modelling the effect of nearshore detached breakwaters

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts Project: SC060026/R2

ii Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

The Environment Agency is the leading public body protecting and improving the environment in England and Wales.

It’s our job to make sure that air, land and water are looked after by everyone in today’s society, so that tomorrow’s generations inherit a cleaner, healthier world.

Our work includes tackling flooding and pollution incidents, reducing industry’s impacts on the environment, cleaning up rivers, coastal waters and contaminated land, and improving wildlife habitats.

This report is the result of research commissioned by the Environment Agency’s Evidence Directorate and funded by the joint Environment Agency/Defra Flood and Coastal Erosion Risk Management Research and Development Programme.

Published by: Environment Agency, Rio House, Waterside Drive, Aztec West, Almondsbury, Bristol, BS32 4UD Tel: 01454 624400 Fax: 01454 624409 www.environment-agency.gov.uk ISBN: 978-1-84911-181-2 © Environment Agency – February, 2010 All rights reserved. This document may be reproduced with prior permission of the Environment Agency. The views and statements expressed in this report are those of the author alone. The views or statements expressed in this publication do not necessarily represent the views of the Environment Agency and the Environment Agency cannot accept any responsibility for such views or statements. This report is printed on Cyclus Print, a 100% recycled stock, which is 100% post consumer waste and is totally chlorine free. Water used is treated and in most cases returned to source in better condition than removed. Email: [email protected]. Further copies of this report are available from our publications catalogue: http://publications.environment-agency.gov.uk or our National Customer Contact Centre: T: 08708 506506 E: [email protected].

Author(s): Hakeem Johnson, Jort Wilkens, Andy Parsons, Tim Chesher Dissemination Status: Publicly available Released to all regions Keywords: Nearshore Breakwaters, Beach control, Morphological modelling, guidance, macro-tidal sites Research Contractor: Halcrow Group Ltd., Burderop Park, Swindon, Wiltshire, SN4 0QD, Tel: +44 (0)1793 812479; Fax +44 (0)1793 812089 HR Wallingford Ltd., Howbery Park, Wallingford Oxon OX10 8BA, Tel: +44 (0)1491 835381; Fax: +44 (0)1491 832233 Environment Agency’s Project Manager: Eleanor Heron, Evidence Directorate Theme Manager: Geoff Baxter, Sustainable Asset Management (SAM) Project Number: SC0600026 Product Code: SCHO0210BRYN-E-P

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts iii

Evidence at the Environment Agency Evidence underpins the work of the Environment Agency. It provides an up-to-date understanding of the world about us, helps us to develop tools and techniques to monitor and manage our environment as efficiently and effectively as possible. It also helps us to understand how the environment is changing and to identify what the future pressures may be.

The work of the Environment Agency’s Evidence Directorate is a key ingredient in the partnership between research, policy and operations that enables the Environment Agency to protect and restore our environment.

The Research & Innovation programme focuses on four main areas of activity:

• Setting the agenda, by informing our evidence-based policies, advisory and regulatory roles;

• Maintaining scientific credibility, by ensuring that our programmes and projects are fit for purpose and executed according to international standards;

• Carrying out research, either by contracting it out to research organisations and consultancies or by doing it ourselves;

• Delivering information, advice, tools and techniques, by making appropriate products available to our policy and operations staff.

Miranda Kavanagh

Director of Evidence

iv Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

Executive summary

Background

Nearshore detached breakwaters are often considered an option for beach erosion control as part of coastal defence schemes. The dominant effect of a detached breakwater is to reduce the incident wave energy on a section of the coast and thereby reduce the sediment transport capacity in the sheltered region. In this way, detached breakwaters promote shoreline accretion in their lee.

Detached breakwaters have been used extensively in Japan, the US, Singapore and the Mediterranean, mainly along micro-tidal coasts (tidal range < 2m). Their use in the UK, where there are many meso-tidal (tidal range between 2m and 4m) and macro-tidal coasts (tidal range >4m), is relatively recent (from 1980; CIRIA 1996). More than 75 per cent of the UK coastline can be classified as meso- or macro-tidal coasts (see co-tidal chart 5058 from UK Hydrographic Office).

Rogers et al. (2006) carried out a review of existing design guidance for determining the geometrical layout of breakwater schemes to achieve a desired level of coastal protection. They concluded that the existing guidance is largely based on empirical data from micro-tidal coasts, and that the understanding of beach-breakwater interaction along meso- or macro-tidal coasts is very limited. In order to bridge this gap, the present research study was commissioned under the joint Environment Agency/ Department for Environment, Food and Rural Affairs Flood and Coastal Erosion Risk Management (FCERM) research programme. The aim of the study was to help improve practical design guidance for determining the geometrical layout of breakwater schemes on sandy macro-tidal coasts.

The development of this guidance has also been informed by the collaboration of the project team with members of the academic research community that carried out the LEACOAST2: scientific study between 2005 and 2008. Eight short papers from the study, which was funded by the Engineering and Physical Sciences Research Council (EPSRC), are collected in Appendix A.

The present report documents the scientific findings from this study. A companion report entitled Guidance for outline design of nearshore detached breakwaters on sandy macro-tidal coasts (Environment Agency 2009) documents the guidance drawn up from the findings of this study. The guidance report is aimed at assisting coastal practitioners who need to determine geometrical layout of breakwater schemes at the option appraisal stage.

Objectives

The main objectives of this study are summarised below.

• To provide an engineering steer to the EPSRC-funded scientific study.

• To undertake generic modelling under a variety of forcing conditions, focusing on tidal processes in combination with wave effects for arrays of nearshore detached breakwaters on a straight coastline.

• To conduct a detailed analysis of the output from the generic modelling in order to assess the effects of tidal processes on the influence that breakwaters have on the behaviour of the shoreline.

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts v

• To produce practical guidance for outline design of nearshore detached breakwaters in the macro-tidal environment.

• To disseminate the study findings.

Although the original objective was focused on determining the effect of tides on morphological response, it became clear during the study that existing guidance for micro-tidal beaches does not include all the important dimensionless parameters. Thus, work was also carried out to investigate the morphological response for non-tidal beaches.

Results and conclusions

For the non-tidal (or micro-tidal) case, it was shown using dimensional analysis that the beach response is dependent on Ls/X and X/Xb

1 when both the breakwater cross-section and the gap width between breakwaters are fixed. Furthermore, using model results from this study together with laboratory experiments compiled by Suh and Dalrymple (1987), the following trends were identified.

• For a given breakwater location in the surf zone (X/Xb), the salient length2 (S/X) increases as the dimensionless breakwater length (LS/X) increases.

• For a given LS/X, the relative salient length increases for low values of X/Xb and thereafter decreases, as should be expected for breakwaters located far away from the surf zone.

• Depending on the relative location of the breakwater in the surf zone, tombolo formation can occur for LS/X > 0.8. The limiting conditions for tombolo formation are postulated as: Ls/X > 2.8 – 1.6(X/Xb), X/Xb ≤ 1.25

Ls/X > -10.2 + 8.8(X/Xb), 1.25 < X/Xb< 2.0.

The main conclusions from the investigation of the effects of tides on the morphological response in the lee of breakwaters are summarised in the paragraphs below.

It was found that the relative salient lengths reduce as the tidal range increases for shore normal waves. However, for large values of Ls/X (>1.3) the influence of tidal range is not significant (if the breakwater is emergent through the tidal cycle). Furthermore, the base of the salient is wider for tides compared with the non-tidal case.

Progressive tides (where the maximum current speed occurs near high water) result in deflection of the nearshore bathymetry in the direction of high water flow. For the same tidal range, the salient length is slightly increased for standing tides (where the maximum current speed occurs near mean sea level) compared with progressive tides.

For oblique wave cases (wave direction at 45o to shore normal in 15m depth of water), the two numerical models used in the study (PISCES and MIKE 21 CAMS) show conflicting trends of the salient length with tidal range. This is considered to be due to the differences in the processes represented in the models. However, it is noted that the incident wave conditions at any given site typically consist of a range of wave directions. Furthermore, detached breakwaters are typically oriented to be shore

1 Ls is length of breakwater, X is the distance from the shoreline to the breakwater and Xb is the distance from the shoreline to the location of wave breaking. 2 The salient length S is the offshore extent of the salient (bulge in shoreline behind the breakwater) from the initial shoreline.

vi Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

parallel, which is typically at a small angle to the dominant wave direction. Thus, it is recommended that the indicative trend for shore-normal waves should be used for practical cases. Both models show a downdrift shift in the location of the salient with oblique wave incidence.

The relative salient length reduces as the breakwater crest level is reduced. The effect is more pronounced for cases where the breakwater is relatively close to the shoreline (Ls/X ≥1.3). Furthermore, beach levels are lower as the breakwater crest level is reduced. This is also confirmed by observations made at Sea Palling on the Norfolk coast of the UK during the EPSRC research project (see Appendix A). These observations show that the salient lengths behind low crested breakwaters are significantly shorter and the beach levels are lower compared with high crested breakwaters.

The seabed erosion across the nearshore profile in the breakwater bay generally reduces with increasing tidal range. There is more movement of the beach contours above mean sea level and, in particular, more erosion above mean sea level with increasing tidal range.

It was found that the simulated mean sea level shorelines in the bay for emerged breakwaters in tidal cases agree reasonably well with the bay shoreline planforms predicted using the method of Silvester and Hsu (1997). It is noted, however, that this result is obtained for test cases with a large gap width between the breakwaters. In the general case where the gap widths significantly influence the wave conditions in the bay, it is presently not clear if the same result will be obtained.

Based on the results above, two preliminary design graphs for determining the effects of tidal range and breakwater crest on salient length are proposed and incorporated in a companion design guidance report (Environment Agency 2010). The guidance report also contains worked examples illustrating the application of the design graphs in outline design.

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts vii

Acknowledgements We hereby express our thanks to Steve Jenkinson, Uwe Dornbusch, Geoff Baxter and Eleanor Heron (all of the Environment Agency) and Emeritus Professor B. A. O’Connor for their constructive comments on the first draft of this report.

We would also like to thank all members of the parallel LEACOAST2 research consortium, funded by the EPSRC, for the constructive discussions over the course of the project. The extensive field and numerical modelling work carried out by the research consortium to study the nearshore breakwater scheme at Sea Palling in Norfolk provided a good foundation for the work carried out in this study. We would also like to thank them for their contribution in the form of short technical papers to this report (Appendix A).

Lastly, we would also like to thank the following for their valuable contributions to the project: Suresh Surendran, Stefan Laeger, Ben Hamer, Jonathan Rogers and Adam Davidson.

viii Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

Abbreviations 2D Two-dimensional 2DH Two-dimensional, horizontal (depth averaged 2D model) 3D Three-dimensional ABS Acoustic backscatter Defra Department for Environment, Food and Rural Affairs EDF Electricité de France EOF Empirical orthogonal function EPSRC UK Engineering and Physical Sciences Research Council FCERM Flood and Coastal Erosion Risk Management HW High water LNH EDF’s National Hydraulics and Environmental Laboratory MSL Mean Sea Level SSPB Segmented shore-parallel breakwaters SWAN Simulating WAves Nearshore

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts ix

Contents 1 Introduction 1 1.1 Background 1 1.2 Effect of detached breakwaters 3 1.3 Outline design of detached breakwaters 4 1.4 This report 6 1.5 Convention 6

2 Literature review 7 2.1 Existing design guidance 7 2.2 Inventory of UK breakwaters 12

3 Dimensional analysis 16 3.1 Characteristic parameters 16 3.2 Dimensionless parameters 16 3.3 Generic test cases 18

4 Morphological modelling using PISCES 21 4.1 Simulated test cases 21 4.2 Model description 22 4.3 Set-up of numerical process models 26 4.4 Results and discussion 30 4.5 Summary 33

5 Morphological modelling using MIKE 21 CAMS 49 5.1 Simulated test cases 49 5.2 Model description 50 5.3 Setup of numerical process models 52 5.4 Results and discussion 56 5.5 Summary 68

6 Analysis of morphological model results 70 6.1 Key dimensionless parameters 70 6.2 Non-tidal cases 70 6.3 Tidal cases 72 6.4 Comparison of processes in model systems 74

7 Conclusions and future work 77 7.1 Conclusions 77 7.2 Future work 79

References 81

x Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

Appendices 85

A LEACOAST2 short papers 86 A.1 Overview of the LEACOAST2 project 88 A.2 Medium-term shoreline evolution 93 A.3 Video-based analysis of morphological changes 97 A.4 Wave, currents and sediment transport observed during the

LEACOAST2 experiment 100 A.5 Marine radar monitoring 103 A.6 Coastal area process modelling – 1 107 A.7 Coastal area process modelling – 2 112 A.8 Probabilistic modelling 116

B Results of generic test cases using PISCES 122

C Results of generic test cases using MIKE 21 CAMS 144 Table 1.1 Description of accreted shorelines in the lee of a detached breakwater 3 Table 1.2 Characteristic parameters 5 Table 2.1 Inventory of breakwater schemes around UK (T■=Tombolo, S∆=Salient, N×=No sinuosity) 13 Table 3.1 Dimensionless parameters for morphological response behind shore parallel breakwaters 17 Table 3.2 Key dimensionless parameters for morphological response for different nearshore breakwater

schemes 17 Table 3.3 Generic modelling test cases 19 Table 4.1 Generic test cases simulated using PISCES 21 Table 4.2 Wave model parameters and settings 27 Table 4.3 Flow model parameters and settings 28 Table 4.4 Sand transport model parameters and settings 28 Table 4.5 Morphodynamic model parameters and settings 30 Table 5.1 Generic test cases simulated using MIKE 21 CAMS 49 Table 5.2 Wave model parameters used in the morphological simulations 53 Table 5.3 Flow model parameters used in the morphological simulations 54 Table 5.4 Sand transport model parameters used in the morphological simulations 55 Table 5.5 Frequency of updates during the morphological simulation 55 Table 5.6 Generic non-tidal test cases simulated using MIKE 21 CAMS 58 Table 5.7 Generic tidal test cases simulated using MIKE 21 CAMS. (tide type: S=standing tides, P=progressive

tides) 63 Table 6.1 Representation of key physical processes in PISCES and MIKE 21 CAMS 75 Table 7.1 Additional tasks to improve further the outline design guidance for breakwaters on macro-tidal coasts 80 Table A.1 Overview of EPSRC short papers 86 Figure 1.1 Example breakwater schemes around UK coasts (1). 1 Figure 1.2 Example breakwater schemes around UK coasts (2). 2 Figure 1.3 Definitions of key variables for nearshore breakwater scheme (adapted from USACE 2003). 4 Figure 2.1 Existing design guidance for morphological response in the lee of detached breakwaters. 9 Figure 2.2 Dimensionless plot of nearshore breakwater projects (from USACE 2003). 9 Figure 2.3 Outline design procedure suggested by Fleming and Hamer (2000). 12 Figure 2.4 Existing design guidance for assessing possible shoreline erosion in the gaps between nearshore

breakwaters. 12 Figure 2.5 Comparison of field data around UK with existing design graph (after Pope and Dean, 1986;

■=Tombolo, ∆=Salient, ×=No sinuosity). 14 Figure 2.6 Comparison of field data around UK with existing design graph (after Rosati 1990; ■=Tombolo,

∆=Salient, ×=No sinuosity). 15 Figure 3.1 Simulated breakwater layouts. 20 Figure 4.1 Overview of the interlinking of process modules and information exchange. 30 Figure 4.2 Model bathymetry and details of the finite element mesh – Layout 1. 35 Figure 4.3 Wave height pattern (left) and wave force pattern (right) based on TOMAWAC. 36 Figure 4.4 Wave height pattern (left) and wave force pattern (right) based on SWAN. 36

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts xi

Figure 4.5 Initial wave, flow and sediment transport patterns for shore-normal waves and non-tidal conditions – Layout 1. 37

Figure 4.6 Initial wave, flow and sediment transport patterns for shore-normal waves and a 3m standing tide – Layout 1. 37

Figure 4.7 Simulated bathymetry after 60 days – Layout 1, shore-normal waves. 38 Figure 4.8 Simulated bathymetry after 60 days – Layout 2, shore-normal waves. 39 Figure 4.9 Simulated bathymetry after 60 days – Layout 1, oblique waves. 40 Figure 4.10 Simulated bathymetry after 60 days – Layout 2, oblique waves. 41 Figure 4.11 Simulated bathymetry after 60 days – Layout 3, oblique waves. 42 Figure 4.12 Bed level changes after 60 days – Layout 1, shore-normal waves. 43 Figure 4.13 Bed level changes after 60 days – Layout 2, shore-normal waves. 44 Figure 4.14 Bed level changes after 60 days – Layout 1, oblique waves. 45 Figure 4.15 Bed level changes after 60 days – Layout 2, oblique waves. 46 Figure 4.16 Bed level changes after 60 days – Layout 3, oblique waves. 47 Figure 4.17 Cross-shore bathymetry profile after 60 days along the centreline of the second breakwater –

Layout 2, shore-normal waves. 48 Figure 4.18 Cross-shore bathymetry profile after 60 days along the centreline of the second breakwater bay –

Layout 2, shore-normal waves. 48 Figure 5.1 Block flow diagram for MIKE 21 CAMS (from DHI Software 2008a). 52 Figure 5.2 Initial wave (top), flow (middle) and sand transport patterns (bottom) for non-tidal cases. 57 Figure 5.3 Simulated bathymetry after 60 days of morphological simulation (top: bathymetry contours; bottom:

profiles across section A and B). 59 Figure 5.4 Simulated bathymetry after 28 days of morphological simulation (top: bathymetry contours; bottom:

profiles across section A and B.). 60 Notes: Waves are normally incident to the shoreline (Hm0=1m, Tp=5s at depth of 15m) and no tides. 60 Figure 5.5 Simulated bathymetry after 60 days of morphological simulation – Layout L1 (Ls/X=0.8) and no tides

(top: bathymetry contours; bottom: profiles across section A and B). 61 Figure 5.6 Initial wave (top), flow (middle) and sand transport patterns (bottom) at four different stages of the 5m

progressive tide. 62 Figure 5.7 Simulated bathymetry contours after 60 days of morphological simulation. 64 Figure 5.8 Simulated nearshore profiles across section A and B after 60 days of morphological simulation. 65 Figure 5.9 Simulated bathymetry after 60 days of morphological simulation – layout L1, Hm0=2m, Tp=8s, oblique

wave incidence (top: bathymetry contours; bottom: profiles across section A and B). 66 Figure 5.10 Simulated bathymetry after 60 days of morphological simulation (top: bathymetry contours; bottom:

profiles across section A and B). 67 Figure 5.11 Simulated bathymetry after 60 days of morphological simulation (top: bathymetry contours; bottom:

profiles across section A and B). 68 Figure 6.1 Non-tidal cases from the numerical simulations and laboratory data from Suh and Dalrymple (1987). 71 Figure 6.2 Cross-shore profile evolution over 60-day simulation along the centre-line across the second

breakwater, for layout L1, shore-normal waves and 3m standing waves. 72 Figure 6.3 Effect of breakwater length for different dimensionless tidal ranges (Rtide/Hm0). 72 Figure 6.4 Effect of breakwater crest level (relative submerged depth at HW, dcr/Hm0) for different breakwater

length and dimensionless tidal ranges (Rtide/Hm0). 74 Figure A1.1 LEACOAST2 study site: Sea Palling, Norfolk. 89 Figure A1.2 Remote sensing equipment – a) ARGUS Video system; b) X-Band Radar. 90 Figure A1.3 a) Measurement locations; b) Instrument Frame. 91 Figure A1.4 Deploying and recovering instrument frames – a) by boat; b) by machinery. 91 FigureA2.1 Medium-term (1991-2005) rate of shoreline change. 94 Figure A2.2 Shoreline width adjacent to breakwaters (tombolos and salients) and gaps (embayments). 95 Figure A3.1 Bay shoreline positions for phase one (upper panel) and phase two (lower panel). 98 Figure A4.1 Experiment 1 and Experiment 2 showing locations of in-situ tripods (purple circles). 100 Figure A4.2 Examples of bed ripples. 101 Figure A4.3 ADCP record from F1 Expt 2.1. 102 Figure A4.4 ABS concentrations for F1 Expt. 2.1. 102 Figure A5.1 The X-band radar mounted on the flat roof of the lifeboat station at Sea Palling. 103 Figure A5.2 The recording system housed in a mobile rack inside the building. 104 Figure A5.3 A radar snapshot of the sea surface showing waves propagating through the breakwaters. 104 Figure A5.4 A short range high resolution radar derived bathymetric map of the breakwaters area. 105 Figure A5.5 A long range medium resolution radar derived bathymetric map extending almost 4km from the radar. 105 Figure A5.6 A radar derived water depth and current vector map. 105 Figure A5.7 Radar derived map of submarine dune features. 106 Figure A6.1 Breakwater scheme at Sea Palling and computational domain. 107 Figure A6.2 The measured waves and tides at F1 location. 108 Figure A6.3 Computed volumetric changes for each embayment and the nearshore area. 108 Figure A6.4 Computed & measured volumetric changes after 670hrs. 109 Figure A6.5 Computed volumetric changes/tide in relation to surge level (All bays). 110 Figure A6.6 Computed volumetric changes/tide in relation to surge level (Bay 0). 110 Figure A6.7 Computed volumetric changes/tide in relation to surge level (Bay Low). 111 Figure A7.1 Computational domain of the Sea Palling site. 113 Figure A7.2 Tidal water level and significant wave height used in E27 storm simulation. 113

xii Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

Figure A7.3 Computed ripple height and transport vector distribution at 15hrs (A) and 22hrs (B) around Reef 6 and Reef 7 114

Figure A7.4 Comparison of measured (A) and computed (B) ripple length around Reef 6 and 7. 114 Figure A7.5 Computed bed level change (BLC) after 1 tidal cycle based on ripple bed in A and the differences

between the model predictions using rippled bed and plane bed in B. 115 Figure A8.1 Location map 116 Figure A8.2 The Sea Palling scheme 116 Figure A8.3 Marginal distributions of wave height, period and direction of the original (lefthand column) and

simulated (righthand column) time series. 118 Figure A8.4 Cross-correlation functions of wave parameters for original (lefthand column) and simulated (righthand

column) series. 119 Figure A8.5 Plot of the mean shoreline position (time and ensemble averaged), together with the envelope of

minimum and maximum excursions throughout the scheme. 119 Figure B.1 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 123 Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 01 – no tide,

shore-normal waves, Layout 1. 123 Figure B.2 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 124 Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 02 – no tide,

oblique incident waves, Layout 1. 124 Figure B.3 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 125 Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 03 – 3m

progressive tide, shore-normal waves, Layout 1. 125 Figure B.4 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 126 Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 04 – 3m

progressive tide oblique incident waves, Layout 1. 126 Figure B.5 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 127 Figure B.6 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 128 Figure B.7 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 129 Figure B.8 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 130 Figure B.9 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 131 Figure B.10 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 132 Figure B.11 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 133 Figure B.12 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 134 Figure B.13 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 135 Figure B.14 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 136 Figure B.15 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 137 Figure B.16 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 138 Figure B.17 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 139 Figure B.18 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 140 Figure B.19 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 141 Figure B.20 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 142 Figure B.21 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. 143 Figure C.1 Simulated bathymetry contours after 60-day morphological simulation (bottom: profiles at 0, 7, 14,

21 and 28 days across A and B). 144 Figure C.2 Simulated bathymetry contours after 60-day morphological simulation (bottom: profiles at 0, 7, 14,

21 and 28 days across A and B). 145 Figure C.3 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0,

15, 30, 45 and 60 days across A and B). 146 Figure C.4 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0,












15, 30, 45 and 60 days across A and B). 158

Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts xiii

Figure C.16 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). 159



Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts 1

1 Introduction

1.1 Background Nearshore detached breakwaters are often considered an option for beach erosion control as part of coastal defence schemes. The dominant effect of a detached breakwater is to reduce the incident wave energy on a section of the coast and thereby reduce the sediment transport capacity in the sheltered region. In this way, detached breakwaters promote the accumulation of sediments, and hence shoreline accretion, in their lee.

Detached breakwaters have been used extensively in Japan, the US, Singapore and the Mediterranean. Their use in the UK is relatively recent (from 1980; CIRIA Beach Management Manual 1996). Examples of nearshore detached breakwaters around UK coasts are shown in Figure 1.1 and Figure 1.2.

Sea Palling, Norfolk

Elmer, West Sussex

Figure 1.1 Example breakwater schemes around UK coasts (1).

2 Modelling the effect of nearshore detached breakwaters on sandy macro-tidal coasts

Sidmouth, Devon

Jaywick, Essex

Figure 1.2 Example breakwater schemes around UK coasts (2).

Rogers et al. (2006) carried out a review of the existing design guidance for determining the geometrical layout of breakwater schemes and concluded that the existing guidance is largely based on empirical data from micro-tidal coasts (tidal range <2m). These data may not be applicable for meso- and macro-tidal coasts, which are common along UK coasts1. In an attempt to bridge this gap, the present research study was commissioned under the joint Environment Agency/Department for Environment, Food and Rural Affairs (Defra) Flood and Coastal Erosion Risk Management (FCERM) research programme. The aim of the study was to help improve practical design guidance for determining the geometrical layout of breakwater schemes on macro-tidal coasts.

1 More than 75 per cent of the UK coastline can be classified as meso- or macro-tidal coasts (see co-tidal chart 5058 from UK Hydrographic Office).


1.2 Effect of detached breakwaters The effect of a detached breakwater is to reduce the incident wave energy on a section of the coast in its lee. This reduction of wave energy in the lee of a breakwater scheme induces complex flow circulation patterns due to gradients in the wave set-up, wave-driven longshore flow and tidal flows, resulting in complex sediment transport patterns. In a meso- or macro-tidal environment, the littoral zone2 is continually changing as the water level changes with the tide. Furthermore, tidal currents also interact with wave-driven currents, leading to more complex flow and sediment transport patterns. These complex sediment transport patterns produce morphological changes in the vicinity of the breakwater. These changes include: a) sediment deposition in the lee of the breakwater; b) erosion in the breakwater bays; and c) scour near the breakwater heads.

Thus, understanding the likely incident wave and water level conditions, how the breakwater will influence the incident wave energy distribution and tidal flows on the beach, and the beach’s response to the new conditions are the three key elements for selecting an appropriate geometrical layout of a breakwater scheme. The shoreline response in the lee of a breakwater is typically classified as shown in Table 1.1.

Table 1.1 Description of accreted shorelines in the lee of a detached breakwater.

Shoreline response Description Example1

Limited response Limited changes in the shoreline planform due to sediment deposition leeward of breakwater.

