Geomorphology 199 (2013) 8–21

Contents lists available at ScienceDirect

Geomorphology

j ourna l homepage: www.e lsev ie r .com/ locate /geomorph

Storms, shoreface morphodynamics, sand supply, and the accretion and erosion ofcoastal dune barriers in the southern North Sea

Edward J. AnthonyAix Marseille Univ, Institut Universitaire de France, CEREGE, UMR CNRS 7330, Europôle Méditerranéen de l'Arbois, B.P. 80, 13545 Aix en Provence, France

E-mail address: anthony@cerege.fr.

0169-555X/$ – see front matter © 2012 Elsevier B.V. Aldoi:10.1016/j.geomorph.2012.06.007

a b s t r a c t

a r t i c l e i n f o

Article history:Received 9 November 2011Received in revised form 17 May 2012Accepted 6 June 2012Available online 16 June 2012

Keywords:StormsShorefaceTidal banksSand supplyCoastal dunesNorth Sea

The coast of the southern North Sea is bound by dune barriers that have developed adjacent to a shallowstorm- and tide-dominated shoreface comprising numerous shore-parallel to sub-shore-parallel tidal sandbanks. The banks evolve under the joint control of tide-, wave- and wind-induced shore-parallel currents,which tend to ‘stretch’ them, eventually leading to bank division, and to shoaling and breaking stormwaves, which tend to drive them ashore. The banks, thus, modulate the delivery of storm wave energy tothe coast, redirect currents alongshore and are the sand sources for the accretion of coastal dunes. Foreduneaccretion occurs where major sand banks have migrated shoreward over the last centuries to be finally driv-en ashore and weld under the impact of storm waves. Morphological changes in the bank field can impact onshoreline stability through dissipation or enhanced shoreward transmission of storm wave energy and effectson radiation stress, particularly when waves are breaking over the banks. Where banks are close to the shore,mitigation of offshore sediment transport, especially during storms, can occur because of gradients in radia-tion stress generated by the complex 3D bank structure. These macro-scale mechanisms involve embeddedmeso-scale interactions that revolve around the mobility of sand waves, mobility of beach bars and troughsand foredune mobility, and micro-scale processes of bedform mobility in the subaqueous and intertidal do-mains, and of swash and aeolian beach–dune sand transport. These embedded interactions and the mor-phodynamic feedback loops illustrate the importance of synchroneity of sand transport from shoreface todune on this coast.Large stretches of the foredunes show either signs of stability, or mild but chronic erosion. Furthermore, ademonstrated lack of a clear relationship occurs between storminess and coastal response over the secondhalf of the 20th century. The present situation may be indicative of conditions of rather limited sand supplyfrom offshore, notwithstanding the abundance of sand on the nearby shallow shoreface, except in areaswhere a nearshore storm-driven tidal sand bank has become shore-attached. Apart from the important influ-ence of shoreface sand banks and of wave–bank interactions, foredune accretion and erosion also depend onvarious context controls that include individual storm characteristics, wind speed and incidence relative tothe shore, tidal stage during storms, and direct human intervention on the shore through foredune andbeach management. The bewildering variability inherent in these intricately related parameters may also ex-plain the poor relationship between storminess and barrier shoreline change and will still continue to renderunpredictable the response of shores to individual storms.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Storm effects on coastal sand barriers have received wide attention,in part because of increasing human occupation of storm-exposedcoasts. Much of the literature has focussed on hurricane impacts onthe North American coast (e.g., Morton, 2002; Stone et al., 2004;Zhang et al., 2005; Sallenger et al., 2006; Wang et al., 2006; Claudino-Sales et al., 2008; Houser et al., 2008), especially on the low-lying Atlan-tic seaboard where the transgressive history of much of the coast ren-ders storm impacts pervasive (Forbes et al., 2004). In Europe, efforts

l rights reserved.

