Geoscience Reference
In-Depth Information
Williams,
1984
; De Deckker and Williams,
1993
; Wendorf et al.,
1993
). Gaven et al.
(
1981
) and Petit-Maire (
1982
) considered Pleistocene Lake Shati in south-east Libya
to be about 130 ka in age, that is, Marine Isotope Stage 5 (MIS 5). However, Williams
(
1984
, p. 440) pointed out that the uranium-series (U-series) ages obtained from
this site by Gaven (
1982
)on
Cerastoderma glaucum
shells showing little or no
recrystallization (Icole,
1982
) in fact fall into four distinct groups: 173-158, 136-132,
around 90 and 40
2 ka, suggesting that there were four lake phases and not just the
single episode inferred by Petit-Maire (
1982
)andGavenetal.(
1982
).
Dating lake carbonates is seldom a straightforward exercise. Causse et al. (
1988
)
corrected for the effects of detrital thorium and obtained U-series ages of 100-80 ka
for lake sediments in the west Sahara that were thought to belong to the last major wet
phase in that area, regarded until then as early Holocene. Szabo et al. (
1995
) reported
U-series ages for lake carbonates from Bir Sahara and Bir Tarfawi in the Western
Desert of Egypt and other Pleistocene lakes in the eastern Sahara. They recognised
five discrete lake phases dated to around 320-250, 240-190, 155-120, 90-65 and 10-
5 ka. Crombie et al. (
1997
) obtained U-series ages on travertines from Kurkur Oasis
in the Western Desert of Egypt that fell into three broad groups:
±
260, 220-191 and
160-70 ka. However, Wendorf et al. (
1993
, pp. 552-573) used a variety of dating
methods to date a series of middle Pleistocene lakes at Bir Sahara and Bir Tarfawi
associated with Acheulian and middle Palaeolithic artefacts, including luminescence
(TL and OSL), uranium-series, amino acid racemisation and electron spin resonance.
Only the OSL ages yielded stratigraphically consistent results, with ages between 175
and 80-70 ka for the various lakes and associated Middle Palaeolithic sites.
In the presently hyper-arid Murzuq Basin of southern Libya, at least four large
lakes were present between 500 and 100 ka, with U-series ages of around 415 ka
(MIS 11), 320-300 ka (probably MIS 9), 285-205 ka (MIS 7) and 138-128 Ka (MIS
5e) (Geyh and Thiedig,
2008
). The lakes became progressively smaller during each
successive wet phase, indicating progressively less humid interglacial conditions from
MIS 11 onwards. Maxwell et al. (
2010
) identified other Middle to Late Pleistocene
lakes immediately west of the Saharan Nile that were fed in part by overflow from the
Nile.
In the Kenya Rift, Lake Naivasha showed three high lake level episodes between
175 and 60 ka (Bergner and Trauth,
2004
). Before that time, during the late Pliocene
and Early to Middle Pleistocene, Trauth et al. (
2010
) believed that the lakes of
the Kenya Rift acted as amplifier lakes (see
Chapter 11
), showing an exaggerated
response to minor changes in precessional forcing. The reason for this lies in the
particular geomorphic setting of these lakes, which are located in low-lying sites with
very high rates of evaporation but are fed from elevated catchments with high rates of
precipitation.
Just as the occurrence of high lake levels may denote past intervals of higher
rainfall, so too can the drying out of former lakes be used to reconstruct past episodes
>