Sidmouth, Devon

Salient Noticeable bulge in the shoreline planform due to sediment deposition leeward of breakwater.

Elmer, West Sussex; Jaywick, Essex

Tidal Tombolo Tombolo at low water, but salient at higher tide levels.

Sea Palling, Norfolk Second, third and fourth breakwaters (from bottom right) in Figure 1.1.

Tombolo Shoreline that has connected to breakwater due to sediment deposition leeward of breakwater.

Sea Palling, Norfolk First breakwater (from bottom right) in Figure 1.1.

1See the photographs of the example breakwater schemes in Figure 1.1. The morphological changes described above are controlled by the incident wave and water level conditions, the sediment characteristics and the geometrical layout of the breakwater scheme.

2 The littoral zone refers to the zone between the shoreline and a location offshore where significant longshore sediment transport takes place.


1.3 Outline design of detached breakwaters The outline design3 of a breakwater scheme consists of specifying the key geometrical parameters required in order to obtain a desired beach response under the prevailing wave and tidal conditions at the specified beach. The key geometrical parameters for a breakwater scheme that should be specified in an outline design are illustrated in Figure 1.3 and summarised in Table 1.2.

Figure 1.3 Definitions of key variables for nearshore breakwater scheme (adapted from USACE 2003).

3 Also known as functional design in shore protection literature from North America, as it relates to the design of a breakwater scheme that will serve the function of protecting a section of the shoreline.


Table 1.2 Characteristic parameters.

Parameter Description Hb Characteristic breaking wave height

T Characteristic wave period (or deep water wave length, L0)

θ Characteristic wave direction

Rtide Characteristic tidal range

Ttide Tidal period

Utide Characteristic tidal current

Inci

dent

wav

e an

d tid

e pa

ram

eter

s

φ Phase shift between maximum tidal current and high water

LS Length of the breakwater

X Cross-shore distance of the breakwater relative to a characteristic initial shoreline (mean sea level shoreline)

G Gap distance between adjacent breakwaters

hcr Elevation of the breakwater crest relative to mean sea level.

dcr An alternative parameter to hcr is the depth of water at the breakwater crest during high water, denoted as dcr

This parameter is used in the dimensional analysis (Chapter 3) as it is more meaningful in terms of wave transmission over the breakwater. Note that dcr is either zero (emergent breakwater over tidal cycle) or positive (breakwater submerged at high water)

Bre

akw

ater

par

amet

ers

B Breakwater crest width m Average beach slope (or, when combined with Hb, this parameter can

be represented by Xb = distance from the shoreline to breaker line; Xb is physically more meaningful and so is used hereafter)

D50 Median sediment size

Bea

ch p

aram

eter

s

Sg Sediment gradation 1684 / DD

The outline design is different from the structural design of the breakwater, which is aimed at ensuring that the materials and cross-section are selected such that the breakwater maintains its structural integrity. Information on the structural design of breakwaters can be obtained from the CIRIA Rock Manual (2007).

The FCERM research programme commissioned the present research study with the objective of improving practical design guidance for determining the geometrical layout of breakwater schemes on macro-tidal coasts. The approach used in this study consists of the following components.

1) Dimensional analysis to identify the important parameters for describing the bathymetry changes in the vicinity of nearshore detached breakwaters on a macro-tidal coast.

2) Numerical simulation of bathymetry changes for several generic test cases (a total of 30 test cases were investigated) using coastal area morphological models to understand the impact of breakwaters.


3) Analysis of the results to obtain trends in the morphological response and provide improved guidance for the outline design of breakwaters on macro-tidal coasts.

Furthermore, this study builds on parallel studies funded by the UK Engineering and Physical Sciences Research Council (EPSRC), which investigated coastal processes near the breakwater scheme at Sea Palling on the North Norfolk coast.

1.4 This report This report presents the main scientific results of the research study. An accompanying guidance note has also been prepared, which can be used by practitioners to develop an outline design for breakwater schemes. The remainder of this report is organised as follows.

Chapter 2 contains an extended summary of the literature review on existing design guidance carried out by Rogers et al. (2006). The review is updated with additional information and an inventory of breakwaters around UK coasts.

Chapter 3 contains details of the dimensional analysis carried out to identify the important parameters for describing the bathymetry changes in the vicinity of nearshore detached breakwaters on a macro-tidal coast.

Chapters 4 and 5 contain descriptions of the morphological modelling of the generic test cases carried out using PISCES and MIKE 21 CAMS respectively. Each of the two chapters also contains an analysis of the simulation results and details the main trends that can be inferred from the simulations.

Chapter 6 contains an analysis of the simulation results presented and discussed in chapters 4 and 5. The aim of this analysis is to establish the main trends in the beach response and to develop design graphs that can be used to conduct an outline design of breakwaters in macro-tidal environments.

Chapter 7 contains a summary of the overall conclusions and suggestions for further work.

Appendix A contains a collection of short papers on the EPSRC study, which investigated coastal processes near the breakwater scheme at Sea Palling, UK.

Appendix B and Appendix C contain the morphological modelling results from PISCES and MIKE 21 CAMS respectively.

1.5 Convention In this report, the following conventions are used.

• Unless stated otherwise, reef is used to describe a breakwater that is placed some distance offshore in order to shelter the shoreline from wave action and to promote sediment accretion in the lee. The term is used interchangeably with nearshore breakwater throughout the report.

• Unless explicitly stated otherwise, the sediment considered in this report is sand – sediments with a grain size coarser than 0.063mm and finer than 2mm.


2 Literature review This chapter presents a synopsis of the literature review developed at the scoping stage of the project by Rogers et al. (2006), supplemented with further information as appropriate.

2.1 Existing design guidance

2.1.1 Introduction to beach control breakwaters

Detached breakwaters are generally built parallel to the shore in shallow water (typically water depths of 5m or less relative to mean sea level (MSL)). They have no structural connection to the shoreline and so currents and sediment can pass between the structure and the waterline, although this transport may be significantly less than on the original open coast.

The reduction in wave energy in the lee of the breakwaters slows the littoral drift and encourages the formation of a shoreline bulge or other salient feature in the sheltered area behind the breakwater. This occurs as a result of local changes in littoral transport relating to wave diffraction, refraction and currents (Ilic et al. 2005 a,b). Under certain breakwater geometry and coastal conditions, the salient can extend seaward sufficiently to reach the breakwater, in which case it is usually termed a tombolo. The complex wave and flow patterns in the gaps between the breakwaters also lead to shoreline erosion and the formation of a bay within the gap.

The preferred shoreline response in the lee of a detached breakwater system is normally a salient rather than a tombolo (USACE 2003). This is partly because salients potentially allow alongshore sediment to continue to move through the project area to downdrift beaches, minimising the wider impact of the scheme. Another reason for preferring salient to tombolo formation is that the former naturally limits risks related to public access to coastal structures. However, when assessing the feasibility of new breakwater schemes the required beach response needs to be assessed from actual project requirements.

Salients are likely to predominate when the breakwaters are sufficiently far from shore, short relative to the incident wavelength and relatively transmissible (low crested or large gaps with low sediment input). Wave action and longshore currents tend to keep the salient from connecting to the structure.

Nearshore breakwaters can be designed to perform a similar function to natural bars, reefs or nearshore islands that dissipate wave energy, reduce erosion and encourage sedimentation. Multiple detached breakwaters spaced along the shoreline can provide substantial protection to shoreline frontages. For example, the Sea Palling scheme on the Norfolk Coast described by Fleming and Hamer (2000) provides protection for approximately 4000m of shoreline frontage.

As described in section 2.2 below, nearshore breakwaters have been used as part of shoreline protection schemes at several locations around the UK coast. When properly designed, detached breakwaters can be very effective at reducing erosion and in building up beaches using natural littoral drift. Moreover, they can also help to retain artificially-nourished beach material.


2.1.2 Structural design of breakwaters

Detached breakwaters are normally built as rubble-mound structures. They may typically be constructed from rock armour or from proprietary concrete armour units that are designed to interlock and efficiently dissipate wave energy. Alternative forms of construction that have been considered for use in nearshore detached breakwaters (for beach stabilisation) include large geotextile sand bags, sunk scrap barges and small ships, scrap sections of offshore oil industry structures and scrap tyres. However, no attempt to assess the viability of any of these alternatives has been attempted under this project.

This project and the design guidance focus on the geometric layout of nearshore detached breakwater schemes rather than the structural design of the breakwater structures. Other guidance should be referred to for structural design. For example, the design of rock armour breakwater cross-sections is dealt with in the Rock Manual (CIRIA 2007). The Coastal Engineering Manual (CEM; USACE 2003) describes the design of both rock armour and concrete armour units. When designing concrete armour units, it is usual to refer to BS6349 Part 7 and proprietary information from those companies that license the use of specific armour units.

2.1.3 Design guidance for non-tidal or micro-tidal locations

A comprehensive review of current knowledge on the design of nearshore detached breakwaters was undertaken during the scoping phase of this project by Rogers et al. (2006). This review highlighted a number of sources that are available to assist with the design of detached breakwaters. These sources are largely based on worldwide experience of built schemes, the majority of which are in non-tidal or micro-tidal situations.

For micro-tidal coasts, the current state of understanding and recommendations for design are described comprehensively in the CEM. That guidance is based on prototype experience, which in the US is generally limited to sediment-starved shores with fetch-limited wave climates on the Great Lakes, Chesapeake Bay and the Gulf of Mexico. Additional experience is quoted from extensive use of breakwaters for shore protection in Japan, Israel, Denmark, Singapore and Spain. The CEM quotes several studies that have reviewed the schemes, a key aspect of which was determining the conditions that lead to either salients or tombolos forming behind the breakwaters. Figure 2.1, based on Rosati (1990), summarises existing guidance for micro-tidal situations.

An alternative design guidance based on the observed beach response at various nearshore breakwater project sites along US micro-tidal coasts is shown Figure 2.2.


L1, L2 and L3 are the 3 generic test layouts modelled in this study.

Figure 2.1 Existing design guidance for morphological response in the lee of detached breakwaters.

Ls / G

X / d

s

Figure 2.2 Dimensionless plot of nearshore breakwater projects (from USACE 2003).


2.1.4 UK design guidance for shingle beaches

The studies for this project have only considered sand beaches. However, for those readers concerned with shingle beaches, a preliminary design method for single nearshore breakwaters on shingle beaches is presented in Box 8.3 of the CIRIA Beach Management Manual. That approach, based on Coates (1994), shows that, in theory, single detached breakwaters can be designed to retain shingle recharge material while allowing natural drift to continue on a beach with a strongly dominant drift direction. The method helps to determine a suitable range of breakwater length, freeboard and offshore distance to achieve a dynamically stable shingle beach. However, the manual makes it clear that ‘Design procedures for multiple breakwaters on shingle beaches have only reached an embryonic stage’ and that ‘At present this design approach is embryonic, but further work could provide a general method’.

Physical scale model investigations of hydrodynamics and morphodynamics at the Elmer detached breakwater scheme in West Sussex are reported in Ilic et al. (2005a,b). At Elmer, the beach material is coarse and the investigations reported by Ilic et al. did not consider the importance of tidal range or tidal currents and so cannot feed directly into the current work. However, many of their findings are still relevant to the design of detached breakwaters for UK coastal sites. The physical model studies found that wave direction, directional spreading and wave transmission through the breakwaters are important processes that influence the hydrodynamics in the embayments. They found that many complex hydrodynamic processes are induced by the construction of offshore detached breakwater schemes, including gyres and long period oscillations due to surf beat.

Although the studies did not report numerical modelling, Ilic et al. suggested that most of the numerical models available for use in design at the time did not include all of the necessary processes. They recommended that coastal area modelling should employ a wave model that takes account of wave refraction, diffraction, reflection, transmission and dissipation, as well as wave–current interaction and the effect of random multidirectional waves.

Ilic et al. (2005b) confirmed that very little is known concerning the effect of tidal level variations and tidal flow on the formation of salients and tombolos at detached breakwater schemes. They also confirmed that the design process for macro-tidal environments and mixed sediment beaches is more demanding than for micro-tidal locations.

2.1.5 Design guidance for UK meso-tidal or macro-tidal locations

The scoping phase of this project found that there is very limited design guidance for meso or macro-tidal coasts. Much, if not all, of the outline design guidance prior to the early 1990s was developed from analysis and observations of beaches in relatively sheltered and micro-tidal situations. The introduction of tides adds significant complications to the design process. In addition to the varying tidal level at which the wave energy is applied to the shoreline, tides add the complicating factor of tidal currents, which interact with the wave-driven currents created in the lee of the breakwater system.

The CIRIA Beach Management Manual highlights the fact that detached breakwaters can induce very strong changes to the coastal processes regime and so a thorough appreciation of their likely impact is a prerequisite for their use. This is especially the case for the macro-tidal conditions found around the UK coast, where experience of detached breakwaters is limited.


The CEM states that: ‘Optimizing detached breakwater designs is difficult when large water level variations are present, as is the case on coastlines with a large tidal range or in portions of the Great Lakes, which may experience long-term water level fluctuations.’

Fleming and Hamer (2000) evaluated the performance of the nearshore breakwater system at Sea Palling on the Norfolk coast and provided some preliminary guidance for the functional design of breakwater schemes in similar tidal regions. At Sea Palling, the spring tidal range is around 3m (meso-tidal), there is exposure to significant wave heights with an annual average of close to 2m and the tides are progressive, such that maximum tidal currents occur near high water (Fleming and Hamer 2000).

In the case of the Sea Palling nearshore breakwater scheme, Fleming and Hamer compared the field response to the two nearshore breakwater configurations used at Sea Palling. They found that the existing design guidance in Figure 2.2 was inadequate to distinguish the actual response and concluded that this was largely due to the complications added by the increased tidal range and the increased currents. However, the relationships presented in Figure 2.1 could give preliminary guidance for determining Ls/X (based on the desired beach response in the lee of the breakwater). They also suggested another design graph to determine the spacing of the breakwaters (G/X) based on the maximum erosion in the bay. Their method is summarised in Figure 2.3 and can also be described as follows.

• First, fix the offshore distance by reference to the sediment transport pathways. If it is not desirable for the nearshore breakwater system to have a major impact on the nearshore – as opposed to beach face – longshore sediment transport, then it should be located inshore of any nearshore features that may be primary sediment pathways.

• Secondly, having decided upon an optimum offshore distance, relationships based on previous field experience, where relevant (see Figure 2.1), can be used to determine the nearshore breakwater lengths required to produce different forms of beach response. Depending on the desired result, decisions may be taken to allow the preferred beach response shape to develop, forming either minimal response, salients or tombolos. Clearly, tombolos will be more disruptive than salients to the longshore movement of sediment, but they will offer more protection during severe storms4 and will offer a greater amenity area.

• Lastly, the gap width between the nearshore breakwaters is determined based on a consideration of the maximum erosion that is acceptable within the breakwater bays.

The above approach, while useful for determining the outline design of a nearshore breakwater scheme, omits other parameters (such as tidal parameters and the relative location of the breakwater in the surf zone) that affect the morphological response in the vicinity of nearshore breakwaters. The effects of these additional parameters are investigated in the present study.

4 A tombolo offers better protection than a salient during a storm as the wave energy is dissipated over a larger area, reducing the incident waves at flood defences. Furthermore, since more sediment is stored in a tombolo, it will take longer (compared to salient) for erosion to expose any flood defence.


Figure 2.3 Outline design procedure suggested by Fleming and Hamer (2000).

Figure 2.4 Existing design guidance for assessing possible shoreline erosion in the gaps between nearshore breakwaters.

2.2 Inventory of UK breakwaters In order to assess previous nearshore breakwater schemes around the UK, we have collated data on the geometric parameters of selected key beach control breakwater schemes.

STAGE 1: Determine X from amount of longshore transport to be bypassed to downdrift beaches. Fix X by reference to sediment pathways.

STAGE 2: Determine LS based on desired beach response (salient or tombolo) Use Figure 2.1 to determine LS/X and hence determine LS.

STAGE 4: Determine G (gap width) based on consideration of erosion in breakwater bays. Use Figure 2.4 to determine G/X and hence determine G.


The data were collated from several sources:

• internal reports and information available to the project team;

• papers in proceedings of conferences and journals;

• information sourced from the Environment Agency/Arun district council;

• information sourced from other consultants (Royal Haskoning);

• Google Earth;

• Admiralty tide tables;

• Admiralty charts.

A summary of the collated breakwater schemes and dimensions is shown in Table 2.1. We found it difficult to determine the location of the initial shoreline5 (which was required to determine X = distance between breakwater and shoreline). Hence, engineering judgement has been used to estimate this parameter. In Table 2.1, the breakwater response is classified as salient (S), tombolo (T) or no sinuosity (N). These breakwater schemes are all located in meso- and macro-tidal coasts.

Table 2.1 Inventory of breakwater schemes around UK (T■=Tombolo, S∆=Salient, N×=No sinuosity).

ID Scheme location Ls (m)

G (m)

X (m)

ds (m) Ls/G X/ds

Rtide1

(m)

Shoreline response (S, T, N)

1 & 2 80 75 130 4.3 1.1 30.2 S∆ 3 to 5 140 75 130 4.3 1.9 30.2 S∆ 6 80 130 130 4.3 0.6 30.2 S∆ 7 80 100 100 4.3 0.8 23.2 S∆

1 Elmer, West Sussex

8 80 80 80 4.3 1.0 18.6

5.38

T■ 1 90 65 100 4.0 1.4 25.0 N× 2 Sidmouth, Devon2 2 30 65 150 4.0 0.5 37.5 5.2 N×

3 Jaywick, Essex 180 250 170 6.0 0.7 28.3 4.0 S∆

4 Monk's Bay, Bonchurch, Isle of Wight 50 5003 50 3.1 1.0 16.1 3.1 T■

Western 150 500 150 3.0 0.3 50.0 S∆ 5 Leasowe Bay, Wallasey, Wirral Eastern 130 500 125 3.0 0.3 41.7 8.4 T■

Stage 1 195 240 250 4.0 0.8 62.5 2.6 T■ 6 Sea Palling, Norfolk Stage 2 140 160 250 4.0 0.9 62.5 S∆

7 Tain, Dornoch Firth, Scotland

8 South of Harwich, North of Walton on the Naze, Essex

9 Directly south of West Mersea, Essex

No data found.

Notes: 1 Rtide = tidal range (mean high water spring minus mean low water spring) 2 Breakwater scheme is oblique to shore. 3 Single breakwater, G set to distance to adjacent groin.

5 The shoreline is defined as the intersection of the mean high water spring line with the coastal land.


The data has been plotted on the existing design graphs (Figure 2.1 and Figure 2.2) and are shown in Figure 2.5 and Figure 2.6. These figures show that the expected shoreline response is not always the same as the actual response in the field. This is possibly because of the influence of tides, but could also be due to the influence of other parameters of the beach-breakwater interaction that are not included in the design graphs (such as the relative location of the breakwater in the surf zone, X/Xb). However, Figure 2.5 is able to predict reliably the shoreline response at most of the Elmer breakwaters (most of the breakwater dimensions fall within the range for salients, which is the same as the observed response).

Overall, the figures illustrate the difficulty of using existing design graphs to predict the expected morphological response of a breakwater scheme along the meso- and macro-tidal coasts of the UK.

0

10

20

30

40

50

60

70

80

0 1 2 3 4 5

Ls/G

X/ds

SP Stage 1SP Stage 2Leasowe WesternLeasowe EasternElmer 1&2Elmer 3-5Elmer 6Elmer 7Elmer 8Sidmouth 1Sidmouth 2Monks BayJaywick

Salients

No Sinuosity

Tombolos

Figure 2.5 Comparison of field data around UK with existing design graph (after Pope and Dean, 1986; ■=Tombolo, ∆=Salient, ×=No sinuosity).


Tombolos

Gourlay (Field) (1981)

SPM (1984)

Toyoshima (1972)

Ahrens & Cox (1990)

0 1 2 3 4 5Ls/Xi,Ls/X

SP Stage 1SP Stage 2Leasowe WesternLeasowe EasternElmer 1&2Elmer 3-5Elmer 6Elmer 7Elmer 8Monks BaySidmouth 1Sidmouth 2Jaywick

Salients

Gourlay (Lab)(1981)

SPM (1984)

Dally & Pope (1986)

Ahrens & Cox (1990)

0 1 2 3 4 5Ls/Xi,Ls/X


Limited Response

NIR (1982)

Ahrens & Cox (1990)

0 1 2 3 4 5Ls/Xi,Ls/X


Figure 2.6 Comparison of field data around UK with existing design graph (after Rosati 1990; ■=Tombolo, ∆=Salient, ×=No sinuosity).


3 Dimensional analysis In this section, the dimensional analysis carried out to identify the important dimensionless parameters governing the morphological response behind detached breakwaters is presented.

A partial list of dimensionless parameters governing morphological behaviour behind emerged breakwaters has previously been presented by Perlin (1979), Hanson and Kraus (1990) and Johnson et al. (1995). The work presented here extends the earlier work by including the effect of tides.

3.1 Characteristic parameters The three key elements for selecting an appropriate geometrical layout of a breakwater scheme are: 1) the likely incident wave and water level conditions; 2) how the structure will influence the incident wave energy distribution and tidal flows on the beach; and 3) the beach’s response to the new conditions.

The independent parameters representing incident wave and tide conditions, breakwater layout and beach conditions are summarised in Table 1.2. The key breakwater parameters are illustrated in Figure 1.3.

3.2 Dimensionless parameters The independent characteristic parameters are suitably combined to form dimensionless parameters using dimensional analysis. The main result is that the morphological response behind multiple detached breakwaters on a macro-tidal coast depends on the list of parameters shown in equation 3.1.

),,',,,,,,,,(05000

MWDgH

ULgT

HR

SDH

Hd

LB

LG

XX

XL

fb

tidetide

b

tideg

b

b

cr

b

SA ϕ=Π (3.1)

A physical interpretation of the dimensionless parameters is given in Table 3.1. Not all the dimensionless parameters will be important in some situations, leading to a reduced number of parameters for describing the morphological response. Example simplifications are presented in Table 3.2 for fixed sediment parameters and fixed tidal period (not considering Hb/D50, Sg and 0/ LgTtide ).


Table 3.1 Dimensionless parameters for morphological response behind shore parallel breakwaters.

Parameter Description LS/X Breakwater blocking efficiency

X/Xb Percentage of littoral drift affected by breakwaters (a measure of the

relative location of breakwater in surf zone)

G/L0 Wave penetration through gaps

B/L0 Wave energy dissipation distance over the breakwater crest

dcr/Hb Wave energy dissipation rate over breakwater

Hb/D50 Sediment mobility parameter

Sg Sediment grading

Rtide/Hb Effect of tide range on surf zone

0/ LgTtide Tidal period relative to characteristic wave period

btide gHU / Effect of tidal current relative to wave induced current

ϕ Effect of type of tidal regime

θ Effect of wave direction

Table 3.2 Key dimensionless parameters for morphological response for different nearshore breakwater schemes.

Nearshore breakwater scheme

Controlling dimensionless parameters

Multiple breakwaters ,,,,,,,,(

000

ϕb

tidetide

b

tide

b

cr

b

SA gH

ULgT

HR

Hd

LB

LG

XX

XL

f=Π

Multiple breakwaters, no tides, submerged

),,,,,(00

θb

cr

b

SA H

dLB

LG

XX

XL

f=Π

Multiple breakwaters, no tides, emergent

),,,(0

θLG

XX

XL

fb

SA =Π

Single breakwater, no tides, submerged

),,,,(0

θb

cr

b

SA H

dLB

XX

XL

f=Π

Single breakwater, no tides, emergent

),,( θb

SA X

XXL

f=Π

Single breakwater, no tides, emergent, constant wave direction

),(b

SA X

XXL

f=Π


3.3 Generic test cases Twenty generic test cases were established in the first year of the project to investigate the morphological response to various combinations of breakwater layouts and wave and tidal conditions. The 20 cases are shown in Table 3.3 (Simulation nos. 01 to 20) and consist of two breakwater layouts x two wave directions x five tidal conditions. The five tidal conditions consist of one non-tidal case, and two tidal ranges (3m and 5m) x two tidal types (progressive and standing tides). These are the 20 test cases envisaged in the original proposal for the study.

Using these test cases, the effect of the dimensionless parameters on the morphological response can be studied, except for the parameters G/L0, B/L0 and dc/Hb. However, the number of cases for each parameter is rather limited. For instance, the two breakwater layouts implied that we have only two Ls/X points that can be extracted for each incident wave and tidal condition, which makes it difficult to derive general trends. Thus, 10 additional test cases were added to include an additional layout (layout 3) and to investigate the impact of breakwater crest level (dc/Hb) and reduced incident wave height conditions.

All the test cases start from a 1:50 plane sloping beach extending to an offshore depth of 15m relative to MSL. Each breakwater layout consists of four shore-parallel breakwaters, each with a length of 200m and a gap width between them of 300m. The distance from the initial MSL shoreline to the centre of the breakwater is 250m (layout 1), 150m (layout 2) and 350m (layout 3). The three layouts are shown schematically in Figure 3.1. The tidal ranges investigated are 0m (no tides), 3m and 5m, while the tidal period is selected as 12.42hrs.


Table 3.3 Generic modelling test cases.

Waves6 Simulation no.

Breakwater crest level, hcr (m MSL)

Hm0 (m)

Tp (s)

θ7 (deg)

Tides Layout

01 2 2 8 90 02 2 2 8 45 No tides

03 2 2 8 90 04 2 2 8 45

3m, Progressive

05 2 2 8 90 06 2 2 8 45

3m, Standing

07 2 2 8 90 08 2 2 8 45

5m, Progressive

09 2 2 8 90 10 2 2 8 45

5m, Standing

L1

11 2 2 8 90 12 2 2 8 45 No tides

13 2 2 8 90 14 2 2 8 45

3m, Progressive

15 2 2 8 90 16 2 2 8 45

3m, Standing

17 2 2 8 90 18 2 2 8 45

5m, Progressive

19 2 2 8 90

Initi

al te

st c

ases

20 2 2 8 45 5m,

Standing

L2

A 2 1 5 90 L1 B 2 1 5 90

No tides L2

07B 1 2 8 90 07C 3 2 8 45

5m, Progressive

L1

19B 1 2 8 90 19D 0 2 8 45

5m, Standing

L2

21 2 2 8 90 No tides 24 2 2 8 45 3m,

Progressive 25 2 2 8 90 3m,

Standing Add

ition

al te

st c

ases

29 2 2 8 90 5m, Standing

L3

6 The offshore wave parameters (wave height Hm0, peak period Tp and wave direction θ) are all specified at a water depth of 15m relative to MSL. 7The wave direction (θ) is the angle in degrees between the initial shoreline and the direction of wave propagation.