have focussed on the various modes of coastal response to storms in acontext of marked morphological variability, and have included workby Cooper et al. (2004), Regnauld et al. (2004), Ferreira (2006),Mendoza and Jiménez (2006), Ciavola et al. (2007), Chaverot et al.(2008), Sabatier et al. (2009), Gervais et al. (2012), Haerens et al.(2012), and Suanez et al. (2012). These efforts clearly show that theconceptual framework and capacity for accurate prediction of storm-induced coastal change remain incomplete (Forbes et al., 2004). Theprecise response of a coastal stretch to each stormevent depends on nu-merous inter-related parameters.Whereas coastal erosion is a commonbut not exclusive result of storms, interactions between storms and theshoreline are complex. This complexity is inherent to the functional

9E.J. Anthony / Geomorphology 199 (2013) 8–21

mechanism of storms, which includes various parameters such as airpressure, mean water level, wind speed, wind direction relative to thecoast, and waves (e.g., Betts et al., 2004; Backstrom et al., 2008, 2009;Haerens et al., 2012). Coastal response to storms is also complex, be-cause it is controlled by multiple factors related to incident storm char-acteristics and to coastal and shoreface morphology and sedimentsupply. These latter factors integrate a longer-term perspective onstorm impacts on the coast.

Of course, a commonly strong link exists between the beach andthe shoreface in terms of sediment exchanges induced by storms, al-though Houser (2009) has recently pointed out the insufficiencies inrecognizing this synchroneity. Although coastal erosion is a com-mon outcome of storms, storms may also rework sediment ontothe beach, thus, generating accretion. For example, on the Atlanticcoast of the US, northeast storms produce downwelling that resultsin offshore-directed sediment transport, whereas upwelling createdby southwest storms results in onshore-directed sediment transport(Wright et al., 1994; Hill et al., 2004). Stone et al. (2004) alsoshowed that barrier islands can conserve mass during catastrophiccyclones, and that less severe cyclones and tropical storms can pro-mote rapid dune aggradation and can contribute sediment to theentire barrier system.

Whereas the general link between storms and the response ofsand barrier shorelines has been recognised in the literature, therole of shoreface modulation of storm impacts on the shore has re-ceived much less attention. Among the few studies having shownthat the morphology of the shoreface can strongly influence coastalresponse to storms and sediment dispersal on the shoreface arethose of Field and Roy (1984), Hequette et al. (2001), andBackstrom et al. (2008, 2009). In this paper, the morphological re-sponse of a dune barrier coast in a storm-dominated macrotidal set-ting on the southern coast of the North Sea is examined withreference to storm impacts at long (order of decades) to short(order of years) timescales. The morphodynamic connections be-tween storms and barrier shoreline response and the embeddedmacro-, meso- and micro-scale levels of interaction are reviewed.The essential role of storms in patterns of shoreline change comes

Line bankWissant Bay

Fig. 1. The dune barriers and shallow shoreface of the southern N

out as being strongly influenced by the conditions prevailing on theinner shoreface.

2. The southern North Sea barrier and shoreface system

2.1. Barrier and shoreface morphology

The North Sea coast of France from Cape Gris Nez to Belgium(Fig. 1) comprises two sand barriers, respectively, in Wissant Baybetween Capes Gris Nez and Blanc Nez, and from Cape Blanc Nezalong the southern North Sea as far as the Netherlands (Fig. 1). Eachbarrier consists of two to three generations of sub-shore-paralleldunes 100 to 600 m wide and with a maximum inland height of25 m, impounding former tidal embayments. The barrier foredune isassociated with beaches exhibiting multiple bar-trough (ridge-and-runnel) couplets that are widely exposed at low tide. The WissantBay barrier comprises variably eroding or accreting foredune sectors.The barrier stretching from Cape Blanc Nez to the Netherlands boundsthe empoldered North Sea coastal plain, large parts of which lie belowsea level, and face risks of flooding from storms. Large stretches of thislatter barrier have been massively transformed or obliterated byurban and port development. The foredunes in both barrier sectors ex-hibit blowouts and deflation corridors.

