Speciation of selected trace elements in three Ethiopian Rift Valley Lakes (Koka, Ziway, and Awassa) and their major inflows

The Ethiopian Rift Valley Lakes (ERVLs) are water resources which have considerable environmental, economic and cultural importance. However, there is an increasing concern that increasing human activities around these lakes and their main inflows can result in increased contamination of these water bodies. Information on total concentrations of some trace elements is available for these lakes and their inflows; however, data on the trace element speciation is lacking. Therefore, the objective of this study was to determine the low molecular mass (LMM) trace element species and also, evaluate the influence of flooding episodes on the LMM trace element fractions. At-site size and charge fractionation system was used for sampling of water from the lakes Koka, Ziway and Awassa and their main inflows during the dry and wet seasons. The results showed that chromium (Cr), manganese (Mn), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), and lead (Pb) in Lake Koka and its inflows as well as in Lake Ziway were predominantly present as HMM (high molecular mass, i.e., > 10 kDa) forms, while arsenic (As), selenium (Se), cadmium (Cd) were more mobile during the dry season. In Lake Awassa, all except Cr and Mn were predominantly found as LMM species (low molecular mass, i.e. < 10 kDa) which can be attributed to the high concentrations of LMM DOC (dissolved organic carbon). During the wet season, results from the Lake Koka and its inflows showed that all trace elements were predominantly associated with HMM forms such as colloids and particles, demonstrating that the mobility of elements was reduced during the wet season. The colloidal fraction of elements such as Cr, Ni, and Cd was also correlated with dissolved Fe. As the concentration of LMM trace element species are very low, the mobility, biological uptake and the potential environmental impact should be low.

► Relevance of using at-site fractionation for trace element speciation demonstrated.
► Some elements > WHO limit, but speciation indicated low risk in dry and wet seasons.
► Multivariate statistics used to identify factors controlling element speciation.
► Variation in total Fe and TOC explained significantly the size distribution pattern.
► Variation in conductivity and TOC explained significantly the charge distribution.