Climate-vegetation dynamics in western tropical Africa: 150,000 years of past environmental change from Lake Bosumtwi (Ghana)

Observations indicate that the extent of the tropical climate zone is more sensitive to global climate change than model predictions suggest and ecological investigations are finding variation within both forest and savannah biomes to be extremely sensitive to climate change. However, the dynamics of how tropical vegetation responds to climate change remains controversial. To explore the relationship between the global climate system and tropical vegetation it is necessary to examine past records of vegetation change through periods of known global climate change, e.g. through glacial-interglacial cycles. The most effective way of examining terrestrial vegetation change is through fossil pollen records from continuous, and chronologically well constrained, lake sediments. However, few fossil pollen lake records exist from the tropics which cover one, or more, complete glacial-interglacial cycle. Pollen records documenting past vegetation change in the lowland tropics of Africa are particularly scarce with only three records from terrestrial lakes which extend beyond the current interglacial published. These are: 1) Bosumtwi (Ghana) which extends back c. 28 thousand calendar years Before Present (kyr BP), 2) Barombi Mbo (Cameroon) covering c. 28 kyr BP, and 3) Ngamakala (Congo) c. 24 kyr BP. Terrestrial vegetation change in tropical Africa has been inferred through glacial-interglacial cycles from pollen analysis on marine cores in the Gulf of Guinea (c. 150 kyr BP) and off the cost of north west Africa (>500 kyr BP; ODP 658). However, interpretation of these data is complicated by the need to understand changes in ocean currents, wind fields and depositional patterns. This paucity of data means that the likely nature of vegetation response to predicted future changes in tropical climate remains essentially unexplored. To shed light on the likely response of tropical vegetation to global climate change and to provide the first data from a terrestrial record in tropical Africa older than 30 kyr I propose to examine fossil pollen from lake sediments raised from Lake Bosumtwi (Ghana) in 2004 at two key time scales and resolutions: 1) to examine vegetation response to major, astronomically forced, global climate change events through the last glacial-interglacial cycle (c. 150 kyr BP) at sub-millennial resolution, and 2) to explore vegetative response to rapid global climate change, e.g. Heinrich events, during the period 30-10 kyr BP at centennial resolution.