Soil acidification, an important type of land degradation in Rwanda, and more specifically in the
central plateau, is challenging the sustainable use of cropping systems. The aim of this thesis is the
identification of acidification risk areas and hotspots, and gaining insight in the acidification process,
its causes and consequences. To achieve this, the legacy soil database is exploited to reveal
relationships between pH and other environmental variables (land use, topographic position, soil
types, parent materials, development stage, and altitude) and assess the spatial distribution of soil
acidity and acidification risk at national and regional scale.
The soil profile database has proved to be a useful tool to this study because it contains all the
information we needed. We firstly checked the correlations between acidity parameters themselves
and they were found strong (r>0.8) to moderate strong (0.6 For the correlations between acidity parameters and other soil properties, they were very weak
(r<0.4), thus less important to consider. The dependency of acidity parameters on environmental
variables were checked and all of them had a significant effect (P<0.01) on acidity parameters. All
these statistical analyses were carried out using SPSS. Besides the effect of each environmental effect,
except for drainage (P>0.05), the interaction effect between them was checked and was found very
significant as well (P<0.01). Thereafter a classification tree test has been done to predict, in an
hierarchical way, the influence of both soil and environmental parameters considered as independent
variables on soil pH considered here as dependent variable. At the first level was the ACZ; at the
second level, we had the influence of landuse and development stage. Finally the outcome showed the
influence of OC at the last stage. From this we noticed that the diversity of climatic characteristics
(altitude, temperature and rainfall) significantly influences the spatial prediction of soil pH in
Rwanda. The structure of the soil profile database also allowed making maps, using ArcGIS software,
in order to identify risk areas and hotspots whereby the Central Plateau of Rwanda was the ACZ
which behaved as having high risk of soil acidification. The most striking observation in this study is
that the soil pH itself is not the only parameter to take into account while assessing the risk for soil
acidification but also the pH buffer capacity of the soil must be taken into account as well. Finally
after assessing acidity status and identifying risk areas for soil acidification, we provided optimized
management practices and the most important one is liming which aims at raising soil pH to render
Aluminium harmless for plant growth. For this, we only calculated LR for soils having pH lower than
5.5 because dissolved Al3+ ,which is toxic to plants, is most soluble at pH lower than 5.5. In all soil
profiles analyzed during the survey, 67% showed pH lower than 5.5. The lime requirement has been
mapped as well to show area by area where one has to pay too much attention. Soil series occupying
15% of total land surface had LR greater than 5ton/ha and were mostly located in Congo Nile
Watershed Divide, Central Plateau, Eastern Plateau and Buberuka Highlands ACZs.