Abstract:
Schistosomiasis is the most widespread water-based disease in sub-Saharan Africa. Transmission is governed by the spatial distribution of freshwater snails that act as inter- mediate
hosts and human water contact patterns. The aim of this study was to identify factors that suites
freshwater snails which are an intermediate hosts of schistosomiasis in Omo-Gibe basin,
southwest Ethiopia. A cross-sectional study design was employed with one hundred-thirty (130)
sampling sites in the study areas were selected. Ordination analysis by applying CANOCO for
windows version 4.5 software was employed and since the gradient of environmental variable
was greater than three (> 3SD), canonical correspondence analysis (CCA) was used for
ordination plot. Classification and regression tree models (CART) was applied to develop the
models. The classification tree models were built using the J48 algorithm a java reimplementation of the C4.5 algorithm in WEKA. Regression tree models were built using the M5
algorithm in WEKA and in order to relate the abundance of snail intermediate host to habitat
and water quality variables. The ordination analysis indicated that pool habitat condition favors
the abundance of most of the snail species such as Lymnaea natalensis, Biomphalaria sudanica,
Bulinus globosus, and Biomphalaria pfeifferi. The classification tree also demonstrated that
presence absence of predators was the most prominent biotic variable influencing the presence
or absence of freshwater snails. This variable was selected in 75% of the classification tree
models. Canopy cover, habitat type and pH were the most important variables determining the
presence or absence of freshwater snail species. In addition the conditional analyses clearly
indicated that the abundance of most of the snail species was decreased when the abundance of
competitors’ macroinvertebrates namely Heptageniidae and Helodidae abundance was greater
than 40 individuals
