17_026_1 FHR reports
Manning’s roughness coefficient in SWASH Application to overtopping calculation
www.flandershydraulicsresearch.be
Manning’s roughness coefficient in SWASH Application to overtopping calculation
Suzuki, T.; Altomare, C.;De Roo, S.; Vanneste, D.; Mostaert, F.
Cover figure © The Government of Flanders, Department of Mobility and Public Works, Flanders Hydraulics Research Legal notice Flanders Hydraulics Research is of the opinion that the information and positions in this report are substantiated by the available data and knowledge at the time of writing. The positions taken in this report are those of Flanders Hydraulics Research and do not reflect necessarily the opinion of the Government of Flanders or any of its institutions. Flanders Hydraulics Research nor any person or company acting on behalf of Flanders Hydraulics Research is responsible for any loss or damage arising from the use of the information in this report. Copyright and citation © The Government of Flanders, Department of Mobility and Public Works, Flanders Hydraulics Research 2018 D/2018/3241/163 This publication should be cited as follows: Suzuki, T.; Altomare, C.;De Roo, S.; Vanneste, D.; Mostaert, F. (2018). Manning’s roughness coefficient in SWASH: Application to overtopping calculation. Version 2.0. FHR Reports, 17_026_1. Flanders Hydraulics Research: Antwerp. Reproduction of and reference to this publication is authorised provided the source is acknowledged correctly. Document identification Customer:
Author(s)
Agentschap Maritieme Dienstverlening en Ref.: WL2018R17_026_1 Kust, Afdeling Kust SWASH; bottom friction; wave run‐up 36 Appendices (p.): 17 ☒ Yes Released as from: 01/01/2020 ☒ The Government of Exception: Flanders Suzuki, T.; Altomare, C.; De Roo, Sieglien
Control
Name
Reviser(s):
Vanneste, D.
Project leader:
Suzuki, T.
Keywords (3‐5): Text (p.): Confidentiality:
Signature
Approval Head of Division:
F‐WL‐PP2.1.3‐2 Version 1 Valid as from 1/09/2017
Mostaert, F.
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Abstract The wave overtopping calculation method for the Safety assessment 2015 was defined in Suzuki et al. (2016). In order to get a conservative overtopping value for the safety assessment, the Manning’s roughness coefficient n in the SWASH 1D overtopping calculation was deliberately set as zero. This choice however impacts the estimation of wave overtopping over a cross shore profile which includes a foreshore having a ‘dry beach’ part in front of the dike. Normally, a sandy beach should add dissipation to wave propagation because of its roughness. Being excluded in the aforementioned methodology, this leads to an overestimation of wave overtopping. In this study, the bottom friction parameter in SWASH is further explored for overtopping calculation under a condtion with long berm (e.g. 50 m) in front of the dike. After the study it has been concluded that material based n value gives good estimation for wave transformation and run-up, for all the tested cases. Taking into account the fact that the Manning coefficient n for the sand and promenade material would be around 0.019, and therefore the default bottom friction value n=0.019 is recommended for overtopping calculation in SWASH.
Final version
WL2018R17_026_1
III
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Contents Abstract ............................................................................................................................................................ III Contents ............................................................................................................................................................ V List of tables..................................................................................................................................................... VII List of figures .................................................................................................................................................. VIII 1
Introduction ............................................................................................................................................... 1
2
Literature review ....................................................................................................................................... 2
3
Data screening ........................................................................................................................................... 4
4
5
3.1
Description of available datasets ...................................................................................................... 4
3.2
Discussions and conclusions ............................................................................................................ 13
Model....................................................................................................................................................... 14 4.1
SWASH model .................................................................................................................................. 14
4.2
Bottom friction approaches in SWASH ............................................................................................ 15
4.3
Manning approach .......................................................................................................................... 15
4.3.1
Manning coefficient................................................................................................................. 15
4.3.2
Scaling ...................................................................................................................................... 17
In-depth analysis of bottom friction settings in SWASH ......................................................................... 18 5.1
Wenduine ........................................................................................................................................ 18
5.1.1
Test cases................................................................................................................................. 18
5.1.2
Results ..................................................................................................................................... 19
5.2
Chen ................................................................................................................................................. 20
5.2.1
Test cases................................................................................................................................. 20
5.2.2
Results ..................................................................................................................................... 20
5.3
ICODEP ............................................................................................................................................. 22
5.4
WaLoWa .......................................................................................................................................... 23
5.4.1
Test cases................................................................................................................................. 23
5.4.2
Results ..................................................................................................................................... 24
5.5
Park & Cox ....................................................................................................................................... 29
5.5.1
Model settings ......................................................................................................................... 29
5.5.2
Implementation of time dependent wave run-up value in SWASH ........................................ 30
5.5.3
Post-processing of wave run-up calculation results ................................................................ 31
5.5.4
Results ..................................................................................................................................... 32
Final version
WL2018R17_026_1
V
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.6 6
Further discussion of Fielder et al. (2018) ....................................................................................... 34
Conclusions .............................................................................................................................................. 35 6.1
Bottom friction n value.................................................................................................................... 35
6.2
Wave generation and number of waves ......................................................................................... 35
References ....................................................................................................................................................... 36 Appendix A: WaLoWa SWASH results ............................................................................................................. A1 Bi_01_4 ........................................................................................................................................................ A1 Bi_01_5 ........................................................................................................................................................ A2 Bi_01_6 ........................................................................................................................................................ A3 Bi_02_4 ........................................................................................................................................................ A4 Bi_02_5 ........................................................................................................................................................ A5 Bi_02_6 ........................................................................................................................................................ A6 Bi_02_6_R .................................................................................................................................................... A7 Bi_01_6_R .................................................................................................................................................... A8 Bi_03_6 ........................................................................................................................................................ A9 Bi_03_6_1 .................................................................................................................................................. A10 Bi_03_6_2 .................................................................................................................................................. A11 Appendix B: Test case calculation ................................................................................................................. A12 Cross section 233 (XB raai 140) ................................................................................................................. A12 Cross section 234 (XB raai 145) ................................................................................................................. A15
VI
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
List of tables Table 3-1: Available dataset for the calibaration of bottom friction in SWASH................................................ 4 Table 3-2: Details of SCANDURA project ........................................................................................................... 5 Table 3-3: Details of SUSCO project .................................................................................................................. 6 Table 3-4: Details of WISE project ..................................................................................................................... 7 Table 3-5: Details of ICODEP project ................................................................................................................. 8 Table 3-6: Details of WALOWA project ............................................................................................................. 9 Table 3-7: Details of CHEN project .................................................................................................................. 10 Table 3-8: Details of Wenduine project........................................................................................................... 12 Table 4-1: Manning’s roughness coefficient of smooth concrete ................................................................... 15 Table 4-2: Manning’s roughness coefficients of plane wood, unfinished concrete and masonry cemented rubble .............................................................................................................................................................. 16 Table 4-3: Manning’s roughness coefficients of sand ..................................................................................... 17 Table 5-1: Overtopping test parameters for selected cases (1/25 model scale) ............................................ 18 Table 5-2: Comparison of wave transformation (n=0.000 vs n=0.012)........................................................... 21 Table 5-3: Bi-chromatic wave test cases (11 Cases) ........................................................................................ 23 Table 5-4: Test conditions for Case 1A-1C ....................................................................................................... 30
Final version
WL2018R17_026_1
VII
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
List of figures Figure 3-1: Sketch of the physical flume as in SCANDURA ................................................................................ 5 Figure 3-2: Scheme of the instruments along the beach in SUSCO .................................................................. 6 Figure 3-3: CIEM configuration for the WISE Benchmark experiments with the initial profile ........................ 7 Figure 3-4: Layout of the ICODEP engineered beach and instruments. ............................................................ 8 Figure 3-5: Model geometry WaLoWa ............................................................................................................ 10 Figure 3-6: Experimental set-up of the CHEN project ..................................................................................... 11 Figure 3-7: Physical and numerical model domains for Wenduine physical model test. ............................... 12 Figure 4-1: Surface texture of unfinished concrete (left) and masonry cemented rubble (right) .................. 17 Figure 5-1: Comparison of wave overtopping discharge with different Manning’s roughness coefficient (Wenduine case) .............................................................................................................................................. 19 Figure 5-2: Comparison of wave overtopping discharge with different Manning’s roughness coefficient (Chen’s case) .................................................................................................................................................... 21 Figure 5-3: Physical model settings (top) and SWASH result (below) ............................................................. 22 Figure 5-4: Wave gauge numbering in SWASH ............................................................................................... 24 Figure 5-5: Comparison of time series in Bi_01_4 case (SWASH vs Physical model) ...................................... 25 Figure 5-6: Comparison of time series in Bi_02_6_R case (SWASH vs Physical model) .................................. 25 Figure 5-7: Quality of time transformation from WG1 to WG3 between ‘good’ case (upper 3 panels) and ‘excellent’ case (lower 3 panels) and WG7s. ................................................................................................... 26 Figure 5-8: WG position on the dike in SWASH ............................................................................................... 27 Figure 5-9: Comparison of time series of eta and force in Bi_02_6_R case (SWASH vs Physical model) ....... 27 Figure 5-10: Comparison of wave transformation and force estimation with different bottom friction parameters. ..................................................................................................................................................... 28 Figure 5-11: Example of numerical model bottom setting in Pax&Cox’s study .............................................. 29 Figure 5-12: Wave run-up criteria (i.e. minimum water level)........................................................................ 30 Figure 5-13: Output example of wave run-up time series (blue) and smoothed one (red) ............................ 31 Figure 5-14: Output example of spectrum at three wave gauge locations and entire and part of wave run-up time series ....................................................................................................................................................... 31 Figure 5-15: SWASH wave run-up result vs Stockdon model (n=0.019) ......................................................... 32 Figure 5-16: Comparison of normalized wave run-up between SWASH and empirical models ..................... 33 Figure 5-17: Estimation of wave run-up conducted by Fiedler et al. (2018)................................................... 34
VIII
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
1 Introduction The wave overtopping calculation method for the Safety assessment 2015 was defined in Suzuki et al. (2016). In this methodology, it is stipulated that the numerical model SWASH (Zijlema et al., 2011b) and an empirical equation (Altomare et al., 2016) need to be used for wave overtopping calculation. SWASH is a process-based numerical wave model, capable of calculating a series of wave overtopping events during a storm. Therefore, it has more applicability compared to an empirical equation, which gives only an average wave overtopping estimate based on physical model tests results (using simplified geometries), for estimation of wave overtopping over a realistic configuration such as a dike with a vertical storm wall on its crest (Suzuki et al., 2017a). Even though SWASH is a process-based model, not all physical processes are included in the methodology for wave overtopping calculation, partly on purpose. One example is the exclusion of bottom friction. In order to get a more conservative overtopping value for the safety assessment, the Manning’s roughness coefficient n in the SWASH 1D overtopping calculation was deliberately set as zero. This choice however impacts the estimation of wave overtopping over a cross shore profile which includes a foreshore having a ‘dry beach’ part in front of the dike. Normally, a sandy beach should add dissipation to wave propagation because of its roughness. Being excluded in the aforementioned methodology, this leads to an overestimation of wave overtopping. Note that infiltration which might influence to the overtopping discharge is not dealt in this report. In this study, the bottom friction parameter in SWASH is further explored for overtopping calculation under a condition with long berm (e.g. 50 m) in front of the dike. The methodology of this study is as follows. First, a literature review on the relevant study is conducted in Chapter 2. One of the purpose of the literature review is to find material to study on the bottom friction effect. Second, available data is explored in order to find extra material to study in Chapter 3. Next, an overview is presented of the SWASH model and Manning’s bottom friction parameters in open-channel flow in different materials. This Chapter 4 also includes a relevant discussion on the scale and possible ranges of Manning’s coefficient values representing the roughness of sand and a dike promenade. After that, the SWASH model is used to study bottom friction effect (mainly by sensitivity analysis), including some extra validation of SWASH in Chapter 5. Finally some conclusions are drawn from this study in Chapter 6.
