Geoscience Reference
In-Depth Information
in the past (see, for example, Servant and Servant-Vidary,
1980; Roberts and Wright, 1993; Metcalfe
et al.
, 1994;
Drake
et al.
, 2008; Burrough, Thomas and Bailey, 2009).
In the southwestern United States, presently hyper-arid
to semi-arid, evidence for more humid conditions dur-
ing the last Pleistocene is found in the tectonically con-
trolled 'basin and range' physiographic province (Smith
and Street-Perrott, 1983). Shoreline and sedimentary fea-
tures show that many individual basins were once occu-
pied by lakes, some of which were linked together to form
huge palaeolakes such as Bonneville (51 700 km
2
in area)
and Lahontan (22 442 km
2
) (see Chapter 4).
Palaeolake Mega Chad in West Africa (Grove and
Warren, 1968) and megalake Makgadikgadi in Botswana
(Grey and Cooke, 1977; Burrough, Thomas and Bailey,
2009) were very extensive (350 000 km
2
and 66 000 km
2
,
respectively) and relatively shallow at their maximum ex-
tents. Their limits are defined by relict shorelines that can
occur as simple single beach ridges or in groups (beach
ridge plains or strand plains) interspersed by swales (nar-
row, shallow trough-like depressions), such as the Dautsa
ridge complex, Lake Ngami (Shaw, 1985). These land-
forms are highly dependent on the individual hydrological
system and can be built in a single event or can exist as
composite features accumulating sediment over multiple
lake high stands in response to climate reorganisations
at both the glacial (Rognon and Williams, 1977; Street,
1981) and sub-Milankovitch scale (Burrough
et al.
, 2007;
Thomas
et al.
, 2009). While these geoproxies can be used
to identify positive hydrological excursions within a basin,
beach ridges are also subject to erosion and migration in
response to changing sediment and water supply, so that
gaps in the record of beach ridge accretion do not neces-
sarily equate to times of low levels and could instead be a
function of differential preservation.
In Australia, groundwater
and
surface-water fluctua-
tions have been identified as important controls of lake-
basin development (Bowler, 1986). The presence of la-
custrine features and fringing transverse or lunette dunes
in lake basins is not necessarily indicative of distinct pe-
riods of positive and negative water balance, for seasonal
climatic regimes can result in the contemporaneous de-
velopment of such landforms (Bowler, 1976, 1978). Al-
though fluctuating lake levels produced active fringing
lunettes during the last glacial period, the major deflation
of dry lakebeds was broadly coincident with the LGM in
the arid interior of Australia (Bowler, 1998; Bowler and
Price, 1998; Magee and Miller, 1998), implying highly
seasonal or arid conditions.
In north and east Africa and the interior of Aus-
tralia, many lake basins were dry during the LGM, sup-
time (Rognon and Williams, 1977). In contrast, the lake
basins in the southwestern United States preserve evi-
dence that indicates that enhanced moisture availability
characterised these areas during much of the Late Pleis-
tocene due, with varying temporal significance, to in-
creased precipitation and/or ice sheet meltwater inputs.
During the early to mid-Holocene lacustrine phases have
been documented in north Africa. Together with other
offshore marine core evidence that suggests an abrupt
change to a wet Sahara, this has given rise to the term
the 'African Humid Period' (DeMenocal
et al.
, 2000), be-
lieved to be controlled by nonlinear feedback responses
to monsoon dynamics, ultimately driven by precession-
controlled changes in solar radiation receipt. Evidence
that this period of desert contraction extended into Ara-
bia has also been established using lake basin and sedi-
mentary evidence (McClure, 1976; Parker
et al.
, 2006),
though there are regional differences in the timings of
lake rises and falls (Street, 1981). With improved dat-
ing and analysis of proxy records, it has been shown that
many lacustrine basins have responded to hydrological
changes more frequently and with greater magnitude than
previously recognised (e.g. Scholz
et al.
, 2007; Thomas
et al.
, 2009; Burrough, Thomas and Bailey, 2009), their
response reflecting not only the basin sensitivity to hy-
drological change but also the background sensitivity of
the environment to abrupt regional or global scale climate
shifts.
3.2.9
Ecological evidence
The oxidising conditions of contemporary arid zones re-
sults in a paucity of well-preserved organic records so that
ecological proxy evidence, commonly utilised in temper-
ate humid environments, is often lacking.
Vegetation change in response to changes in humid-
ity, temperature and CO
2
levels are, however, commonly
found in archives of pollen assemblages both outside the
desert belt (Kiage and Liu, 2006) and laid down within off-
shore ocean cores. In addition, novel palaeoenvironmen-
tal archives such as speleothems (e.g. Brook, Burney and
Cowart, 1990; Holmgren, Carless and Shaw, 1995) and
fossilised rock hyrax middens (Scott, Marais and Brook,
2004; Chase
et al.
, 2009) facilitate pollen preservation
even under arid conditions, and can be used to provide
information on changing environmental conditions and
the contraction and expansion of the arid zone during the
Quaternary.
In addition to pollen itself, analyses of other, more per-
sistent, vegetation-based biomarkers have provided an ad-