Figure 3.1 Simulated breakwater layouts.

Two modelling systems (PISCES and MIKE 21 CAMS), described in Chapters 4 and 5 respectively, were used to simulate all the 30 test cases. The original 20 test cases plus simulation no. 24 were simulated using the PISCES morphological model (described in Chapter 4). The additional test cases (except simulation no 24) and a subset of the original test cases were simulated using MIKE 21 CAMS (described in Chapter 5). The two different model systems provide additional insight into the effect of representing different processes on the morphological response.


4 Morphological modelling using PISCES This chapter describes the study of the tidal influence on the morphological evolution in the vicinity of detached nearshore breakwaters using HR Wallingford’s numerical modelling system PISCES. Section 4.1 summarises the simulated test cases, while section 4.2 provides a description of PISCES. Section 4.3 presents details of the set-up of the individual process modules and their coupling. In Section 4.4, the results are presented and discussed, followed by a summary in Section 4.5.

4.1 Simulated test cases The PISCES system was used to simulate the 20 generic test cases defined at the outset of this project, as well as one additional case for a third layout in which the breakwaters were located 350m from the initial MSL shoreline.

Table 4.1 Generic test cases simulated using PISCES.

Waves Simulation no.

Breakwater crest level, hcr (m MSL)

Hm0 (m)

Tp (s)

θ (deg)

Tides Layout

01 2 2 8 90 02 2 2 8 45 No tides

03 2 2 8 90 04 2 2 8 45

3m, Progressive

05 2 2 8 90 06 2 2 8 45

3m, Standing

07 2 2 8 90 08 2 2 8 45

5m, Progressive

09 2 2 8 90 10 2 2 8 45

5m, Standing

L1

11 2 2 8 90 12 2 2 8 45 No tides

13 2 2 8 90 14 2 2 8 45

3m, Progressive

15 2 2 8 90 16 2 2 8 45

3m, Standing

17 2 2 8 90 18 2 2 8 45

5m, Progressive

19 2 2 8 90

Initi

al te

st c

ases

20 2 2 8 45 5m,

Standing

L2

24 2 2 8 45 3m, Progressive

L3

Each of the three layouts comprised an initial bathymetry defined as a plane sloping beach with a slope of 1:50 and included an array of four shore-parallel detached breakwaters with a length of 200m, a gap width of 300m and a crest level of 2m above MSL. The breakwaters were located 250m (L1), 150m (L2) and 350m (L3) from the initial MSL shoreline, in an initial water depth of 5m, 3m and 7m respectively. Layout 1 is presented in Figure 4.2.


For Layouts 1 and 2, the simulations were carried out for shore-normal and 45° oblique incident offshore wave directions, with an offshore significant wave height of 2m and a peak period of 8s. The simulations were carried out for non-tidal and four tidal conditions, consisting of progressive (with maximum velocities occurring near high water) and standing (with maximum velocities occurring near mean sea level) tides with a tidal range of either 3m or 5m. One additional test case was later added for Layout 3: a 3m progressive tide and oblique incident waves.

All cases were simulated for uniform sediment with a grain size of 0.2mm, as defined at the outset of this project.

4.2 Model description The developed test cases were simulated using the PISCES finite element morphodynamic modelling system. PISCES (developed by HR Wallingford) consists of individual modules for waves, flow, sediment transport and morphological evolution. These modules are linked, allowing them to exchange information and work interactively. Amongst other applications, PISCES has been applied successfully to: simulate morphology in the lee of nearshore detached breakwaters (Nicholson et al. 1997); investigate nearshore bar formation and evolution (Damgaard et al. 2003); simulate morphology in a complex estuary mouth environment (Sutherland et al. 2004); and investigate infill of dredged sand pits (Van Rijn et al. 2005).

4.2.1 Wave model

Two separate wave models are available within the PISCES modelling system: the second generation wave model TOMAWAC (developed by the National Hydraulics and Environment Laboratory (LNH, Paris) of Electricité de France (EDF)); and the third generation wave model SWAN (developed by Delft University of Technology in the Netherlands).

TOMAWAC wave model

The second generation wave model TOMAWAC was used to model the instantaneous wave fields over the model domain. TOMAWAC computes the generation and propagation of waves in coastal waters for given wind, seabed and current conditions. It is particularly suited to modelling the generation and transformation of waves over relatively large coastal areas.

TOMAWAC is part of EDF’s TELEMAC suite and is based on the finite element method used throughout TELEMAC. TOMAWAC is a second generation wave model, in that it restricts the two-dimensional (2D) energy density to a pre-defined spectral shape in the frequency range (for example, a JONSWAP spectrum). The model is based on the numerical solution of the wave action density balance equation, given in terms of unknown quantities m0 and m1, where m0 and m1 are the zeroth and first moments of spectral density respectively.

TOMAWAC is based on the assumption of steady state conditions – no variation in time. These features result in a model that is computationally efficient in terms of both memory and floating point operations. The model has been validated for a variety of test cases and has already been used in numerous studies.


It is designed to represent the following wave propagation processes:

• shoaling due to spatial variations in seabed and current;

• refraction due to spatial variations in seabed and current;

• blocking by opposing currents;

• generation by wind;

• dissipation by whitecapping;

• dissipation by depth-induced wave breaking;

• dissipation by seabed friction or percolation;

• wave-wave interactions (quadruplets).

Diffraction by coastal structures or features of the seabed is not modelled in TOMAWAC, nor are reflections due to structures or significant seabed irregularities such as large sand banks.

More details on the TOMAWAC wave model can be found in Benoit et al. (1996).

SWAN wave model

SWAN (Simulating WAves Nearshore) is a computational spectral wave transformation model. It can be used to obtain realistic estimates of wave parameters in coastal areas, lakes and estuaries from given wind, seabed and current conditions. The model was developed by Delft University of Technology.

SWAN is based on a fully spectral representation of the wave action balance equation (or energy balance in the absence of currents) with all physical processes modelled explicitly. No a priori limitations are imposed on the spectral evolution. This makes SWAN a third-generation wave model.

The model has been used successfully at numerous sites around the UK and in other parts of the world. It is designed to represent the following wave propagation processes:

• refraction due to spatial variations in seabed and current;

• shoaling due to spatial variations in seabed and current;

• blocking and reflections by opposing currents;

• transmission through, blockage by or reflection from obstacles (such as coastlines or breakwaters).

The following wave generation and dissipation processes are also represented in SWAN:

• generation by wind;

• dissipation by whitecapping;

• dissipation by depth-induced wave breaking;

• dissipation by seabed friction;

• wave-wave interactions (quadruplets and triads);


• energy transmission through and reflection against obstacles;

• Diffraction (phase-decoupled approach).

In SWAN, diffraction is modelled in a phase-decoupled approach that results in the same qualitative behaviour of spatial redistribution and changes in wave direction as would result from a full, computationally-expensive solving of diffraction.

The SWAN wave model was conceived to be a computationally feasible third-generation spectral wave model for waves in shallow water (including the surf zone) with ambient currents.

More details on the SWAN wave model can be found in sources such as Booij et al. (1999) and Ris et al. (1999).

Comparison of TOMAWAC and SWAN

In general, when HR Wallingford’s PISCES system is applied, the TOMAWAC wave model is used, This wave model has been shown to produce realistic area-wide wave patterns for a large range of coastal area studies. As it is part of the TELEMAC modelling framework, it works on the same finite element mesh and uses the same conventions as the TELEMAC flow model and the SANDFLOW sand transport model, making transfer of information between the models straightforward and efficient.

For the present project, the possibility of applying the SWAN wave model was also investigated, as it is thought that diffraction of wave energy (a process not included in the TOMAWAC model) may play an important role in the morphological response in the lee of detached offshore breakwaters. This approach involves developing automated operations to allow information on water level and current fields, wave force and wave orbital velocity fields and the updated model bathymetry to be transferred between the models. This includes transferring the information between the finite-element TELEMAC mesh and the finite-difference SWAN grid, and modifying the transferred variables so that they are defined according to the definitions and conventions of each of the applied models.

Both wave models were applied to simulate the wave patterns on the basis of the initial model bathymetry, non-tidal conditions and shore-normal waves with an offshore significant wave height of 2.0m and peak period of 7.0s. The results of these simulations are shown in Figure 4.3 and Figure 4.4.

The wave model results show similar overall patterns and both clearly show the sheltered areas behind the breakwaters. Differences are seen in the nearshore zone, where somewhat higher wave heights are seen in the TOMAWAC results compared to the SWAN results, particularly in the gaps and nearer the shoreline behind the breakwaters (the sheltered zone is smaller in the TOMAWAC results). The relatively high wave forces seen on the offshore slope of the breakwater are the result of waves breaking on this steep slope and are of a similar order of magnitude in both models. It is noted that, as a result of post-processing, the wave force vectors were not plotted in the SWAN results.

For simulating wave-driven currents, the gradients in wave energy are the most significant parameter, as the wave forces that induce wave-driven currents are directly related to these gradients. The wave forces computed by the models from these wave energy gradients (expressed as an acceleration term of the water column) are also presented in Figure 4.3 and Figure 4.4. These show that: a) the wave forces are larger in the TOMAWAC results; and b) the direction of the wave forces is more strongly directed toward the sheltered areas in the lee of the breakwaters in the SWAN results.


These differences are the result of: differences in the formulation of the underlying processes; a different numerical schematisation; and the fact that the SWAN model includes diffraction. Without diffraction, wave energy levels in the lee of the breakwater may be under-estimated. Conversely, the spreading of wave energy behind the breakwater will also depend on other processes, including directional spreading (included in both wave models; see, for example, Enet et al. 2006) and numerical diffusion, which will have a similar effect to diffraction even though it is an artificial process.

For the simulated generic cases, where there is an absence of specific validation data, it is not possible to state whether or not the combined effects detailed above lead to an over- or under-estimate of the wave energy in the lee of the breakwaters.

In this project, the investigations of tidal impact on morphological response have been carried out by comparing morphological model results for the various tidal conditions against the non-tidal scenario. As a result, it was considered that, when comparing those results from the simulations that are identical except for the varying tidal conditions, the described effects of diffraction, directional spreading and numerical diffusion are of secondary concern. Therefore, and in the light of the computationally expensive simulations to be undertaken (with simulation times of the order of three weeks), the more efficient and accurate (due to avoidance of required interpolations) coupling of the TOMAWAC wave model to the TELEMAC and SANDFLOW models was selected for the morphodynamic simulations.

4.2.2 Flow model

The TELEMAC2D flow model, part of the TELEMAC modelling system developed by EDF-LNH in Paris, has been used to simulate the tidal and wave-driven flows.

TELEMAC2D is a sophisticated flow model for free surface flows. It solves the 2D depth-integrated shallow water equations used to model flows in rivers, estuaries and seas.

TELEMAC2D employs finite element techniques, allowing the use of very flexible unstructured triangular meshes. It has been developed under a quality assurance system that includes application to a standard set of validation tests.

The model can simulate depth integrated tidal flows in estuaries and seas, including the presence of drying banks. It can also simulate flows in rivers, including turbulence structures resulting from flow obstructions and trans-critical flows (Froude numbers of about 1).

More details about the TELEMAC2D model can be found in Hervouet (1997).

4.2.3 Sand transport model

HR Wallingford’s non-cohesive sediment transport model SANDFLOW was used for the present studies.

SANDFLOW is a sand transport modelling module developed by HR Wallingford for use with the TELEMAC system. SANDFLOW uses flows calculated by TELEMAC2D and wave orbital velocities calculated by the TOMAWAC wave model to calculate the transport, deposition and erosion of non-cohesive (sandy) sediment, and thereby identify areas of potential siltation and erosion.

SANDFLOW uses the same finite element mesh as the hydrodynamic modules. More details about the SANDFLOW sand transport model can be found in Miles (1991).


4.2.4 Bed level updating

The process modules for waves, flow and sediment transport are linked through a morphodynamic modelling script that keeps track of the time management and information exchange between the modules. This script can be adapted for specific morphological studies according to the characteristics of the case being investigated and the type and sequence of the conditions being simulated. The script used for the model investigations in this study is described in more detail in Section 4.3.4.

4.3 Set-up of numerical process models Separate finite-element model meshes and geometries were set-up for the three layouts, starting from a domain size of 1000m cross-shore by 8000m alongshore. The meshes were optimised to represent accurately the bathymetry and processes in the vicinity of the breakwaters (10m node spacing). They were also set-up to minimise computational requirements by reducing the resolution near the offshore and cross-shore open boundaries (40m spacing), as computationally expensive long-term morphodynamic simulations were to be made. These meshes have approximately 24,000 elements and 12,000 nodes. The mesh and bathymetry for Layout 1 are shown in Figure 4.2.

4.3.1 Wave model

The model boundaries were defined such that wave energy could enter the model domain through the top, right and bottom model boundaries. The conditions were imposed as a combination of significant wave height, peak period, mean wave direction and directional spreading. The wave conditions along the right boundary have been imposed uniformly as specified, as this boundary is located in a still water depth of 15m and thus waves along this boundary are not affected by the seabed. Whilst progressing from the right offshore boundary toward the coast, a point is reached where the propagation of wave energy is influenced by the gradually decreasing depth (shoaling, wave breaking and refraction).

In the model set-up used for these studies, the wave conditions along the top and bottom cross-shore model boundaries were estimated following the method described by Goda (1985). This method accounts for the effects of wave shoaling, wave breaking and wave refraction. Application of this method results in a cross-shore varying wave condition that depends on the local water depth (which varied during the morphodynamic simulation).

A summary of the parameters and settings used in the wave model is given in Table 4.2.


Table 4.2 Wave model parameters and settings.

Parameter Description Type of model Spectral wave model TOMAWAC-2G Mesh spacing 10m to 40m Refraction/shoaling Yes Diffraction No Directional spreading 36 discrete directions (cos2 spreading function) Frequency spectrum JONSWAP spectrum solved for m0 and m1 Wave breaking Battjes and Janssen (1978)

α=1, γ1=0.88, γ2=0.8 Bottom friction Quadratic friction law Wave-current interaction Yes

4.3.2 Flow model

The top and bottom cross-shore boundaries were defined as open, whereas the right offshore boundary was defined as closed (no flow was allowed through this boundary). The conditions imposed on these boundaries depended on the simulated tidal case. Ignoring the variation in tidal range, three different tidal conditions were imposed – non-tidal, progressive and standing tides.

For all simulations, water level conditions were defined along the cross-shore boundaries. In addition, results from the wave model along these boundaries were used to provide an estimate of the wave-induced set-up.

For the non-tidal case, the specified water level was set at a constant value of 0.0m, constituting the still water level.

For the progressive tidal condition, boundary settings needed to be defined such that the maximum current velocities occurred near high and low water. This was realised by defining two sinusoids with slightly different amplitudes but identical phase. Both sinusoids being in phase, maxima and minima occurred at the same time at both boundaries. The slight variation in amplitude causes the maximum level at one boundary to be slightly lower than at the opposite boundary, and the minimum at the same boundary to be somewhat higher than at the opposite boundary. These water level gradients induce currents alongshore, with a maxima when the gradients are largest – around high and low water.

The standing tide conditions are characterised by maximum flow velocities near mid-tide and were thus defined by two sinusoids with a slight phase difference but identical amplitudes. Similarly to the progressive tide condition, the largest velocities are induced by the largest water level gradients and thus occur around mid-tide. The amplitude of the sinusoidal boundary settings are varied to correspond to the two tidal ranges simulated (3m and 5m) for both the progressive and standing tidal conditions.

A summary of parameters and settings used in the flow model is given in Table 4.3.


Table 4.3 Flow model parameters and settings.

Parameter Description Type of model 2DH Flow model TELEMAC2D Mesh spacing 10m to 40m Time step 1s Wave driving forces Gradients in radiation stresses Tide driving forces Gradients in tide levels at lateral boundaries Eddy viscosity Constant value used; E=1m/s2 Flooding/drying Yes; minimum water depth = 0.1m

4.3.3 Sand transport model Similar to the flow model, the top and bottom cross-shore boundaries were defined as open and the right offshore boundary was set as a closed. Thus, sediment could freely enter and leave the domain only through the cross-shore boundaries.

At these open boundaries, the equilibrium suspended sediment concentration was computed using the Soulsby-Van Rijn (1993) total load sediment transport formula, while the instantaneous flow velocities and wave orbital velocities were taken from the hydrodynamic models. The same formula is applied throughout the internal model domain.

The availability of erodible material was set to be unlimited throughout the model domain with the exception of the breakwaters, which were defined as fixed bed. All simulations were carried out on uniform sediment with a grain size of 0.2mm, as defined at the outset of the project.

A summary of the parameters and settings used in the sand transport model is given in Table 4.4.

Table 4.4 Sand transport model parameters and settings.

Parameter Description Type of model Non-cohesive sediment transport model SANDFLOWMesh spacing 10m to 40m Time step 1s Sand characteristics Uniform grain size D50=0.20mm Intra-wave calculations No; wave period averaged conditions used in sand

transport calculations Non-equilibrium transport Yes; advection-diffusion equations solved to include

settling lag Quasi-3D effects (wave asymmetry, undertow)

Not included in these simulations

Bed slope terms No bed slope effects included in computation


A schematic overview of how the morphodynamic modelling script works is presented below. The morphodynamic simulations consisted of 720 loops, equivalent to 1440hrs. Two consecutive loops are shown in Figure 4.1 below. Each loop consisted of one hour of simulation, with an additional 30 minutes incorporated into the flow model (for each loop) to allow the model to adapt to the new bathymetry. Only the flow model results from the last hour of simulation (i.e excluding the first 30 minutes) were passed on to the sand transport model.


A loop starts with a wave simulation that receives input water level and velocity fields from the final time step of the flow simulation of the previous loop, indicated by the red arrow. It also receives the updated model bathymetry from the previous loop (orange arrows). Hence, water level set-up is passed to the wave model and refraction due to currents is also modelled.

Following completion of the wave simulation, the resulting wave forces (the forces that the wave field exert on the water body, particularly due to wave breaking) are passed on to the flow model (black arrows). Because the wave model is a stationary model, the wave forces are constant over the duration of the flow simulation. The flow model also receives the updated model bathymetry from the previous loop (orange arrows), together with an estimate of the updated flow field based on flow continuity. The first 30 minutes in the flow simulation are included to allow the flow model to adapt to the new bathymetry.

The velocity fields over the final hour of flow simulation were subsequently passed on to the sand transport model (green arrows). This model also received wave orbital velocity fields from the wave model (blue arrows), as well as the updated bathymetry from the previous loop (orange arrows). Furthermore, the sand transport model received an updated concentration field, based on the final time step of the previous loop and the updated bathymetry, as initial conditions.

Following computation of the sediment transport rates over the simulated period of one hour, the bathymetry was updated on the basis of the amount of sediment entrained or deposited. The changes in bathymetry were multiplied by a morphological factor of two, in order to overcome excessively long computation times. Effectively, this means that one hour of simulation represents two hours of morphological time. This is a widely-used method for long-term morphodynamic simulations. It is applicable when bed changes over a single loop do not alter the flow and wave fields to a sufficient extent that a new hydrodynamic computation is required to properly represent these (see, for example, De Vriend et al. 1993). It is noted that erosion of dry areas – above the instantaneous water level – was not included in the morphodynamic model.

Using the updated bathymetry, an initial flow field was calculated for the following loop according to the continuity method. This method assumes that horizontal flow patterns are likely to be similar as long as bed changes are small, and that local flow rates do not therefore change. A small increase in water depth thus results in a small decrease in flow velocity, and vice versa. Continuity of sediment mass is obtained by reducing the suspended sediment concentration as the water depth increases, and vice versa. Thus the same amount of suspended sediment is maintained from the end of one loop to the start of the subsequent loop.


Water level and flow velocities

Waves

Flow

Sand transport

Time

Loop i Loop i+1

Wave forces

Flow velocities

Orbital velocities

Updated bathymetry

Flow adaptation to updated bathymetry

1 hour0.5 hour

2 hours

Figure 4.1 Overview of the interlinking of process modules and information exchange.

Table 4.5 Morphodynamic model parameters and settings.

Parameter Description Type of model Morphodynamic model system PISCES Duration of process simulations in a loop 1.0hr Flow model warm-up duration 0.5hr Morphological acceleration factor 2 Frequency of updates for: - bathymetry files - wave model results used in flow and sand transport calculations - flow model results used in sand transport calculations

2.0hr 1.0hr 1.0s

4.4 Results and discussion Results of the modelling are presented in the following sections. The combined wave-induced and tidal currents lead to complex sediment transport patterns for the simulated conditions, resulting in varying morphological responses. The feedback of the updated bathymetry on subsequent wave, flow and sediment transport simulations further enhances the complex behaviour of the hydrodynamics and sediment dynamics.

Results from the individual process models are described in the following section on the basis of the initial bathymetry. These results are included to show the complexity of the varying wave, flow and sand transport patterns that occur in the vicinity of the breakwaters over a tidal cycle, eventually leading to the predicted final bathymetries.


Finally, the resulting final bathymetry for each of the simulated cases after 60 days are discussed and inter-compared to assess the effects of cross-shore distance of the breakwater scheme, type and range of the imposed tidal wave, and oblique incident wave direction versus shore-normal waves. There is also an assessment of erosion in the breakwater bays.

4.4.1 Wave, flow and sand transport results on initial bathymetry

The simulated morphological response varies due to the variation in the patterns of waves, as well as combined wave-driven and tidal flows and sediment fluxes. By way of example, these patterns are shown for the initial bathymetry in Layout 1 for non-tidal conditions (Figure 4.5) and for a 3m standing tide (Figure 4.6), both for shore-normal waves.

Comparison reveals that the patterns near high and low water for the 3m standing tide resemble those for non-tidal conditions, with typical circulation cells in the lee of the breakwater. Near high and low water tidal velocities for the 3m standing tide case (Figure 4.6) are small and therefore the flow and sediment flux patterns are dominated by wave action and the instantaneous water depth. During the rising and falling tide (Figure 4.6), the patterns are noticeably different between the two cases as a result of tidal currents that deflect the circulation cells in the direction of the tidal current.

The tide-induced changes result in highly complex flow and sand flux patterns that continuously adapt to the changing morphology. The various investigated tides (standing and progressive, 3m and 5m tidal range) all lead to similarly complex changes in the hydrodynamic and sediment flux patterns, which in turn result in variations in the simulated morphological response.

4.4.2 Effect of cross-shore distance

The resulting morphologies after 60 days for each of the simulated cases are presented in Figure 4.7 to Figure 4.11. Figure 4.7 and Figure 4.8 show the results for shore-normal waves for the non-tidal and four tidal conditions, for Layout 1 and Layout 2 respectively. In Figure 4.9 and Figure 4.10, the same results are shown for oblique incident waves. Figure 4.11 shows the results for oblique incident waves and Layout 3. The bed level changes resulting from these simulations are shown in Figure 4.12 through to Figure 4.16 in the same order.

In order to assess the effect of the cross-shore distance of the breakwater schemes, the results for identical wave and tidal conditions have been compared for the three investigated layouts.

The resulting bathymetries after 60 days for Layout 2 (shown in Figure 4.8 and Figure 4.10) show trends that are similar to those observed for Layout 1 (Figure 4.7 and Figure 4.9), albeit the -1m depth contours of the salients extend close to the breakwaters for all simulated cases. The extent of the salients is also evidenced by the amount of deposition occurring in the lee of the breakwaters, as shown in Figure 4.12 through to Figure 4.16. Comparing the results for Layout 3 (Figure 4.11 and Figure 4.16) to those for Layout 1 (Figure 4.9b and Figure 4.14b) reveals relatively less bypassing offshore of the up-drift breakwater and a reduced salient length in the lee of the down-drift breakwaters.

The reduced distance offshore in Layout 2 lessens the sediment volume required to build up the salients to this level and the closer proximity of the scheme to the shoreline increases the sheltering effect in the lee. For oblique incident waves, sediment


bypassing is enhanced as a larger proportion of the littoral zone extends beyond the breakwaters, particularly near low water and for larger tidal ranges. The more pronounced area of deposition (yellow/red) offshore of the breakwaters for Layout 2 and oblique waves (Figure 4.13) is evidence for the enhanced bypassing of sediment offshore of the scheme. Less bypassing is seen for Layout 3, where the breakwaters are further offshore, and the resulting salient has a reduced length in the lee of the down-drift breakwaters. This may be explained by the reduced supply from up-drift and the larger distance between the shoreline and the breakwaters, allowing for increased tidal flow through this area.

4.4.3 Effect of type and range of tidal wave

The effect of the type and range of the imposed tidal conditions has been assessed by comparing the resulting morphologies for each combination of layout and incident wave direction.

The general effect of including tidal action in the morphodynamic simulations is that the presence of tides leads to overall smoothing of the bathymetry contours. This is due to changes in the littoral zone width with tidal level and the additional shore-parallel currents induced by tides. Furthermore, the changing water levels lead to a change in the location of the wave-breaking zone and a change in the distribution of wave heights in the lee of the breakwaters.

Figure 4.7 through to Figure 4.10 show that the base of the salient is wider for tides compared to the non-tidal case and that the bay is not as deep as in the non-tidal case. This is further illustrated in Figure 4.17, which shows the cross-shore bathymetric profiles along the centreline of the second breakwater bay for the investigated tidal conditions (for Layout 2 and shore-normal waves).

With increasing tidal range, the amount of sediment and the accretion height immediately behind the breakwater becomes less and the width of the salient becomes larger (see Figure 4.12 through to Figure 4.15). This is also illustrated in Figure 4.18, which shows the cross-shore bathymetric profiles along the centreline of the second breakwater (for Layout 2 and shore-normal waves). This result may be explained by an increased smoothing of the bed due to enhanced tidal currents and a larger variation in the local wave conditions over a tidal cycle. Furthermore, the extent of erosion in the bays reduces with tidal range, particularly for Layout 1. For cases with a large tidal range, there is more movement of the beach contours above MSL.

Comparing the results for standing and progressive tides indicates that progressive tides push the apex of the salients in the direction of flow around high water (down the page) for both layouts. For the defined progressive tides, the larger water depth and maximum current speeds around HW are likely to promote this deflection.

4.4.4 Effect of oblique wave incidence

An assessment of the effect of oblique incidence waves versus shore-normal waves has been carried out by comparing the results for both incident wave directions, for combinations of layout and imposed tidal conditions.