Final version
WL2018R17_026_1
1
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
2 Literature review The purpose of this study is to investigate Manning’s bottom friction parameter for wave overtopping calculation in SWASH under a berm condition (i.e. long sandy beach in front of the dike). Even though the ultimate target is to estimate accurate overtopping discharge, a study on wave run-up is also relevant for this study since the main concern is the behavior of wave or bore running up on sand. Therefore the literature review is specially focused on wave run-up. Note that the literature review on Manning coefficient and relevant studies, and also SWASH, are conducted in Chapter 4. In general, overtopping occurs when maximum wave run-up exceeds the crest of dune or structure. Therefore run-up is of great importance for coastal protection and many research has been done by the Coastal Engineering community. Due to its complex feature (e.g. the interaction of run-up and run-down) mostly the relationship between wave parameters and run-up is expressed by empirical equations , similar as for wave overtopping. One of the popular methods is to relate wave run-up with the Iribarren number (in other words: surfsimilarity parameter). The Iribarren number is expressed as đ?œ€đ?œ€ =
đ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ą
(1)
ďż˝đ??ťđ??ť/đ??żđ??ż
where β is the characteristic slope angle, H is wave height and L is wave length. Holman (1986) developed empirical wave run-up model using this expression based on field measurement conducted at Duck, NC, USA. The equation is expressed as follows. đ?‘…đ?‘…2% đ??ťđ??ť0
(2)
= 0.83đ?œ€đ?œ€ + 0.2
where R2% is the run-up value exceeded by 2% of the run-up events and H0 is significant wave height. Mase (1989) also developed a similar run-up equation based on the physical model test in a wave flume with plane slopes ranging from 1/30 to 1/5. The equation is as follows. đ?‘…đ?‘…2% đ??ťđ??ť0
= 1.86đ?œ€đ?œ€ 0.71
(3)
More recently Stockdon et al. (2006) developed another wave run-up model as below. đ?‘…đ?‘…2% = 1.1ďż˝0.35đ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąďż˝đ??ťđ??ť0 đ??żđ??ż0 + 0.5ďż˝đ??ťđ??ť0 đ??żđ??ż0 (0.563đ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ąđ?‘Ą2 + 0.0004)ďż˝
(4)
In addition to those studies, a number of equations are proposed in literature, such as van der Meer & Stam (1993), De Waal and Van der Meer (1992) and Vandermeer (1998). Those are summarized in e.g. EurOtop II manual (Allsop et al., 2016). Apart from articles dealing with the empirical equations, some numerical studies on wave run-up are also found in literature. Those are more relevant to this study since the final purpose of this study is to find appropriate bottom parameter to be used in the numerical model, SWASH. From those numerical studies, Park & Cox (2016) is particularly interesting for this study since it investigates wave run-up on a long berm. Empirical equations were developed based on their numerical model (Boussinesq equation based model) which has been validated with field measurement. Tested cases have similar dry beach length as can be found in Belgian coast after beach nourishment (cfr Knokke beach nourishment). This case will be further explored using SWASH in this study. The study result is shown in Chapter 5. Fiedler et al. (2018) is also a very relevant article to this study since they used the SWASH model and have proven that SWASH can be applicable to wave run-up estimation on steep and mildly sloping natural
2
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
beaches. In the end they suggested to use n=0.019 as Manning bottom friction parameter for the estimation of wave run-up in the natural beach. This study provides an interesting scope but some important discussion about directional spreading effect is still missing The detailed discussion is found in the last chapter of this report.
Final version
WL2018R17_026_1
3
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
3 Data screening 3.1 Description of available datasets Seven different datasets have been identified (Table 3-1) as potential datasets for the calibration of bottom friction in SWASH. Most of the campaigns have been carried out in large-scale facilities. Five of them have used mobile bed, meanwhile two are characterised by fixed bottom over the whole length of the wave facility. Three of seven datasets refer to experimental campaign carried out in the large-scale wave flume CIEM at Universitat Politecnica de Catalunya (UPC) in Barcelona, Spain. One experimental campaign has been carried out in the Grosser Wellenkanal (GWK) in Hannover, Germany and one in the new Delta Flume at Deltares, in The Netherlands, these are one of the largest wave facilities worldwide. The last two datasets come from physical model tests carried out at Flanders Hydraulics Research (FHR). Chen’s dataset was developed in a STW-programme on integral and sustainable design of multifunctional flood defences, Project No. 12760. The Wenduine test case was carried out in FHR for the design of new storm wall in Wenduine. The name and facility of each dataset are listed hereafter: Table 3-1: Available dataset for the calibaration of bottom friction in SWASH
Project name (facility)
Facility size
Mobile/fixed bed
SCANDURA (UPC)
Large
Mobile
SUSCO (UPC)
Large
Mobile
WISE (UPC)
Large
Mobile
ICODEP (GWK)
Large
Mobile
WALOWA (Deltares)
Large
Mobile
CHEN (FHR)
Non-large
Fixed
WENDUINE (FHR)
Non-large
Fixed
SCANDURA The SCANDURA project aimed at obtaining new data about flow velocity in a large scale wave flume where the bottom boundary layer is in the turbulent regime. The measurements provided instantaneous velocities along the vertical, offshore of the breaker line, in presence of an erodible bed and, in turn, in presence of small scale bed forms. The experiments have been carried in the CIEM large wave flume at the UPC (Universitat Politècnica de Catalunya). The wave flume is 100 m long, 3 m wide and 5 m deep. A sketch is reported in Figure 3-1. The profile follows the geometry of an initial slope (0.3 m height in 4 m length), followed by a flat profile along the next 15 meters and a final second slope (1/15) 43 m long. The velocity measurements have been carried out at 4 different positions along the channel, two positions along the slope, one in the horizontal part and a last one close to the beginning of the slope. The velocities have been measured along the water depth by using an array of 8 ADVs, while close to the bottom an UVP (Ultrasonic velocimeter profiler) have
4
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
been used. The location of these instrumentation changed during the experiments in order to study the profile at different flume locations. Water surface elevation has been measured along the flume by means of resistance type wave gauges and Micro Acoustics wave-gauges (AWGs). Some of the AWGs are located in the surf zone and swash zone (between 50 and 67m from the wave paddle position at rest). Table 3-2: Details of SCANDURA project
Infrastructure
CIEM
Flume bottom
mobile bed (sand, d50=246μm)
Generated waves
2nd order monochromatic waves with AWAS
Figure 3-1: Sketch of the physical flume as in SCANDURA
SUSCO The SUSCO project aimed at using the Hydralab III facility at UPC to investigate shoreline response and the swash zone hydrodynamics when grouping waves, able to generate free waves and energy in the high frequency part of the spectra, impact on the controlled area. The experiments have been carried in the CIEM large wave flume at the UPC (Universitat Politècnica de Catalunya). The types of measurements that have been performed are: • • • • • •
Beach profile evolution; Wave characteristics; Run and set-up; Velocities inside the SZ; High frequency measures of longshore-crosshore sediment concentration within the SZ; High frequency measures of height bed bottom over the entire SZ.
