714
21.4 Methods, Results and Discussion
21.4.1 How Did We Meet Our Criteria for Planning
the Breaching of the Dikes?
As outlined in the previous section, our task was to design dike breaches that would
allow the restoration cells to fully flood, result in maximum sediment deposition,
and minimize wave impact on the new dike landward of the failing dike, etc. To do
this, a data set was assembled as follows.
• K.E. Millard provided us with the 2006 LiDAR-derived digital elevation model
that she had created for the sites for her M.Sc. thesis work (Millard 2008).
• Water level data were recorded at the sites for multiple spring-neap tidal cycles
using a self-logging pressure sensor (Onset Computer Corporation). Data were
relative to the Canadian Geodetic Vertical Datum of 1928 (CGVD28).
• Tide height predictions for Canadian Hydrographic Service (CHS) station 215
(Joggins, NS) relative to Chart Datum were obtained for the period coinciding
with the recorded water level data.
• Aerial photography was acquired for the sites in August 2009 from Nortek
Resource Solutions Inc. with 15 cm pixel resolution.
• Field observations and ground-level photographs were collected in the summer
of 2009.
The measured and predicted water level data were then analyzed to develop a
tidal transformation equation that allowed us to convert predicted tide height at CHS
station 215 to predicted water level at the sites, relative to the restoration cell surfaces and reference marshes, for any time. The resulting data allowed us to calculate
flooding frequencies at various water depths for the proposed restoration cells for a
multi-year period. Maximum flooding depth was over 2 m at peak spring tide.
Sanderson created a two-dimensional wetting-drying model for the site using
standard approaches (similar to Sanderson 2011; reports by Sanderson in Ollerhead
et al. 2010). His hydrodynamic model solved the shallow water equations and
included an algorithm for wetting and drying surfaces as the tide rose and fell. Field
accelerations were computed using a semi-Lagrangian method (McGregor 1993)
with bottom drag calculated implicitly (the drag coefficient parameterizes the magnitude of friction between the water column and the underlying bed). The barotropic
mode was calculated using iterative under-relaxation. A small number of iterations
corresponded to treating all but the shortest waves as being explicit.
Possible openings in the old dike were then simulated and run in the model.
Possibilities explored included different numbers of openings, different widths,
openings with and without notches, and various placements around the restoration
cells. See Fig. 21.3 for an example of a design that was tested but not used. All outputs were evaluated against the design criteria (listed above). If one or more of the
criteria were violated, then the design was discarded. For example, a narrow opening
L.K. Boone et al.
21.4 Methods, Results and Discussion
21.4.1 How Did We Meet Our Criteria for Planning
the Breaching of the Dikes?
As outlined in the previous section, our task was to design dike breaches that would
allow the restoration cells to fully flood, result in maximum sediment deposition,
and minimize wave impact on the new dike landward of the failing dike, etc. To do
this, a data set was assembled as follows.
• K.E. Millard provided us with the 2006 LiDAR-derived digital elevation model
that she had created for the sites for her M.Sc. thesis work (Millard 2008).
• Water level data were recorded at the sites for multiple spring-neap tidal cycles
using a self-logging pressure sensor (Onset Computer Corporation). Data were
relative to the Canadian Geodetic Vertical Datum of 1928 (CGVD28).
• Tide height predictions for Canadian Hydrographic Service (CHS) station 215
(Joggins, NS) relative to Chart Datum were obtained for the period coinciding
with the recorded water level data.
• Aerial photography was acquired for the sites in August 2009 from Nortek
Resource Solutions Inc. with 15 cm pixel resolution.
• Field observations and ground-level photographs were collected in the summer
of 2009.
The measured and predicted water level data were then analyzed to develop a
tidal transformation equation that allowed us to convert predicted tide height at CHS
station 215 to predicted water level at the sites, relative to the restoration cell surfaces and reference marshes, for any time. The resulting data allowed us to calculate
flooding frequencies at various water depths for the proposed restoration cells for a
multi-year period. Maximum flooding depth was over 2 m at peak spring tide.
Sanderson created a two-dimensional wetting-drying model for the site using
standard approaches (similar to Sanderson 2011; reports by Sanderson in Ollerhead
et al. 2010). His hydrodynamic model solved the shallow water equations and
included an algorithm for wetting and drying surfaces as the tide rose and fell. Field
accelerations were computed using a semi-Lagrangian method (McGregor 1993)
with bottom drag calculated implicitly (the drag coefficient parameterizes the magnitude of friction between the water column and the underlying bed). The barotropic
mode was calculated using iterative under-relaxation. A small number of iterations
corresponded to treating all but the shortest waves as being explicit.
Possible openings in the old dike were then simulated and run in the model.
Possibilities explored included different numbers of openings, different widths,
openings with and without notches, and various placements around the restoration
cells. See Fig. 21.3 for an example of a design that was tested but not used. All outputs were evaluated against the design criteria (listed above). If one or more of the
criteria were violated, then the design was discarded. For example, a narrow opening
L.K. Boone et al.