The resulting bathymetries after 60 days for the oblique wave cases (Figure 4.9 through to Figure 4.11) show that accumulation of sediment in the lee of the up-drift breakwater (top breakwater in the figures) is much larger than at the down-drift breakwaters. This trend can also be seen from the bed level change plots in Figure 4.14 and Figure 4.15. The bypassing offshore of the breakwaters, evidenced by the


accretion (yellow/red) in Figure 4.14 and Figure 4.15, is enhanced for higher tidal ranges. The result is somewhat less extensive accretion offshore of the up-drift breakwater as more sediment is transported towards the down-drift breakwaters. Figure 4.16 does not show the same pattern to the same extent for Layout 3, as the distance between the shoreline and the breakwaters is larger and the up-drift breakwater is able to trap more sediment.

Some of the increased amount of bypassed sediment provides an additional supply of sediment for the salients in the lee of the down-drift breakwaters, leading to a more pronounced area of accretion in these protected areas. This bypassing leads to an increase in salient length at the down-drift breakwaters with increasing tidal range. In addition, the increased area in the lee of the breakwaters near HW may allow for circulation cells to develop and thus cause the re-distribution of accumulated sediment in the lee of the breakwaters. The described effect on the resulting morphological response is stronger for standing tides than for progressive tides.

4.4.5 Erosion in breakwater bays

Assessing the extent of erosion in the breakwater bays is facilitated by the bed level change plots in Figure 4.12 through to Figure 4.16. It can be seen that the extent of erosion is larger for non-tidal conditions than for tidal conditions, and reduces further with increasing tidal range. The erosion tends to be slightly more pronounced for progressive tides than for standing tides, which is most clearly seen when comparing the area of erosion for 5m progressive and standing tide conditions for Layout 2. Furthermore, the extent of erosion is larger for cases with oblique incident waves than shore-normal waves.

The results presented in this section consist of the modelled bathymetries after 60 days of simulation. The modelled bathymetries after 15, 30 and 45 days are included in Appendix B to illustrate the varying time scales of the characteristics identified in the predicted morphological response for the investigated test cases.

4.5 Summary The morphological evolution of the seabed in the vicinity of four shore-parallel breakwaters was simulated with the PISCES modelling system over a period of 60 days for non-tidal and various tidal conditions. The aim was to investigate the effect of tides on the morphological response for shore-normal and oblique incident waves. Three different layouts were investigated in order to assess the dependency of the identified tidal effects on the distance between the breakwaters and the initial MSL shoreline.

The model results for non-tidal conditions illustrated the typical morphological response of sediment accumulation in the lee of the breakwaters as a result of circulation patterns developing due to the modified wave conditions in the vicinity of the breakwaters. These results also illustrated the interception of the littoral drift induced by oblique incident waves and the resulting longshore currents. The largest accumulation occurred in the lee of the first (up-drift) breakwater, while the largest response occurred at a location down-drift from the breakwater centreline as a result of the modified circulation patterns. The model results show significant accumulation on the offshore side of the first breakwater and extending into the first breakwater bay, indicating bypassing of sediment toward the down-drift breakwaters.

The general effect of including tidal action in the simulations was to smoothen the bathymetry contours. This was a result of changes to the littoral zone width with tidal


level and the additional, generally shore-parallel, tidal currents, together with the changing distribution of wave energy and wave breaking.

For oblique incident waves, it was further found that with increasing tidal range the bypassing seen for non-tidal conditions was enhanced, leading to a larger supply of sediment to the salients in the lee of the down-drift breakwaters. This increased bypassing and the increased wetted area near high water, which may allow for larger circulation cells to develop in the lee of the breakwaters, results in an increase in salient length with increasing tidal range at these breakwaters. This effect on the morphological response was predicted to be stronger for standing tides than for progressive tides.

When the breakwaters are placed closer to the shore, a larger proportion of the littoral drift (for oblique incident waves) bypasses the breakwaters on their offshore side, leading to a faster response in the lee of the down-drift breakwaters. This leads to a more similar morphological response in the lee of the breakwaters for the oblique incident waves.


Figure 4.2 Model bathymetry and details of the finite element mesh – Layout 1.


Figure 4.3 Wave height pattern (left) and wave force pattern (right) based on TOMAWAC.

Figure 4.4 Wave height pattern (left) and wave force pattern (right) based on SWAN.

3000

3500

4000

4500

5000

0 100 200 300 400 500 600 700 800 900 1000

Hm0 (m)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1.30

1.40

1.50

1.60

1.70

1.80

1.90

2.00

2.10

2.20

3000

3500

4000

4500

5000

0 100 200 300 400 500 600 700 800 900 1000

Wave force(m/s2)

0.00

0.00

0.01

0.01

0.02

0.03

0.03

0.04

0.04

0.05

0.05

3000

3500

4000

4500

5000

0 100 200 300 400 500 600 700 800 900 1000

Wave forcevectors

0.10 m/s2

3000

3500

4000

4500

5000

0 100 200 300 400 500 600 700 800 900 1000

Hm0 (m)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1.30

1.40

1.50

1.60

1.70

1.80

1.90

2.00

2.10

2.20

3000

3500

4000

4500

5000

0 100 200 300 400 500 600 700 800 900 1000

Wave force(m/s2)

0.00

0.00

0.01

0.01

0.02

0.03

0.03

0.04

0.04

0.05

0.05

3000

3500

4000

4500

5000

0 100 200 300 400 500 600 700 800 900 1000

Wave forcevectors

0.10 m/s2


Wave height Velocity Sand flux

Figure 4.5 Initial wave, flow and sediment transport patterns for shore-normal waves and non-tidal conditions – Layout 1.

MSL, Rising tide High Water Wave height Velocity Sand flux Hs Velocity Sand flux

MSL, Falling tide Low Water

Wave height Velocity Sand flux Hs Velocity Sand flux

Figure 4.6 Initial wave, flow and sediment transport patterns for shore-normal waves and a 3m standing tide – Layout 1.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal

Figure 4.7 Simulated bathymetry after 60 days – Layout 1, shore-normal waves.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal

Figure 4.8 Simulated bathymetry after 60 days – Layout 2, shore-normal waves.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal

Figure 4.9 Simulated bathymetry after 60 days – Layout 1, oblique waves.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal



3m P

rogr

essi

ve

tide



e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal

Figure 4.12 Bed level changes after 60 days – Layout 1, shore-normal waves.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal

Figure 4.13 Bed level changes after 60 days – Layout 2, shore-normal waves.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal

Figure 4.14 Bed level changes after 60 days – Layout 1, oblique waves.


e) 5

m S

tand

ing

tide

d) 5

m P

rogr

essi

ve

tide

c) 3

m S

tand

ing

tide

b) 3

m P

rogr

essi

ve

tide

a) N

on-ti

dal



3m P

rogr

essi

ve

tide



-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0

6.0

8.0

0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0

non-tidal3m progressive tide3m standing tide5m progressive5m standing tide

Figure 4.17 Cross-shore bathymetry profile after 60 days along the centreline of the second breakwater – Layout 2, shore-normal waves.

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0

6.0

8.0

0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0

non-tidal3m progressive tide3m standing tide5m progressive5m standing tide

Figure 4.18 Cross-shore bathymetry profile after 60 days along the centreline of the second breakwater bay – Layout 2, shore-normal waves.


5 Morphological modelling using MIKE 21 CAMS In this section, the study carried out using the coastal area morphological model MIKE 21 CAMS is presented. The outline of this chapter is as follows: first, the simulated test cases are presented in Section 5.1, followed by a short description of MIKE 21 CAMS in Section 5.2. Next, there is a description of the setup of the numerical process models in Section 5.3, followed by presentation and discussion of the model results in Section 5.4. Finally, a summary of the work is presented in Section 5.5.

5.1 Simulated test cases MIKE 21 CAMS was used to simulate 18 out of the 30 generic test cases described in Chapter 3. These test cases are summarised in Table 5.1.

Table 5.1 Generic test cases simulated using MIKE 21 CAMS.

Waves Simulation no.

Breakwater crest level, hcr

(m MSL) Hm0 (m)

Tp (s)

θ (deg)

Tides Layout

01 2 2 8 90 02 2 2 8 45 No tides

05 2 2 8 90 3m, Standing

07 2 2 8 90 08 2 2 8 45

5m, Progressive

09 2 2 8 90 5m, Standing

L1

11 2 2 8 90 No tides

15 2 2 8 90 3m, Standing

Initi

al te

st c

ases

19 2 2 8 90 5m, Standing

L2

A 2 1 5 90 L1 B 2 1 5 90 No tides L2

07B 1 2 8 90 07C 3 2 8 45

5m, Progressive L1

19B 1 2 8 90 19D 0 2 8 45

5m, Standing L2

21 2 2 8 90 No tides

25 2 2 8 90 3m, Standing

Add

ition

al te

st c

ases

29 2 2 8 90 5m, Standing

L3

The three breakwater layouts (L1, L2 and L3) in Table 5.1 consist of four breakwaters, each with a length of 200m and a gap width of 300m between the breakwaters. The breakwater crest width is taken as 5m, while the crest level varies as shown in the table. The distance from the initial MSL shoreline to the centre of the breakwater is


250m (L1), 150m (L2) and 350m (L3). The tidal ranges investigated are 0m (no tides), 3m and 5m; while the tidal period is selected as 12.42hrs.

Most of the additional test cases (nine out of 10 cases) were simulated using MIKE 21 CAMS. Furthermore, nine of the original 20 test cases were simulated again using MIKE 21 CAMS.

5.2 Model description MIKE 21 CAMS (developed by DHI, Denmark) combines standard MIKE 21 modules for waves, flow and sand transport, as well as a bed level update scheme, into an automated system for the simulation of spatial and temporal changes in nearshore morphology. Execution of the modules is controlled by a shell, which also ensures a flow of information among the components of the modelling system.

MIKE 21 CAMS has been used successfully to simulate the morphological evolution (bed level changes) in the vicinity of breakwaters in the field. Johnson et al. (2005) describes the validation of the model against the morphological evolution in the vicinity of a breakwater on the Jumerah coast of Dubai, UAE, over a period of 21 months (March 1995–January 1997). The results showed very good agreement between the measured and simulated changes to the beach profiles in the lee of the breakwater (accretion) and in the middle of the breakwater bay (erosion). Zyserman et al. (2005) used MIKE 21 CAMS to illustrate the dependence of far-field erosion near edges of a detached breakwater scheme on the freeboard of the coastal protection structures.

5.2.1 Wave model

The wave model used is MIKE 21 PMS – a refraction/diffraction model based on parabolic approximation to the mild slope equation. It includes the wide-angle parabolic approximation equations (Minimax approximations) of Kirby (1986). The model accounts for the influence of refraction, shoaling, diffraction, wave breaking, bottom friction, frequency and directional spreading. Further details about the model can be found in DHI Software (2008b)

5.2.2 Flow model

The flow model used is MIKE 21 Flow Model – a 2D depth-averaged flow model for simulating water levels and depth-integrated fluxes driven by wave breaking (radiation stresses), tides, wind and atmospheric pressure conditions. The main features of the model are described by Abbott et al. (1973). Further details about the model can be found in DHI Software (2008c).


The sand transport model used is MIKE 21 ST. MIKE 21 ST uses DHI’s deterministic intra-wave-period sediment transport model STP (Deigaard et al. 1986a,b) to calculate the total transport rates (bed load + suspended load) of non-cohesive sediment (sand) under the combined influence of waves and current.

The model calculates the rates of bed level changes from the equation of conservation of sediment mass, which is solved using the modified Lax Wendroff scheme. The main features of the bed level update scheme are described in Johnson and Zyserman


(2002). Subsequent improvements in the bed level update scheme include the capability to erode dry beach and the inclusion of longitudinal and transverse slope effects on total sediment transport rates as diffusion terms. The latter addition significantly improves the stability of the morphological calculations. Further details can be found in DHI Software (2008d).


A block flow diagram summarising the flow of information in MIKE 21 CAMS is shown in Figure 5.1.

The input data for the morphological simulation consist of: a) initial bathymetry; b) time series of wind fields (if applicable), waves and flow boundary conditions; and c) sediment data. Using this input data, the wave and flow modules are used separately to determine the initial wave (maps of wave parameters and radiation stresses) and flow conditions (conditions at computational timestep i=1). The given input data, together with the initial wave and flow conditions and specification parameters for the various process modules, are used as input to MIKE 21 CAMS. The flow of information in MIKE 21 CAMS is briefly described below (adapted from DHI Software 2008a).

1. Using the calculated wave and flow conditions at computational step i (i=1 for the initial conditions), the sand transport model is used to simulate the sand transport rates and the associated rates of bed level changes dz/dt. The bathymetry corresponding to the next computational step, i+1, is also calculated. The actual time interval from computational step i to computational step i+1 is selected as the lowest of: a) maximum morphological time step based on the Courant stability condition for the bed update scheme; b) user-selected maximum morphological time step; and c) time interval to the next output time for the morphological results. The morphological output files are updated at this stage, depending on the selected output frequency.

2. Next, the wave model is used to determine the wave and radiation stress maps for computational step i+1 based on the new bathymetry calculated in step A for computational step i+1. Note that for the calculations in this study the wave and radiation stress fields are updated after every update of the sand transport field (k=1 in Figure 5.1).

3. Next, the apparent bed resistance (experienced by the flow) due to combined waves and currents is calculated using the MIKE21 bed resistance tool. Furthermore, the wave-driven flow boundary conditions at the model lateral boundaries are calculated and combined with the prescribed tidal boundary conditions (if relevant).

4. Next, the flow model is used to simulate the flow conditions from computational step i to step i+1. The bed level is changed at the rate of dz/dt calculated in (1) during the flow model calculations. Thus, the final bathymetry in the flow model at computational step i+1 is identical to the new bathymetry calculated in (1).

5. Next, the computational step number i is incremented by 1, and steps 1 to 4 are repeated until the specified end date for the morphological simulation is reached.


Figure 5.1 Block flow diagram for MIKE 21 CAMS (from DHI Software 2008a).

5.3 Setup of numerical process models As stated in the preceding section, MIKE 21 CAMS uses process modules for waves, flow and sediment transport, together with a bed level update scheme for the morphological simulations. The setup parameters used for the process modules are summarised in the sub-sections below.

5.3.1 Wave model

The model open boundaries are at the offshore boundary (depth of 15m relative to MSL) and the two lateral boundaries that are perpendicular to the shoreline. For each test case, the offshore wave parameters are specified as a 2D energy spectrum (in frequency and direction) at the offshore boundary, while a symmetric boundary condition is specified at the lateral boundaries.


The parameters used in the model setup are summarised in Table 5.2. The wave breaking parameter (γ2) over the breakwaters was specified as 1.4. Johnson (2006) found that such a high value is more appropriate for obtaining wave transmission coefficients over submerged breakwaters that are in reasonable agreement with values obtained from empirical formulas, such as those developed by d’Angremond et al. (1996).

Table 5.2 Wave model parameters used in the morphological simulations.

Parameter Description Type of model Parabolic mild slope model, MIKE 21 PMS Parabolic approximation Minimax 50 Grid spacing 5m Refraction/shoaling Yes Diffraction Yes Directional spreading Five discrete directions (cos5 spreading function) Frequency spectrum 10 discrete frequencies (JONSWAP) Wave breaking Battjes and Janssen (1978)

α=1, γ1=1, γ2=0.8 (γ2=1.4 over breakwaters) Bottom friction Quadratic friction law (kN=2mm) Wave-current interaction No

5.3.2 Flow model

The model open boundaries are at the offshore boundary and the two lateral boundaries that are perpendicular to the shoreline. For each test case, the offshore boundary is specified as a streamline (no flow across the boundary).

At the lateral boundaries, water level boundary conditions are specified. The specified water level variation across the lateral boundary is calculated using a utility program8 to determine the wave setup and setdown, which is combined with tidal levels for the relevant cases. The specified tidal levels at the lateral boundaries depend on the type of tidal conditions.

For progressive tides, boundary conditions needed to be defined such that the maximum current velocities occurred near high and low water. This was achieved by defining two sinusoids with slightly different amplitudes but identical phase (using the same method as described in Section 4.3.2). Slight changes in the tidal amplitude at the two lateral boundaries are made such that the required tidal range (3m or 5m) is achieved at the centre of the model area.

The standing tide conditions are characterised by maximum flow velocities near mid-tide, and were thus defined by two sinusoids with a slight phase difference and identical amplitudes (using the same method as described in Section 4.3.2). In this case, the maximum velocities are induced by the largest water level gradients, which occur around mid-tide.

The parameters used in the model setup are summarised in Table 5.3. The apparent bed resistance in combined waves and currents is calculated using the Fredsoe (1984) model. Ideally, these calculations should be included in the morphological loop. However, this was not done due to computational difficulties experienced during the initial simulations with tides. Thus, the apparent bed resistance map is calculated at

8 The utility program assumes straight and parallel contours in the vicinity of the lateral boundary; its results are automatically updated to account for varying tidal levels and bathymetry changes across the lateral boundary during the morphological simulation.


HW and the calculated bed resistance is schematised and used as a fixed resistance map during the numerical calculations.

Table 5.3 Flow model parameters used in the morphological simulations.

Parameter Description Type of model 2DH Flow model, MIKE 21 Flow model Grid spacing 5m Time step 2s Wave driving forces Gradients in radiation stresses Tide driving forces Gradients in tide levels at lateral boundaries Eddy viscosity Constant value used; E = 1m/s2 Bottom friction Wave current shear stress (Fredsoe 1984)

(kN=0.01m, kN=3.75m over breakwaters) Flooding/drying Yes; drying depth=0.2m; flooding depth=0.3m


The sand transport rate at each model grid point is calculated based on the assumption that it is in equilibrium with local hydrodynamic (waves and flow) conditions. Thus, the sand transport at each grid point is determined explicitly by the local hydrodynamic conditions and there is no need to specify any sediment concentration at the model open boundaries.

For calculating bed level changes, the gradient in bed level changes across each of the lateral boundaries is specified as zero.

All simulations were carried out for moderately graded sediment with a median grain size of 0.25mm and a gradation of 1.5. The breakwater areas are represented as non-erodible (fixed bed) in the model.

Quasi-three dimensional (3D) effects (wave asymmetry, undertow) are not included in the sand transport model for the morphological calculations carried out in this study. In previous work carried out at DHI (work leading to Johnson et al. 2005), it was found that including these effects in simulations where small beach building waves are not included was inappropriate. A similar conclusion was reached by Van Rijn (2005), who reviewed the results of work carried out by Bos et al. (1996) and Sayed (1997) to simulate the bed evolution behind a single detached breakwater using a 2DH (two-dimensional, horizontal) model and a quasi-3D model respectively. Van Rijn concluded that the 2DH model produces realistic erosion/deposition patterns, while the quasi-3D model produces less realistic results – rather irregular contour lines and a double salient.

The parameters used in the model setup are summarised in Table 5.4.


Table 5.4 Sand transport model parameters used in the morphological simulations.

Parameter Description Type of model MIKE 21 ST Sand characteristics D50=0.25mm, sqrt(D84/D16)=1.5 Intra-wave calculations Yes; DHI’s STP method used with intra-wave sand

transport calculations Non-equilibrium transport No; transport rates depend only on local

hydrodynamic conditions Quasi-3D effects (wave asymmetry, undertow)

Not included for these simulations

Bed slope terms Included in conservation of sediment mass equation


A key parameter in morphological simulations is the frequency with which the bathymetry is updated in the different process models (waves, flow and sand transport), as well as the frequency of transferring information from one model to the next model in the morphological loop (such as radiation stress data from the wave model to the hydrodynamic model).

The frequency of updates used in the morphological simulations is summarised in Table 5.5. Note that no morphological acceleration is used for these calculations.

Table 5.5 Frequency of updates during the morphological simulation.

Parameter Maximum time-interval before updating Bathymetry files in: Wave model Flow model Sediment transport model

1hr 2s [The bathymetry in the flow model is updated using the calculated dz/dt based on flow conditions at the end of the previous computational step. The dz/dt map is updated every hour or less] 1hr

Wave model results used in flow and sand transport calculations

1hr

Flow model results used in sand transport calculations

1hr

All the simulations were carried out for 60 days, except the simulations with low waves (A and B in Table 5.1), which were carried out for 28 days.


5.4 Results and discussion

5.4.1 Non-tidal cases: wave, flow and sand transport results on initial bathymetry

The effect of the breakwaters on the wave, flow and sand transport pattern is illustrated in Figure 5.2 for the cases with normal and oblique wave incidence. Figure 5.2 shows that the breakwaters cause reduced wave heights in their lee. For normal wave incidence, the shoreline section with reduced wave heights is symmetrical about the breakwater, while for oblique wave incidence it is shifted further down-wave.

For normal wave incidence, the reduced wave heights in the lee of the breakwater result in gradients in wave heights and the associated wave setup. This drives the two symmetrical circulation cells (flow towards the centre of the breakwater) in the lee of the breakwater. The combined wave and flow pattern leads to the sand transport pattern shown in Figure 5.2. It also leads to the resulting redistribution of sediments in the vicinity of the breakwater, as sand is carried from the unprotected adjacent shoreline into the area directly in the lee of the breakwater. In the exposed region outside the influence of the breakwaters, there is no longshore flow or longshore sand transport.

For oblique wave incidence, the wave shadow region is not symmetrical about the breakwater centre-line, but is shifted further down-wave behind the breakwater. Thus, the section of the shoreline with longshore gradients in wave heights (and wave setup) is also shifted down-wave behind the breakwater.


Normal wave incidence Oblique wave incidence

Figure 5.2 Initial wave (top), flow (middle) and sand transport patterns (bottom) for non-tidal cases.

Along the exposed section of the shoreline, wave-driven longshore currents are generated due to oblique breaking waves. The combination of longshore currents and wave setup gradients in the shadow region of the breakwater results in a complex flow pattern in the lee of the breakwaters. The incoming longshore flow is accelerated in the section with maximum positive gradient in wave height (wave height reducing in the same direction as longshore flow). At the down-wave edge of the shadow zone, where


there is a maximum negative gradient in wave height (wave height increasing in the same direction as longshore flow), the longshore flow is deflected offshore, due to the clockwise circulation induced by wave setup gradients in this section.

The combined wave and flow pattern leads to a sand transport pattern with two distinct areas of reducing sediment transport rates (and hence sediment deposition). These areas are: 1) a ‘deep’ water area in the immediate lee of the breakwater; and 2) an area closer to the shoreline, where the longshore transport is deflected offshore. Finally, away from the influence of the breakwater scheme, the longshore transport rate increases towards the uniform longshore drift for an exposed shoreline. This results in down-drift shoreline erosion.

5.4.2 Non-tidal cases: morphological evolution results

An overview of simulated non-tidal test cases is shown in Table 5.6. The bathymetry contours, profiles across the centreline of the first breakwater and profiles across the middle of the first bay are shown in Appendix C for different time intervals. For the 60 day morphological simulations, the results are shown at 0, 15, 30 and 60 days, while for the 28-day simulations the results are shown at 0, 7, 14 and 28 days.

Table 5.6 Generic non-tidal test cases simulated using MIKE 21 CAMS.

Waves Layout Sim no. Hm0

(m) Tp (s)

θ (deg) Plan Ls/X X/Xb

Sim period (days)

Morph response

A 1 5 90 L1 0.80 2.50 28 Salient B 1 5 90 L2 1.33 1.50 28 Salient 01 2 8 90 L1 0.80 1.25 60 Tombolo 02 2 8 45 L1 0.80 1.25 60 Salient /

Tombolo 11 2 8 90 L2 1.33 0.75 60 Tombolo 21 2 8 90 L3 0.57 1.75 60 Salient /

Tombolo Notes: Breakwater crest level, hcr ≥ 2m MSL; Xb=200m for Hm0=2m and Xb=100m for Hm0=1m.

5.4.2.1 Effect of breakwater cross-shore distance

The effect of varying breakwater cross-shore distance on the morphological response is illustrated in Figure 5.3 for Hm0=2m and Figure 5.5 for Hm0=1m.

Figure 5.3 shows that the profiles in the lee of the breakwater are characterised by a flat profile section and a steep section closer to the breakwater. The width of the flat section of the profile can be used as a measure of the width of the salient. It is noted, however, that the flat section is typically below the MSL (as model computations are only carried out at wet points, with a water depth of at least 0.3m).

Figure 5.3 shows that layout L3 forms a salient (it is also starting to form a bell-shaped tombolo), L1 forms a tombolo (in the shape of an hourglass), while L2 forms a more uniform tombolo in the lee of the breakwater.


21:L3, Ls/X =0.57 01:L1, Ls/X =0.80 11:L2, Ls/X =1.33

Section A: Centreline of 1st breakwater Section B: C/L of 1st bay

Figure 5.3 Simulated bathymetry after 60 days of morphological simulation (top: bathymetry contours; bottom: profiles across section A and B). Notes: Waves are normally incident to the shoreline (Hm0=2m, Tp=8s at depth of 15m) and no tides.

The influence of reduced wave exposure on the effect of a breakwater scheme on the shoreline is illustrated in Figure 5.4. In this case, both L1 and L2 form a salient, as the breakwaters are located offshore from the zone of active sediment transport. Furthermore, the size of the circulation cells is less constrained in L1 than L2, leading to a greater re-distribution of sediments in L2. Lastly, practically no bed level change is seen along the profile in the middle of the breakwater bays (B), as the sediment redistribution is limited to an area closer to each breakwater.

S

A

B


A: L1, Ls/X =0.80 B:L2, Ls/X =1.33

Section A: Centreline of 1st breakwater Section B: C/L of 1st breakwater bay

Figure 5.4 Simulated bathymetry after 28 days of morphological simulation (top: bathymetry contours; bottom: profiles across section A and B.). Notes: Waves are normally incident to the shoreline (Hm0=1m, Tp=5s at depth of 15m) and no tides.

5.4.2.2 Effect of oblique wave incidence

The effect of oblique wave incidence is illustrated in Figure 5.5. Considering the flat part of the profile in the lee of the breakwater, the morphological response for the two cases can be classified as a tombolo. However, the tombolo is at a higher bed level for normal wave incidence than for the oblique wave incidence.


01:Normal wave incidence 02:Oblique wave incidence


Figure 5.5 Simulated bathymetry after 60 days of morphological simulation – Layout L1 (Ls/X=0.8) and no tides (top: bathymetry contours; bottom: profiles across section A and B).