The following instrument have been used during the experimental campaign: • • • • • • • •
1 Profiler 6 ADVs 10 Resistant Wave Gauges 8 Micro Acoustic Wave Gauges 4 Acoustic Wave gauges 8 Optical Backscatter Sensor 6 Electromagnetic Current Meters 6 Pressure Sensors.
Location of the instruments along the beach is depicted in Figure 3-2.
Final version
WL2018R17_026_1
5
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Table 3-3: Details of SUSCO project
Infrastructure
CIEM
Flume bottom
mobile bed (sand, d50=246μm)
Generated waves
2nd order regular monochromatic, combination of free standing long waves plus monochromatic short waves, bi-chromatic waves, random waves with AWAS
Figure 3-2: Scheme of the instruments along the beach in SUSCO
WISE These tests have been done in order to study the bottom evolution when reproducing 36 times the same random time series. Wave conditions are Hs = 0.47 m with Tp = 3.7 s following a Jonswap Spectrum with gamma 3.3 using always the same seeding number. A profile has been taken after each time series, while wave evolution, velocity and suspended sediment concentrations where measured mainly on the bar and in the swash zone. The experiments have been carried in the CIEM large wave flume at the UPC (Universitat Politècnica de Catalunya). A sketch of the flume with location of all instruments is reported in Figure 3-3. The following instrument have been used: • • • • • • 6
ADV Resistance Wave Gauges Acoustic Wave Gauges (medium range) Profiler Optical Backscatter Sensor (OBS) Pressure Sensors. WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Table 3-4: Details of WISE project
Infrastructure
CIEM
Flume bottom
mobile bed (sand, d50=246μm)
Generated waves
2nd order random waves with AWAS
Figure 3-3: CIEM configuration for the WISE Benchmark experiments with the initial profile (solid black line). The marks show the position for Resistive Wave Gauges (solid blue lines), ADVs (solid black pentagram), OBSs (empty red circles), PPT (solid black squares) and AWGs (empty blue squares).
ICODEP The experimental investigation focuses on the analysis of the influence of bed mobility on wave overtopping and wave loads on a flood defense structure. A large- scale model of a sloped seawall and a foreshore is set up in a wave flume. The incoming wave and mean water level conditions are chosen in order to be representative of a typical storm in a macro tidal regime. The experimental set-up consists of a 10/1 sloped steel seawall and a natural sandy foreshore with an initial 1/15 steepness. The sand available at the GWK was used with nominal diameter (d50) of 0.30 mm. A layout of the experimental set up is shown in Figure 3-4. The measured parameters are listed as follows: • •
Incident and reflected wave measured by means of Resistance Wave Gauges (WG) and Acoustic Wave Gauges (US). Morphodynamics of the sandy beach, through: measurements of the beach profiles by means of a mechanical profiler and of the 3D laser scanner, measurements of the velocity field and the sediment concentration along the beach (ADV, ABS). Information on the beach saturation condition
Final version
WL2018R17_026_1
7
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
• • •
are also acquired by the pore pressure sensors (PPS) located on the sand in front of the structure below the toe level. Morphology of the scour at the toe of the structure with laser scanner (3D and s2D). Impacts on the structures are measured with pore pressure transducers (PS) and load cells. Overtopping volume, from the analysis of the load cells located below the overtopping tank coupled with measurement of the pumping function and time operation. Table 3-5: Details of ICODEP project
Infrastructure
Grosser Wellenkanal (GWK)
Flume bottom
mobile bed (sand, d50=300μm) and a steel wall
Generated waves
2nd order random waves with AWAS
Figure 3-4: Layout of the ICODEP engineered beach and instruments. Upper panel: overview of the flume. Lower panel: close up on the structure.
WALOWA WALOWA aimed at studying the overtopped wave loads on walls in shallow foreshore conditions with particular focus on the impact mechanism and on the scale effects, by comparing the results from small
8
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
scale tests (1/25) with those obtained in the Delta Flume (model scale 1/4.3). The study of the scale effect is now on-going, so the report on this item has not been available yet. The experiments were carried out in the new Delta Flume at Deltares, in Delft (The Netherlands). A sketch of the flume with wave gauge location is depicted in Figure 3-5. A foreshore of compacted sand was constructed within the Delta Flume. The d50 was 320μm. This sand was only installed for the 0.4m top layer over the entire foreshore. Below the 0.4 m top layer was a sand with d50 = 230μm installed. The erosion depth over the entire foreshore never exceeded 0.4m during the tests. The foreshore starts at location X = 93.98 m from the paddle and is comprised of a transition slope (1/10) until 1.95 m flume height and attached a beach slope (1/35) until 3.71 m flume height. The end of the foreshore is the dike toe location X = 175.08 m from the paddle. Connected to the foreshore is a 0.53 m high dike structure with a 1/2 slope. On the dike a 2.32m wide promenade is built and at the end of the promenade a non-overtopped wall of 1.6 m height is constructed. The promenade has a 1/100 seaward slope to facilitate drainage of water. The following instrument were used during the experimental campaign: • • • • • • • • • • • • •
WaveGuide wave radar 7 Resistance Wave Gauges (WHM) along the wave flume Kulite HKM-379 (M) pressure sensors HBM U9 load cells Wave gauges on promenade MaxSonar HRXL/ Honeywell 943 M18 ultrasonic distance sensors Valeport 802 electro-magnetic current meter Airmar flow meter S300 paddle wheels High speed camera Void-fraction meter SICK LMS511 laser profiler ASM-IV-N argus surface meter Mechanical profiler
Using the above listed measurements, the following parameters were measured during the campaign: • • • • • • • • • • •
water surface elevation sand concentration bottom profile layer thickness on the promenade layer velocity on the promenade impact forces on wall impact pressure on wall void-fraction in front of the wall High speed camera recording from behind wall Laser profile of dike, promenade and wall Video recordings from top, side and behind wall Table 3-6: Details of WALOWA project
Infrastructure
New Delta Flume
Flume bottom
mobile bed (sand, d50=320μm for the foreshore 40cm-top-layer and d50=230μm for the foreshore bottom layer) with concrete dike after the foreshore
Generated waves
1st and 2nd order random and bichromatic waves with AWAS
Final version
WL2018R17_026_1
9
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Figure 3-5: Model geometry WaLoWa
CHEN Physical model tests were performed in a 4 m wide, 1.4m deep and 70 m long wave flume at Flanders Hydraulic Research, Antwerp, Belgium. A piston-type wave generator with a stroke length of 0.6 m was used generating monochromatic, bi-chromatic and random waves, but without an active wave absorption system. The wave flume was split into four sections (1 m for each), as shown in in Figure 3-6. The wave overtopping was measured in section A meanwhile the incident wave boundary conditions were measured in the outer section. Section A was used to measure the unobstructed overtopping flow features along the dike crest and Section B was used to measure the impact force of overtopping flows. Layer thickness on the sea dike was measured by means of 6 resistive wave gauges, 3 in Section A and 3 in Section B. To characterize the velocity field close to the wall, video camera recording together with Bubble Image Veocimetry was implemented. Table 3-7: Details of CHEN project
Infrastructure
Large wave flume at FHR
Flume bottom
fixed concrete bottom and a wooden dike
Generated waves
1st order regular monochromatic, bi-chromatic waves, random waves without AWAS
10
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Figure 3-6: Experimental set-up of the CHEN project
WENDUINE The purpose of the study is to provide necessary information for the design of storm wall in Wenduine. Tests were performed at 1/25 scale in the large wave flume at Flanders Hydraulics Research (Antwerp, Belgium). The flume is 70 m long, 1.5 m high and 4 m wide. A piston-type wave generator with a stroke of 0.5 m was used for wave generation with a passive wave absorption system located downstream of the sea-dike. The maximum water depth at the wave generator is 1.2 m. For physical model tests a JONSWAP wave spectrum with γ =3.3 was generated with the wave paddle and the total number of waves generated was about 1000. Wave height measurements were obtained with twelve resistance type wave gauges installed at the locations summarized in Figure 3-7. Mean wave overtopping discharge was obtained by dividing the total volume of water collected in an overtopping box during a test by the total duration of the test. A limitation in this experimental campaign is the use of first-order wave generation and the lack of active wave absorption. However, as proven in Altomare et al. (2016), the influence of the lack of active wave absorption is negligible. Also, the influence of cross waves was also found to be negligible as detailed in Altomare et al. (2016). Note that incident time series estimated from three offshore wave gauges were used in the validations, therefore the first-order wave generation and lack of active wave absorption are not really limitations for those validation cases. As can be seen in Figure 3-7, storm walls are positioned at different location on the dike: ‘A’ configuration denotes a vertical dike plus a vertical wall immediately behind the 1/35 shallow foreshore slope; ‘B’ configuration is a 1/2 sloping dike and a vertical wall at the end of the dike slope; and ‘C’ configuration is a vertical wall in the middle of the dike. The vertical wall was omitted in three test cases (refer as
Final version
WL2018R17_026_1
11
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
“WEN_004”, “WEN_024” and “WEN_026”). The mean overtopping discharge for these test cases cannot be estimated by existing semi-empirical equations due to the relatively complex configuration of a promenade combined with a storm wall in a very shallow foreshore. Note that if this were a deep water condition, the equation of (Van Doorslaer et al., 2015) could be used to estimate mean wave overtopping discharge, or if the storm wall was not present on the dike the empirical equation introduced by Altomare et al. (2016) would be applicable. Table 3-8: Details of Wenduine project
Infrastructure
Large wave flume at FHR
Flume bottom
fixed bottom with wooden dike
Generated waves
1st order irregular waves without AWAS
Figure 3-7: Physical and numerical model domains for Wenduine physical model test. Upper panel shows the entire model domain including location of wave gauges 1 to 12. Lower panel shows close up of sea dike and sea wall configuration “A”, “B” and “C” with dashed lines.
