Sediment and pollen evidence for an early to mid-Holocene humid period in the eastern Sahara

LATE QUATERNARY HISTORY OF VEGETATION AND CLIMATE OF TROPICAL NORTH AFRICAThe critical examination o f available pollen data from the vegetation of the Sahara allows one to conclude that this vegetation has gone through but few qualitative changes during the last twenty thousand years. In particular, one notices an extension in the Sahara of tropical Sahel taxa about the middle of Holocene. Quantitatively, some pollen and geological data converge to Show that the Saharian plains were extremely arid between about 20 000 and 15 000 years BP and that on the mountains the vegetation became very sparse. A new colonization began on the mountains about 15 000 years ago.The pollen study of Holocene sediments from the central part o f the Chad basin was done in the Tjéri station. The results of this study exhibit a major change near 7 000 years BP, characterized in the Sahel zone by a dramatic extension o f arboreal taxa until about 5 000 years BP, probably corresponding to northward extension of the sahel savanna. One important change took place also at the same time in the wet north tropical zone where, between about 7 000 and 4 000 years BP, there occurred an extension of taxa growing presently on the well-leached soils of interfluves.Such a change near 7 000 years BP also appears in the available stratigraphical, sedimentological and pedological data from tropical north Africa. One observes particularly that, between 15 000 and 7 000 years BP, the

During Pleistocene pluvial precipitation was sufficient for the maintenance of groundwater supported lakes and for the accumulation of playa lakes in wind-scoured depressions during the early Holocene pluvial. At places where ground water reaches near to the surface, wells (birs) have been dug and maintained in historic times. These birs have been used as sampling sites for water analyses, including carbon-14 levels, carbon and oxygen stable isotope ratios, tritium concentrations, and chemical data. All the waters from birs analyzed to date produced apparent radiocarbon ages ranging from late historic to early Holocene, and tritium analyses on some of these indicate no recharge during the Atomic age. Sources of error for the radiocarbon analyses, including exchange with atmospheric CO 2 , respiration by plant roots, and contact with carbonates of considerably older age, were evaluated. None of these factors have such an extreme impact on the measurements as to render the result invalid. Two trends revealed by these data are an increase in apparent age from northwest to southwest and with subsequent extractions at the same site where the hand-dug well was bailed out and sampled four times within two days. We conclude that recharge of shallow ground water occurred in early Holocene time, and some recharge of deeper aquifers may have occurred where infiltration paths permitted. Some recharge occurred in late Holocene (post pluvial) time, but the net trend has been toward hyper-aridity that characterizes the area today.