The gently sloping shallow shoreface, extending seaward of thebeach bars and troughs, is characterised by an important field ofprominent tidal sand banks and ridges (Fig. 1). The bank field offthe coasts of France and Belgium, which forms the Flemish Bankscomplex, is particularly well developed as the narrow Dover Straitopens up on the epicontinental southern North Sea. These banksare several kilometres long and have heights of up to 10 m. The crestsof sand banks closest to the shoreline lie at depths of less than 5 mbelow the mean low water line (Fig. 1). They practically impingeon the beach in places. These elongated sand bodies are commonlyoriented WSW–ENE, roughly parallel to sub-parallel to the coastline.Sediment distribution in the bank field is strongly related to the ba-thymetry, with fine to coarse sand in the interbank troughs, whereas

Cliffs

Westhinder

Trapegeer

A2-B

orth Sea. Bathymetric contours are from Augris et al. (1995).

10 E.J. Anthony / Geomorphology 199 (2013) 8–21

the shallower sectors of the banks and inner shoreface generally con-sist of fine to medium sand (Corbau et al., 1999; Van Lancker, 1999).

2.2. Winds, waves and tides

The southern North Sea coast of France may be described as typicalmixed storm-wave- and tide-dominated (Anthony and Orford, 2002),subject to a complex pattern of time-varying influences of tides andstorm waves, in addition to wind-forced flows. Winds mainly comefrom the southwest and northeast, but the strongest winds mostly origi-nate fromwest to southwest (Fig. 2). The overall aeolian dynamic regimeof this North Sea coast operates within the framework of a relatively lowfrequency of strong (>12m/s) onshore/offshorewinds (1.2–3.8%), and ofa relatively balanced wind regime, in which the proportion of ‘counter-active’ offshore winds in the significant wind category (>8m/s) attains41% and 34% in Calais and Dunkerque (Dunkirk), respectively. Overall,combined offshore and shore-parallel winds account for 71% and 57% ofwinds >8m/s, respectively, in these two sites.

The hydrodynamic context is that of a short-fetch, storm–waveenvironment, characterised by marked short-term (order of days toweeks) fluctuations in wave height. The dominant waves are fromsouthwest to west, originating from the English Channel, followedby waves from the northeast to north, generated in the North Sea.Long-term (Jan. 1977–Dec. 2002) records of significant wave heights,obtained fromwaverider bouys, are shown (Fig. 3a) for two locations,Westhinder and A2-B, respectively, about 35 km and 5 km offshore ofthe neigbouring coast of Belgium (see Fig. 1 for locations). Maximumsignificant wave heights are highest in autumn and winter but meanand minimum significant wave heights show smoother trends. Waveheights are lower at A2-B closer to the shoreline because of refractionand shoaling over the sand banks. These differences are furtherhighlighted by a recent record (10 March to 09 April 2012) of meansignificant wave heights at Westhinder and at Trapegeer (Fig. 3b), an-other station located close to the coast and nearer to the study area(Fig. 1). This graph also shows the short wave periods and the sharpshort-term fluctuations in significant wave heights typical of thisstorm–wave environment. Records from simultaneous deploymentsof pressure sensors on the inner shoreface and intertidal beach justeast of Dunkerque, compared to Westhinder, further highlight thesignificant drop in wave heights caused by the sand bank field(Fig. 3c). In conjunction with the nearshore sand bank morphology,tidal modulation may result in a highly variable wave field close tothe beach (Héquette et al., 2009). Breaking waves are essentiallyfrom a north-northeast to northwest window, although the dominantdeepwater directions are from north and west.