12
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
3.2 Discussions and conclusions Most of the experimental test campaigns (i.e. SCANDURA, SUSCO, WISE, ICODEP, WALOWA) used sand as bottom and therefore there were changes in bathymetry during one test. On top of that, none of them has a long and dry berm in those tests. Therefore those would be difficult to be used for the validation/tuning a fixed bed model SWASH for overtopping over sand berms. On the other hand, FHR and the associated research group have conducted preliminary SWASH calculations and good scopes how to use SWASH are available for ICODEP and WALOWA cases even though they used mobile bed. Data from WALOWA will be most useful since some test cases were using bi-chromatic waves (i.e. short duration test) and thus the bathymetry change during the tests is limited. Even though the berm effect cannot be fully discussed, we decided to use WALOWA tests for the further investigation of SWASH settings. The ICODEP case is also used to see the behaviour of SWASH. Apart from those cases, the CHEN and WENDUNE test will also be used since FHR has a good knowledge of those physical models (Altomare et al., 2016) and also SWASH modelling (Suzuki et al, 2017b) for those datasets. All relevant calculation and results are shown in Chapter 5.
Final version
WL2018R17_026_1
13
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
4 Model 4.1 SWASH model SWASH is an open source deterministic time domain wave model. The governing equations of the model are the non-linear shallow water equations with added non-hydrostatic effects. The one-dimensional, depth-averaged shallow water equations in non-conservative form are shown as follows:
(5)
(6)
(7) (8) where t is time, x the horizontal coordinate, u the depth averaged velocity in x-direction, ws and wb the velocity in z-direction at the free surface and at the bottom, respectively. Îś is the free-surface elevation from still water level, d is the still water depth and h the total depth. pb is the non-hydrostatic pressure at the bottom, g the gravitational acceleration and cf the dimensionless bottom friction coefficient. The bottom friction coefficient cf is expressed by Manning's roughness coefficient n as follows: .
(9)
Note that Manning coefficient n is a dimensional parameter [s/m^1/3]. Eqs. (5) and (8) are the global and local continuity equations, respectively, to assure both local and global mass conservation. Eq. (6) is the momentum equation for the u-velocity which includes the effect of nonhydrostatic pressure and bottom friction. Note that momentum conservation is obtained at the discrete level in line with (Stelling & Duinmeijer, 2003). First equation of Eq. (7) is the momentum equation for the vertical velocity at free surface ws. The vertical velocity at the bottom wb is described by means of the kinematic condition as presented by the last part of Eq. (7). Note that the governing equations are based on the incompressible Navier-Stokes equations when multiple layers in the vertical are considered. In this way we take into account the vertical structure of the horizontal flow. In this study all calculations have been conducted in one single layer, i.e. depth-averaged, which appeared to be sufficient with respect to frequency dispersion related to wave transformation and suitable for wave overtopping calculation in terms of computational stability. A full description of the numerical model based on a staggered, conservative, finite-difference scheme, different kinds of boundary conditions, and different types of applications are given in Rijnsdorp & Zijlema (2016), Smit et al. (2013) and Zijlema et al. (2011).
14
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
4.2 Bottom friction approaches in SWASH In the user manual of SWASH describes the bottom friction approaches in SWASH as follows: For typical depth-averaged calculations, four different bottom friction values are available, i.e., constant, Chezy, Manning and Colebrook-White values. Note that the Colebrook-White friction value equals the Nikuradse roughness height. Although they are associated with depth-averaged flow velocities, they may be applied in the multi-layered mode as well. However, some inaccuracies may occur in the vertical structure of the velocity, in particular when the depth-averaged velocity is zero. Alternatively, the logarithmic wall law may be applied. In this case, a distinction is made between smooth and rough beds. For rough beds, the user must apply a Nikuradse roughness height. The aforementioned friction formulations are usually derived for quasi-steady flow condition (e.g. flow in a river). However, numerical experiments have indicated that the Manning formula provides a good representation of wave dynamics in the surf zone, and even better to that returned by other friction formulations. In this study we only applied Manning approach to include bottom friction effect in line with Suzuki et al. (2017b) and Zijlema et al. (2011). The reasoning to choose the Manning approach is that this expression provides a better representation of wave dynamics in the surf zone to that returned by other well-known friction formulations such as the one in terms of the Chézy coefficient and the Colebrook–White equation, according to Zijlema et al. (2011). As described above, Manning bottom friction approach can be applied to multi-layered mode as well. The applicability and results will be shown in Chapter 5.
4.3 Manning approach 4.3.1
Manning coefficient
The Manning formula is developed by R. Manning in 1891 which characterizes the flow in channels. According to that theory flow velocity characteristics of open channels are dependent on the slope of the bed, the roughness of the channel, and the hydraulic radius of the channel. The empirical value of the Manning coefficient has been studied for long time and we can access to those in literature. Some examples are shown below. Note that application of those values are discussed in the next section. Table 4-1: Manning’s roughness coefficient of smooth concrete
Source: https://serc.carleton.edu/details/images/9159.html
Final version
WL2018R17_026_1
15
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Table 4-2: Manning’s roughness coefficients of plane wood, unfinished concrete and masonry cemented rubble
Source: http://www.fsl.orst.edu/geowater/FX3/help/8_Hydraulic_Reference/Mannings_n_Tables.htm
16
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Table 4-3: Manning’s roughness coefficients of sand
Source: http://www.engr.colostate.edu/~bbledsoe/CIVE413/Arcrement&Schneider.pdf
4.3.2
Scaling
As noticed before, the Manning roughness coefficient is a dimensional parameter (unit is [s/m^1/3]). Thus scaling factor can be applied if one follows the Froude scaling law. If one follows this, scaling up from 1/25 model scale (smooth wood for dike and concrete for foreshore bottom: both n=0.012 according to Table 4-1 and Table 4-2) to prototype scale is shown below. 1/25 scale
1/1 scale
dike length 1 [m]
dike length 25 [m] (factor 25)
time 2.4 [s]
time 12 [s] (factor 25^1/2)
n=0.012 [s/m^1/3] (mainly used in SWASH)
n=0.0205 [s/m^1/3] (factor 25^1/6)
It is worth to mention that using smooth wood and concrete in 1/25 scale (n=0.012) was more or less a good representation of sand and promenade in prototype (n=0.0205): sand’s n is around 0.017 assuming that the d50 is 0.3mm and promenade’s n can be 0.017-0.025 assuming that the material on the promenade is more rough than unfinished concrete and less rough than masonry cemented rubble, see surface material below. Figure 4-1: Surface texture of unfinished concrete (left) and masonry cemented rubble (right)
Final version
WL2018R17_026_1
17
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5 In-depth analysis of bottom friction settings in SWASH In order to have a comprehensive overview of parameter settings of SWASH for overtopping calculation, different parameters/methods are explored here while main focus stays on bottom friction parameter, Manning roughness coefficient n.
5.1 Wenduine The physical model test data of Wenduine are used for sensitivity analysis of bottom friction parameter in the SWASH model. These tests feature a smooth, impermeable sea dike (made of wood in the physical model) setup on a concrete foreshore with a constant 1/35 slope. 5.1.1
Test cases
In total, eleven physical model tests, as listed in Table 5-1, were simulated with the SWASH model, where Hm0, Tp are the offshore measured wave parameters at wave gauge 6 (refer to Figure 3-7). The still water levels (SWL) are measured values in the flume before the physical model test began. In these tests the SWL, wave conditions, foreshore level at the toe of the dike, wall heights and dike slopes are varied. Vertical wall elements on top of the dike, at the locations denoted ‘A’, ‘B’ and ‘C’, were also tested. Table 5-1: Overtopping test parameters for selected cases (1/25 model scale)
TEST NAME
SWL
Offshore Hm0
Offshore Tp
DIKE SLOPE
DIKE TOE LEVEL
DIKE CREST LEVEL*
CREST LEVEL
WALL CONFIGRATION
[-]
[m]
[m]
[s]
[-]
[m]
[m]
[m]
[-]
WEN_004
0.942
0.189
2.16
1/2
0.9352
1.0024
1.0096
No wall (Dike + Prom.)
