Detection of disturbances in natural vegetation
Institute of Surveying, Remote Sensing & Land Information (IVFL)
University of Natural Resources and Life Sciences, Vienna (BOKU)
Peter Jordan Strasse 82
Prof. Clement Atzberger
Head of Institute
Tel: +43 1 47654 5101 (direct line)
Tel: +43 1 47654 5100 (secretariat)
Forest disturbances occur naturally and are important for forest ecosystems. With changing climate these biotic (e.g. insect outbreak, fungi and nematode) and abiotic (e.g. windthrow, forest fires and snow breakage) calamities are occurring more frequently (Seidl et al., 2011, Seidl et al. 2014). In addition, human induces impacts (like illegal logging and pollution) are affecting forests. This is a challenging task for forestry since the disturbances cause ecological and economical loss (Hanewinkel et al, 2013) and it is insecure where and how intense the forest is affected.
The objective of this activity is to demonstrate a new approach to detect disturbances in forested areas of Europe (e.g., caused by forest clearing, windthrow and pathogens) using change-detection analysis of multi-sensor temporal series of EO data. Contrary to classical approaches using images acquired at different times by the same sensor, we propose to combine Normalized Difference Vegetation Index (NDVI) data/information from two sources:
- MODIS (NDVI at 250 m GSD)
- Proba-V (NDVI at 100 m GSD)
The idea is to leverage, on the one hand, the length of the available time series from MODIS (starts in 2002) and, on the other hand, use the higher spatial resolution of Proba-V with up to 100m pixel size (through its central of three cameras). This resolution is significantly better than the 250m (nominal) of MODIS.
NRT disturbance detection on the PV MEP VM for selected forested areas of Europe combining MODIS and PROBA-V time series.