The tidal regime in the region is semi-diurnal and macrotidal, thetidal range decreasing from 6.4 m in Calais to 5.6 m in Dunkerque dur-ing spring tides (Service Hydrographique et Océanographique de laMarine, 1968). Because of the large tidal range, tidal currents are strong

Fig. 2. Wind roses for Ca

on the shoreface (Fig. 4) and over the intertidal beach (Fig. 5). In calmweather, current directions are closely conditioned by the tide, withdominantly longshore east-northeast flood directions and west-southwest ebb directions. Current reversals do not occur at high orlow tide, but are typically retarded by up to 2 h in Dunkerque and 2 to3 h in Calais because of the coexistence of stationary and progressivetidal waves. Because of the high velocity magnitude, tidal currents areonly slightly modified in the shallow water column of the shorefacewhere the directions remain constant from the surface to the seabed, al-though speeds decrease bottomward. Current speeds generally dimin-ish eastward from Calais towards Belgium from decreasing tidalamplitude in the same direction. Note, in Figs. 4 and 5, the influenceof stormwind conditions on currents andwaves. Strongwinds enhancethe strength of ebb or flood currents when blowing in the same direc-tion, or limit, and even prevent tidal reversalwhen blowing in the oppo-site direction (Reichmüth and Anthony, 2007; Héquette et al., 2008a,2008b), but flow is more commonly flood-dominated. During condi-tions of significant wind stress (sustained wind speeds >10 m s−1),the peak current speeds can be two to three times higher than ‘normal’(tide-generated) peak spring tide speeds. Longshore currents can be-come particularly strong during storms because of direct wind stressand gradients in radiation stress that divert, alongshore, offshoremean currents generated bywave breaking and onshoremass transportas waves pass over or break over the sand banks. Stormsmay add up to1 m of surge above high-tide swash excursion levels.

3. Storms, shoreface morphodynamics and the shoreline

3.1. Storms and shoreface morphodynamics: the big picture

The abundant accumulation offine tomediumsandon the shorefaceof the southern North Sea is considered as the product of large-scalewind- and tide-dominated hydrodynamic circulations in the easternEnglish Channel that have gradually sorted out, in the course of the Ho-locene, heterogeneous sea-bed sediments that accumulated under pastchanging sea-level and land drainage conditions (Anthony, 2000). Thesouthern extremity of theNorth Sea is a net depocentre of sand circulat-ing along coastal sediment pathways linking the eastern English Chan-nel to the southernNorth Sea (Beck et al., 1991; Grochowski et al., 1993;Anthony, 2000, 2002). In the course of theHolocene, this incoming sandhas been reworked by the interplay of tidal currents and storm wavesinto the impressive jumble of tidal sand ridges and banks that haveserved as sources for coastal (foredune, estuarine and back-barrier) ac-cretion. Sediment supply for the accumulation of the coastal dune bar-riers in the southern North Sea has depended essentially on stormsdriving sand from these shoreface banks. From high-resolution seismicprofiles of these sand banks, Tessier et al. (1999) identified a pattern ofbank mobility involving storm waves dominantly orthogonal to thecrests and seaward flanks of the tidal banks, which tend to migrate

lais and Dunkerque.

Sig

nific

ant w

ave

heig

ht (

m)

Sig

nific

ant w

ave

heig

ht (

m)

Sig

nific

ant w

ave

heig

ht (

m)

0

0.5

1

1.5

2

2.5

14/02 15/02 16/02 17/02 18/02 19/02 20/02 21/02 22/02 23/02 24/02 25/02 26/02 27/02 28/02

Offshore (-27 m) WesthinderNearshore (-1.9 m) site 2 Nearshore (-1.4 m) site 1 Beach (+ 2.2 m) site 2

2

1

1.5

0.5

0

2012

03-1

1-19

:30

03-1

0-11

:30

03-1

3-3:

3003

-14-

11:3

003

-15-

19:3

003

-17-

3:30

03-1

8-11

:30

03-1

9-19

:30

03-2

1-3:

3003

-22-

11:3

003

-23-

19:3

003

-25-

3:30

03-2

6-11

:30

03-2

7-19

:30

03-2

9-3:

3003

-30-

11:3

003

-31-

19:3

004

-02-

3:30

04-0

3-11

:30

04-0

4-19

:30

04-0

8-16

:30

04-0

7-8:

30

04-0

6-00

:30

1

2

3

4

5

6

7

8

Westhinder

Trapegeer

Wav

e pe

riod

6

5

4

3

2

1

0

mean

minJ F M A M J J A S O N D

maxWesthinder

A2-B

a

b

c

Fig. 3. Wave conditions in the southern North Sea: (a) a 26-year (1977–2012) record of maximum, mean and minimum significant wave heights (Hs) from two waverider bouystations off the Belgian coast (see Fig. 1 for locations); (b) a typical recent (10 Mar.–09 Apr., 2012) record of mean significant wave heights (Hs) from the Westhinder (in red)and Trapegeer (in blue) (see Fig. 1) bouys; the same record also shows typical wave periods at Westhinder (in black); (c) a comparaison of mean significant wave heights overa two-week period (14 Feb.–28 Feb., 2007) at Westhinder and from inner shoreface and beach deployments near Dunkerque, showing the significant drop in wave heights dueto dissipation on the shoreface. (a, b) from Administration Waterways and Marine Affairs of Flanders, and Flanders Marine Institute, Belgium (http://www.vliz.be/vmdcdata/midas/mvb.php), (c) from Héquette et al., 2009.

11E.J. Anthony / Geomorphology 199 (2013) 8–21

shoreward, whereas tidal currents, enhanced by wind stress duringstorms, tend to ‘stretch’ the banks alongshore. The stretching processmay eventually result, where the crest becomes very narrow, in bankdivision. Shoreward mobility may involve crest lowering as sedimentis transferred through bedform (dune) migration from the exposed tothe more sheltered landward flank of the bank (Houthuys et al.,1994). Sandmoving towards the landwardflanksmay also be depositedin troughs dominated by tidal currents. These ‘orthogonal’ directions ofwave-versus tidal current influence on sand banks have been corrobo-rated by Héquette et al. (2008a) from measurements of bedload trans-port on the inner shoreface off Dunkerque (Fig. 6). Giardino et al. (2010)have also shown, in a morphodynamic modelling study of sand banks,that the interaction between wave activity and tidal currents leads toa high increase in bottom shear stress, especially at the sand bank crestsand, as a consequence, to an increase in sand tranport. The studyshowed that wave activity is also responsible for a change in directionof the net flux of sediments during the tidal cycle. The actual mecha-nisms of onshore bankmigration, however, still remain elusive. The on-shore sand transport occurring during storms is probably related to

offshore currents being weaker across a shallow sand bank and/or tothe alongshore diversion of these currents by radiation stress gradientsgenerated by the 3D-bathymetry, such that waves become thedominant (onshore) transport mechanism. Where banks are veryclose to the shoreline, these diverted currents may also transportalongshore sediment eroded from the beach and foredune duringstorms, mitigating, in this way, sediment losses offshore.

Long-term onshore bankmigration under impinging stormwavesappears to be modulated by: (1) shoreface morphology, with migra-tion occurring where the seaward flanks of the banks are directly ex-posed to storm waves; and (2) tidal stage. These two points alsohighlight the important role of the bank field in modulating onshorewave energy transmission. Point 1 implies that the dissipation ofstorm wave energy over the most seaward banks may significantlylimit the propensity for the migration of banks closer to the shore,whereas bank stretching, and eventual division, induced bylongshore currents, become more prevalent. Point 2 concerns thecontrasting effects of neap and spring tides on bank migration andstretching. Neap tides and low waters at spring tides are generally

Fig. 4. A typical shoreface record of hydrodynamic and meteorological parameters recorded 2 km offshore of Dunkerque (26/03 to 09/04, 2004), showing from top to bottom: waterdepth, mean near bottom current direction, mean near-bottom current speed, significant wave height recorded in 5 m water depth (relative to Hydrographic Datum), and windspeed and direction measured in Dunkerque. The record shows the conjugate strengthening effect of spring tides and wind stress on tidal currents. Also note the clear relationshipbetween wind speed and wave height.From Héquette et al. (2008a).