WEN_017
0.989
0.186
2.19
1/2
0.9352
1.0024
1.0504
B
WEN_018
0.989
0.185
2.19
1/2
0.9352
1.0024
1.0504
C
WEN_024
0.967
0.187
2.19
1/2
0.9352
1.0024
1.0676
No wall (only dike)
WEN_026
0.952
0.198
2.63
1/2
0.9352
1.0024
1.0676
No wall (only dike)
WEN_027
0.943
0.196
2.63
1/2
0.9352
1.0024
1.0504
C
WEN_041
0.943
0.188
2.19
1/2
0.9352
1.0024
1.0744
C
WEN_042
0.943
0.186
2.19
1/2
0.9352
1.0024
1.0504
B
WEN_124
0.952
0.186
2.19
0
0.9352
1.0104
1.0343
A
WEN_125
0.952
0.186
2.19
0
0.9352
1.0104
1.0583
A
WEN_126
0.952
0.194
2.19
0
0.9352
1.0104
1.0823
A
*at the seaward edge of the dike
18
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.1.2
Results
In the safety assessment methodology book (Suzuki et al., 2016), a conservative approach, namely using bottom friction parameter n=0.000 (zero-friction), has been implemented. Looking at Figure 5-1 (also in the background report Suzuki et al., 2017a) the result of n=0.012 is slightly underestimating the overtopping discharge compared to n=0.000. Indeed n=0.000 gives the best result in terms of the Geo parameter (The geometric mean, see equation 9). However actually both values 0.87 (n=0.000) and 0.70 (n=0.012) are very good value compared to other results reported in Suzuki et al. (2017a), and thus we can still concluded that n=0.012 is a good parameter setting. Note that Geo=1 is the perfect match. In other words, it can be concluded that the material based n value (smooth concrete as foreshore and smooth wood as dike, both n=0.012) gives a good result in terms of overtopping estimation using SWASH.
Figure 5-1: Comparison of wave overtopping discharge with different Manning’s roughness coefficient (Wenduine case)
Geo
GSD
Final version
WL2018R17_026_1
19
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.2 Chen The physical model test data of Chen are used for sensitivity analysis of bottom friction parameter in the SWASH model. These tests feature a smooth, impermeable sea dike (made of wood in the physical model) setup on a concrete foreshore with a constant 1/35 slope. 5.2.1
Test cases
In total, 10 physical model tests from Chen’s test were simulated with the SWASH model. SWL, dike slope, dike toe level, dike crest level were fixed as 0.960 m, 1/3, 0.910 m and 1.011 m, respectively. Offshore significant wave heights were varied from 0.12 to 0.20 m and offshore peak wave period were varied from 2.0 to 2.4 s. 5.2.2
Results
The wave transformation and overtopping results (comparison between n-0.000 and 0.012) are shown in Figure 5-2 and Table 5-2, respectively. Surprisingly the result with n=0.000 gives slightly lower overtopping discharge (Figure 5-2) while wave properties are almost the same (Table 5-2). The normal expectation is that the higher bottom friction case gives lower overtopping discharge. However in this case the result is opposite. This has not been well understood so far. However, again the geo values for both cases are in a very good range (in general better than ones in Wenduine) thus we can still conclude that n=0.012 is a good parameter setting. It can be concluded that the material based n value (smooth concrete as foreshore and smooth wood as dike, both n=0.012) gives a good result in terms of overtopping estimation using SWASH.
20
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Table 5-2: Comparison of wave transformation (n=0.000 vs n=0.012)
Case
Hm0_n0.000/n0.012
Tm-1,0__n0.000/n0.012
Set-up_differece
134
1.01
1.03
-0.001
135
1.01
1.01
-0.001
143
1.00
0.97
-0.001
144
0.99
0.94
-0.001
153
1.00
0.95
-0.002
164
1.01
1.03
0.000
174
1.02
0.99
0.000
187
1.00
0.95
0.000
189
1.02
1.03
0.000
188
1.01
0.96
0.000
Figure 5-2: Comparison of wave overtopping discharge with different Manning’s roughness coefficient (Chen’s case)
Final version
WL2018R17_026_1
21
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.3 ICODEP The physical model test data of ICODEP project are used for sensitivity analysis of bottom friction parameter in the SWASH model. This experimental investigation focuses on the analysis of the influence of bed mobility on wave overtopping and wave loads on a flood defense structure. See details in Section 3.1. The SWASH modelling was conducted by Giulia Mancini from Nottingham University, supervised by Dr. Ricardo Briganti. FHR also gave some advice on her SWASH modelling. Here below is from ones of her presentation ppt slide, showing the overview of the physical model. According to her work, overtopping estimation is more accurate when the end bathymetry of each segments is used in SWASH. Note that one run is not too long (around 30 min in model scale), hence the bathymetric change within a segment is not comparable to e.g. the runs carried out for the safety assessment (45 hours in prototype). Figure 5-3 shows the sensitivity analysis of the bottom friction parameter for overtopping estimation. The overtopping estimation result is changed by the bottom friction setting ranging from n=0.010 to n=0.014 at the 10:1 sloped steel wall. From this analysis n=0.014 gives the best result. From this study also, it can be concluded that the material based n value (steel’s n would be around 0.014) gives a good result in terms of overtopping estimation using SWASH. Figure 5-3: Physical model settings (top) and SWASH result (below)
22
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.4 WaLoWa The physical model test data of WaLoWa project are used for further validation of SWASH (wave generation, transmission, wave transformation on the dike and force estimation) and sensitivity analysis of bottom friction parameter. These tests feature a smooth, impermeable sea dike (made of concrete in the physical model) setup on a mobile bed with initial slope of 1/35. See details in Section 3.1. 5.4.1
Test cases
As stated before, one of the good aspects of the WaLoWa test for the present study is that it has bichromatic wave test cases. The run time of the bichromatic wave cases are much shorter (e.g. 5 min) than one in irregular wave cases (3 hours). This data set provides an opportunity to compare wave generation, transformation, overtopping and force in a very detailed manner (i.e. irregular wave cases are typically evaluated by statistical wave parameter such as Hm0 and Tm-1,0, on the other hand bichromatic wave cases can be evaluated by time series). Accordingly, possible weak points of the numerical model can be detected relatively easier. In this study, all the bichromatic wave cases (see Table 5-3) are simulated using SWASH. The input of the wave boundary condition in SWASH is the water surface elevation obtained at WG2 (x=43.5 m). Note that reflection analysis was not conducted to get incident time series since the number of bichromatic waves were limited to three (reflected waves reached at WG2 right after the three bichromatic waves). Table 5-3: Bi-chromatic wave test cases (11 Cases) Test ID
a2/a1
f1 [Hz]
a1 [m]
omega [rad/s]
phase [°]
f2 [Hz]
a2 [m]
Omega [rad/s]
Phase [°]
Bi_01_4
1
0.174
0.5
1.0895
0
0.142
0.4
0.891
0
Bi_01_5
1
0.174
0.5
1.0895
0
0.142
0.45
0.891
0
Bi_01_6
1
0.174
0.5
1.0895
0
0.142
0.5
0.891
0
Bi_02_4
1
0.19
0.45
1.1931
0
0.155
0.36
0.976
0
Bi_02_5
1
0.19
0.45
1.1931
0
0.155
0.405
0.976
0
Bi_02_6
1
0.19
0.45
1.1931
0
0.155
0.428
0.976
0
Bi_02_6_R
1
0.19
0.45
1.1931
0
0.155
0.428
0.976
0
Bi_01_6_R
1
0.174
0.5
1.0895
0
0.142
0.5
0.891
0
Bi_03_6
1
0.1735
0.3
1.0896
0
0.142
0.3
0.891
0
Bi_03_6_1
1
0.1735
0.35
1.0896
0
0.142
0.35
0.891
0
Bi_03_6_2
1
0.1577
0.35
0.9910
0
0.129
0.35
0.811
270
The SWASH domain is modelled from x=43.5 to 206.4 m in the flume. The grid size of 0.2 m is used as a standard case. The vertical wall is modelled by activating BOTCel SHIFT mode in SWASH, so that the wall is expressed correctly while the default mode without BOTCel gives one extra step in front of the vertical wall due to interpolation. As stated above, the boundary input comes from time series of water surface elevation measured at WG2. The wave boundary in SWASH is the weakly reflective boundary which theory is the same as AWAS in the wave flume. In the standard case manning bottom friction parameter n=0.019 is used. The used SWASH version is always 4.01 unless otherwise stated. Wave gauges placed in the SWASH domain is 11 (7 on the flat bottom and foreshore, and 4 on the dike). The location of the seven wave gauges are depicted in Figure 5-4. To distinguish SWASH and physical model, the wave gauges in SWASH is described as e.g. ‘WG7’ while the physical model ones are described as e.g. ‘wave gauge 14’.
