is mostly used as classification tool. Because of it’s high temporal resolution the MODIS terra satellite could be used very well to monitor habitat types. This thesis tries to discover the possibilities of the use of MODIS data for habitat classification and monitoring. This is done for the Veluwe area in the Netherlands. The phenological cycles for the habitat types heath land, grassland and forest were investigated and the spatial distribution was classified in two different ways. One method uses only the MODIS data and the other method uses also the LGN5 database of the Netherlands. It is not yet possible to classify, and thus monitor, areas with the same order of size as the Veluwe well with the MODIS data alone. Though, when used together with the LGN5 database, MODIS images have possibilities for Natura 2000 habitat monitoring. The accuracy of this method is the highest with an overall accuracy of 64.8 %, when it is based on the MODIS EVI values, and 56.8 % when it is based on the MODIS NDVI values.