Section: Application Domains
Remote Sensing for natural risks and agricultural monitoring
This area concerns specific case studies on crucial environmental topics, involving strong cooperation with environmental experts and access to multiple data sources (satellite images, ground truth, validation data, models). Currently the Clime team is involved in five activities, which are strongly related from a methodological point of view:
- Soil degradation:
AVHRR and MODIS image sequences are used to assess the changes of land use and land cover in the Pantanal area (Brazil), for monitoring of soil degradation, erosion, deforestation and sustainable agricultural practices.
the investigations concern the characterisation of long-term time series of vegetation indices (computed from AVHRR or MODIS acquisitions) for the early detection of desertification in Northern Africa and Southern Italy.
- Land cover changes and early classification:
the goal is to assess the potential of SPOT sequences for the detection of land use changes in agricultural areas, and for obtaining an early land use classification with respect to the vegetation cycle.
- Hail damage assessment:
the purpose is to define a method for the detection and quantification, from satellite images, of damages caused by hail to vineyards.
- Fire plume detection:
it aims at the definition of methods for fire plumes detection from multispectral satellite images (NOAA-AVHRR, METEOSAT). The investigation concerns the definition of spatial and temporal constraints for improving the robustness of pixel-based detection methods.