Final version
WL2018R17_026_1
23
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Figure 5-4: Wave gauge numbering in SWASH 6
Level [m]
1 2
3
4
5
6
7
4 2 0 0
50
100
150
200
250
x [m]
1-7 are represented of wave gauges placed in SWASH: WG1 –WG7 (top)
5.4.2
Results
Wave generation and wave transformation The quality of the result is roughly categorized into 2, namely ‘good’ and ‘excellent‘ results in terms of wave transformation (evaluated at WG7 in SWASH, i.e. ‘wave gauge 14’ in the physical model). One example of the ‘good’ results is shown in Figure 5-5 (case Bi_01_4). As can be seen the time series calculated using SWASH follow the general trend of the time series of the physical model. However, if the time series are critically evaluated, some discrepancy can be recognized e.g. WG7. The time series of WG7 is especially important since the quality of WG7 decides the quality of the estimation of the force (evaluation of the force is described later). Similar quality results are found in Bi_01_4, Bi_01_5 and Bi_01_6 (3 cases) One example of the ‘excellent’ results is shown in Figure 5-6 (case Bi_02_6_R). As can be seen the time series calculated using SWASH follow the trend of the time series of the physical model until toe of the dike (i.e. WG7). Similar quality results are found in Bi_02_4, Bi_02_5, Bi_02_6, , Bi_01_6_R, , Bi_02_6_R, , Bi_03_6, , Bi_03_6_1, and Bi_03_6_2 (8 cases). See all the details of the results in Appendix A. Note that the some extra oscillation found in WG7 in SWASH after the third/fourth bichromatic wave could be due to input signal. As explained earlier we used measured water surface elevation from WG1 (wave gauge 2 in physical model) and therefore it includes reflected waves. On the other hand SWASH recognized this as all incident wave and therefore SWASH generated wrong waves after reflected waves arrived at WG1. However this is not a problem since the signal after third/fourth bichromatic wave is not our interest. It could be corrected if incident wave information was given (e.g. by reflection analysis from three wave gauges). However in this study the position of the three wave gauges were not optimized for the bichromatic wave condition, so further treatment is not conducted. Closely looking at the two different results, it can be found that the quality of time series at WG7 is decided by the quality of wave transformation close to the wave genaration boundary, in other word, wave transformation from WG1 to WG3 (Figure 5-7). As can be seen in the blue circles, the wave transformation of Bi_01_4 is much worse than one of Bi_02_6_R. This eventually resulted in less good correspondence of time series between SWASH and physical model in WG7. However the reason is not sure why the underestimation of the crest is happened. It might be due to the boundary problem: in general SWASH gives higher wave height (around 10%) at the wave boundary while this is rapidly decreased in space. However this higher wave or higher wave steepness would cause extra wave breaking. In this test program, the problem is only seen in three cases, Bi_01_4, Bi_01_5 and Bi_01_6. As a summary, the quality of the initial part of the input wave time series (WG1-3) is very important and if it is fine the time series at WG7 (at the toe of the dike) is predicted very well.
24
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Figure 5-5: Comparison of time series in Bi_01_4 case (SWASH vs Physical model) 1
2 1
eta [m]
eta [m]
1 0 -1 50
100
150
200
250
50
100
t [s]
3
4
150
200
250
150
200
250
150
200
250
1
eta [m]
eta [m]
Model
-1
t [s]
1 0 -1 50
100
150
200
0 -1
250
50
100
t [s]
t [s]
5
6
1
1
eta [m]
eta [m]
SWASH
0
0 -1 50
100
150
200
0 -1
250
50
100
t [s]
t [s] 7
eta [m]
1 0 -1 50
100
150
200
250
t [s]
Figure 5-6: Comparison of time series in Bi_02_6_R case (SWASH vs Physical model) 1
2 1
eta [m]
eta [m]
1 0 -1 50
100
150
200
Model
50
100
t [s]
t [s]
3
4
1 0 -1 50
200
100
150
150
200
150
200
0 -1
200
50
100
t [s]
t [s]
5
6
1
1
eta [m]
eta [m]
150
1
eta [m]
eta [m]
SWASH
0 -1
0 -1 50
100
150
0 -1
200
t [s]
50
100
t [s]
7
eta [m]
1 0 -1 50
100
150
200
t [s]
Final version
WL2018R17_026_1
25
Manning’s roughness coefficient in SWASH - Application to overtopping calculation Figure 5-7: Quality of time transformation from WG1 to WG3 between ‘good’ case (upper 3 panels) and ‘excellent’ case (lower 3 panels) and WG7s. 1
eta [m]
1 0 -1 50
100
150
200
250
150
200
250
150
200
250
t [s] 3
eta [m]
1 0 -1 50
100
t [s] 5 7
eta [m]
1 0 -1 50
100
t [s] 1
eta [m]
1 0 -1 50
100
150
200
150
200
150
200
t [s] 3
eta [m]
1 0 -1 50
100
t [s] 5 7
eta [m]
1 0 -1 50
100
t [s]
26
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Wave transformation on the dike and wave force estimation In this section wave transformation on the dike (from WG7 to WG12 in SWASH, see Figure 5-8) and force estimation are discussed. Figure 5-8: WG position on the dike in SWASH
Level [m]
6 9
8
7
5
10
11
12
13
4 3 170
171
172
173
176
175
174
177
178
x [m]
9-12 are WGs to measure layer thickness in SWASH.
The wave transformation on the dike and force estimation from the ‘excellent’ case in the previous section is shown in Figure 5-9 (case Bi_02_6_R). As can be seen a good correspondence is found between SWASH and measurement not only wave time series at toe of the dike (i.e. WG7) but also wave time series in front of the wall (i.e. WG12) and force acting on the wall. Figure 5-9: Comparison of time series of eta and force in Bi_02_6_R case (SWASH vs Physical model)
1.5
b2R toe
SWASH
1
Measurement
eta [m]
0.5 0 -0.5 40
60
80
100
120
140
160
180
200
140
160
180
200
140
160
180
200
t [s]
eta [m]
1
in front of wall
0.5
0 40
60
80
100
120
t [s] 10
F [kN/m]
5
0
-5 40
60
80
100
120
t [s]
Final version
WL2018R17_026_1
27
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Sensitivity analysis on bottom friction Figure 5-10 shows comparison of wave transformation and force estimation with different bottom friction parameters. As can be seen in the figure, there is no significant influence of bottom friction while small difference can be seen in force estimation. From those results, it can be concluded that the influence of bottom friction is somewhat limited when the dike length (i.e. dry part) is limited. However, it is expected that the bottom friction plays an important role for wave run-up (eventually overtopping/force) when there is a long dry beach (or berm) in front of the overtopping measurement point (e.g. safety line in the safety assessment). Figure 5-10: Comparison of wave transformation and force estimation with different bottom friction parameters.
Top figures shows wave signal at the toe, the second ones are measured on the dike and bottom ones are force measurement.
28
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.5 Park & Cox 5.5.1
Model settings
Figure 5-11 illustrates an example of the idealized profile defined by Park & Cox (2016). They developed a simplified dune–berm profile based on the assumptions that the berm width is flat and that the dune follows a Gaussian shape. The beach profile is based on an equilibrium beach profile which Romańczyk et al. (2005) defined. In principle the beach slope is determined by d50 size of sand. Park & Cox’s study consists of 3 cases. The first one, namely Case 1A-1C , is a fixed surge level case and does not use the berm and dune but only extension of the foreshore slope on the dry beach following Romańczyk et al. (2005). The second uses the same bathymetry but uses different surge levels and the third one uses the berm and dunes. Our study only uses only the first case (i.e. Case 1A-1C) for two reasons. First, the third case of Park & Cox’s study (using berm and dune) is based on only their numerical model, so the proposed equation might be including some model effects. On the other hand Case 1A-1C compare with empirical equations based on field measurement and physical model tests in literature, and therefore the case is more reliable. Secondly the 1A case still has a relatively long run-up distance (c.a. 50 m) and this is similar dry beach distance for our application (cfr Knokke beach nourishment project). Therefore the berm and dune configuration defined in the third case is not necessary to be used. Figure 5-11: Example of numerical model bottom setting in Pax&Cox’s study
B=3.5m, S=0m,WB=0m, hD=0m, WD=0m in Case 1A-1C
Final version
WL2018R17_026_1
29
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
The tested parameters for Case 1A-1C are listed below in combination with Table 5-4. In total around 400 runs have been conducted. Note that some cases are not successfully executed due to instability of the SWASH runs. • • •
Bottom friction: n=0.000, 0.019 and 0.025 Number of calculation layers in SWASH: 1 layer and 2 layers Wave generation: First order and Second order waves (only sub-harmonic) Table 5-4: Test conditions for Case 1A-1C
5.5.2
Implementation of time dependent wave run-up value in SWASH
Park & Cox’s study focused on wave run-up. However the present SWASH model (version 4.01) does not have a function to output time dependent wave run-up value (from which representative wave run-up parameter can be calculated, e.g. R2%), while which only has output of maximum wave run-up value (i.e. Rmax) during the entire run. Therefore in this study the time-dependent wave run-up value is implemented in SWASH version 4.01. Note that this implementation will be officially released in the future SWASH version including 2DH (i.e. basin like) application, however this implementation is only limited to 1D (2DV; i.e. flume like) application at this moment. The algorithm is shown as follows. 1) Decide run-up criteria value (height above the bottom level). Note that Fiedler et al. (2018) uses 0.1 m as the run-up criteria. 2) Detect the first grid point from the last calculation grid in x direction which exceeds the run-up criteria. 3) Calculate exact run-up value from the equation below (run-up criteria 0.01 is used in the equation). 4) Output wave run-up value at each time step. Figure 5-12: Wave run-up criteria (i.e. minimum water level)
30
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.5.3
Post-processing of wave run-up calculation results
Output example of the time series of the wave run-up is shown in Figure 5-13. As can be seen, time series present some noise close to each crest level. The reason of the noise could be originated due to numerical issue but the exact reason has not been understood yet. For an easier processing of the signal, a low-pass filter is applied after some testing with different frequencies. Eventually, a low-pass filter of 0.1 Hz gives a reasonable output (as shown with red line in Figure 5-13) while it gives some small underestimation. We accept this underestimation as a model effect. Figure 5-14 shows an output example of spectrum at three wave gauge locations in the numerical domain as defined in (Park & Cox, 2016) and entire and part of wave run-up time series. As can be seen, all the run-up points are well captured thanks to the post-processing (i.e. smoothing through the low pass filter). Figure 5-13: Output example of wave run-up time series (blue) and smoothed one (red)
