Planar Beach: Difference between revisions

From CIRPwiki
Jump to navigation Jump to search
No edit summary
No edit summary
 
(4 intermediate revisions by the same user not shown)
Line 24: Line 24:
a concrete bed and Case 7 was run over a thin 0.005-0.01 m layer of
a concrete bed and Case 7 was run over a thin 0.005-0.01 m layer of
gravel grouted onto the concrete floor. A summary of the wave
gravel grouted onto the concrete floor. A summary of the wave
conditions is provided in Table 21.
conditions is provided in Table 1.


{|border="1"
{|border="1"
|+ Table 21. Wave conditions for the Visser (1991) test cases.
|+ Table 1. Wave conditions for the Visser (1991) test cases.
!Parameter          !! Value
!Parameter          !! Value
|-
|-
Line 39: Line 39:
= Model Setup =
= Model Setup =


The computational grid (Figure 22) consists of 84 rows and 147 columns
The computational grid (Figure 1) consists of 84 rows and 147 columns
with a constant grid resolution in the longshore direction of 0.15 m
with a constant grid resolution in the longshore direction of 0.15 m
and a variable grid resolution between 0.04 and 0.15 m in the
and a variable grid resolution between 0.04 and 0.15 m in the
Line 57: Line 57:
and the roller surface stresses were then added to the wave radiation
and the roller surface stresses were then added to the wave radiation
stresses before running CMS-Flow. A summary of the important
stresses before running CMS-Flow. A summary of the important
simulation settings for CMS-Flow and CMS-Wave is given in Tables 22
simulation settings for CMS-Flow and CMS-Wave is given in Tables 2
and 23 respectively. The experiments was simulated in laboratory
and 3 respectively. The experiments was simulated in laboratory
scale, which is why some of the parameters like the wetting/drying
scale, which is why some of the parameters like the wetting/drying
depth were decreased.
depth were decreased.




[[File:Visser_test_grid.png ||thumb|right|600px|Figure 1. CMS computational grid for the Visser (1991) test cases.]]
[[File:Visser_test_grid.png |thumb|right|400px|Figure 1. CMS computational grid for the Visser (1991) test cases.]]




{|border="1"
{|border="1"
|+ Table 22. CMS-Flow settings for the Visser (1991) test cases.
|+ Table 2. CMS-Flow settings for the Visser (1991) test cases.
!Parameter                  !! Value
!Parameter                  !! Value
|-
|-
Line 84: Line 84:


{|border="1"
{|border="1"
|+ Table 23. CMS-Wave settings for the Visser (1991) test cases.
|+ Table 3. CMS-Wave settings for the Visser (1991) test cases.
!Parameter              !! Value
!Parameter              !! Value
|-
|-
Line 110: Line 110:


{|border="1"
{|border="1"
|+ Table 24. Calibration parameters for the Visser (1991) test cases.
|+ Table 4. Calibration parameters for the Visser (1991) test cases.
!Parameter          !! Case 4    !!    Case 7  !!    Default
!Parameter          !! Case 4    !!    Case 7  !!    Default
|-
|-
Line 130: Line 130:


The measured and computed wave heights, longshore currents, and water
The measured and computed wave heights, longshore currents, and water
levels for Case 4 are compared in Figure 23. Results are shown with
levels for Case 4 are compared in Figure 2. Results are shown with
and without the surface roller. The results are significantly improved
and without the surface roller. The results are significantly improved
when the surface roller is included as demonstrated by the
when the surface roller is included as demonstrated by the
goodness-of-fit statistics shown in Table 25. The NMAE for longshore
goodness-of-fit statistics shown in Table 5. The NMAE for longshore
current was reduced from approximately 20 to 5%. The roller has the
current was reduced from approximately 20 to 5%. The roller has the
effect of spreading the peak longshore current and moving it closer to
effect of spreading the peak longshore current and moving it closer to
Line 144: Line 144:
current-wave interaction is relatively weak in the wave model.
current-wave interaction is relatively weak in the wave model.


[[File:Visser_Wave_Heights_Case4.png ||leftthumb|400px|alt=framework]]
[[File:Visser_Wave_Heights_Case4.png |thumb|right|400px|Figure 2. Measured and calculated wave height (top), longshore
Figure 23. Measured and calculated wave height (top), longshore
current (middle), and water level (bottom) for Visser (1991) Case 4.]]
current (middle), and water level (bottom) for Visser (1991) Case 4.
 