For oblique wave incidence, a salient forms close to the downdrift edge of the breakwater and the maximum shoreline erosion in the breakwater bays occurs near the updrift end of the breakwaters in the bay (see Figure 5.5). These effects are related to the obliquity of the incident waves, as this controls the section of the shoreline in the shadow of the incident waves. It is noted that the locations of the updrift shoreline erosion and downdrift shoreline accretion (salient) coincide respectively with the updrift and downdrift limits of the wave shadow on the shoreline. The above discussion implies that existing infill behind a detached breakwater will be pushed downdrift, if waves are obliquely incident to the breakwater. This behaviour has been documented at the Sea Palling breakwaters (see Fairley and Davidson 2009 in Appendix A2).

5.4.3 Tidal cases: wave, flow and sand transport patterns during the initial tidal cycle

The initial results from the wave, flow and sand transport model during the first few hours of the morphological simulation for test case 07 (normal wave incidence, 5m


progressive tides) are shown in Figure 5.6. These results are used to illustrate the effect of tides on the various processes.

Figure 5.6 Initial wave (top), flow (middle) and sand transport patterns (bottom) at four different stages of the 5m progressive tide.

Figure 5.6 shows that the changing tide levels lead to a changing location of the wave breaking zone and a changing distribution of wave heights in the lee of the breakwater. Furthermore, the combined tide and wave-driven flow pattern for this case show that there is a pronounced lack of symmetry near HW due to the influence of the tidal flow. The modified flow pattern resembles the flow pattern for oblique wave incidence in the lee of the breakwater (see Figure 5.2). A similar lack of symmetry is not seen at low tide, probably due to the weaker tidal speeds. This means that the alongshore gradients in wave height (and the resulting gradients in wave setup) become dominant over tidal forcing. The combined effects of the modified wave and flow pattern are also evident in the sand transport fields.

It is, however, noted that the asymmetry in the flow pattern at HW is destroyed if the breakwater crest level is too low (water depth of 1.5m over the breakwater crest at HW), since the primary wave induced circulation cells for an emergent breakwater are non-existent when the breakwater is significantly overtopped.

5.4.4 Tidal cases: morphological evolution results

An overview of simulated tidal test cases is shown in Table 5.7. The simulated bathymetry contours after 15, 30 and 60 days are shown in Appendix C. The corresponding profiles across the centreline of the first breakwater and in the middle of the first breakwater bay are also shown in Appendix C.


Table 5.7 Generic tidal test cases simulated using MIKE 21 CAMS. (tide type: S=standing tides, P=progressive tides).

Waves Layout Tide Sim no. Hm0

(m) Tp (s)

θ (deg) Plan hcr

(m) Range

(m) Type

Morph response

05 2 8 90 L1 2 3 S Salient 07 2 8 90 L1 2 5 P Salient

07B 2 8 90 L1 1 5 P Salient 07C 2 8 90 L1 3 5 P Salient 08 2 8 45 L1 2 5 P Tidal tombolo 09 2 8 90 L1 2 5 S Salient 15 2 8 90 L2 2 3 S Salient 19 2 8 90 L2 2 5 S Salient

19B 2 8 90 L2 1 5 S Salient

19D 2 8 90 L2 0 5 S Limited response

25 2 8 90 L3 2 3 S Salient 29 2 8 90 L3 2 5 S Salient

Effect of tidal range

The effect of varying tidal range on the morphological response in the vicinity of breakwaters for normal wave incidence is illustrated in Figure 5.10 and Figure 5.8.

Figure 5.10 shows the simulated bathymetry contours for varying tidal range (Rtide=0, 3m and 5m) after a 60-day morphological simulation, while Figure 5.8 shows the corresponding profiles across sections A and B indicated in Figure 5.3.

The presence of tides leads to overall smoothing of the bathymetry contours, due to changes in the littoral zone width with tidal level and the additional shore-parallel currents induced by tides. Furthermore, the changing water levels lead to changing locations of the wave breaking zone, modifying the distribution of wave heights in the lee of the breakwater (see Figure 5.6).

Figure 5.8 shows that the salient length decreases with tidal range, for a given breakwater cross-shore distance. The amount of erosion in the breakwater bay generally reduces with increasing tidal range. However, there is more movement of the beach contours above MSL (in particular, more erosion above MSL with increasing tidal range) and cross-shore profiles become less flat as the tidal range increases. The influence of varying cross-shore distance is similar to the non-tidal case.

The predicted shoreline planforms using the parabolic bay shape method of Silvester and Hsu (1997) are overlain in the inset boxes in Figure 5.8. The parabolic bay shoreline planforms (taken to be MSL) are calculated separately for each breakwater and combined as shown in the plots. It is interesting to note that the predicted MSL shoreline using the method of Silvester and Hsu agrees fairly well with the simulation results for the tidal cases in the bay between the breakwaters.

The effect of tidal range and oblique wave incidence on morphological response is illustrated in Figure 5.9. This also shows a reduction in salient length (submerged salient) as the tidal range increases. Furthermore, the profile becomes less flat, as also observed for normal incidence.


-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

-1000

100

200

300

400 -2

50-1

50-5

050

150

250

350

450

550

650

750

Figure 5.7 Simulated bathymetry contours after 60 days of morphological simulation. Notes: Waves: Hm0=2m, Tp=8s, normal wave incidence; tides: standing tides)



Figure 5.8 Simulated nearshore profiles across section A and B after 60 days of morphological simulation. Notes: Waves: Hm0=2m, Tp=8s, normal wave incidence; tides: standing tides.


02: No tide 08: Rtide = 5m, progressive


Figure 5.9 Simulated bathymetry after 60 days of morphological simulation – layout L1, Hm0=2m, Tp=8s, oblique wave incidence (top: bathymetry contours; bottom: profiles across section A and B).

Effect of breakwater crest level

The effect of varying breakwater crest level on the morphological response in the vicinity of breakwaters is illustrated in Figure 5.10 (Layout L1) and Figure 5.11 (Layout L2).


07C: hcr=3m 07: hcr=2m 07B: hcr=1m


Figure 5.10 Simulated bathymetry after 60 days of morphological simulation (top: bathymetry contours; bottom: profiles across section A and B). Notes: Waves: Hm0=2m, Tp=8s, normal wave incidence. Tides: Rtide=5m, progressive tides.

Figure 5.10 shows practically no difference in the simulated bathymetries after 60 days for crest levels of 2m and 3m. This is due to the breakwater crest level in both cases being above the tide level for most parts of the tidal cycle. With a crest level of 2m, the breakwater is submerged for only 1.5hr per tidal cycle (tidal range is 5m and period is 12.42hrs), while with a crest level of 3m, the breakwater crest level is always above the tide level.

Figure 5.11 shows a more pronounced difference in the simulated bathymetries when the breakwaters are closer to the shore (for instance, compare the results for crest levels of 2m and 1m in Figure 5.10 and Figure 5.11). The bed levels over the salient reduce considerably as the crest level is reduced to +0m (with water depth of 2.5m


over the breakwater at HW). This shows that the beach level in the lee of a frequently overtopped breakwater is significantly lower than that in the lee of an emerged breakwater. This result is consistent with observations at the phase 2 breakwaters (low crested) at Sea Palling (see Dolphin et al. 2009 in Appendix A2).

19: hcr=2m 19B: hcr=1m 19D: hcr=0m


Figure 5.11 Simulated bathymetry after 60 days of morphological simulation (top: bathymetry contours; bottom: profiles across section A and B). Notes: Waves: Hm0=2m, Tp=8s, normal wave incidence. Tides: Rtide=5m, standing tides.

5.5 Summary The morphological response (bed level changes) to incident waves and tides in the vicinity of breakwaters has been investigated using MIKE 21 CAMS. The model was applied to a number of generic cases including non-tidal/tidal cases with various tidal ranges and tidal types, and varying the cross-shore location of the breakwater scheme and the breakwater crest level.


The simulation results show that for non-tidal cases the modification of the wave conditions in the lee of the breakwater leads to flow circulation and redistribution of sediments in the vicinity of the breakwater.

For non-tidal cases with normal incident waves, the bed level changes are primarily caused by the gradients in wave height (and the associated wave setup), which leads to the formation of symmetrical sediment deposition patterns in the lee of the breakwater. Depending on the cross-shore location of the breakwater, the sediment deposition pattern may be characterised as a salient, tombolo or limited response.

For oblique incident waves, the bed level changes are caused by a combination of the effects of gradients in wave heights and longshore currents induced by the oblique breaking waves. This leads typically to two distinct regions of sediment deposition, as the wave shadow in the lee of the breakwater is now at an angle to the coast. The two sediment deposition regions consist of: 1) a region in the immediate lee of the breakwater where the longshore sediment transport rate reduces due to decreased wave heights; and 2) a region very close to the shoreline where the incoming longshore flow is deflected offshore due to a negative gradient in wave height (and associated wave setup) inducing an opposing flow to the incoming longshore flow.

The simulation results show that tides modify the above basic processes due to changing water levels and tidal currents.

For a given breakwater cross-shore distance and normal wave incidence, the salient length decreases as the tidal range increases. Furthermore, there is more movement of the beach contours above MSL and cross-shore profiles become less flat as the tidal range increases.

For oblique wave incidence, the morphological response is also a reduction in salient length (submerged salient) as the tidal range increases. Furthermore, the profile becomes less flat, as observed for normal incidence.

The morphological simulation results also show that the beach level in the lee of a frequently overtopped breakwater is significantly lower than in the lee of an emerged breakwater. This result is consistent with observations at the Phase 2 breakwaters (low crested nearshore breakwaters) at Sea Palling (see Dolphin et al. 2009 in Appendix A2). Furthermore, the effect of the breakwater crest level is significantly stronger if the breakwater is closer to the shoreline.


6 Analysis of morphological model results

In Chapters 4 and 5, advanced numerical coastal area morphological models (PISCES and MIKE 21 CAMS) were used to simulate 30 combinations of breakwater layouts, wave conditions and tidal conditions in order to understand the effect of tides on the morphological response in the vicinity of breakwaters. The results of these simulations are used in this section to develop design graphs that can be used in the outline design of breakwaters in macro-tidal environments.

The outline of this chapter is as follows: first, a short summary of the key dimensionless parameters is presented in Section 6.1. This is followed by a presentation of design graphs for non-tidal cases and tidal cases in Section 6.2 and Section 6.3 respectively.

6.1 Key dimensionless parameters The dimensionless analysis carried out in Chapter 3 showed that the morphological response in the lee of nearshore breakwaters on macro-tidal sites can be described as a function of the following eight parameters.

1) Ls/X is a measure of the breakwater blocking efficiency. 2) X/Xb is a measure of the percentage of littoral drift affected by

breakwater(s) (or a measure of the relative location of a breakwater in the surf zone).

3) G/L0 is a measure of the wave penetration through gaps. 4) B/L0 is a measure of the wave energy dissipation distance over the

breakwater crest. 5) dcr/Hb is a measure of wave energy dissipation rate over a breakwater. 6) Rtide/Hb is a measure of the effect of tide range on the surf zone.

7) btide gHU / is a measure of the effects of tidal current relative to wave-induced current.

8) φ is a measure of the type of tidal regime. The shoreline response is typically characterised in terms of accretion in the lee of the breakwater (salient/tombolo) and erosion at the shoreline between the breakwater gaps.

6.2 Non-tidal cases The predicted morphological response is compared with an existing design graph from Rosati (1990; see Figure 2.1). This suggests that L1 [Ls/X=0.80] will likely form a salient or tombolo, L2 [Ls/X=1.33] will likely form a tombolo and L3 [Ls/X=0.57] will likely form a salient or limited response. This compares well with the results from the simulations for the high waves case (incident Hm0=2m). However, the predicted response using the design graph is different for the low waves case (incident Hm0=1m).

The result of the dimensionless analysis (Chapter 3) indicates that the morphological response for the non-tidal case is dependent on Ls/X (a measure of the breakwater blocking efficiency) and X/Xb (a measure of the relative location of the breakwater in the surf zone), when the breakwater cross-section and the gap width between


breakwaters are fixed. Thus, even for the non-tidal case, a design graph such as Figure 2.1 cannot be expected to provide accurate predictions when the relative location of the breakwater within the surf zone is varied. Other authors (such as Suh and Dalrymple 1987) have attempted to relate the morphological response to both Ls/X and X/Xb, but the resulting graphs have a lot of scatter. A slightly different approach is used in this study.

The calculated salient lengths (based on changes to the -1m MSL contour) using results from both PISCES and MIKE 21 CAMS are plotted in Figure 6.1, together with data from various laboratory model experiments summarised in Suh and Dalrymple (1987). The combined dataset shows the following trends.

• For a given relative breakwater location in the surf zone (X/Xb), the relative salient length (S/X) increases as the dimensionless breakwater length (LS/X) increases.

• For a given dimensionless breakwater length (Ls/X), the relative salient length increases for low values of (X/Xb) and thereafter decreases, as should be expected for a breakwater located far away from the surf zone.

• Depending on the relative location of the breakwater in the surf zone, tombolo formation can occur for LS/X > 0.8.

Using Figure 6.1, the limiting conditions for tombolo formation are postulated as:

Ls/X > 2.8 – 1.6(X/Xb), X/Xb ≤ 1.25 (6.1a)

Ls/X > -10.2 + 8.8(X/Xb), 1.25 < X/Xb< 2.0 (6.1b).

If the breakwater is located well beyond the limiting depth for littoral drift, it is very unlikely that a tombolo will form. Thus, equation 6.1 should not be used if X/Xb>2. Beyond this value, it is postulated that the beach response will either be a salient or limited response.

S&T:Shinohara & Tsubaki(1966); R&V:Rosen & Vajda(1982); S&D:Suh & Dalrymple(1987)

Figure 6.1 Non-tidal cases from the numerical simulations and laboratory data from Suh and Dalrymple (1987). Notes: Labels on the plot are the dimensionless salient length (S/X).


6.3 Tidal cases

6.3.1 Effect of the type and range of tidal wave

In Chapters 4 and 5, it is shown that the profiles in the lee of the breakwater are characterised by a flat profile section and a steep section closer to the breakwater. For the analysis of trends in salient lengths, the intersection of the flat slope and the steep slope is determined, and the salient length is measured as the change in position of the contour level at this intersection (see Figure 6.2). This approach was found to be the best option for measuring the salient lengths for the range of test cases simulated in this study.

-6.0

-4.0

-2.0

0.0

2.0

400 450 500 550 600 650 700 750 800 850 900 Figure 6.2 Cross-shore profile evolution over 60-day simulation along the centre-line across the second breakwater, for layout L1, shore-normal waves and 3m standing waves.

The relative salient lengths for different relative tidal ranges are plotted against the relative breakwater length in Figure 6.3, which shows that the relative salient length reduces as tidal range increases for shore normal waves. However, for large values of Ls/X (> 1.3), the influence of tidal range is not significant (if the breakwater is emergent through the tidal cycle).

Effect of Tides on Salient length

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.5 0.7 0.9 1.1 1.3 1.5Ls / X

S /

X

CAMS: Rtide/Hm0=0 CAMS: Rtide/Hm0=1.5 CAMS: Rtide/Hm0=2.5PISCES: Rtide/Hm0=0 PISCES:Rtide/Hm0=1.5 PISCES: Rtide/Hm0=2.5

Figure 6.3 Effect of breakwater length for different dimensionless tidal ranges (Rtide/Hm0).

However, for a given tidal range, the models show different responses to a change in geometry (changing Ls/X). The PISCES model results suggest that the dimensionless

S


salient extent is practically unchanged for Ls/X > 0.8. On the other hand, the MIKE 21 CAMS results show an increase in the salient length with increasing Ls/X. These differences are probably due to differences in the representation of physical processes (waves, flow and sand transport) in the two models. A comparison of the physical processes in the two models is presented in Section 6.4.

The presence of tides leads to overall smoothing of the bathymetry contours, due to changes in the littoral zone width with tidal level and the additional shore-parallel currents induced by tides. The base of the salient is wider for the tidal cases compared with the non-tidal case and the bay is not as deep as in the non-tidal case.

Progressive tides result in deflection of the nearshore bathymetry in the direction of HW flow. For the same tidal range, the salient length is slightly increased for standing tides compared to progressive tides.

6.3.2 Effect of oblique wave incidence

The model results for the oblique wave cases show that sediment accumulates more in the lee of the up-drift breakwater than at the down-drift breakwaters.

The two numerical models used in the study give conflicting results on the effect of tidal range on the salient length for oblique wave incidence (45o at a depth of 15m). The MIKE 21 CAMS model shows a trend of decreasing salient length with tidal range (same as normal incidence), while the PISCES model shows the opposite trend (trend of increasing salient length with tidal range). It is concluded that these conflicting results may be due to differences in the processes represented in the two models. This is further discussed in Section 6.4.

6.3.3 Effect of breakwater crest level

The relative salient lengths (S/X) for different relative breakwater lengths (Ls/X=0.8, 1.33) are plotted against the relative tidal ranges in Figure 6.3, which shows that the relative salient length reduces as tidal range increases for shore normal waves. However, for large values of Ls/X (> 1.3), the influence of tidal range is not significant (if the breakwater is emergent through the tidal cycle).

The simulation results show that the relative salient length decreases as the breakwater crest level is reduced (Figure 6.4). The effect is more pronounced in cases where the breakwater is relatively close to the shoreline (Ls/X of 1.3 or more). As discussed in Chapter 5, the simulation results also show that the bed levels (over the salient) decrease as the breakwater crest level is reduced and more overtopping occurs over the breakwater.

The effect of the breakwater crest level is found to be insignificant if the maximum depth of water at HW (for a semi-diurnal tide) is about 0.5m or lower. In practical cases, it is expected that the limiting depth of submergence for negligible effect of breakwater crest level will also account for storm surge.

The breakwater scheme at Sea Palling, which consists of high crested breakwaters and low crested breakwaters, illustrates this effect. Observations have shown that the salient lengths behind the low crested breakwaters are significantly shorter and the beach levels lower compared to the high crested breakwaters (see Dolphin et al. 2009 in Appendix A2).


Effect of BW crest level on Salient length: R/Hm0=2.5

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 0.2 0.4 0.6 0.8 1 1.2 1.4

dcr / Hm0

S /

X

CAMS: Ls/X=1.33, Standing tides CAMS: Ls/X=0.80, Progressive tides

Figure 6.4 Effect of breakwater crest level (relative submerged depth at HW, dcr/Hm0) for different breakwater length and dimensionless tidal ranges (Rtide/Hm0).

6.3.4 Erosion in the breakwater bays

The results in Chapters 4 and 5 show that the amount of erosion in the breakwater bay generally reduces with increasing tidal range. There is more movement of the beach contours above MSL and, in particular, more erosion above MSL with increasing tidal range. This leads to a flattening of the beach profile.

The simulated MSL shorelines in the bay for emerged breakwaters in tidal cases (and shore normal waves) agree reasonably well with the predicted bay shoreline planforms generated by the method of Silvester and Hsu (1997). This result should, however, be viewed with caution, since the breakwaters in the two generic scheme layouts tested here are largely independent of one another because of the large gap width. In the case of multiple breakwaters, where the gap widths are such that the wave field in the bay is dependent on the gap width, it is not clear if the same result would be obtained.

6.4 Comparison of processes in model systems The two morphological model systems (PISCES and MIKE 21 CAMS) applied in this study showed the same general trends of morphological evolution in the vicinity of emergent breakwaters and the same influence of tidal range on the growth of the salient. However, conflicting results regarding the effect of tidal range on the morphological response (width of salient) for emergent breakwaters were found for oblique wave incidence. The PISCES results show increasing salient length with increasing tidal range, while the MIKE 21 CAMS results show a tendency for decreasing salient length with increasing tidal range. It is concluded that this conflicting result is due to differences in the representation of the physical processes in the model. Representations of key physical processes in the two models are compared in Table 6.1.


The main differences occur in the representation of wave diffraction, wave-current interaction (effect on wave parameters and apparent bed resistance) and graded/uniform sediments, and in the modelling of sand transport using equilibrium or non-equilibrium sand transport models. However, it was not possible to fully identify the reasons for the conflicting results within the scope of the present project.

Table 6.1 Representation of key physical processes in PISCES and MIKE 21 CAMS.

Parameter PISCES MIKE 21 CAMS

Remarks

Type of model

Spectral wave model

Parabolic mild slope model

Model name TOMAWAC-2G MIKE 21 PMS Gridspacing 10m in surf zone

and near breakwaters to 40m in remote areas

5m OK, if there are sufficient grid points (at least six points) to resolve the surf zone in both models, which is the case here.

Refraction/ shoaling

Yes Yes

Diffraction No Yes Diffraction is an important mechanism for wave energy in the lee of the breakwater. Its absence in the model may yield lower wave heights in the lee of the breakwater, and hence an increased potential for deposition.

Directional spreading

Yes (cos2 function) Yes (cos5 function)

Directional spreading is an important mechanism for wave energy in the lee of the breakwater, and if the spreading function is broad it may compensate somewhat for the absence of diffraction.

Frequency spectrum

Partly (JONSWAP);solves for m0 & m1

Yes (JONSWAP)

Wave breaking

Yes Yes

Bottom friction

Yes Yes

Wav

e m

odel

and

pro

cess

es

Wave-current interaction

Yes No

Wave current interaction may significantly affect the wave parameters if the ratio of the current speed to phase celerity is large (say > 0.1).

Type of model

2DH Flow model 2DH Flow model

Model name Telemac2D MIKE 21 Flow Gridspacing 10m in surf zone


5m OK, if there are sufficient grid points (at least six points) to resolve the surf zone in both models, which is the case here.

Flow

mod

el a

nd p

roce

sses

Time step 1s 2s OK, if time step is selected such that it complies with the stability requirements of the relevant model, which is the case here.


Parameter PISCES MIKE 21 CAMS

Remarks

Wave-driving forces

Yes Yes

Tide-driving forces

Yes Yes.

Eddy viscosity

Yes Yes

Wave-current interaction effect on bottom friction

No; constant values Yes This feature results in enhanced apparent bed resistance in the flow model (due to wave boundary layer). This results in reduced flow speeds and impact on sediment transport and morphology.

Flooding/ drying

Yes Yes

Type of model

Non-cohesive sediment transport model

Non-cohesive sediment transport model

Model name SANDFLOW MIKE 21 ST Gridspacing 10m in surf zone


5m OK, if there are sufficient grid points (at least six points) to resolve the littoral zone in both models, which is the case here.

Time step 1s Not applicable Specification of time step is only required in non-equilibrium sand transport models, where the advection-dispersion scheme is solved.

Intra-wave calculations

No Yes

Effect of sediment grading

Optional, but uniform sediment size used

Yes Model including this feature will estimate the gradation of sediment fractions in natural sand samples.

Non-equilibrium transport

Yes; advection-diffusion equations solved

No; equilibrium transport rates

Model including this feature will have spatial lag in the suspended sediment concentrations (especially for fine sediments), possibly leading to deposition over a larger area in an area of reducing sediment transport capacity.

Quasi-3D effects

Not included Not included

San

d tra

nspo

rt m

odel

and

pro

cess

es

Bed slope terms

No Yes This feature mainly affects the smoothness of the morphological response.


7 Conclusions and future work The outline design of a breakwater scheme consists of determining the geometrical parameters (Ls, X, G, hcr, B) needed to obtain a desired shoreline response. While empirical equations and design curves are available for determining the type of morphological response and salient length in non-tidal (or micro-tidal) cases (such as the design graphs discussed in Chapter 2), this is the first detailed study (to the knowledge of the authors) to focus on developing guidance on the effect of breakwaters on macro-tidal coasts.

The main conclusions from this study are summarised in Section 7.1, while suggestions for future work are summarised in Section 7.2. A companion report entitled Guidance for outline design of nearshore detached breakwaters on sandy macro-tidal coasts (Environment Agency 2009) documents the guidance derived from this study. The guidance report is aimed at assisting coastal practitioners who need to determine the geometrical layout of breakwater schemes at the option appraisal stage.

7.1 Conclusions

7.1.1 Non-tidal beaches

Using dimensional analysis, it was shown that when the breakwater cross-section and gap width between breakwaters are fixed the beach response for the non-tidal case is dependent on Ls/X and X/Xb. This shows that the use of Ls/X alone in typical design guidance for non-tidal beaches has limitations. Thus, this study also included a re-evaluation of the effect of breakwaters on micro-tidal beaches.

Using a composite dataset of morphological model results and laboratory experiments compiled by Suh and Dalrymple (1987), the following trends were identified.

• For a given relative breakwater location in the surf zone (X/Xb), the relative salient length (S/X) increases as the dimensionless breakwater length (LS/X) increases.

• For a given dimensionless breakwater length (LS), the relative salient length increases for low values of X/Xb and thereafter decreases, as should be expected for breakwaters located far away from the surf zone.

• Depending on the relative location of the breakwater in the surf zone, tombolo formation can occur for LS/X > 0.8.

• The limiting conditions for tombolo formation are postulated as: Ls/X > 2.8 – 1.6(X/Xb), X/Xb ≤ 1.25 Ls/X > -10.2 + 8.8(X/Xb), 1.25 < X/Xb ≤ 2.

• If the breakwater is located well beyond the limiting depth for littoral drift (X/Xb>2), it is postulated that a tombolo will not form.


7.1.2 Macro-tidal beaches

Effect of the type and range of tidal wave

It was found that the relative salient lengths reduce as the tidal range increases for shore normal waves. However, for large values of Ls/X (>1.3) the influence of tidal range is not significant (if the breakwater is emergent through the tidal cycle). Furthermore, the base of the salient is wider for the case of tides compared with the non-tidal case, but the bay is not as deep as in the non-tidal case.

Progressive tides (where the maximum current speed occurs near HW) result in deflection of the nearshore bathymetry in the direction of HW flow. For the same tidal range, the salient length is slightly increased for standing tides (where the maximum current speed occurs near MSL) compared with progressive tides.

Effect of oblique wave incidence

The model results for the oblique wave cases show that sediment accumulates more in the lee of the up-drift breakwaters than at the down-drift breakwaters. The two numerical models used in the study show conflicting trends for the effect of tidal range on salient length. Further work is required to clarify the effect of oblique wave incidence, especially in the case of frequently overtopped breakwaters, which was not within the scope of the present study.

Fortunately, the incident wave conditions at any given site typically consist of a range of wave directions. Furthermore, detached breakwaters are typically oriented to be shore parallel, which usually means they are at a small angle to the dominant wave direction. Thus, it is suggested that for practical cases the indicative trend for shore-normal waves should be used.

Effect of breakwater crest level

The model results show that the relative salient length reduces as the breakwater crest level is reduced. The effect is more pronounced for cases where the breakwater is relatively close to the shoreline (Ls/X ≥1.3). Furthermore, the beach levels fall as the breakwater crest level is reduced. This is also confirmed by observations at Sea Palling, which show that the salient lengths behind low crested breakwaters are significantly shorter and the beach levels are lower compared with high crested breakwaters.

