Geoscience Reference
In-Depth Information
derive many types of information layers. In order for this to take place, each of the
pixels of the DEM is successively considered as being the center of an nun pixels
window. The altitude values of this window are adjusted by regression to obtain a
plan from which it is possible to calculate different topographic variables such as:
- altitude (value of the window's central pixel);
- gradient of the slope, which corresponds to the incline of the adjustment plan
that is to be carried out on the window;
- orientation of the slope;
- topographic roughness, which corresponds to the standard deviation of the
residuals within the adjustment in altitude; this topographic roughness highlights the
importance of the irregularities in relief that exist (in flat regions or on a straight
mountain face there may be no irregularities in relief);
- the theoretical model for global solar radiation is calculated for different key
moments of the year (summer or winter solstice, spring and fall equinox), and also
takes certain hidden topographic factors into consideration (e.g. the effect that
shadows have on the model is particularly significant, especially in winter).
In this way, other information layers can be provided by the DEM.
2.2.3.2.
Land cover
Land cover, from satellite or aerial images, provides information on how the
landscape is made up of different classes of objects (vegetation, developed sites,
networks, crops, etc.). Land cover is the second most useful, if not essential, source
of information used to model the climate in its different spatial variations. With the
use of relevant calculations, it is possible to derive other data from information
relating to land cover:
- with the help of TM imagery it is possible to calculate the Normalized
Difference Vegetation Index (NDVI). This is the index that estimates biomass
potential. The NDVI is calculated from channels 3 and 4 of the Landsat TM images
by using the following formula: (TM3-TM4) / (TM3+TM4);
- from a simple classification it is possible to calculate the distance that exists
between each pixel of different objects such as the sea, the forest or the towns.
The land cover of a particular country can also be provided by other sources of
information. CLC (supplied by the French Institute for the Environment) observes
land cover on a scale of 1:100,000. The main aim of CLC is to respond to both
French and European needs. The minimum surface area that can be considered
(description threshold) is 25 hectares; the minimum width of the network elements
that are to be observed (hydrography in particular) is 100 m. This information, is
suited to a raster GIS with a resolution of 250 m and could be if it is a question of
modeling small-scale climatic spatial variations, for example the variations that
occur in France.
