4.2 AQUEOUS PROCESSES
103
respectively (d 2 > d 1), g., is the acceleration due to gravity, h., is the height of the flow,
and a is the bottom slope. The relationship between the speed of a density flow and
other parameters may be expressed as:
V
2(d2- dl)
dl
g.h.,
where V is the velocity. Derivations and discussions of these equations can be found in
Middleton (1966a,b), Allen (1985a), and Baum and Vail (1998). These formulas show
that the behavior of a turbid flow is governed by the difference in density between it and
the ambient fluid, by the shear stresses of its upper and lower boundary, by its height,
and the angle of slope down which it flows. Additional important factors are whether it
is a steady flow, such as a turbid river entering a lake, or whether it is a unique event of
limited duration as in the case of a liquefied slump.
Turbidity flows have been modeled mathematically by Zeng and Lowe (1997a,b).
They have also been studied experimentally in the laboratory and in present-day lakes,
seas, and oceans. The early experimental studies of Bell (1942) and Kuenen (1937,1948)
showed that when muddy suspensions of sand were suddenly introduced to a flume they
rushed downslope in a turbulent cloud to cover the bottom. Sand settled out first, followed by silt and then clay. Thus beds were deposited with sharp basal contacts that
showed an upward-fining grain-size profile from sand to clay in the space of a few centimeters. Additional experiments by Kuenen (1965) produced laminated and rippled
graded sands from turbid flows in a circular flume. Experiments by Dzulinski and Walton (1963) showed how small turbidity flows could, in the laboratory, generate many
of the erosional features scoured beneath ancient turbidite sands. Experiments such
as these have been criticized because particle fall diameter and fluid viscosity were
not proportional to each other and to the flow size. Subsequently, however, Middleton (1966a,b, 1967) carried out carefully scaled experiments with plastic beads with a
density of 1.52 g/cm 3 and diameters of about 0.18 mm. Suspensions of beads released
into a standing body of water generated graded beds similar to those of less scientific
experiments.
Turning from the laboratory to the outside world, numerous cases of modern turbidity flows are seen. They have been described from lakes, such as Lake Mead, by Gould
(1951) to Norwegian fjords by Holtedahl (1965). In these instances it was possible to
demonstrate direct relationships between inflows of muddy river water and extensive
layered deposits on the floors of the lakes and fjords. The evidence for the existence of
modern marine turbidity flows is equally impressive. A common feature of continental
shelves and delta fronts is that they are incised by submarine valleys. Where these terminate at the base of the slope it is common to find a radiating fan-shaped body of sediment. Submarine telegraph cables that cross these regions tend to be broken rather frequently. Daly (1936) postulated that these submarine canyons were eroded by turbidity
currents, which snapped the cables. One famous and oft-quoted instance of this was the
celebrated Grand Banks earthquake of 1929 (Piper et al., 1999). On 18 November there
was an earthquake with an epicenter at the edge of the Grand Banks off Nova Scotia.
Within the next few hours, 13 submarine telegraph cables were broken on the slopes and
ocean floor at the foot of the Banks. No cables were broken on the Banks themselves.
103
respectively (d 2 > d 1), g., is the acceleration due to gravity, h., is the height of the flow,
and a is the bottom slope. The relationship between the speed of a density flow and
other parameters may be expressed as:
V
2(d2- dl)
dl
g.h.,
where V is the velocity. Derivations and discussions of these equations can be found in
Middleton (1966a,b), Allen (1985a), and Baum and Vail (1998). These formulas show
that the behavior of a turbid flow is governed by the difference in density between it and
the ambient fluid, by the shear stresses of its upper and lower boundary, by its height,
and the angle of slope down which it flows. Additional important factors are whether it
is a steady flow, such as a turbid river entering a lake, or whether it is a unique event of
limited duration as in the case of a liquefied slump.
Turbidity flows have been modeled mathematically by Zeng and Lowe (1997a,b).
They have also been studied experimentally in the laboratory and in present-day lakes,
seas, and oceans. The early experimental studies of Bell (1942) and Kuenen (1937,1948)
showed that when muddy suspensions of sand were suddenly introduced to a flume they
rushed downslope in a turbulent cloud to cover the bottom. Sand settled out first, followed by silt and then clay. Thus beds were deposited with sharp basal contacts that
showed an upward-fining grain-size profile from sand to clay in the space of a few centimeters. Additional experiments by Kuenen (1965) produced laminated and rippled
graded sands from turbid flows in a circular flume. Experiments by Dzulinski and Walton (1963) showed how small turbidity flows could, in the laboratory, generate many
of the erosional features scoured beneath ancient turbidite sands. Experiments such
as these have been criticized because particle fall diameter and fluid viscosity were
not proportional to each other and to the flow size. Subsequently, however, Middleton (1966a,b, 1967) carried out carefully scaled experiments with plastic beads with a
density of 1.52 g/cm 3 and diameters of about 0.18 mm. Suspensions of beads released
into a standing body of water generated graded beds similar to those of less scientific
experiments.
Turning from the laboratory to the outside world, numerous cases of modern turbidity flows are seen. They have been described from lakes, such as Lake Mead, by Gould
(1951) to Norwegian fjords by Holtedahl (1965). In these instances it was possible to
demonstrate direct relationships between inflows of muddy river water and extensive
layered deposits on the floors of the lakes and fjords. The evidence for the existence of
modern marine turbidity flows is equally impressive. A common feature of continental
shelves and delta fronts is that they are incised by submarine valleys. Where these terminate at the base of the slope it is common to find a radiating fan-shaped body of sediment. Submarine telegraph cables that cross these regions tend to be broken rather frequently. Daly (1936) postulated that these submarine canyons were eroded by turbidity
currents, which snapped the cables. One famous and oft-quoted instance of this was the
celebrated Grand Banks earthquake of 1929 (Piper et al., 1999). On 18 November there
was an earthquake with an epicenter at the edge of the Grand Banks off Nova Scotia.
Within the next few hours, 13 submarine telegraph cables were broken on the slopes and
ocean floor at the foot of the Banks. No cables were broken on the Banks themselves.