{|border="1"
{|border="1"
|+ Table 25. Goodness-of-fit statistics for the Visser (1991) Case 4.
|+ Table 5. Goodness-of-fit statistics for the Visser (1991) Case 4.
!            !!              !!  NRMSE, %  !!NMAE,%!!R2    !!Bias
!            !!              !!  NRMSE, %  !!NMAE,%!!R2    !!Bias
|-
|-
Line 170: Line 170:
The measured and computed wave heights, longshore currents, and water
The measured and computed wave heights, longshore currents, and water
levels with and without the roller for Case 7 are shown in
levels with and without the roller for Case 7 are shown in
Figure 24. It is interesting to note that although the offshore wave
Figure 3. It is interesting to note that although the offshore wave
height, period and direction are the same as Case 4, the location of
height, period and direction are the same as Case 4, the location of
the breaker for Case 7 is significantly further offshore. It is
the breaker for Case 7 is significantly further offshore. It is
Line 184: Line 184:




[[File:Visser_Wave_Heights_Case7.png ||leftthumb|400px|alt=framework]]
[[File:Visser_Wave_Heights_Case7.png |thumb|right|400px|Figure 3. Measured and computed longshore currents (top), water
Figure 24. Measured and computed longshore currents (top), water
levels (middle) and wave heights (bottom) for Visser (1991) Case 7.]]
levels (middle) and wave heights (bottom) for Visser (1991) Case 7.
 




{|border="1"
{|border="1"
|+ Table 26. Goodness-of-fit statistics for the Visser (1991) Case 7.
|+ Table 6. Goodness-of-fit statistics for the Visser (1991) Case 7.
!!Roller!!Variable      !!NRMSE, %    !!NMAE, %  !!<math>R^2</math>!!Bias
!!Roller!!Variable      !!NRMSE, %    !!NMAE, %  !!<math>R^2</math>!!Bias
|-       
|-       
Line 222: Line 222:
the comparison between measured and simulated longshore current
the comparison between measured and simulated longshore current
velocities and water levels under strong wave forcing.
velocities and water levels under strong wave forcing.
= References =
* Fredsoe, J. (1984). “Turbulent boundary layer in wave-current motion,” Journal of Hydraulic Engineering, ASCE, 110, 1103-1120.
* Stive, M.J.F., and De Vriend, H.J. (1994). “Shear stress and mean flow in shoaling and breaking waves,” in Proceedings 24th International Coastal Engineering Conference, ASCE, New York, 594-608 pp.
* Visser, R. J. (1991). “Laboratory measurements of uniform longshore currents”. Coastal Engineering, 15, 563-593.
----
[[Test_Cases]]

Latest revision as of 21:13, 17 April 2014

Test C2-Ex3: Planar sloping beach with oblique incident regular waves

Purpose

The CMS is applied to a laboratory experiment of wave-induced currents and water levels due to regular waves. The large cross-shore gradient of wave height in the surf zone produces a large forcing useful for testing hydrodynamic model stability and performance under strong wave forcing. The specific CMS-Flow features tested are the surface roller, cross-shore boundary conditions, and combined wave-current bottom shear stress parameterization.


Experiment

In 1991, Visser conducted eight laboratory experiments of monochromatic waves on a planar beach and collected measurements on waves, currents and water levels. In this report, experiments (Cases) 4 and 7 are selected as representative test cases. The bathymetry consisted of a 1:10 slope for the first 1 m from shore, a 1:20 slope for the next 5 m, followed by 5.9-m flat bottom to the wave generator. Cases 4 and 7 had an incident wave height of 0.078 m, peak period of 1.02 s and incident wave angle of 15.4°. Case 4 was run over a concrete bed and Case 7 was run over a thin 0.005-0.01 m layer of gravel grouted onto the concrete floor. A summary of the wave conditions is provided in Table 1.

Table 1. Wave conditions for the Visser (1991) test cases.
Parameter Value
Wave height (regular) 0.078 m
Wave period 1.02 s
Incident wave angle 15.4º

Model Setup

The computational grid (Figure 1) consists of 84 rows and 147 columns with a constant grid resolution in the longshore direction of 0.15 m and a variable grid resolution between 0.04 and 0.15 m in the cross-shore direction. A constant zero water level was forced at the offshore boundary and cross-shore boundaries were applied on each side of the shoreline. The boundary type solves the 1-D cross-shore momentum equations for the longshore current and water level and applies a flux boundary condition for inflow conditions and a water level condition for outflow conditions. The combined wave-current bottom shear stress model of Fredsoe (1984) is used. The cases were simulated as steady-state solutions with pseudo-time stepping to reach steady-state while coupling waves, currents and water levels. The initial condition was specified as zero current velocity and water level for the whole domain. Waves and hydrodynamics were coupled every 20 min (steering interval) and run until steady-state. The surface roller model (Stive and De Vriend 1994)was run after each CMS-Wave run and the roller surface stresses were then added to the wave radiation stresses before running CMS-Flow. A summary of the important simulation settings for CMS-Flow and CMS-Wave is given in Tables 2 and 3 respectively. The experiments was simulated in laboratory scale, which is why some of the parameters like the wetting/drying depth were decreased.


Figure 1. CMS computational grid for the Visser (1991) test cases.


