Late Quaternary palaeohydrological changes in the northern Namib Sand Sea: New chronologies using OSL dating of interdigitated aeolian and water-lain interdune deposits

Abstract

Ephemeral rivers today terminate in playas adjacent to the eastern part of the Namib Sand Sea, although there is evidence that these fluvial systems penetrated further west in the past. There are water-lain sediments exposed in interdune areas of the northern Namib Sand Sea associated with the former course of the Tsondab River, which are central to palaeoenvironmental reconstruction in the region. These deposits indicate that there was increased moisture availability in this presently hyper-arid region and/or the catchments of the rivers to the east of the Sand Sea in the Great Escarpment. Until now the chronology for the deposition of these sediments has been based on radiocarbon ( 14C) dates on carbonate materials. The validity of these ages is problematic and 14C limits the timescale to which these sediments can be dated. The water-lain units are bracketed by aeolian sand units, which are highly suitable for optically-stimulated luminescence (OSL) dating. Here, 24 OSL ages from three sites associated with the former course of the Tsondab River are presented, allowing a reassessment of the timing of sediment deposits and the west to east progression of the terminus of the Tsondab River during the late Quaternary. Narabeb, which is 60 km north-west of the current end point of the Tsondab River at Tsondab Vlei, contains eight mud units, six of which are shown to be of marine oxygen isotope stage (MIS 5, 128–75 ka) age. Water-lain units at Hartmut Pan, 55 km north-west of Tsondab Vlei, are dated by OSL to between 16.9 ± 0.9 ka and 12.7 ± 0.1 ka. At Ancient Tracks, closest to the Tsondab Vlei (6 km west), mud units are younger, and placed between 12.0 ± 0.7 and 12.8 ± 0.8 ka, 11.5 ± 0.5–10.5 ± 0.5 ka and after 10.5 ± 0.5 ka. This new chronology is compared with other key regional records to identify potential regional relationships and forcing factors. There is greatest correspondence with the records in the north of the summer rainfall zone. A revision of both old and young 14C ages on carbonate materials calls for revision of other 14C chronologies in the region and highlights the problems with the continued used of histograms of 14C-ages to indicate periods of increased humidity in southern Africa.