2.3. Some landscape elements which are not visible 
for the normal human viewing can become visi- 
ble upon aerial photographs or upon satellite ima- 
ges. This "visibility" problem is double. 
Some surface phenomena as f.i. shadow marks (mi- 
cro-relief). soil marks (drainage) or vegetation 
marks are difficultly observable in the field but 
became obvious by a bird view. 
These marks are very often ephemeral and are only 
visible when very specific conditions of soil 
moisture and plant growth are present (fig. 4). 
  
Fig. 4. Oblique aerial photograph of the Waardamme 
area (J.SEMEY) 
This oblique aerial photograph is taken near to 
the village of Waardamme in sandy Flanders. A 
strange design of concentric circles appears. It 
remains an enigma. In analogy with other similar 
sites it looks like a cult place of the Bronze Age. 
Other landscape informations, as vegetation stress 
of vegetation sickness are characterized by chan- 
ges in reflection, especially in the near-infrared 
part of the spectrum, not visible for our eyes. 
These informations become obvious on the colour 
infra-red emulsion, or on the near Infra-red bands 
of the satellite images. 
Thermal characteristics, also an important compo- 
nent of the environment, can become visible by a 
thermal scanning, as f.i. on band 6 of the Landsat 
T.M. images. 
2.4. Aerial photographs and especially the satelli- 
te images allow a temporal analysis. The com- 
parison between images taken during different sea- 
sons or at different years can show different as- 
pects of a landscape or make clear the evolutions 
which are going on in a landscape. This possibili- 
ty is especially interesting in areas were quick 
changes are occuring as f.i.in the fringes of the 
deserts or in the periphery of the big cities. 
2.5. The multispectral nature of the satellite ima- 
ges makes possible to perform digital classi- 
115 
fications, using the appropriate algorithms. 
These classifications are touching the very funda- 
menta] behavior of the surface phenomena, they are 
indeed based upon the exchange of energy, incident 
and reflected, showing the specific spectral re- 
flectance characteristics of the objects (fig. 5). 
  
Fig. 5. Supervised image classification of S.W. 
Messinia (Greece). 
Here an example of supervised image (SPOT) classi- 
fication in order to identify the soil forming se- 
diments in a part of S.W.Messinia (Greece). 
Black - water 
Dark grey = soils developed on alluvium 
Grey = soils developed on colluvium 
Light grey = soils developed on flysch material 
White = soils developed on limestone material. 
Different classification methods were developed, 
each adapted to specific interpretation needs 
(LILLESAND and KIEFER, 1989). 
The meaning of this enumeration of some possibili- 
ties of the Remote Sensing documents is to show 
that these documents form a powerful tool for the 
study of landscapes, the associated settlements and 
the spatial relationships between them. The Remote 
Sensing enables as well the analytical as the ho- 
lostic approach. 
3. The landscape as archives. 
This statement means simply that the landscape con- 
tains a lot of information, but, exactly the same 
as for archives, it needs serious work and skill 
to extract this information.