Erosion in the breakwater bays

The model results show that the amount of erosion in the breakwater bay generally reduces with increasing tidal range. However, there is more movement of the beach contours above MSL and, in particular, more erosion above MSL with increasing tidal range.

It was found that the simulated MSL shorelines in the bay for emerged breakwaters in tidal cases agree reasonably well with the predicted bay shoreline planform generated by the method of Silvester and Hsu (1997). It should, however, be considered that this result is obtained for test cases where the breakwaters are largely independent of one another because of the large gap width. In the general case of multiple breakwaters


where the gap widths significantly influence the wave conditions in the bay, it is presently not clear if the same result would be obtained.

7.2 Future work This study is a step towards improving our understanding of the effect of tidal processes on beach changes in the lee of breakwaters. The results from the study have been used to propose design guidance that accounts for the influence of tides on beach response in the lee of breakwaters. However, to realise fully the goal of improved guidance on the various geometrical parameters required for outline guidance, we have identified some gaps in knowledge that can be filled by additional work. The proposed additional work is summarised in Table 7.1.


Table 7.1 Additional tasks to improve further the outline design guidance for breakwaters on macro-tidal coasts.

Task Benefits Investigation of the effect of varying wave obliquity for different submerged depths (dcr/Hm0) at the breakwater crest during HW.

The present study has investigated the effect of normal wave incidence on low-crested breakwaters in macro-tidal areas, and used this to obtain design guidance on breakwater crest level and salient length. The effect of oblique wave incidence in this situation is presently unclear. The main benefit of this work will be to clarify the effect of oblique wave incidence and thereby enhance the value of the design guidance.

Investigation of the effect of varying gap width on beach erosion within the breakwater bays.

The effect of varying gap width was not considered in the present study. It is noted that guidance on the gap width is required for selecting the length and number of breakwaters required to protect a given beach frontage. Thus, the main benefit will be to provide additional information for the guidance that is presently not available.

Further investigation of the effect of tidal range and emerged breakwaters for additional cases of R/Hm0 and L/X.

This task will add more points to the design graph on the effect of tidal range, thus resulting in improved guidance for outline design.

Further investigation of the effect of tidal range (Rtide/Hm0) and breakwater crest submergence at HW (dcr/Hm0) on beach response.

This task will add more points to the design graph regarding the effect of tidal range on low crested breakwaters, resulting in improved guidance for outline design.

Consideration should be given to carrying out a validation study at a macro-tidal coast in the UK.

The numerical models used in this study are state-of-the-art morphological models. However, the models have not yet been validated against beach response in the lee of breakwaters in a macro-tidal beach.

Thus, the aim of the validation exercise would be to increase confidence in the model studies and thus in the outline design guidance.

At completion of the proposed future work, consideration should be given to developing a simple tool (such as a java applet) using the updated design curves to aid in the outline design stage.

This will aid coastal practitioners in the outline design stage and minimise the risk of incorrect usage of the design curves.


References ABBOTT, M.B., DAMSGAARD, A. AND RODENHUIS, G.S., 1973. System 21, Jupiter. A design system for two-dimensional nearly-horizontal flows. Journal of Hydraulic Research, 11 (1), 1-28.

BATTJES, J.A. AND JANSSEN, J.P.F.M., 1978. Energy loss and set-up due to breaking of random waves. In: Proceedings of the 16th international conference on coastal engineering, 1978 Hamburg, Germany. New York: American Society of Civil Engineers, 569–578.

BENOIT, M., MARCOS, F. AND BECQ, F., 1996. Development of a third generation shallow-water wave model with unstructured spatial meshing. In: Proceedings of the 25th international conference on coastal engineering, 2–6 September 1996 Orlando, Florida. New York: American Society of Civil Engineers, 465–478.

BOOIJ, N., RIS, R.C. AND HOLTHUIJSEN, L.H., 1999. A third-generation wave model for coastal regions 1. Model description and validation. Journal of Geophysical Research, 104 (C4):7649–7666.

BOS, K.J., ROELVINK J.A. AND DINGEMANS, M.W., 1996. Modelling the impact of detached breakwaters on the coast. In: Proceedings of the 25th international conference on coastal engineering, 2–6 September 1996 Orlando, Florida (cited in Van Rijn 2005).

BSI, 1991. Maritime structures. Guide to the design and construction of breakwaters. London: BSI, (BS 6349-7).

CIRIA, 1996. Beach management manual. London: CIRIA, (CIRIA Report 153).

CIRIA, CUR, CETMEF, 2007. The rock manual: the use of rock in hydraulic engineering. 2nd ed. London: CIRIA, (C683).

COATES, T., 1994. Effectiveness of control structures on shingle beaches, physical model studies. Wallingford, UK: HR Wallingford, (Report SR288).

D’ANGREMOND, K., VAN DER MEER, J.W. AND DE JONG, R.J., 1996. Wave transmission at low-crested structures. In: Proceedings of the 25th international conference on coastal engineering, 2–6 September 1996 Orlando, Florida. New York: American Society of Civil Engineers, 2418– 2427.

DAMGAARD, J., DODD, N., HALL, L. AND CHESHER, T.J., 2003. Morphodynamic modelling of rip channel growth. Coastal Engineering, 45, 199–221.

DEIGAARD, R., FREDSØE, J. AND HEDEGAARD, I.B., 1986a. Suspended sediment in the surf zone. Journal of Waterway, Port, Coastal, and Ocean Engineering, 112 (1), 115–128.

DEIGAARD, R., FREDSØE, J. AND HEDEGAARD, I.B., 1986b. Mathematical model for littoral drift. Journal of Waterway, Port, Coastal, and Ocean Engineering, 112 (3), 351– 369.

DHI SOFTWARE, 2008a. MIKE 21 coastal area morphological modelling shell, user guide and reference manual. Available from DHI Water and Environment, Denmark.

DHI SOFTWARE, 2008b. MIKE 21 parabolic mild-slope module, user guide and reference manual. Available from DHI Water and Environment, Denmark.

DHI SOFTWARE, 2008c. MIKE 21 flow model, user guide and reference manual. Available from DHI Water and Environment, Denmark.


DHI SOFTWARE, 2008d. MIKE 21 non-cohesive sediment transport module, user guide and reference manual. Available from DHI Water and Environment, Denmark

ENET, F., NAHON, A., VAN VLEDDER, G. AND HURDLE, D., 2006. Evaluation of diffraction behind a semi-infinite breakwater in the SWAN wave model. In: Proceedings of the 9th international symposium on ocean wave measurement and analysis – WAVES06, 2006 Victoria, Canada.

ENVIRONMENT AGENCY, 2009. Guidance for outline design of nearshore detached breakwaters on sandy macro-tidal coasts. Bristol, UK: Environment Agency, (SC0600026/SR2).

FLEMING, C. AND HAMER, B., 2000. Successful implementation of an offshore reef scheme. In: Proceedings of the 27th international conference on coastal engineering, 16–21 July 2000 Sydney, Australia. New York: American Society of Civil Engineers.

EurOtop, 2007. Wave overtopping of sea defences and related structures: assessment manual. Wallingford, UK: H.R. Wallingford.

FREDSØE, J., 1984. Turbulent boundary layer in wave-current motion. J. Hydraul. Eng., 110 (8), 1103–1120.

GODA, S., 1985. Random seas and design of maritime structures. Tokyo: University of Tokyo Press.

HANSON, H. AND KRAUS N.C., 1990. Shoreline response to a single transmissive detached breakwater. In: Proceedings of the 22nd international conference on coastal engineering. New York: American Society of Civil Engineers, 2034-2046.

HENLEY, M.A., 1998. Approaches to data collection. Scientific Research Matters, 7 (2), 152–165.

HERVOUET, J.-M., 1997. Hydrodynamics of free surface flows: modelling with the finite element method. Chichester, UK: John Wiley & Sons.

ILIC, S., CHADWICK, A. AND FLEMING, C., 2005. Investigation of detached breakwaters. Part 1 – hydrodynamics. Proceedings of the Institution of Civil Engineers – Maritime Engineering, 158 (3), 91–102.

ILIC, S CHADWICK, A, FLEMING, C., 2005. Investigation of detached breakwaters. Part 2 – morphodynamics. Proceedings of the Institution of Civil Engineers – Maritime Engineering, 158 (4),163–172.

JOHNSON, H.K., BRØKER, I., ZYSERMAN, J.A. AND MANGOR, K., 1995. Morphological response in the vicinity of offshore breakwaters. In: Proceedings of the 4th international conference on coastal and port engineering in developing countries, 25–29 September 1995 Rio de Janeiro, Brazil.

JOHNSON, H., WILKENS, J., PARSONS, A., CHESHER, T. AND HERON, E., 2009. Towards design guidance for nearshore detached breakwaters on macro-tidal coasts. Submitted to ICE conference on coasts, marine structures and breakwaters, 16-18 September 2009.

JOHNSON, H.K., 2006. Wave modelling in the vicinity of submerged breakwaters. Coastal Engineering, 53 (1), 39–48.

JOHNSON, H.K., ZYSERMAN, J.A., BRØKER, I., MOCKE, G., SMIT, F. AND FINCH, D., 2005. Validation of a coastal area morphological model along the Dubai coast. In: Proceedings of Arabian Coast 2005, 27–29 November 2005 Dubai, UAE. See also: www.encora.eu/coastalwiki/Process-based_morphological_models [Accessed 10 September 2009).


JOHNSON, H.K. AND ZYSERMAN, J.A., 2002. Controlling spatial oscillations in bed level update schemes. Coastal Engineering, 46 (2), 109–126.

KIRBY, J.T., 1986. Rational approximations in the parabolic equation method for water waves. Coastal Engineering, 10, 355–378.

MILES, G., 1991. Transport of sand mixtures. In: R.L.S. SOULSBY AND R. BETTESS, eds. Sand transport in rivers, estuaries and the sea: proceedings of the Euromech 262 Colloquium 26–29 June 1990 Wallingford, UK. Rotterdam, NL: Balkema Publications, 253–257.

NICHOLSON, J., BROKER, I., ROELVINK, J.A., PRICE, D., TANGUY, J.M. AND MORENO, L., 1997. Intercomparison of coastal area morphodynamic models. Coastal Engineering, 31, 97–123.

O’CONNOR, B., NICHOLSON, J. AND MACDONALD, N., 1995. Modelling morphological changes associated with an offshore breakwater. In: C.A. Bebbia, L. Traversoni and L.C. Wobel, eds. Computer modelling of seas and coastal regions. Southampton, UK: WIT Press, 215–272.

PERLIN, M., 1979. Predicting beach planforms in the lee of a breakwater. In: Proceedings of Coastal Structures’79, American Society of Civil Engineers, 792-808.

PILARCZYK, K.W. AND ZEIDLER, R. B., 1996. Offshore breakwaters and shore evolution control. Rotterdam, the Netherlands: A. A. Balkema.

RIS, R.C., BOOIJ, N. AND HOLTHUIJSEN, L.H., 1999. A third-generation wave model for coastal regions 2. Verification. Journal of Geophysical Research, 104 (C4), 7667–7682.

ROELVINK, J.A., 2005. Coastal morphodynamic evolution techniques. Coastal Engineering, 53, 277–287.

ROGERS, J., CHESHER, T. AND HAMER, B., 2006. LEACOAST2: Practical guidance scoping study (R&D technical report MP0107/TR1 March 2006). Available from: Defra, Flood Management Division, Ergon House, Horseferry Road, London SW1P 2AL.

ROSEN, D.S. AND VAJDA, M., 1982. Sedimentological influences of detached breakwaters. In: Proceedings of the 18th international conference on coastal engineering 1982 Cape Town, South Africa. New York: American Society of Civil Engineers, 1930–1949 (cited in Suh and Dalrymple 1987).

SAYED A., 1997. 2D and 1D numerical model simulations for the effect of a single detached breakwater on the shore. Report Q1800.22. Delft, The Netherlands: Delft Hydraulics, (cited in Van Rijn 2005).

SHINOHARA K. AND TSUBAKI, T. 1966. Model study on the change of shoreline of sandy beach by the offshore breakwater. In: Proceedings of the 10th international conference on coastal engineering 1966, 550-563 (cited in Suh and Dalrymple 1987).

SILVESTER, R. AND HSU, J.C.R., 1997. Coastal stabilization: Advanced series on ocean engineering, Vol. 14. Singapore: World Scientific.

SIMM, J.D., BRAMPTON, A.H., BEECH, N. AND BROOKE, J., 1996. Beach management manual. London: CIRIA, (CIRIA Report 153).

SMITH, G.A. AND KEENE, B., 1995. Habitat decline in the UK. 2nd ed. London: Collins.

SUH, K. AND DALRYMPLE, R.A., 1987. Offshore breakwaters in laboratory and field. Journal of Waterway, Port, Coastal, and Ocean Engineering, 113 (2), 105–121.


SUTHERLAND, J., WALSTRA, D.J.R., CHESHER, T.J., VAN RIJN, L.C. AND SOUTHGATE, H.N., 2004. Evaluation of coastal area modelling systems at an estuary mouth. Coastal Engineering, 51, 119-142.

VAN RIJN, L.C., SOULSBY, R.L., HOEKSTRA, P. AND DAVIES, A.G., eds., 2005. Sandpit: sand transport and morphology of offshore sand mining pits. Blokzijl, The Netherlands: Aqua Publications.

VAN RIJN, L. C., 2005. Principle of sedimentation and erosion engineering in rivers, estuaries and coastal seas. Blokzijl, The Netherlands: Aqua Publications.

U.S. ARMY CORPS OF ENGINEERS, 2003. Coastal engineering manual. Washington, D.C: U.S. Army Corps of Engineers, (Engineer Manual 1110-2-1100).

ZYSERMAN, J., BRØKER, I., JOHNSON, H., MANGOR, K. AND JORGENSEN, K., 1998. On the design of shore-parallel breakwaters. In: B.L. Edge, ed. Proceedings of 26th coastal engineering conference 22–26 June 1998 Copenhagen, Denmark. New York: American Society of Civil Engineers, 1693–1704.

ZYSERMAN, J.A. AND JOHNSON, H.K., 2002. Modelling morphological processes in the vicinity of shore-parallel coastal breakwaters. Coastal Engineering, 45, 261–284.

ZYSERMAN, J.A., JOHNSON, H.K., ZANUTTIGH, B. AND MARTINELLI, L., 2005. Far-field erosion and morphological response induced by low-crested rubble-mound structures. Coastal Engineering, 52, 977–994.

.


Appendices


A LEACOAST2 short papers This appendix contains eight short papers prepared by the participants in the parallel LEACOAST2 research consortium funded by the EPSRC. The LEACOAST2 research focused on understanding the long-term morphological evolution of the breakwater scheme at Sea Palling. A list of the short papers is given in Table A.1.

In the main report, morphological modelling of various combinations of generic cases are used as the basis for determining outline design guidance for detached nearshore breakwaters on sandy coasts. This appendix complements the report by providing detailed information and analysis at a specific site.

Table A.1 Overview of EPSRC short papers.

Paper Author(s) Title of paper A1

Shunqi Pan4 et al. Overview of the LEACOAST2 project

A2 Dolphin, T.J.1, Vincent, C.E.1, Dumont, E.1 and Dufflo, C.1

Medium-term shoreline evolution in the vicinity of shore-parallel breakwaters at Sea Palling

A3 Fairley, I.2 and Davidson, M.2

Video-based analysis of morphological changes behind the Sea Palling breakwaters

A4 Judith Wolf3, Peter Thorne3, Paul Bell3, Richard Cooke3 and Alejandro Souza3

Wave, currents and sediment transport observed during the LEACOAST2 experiment

A5 Paul Bell3 Remote bathymetry, bedforms and current mapping using marine radar during the LEACOAST2 experiment

A6 Shunqi Pan4 and Yanliang Du4

A short report on LEACOAST2 process modelling

A7 Ming Li5, Nichols Spanakis5 and Brian A. O’Connor5

Numerical study of ripple dynamics and its impacts on morphodynamics at Sea Palling

A8 Dominic Reeve4 and Baoxing Wang4

Probabilistic simulation of beach morphodynamics within a flood defence scheme

Notes: 1School of Environmental Sciences, University of East Anglia, Norwich 2Coastal Processes Research Group, University of Plymouth, Plymouth 3Proudman Oceanographic Laboratory (POL), Joseph Proudman Building, Liverpool, L3 5DA 4University of Plymouth, Plymouth 5University of Liverpool, Liverpool


An overview of the LEACOAST2 project is provided in paper A1. In general, the LEACOAST2 research tasks comprised:

• observations and analysis of morphological response using aerial photos (paper A2 – Dolphin et al.) and images obtained from remote sensing (paper A3 – Fairley and Davidson);

• measurements of waves, flow, sediment transport and bathymetry (paper A4 – Wolf et al.; paper A5 – Bell); and

• morphological modelling of the observed response using a process-based coastal area model (paper A6 – Pan and Du; paper A7 – Li et al.) and a shoreline evolution model (paper A8 – Reeve and Wang).

The observed morphological response in papers A2 and A3 broadly confirms some of the trends in the simulated morphological response from the generic modelling. For instance, the different observed responses in the lee of the high breakwaters and the low-crested breakwaters are also reproduced in the generic modelling.

Papers A4 and A5 summarise the data collected, which are used in the morphological modelling studies presented in papers A6 through A8.

Papers A6 and A7 show some results from the coastal area modelling studies carried out at Sea Palling. Paper A6 illustrates the effect of enhanced water levels due to storm surge on the morphological changes at Sea Palling. It is also noted that the surge level correlates well with high values for wave heights. The paper demonstrates that significant sediment volume changes correlate well with these large events. This provides some support for using the moderately high offshore wave heights (Hm0=2m) in the generic modelling study, as these conditions generally correspond to large morphological changes. Paper A7 analyses the impact of ripple dynamics on the morphological evolution at the site.

Paper A8 presents the results of applying one-line shoreline modelling concept with probabilistic techniques at the Sea Palling site.


A.1 Overview of the LEACOAST2 project Larger-scale Morphodynamic Impacts of Segmented Shore-Parallel

Breakwaters on Coasts and Beaches: An Overview of the LEACOAST2 Project

Shunqi Pan1, Brian O’Connor2, Chris Vincent3, Dominic Reeve1, Judith Wolf4, Tim Chesher5, Hakeem Johnson6, Mark Davidson1, Tony Dolphin3, Pete

Throne4, Paul Bell4, Adam Leadbetter7

1School of Engineering, University of Plymouth 2Department of Engineering, University of Liverpool

3School of Environmental Sciences, University of East Anglia 4Proudman Oceanographic Laboratory

5HR Wallingford 6Halcrow Group

7Bristish Oceanographic Data Centre

Introduction Coastal structures, such as detached nearshore breakwaters, have been widely used. The UK applications include the breakwaters at Kings Parade, Wirral; Elmer, Sussex; and Sea Palling, Norfolk. However, most current UK structures were designed a decade or more ago with the object of providing appropriate levels of flood protection as well as resisting the worst-storm conditions likely to be experienced over the lifetime of the structures and also minimizing the long-term (25-50 years) impact of the structures on adjacent coastlines. Unfortunately, existing design guidelines rely heavily on micro-tidal experience, and even this experience is imperfect as demonstrated by the removal of structures in the USA and the use of modern computer methods, which show the inability of some engineering criteria to correctly predict the formation of salients and tombolos in the lee of such structures, O’Connor et al (1995).

The LEACOAST2 project attempts to address the knowledge gaps identified above, so that enhanced design tools and integrated monitoring approaches can be further developed to assist future engineering studies and coastal planning projects with additional information to be gathered and analyzed under a parallel companion project funded by Defra/EA.

This paper provides an overview of the main tasks carried out in the LEACOAST2 project, whilst further details on the different aspects of the project are given in other papers contained in this report.

Research Objective The main objective of the LEACOAST2 research is to evaluate the generic effects of shore-parallel breakwaters in tidal conditions on coastal morphology on scales of kilometres and years, using a combination of deterministic and probabilistic morphological modelling and new longer-term hydrodynamic and morphological data, as well as to provide enhanced tools to improve the design guidelines which is undertaken in parallel by the industrial partners of the project funded by Defra/EA.

In order to achieve the research objective, the project makes use of an extensive data base of information which has been built up for the study site at Sea Palling on the Norfolk coastline from past research. These structures, which have been in operation


for some twenty years, were built in two phases including 4 surface-piercing breakwaters (high-crested) in Phase I and 5 over-topped breakwaters (low-crested) in Phase II, as shown in Figure A1.1. The high-crested breakwaters built in Phase I are longer and wider spaced in comparison with the low-crested breakwaters built in Phase II.

This research builds upon an earlier research project at the same site – LEACOAST also funded by the EPSRC, which used a combination of process-based computer models and newly-obtained local area field data to study the storm-scale response of a particular representative breakwater embayment (Bacon et al 2004; Dolphin et al 2004, Pan et al 2004) over the medium term period 2002-2004. The LEACOAST2 project is with much extended study area, computer modelling and field work.

Figure A1.1 LEACOAST2 study site: Sea Palling, Norfolk.

Aerial photographs & Remote sensing monitoring Aerial photographs provide a means to monitor shoreline positions over a given period of time. Shoreline positions were obtained by scanning and geo-referencing Environment Agency aerial photographs (1991-2005). The instantaneous shoreline are digitised from the photographs and adjusted to account for the difference between the tidal level at the time of photography and mean sea level. The results of this investigation is summarised in Appendix A.2 (Dolphin et al., 2009).

Video remote sensing provides a unique methodology for monitoring morphodynamic evolution over distances of a kilometre, at a frequency that is simply not practical using conventional surveying techniques. For the LEACOAST2 project, a 5-camera video-system operated by the University of Plymouth (UP) team, see Figure A1.2(a), was used to resolve changes in morphology at daily, storm event and seasonal time-scales. Combining the video data from ARGUS with regular beach and bathymetry surveys, and the extensive archive of survey data already available for the site from the LEACOAST project and the EA, made it possible to study the long-term (>10 years) variability of the system. The results of this investigation is summarised in Appendix A.3 (Fairley and Davidson, 2009).

Video Cameras

X-Band Radar


Radar systems also provide a convenient imaging system, allowing large areas of the sea surface to be imaged at relatively shallow grazing angles. It has been demonstrated, Wolf & Bell (2001), that marine X-band radar could be used with appropriate digital recording systems to provide the required image sequences for bathymetric inversions. Both linear and non-linear wave theories can be used to produce a depth inversion algorithm. A new millimetre wave (MMW) radar system nested within the X-band radar can provide fine details of surf zone and swash processes to a range of O(200m), while the marine radar provides lower resolution images of O(10m) to longer ranges of O(2km) (Bell et al, 2004), as shown in Figure A1.2(b), operated by POL. The results of this investigation is summarised in Appendix A.5 (Bell, 2009).

a)

b)

Figure A1.2 Remote sensing equipment – a) ARGUS Video system; b) X-Band Radar.

Field measurements Large scale field measurements were planned and executed over the two winter periods (Sept-Oct 2006 and Oct-Dec 2006). The fieldwork includes the measurements of hydrodynamics, sediment concentrations, particle size and bedforms using the tripod frames at various locations in the vicinity of the breakwaters, as shown in Figure A1.3. New marine acoustic instrumentation, developed in the UK over the last few years, were used to make direct measurements of the sediment transport and associated hydrodynamic forcing parameters at a number of points located both within the SSPB system and just outside the SSPBs in the region where sand by-passing of the system may be occurring. High-frequency acoustic backscatter instruments (ABS) measured the suspended sand concentration at intra-wave timescales while acoustic Doppler systems profiled the water column and measure turbulent intensities and stresses. Rotary acoustic scatters were used to measure bedforms and bedform migration rates at the same time (essential for the estimation of bedload transport).

Deployment of the instrument frames was a critical part of the field measurement, and was also the most hazardous. The offshore frames were deployed and recovered by a boat, see Figure A1.4(a) and the nearshore frames were deployed by the heavy duty machineries (JCBs), see Figure A1.4(b).

The field measurements are summarised in Appendix A.4 (Wolf et al., 2009).


a)

b)

Figure A1.3 a) Measurement locations; b) Instrument Frame.

a)

b)

Figure A1.4 Deploying and recovering instrument frames – a) by boat; b) by machinery.

Numerical modelling Both process-based and probabilistically modelling techniques were employed in the project with a view to achieve better understanding of impact of structures on the adjacent beaches under the short-term (storm) conditions with detailed nearshore processes and longer term (years) impact on shoreline changes.

The process-based model studies are summarised in Appendix A.6 (Pan and Du, 2009) and Appendix A.7 (Li et al, 2009), while the probabilistic modelling study is summarised in Appendix A.8 (Reeve and Wang, 2009).

Acknowledgments

This work is partly supported by the EPSRC under grant numbers: EP/C010965, EP/C010930 & EP/C013085. Support from the EA/Defra during the course of project is


also acknowledged. The authors would like to thank the followings for their valuable contributions to the project: Ben Hamer, Steve Hayman, Noel Beech, Alex Souza, John Huthnance, Jonathan Rogers, Jon Williams, Ming Li, John Bacon, Prem Fernando, Robin McCandliss, Dick Weight, Clare Coughlan, Yanliang Du, Yongping Chen, Baoxing Wang, Iain Fairley, Rodolfo Bolanos, Stefan Laeger, Andy Parsons, Jort Wilkens, Isabel Garcia Hermosa, Roger Phillips, Ben Moate, Philip Staley, Joanne Parry and Estelle Dumont.