Table 2. CMS-Flow settings for the Visser (1991) test cases.
Parameter Value
Solution scheme Implicit
Time step 1 min
Wetting/drying depth 0.006 m
Simulation duration 3 hr
Ramp duration 2 hr
Wave-current bottom friction Fredsoe (1984)
Table 3. CMS-Wave settings for the Visser (1991) test cases.
Parameter Value
Wave breaking formulation Battjes and Janssen (1981)
Bottom friction Off (default)
Steering interval 20 min

All of the wave breaking formulations in CMS-Wave are designed for random waves. However the Visser (1991) laboratory experiments were run with regular (monochromatic) waves which are not useful for validating the CMS-Wave. Since the objective of this test case was to assess the performance of the hydrodynamics, it was necessary to calibrate the waves to obtain the most accurate wave results in order to analyze the performance of the hydrodynamic model by itself and not have the analysis impacted by the results from an inadequately calibrated wave model. The calibration procedure consisted of first calibrating the location of the breaker using the breaker index γ. The flow was then calibrated using the Manning's coefficient and roller efficiency coefficient (Stive and De Vriend 1994). Additional tests were run for comparison with the same settings except the roller model was turned off.

Table 4. Calibration parameters for the Visser (1991) test cases.
Parameter Case 4 Case 7 Default
Manning’s coefficient,s/m1/3 (flow only) 0.0115 0.018 None
Breaker coefficient 0.64 0.9 Automatic(random waves)
Roller dissipation coefficient 0.1 0.1 0.1
Roller efficiency factor 0.8 0.8 1.0


Results and Discussion

Case 4

The measured and computed wave heights, longshore currents, and water levels for Case 4 are compared in Figure 2. Results are shown with and without the surface roller. The results are significantly improved when the surface roller is included as demonstrated by the goodness-of-fit statistics shown in Table 5. The NMAE for longshore current was reduced from approximately 20 to 5%. The roller has the effect of spreading the peak longshore current and moving it closer to the shore. The surface roller also reduces the setup at the breaker and increases it in the surf zone and near the shoreline. Although the water levels and currents are significantly affected by the surface roller the wave height profile shows only minor differences when the roller is included. This is due to the fact that in this case the dominant wave process in the surf zone is the wave breaking and the current-wave interaction is relatively weak in the wave model.

Figure 2. Measured and calculated wave height (top), longshore current (middle), and water level (bottom) for Visser (1991) Case 4.


Table 5. Goodness-of-fit statistics for the Visser (1991) Case 4.
NRMSE, % NMAE,% R2 Bias
No Roller Wave height 7.10 5.35 0.985 0.002 m
Longshore current 22.59 19.91 0.609 0.018 m/s
Water level 13.95 11.66 0.954 0.000 m
Roller Wave height 6.70 5.11 0.985 0.002 m
Longshore current 7.14 5.11 0.962 0.007 m/s
Water level 9.04 7.38 0.957 0.000 m

Case 7

The measured and computed wave heights, longshore currents, and water levels with and without the roller for Case 7 are shown in Figure 3. It is interesting to note that although the offshore wave height, period and direction are the same as Case 4, the location of the breaker for Case 7 is significantly further offshore. It is suspected that Case 7 actually had a larger wave height than Case 4 which produced a larger breaker further offshore. However, because no measurements were available further offshore of the breaker, no changes were made to the incident wave height. The results are similar to those of Case 4 in that the longshore current velocities are significantly improved when the roller is included (see Table 26). No measurements of water levels were available for Case 7. Similarly to Case 4 the longshore current is well predicted when the roller is included except for the first 1 m from the shoreline where the current velocity is overpredicted.


Figure 3. Measured and computed longshore currents (top), water levels (middle) and wave heights (bottom) for Visser (1991) Case 7.


Table 6. Goodness-of-fit statistics for the Visser (1991) Case 7.
!Roller Variable NRMSE, % NMAE, % Bias
Off Wave height 7.10 5.35 0.985 0.002 m
Longshore current 22.59 19.91 0.609 0.018 m/s
Water level 13.95 11.66 0.954 0.000 m
On Wave height 6.70 5.11 0.985 0.002 m
Longshore current 7.14 5.11 0.962 0.007 m/s
Water level 9.04 7.38 0.957 0.000 m

Conclusions and Recommendations

Wave-induced currents and water levels were simulated with the CMS for the case of monochromatic waves over a planar bathymetry. Results were calculated with and without the surface roller and the best results were obtained with the roller turned on, using a roller dissipation coefficient of 0.1 and a roller efficiency factor of 0.8. Both currents and water levels were predicted with errors less than 10%. Additional tests will be conducted in the future to show model sensitivity to the calibration parameters and to better determine these parameters based on field conditions. The wave calibration and results shown here are related to regular waves and are not directly applicable to field conditions. However, the purpose of these tests was to test the performance of the hydrodynamic model as quantified by the comparison between measured and simulated longshore current velocities and water levels under strong wave forcing.


References

  • Fredsoe, J. (1984). “Turbulent boundary layer in wave-current motion,” Journal of Hydraulic Engineering, ASCE, 110, 1103-1120.
  • Stive, M.J.F., and De Vriend, H.J. (1994). “Shear stress and mean flow in shoaling and breaking waves,” in Proceedings 24th International Coastal Engineering Conference, ASCE, New York, 594-608 pp.
  • Visser, R. J. (1991). “Laboratory measurements of uniform longshore currents”. Coastal Engineering, 15, 563-593.

Test_Cases