Departments

• its attribute structure must be flexible
• it must be easily extensible with additional attributes
• though at first it will be used solely on the ITC it must not raise any barriers that make deployment on the internet
• support large quantities of data (e.g. million records, 200 GB of data)
• be compatible with our existing (in-house) analytical software tools (e.g. ArcGIS, Ilwis, ERDAS,…) 
  

High spectral resolution, or hyperspectral, imagery combines spatial imaging with a spectrometer. A spectrometer is a device which records up to several hundred narrow spectral bands with a spectral resolution of 10 nm or narrower. In other words, rather than having a few wide bands for each pixel, imaging spectrometers produce a more complete spectrum for every pixel of the image.

Unfortunately, broad band scanners tend to average out important differences in reflectance that would allow vegetation species and types to be identified. In addition, spectral ranges where the broad bands are placed may not coincide with the areas of maximum difference in the spectral curves for vegetation. There is great potential for hyperspectral remote sensing in sustainable land management. Materials and cover types may be identified, permitting an improved ability to:

• map and monitor land cover and materials
• monitor land degradation (changes in vegetation and structure)
• measure evapo-transpiration
• assess and monitor environmental degradation and fragmentation.

The rationale behind the GIS approach to species' distribution modelling is simple: the database contains a large number of data sets (layers), each one of which describes the distribution of a given measurable and mappable environmental variable. The ecological requirements of the species are defined according to the available layers. The combination of these layers and the subsequent identification of the areas that meet the species' requirements identify the species' distribution range, either actual, if there is evidence of presence, or potential, if the species has never been observed in that area. This basic scheme can be implemented using different approaches.

The pictures show a GLM model based on bioclimatic data. The GLM uses random points sampled within the known distribution range of the species as training set. The output is a map of the probability of occurrence of the species (here the African Buffalo - Syncerus caffer). The graph illustrates the relationship between probability of occurrence and one of the parameters used (length of dry season in days).

In response to the continuing loss of forest resources for rural people in many parts of the world, the role of tree resources outside forests (“trof” in short) is becoming increasingly more important.
Trof has various productive/economic, protective / ecological and social functions.

With the growing importance of Trof, also a growing interest in information about them is emerging.

http://www.itc.nl/departments/nrs/eia/eia-sea.html

Quantitative assessment of effects of climate change on mammals biodiversity in Africa in the years 2050 and 2080.

Current species richness is displayed at the centre of the figure to related number of losses with initial number of species.

Generalised additive models (GAM) relating the mammal species distributions to the six bioclimatic variables were calibrated using a random sample of the initial data (70%) and a stepwise selection methodology with the most parsimonious model being selected using the Akaike Information Criterion (AIC). Individual GAM were then used to predict future extent of the species distributions under the two scenarios.

The area of the Phu Wiang watershed is monitored, with the help of aerial photographs and satellite images. Aerial photographs of 1954, 1976 and 1984 were interpreted. The resulting maps were digitised. Landsat TM image of 1988 was classified. Field data was collected in 1989 to validate the 1988 classification. The land cover/use information of the different years was superimposed with GIS overlay to detect changes in land cover/use. Deforestation in the area was seen from 1954 till 1984. Protection and forest plantations started in 1981. Since then the forest area did not decrease.

Using data assimilation approaches is possible to map wheat yield in the Camargue region using multi-temporal SPOT imagery.