References BACON, J.B., VINCENT, C.E., DOLPHIN, T.J. AND TAYLOR, J., 2004. The offshore breakwater scheme at Sea Palling, England: Sand transport generated by tidal currents. In Proceedings of the 29th International Conference on Coastal Engineering, Lisbon, Portugal, 2004, 2: 1896-1908. BELL, P. S., WILLIAMS, J. J., CLARKE, S., MORRIS, B. AND VILA CONCEJO, A. (2006), “Nested radar systems for remote coastal observations”, Journal of Coastal Research, SI39, 483–487 DOLPHIN, T.J., TAYLOR, J.A., VINCENT, C.E., BACON, J.C., PAN, S. AND O'CONNOR, B.A. (2004), “Storm-scale effects of shore-parallel breakwaters on beaches in a tidal setting (LEACOAST)”, Book of Abstracts, ICCE2004, Lisbon O'CONNOR B. A., NICHOLSON J., & MACDONALD N. J. (1995), “Modelling morphological changes associated with an offshore breakwater”, In Computer Modelling of Seas and Coastal Regions II, CA Brebbia, L Traversoni and LC Nrobel (eds), Comp. Mech. Pubs. S'ton, pp. 215-272. PAN, S., VINCENT, C. E., FERNANDO, P. T. , LI, M., ZHU, Y., TAYLOR, J.A., DOLPHIN, T. J., BACON, J.C. AND O'CONNOR, B. A. (2004), “Effect of shore parallel breakwaters on coastal morphology under storm conditions”, Abstract accepted for The Coastlines, Structures and Breakwaters Conference, London, April 2005 WOLF, J. and BELL P. S. (2001), “Waves at Holderness from X-band radar”, Coastal Engineering, Vol. 43(3-4), pp. 247-263


A.2 Medium-term shoreline evolution

Medium-term shoreline evolution in the vicinity of shore-parallel breakwaters, Sea Palling, UK

Dolphin, T.J., Vincent, C.E., Dumont, E. and Dufflo, C.

School of Environmental Sciences, University of East Anglia Introduction This short paper describes the shoreline response to the construction of the so called phase-1 and phase-2 breakwaters in 1995 and 1997 at Sea Palling, Norfolk. Detail can be found in Dolphin et al. in prep). Shoreline positions were attained by scanning and geo-referencing Environment Agency aerial photographs (1991-2005), digitising the instantaneous shoreline, and adjusting the shoreline to account for the difference between the tidal level at the time of photography and mean sea level datum. The adjustment of shoreline positions to the common mean-sea-level datum was achieved by translating the shoreline a distance Δx using the datum – tide level difference and the beach slope taken from measured Environment Agency beach profiles. The data are presented as time-series of distances from a baseline along each of 70 shore-parallel transects spaced every 50 m. The ‘average’ rates of change along each transect were determined by linear fit to the time-series of shoreline positions (refer Thieler et al., 2003).

Results and discussion The persistent spatial pattern in the shoreline rate-of-change is of alternating local maxima and minima associated with breakwaters (tombolos and salients) and breakwater gaps (embayments)(Figure A2.1). The gap to breakwater differences are large in phase 1 (4.8 – 9.7 m/yr), especially Bay B, whilst in phase 2 the gap to breakwater difference rates are smaller (0.7 – 3.7 m/yr). These differences reflect formation of the wider tidal-tombolos in phase 1 (250 m wide) and the more subdued salients (75-95 m wide) in phase 2.

A 500-m running mean, corresponding to the length scale of a phase 1 breakwater and gap (red dashed line, Figure A2.1), highlights the broad spatial pattern of accreting shorelines at either end of the system and a global minima in the centre (-6.03 m/yr in Bay E) near the junction between the phase 1 and phase 2 breakwaters. Both embayments and tombolos/salients follow the trend of decreasing accretion to increasing erosion toward Bay E near the centre of the breakwater system. The lower beach width toward the centre of the system suggests littoral drift sediments are yet to be deposited and retained there. However, wider accreting beaches at either end of the system trap and retain sediment moving alongshore during individual storm events as it encounters the breakwater system. As the net littoral drift is from the north (Vincent, 1979), wider beaches have formed at the north end. The higher breakwaters in the north are also responsible for the wider beaches there. End-effects can also be observed at Tombolo 8 (Figure A2.1), which was the southern end of the breakwater system in 1995 – 1997, prior to phase-2 breakwater construction.


9 10 11 12 138765A B C D E F G H

100 105 110 115 120 125 130 135 140 145 150 155 160 165Transect ID

-5

0

5

10S

hore

line

chan

ge (m

/yr)

FigureA2.1 Medium-term (1991-2005) rate of shoreline change.

Transect spacing is 50 m. The red dashed line is a 500 m running mean. Following the LEACOAST project convention the breakwaters are numbered 5 – 13 and embayments are labelled A – H, from north to south.

Inspection of the time-series (Figure A2.2) reveals the broad shoreline response that controls the rate of change statistics shown in Figure A2.1. In all cases the greatest shoreline changes occurred in response to a 300,000 m3 recharge in 1996 (phase-2 only) followed by a 1,000,000 m3 recharge in 1997 (all breakwaters). On average, the shorelines advanced 160 m behind phase-1 breakwaters, 50 m behind phase-2 breakwaters, 30 m in phase-1 gaps and 15 m in phase-2 gaps. These results highlight variation in shoreline response (trapping and retention of natural and recharge sediments) as a result of differences in breakwater design. The longer and higher phase-1 breakwaters have low transmission (overtopping is insignificant) and create a large deposition zone with wider beaches in comparison to the phase-2 breakwaters where the lower and shorter breakwaters have higher transmission (breakwaters are submerged at high tide) and result in narrower beaches and shorter residence times for recharge sediments (Figure A2.2).

High growth rates and shoreline advances resulting from recharge led to the formation of an all-tide tombolo behind breakwater 5, and three tidal tombolos (tombolos at low tide, salients at high tide) behind breakwaters 6 – 8 (Figure A2.2). Only salients, not tombolos, formed during the 1995 – 1997 period prior to recharge. Following recharge the shorelines in the lee of the phase-1 breakwaters advanced and became stable, with the exception of tombolos 6 and 7 that shrank and retreated in 2005 (Figure A2.2). At this time a channel developed at the junction of the Bay A beach face and the flat and dissipative bay floor. The channel lengthened and eventually cut through tombolo 6, following which the elevation of the separated seaward part fell until it was level with, and part of, the adjacent Bay A. At the same time tombolo 7 also retreated.

The recharge-induced formation of tombolos is significant to the sediment transport regime. Tombolo 5 acts as a 250 m long groyne, prohibiting longshore sediment transport except during submergence associated with large storm surges and high spring tides. The tidal tombolos (6 – 8) have a much lower elevation and width, but they also inhibit sediment transport: the tombolos emerge at low tide, during which ebb currents peak, and significantly reduce ebb-directed sediment transport (Bacon et al., 2004).

In the breakwater gaps of phase-1, shorelines erode post-recharge, with the exception of Bay A, which is relatively stable. The other gap shorelines in phase-1 have differing responses to recharge, but all erode post-recharge, contributing to the net erosion


rates shown in Figure A2.1. Bays B and C cannot retreat further as they are anchored by two rock ‘windrows’, parallel and in front of the sea wall. The windrows prevent beach lowering in front of the sea wall.

Tombolo 5 (T100)Tombolo 6 (T110)Tombolo 7 (T120)Tombolo 8 (T130)Mean

1/1/911/1/93

1/1/951/1/97

1/1/991/1/01

1/1/031/1/05

150

200

250

300

350

Dis

tanc

e (m

)

21

1/1/911/1/93

1/1/951/1/97

1/1/991/1/01

1/1/031/1/05

150

200

250

300

350

Dis

tanc

e (m

)

Salient 9 (T136)Salient 10 (T143)Salient 11 (T150)Salient 12 (T156)Salient 13 (T162)Mean

21

Bay A (T105)Bay B (T115)Bay C (T125)Bay D (T134)Mean

1/1/911/1/93

1/1/951/1/97

1/1/991/1/01

1/1/031/1/05

80

120

160

200

240

Dis

tanc

e (m

)

21

1/1/911/1/93

1/1/951/1/97

1/1/991/1/01

1/1/031/1/05

80

120

160

200

240

Dis

tanc

e (m

)

Bay E (T140)Bay F (T146)Bay G (T153)Bay H (T159)Mean

21

Figure A2.2 Shoreline width adjacent to breakwaters (tombolos and salients) and gaps (embayments). The data are from the breakwater and gap centres, and the T numbers in brackets are the transect ID’s as shown on Figure A2.1. The construction phases of the breakwaters are labelled and marked as .The proportional circles on the time axis indicate the timing (1996 and 1997) and magnitude (300,000 m3 in phase-2 followed by 1,000,000 m3 in phase 1 and 2) of sediment recharge. Note that the distance scales differ between the phase-1 shorelines (top panel) and the phase-2 shorelines (bottom panel). The grey line is the mean response in each graph. The spike in the Bay A record (2004) is an intertidal bar exposed during aerial photography. Behind phase-2 breakwaters, a subdued shoreline response to the 1996/7 recharge is followed by gradual retreat, with the exceptions of Bay H and salients 12 and 13 at the southern end of the breakwater system. In the case of salients 9 – 11 the shorelines erode back to their pre-recharge positions, more or less, over the 1998-2004 period (residence time of up to 6 years). The phase-2 embayments follow a similar pattern, only shoreline retreat is equivalent to, or larger than, the recharge gain. The exceptions – bay H and salients 12 and 13 – are likely to receive littoral drift sediments that appear, in this case, to balance losses (offshore?) experienced in the other parts of phase 2.

The aerial photo data suggest that the mean sea level shoreline in bay A stabilised in 1998 and has remained more or less constant since. However, Environment Agency beach profiles and 3-dimensional beach surveys conducted at various times between 2003 and 2007 (LEACOAST) suggest that the bay floor has gradually infilled (1997 –


2005), and was in dynamic equilibrium (bed level range < ~ 0.6 m) between 2005 – 2007. Bay B is much less infilled than bay A, with depths typically 2-3 m greater in Bay B. Retention of recharge sediments here, as with other embayments, is low and difficult to distinguish from natural variability in bed levels from beach profiles. In Bay B, the sea floor elevation fluctuated until 2005 and appears to be relatively stable over the 2006-2007 period. EA profiles and those extracted from 3D beach surveys indicate a recent period of stability that could suggest an equilibrium condition has been approached. However, a longer record is required to substantiate this suggestion.

References BACON, J.B., VINCENT, C.E., DOLPHIN, T.J. AND TAYLOR, J., 2004. The offshore breakwater scheme at Sea Palling, England: Sand transport generated by tidal currents. In Proceedings of the 29th International Conference on Coastal Engineering, Lisbon, Portugal, 2004, 2: 1896-1908. DOLPHIN, T.J., VINCENT, C.E., BACON, J.C., DUMONT, E. AND TERENTJEVA, A.under review. Medium-term impacts of a segmented, shore-parallel breakwater system, Coastal Engineering. THIELER, E. R., MARTIN, D., AND ERGUL, A., 2003, The Digital Shoreline Analysis System, version 2.0: Shoreline change measurement software extension for ArcView. U.S. Geological Survey Open-File Report 03-076. THOMALLA, F. AND VINCENT, C.E., 2003. Beach response to shore-parallel breakwaters at Sea Palling, Norfolk, UK. Estuarine, Coastal and Shelf Science 56 (2003) 203–212. VINCENT, C. E., 1979. Longshore sand transport rates—a simple model for the East Anglian coastline. Coastal Engineering 3, 113–136.


A.3 Video based analysis of morphological changes

Video based analysis of morphological changes behind the Sea Palling breakwaters

Fairley I., Davidson, M.

Coastal Processes Research Group, University of Plymouth. Introduction This paper describes the results of investigations into the beach morphodynamics at Sea Palling based on video analysis from 6 Argus cameras (Holman and Stanley 2007). Two datasets have been collected: storm scale beach changes and mean sea level (MSL) shoreline changes over a 2.5yr period. Shorelines at known water levels were manually picked from the oblique images and rectification routines (Holland 1997) used to transform these to real world co-ordinates. Collation and interpolation of shorelines over a tidal cycle was used to obtain the intertidal morphology. Diminishing pixel resolution with distance meant analysis was confined to 6 embayments which are closest to the cameras.

Storm scale changes Storms with various wave and surge conditions have been analysed (Fairley in press). Obliquely incident storms were found to move the salients and tidal tombolos in a down drift direction. This change was more defined in phase I than in phase II. Accretion was observed in the intertidal region over storms although it is uncertain whether this was due to the structures trapping longshore sediment transport or erosion of material from the supra-tidal beach caused by larger wave set up and surge effects. Cross-shore beach profiles remained similar in form and gradient over storms for phase two whilst phase one showed a reduction in gradient. This difference is believed to be due to the narrower phase II beach. Storms associated with large surges exhibited greater flattening of the beach gradient.

Mean sea level contour dataset. A range of simple analysis were applied to the MSL contour dataset before EOF analysis was used to decompose the dataset into its constituent modes of change (Fairley in Prep.). Examination of the standard deviation of the dataset showed that the salients and tidal tombolos displayed greater variability than the bay centres. The position of the different bay centres through time is shown in Figure A3.1, it can be seen that for phase I the two bays show opposite underlying behaviour with bay B, the northern bay, gradually prograding and the bay C gradually receding. In phase II the three bays display similar behaviour. The bay shoreline positions were all significantly correlated with both wave height and wave steepness. For bay B larger waves lead to a progradation of the shoreline, whilst for the other bays shoreline recession occurs under larger steeper waves. In phase II all the bays respond in the same way to wave forcing and in a similar way to a natural beach would be expected to behave (erosion under larger waves). The difference between the phase I bays is due to bay B effectively trapping sediment supply from the north.

Correlation with estimated rates of longshore sediment transport (Q) showed that bay B was significantly correlated with a directional estimate of Q such that the shoreline prograded under transport from the north and receded under transport from the south. When sediment transport is from the south, available sediment will be trapped by other embayments and not affect bay B. The other bays were not significantly correlated with Q. It is thought that for bay C this is due to bay B trapping the majority of the


southward incident sediment. Phase two is less effective in trapping sediment and thus does not show significant correlation between sediment transport and shoreline change.

Figure A3.1 Bay shoreline positions for phase one (upper panel) and phase two (lower panel).

The MSL contour dataset was split spatially into phases I and II and EOF analysis applied. For phase I, the first two modes made up 75% of the variance in the dataset. The first mode describes the progradation and recession of the shoreline, which exhibits greater amplitude of change on the tombolo horns. The temporal signal of this mode is significantly correlated with the wave parameters, but more significantly correlated with the maximum tidal height and a combination of tidal height and wave height. There is no statistical difference in these two correlations. Large spring tides and surge events facilitate wave transmission into the embayments via overtopping and lesser dissipation over the bay floors. Waves reach the supra tidal beach mobilising sediment that is normally above the high tide line. Under surge events a sediment transport pathway over the breakwater 5 tombolo opens injecting additional sediment into the system. Under larger waves the amount of longshore sediment increases which means that sediment is more abundant and available to be trapped by the breakwaters.

The second mode for phase one describes the longshore displacement of the tidal tombolos, a positive temporal component moves the features in one direction and a negative temporal component moves the features in the opposite direction. This movement has been observed in the storm-scale analysis for oblique waves but the temporal component does not correlate with wave direction or Q. It is believed that this is due to the importance of the tombolo starting position to longshore movement under oblique waves.


In phase II the first three modes of change make up 75% of the dataset variability. The first mode is similar to the first mode for phase I, dealing with the progradation and recession of the shoreline. The change is in the same direction throughout phase II and does not have the same concentration on the salients. Like phase I this mode is correlated with the sum of wave height and daily maximum tidal elevation, for larger tides and waves the shoreline progrades, for smaller waves and tides the shoreline recedes. The second mode for phase two again relates to the longshore movement of the salient but rather than being a translation (like phase I) the second mode shows a beach rotation with erosion from one half of the embayment and accretion on the other. It cannot be said whether sediment is exchanged over the salient, across the bay or a combination of the two. Inverse correlations with wave height suggest that larger waves produce erosion on the southern half of the tombolo and accretion on the northern half, while smaller waves produce the reverse effect. The third mode describes the increase and decrease of bay curvature effected via the erosion of the salients and infilling of the bay centres or conversely, bay centre erosion and amplification of the salients. This mode of change is correlated with the wave period: higher than average wave periods lead to salient erosion and bay infilling. Previous work (Fairley 2007) has shown that longer period waves are more readily transmitted through the phase II breakwaters. Waves transmitted into the lee of the breakwaters could both erode the salients and retard the formation of circulation patterns that maintain the salient structure.

Conclusions The individual bays in phase I behave differently under the same wave forcing, this is largely due to differences in sediment supply due to the tidal tombolos blocking a considerable portion of the longshore sediment potential. Bay B is largely accreting whilst bay C is still eroding. It is thought that once bay B reaches a certain equilibrium level, bay C will start to accrete too as sediment passes through bay B. The phase II bays behave more similarly due to open exchange of sediment between embayments; they also behave similarly to expected behaviour for an unprotected beach. The EOF analysis shows both similarities and differences between the two phases. The first two modes of change for both phases are similar. The cross shore movement of the MSL contour is the most dominant change followed by the longshore movement of the salients and tidal tombolos. However, for phase two the change is much more spread over the entire embayment whilst for phase I the changes are focussed on the tidal tombolos. The amount of variance described by each mode varies between phases with much more variance attributed to the first mode for phase one.

References FAIRLEY, I., DAVIDSON, M., KINGSTON, K., (2007). Video monitoring of overtopping of detached breakwaters in a mesotidal environment. Coastal Structures '07, Venice. FAIRLEY, I., DAVIDSON, M., KINGSTON, K., (in press). A video based investigation into the morphological impacts of storms behind a series of detached breakwaters. International Conference of Coastal Engineering. Hamburg, ASCE. Poster Proceedings. FAIRLEY, I., DAVIDSON, M., KINGSTON, K., DOLPHIN, T., PHILLIPS, R. (in Prep.). "Empirical orthogonal function analysis of shoreline changes behind two different designs of detached breakwaters." submitted to Coastal Engineering. HOLLAND, K. T., HOLMAN, R.A., LIPPMAN, T.C., STANLEY, J., PLANT, N., (1997). "Practical use of video imagery in nearshore oceanographic field studies." IEEE Journal of Oceanic Engineering 22(1): 81-92. HOLMAN, R. A. AND J. STANLEY (2007). "The history and technical capabilities of Argus." Coastal Engineering 54(6-7): 477-491.


A.4 Wave, currents and sediment transport observed during the LEACOAST2 experiment

Wave, currents and sediment transport observed during the LEACOAST2

experiment

Judith Wolf, Peter Thorne, Paul Bell, Richard Cooke, Alejandro Souza POL, Joseph Proudman Building, Liverpool, L3 5DA. email: [email protected] ABSTRACT As part of the LEACOAST2 project, data were collected from 3 instrumented tripods deployed near the breakwaters, during 2 field campaigns (March-May 2006 and October 2006 – January 2007). The instrumentation included acoustic and optical instrumentation (ADCP, ADV, LISST and bedform scanners). A large dataset has been successfully obtained. Here we present an overview of the dataset, the observed hydrodynamic conditions, sediment transport and bedforms from this experiment.

Data collection Data were collected from 3 instrumented tripods deployed near the breakwaters (F1, F2, F3, see Figure A4.1), during 2 experiments: Experiment 1 (March-May 2006), Experiment 2.1 (October-December 2006) and 2.2 (December 2006 – January 2007). The instrumentation included acoustic and optical instrumentation (ADCP, ADV, LISST and bedform scanners). One tripod (F2) was deployed in the intertidal zone. The others were deployed in water depths of 6-8m (mean tidal range1 in the area is about 2m).

Figure A4.1 Experiment 1 and Experiment 2 showing locations of in-situ tripods (purple circles).

1 The mean tidal range is (Mean High Water – Mean Low Water). This is different from mean spring tidal range (Mean High Water Springs – Mean Low Water Springs) referenced in Section 2 of the main report. The mean spring tidal range in the area is about 3m.


The ADCP gave current profiles, surface elevation and wave data for Experiment 2 (it failed in Experiment 1). The ADV gave wave parameters and point measurements of 3-axis high frequency currents near the bed (at a single depth at F2 and F3 and 3 levels at F1). The ABS gave information on sediment concentration profiles near the bed. The mobility of the bed under combined wave and current conditions is illustrated by the variety of bedforms observed by the ripple scanners e.g. see Fig. A4.2.

Figure A4.2 Examples of bed ripples.

From observed and model winds for 2006-7, it can be seen that the strongest winds are from the west, north and north-east. Westerly winds are blowing offshore and in these conditions wave heights are negligible. Northerly and NE storms can generate high waves and storm surges. The maximum wave height is from just north of shore-normal, suggesting wave-induced long-shore currents will be to the south.

1st November storm During Expt. 2.1 the largest storm occurred with wave heights at F1 reaching 3m on 1 November as shown by the ADCP record (Fig. A4.3). A positive surge of 1.87m occurs on 31 October. It may also be seen that the surge current overwhelms the normal tidal reversal leading to persistent southerly flow for several tidal cycles.

Fig. A4.4 shows the ABS concentrations of suspended sediment at 0.075m above the bed level for F1 for Expt. 2.1. The concentrations are well-correlated with the wave height as illustrated in Fig.A4.3 but also exhibit tidal fluctuations during calm periods.

Summary Tides, waves and storm surges have been observed during the experimental periods in 2006-7. Tidal conditions can be seen to move the sediment but episodic storm events increase the sediment flux by several orders of magnitude. A summary of the dataset and data may be obtained via the British Oceanographic Data Centre () or the project web site (). Some results have been presented by Pan et al. (2007) and Wolf et al. (2008).

References PAN, S., WOLF, J., CHEN, Y., BELL, P. S., DU, Y., FERNANDO, P. AND LI, M. 2007 Modelling nearshore waves with presence of shore-parallel breakwaters Coastal Structures 2007, Australia. WOLF, J., SOUZA, A. J., BELL, P. S., THORNE, P.D., COOKE, R.D. AND PAN, S. 2008 Wave, currents and sediment transport observed during the LEACOAST2 experiment. PECS 2008, Liverpool.


Figure A4.3 ADCP record from F1 Expt 2.1.

Figure A4.4 ABS concentrations for F1 Expt. 2.1.


A.5 Marine radar monitoring

Remote bathymetry, bedforms & current mapping using marine radar during the LEACOAST2 experiment

Paul S. Bell

POL, Joseph Proudman Building, Liverpool, L3 5DA. email: [email protected] ABSTRACT During the LEACOAST2 project, a marine radar was deployed overlooking the shore parallel breakwaters at Sea Palling in East Anglia. This provided hourly records of sea surface conditions from March 2006 – July 2008. The radar data record the patterns of sea surface roughness over a range of 4km on which sea surface waves are visible as well as hard targets such as ships, breakwaters and jet skis. The wave patterns can be analysed to infer the underlying water depth map and current vectors that caused the observed wave behaviour.

Data collection The system deployed at Sea Palling Lifeboat Station consisted of a Kelvin Hughes 10kW X-band radar with a 2.4m antenna operating on short (60ns) pulse setting, see Figure A5.1.

Figure A5.1 The X-band radar mounted on the flat roof of the lifeboat station at Sea Palling.

This was coupled to a PC based digitization system (Figure A5.2) that recorded sequences of 256 images of the sea surface (Figure A5.3) with ~2.7 second intervals spanning about 12 minutes every hour. Summary images were uploaded to the POL website within 30 minutes of recording allowing the status of both the system and the sea state to be monitored. Other data used include the significant waveheight monitored using an Acoustic Doppler Current Profiler and tide gauge data available


from the National Tide and Sea Level Facility for the Cromer and Lowestoft gauges for estimating the surge component.

Figure A5.2 The recording system housed in a mobile rack inside the building.

Figure A5.3 A radar snapshot of the sea surface showing waves propagating through the breakwaters.

Analysis By mapping the wave properties across the radar images and fitting that behaviour to an equation governing wave propagation it is possible to infer the water depth (Figures A5.4 & A5.5) and current (Figure A5.6) causing the observed wave behaviour. Further, by using tide predictions together with a surge component derived from the NTSLF tide gauge data it is possible to reference these water depth maps to the local datum. The derived water depths are generally within 1m in elevation of comparable survey data.

A minimum significant waveheight of about 1m is necessary for waves to be visible on the radar, so mapping is limited to such wave events.


Figure A5.4 A short range high resolution radar derived bathymetric map of the breakwaters area.

Figure A5.5 A long range medium resolution radar derived bathymetric map extending almost 4km from the radar.

Figure A5.6 A radar derived water depth and current vector map.


A recent advance has been the development of a method to extract maps of the locations of submarine bedforms with wavelengths of 10s-100s of metres based on the modulation of sea surface roughness from the convergence and divergence of currents flowing over the bedforms. This method can be used to maps the location of large dune features to a higher horizontal resolution than the wave inversion technique described above. (Figure A5.7)

Figure A5.7 Radar derived map of submarine dune features.

Summary Marine X-band radar can be used to monitor remotely the spatial variations of waves, currents, bathymetry and bedforms over ranges of several km from a shore based station.

References BELL, P.S., WILLIAMS, J.J., CLARK, S., MORRIS, B.D. AND VILA-CONCEJO, A., 2006. Nested radar systems for remote coastal observations. Journal of Coastal Research, SI39 483-487. BELL, P.S., 2008, Mapping of bathymetry and tidal currents in the Dee Estuary using marine radar data. Proceedings of PECS 2008: Physics of Estuaries and Coastal Seas, 5 -29 August 2008, Liverpool (UK), pp 175–178.


A.6 Coastal Area Process Modelling - 1

A Short Report on LEACOAST2 Process Modelling

Shunqi Pan & Yanliang Du School of Engineering, University of Plymouth, Plymouth, PL4 8AA

Introduction The main objective of the process-based modelling work within the LEACOAST2 project was to implement and test the important coastal processes in the nearshore area which are interacting with the coastal defence structures, with an existing coastal model. The calibrated model was then applied to the Sea Palling site to study the effects of waves, tides and surge on the beach morphology with the presence of shore-parallel breakwaters, particularly, under storm conditions. The model results were used to identify the dominant processes in such an environment. The work was also extended to study the longer term effects by applying the model generically to the site with a series of scenarios.

A number of sub-set tests have been carried out to examine the impacts of tides, wave overtopping and storm surge. This report is to summarise the results of model tests regarding the impacts of storm surge on the nearshore sediment transport at Sea Palling.

Figure A6.1 Breakwater scheme at Sea Palling and computational domain.

Model Setup The process-based model was set at the Sea Palling site, covering an area of 5km in longshore direction and 1.8 km in the cross-shore direction. The model domain includes all 9 shore-parallel breakwaters with the offshore mean water depth of approximately 20 m, see Fig A6.1(a).

The computational domain was rotated, such that the offshore boundary was aligned at the left hand side and shoreline at the right as shown in Fig. A6.1(b). Wave and tide


conditions measured at F1 location, see Fig. 6.1(a), were modified and imposed at the offshore boundary (for waves) and lateral boundaries (up and down-drifts, for progressive tides and surge). The measurements in the Experiment 2.1, as shown in Fig. A6.2, which lasted approximately 670 hours from 22nd Oct to 22nd Nov 2006, were used in the model simulations.

