Geoscience Reference
In-Depth Information
40
8
30
6
Shear stress
C
20
4
L
2
10
R
0
0
0
5
10
15
20
25
30
35
Time, minutes
Figure 13.25
Unit bedload flux and channel average boundary shear stress during the rising limb of a flash flood of 8 February
1991 in the gravel bed at Nahal Yatir, northern Negev, Israel, showing the sympathetic response of bedload to increasing hydraulic
stress and the impact of sidewall drag on sediment flux. Birkbeck-type samplers were located at: C, channel centreline; L, leftward
half-channel; R, rightward half-channel (after Powell
et al
., 1998).
floods of modest duration will be finer than those laid
down by a spectrum of floods that includes larger flow
durations and flow magnitudes that ensure some periods
when there is equal mobility of all the grain sizes available
for transport.
Because of its relative importance, bedload cannot be
ignored as a component of sediment yield in arid environ-
ments in the same way that it often and conveniently is in
more humid environments.
bed, so reducing armour development and permitting
high
bedload
transport
rates
through
an
absence
of
protection (Dietrich
et al
., 1989; Laronne
et al
., 1994).
The virtually unique bedload data sets of the Nahal
Yatir and Nahal Eshtemoa have also provided conclusive
evidence that settles a controversy that has exercised river
scientists for several decades. It has long been known that
sands and granules may be moved at a comparatively low
transport stage that just exceeds the shear stress associ-
ated with initial entrainment of bedload. There have been
flume experiments that show that increasing shear stress
brings progressively larger grains into motion until all
grain sizes of the bed material are mobile. One data set,
for a perennial stream, has provided a hint that this con-
dition of 'equal mobility' is approached as discharge in-
creases, but, unfortunately, the maximum measured trans-
port stage fell short of achieving this condition. Using the
Eshtemoa bedload flux data, Powell, Reid and Laronne
(2001) have shown that equal mobility of all grain sizes
is achieved when boundary shear stresses rise above four-
and-a-half times that associated with initial entrainment.
The coarseness of bedload sediment transported by each
flash flood is directly related to the duration of flood flow
that exceeds this threshold. This has implications for the
coarseness of sediment transported into water reservoirs,
all of which provide a vital resource in drylands and all
of which have comparatively small half-lives because of
sedimentation (Syvitski, 2003). The spectrum of flash-
flood magnitude therefore controls the growth and nature
of reservoir deltas through its impact on the calibre of
the bedload. It also carries implications for the grain-size
distribution of ancient desert fluvial sediments. Deposits
13.5
Desert river deposits
It is perhaps curious that much of what has been written
about the
dynamic
behaviour of desert streams is spec-
ulative and comes from deductions about
static
fluvial
deposits. Perhaps even more curious is that there is more
written about the dynamics of desert sediments that may
be as much as 400 million years old (see, for example,
Allen, 1964; Schumm, 1968a; Tunbridge, 1984; Olsen,
1987; Frostick
et al
., 1988; Frostick, Linsey and Reid,
1992; North and Taylor, 1996) than about those of mod-
ern streams, for which there is more information about
the likely character of flows that were responsible, the na-
ture of the channel and so on. This might reflect Schumm's
(1968b) interesting observation that, prior to development
of higher land plants during the Devonian, all fluvial sed-
iments were derived in landscapes akin to deserts, regard-
less of how much rainfall they received and where they
were in the world. There are, however, a few studies of
modern flash-flood deposits. They are often associated
with high-magnitude events that have achieved local no-
