Geoscience Reference
In-Depth Information
weathering regime in the headwaters. In many parts of the desert world, the late Qua-
ternary alluvial records show a pattern of upward-fining alluvial sequences, indicative
of a change from a high-energy and ephemeral or highly seasonal flow regime to a
low-energy and less variable flow regime. More useful in certain contexts is the pres-
ence of freshwater snail shells within the alluvium (Kropelin,
1993
). Provided they
are carefully tested for possible recrystallization, shells within the time range 0-50 ka
are amenable to radiocarbon dating. In addition, the stable carbon and oxygen isotopic
composition of the shell carbonate can throw some light on the type of environment
in which the shells were living (Abell et al.,
1996
; Abell and Hoelzmann,
2000
; Wil-
liams et al.,
2000
; Williams et al.,
2006c
). Analysis of the
87
Sr/
86
Sr ratios can clarify
the source of the water in which the shells were living and can indicate when and
whether certain sub-basins within the main basin were connected to the main drain-
age basin (Talbot et al.,
2000
). This applies equally to alluvial clays (Stanley et al.,
2003
).
However, there are certain limitations involved in using river sediments to recon-
struct past climatic changes in deserts, of which the principal one is the inherently
fragmentary nature of the alluvial record. This limitation can be offset to some extent
by consulting the offshore record (as in the case of the Nile submarine cone), but many
desert rivers fail to reach the coast, and in the more arid areas, the alluvium is subject
to reworking by wind. A final limitation involved in using fluvial evidence to infer past
environmental changes concerns the problems involved in dating alluvial deposits. In
large catchments, deposition of alluvial sediments may often be time-transgressive
(Vita-Finzi,
1973
; Vita-Finzi,
1976
), which means that alluvial terraces upstream
may differ significantly in age from superficially similar terraces downstream. The
solution to this problem is to date the alluvial sequences at a series of sites along
the valley using as many independent dating methods as possible. In this context, it
is worth remembering that charcoal fragments within the alluvium may have been
reworked from previous sediment stores, so the radiocarbon age of the charcoal may
be hundreds or even thousands of years older than the time of deposition of the river
sediment (Blong and Gillespie,
1978
).
In a comprehensive review of process, form and change in arid land rivers, Tooth
(
2000
) illustrates how our understanding of desert river systems is still very limited.
In particular, the influence of major flood events on channel form may mean that
many such systems are in a state of non-equilibrium. In fact, the concept of equi-
librium as applied to rivers in humid areas may not be applicable to desert rivers.
A particular gap in studies of desert rivers identified in Tooth's review concerns
the almost complete lack of integration between short-term process studies and
studies of Quaternary (and older) river history. Tooth concludes with a plea for
devising a stronger theoretical basis in regard to dryland river systems in order to
improve attempts to manage these systems into the future. This requirement applies