Figure A6.2 The measured waves and tides at F1 location.

Preliminary Results Due to the complex coastal processes in the nearshore area of the study site, the model results were analyzed in different ways to identify the impacts of each dominant process, as previously mentioned. The results shown here were to quantify the impacts of tidal surge on the overall morphological change in the nearshore area, as well as that in each embayment. As shown in Fig.A6.1(b), the nearshore area was divided into a number of sub-areas, referred as “boxes” hereafter. In addition to each embayment between Reefs 5 & 8, the area for all 5 low-crested breakwaters (Reefs 9-13) was combined as “Bay Low”, see Fig. A6.1(b). The measured waves were also statistically sorted to two classes, namely: waves from up-drift to down-drift, referred as “+wd” in Fig. A6.1(b), which follows flood tidal flows, and wave in opposite direction, referred as “-wd” in Fig. 1(b), which follows the ebb tidal flows. Mean water levels were averaged over each tidal cycle (M2, 12.42 hours, or nearest model output), in order to measure the surge level individually.

Fig. A6.3 shows the time series of the computed volumetric changes for each embayment and the combined area which includes all embayments. It can be clearly seen that the morphological change in the nearshore area responded to the wave and tidal conditions at different degrees, in particular during the storm events. Considerable morphological changes occurred after the main storm event on 1st Nov 2006 in the entire area. In general, the storm events have profound effects on Bay 0 and Bay Low, but much less impacts on the other embayments, namely Bays A, B, C and D. The comparison of total volumetric changes for each embayment and their combination (denoted as “Total”) with the survey data is shown in Fig. A6.4.

Figure A6.3 Computed volumetric changes for each embayment and the nearshore area.


Figure A6.4 Computed & measured volumetric changes after 670hrs.

The results show that the computed total volumetric changes agree largely with the measurements. However, due to the complexity of the processes involved, there are clearly discrepancies between the computed and measured volumetric changes, particularly in Bays A and C, which show opposite trends, although the magnitudes of these volumetric changes are relatively small. The largest discrepancy appears to occur in Bay 0, the area just in the up-drift of the first breakwater.

Further examining the model results appears to suggest that morphological changes are not only affected by waves and tidal ranges, but also the surge levels, particularly, the acceleration of the surge, i.e., the rate and direction of the surge level changes. To this end, the mean water level for each tide was derived from the model results and the net volumetric changes in each box, as well as the entire nearshore area were computed. Fig. A6.5 shows the wave height and mean water level averaged over a tidal cycle, together with the time varying accumulative volumetric changes for the entire breakwater scheme area, in the top panel. Shown in the bottom panel of Fig. A6.5 are the mean tidal levels and net volumetric changes per tide, with the wave height and tidal level information in the background. The results show clear correlations between the computed net tidal volumetric changes and mean tidal level, i.e., the surge level. When the surge level increases, the volumetric change increases, vice-verse. The net volumetric changes are also strongly modulated by the wave height. The influence of the tidal range itself appears to be small. Similar results for Bay 0 and Bay Low, where the volumetric changes have been strongly affected during the storm conditions, see Fig. A6.3, are shown in Figs. A6.6 & A6.7.

Conclusions The process-based model has been applied to the LEACOAST2 study site at Sea Palling for a duration of 670 hours. Detailed analysis of volumetric changes was carried out. The results clearly show that the acceleration of surge plays an important and significant role on the morphological changes in the nearshore area. The most significantly affected areas are Bay 0 and Bay Low. Further analysis was also carried out, but not presented here, to quantify the volumetric changes at the site.


Figure A6.5 Computed volumetric changes/tide in relation to surge level (All bays).

Figure A6.6 Computed volumetric changes/tide in relation to surge level (Bay 0).


Figure A6.7 Computed volumetric changes/tide in relation to surge level (Bay Low).


A.7 Coastal Area Process Modelling - 2

Numerical study of ripple dynamics and its impacts on morphodynamics at Sea Palling

Ming Li, Nichols Spanakis, and Brian A. O’Connor

Department of Engineering, the University of Liverpool, Brownlow Street, Liverpool, L69 3GQ, [email protected]

Introduction As part of the LEACOAST2 project, a detailed study of the dynamics of seabed ripple and related sediment entrainment under combined waves and currents, and their impacts on the morphological changes around the breakwaters at Sea Palling was conducted at the University of Liverpool using a quasi-3D computer model system. Field data from the EPSRC-funded LEACOAST and LEACOAST2 projects were used to test the model predictions.

The numerical model The Q3D model system includes a wave module, a current module, a sand transport module and a morphodynamic module. The wave module is based on a hyperbolic type phase-resolving wave model that can simulate wave randomness, refraction, diffraction and reflection from structures in the coastal region. The computed wave information is then used in the current module to derive depth-averaged flow across the model area (Li et al 2007). To take into account wave-structure interactions, wave transmission and dissipation have been added into the existing model. The Wamsley and Ahrens (2003) method was chosen to estimate the transmission and dissipation parameters. At each grid point, a 1DV model is used in the sand transport module to describe the vertical profiles of both flow velocities and related sediment concentration distribution over the water column, which also provides the suspended load transport rate.

A detailed 3D computer model of Li et al (2006) was used to represent the boundary layer processes, including sediment transport above individual vortex ripple under combined wave and current action at an intra-wave-period time scale. The boundary layer model results were then integrated and parameterised into suitable boundary conditions to be used in the 1DV model, such as the bed roughness height and near-bed reference concentration. The enhanced drag force and sediment suspension due to the presence of ripples was also included in the morphological simulation. To investigate ripple evolution around the structures and their influences on the overall morphology, the bedform predictor of Soulsby and Whitehouse (2005), refer as SW05, was used to compute ripple characteristics over the whole study domain.

The model was applied to the offshore breakwater system at Sea Palling (Fig A7.1). The measured wave height was specified along the offshore boundary, together with the tidal water level. A radiation condition was used along both side boundaries. The computational grid of 25mx12.5m covers of 4km in the long-shore direction and 2km in the cross-shore direction. To simplify the computation, a single representative grain size of 0.3mm was used over the whole domain, although the model is capable of handling spatially-varying grain sizes.


Figure A7.1 Computational domain of the Sea Palling site.

Preliminary results A storm event, E27, was used to test the model. The significant wave height varies from about 0.4m to 2.7m during the storm at an angle of 20° relative to the normal direction to the shore. The tidal range was 3.0m as shown in Fig A7.2.

Figure A7.2 Tidal water level and significant wave height used in E27 storm simulation.

Fig A7.3 presents the computed ripple height distribution at 15hrs and 22hrs. The corresponding transport vectors are also shown in the figure.

Due to the asymmetry of the tidal current, the transport rates at these two phases are not symmetrical. However, it is apparent that the ripples are widely spread within the embayment and seawards of the structures. The ripple height correlates closely to the transport rate, i.e. higher ripples exist in each embayment where the transport rate is strong and remains small immediately behind the structure where the transport rate is small. Comparison against the measured ripple lengths were also carried out for the area between Reef 6 and Reef 7 in Fig A7.4. Overall, the computed ripple lengths have a similar pattern to the measurements, i.e. higher values at one side behind the breakwater and lower values at the other side, which indicates the strong influences of wave and tide-induced currents in this region. However, the predicted ripple length tends to be lower than the measured one. It may be due to the spatially varying grain size that is not included in the model. Meanwhile, the SW05 method primarily concerns ripples only. The large lengths in the measurements may suggest that the bed form is


in transition from small scale ripples to dunes in this region, which is not included in the SW05 method.

Figure A7.3 Computed ripple height and transport vector distribution at 15hrs (A) and 22hrs (B) around Reef 6 and Reef 7.

Figure A7.4 Comparison of measured (A) and computed (B) ripple length around Reef 6 and 7.

The computed ripple characteristics were then imported into the Li et al (2006) boundary layer model to obtain the sediment entrainment and bottom drag forces that is needed in the 1DV transport module. Due to the unknown ripple shape, a parabolic profile was used based on a given ripple height and length. In total, around 20 representative ripples were used across the area. The computed bed roughness and near-bed sediment concentration values were then formed into a lookup table and interpolated at each model grid point.

In order to identify the possible impacts from the ripples on morphological changes, the model was then run with the rippled bed condition and with a plane bed. In the rippled bed case, all parameters computed from the boundary layer model of Li et al (2006) were used in the sediment transport module. In the plane bed case, only the equivalent bed roughness height due to ripples was used and the near bed sediment entrainment was computed as if the bed was free from ripples. Such an approach is to minimise the uncertainties involved and focus on the ripple-induced mixing and entrainment from the bed surface. The resultant bed level change after one tidal cycle using the rippled bed together with the differences between the two cases are shown in Fig A7.5. The overall erosion and accretion are approximately in the range of -0.7m to 0.4m in the rippled bed results. It is also obvious that most bed level change takes place in the embayment

A B

A B


behind the breakwaters. The differences between the rippled bed and plane bed are in the range between -0.1m to 0.1m, this is within 20% of the total bed evolution values. In a similar way to the total bed level change figure, the most noticeable difference between the two model results lies in the embayment behind the breakwaters. Outside the breakwaters, the difference is very small.

Figure A7.5 Computed bed level change (BLC) after 1 tidal cycle based on ripple bed in A and the differences between the model predictions using rippled bed and plane bed in B.

Conclusions Initial model results based on the ripple predictor of SW05 seem to suggest a wide spread of ripples around these structures. Their persistent existence affects local flow and wave propagation and hence the sediment transport. Comparing with limited measurements, the ripple length seems to be under-predicted which is possibly due to the fact that the transition from ripple to larger scale bed form is not considered in SW05. Differences in the total transport rate for a non-ripple condition and a rippled condition seems to be small provided a proper bed roughness height is used around the structures. Most differences were found within the embayment behind the structure where the water depth and wave heights are small but the current is strong. However, due to their existences, the long term accumulation effects may be considerable in terms of total morphological variations throughout the region. This work is ongoing to study long term effects from these offshore breakwaters on the morphological changes.

Acknowledgement The current study is partially sponsored by EPSRC LEACOAST, LEACOAST2 and SANTOSS projects. Professor Chris Vincent from UEA, Peter Thorn and Dr Judith Wolf from Proudman Oceanographic Laboratory provided helpful comments.

References LI, M., PAN, S. AND O'CONNOR, B. A. (2006) Modelling coastal boundary layer flows over typical bed-forms, Maritime Engineering, Proceedings of the Institution of Civil Engineers (MA1), 9-24. LI, M., FERNANDO, P. T., PAN, S., O’CONNOR, B. A., CHEN, D. (2007) Development of a Quasi-3D numerical model for sediment transport in the coastal region, Journal of Hydro-environment Research, 1(2), pp 143-156. SOULSBY R. AND WHITEHOUSE, R. (2005) Prediction of ripple properties in shelf seas, Mark 2 Predictor for time evolution, TR154, HR Wallingford. WAMSLEY, V. AND AHRENS, J. (2003) Computation of wave transmission coefficients at detached breakwaters for shoreline response modelling, Coastal Structures 2003, ASCE.

A B


A.8 Probabilistic Modelling

Probabilistic simulation of beach morphodynamics within a flood defence scheme

Dominic Reeve & Baoxing Wang

University of Plymouth, Centre for Coastal Dynamics and Engineering, School of Engineering, Reynolds Building, Drake Circus, Plymouth, Devon, PL4 8AA,

UK.

Introduction This note describes the development and validation of a probabilistic model simulating the changes in a large flood defence scheme. The scheme consists of a series of detached breakwaters together with beach nourishment, is located near the village of Sea Palling in Norfolk, UK and covers a length of ~4km. The aim of the LEACOAST2 project was to investigate the larger scale morphodynamic impacts of nine segmented shore-parallel breakwaters at Sea Palling. The project includes monitoring of beach surveys, water levels and wave conditions; as well as short-term detailed process modelling to predict storm response and probabilistic morphological modelling to investigate the behaviour of the whole scheme over the period of decades.

Field Site Figure A8.1 shows a location map for the site which is on the east coast of the UK.

Figure A8.1 Location map (adapted from GoogleEarth). White triangle shows the location of the study site

Figure A8.2 The Sea Palling scheme, looking north-northwest. (Courtesy of Mike Page)


The Sea Palling defence scheme is shown in Figure A8.2, which is an aerial photograph looking north-northwest. Phase 1 involved the construction of four reefs (numbers 5-8) along the most vulnerable length of coastline, north-west of Sea Palling. The breakwaters were constructed between 1993 and 1995, and Phase 1 consisted of the 4 breakwaters towards the top of Figure A8.2. The Phase 2 works comprised:

• Five more reefs (numbers 9-13) to the south-east of the existing reef system;

• A beach recharge campaign (approx. 1 million cubic metres) behind and between the Phase 1 and 2 reefs and to the south to mitigate the predicted downdrift beach losses;

and were completed in July 1997. A key part of the scheme was an ongoing monitoring programme consisting of aerial photography and in situ surveys. The project study area covers a 6km-long stretch of coastline, consisting of four sub-regions: i) a ~1km up-drift zone to the first breakwater, ii) the 4 breakwaters of Phase 1, iii) the 5 lower breakwaters of Phase 2, and iv) a down-drift zone of ~ 1km.

Model Development The starting point is the one-line model, originally developed by Pelnard-Considère (1956). Simplified versions of this model, which do not account for spatial variation in wave conditions, have provided analytical solutions see eg. Wind (1990), Larson et al (1997), Dean & Dalrymple (2002) and Reeve (2006). Thus, effort has been directed at developing methods that are suitable for solving the one-line beach model numerically. The one line model takes the form of three simultaneous equations governing the conservation of sediment, a longshore sediment transport formula and the geometric relationship between the incoming breaking waves, the baseline and the local alignment of the shoreline. This type of model has been adapted for use in probabilistic simulations (eg. Vrijling & Meijer 1992, Dong & Chen 1999, Pedrozo-Acuna et al 2007, Reeve et al 2009).

We use the sediment transport equation proposed by Hanson et al (2006) which includes the effects of tidal currents. Further, given the dependence of longshore transport on wave angle and the complex wave transformations that are induced by the structures, we have integrated a sophisticated wave model into the 1-line approach.

Numerical solution of the 1-line equations Both explicit and implicit time-stepping finite difference schemes are well-known techniques for the numerical solution of the ‘one-line’ model for shoreline change (see eg. Gravens et al., 1991). Here, a technique known as the Method of Lines, (eg Schiesser 1991), is used to solve the ‘one-line’ model, (see Reeve et al 2009 for further details). The advantages of using this method include: 1) unconditionally stable solutions, 2) accuracy that does not depend on a user-specified computation time-step, and 3) simple and flexible coding.

Nearshore wave transformation To derive the wave conditions at the shoreline, an elliptic wave model (Li 1994) has been dynamically linked with the one-line model. Thus, deep water wave conditions are used as the offshore boundary condition to the wave model, which transforms the wave height, period and direction to gridpoints along the shoreline. These transformed wave conditions are then used as input to the one-line beach model. The one-line model predicts changes in the beach contour configuration which are then fed back into the bathymetry for the wave model before the next wave condition is transformed inshore. Overtopping of the tombolos is estimated using the fomulae given in Eurotop (2007).


Inputs Tide gauge measurements from the field campaign were analysed to determine tidal harmonics. These were used to predict the tide levels over the simulation period. The MSL (+0.24m ODN) line is taken as a representative shoreline. The model was driven by hindcast deepwater wave conditions obtained from the UK Met. Office European wave model, covering a period of almost 13 years, (1995 to 2007), at an interval of 3 hours. A method described by Cai (2005) and Cai et al (2007) was employed to create multiple realisations of 13-year time series with statistical properties (ie. marginal distributions and temporal auto-correlation and cross-correlation behaviour), that mirror those in the original 13-year time series of hindcast waves.

Results Calibration of the model at a regional and local level is reported elsewhere (see Reeve et al 2009). However, computed longshore transport rates were in good agreement with those reported previously by Vincent (1979) and Damgaard et al (1999).

Figure A8.3 Marginal distributions of wave height, period and direction of the original (lefthand column) and simulated (righthand column) time series.

A comparison of the marginal distributions of wave height, period and direction obtained from the original hindcast data and one of the simulated 13-year sequences is shown in Figure A8.3. Agreement is very good. As well as recreating the correct marginal distributions it is also important to simulate the correct second-order statistics such as auto-correlation and cross-correlation. Comparisons between the original and simulated data are shown in Figure A8.4, which show a very good reconstruction of the statistical behaviour of the original series.


Figure A8.4 Cross-correlation functions of wave parameters for original (lefthand column) and simulated (righthand column) series.

Running the beach model repeatedly with each of the different 13-year realisations provides alternative, (but statistically valid), sequences of beach response. These results are used firstly, to validate the Monte Carlo predictions of the positions of the bays and salients against beach surveys taken during the course of the project (covering a period of ~ 6 years), and then to investigate the statistics of predicted beach position throughout the scheme. Two hundred realisations (each containing a 13 year time series of wave sequences at 3 hourly intervals – an equivalent of 2600 years in total) were generated and then used to drive the one-line model to create 200 outcomes of shoreline evolution. From this ensemble of outcomes it is possible to calculate ensemble average statistics of the beach position as a function of time. The envelope of shoreline movement is shown in Figure A8.5. The maximum deviation appears in Bay 1, while significantly lower deviations present in other bays. The plot gives a measure of the variation in beach position throughout the area. The width of the zone varies considerably along the coastline behind the breakwater system. Again, the widest movements are found in Bay 1, the coastline position lies between 330 and 380m, while other bays show a relatively narrow band of change. In general, the shoreline position varies within a range of about 40m. However, the longshore positions of the salients and tombolos in Phase One of the scheme appear very stable, whereas the smaller salients in Phase Two show some propensity to move.

Figure A8.5 Plot of the mean shoreline position (time and ensemble averaged), together with the envelope of minimum and maximum excursions throughout the scheme.


Discussion and Conclusions The development and validation of a probabilistic model simulating the changes in a large flood defence scheme consisting of offshore breakwaters in a strongly tidal environment has been described.

The simulation system consists of an elliptic wave transformation model that treats wave refraction, diffraction, breaking and reflection together with a dynamically coupled one-line model. The model has been successfully calibrated at a regional and local level against published results and survey data gathered during the field campaign. Input to the model has been derived from a 13-year sequence of hindcast deepwater wave conditions. Realisations with statistically similar first and second-order statistics have been generated from this sequence using a new, recently published, technique. The modelling system runs sufficiently fast to enable a meaningful number of realisations to be computed in a reasonable time so that reliable statistics of shoreline position can be obtained. The probabilistic simulations provide a means of estimating the mean shoreline and uncertainty bounds, which are a valuable input to beach management and strategic coastal planning.

Although the model shows some promising performance, the one-line beach level fluctuations due to the cross-shore sediment transport are ignored on the grounds that their long-term average effect on the mean shoreline position is small. The impact of this assumption requires further investigation.

References CAI, Y., 2005. A forecasting procedure for nonlinear autoregressive time series models. The Journal of Forecasting, 24, 335-351. CAI, Y., GOULDBY, B., DUNNING, P. & HAWKES, P., 2007. A simulation method for flood risk variables. The 2nd IMA international conference on flood risk assessment, Plymouth, UK. DAMGAARD, J.S., CHESHER, T.J. & HALL, L.J.,1999. Simulation of the sediment budget for the Happisburgh to Winterton reefs- PISCES application study. Technical report, H.R. Wallingford. DEAN, R. G. AND DALRYMPLE, R. A., 2002. Coastal processes: with engineering applications, Cambridge University Press, Cambridge. DONG, P. AND CHEN, H., 1999. Probabilistic predictions of time-dependent long-term beach erosion risks, Coastal Engineering 36, pp. 243–261. EUROTOP, 2007. Wave Overtopping of Sea Defences and Related Structures: Assessment Manual FLEMING, C. AND HAMER, B., 2000. Successful implementation of an offshore reef scheme. In 27th Coastal Engineering, Sydney. 1813-1820. GRAVENS, M.B., KRAUS, N.C., HANSON, H., 1991. GENESIS: generalized model for simulating shoreline change. Report 2, Workbook and System User’s Manual. U.S. Army Corps of Engineers. HANSON, H. AND KRAUS, N. C. 1989. GENESIS - generalized model for simulating shoreline change. Technical Report CERC-89-19, US Army Engineer Waterways Experiment Station, Coastal Engineering Research Centre. HANSON, H., LARSON, M., KRAUS, N.C. & GRAVENS, M.B. 2006 Shoreline response to detached breakwaters and tidal current: comparison of numerical and physical models. Proceeding of the 30th International Conference of Coastal Engineering. LARSON, M., HANSON, H., KRAUS, N. C., 1997. “Analytical solutions of one-line model for shoreline change near coastal structures.” Journal of Waterway, Port, Coastal, and Ocean Engineering, 123(4), 180-191. LE MEHAUTE, B., WANG, J.D. AND LU, C-C., (1983) Wave data discretization for shore line processes, J. of Waterway, Port, Coastal and Ocean Engineering, ASCE, Vol. 109, No.1, pp. 63-78 Li, B., 1994. An Evolution Equation for Water Waves. Coastal Engineering 23: 227-242.


PAN, S., FERNANDO, P., LI, M., ZHU, Y., O’CONNOR, B., VINCENT, C., TAYLOR, J., DOLPHIN, T. AND BACON, J. 2005. Effect of shore parallel breakwaters on coastal morphology under storm conditions. Coastlines, Structures and Breakwaters 2005, London, UK, Institute of Civil Engineers, London. PEDROZO-ACUNA, A., REEVE, D. E. & SPIVACK, M., 2007. ‘Beach variability near groynes’. Proc. ICCE 2006, San Diego, World Scientific, p3708-3718. PELNARD-CONSIDÈRE, R., 1956. Essai de theorie de l'evolution des forms de rivages en plage de sable et de galets, Fourth Journee de l'Hydralique, les energies de la Mer, Question III, Rapport No. 1, pp. 289–298. REEVE, D. E., 2006. “Explicit expression for beach response to non-stationary forcing near a groyne.” Journal of Waterway, Port, Coastal and Ocean Engineering, 132, 125-132. REEVE, D. E., WANG, B., TOMAS, L. AND ZACHARIOUDAKI, A., 2009. ‘Probabilistic simulation of long-term beach changes within a flood defence scheme’, in Proceedings ICE Conf. Coasts, Marine Structures and Breakwaters, Edinburgh, 2009. SCHIESSER, W.E., 1991. The numerical method of lines: Integration of partial differential equations, Academic Press, Inc, 326 pp. VINCENT, C. 1979. Longshore sand transport rates – a simple model for the East Anglia coastline. Coastal Engineering (3): 113-136. VRIJLING, J.K. AND MEIJER, G.J. 1992. Probabilistic coastline position computations, Coastal Engineering 17, pp. 1–23. WIND, H.G., 1990. “Influence functions.” Proc., 21st International Conference on Coastal Engineering, Costal del Sol-Malaga, Spain, 3281-3294


B Results of generic test cases using PISCES


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0

0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0distance along centreline of 2nd bay (m)

bed

leve

l (m

)

initial15 days30 days45 days60 days

-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0

0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0distance along centreline of 2nd breakwater (m)

bed

leve

l (m

)


Figure B.1 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 01 – no tide, shore-normal waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.2 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 02 – no tide, oblique incident waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.3 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 03 – 3m progressive tide, shore-normal waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.4 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 04 – 3m progressive tide oblique incident waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.5 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 05 – 3m standing tide, shore-normal waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.6 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 06 – 3m standing tide, oblique incident waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)




-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.9 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Note: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 09 – 5m standing tide, shore-normal waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.10 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 10 – 5m standing tide, oblique waves, Layout 1.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.11 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 11 – no tide, shore-normal waves, Layout 2.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.12 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 12 – no tide, oblique incident waves, Layout 2.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)




-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.16 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 16 – 3m standing tide, oblique incident waves, Layout 2.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)




-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


Figure B.20 Initial bathymetry and resulting bathymetry after 15, 30 and 60 days of simulation. Notes: Cross-shore profiles along centrelines of 2nd breakwater and 2nd breakwater bay. Simulation 20 – 5m standing tide, oblique waves, Layout 2.


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)


-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

4.0


bed

leve

l (m

)




C Results of generic test cases using MIKE 21 CAMS


Figure C.1 Simulated bathymetry contours after 60-day morphological simulation (bottom: profiles at 0, 7, 14, 21 and 28 days across A and B). Notes: Run A: Hm0=1m, Tp=5s, θ=90deg (normal to shoreline), no tides, Layout L1, breakwater crest, hcr > 2m.



Figure C.2 Simulated bathymetry contours after 60-day morphological simulation (bottom: profiles at 0, 7, 14, 21 and 28 days across A and B). Notes: Run B: Hm0=1m, Tp=5s, θ=90deg (normal to shoreline), no tides, Layout L2, breakwater crest, hcr > 2m.



Figure C.3 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 01: Hm0=2m, Tp=8s, θ=90deg, no tides, Layout L1, breakwater crest, hcr = 2m.



Figure C.4 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 02: Hm0=2m, Tp=8s, θ=45deg (oblique waves), no tides, Layout L1, breakwater crest, hcr = 2m.



Figure C.5 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 05: Hm0=2m, Tp=8s, θ=90deg, Rtide=3m, standing tides, Layout L1, breakwater crest, hcr = 2m.



Figure C.6 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 07: Hm0=2m, Tp=8s, θ=90deg, Rtide=5m, progressive tides, Layout L1, breakwater crest, hcr = 2m.



Figure C.7 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 07B: Hm0=2m, Tp=8s, θ=90deg, Rtide=5m, progressive tides, Layout L1, breakwater crest, hcr = 1m.



Figure C.8 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 07C: Hm0=2m, Tp=8s, θ=90deg, Rtide=5m, progressive tides, Layout L1, breakwater crest, hcr = 3m.



Figure C.9 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 08: Hm0=2m, Tp=8s, θ=45deg, Rtide=5m, progressive tides, Layout L1, breakwater crest, hcr = 2m.



Figure C.14 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 19B: Hm0=2m, Tp=8s, θ=90deg, Rtide=5m, standing tides, Layout L2, breakwater crest, hcr = 1m.



Figure C.15 Simulated bathymetry contours (top) after 60-day morphological simulation (bottom: profiles at 0, 15, 30, 45 and 60 days across A and B). Notes: Run 19D: Hm0=2m, Tp=8s, θ=90deg, Rtide=5m, standing tides, Layout L2, breakwater crest, hcr = 0m.

modelling the effect of nearshore detached breakwaters

Documents