Figure 5-14: Output example of spectrum at three wave gauge locations and entire and part of wave run-up time series WG1, H
m0
=0.96m, T
=16.7s
m-1,0
WG2, H 2.5
=1.21m, T
m-1,0
=16.9s
WG3, H
1.5
1.5
0.5
E [m
1
0.5
0 0.2
0.4
0.6
0.8
1
m-1,0
=26.3s
1
0.5
0 0
=0.77m, T
2
2
/Hz]
1.5
/Hz]
2
1
m0
2.5
2
E [m
E [m
m0
2
2
/Hz]
2.5
0 0
0.2
0.4
f [Hz]
0.6
0.8
1
0
0.2
f [Hz] PAK1A10175192S, R
0.6
0.8
1
f [Hz] 2%
1.5
0.4
=0.963, n=198
wave run-up [m]
1
0.5
0
-0.5 0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
t [s] 2%
wave run-up [m]
1
0.5
0
1600
1700
1800
1900
2000
2100
t [s]
Final version
WL2018R17_026_1
31
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.5.4
Results
Park & Cox conducted three comparisons with empirical equations. One is with the Stockdon model. The other ones are with Mase and Holman models. The results are shown below. Comparison with Stockdon model Figure 5-15 shows SWASH wave run-up result R2% comparing with empirical equation developed by Stockdon. Generally wave run-up estimated in SWASH is overestimated compared to Stockdon model which is based on field measurement. This means that the Stockdon model includes directional spreading effect while SWASH model is conducted in 1D (i.e.2DV). Therefore the difference can be explained by the directional spreading effect. However, unfortunately, in this study 2DH calculation cannot be conducted due to the limitation of the implementation of run-up in SWASH at this moment. The observation of the different outputs of SWASH is described here below (i.e. First order wave generation with 1 layer, second order in 1 layer and second order 2 layers). First of all, SWASH 1st order wave generation results give the highest wave run-up (up to 30%). This is due to spurious long wave generation which has been observed in Wenduine physical model test. This spurious long wave can be restricted by 2nd order wave generation in theory. However according to Rijsdorp et al. (2014), 2nd-order wave generation with 1 layer also includes some spurious free energy and this can be improved by using more than 2 layers. As shown in the result, indeed SWASH 2nd order wave generation in 2 layer give some differences in run-up compared to one from 2nd order wave generation in 1 layer (up to 15 % difference). Figure 5-15: SWASH wave run-up result vs Stockdon model (n=0.019)
3
2.5 Stockdon model
Stockdon model
First order 1 lay
First order 1 lay
2.5
Second order 1 lay
Second order 1 lay
2
Second order 2 lay
Second order 2 lay
2
[m] R
2%
1.5
R
2%
[m]
1.5
1
1
0.5
1A
0.5
1B 0
0 0
5
10
15
(H
20 o
T ) o
25
30
35
0
0.5
5
10
15
(H
20 o
25
30
35
T ) 0.5 o
3.5 Stockdon model First order 1 lay
3
Second order 1 lay Second order 2 lay
2.5
R
2%
[m]
2
1.5
1
1C
0.5
0 0
5
10
15
(H
32
o
20
25
30
T ) 0.5 o
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Comparison with Mase and Holman model Figure 5-16 shows comparison of normalized wave run-up between SWASH and two empirical models, Mase and Holman. Mase’s model is developed based on their 2DV flume test while Holman’s model is developed based on field measurement. As stated above, directional spreading effect again plays an important role. As can be seen in the figure, cloud of the SWASH results is closer to Mase’s model due to directional spreading effect. SWASH results consist of 3 different bottom friction values, n=0.000, 0.019 and 0.025. In line with the other flume test (i.e. the bottom is impermeable smooth material), the bottom friction parameter for the physical model from Mase would be around 0.012. The result shows that Mase’s model comes in between SWASH results of n=0.000 and n=0.019. This implies that the SWASH performance is promising. Even if we did not conduct the case with n=0.012 in SWASH, it would presumably give a good fit to the empirical equation line. One conclusion that can be drawn from this result is that material based n values give a good estimation in this case. Figure 5-16: Comparison of normalized wave run-up between SWASH and empirical models 3.5 Theory Holman (from field measurement) Theory Mase (from flume test)
3
n=0.000 n=0.019 n=0.025
2
1.5
R
2%
/H
o
[m]
2.5
1
0.5
0 0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
xsi
The spreading for constant xi value is caused by different settings of the SWASH runs st nd (1 order and 2 order, one layer and two layers)
Final version
WL2018R17_026_1
33
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
5.6 Further discussion of Fielder et al. (2018) As can be seen, SWASH performance is promising in the tested cases (Wenduine, Chen, ICODEP, WaLoWa and Mase), and there are obvious directional spreading effect in 1D calculation. However, Fielder’s result indicates there is no directional spreading effect (1D calculation represents the wave run-up almost perfect: with error of 6-8 % error). We do not still understand the mechanism why Fiedler’s 1D SWASH model fits to field data. Even they mention that the 2D dynamics that lead to good 1D model performance are not understood (see details below).
We are not sure but in general field data contains a lot of uncertainties (e.g. bathymetry is not taken during the storm), so it is interesting to investigate this directional spreading effects in a physical model test. This project is ongoing within the CREST project (WL project number 18_039). Looking into some more details of their result, Figure 5-17 still indicates that directional spreading is playing an important role: Cardiff case (directional spreading of 20 deg) is a bit more overestimating run-up compared to the one in Agate (10 deg). As stated above detailed investigation of directional spreading effect will be studied soon in the CREST project. Figure 5-17: Estimation of wave run-up conducted by Fiedler et al. (2018)
34
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
6 Conclusions 6.1 Bottom friction n value This study indicates that material based n value gives good estimation for wave transformation and run-up, for all the tested cases. Taking into account the fact that the Manning coefficient for the sand and promenade material would be around 0.019, the default bottom friction value n=0.019 is recommended for overtopping calculation in SWASH. Note that the value can be changed if the material’s n is much differ from n=0.019.
6.2 Wave generation and number of waves SWASH 2nd order wave generation in 2 layer give some differences in run-up compared to one from 2nd order wave generation in 1 layer (up to 15 % difference). The difference is not that big in the tested configuration while (more than) 2 layers are supporsed to be more accurate in terms of wave generation as mentioned by Rijsdorp et al. (2014). It is noted that this might cause instability problem since in general 2 layers are more unstable than 1 layer. Further validations (e.g. using the 2DH physical model test in CREST project) will be useful in order to select a robust method.
Final version
WL2018R17_026_1
35
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
References Allsop, N.W.H.; Bruce, T.; De Rouck, J.; Kortenhaus, A.; Pullen, T.; Schüttrumpf, H.; Troch, P.; Van der Meer, J.W.; Zanuttigh, B. (2016). EurOtop II. Available at: www.overtopping-manual.com Altomare, C.; Suzuki, T.; Chen, X.; Verwaest, T.; Kortenhaus, A. (2016). Wave overtopping of sea dikes with very shallow foreshores. Coast. Eng. 116: 236–257. doi:10.1016/j.coastaleng.2016.07.002 De Waal, J.P., Van der Meer, J.W. (1992). Wave run-up and overtopping on coastal structures. Coast. Eng. Proc. 1 (23), 1758–1774. Fiedler, J.W.; Smit, P.B.; Brodie, K.L.; McNinch, J.; Guza, R.T. (2018). Numerical modeling of wave runup on steep and mildly sloping natural beaches. Coast. Eng. 131: 106–113. doi:10.1016/j.coastaleng.2017.09.004 Holman, R.A. (1986). Extreme value statistics for wave run-up on a natural beach. Coast. Eng. ISBN 03783839 (ISSN) 9(6): 527–544. doi:10.1016/0378-3839(86)90002-5 Mase, H. (1989). Random wave runup height on gentle slope. J.Waterw. Port Coast. Ocean Eng. 115(5): 649–661 Park, H.; Cox, D.T. (2016). Empirical wave run-up formula for wave, storm surge and berm width. Coast. Eng. ISBN 0378-3839 115: 67–78. doi:10.1016/j.coastaleng.2015.10.006 Rijnsdorp, D.P.; Zijlema, M. (2016). Simulating waves and their interactions with a restrained ship using a non-hydrostatic wave-flow model. Coast. Eng. 114: 119–136. doi:10.1016/j.coastaleng.2016.04.018 Smit, P.; Zijlema, M.; Stelling, G. (2013). Depth-induced wave breaking in a non-hydrostatic, near-shore wave model. Coast. Eng. 76: 1–16. doi:10.1016/j.coastaleng.2013.01.008 Stelling, G.S.; Duinmeijer, S.P.A. (2003). A staggered conservative scheme for every Froude number in rapidly varied shallow water ows. Int. J. Numer. Meth. Fluids 1329–1354 Stockdon, H.F.; Holman, R.A.; Howd, P.A.; Sallenger, A.H. (2006). Empirical parameterization of setup, swash, and runup. Coast. Eng. ISBN 0378-3839 53(7): 573–588. doi:10.1016/j.coastaleng.2005.12.005 Suzuki, T.; Altomare, C.; De Roo, S.; Kolokythas, G.K.; Vanneste, D.; Peeters, P.; Mostaert, F. (2017a). Methodology for safety assessment 2015 - Background report on wave transformation and overtopping calculation by SWASH. FHR reports, 14_014_6. Flanders Hydraulics Research: Antwerp Suzuki, T.; Altomare, C.; Veale, W.; Verwaest, T.; Trouw, K.; Troch, P.; Zijlema, M. (2017b). Efficient and robust wave overtopping estimation for impermeable coastal structures in shallow foreshores using SWASH. Coast. Eng. 122: 108–123. doi:10.1016/j.coastaleng.2017.01.009 Suzuki, T.; De Roo, S.; Altomare, C.; Zhao, G.; Kolokythas, G.K.; Willems, M.; Verwaest, T.; Mostaert, F. (2016). Toetsing kustveiligheid 2015 - Methodologie: toetsingsmethodologie voor dijken en duinen. Versie 10. WL Rapporten, 14_014. Waterbouwkundig Laboratorium: Antwerpen van der Meer, B.J.W.; Stam, C.M. (1993). W A V E RUNUP ON SMOOTH AND R O C K SLOPES OF 118(5): 534–550 VanderMeer, J.W. (1998). Wave run-up and overtopping. In: Pilarczyk, K.W. (Ed.), Dikes and Revetments: Design, Maintenance and Safety Assessment. AA Balkema, Rotterdam, The Netherlands, pp. 145–159. Van Doorslaer, K.; Rouck, J. De; Audenaert, S.; Duquet, V. (2015). Crest modi fi cations to reduce wave overtopping of non-breaking waves over a smooth dike slope. Coast. Eng. 101: 69–88. doi:10.1016/j.coastaleng.2015.02.004 Zijlema, M.; Stelling, G.; Smit, P. (2011). SWASH : An operational public domain code for simulating wave fi elds and rapidly varied fl ows in coastal waters. Coast. Eng. 58(10): 992–1012. doi:10.1016/j.coastaleng.2011.05.015
36
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Appendix A: WaLoWa SWASH results All the time series (wave gauges and force) of bi-chromatic wave cases are presented here. Note that 7 in the figures indicates the time series of water surface at the toe of the dike.
