Geoscience Reference
In-Depth Information
Table 13.1
Attributes of endogenous desert ephemeral streams and rivers.
Upland, headwater channels
a
Lowland, distal channels
b
Attribute
Small to medium, regional;
<
1
0
-10
3
km
2
Large to subcontinental, 10
5
km
2
System size
Terrain
Mountainous, rifted or block-faulted
Peneplaned and cratonic
Moderate to high; 10
−
3
-10
−
2
Low; 10
−
4
Channel slope
Channel type
Single-thread (often straight), braided
Anastomosed, braided, single-thread
Flow regime
Ephemeral; predominantly upper flow regime
Intermittent and ephemeral;
predominantly lower flow regime
Flood timebase
10
0
-10
2
h
10
2
-10
3
h
Sediment concentration:
(a) Suspended load
(b) Bedload
High to very high; 15-285 g/L
High; maximum recorded: 5 g/L (7 kg/m s at a
boundary shear stress of 36 Pa)
Low to moderate; 0.5-12 g/L
Unknown
Bedforms in sand
Plane bed (predominant); dunes (occasional)
Ripples (20-30 %); dunes (25-45 %);
plane bed (7-15 %)
a
Alexandrov
et al
. (2009); Bull (1979); Dunkerley (1992); Frostick, Reid and Layman (1983); Graf (1983); Hassan (1990b); Karcz (1972);
Laronne and Reid (1993); Leopold, Emmett and Myrick (1966); Meerovitch, Laronne and Reid (1998); Powell, Reid and Laronne (2001); Renard
and Keppel (1966); Schick, Lekach and Hassan (1987); Schumm (1961); Sharma and Murthy (1994); Sneh (1983); Thornes (1977).
b
Bourke and Pickup (1999); Maroulis and Nanson (1996); Nanson and Knighton (1996); Tooth (2000); Tooth and Nanson (2004); Williams
(1971).
1967; Williams, 1971; Sneh, 1983), though some are less
specific (such as Picard and High, 1973).
Before assessing what lessons might be gleaned from
the present in order to interpret ancient desert fluvial sed-
iments, it is essential to remind ourselves that research on
modern arid-zone rivers has been conducted largely in two
distinct morphogenetic provinces. The first (and probably
most visited) might be characterised as headwaters, in-
volving small- to medium-sized catchments. Much of this
research has been conducted in or near orogens or rifts of
Cenozoic age and comes from the American southwest,
the Levant and Maghreb, Iberia, East Africa and north-
west India. The second emanates from cratonic lowlands,
often involving distal reaches of systems of regional to
subcontinental scale. Most of this research has been car-
ried out in Australia, though some is located in southern
Africa. Table 13.1 attempts to summarise key attributes
of the drainage systems in these two provinces. Nanson,
Tooth and Knighton (2002) have issued a strong caution
that the riverine processes and landforms of montane and
rifted terrains might only be of marginal relevance when
characterising and understanding systems in cratonic ter-
rain. It is axiomatic that the reverse is true. It behoves the
sedimentologist, therefore, to contextualise the geomor-
phological arena within which sediments were accumu-
lating, where possible keeping in mind that knowledge
about desert stream sediments is, as yet, in its infancy.
Since the hydraulic processes that entrain a clast are uni-
versal regardless of the environmental setting, the question
teristics that can be used to distinguish ephemeral from
perennial stream deposits. Some doubt has been cast (e.g.
North and Taylor, 1996; Martin, 2000). However, we con-
tend that there appear to be at least three attributes that
might especially lead a sedimentologist to conclude that
an ancient sedimentary sequence was laid down by an
ephemeral stream.
13.5.1
Thin beds
There is no quantification that can be used to justify a
claim that desert river deposits are composed charac-
teristically of thin beds, i.e. 0.1-0.3 m thick. However,
deposits of widely differing age - Devonian (Tunbridge,
1984) (Figure 13.26), Triassic (Frostick
et al
., 1988; Reid,
Linsey and Frostick, 1989) and Holocene (Frostick and
Reid, 1986) - undoubtedly have an easily recognis-
able affinity because of the nature of their bedding
(Figure 13.27). There is plenty of evidence of the mi-
nor incision that is associated with scour in the nested
fill-sets of ancient deposits and the range in bed thickness
is consistent with the depth of scour and fill as defined by
field experiments in modern sand-bed ephemeral streams
(Leopold, Emmett and Myrick, 1966; Powell
et al
., 2006)
(Figures 13.17 and 13.27). Modern gravel-bed ephemerals
also exhibit similar thicknesses of scour and fill (Laronne
and Shlomi, 2007; Lekach, Amit and Grodek, 2009).
Hassan, Marren and Schwartz (2009) (Figure 13.28) de-
