Selected Interpretation Results:

EQUATORIAL-GUINEA

Figure 8: Geographic map of Equatorial Guinea

The Republic of Equatorial Guinea covers 28,051 km² and has a population of 360,000. Its major income comes from the exploitation of its forests.

The mangrove forest application of SAR has already been described for the border region between Gabon and Equatorial Guinea. Image 30 shows an area inside Equatorial Guinea which is completely covered by tropical forest. For overview purposes, a simple filtering enhances the visibility of the different features on this large scale. The second basic feature (besides the grey coverage corresponding to tropical forests) is the mountain chains on the left and right of the river that crosses the scene from the upper left to the lower right .

The location of the mountain chains and their internal structure is of major interest for geological purposes such as the exploitation of minerals. This application is described in detail in chapter Central African Republic.

Image 30: ERS overview image (100 x 100 km) not filtered

In general, images with only one type of feature (here tropical forest) are much more difficult to interpret than images with a number of different features. In addition, georeferencing is often almost impossible because it generally tends to be difficult to match satellite data with ground control points on large scale maps. In such cases, rivers are often the only way to match image and map. In areas where these problems are encountered, it is recommended to limit scales to above 1:500,000 in order to avoid location errors which are too great. Often, it may be useful to mosaic a few images with the same orbit conditions (ascending or descending), as georeferencing points might be identifiable on the multitemporal image with greatest precision.

One should however not overestimate this problem as basic features can also be extracted from the original data set through visible interpretation followed by a digitalization of the result, which is finally georeferenced with less difficulty than the original image. In case almost no points are found in the image, it is very useful to look through the data set in detail with a full screen resolution. Image 31 shows an example of what additional features can be seen when such "browsing" is performed. In this image one can not only detect the river bed in great detail, but also see a small village which can provide a new reference point.

Image 31: Subset with almost full screen resolution