Geology Reference
In-Depth Information
11.6.4 Sinuosity
given the indeterminate division of the total tidal prism
between channels in a network:
A relationship exists between the length of meanders
and the channel width (Fig. 13, Marani et al.
2002,
2004
; Dalrymple and Choi
2007
) . This relationship
holds for all meandering channels from fl uvial to tidal,
including salt marsh and tidal fl ats channels, and chan-
nels within estuaries and deltas (Marani et al.
2002
;
Seminara
2006
; Hood
2010
). Salt marsh channels do
not form a distinct population in terms of meander-
to-width geometry as they do for width to depth ratio
(Fig.
11.13
, D'Alpaos et al.
2005
) . This is consistent
with the observations that marsh creeks, which tend to
be narrower, exhibit tighter meanders than channels
over tidal fl ats (Figs.
11.2
and
11.13
) and implies that
depth does not signifi cantly infl uence meander width.
Ωα
Q
c
(11.3)
where, based on observations in 242 cross sections,
c
,
the exponent of
Q,
falls within the range 0.73-1.34,
with an average of 0.96 (i.e. ~1; Friedrichs
1995
) .
The equilibrium theory requires that the peak dis-
charge (
Q
), and thus the peak velocity (
U = Q/A
)
produces a 'stability' shear stress, t
s
, which controls
the sediment transport within the channel. The t
s
will be just greater than the critical shear stress, t
c
,
required for initiation of sediment movement; and
based on laboratory experiments t
c
< t
s
<
0.15 t
c
.
(Diplas
1990
, from Friedrichs
1995
) . However, this
theory is complicated by the variation of sediment
type through tidal systems and it would stand to rea-
son that t
c
in marsh channels will differ to that in
channels on tidal fl ats because of the difference in
grain size, organic content and level of vegetative
stabilization. Lastly, application of this theory is
complicated further by lateral friction, which will
vary with hydraulic radius, itself a function of chan-
nel width and shape; with sediment type, organic
content or biomass; and with the variation of the
hydraulic radius over the tidal cycle (i.e., the ratio of
mean water depth to tidal range).
Further studies have explored the idea that the area,
A
, which a certain channel drains (sometimes called
the
creekshed
) is representative of the volume of water
which fl ows through it. Based on this, an alternative
relationship may be used (Fagherazzi et al.
1999
;
Rinaldo et al.
1999
) :
11.6.5 Stream Order and Drainage Density
Pestrong (
1965
) observed that, unlike fl uvial systems,
neither drainage basin area, nor channel lengths and
widths, scaled with stream order. Knighton et al. (
1992
)
found closer agreement to fl uvial behavior in channels
in the Van Diemen Gulf (Australia), and Novakowski
et al. (
2004
) concluded that tidal networks in South
Carolina, USA were similar but more elongate than
fl uvial networks. The disagreement in these observa-
tions may be explained by variation in scaling from
basin to basin that can be observed in tidal fl ats and salt
marshes (Rinaldo et al.
1999
; Fagherazzi et al.
1999
) .
Within networks, the drainage density is defi ned as
the ratio of total channel length (S
l
) divided by the
watershed area (
A
). This parameter, which provides a
measure of channelization, was examined for tidal
channel networks within salt marshes by Marani et al.
(
2003
). The study considers 136 creeksheds within the
Venice Lagoon, Italy, and makes several poignant
observations; fi rstly the probability distribution of
length of pathways across the unchanneled surface fol-
low an exponential decay, similar to that seen in fl uvial
networks. As with the variation of width along channel
(e-folding lengths), different decay rates were seen
within individual basins. Secondly, a linear relation-
ship exists between total channel length (in any creek-
shed) and tidal prism (for that basin). A similar
correlation, although not specifi ed as linear, was found
by Allen (
1997
). The exact relationship varies between
d
(11.4)
Ωα
A
where
d
is of the order ~1.
This relationship has recently been explored even
further using numerical models of hydro- and morpho-
dynamics and successfully used to represent the evolu-
tion of tidal networks (D'Alpaos et al.
2005,
2010
) .
The validity of the assumption of dynamic equilib-
rium is supported by early observations by Steers
(
1969
) that headward-eroding marsh creeks exhibit a
gradient (albeit low) along the channel bed. This gradient
becomes zero once the creek has stopped extending
and reached equilibrium with its local tidal prism or
drainage area. This suggests further that headward ero-
sion of previously stable creeks is indicative of an
increased tidal prism (Hughes et al.
2009
) .