Bi_01_4
Final version
WL2018R17_026_1
A1
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_01_5
A2
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_01_6
Final version
WL2018R17_026_1
A3
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_02_4
A4
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_02_5
Final version
WL2018R17_026_1
A5
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_02_6
A6
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_02_6_R
Final version
WL2018R17_026_1
A7
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_01_6_R
A8
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_03_6
Final version
WL2018R17_026_1
A9
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_03_6_1
A10
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Bi_03_6_2
Final version
WL2018R17_026_1
A11
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Appendix B: Test case calculation The proposed methodology is tested for the cross section 233 (XB raai 140) and 234 (XB raai 145) (cfr. 16_027 Knokke beach nourishment project, Table 3 at p.25 in Van Santen et al., 2017).
Cross section 233 (XB raai 140) The incident wave properties and overtopping discharge in the project of 16_027 (see cross section in Figure B-1) are: Hs=1.11 m ; Tm-1,0=47.3 s; mean water level at the toe (SWL+setup)=7.20 m TAW and q=5.85 l/s/m Figure B-1: Bathymetry for the overtopping calculation 15
z [m TAW]
10
5
0 1600
1650
1700
1750
1800
1850
1900
1950
2000
x [m]
A12
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
This test is repeated: Calibration (input: SWL=7.02 m TAW, Hm0=3.27 m, Tp=10.9 s) with n=0.000 using the same random number Hm0=1.10 m (-1 %), Tm-1,0=48.03 s (+2%) and h=7.02+0.125=7.15 m (-5 cm) Overtopping discharge q=5.07 l/s/m
Figure B-2: Time series of the repeated overtopping calculation with n=0.000
Layer thickness
h [m]
0.5
0 1000
2000
3000
4000
5000
6000
4000
5000
6000
5000
6000
5000
6000
time [s] Layer speed
u [m/s]
10
0
-10 1000
2000
3000
time [s] Mean wave overtopping discharge: 5.073 [l/s/m]
q [m
3
/s/m]
2 1 0 1000
2000
3000
4000
time [s]
q [m
3
/m]
Cumulative overtopping volume
20
0 1000
2000
3000
4000
time [s]
The incident wave properties are within the acceptable range and therefore the incident waves are fine. Overtopping discharge is slightly underestimated (-13%). The difference can be due to the difference of the SWASH version (SWASH version 4.01.A is used here). See details of this discussion in De Roo (2017).
Final version
WL2018R17_026_1
A13
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
The incident wave calibration is conducted using the bottom friction n=0.019, as proposed in Chapter 6. The results are as follows. Calibration 1 (SWL=7.05 m TAW, Hm0=3.28 m, Tp=10.9 s) with n=0.019 Hm0=1.10 m (+1%), Tm-1,0=39.04s(-11%) and h=7.05+0.196=7.25m (+6 cm) -> the wave period is underestimated and water level is overestimated. Calibration 2 (SWL=6.95 m TAW, Hm0=3.40 m, Tp=10.93 s) with n=0.019 Hm0=1.10 m (-1%), Tm-1,0=46.0 s(-3%) and h=6.95+0.239=7.19m (-2 cm) -> all the parameter is within the acceptable range. The overtopping calculation is conducted and the result is shown below. q= 1.49 l/s/m Figure B-3: Time series of the overtopping calculation with n=0.019
Layer thickness
h [m]
0.4
0.2
0 1000
2000
3000
4000
5000
6000
4000
5000
6000
5000
6000
5000
6000
time [s] Layer speed
u [m/s]
5
0
-5 1000
2000
3000
time [s] Mean wave overtopping discharge: 1.488 [l/s/m]
q [m
3
/s/m]
1 0.5 0 1000
2000
3000
4000
time [s]
q [m
3
/m]
Cumulative overtopping volume
20
0 1000
2000
3000
4000
time [s]
As can be seen in the figure, the smaller overtopping events are disappeared. The average overtopping discharge is lowered around 30% compared to the original one (i.e. 1.49 from 5.14 l/s/m).
A14
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
Cross section 234 (XB raai 145) The incident wave properties and overtopping discharge in the project of 16_027 (see cross section in Figure B-1) are: Hs=1.09 m ; Tm-1,0=43.8 s; mean water level at the toe (SWL+setup)=7.19m TAW and q=6.80 l/s/m Figure B-4: Bathymetry for the overtopping calculation
z [m TAW]
10
5
0 1600
1650
1700
1750
1800
1850
1900
1950
2000
x [m]
Final version
WL2018R17_026_1
A15
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
This test is repeated: Calibration (input: SWL=7.05 m TAW, Hm0=3.28 m, Tp=10.90 s) with n=0.000 using the same random number Hm0=1.09 m (0 %), Tm-1,0=43.12s(-2%) and h=7.05+0.109=7.16m (-3 cm) Overtopping discharge q=5.14 l/s/m Figure B-5: Time series of the repeated overtopping calculation with n=0.000
Layer thickness
h [m]
0.4
0.2
0 1000
2000
3000
4000
5000
6000
4000
5000
6000
5000
6000
5000
6000
time [s] Layer speed
u [m/s]
10
0
-10 1000
2000
3000
time [s] Mean wave overtopping discharge: 5.139 [l/s/m]
q [m
3
/s/m]
2 1 0 1000
2000
3000
4000
time [s]
q [m
3
/m]
Cumulative overtopping volume
20
0 1000
2000
3000
4000
time [s]
The incident wave properties are within the acceptable range and therefore the incident waves are fine. Overtopping discharge is slightly underestimated (-24%).
A16
WL2018R17_026_1
Final version
Manning’s roughness coefficient in SWASH - Application to overtopping calculation
The incident wave calibration is conducted using the bottom friction n=0.019, as proposed in Chapter 6. The results are as follows. Calibration 1 (SWL=7.05 m TAW, Hm0=3.28 m, Tp=10.90 s) with n=0.019 Hm0=1.10 m (+1%), Tm-1,0=39.04s(-11%) and h=7.05+0.196=7.25m (+6 cm) -> the wave period is underestimated and water level is overestimated. Calibration 2 (SWL=6.91 m TAW, Hm0=3.45 m, Tp=10.90 s) with n=0.019 Hm0=1.09 m (0%), Tm-1,0=42.7 s(-2%) and h=6.91+0.259=7.17m (-2 cm) -> all the parameter is within the acceptable range. The overtopping calculation is conducted and the result is shown below. q=2.65 l/s/m Figure B-6: Time series of the overtopping calculation with n=0.019
Layer thickness
h [m]
1
0.5
0 1000
2000
3000
4000
5000
6000
4000
5000
6000
5000
6000
5000
6000
time [s] Layer speed
u [m/s]
10
0
-10 1000
2000
3000
time [s] Mean wave overtopping discharge: 2.654 [l/s/m]
q [m
3
/s/m]
4 2 0 1000
2000
3000
4000
time [s]
q [m
3
/m]
Cumulative overtopping volume
20
0 1000
2000
3000
4000
time [s]
As can be seen in the figure, the smaller overtopping events are disappeared but still big overtopping event is less affected by the bottom friction. The average overtopping discharge is lowered around half compared to the original one (i.e. 2.65 from 5.14 l/s/m) but the cumulative overtopping volume of the first big overtopping event (around t=2100 s) is not reduced (both around 12 m3/m) in this case.
Final version
WL2018R17_026_1
A17
DEPARTMENT MOBILITY & PUBLIC WORKS Flanders hydraulics Research Berchemlei 115, 2140 Antwerp T +32 (0)3 224 60 35 F +32 (0)3 224 60 36 waterbouwkundiglabo@vlaanderen.be www.flandershydraulicsresearch.be