— n 
ie 
ylished 
by 
carpmert) 
ore 
n the 
floor 
shed: 
cal 
1e 
ed 
hrub 
ally 
sts 
)hoto- 
lap. 
rea. 
Land 
lated 
| used 
r for 
| with 
nts, 
'educed 
| ern- 
«x 
          
  
Pu: 
Cpu i m REEL, LE RB SE EE 
detailed interpretation was then carried out with aerial photographs 
at a scale of 1/50000, taken in January 1967 and up dated by landsat 
information. However, major and most distinct habitat types identified 
on the aerial photographs, and recognisable on the Landsat scenes, were 
used as a guide in the interpretation and subsequent delineation of 
vegetal units from the Landsat images. This approach enabled a coinsice 
description of the units based on habitat characteristics, as in 
KREMU's Habitat Classes For Aerial Surveys (October, 1979). Light 
aircraft and ground surveys were also performed to gather plant species 
information. The resulting map was compiled at a scale of 1/125000. 
3 - Geology (Fig. 3) - 
Good maps at a scale of 1/125000 are already published by the 
Mines and Geology Department and were used in the study. 
4 - Soil Mapping - 
A map of part of the area, made by the Kenya Soil Survey in the 
Ministry of Agriculture and published in 1978 was also utilized for 
the purpose of the study. 
RESULTS AND DISCUSSION 
The final land capability map was compiled at a scale of 1/125000, from 
the various sources of information (Fig. 4). Some examples of the extensive 
legend are given in Fig. 5 . The major entrances are slopes and geology, 
followed by indication of main soil units, vegetation types and present land 
use. 
The actual land suitability and land potential is qualitatively eva- 
luated for the use of 1) arable rainfed farming,2) irrigated farming, 5) range, 
4) afforestation and 5) natural cover according to four levels: suitable, 
moderately suitable, marginally suitable and unsuitable. Specific remarks 
about limitation and positive aspects and climatic data are also mentioned when 
available. 
The study gives a simple approach of the land capability with the spatial 
distribution of each class from which total acreage can easily be worked out. 
It shows in particular that the alluvial and colluvial plain offers substatial 
possibilities for agricultural development and grazing provided some measures 
are taken against erosional processes and seasonal flooding phenomena. 
CONCLUSION 
Land capability maps are essentially derived from a detailed geomorphic 
study and a good knowledge of the geology, both types of data being obtained 
through aerial photograph and satellite image interpretation, accompanied by 
ground checking. However, to perform the geomorphic analysis, stereoscopic 
interpretation is compulsory since landsat images cannot be used for this 
purpose. 
Complementary information regarding habitat mapping and current land 
use mapping which provide additional data on the soil characteristics were 
also obtained by means of aerial photograph and satellite image interpretation. 
For the selected test site, it is obvious that the availability of the 
geological map has saved interpretation time and spared ground controls. The 
soil map has also brought a fair contribution. It was noted that the classes 
837 
4 
SS