"Late Quaternary palaeoenvironments and sedimentology of the Magadi-Natron rift basins, Kenya-Tanzania"

Lakes Magadi, Natron and Nasikie Engida, which lie in the rift valley of southern Kenya and northern Tanzania represent three differing responses to evaporation of closed saline water bodies with no outflow. Lake Magadi is dry for much of the year with exposed sodium carbonate (trona), but is flooded seasonally. Lake Natron fluctuates annually in area and salinity but normally has surface water. Nasikie Engida is fed by perennial hot-spring inflow and maintains a shallow saline lake all year. All three lakes have varied in water depth throughout the last one million years as a result of climate change and tectonics, periodically becoming fresher, deeper, and occasionally uniting as a single large lake.

The sediments include several unusual deposits with uncommon minerals. During fresher periods, limestones have formed. Trona precipitated during dry periods and forms extensive deposits today in all three lakes. Siliceous gels have accumulated in parts of the basins near sites of hot-spring inflow. The Pleistocene deposits of each of these lakes contain abundant sodium silicate minerals such as magadiite and abundant quartzose cherts. Although most of these cherts were originally attributed to alteration of a magadiite precursor, a recent re-evaluation has suggested that some might have had a microbiological origin, or were precipitated from silica-rich hot spring waters on periodically emergent mudflats.

This project will revisit questions concerning the origins of sediments in the Magadi- Natron Basins through a systematic study of the ancient and modern deposits and a renewed investigation of the modern lake brines, rivers, springs and groundwater inflows. We will provide field descriptions of new sediment sections and return samples for chemical, sedimentological and microfossil analyses that will improve our understanding of the past environments in which those deposits formed, and refine the geological history for the region with a better constrained chronology of geological events. We will also obtain short cores from the saline lakes and recover modern sediment samples to enhance our understanding of recent sedimentary and early diagenetic processes.

The resulting data will be used to develop models of chemical and biochemical sedimentation in an arid rift basin through time, which will assist in reconstructing the geological history of the Magadi-Natron region, and in assessing the origins of analogous ancient sediments elsewhere (e.g., Eocene Green River Formation, USA). In particular, the new research will help to explain the processes involved in the formation of trona, magadiite and non-marine cherts in general.