M eR EE ERR P 
6. Anttopogene influences(Fig. 8) 
They are easily detectable as their change is frequently and highly correlated 
with the population density and vegetation distribution. 
In the case of the study area an immediate detection of antropogene influences 
lead to the determination of the amount of subsistance fields(shambas). 
The shamba-structure is rotational,less shifting,therefore a tendency of clu- 
stering is obvious. 
This gives an advantage of detection on the satellite images. While a single 
shamba has an average area of 1 to 1.5 ha and a nearly round shape which is 
just a bit larger than a pixel of LANDSAT,several shambas ordered in a rotatio- 
nal shape are easier to detect. 
The three main shamba status can be detected and differentiated on a false 
colour composite image for: 
1. burned and bare soil status by green,blueish and light 
2. fallow status with low density vegetation by light red and brown 
3. full crop density by red and sometimes dark red(the latter is not to exchan- 
ge with the same colour for damaged vegetation in some cases) 
Swampy and flood areas in which rice is planted are difficult to eliminate from 
the described colours as the fields are very limited in’ extension and are scat- 
tered irregularly. 
A shamba statistic carried out on 2 processed scenes which allows a conclusion 
of possible shamba numbers by samples within the area leads to the most proba- 
ble numbers in reality. 
By counting them in their different status and forming ratios a prediction of 
an environmental sparcing and non-sparcing crop production can be made. 
Conclusion to the population density is possible as far as sufficient nutrition 
of the population is given and no inadequate living conditions are dominating, 
like overpopulation,famine,deseases and high birth or death rates. 
In case of the sample area balanced living conditions of the population are 
still given andd sertification is rather low. 
7. Forecasting and predicting environmental development on the basis of peri- 
odic monitoring 
A prediction of the future ecological development of the region was made by 
forecasting the development of the 4 main factors mentioned under point 2. 
using ground datas together with LANDSAT-observations. 
1. Particular drainage basins on different scenes show a decreasing vegetation 
on the slopes of the drainage channels due to rill and gully erosion. 
This can be confirmed by the sharp contrasted crests separating the lowlands 
from the plateau areas. Within these crested parts gully erosion is progressing. 
From the amount of sand and sediment detectable on band 4 in the lower sloped 
regions of the rivers a high erosion on the steeper parts can be concluded. 
Consecutive images from different dates show a tremendous change in the erosio- 
nal pattern. Photographs of ground truth flights confirmed these facts. 
The drainage density in particular areas especially in the lower basin regions 
is fairly low due to extended flat sediment layers. 
In the mountainous region of the plateau areas the drainage density increases. 
The medium sloped regions are highly affected by a mixture of gully and splash 
erosion,enhanced by overgrazing and firewood cutting. 
If no conservation measures are taken these parts will get rid of their humus 
soilcover and gully erosion will progress. 
As the peasants are planting in flat,moderate sloped areas and swampy parts, 
these other parts are neglected. Terracing would be measure against these ne- 
glective developments. 
456