12 E.J. Anthony / Geomorphology 199 (2013) 8–21

more favourable to bank migration as wave energy dissipation be-comes concentrated on the bank field. These two points explain thestrong gradients in wave energy from the deeper shoreface to theshoreline (Fig. 3), whereas high waters during spring tides favourmore efficient shoreward transmission of wave energy (see relation-ship between tide-modulated water depth and wave height in Fig. 5).

A final point to be considered is that the sand bank field may also behaving a large-scale feedback effect on the intensity of incident storms.Notwithstanding the fair frequency of storms in theNorth Sea, storm in-tensity in the southern part of theNorth Sea appears to be lower thanonthemore exposed coasts of Belgium, Holland, Germany andDenmark, ifone is to judge by the effects of historic storms on the countries

0

4590

135180

225

270315

360

1

2

3

4

5

0

0.4

0.8

1.2

1.6

2

00.20.40.60.81.01.21.41.6

04590

135180225270315360

-2.0-1.5-1.0-0.5

00.51.01.52.0

-0.2-0.1

00.10.20.30.4

00.51.01.52.02.53.03.5

02468

1012141618

Win

d di

rect

ion

Offs

hore

wav

e he

ight

(m

)W

ind

spee

d (

m s

-1)

Wat

er d

epth

(m

)M

ean

over

all

curr

ent (

m s

-1)

Mea

n ov

eral

lcu

rren

t dire

ctio

nM

ean

long

shor

ecu

rren

t (m

s-1

)M

ean

cros

s-sh

ore

curr

ent (

m s

-1)

Sig

nific

ant

wav

e he

ight

(m

)

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

14/01 16/01 18/01 20/01 22/01 24/01 26/01 28/01

nd

nd

nd

nd

nd

NE

SW

Offshore

Onshore

nd

SpeedDirection

Bar 2Bar 3

Idemfor 4to 8

16151413121110987654321 17 18 19 20 21

Fig. 5. A typical record of hydrodynamic and meteorological parameters recorded at high tide over two intertidal bars on the beach in Wissant Bay (14/01 to 28/01, 2004). Numbersin the water depth panel correspond to tides; nd: faulty data because of storm overturning and sand burial of sensors. From Sedrati and Anthony (2007). Note, as in Fig. 4, the strongrelationship between wind stress and currents and waves.

13E.J. Anthony / Geomorphology 199 (2013) 8–21

bordering this sea (Lamb, 1991). Such regional differences are, howev-er, still poorly known and await more complete investigation (Stoneand Orford, 2004). Lower intensity of storms in the southern NorthSea may also be, in addition to a probable regional meteorological

storm intensity gradient from the north to the south of this sea, an out-come of significant dissipation of storm wave energy by the impressivesand bank field concentrated in this southern area, as suggested bythe marked cross-shelf wave energy gradients depicted in Fig. 3. This

Fig. 6. Time series of (a) computed shear velocity and (b) modeled sediment transport (bedload qb and total load qt) on the shoreface off Dunkerque (see also Fig. 4) based on theSEDTRANS96 model. The bedload transport directions are closely conditioned by the tide-induced longshore currents (270°) and by onshore (360°) wave-induced bottom stress.Adapted from Héquette et al. (2008a).

Fig. 7. A Formosat 2 image (27/09/2008) of the North Sea coast midway between Calaisand Gravelines (Fig. 1), showing the eastern limit of a sand flat shore that significantlyprograded in the course of the 20th century from the welding of a shoreface bank. Theunvegetated part of the bank is subject to aeolian activity and foredune development,whereas the dark inner part consists of a mudflat that has evolved into a saltmarsh. Thecircular depressions in this saltmarsh are artefacts dug out by wildfowl hunters. In thistransition zone, the present North Sea dune shoreline in the east has been isolated in-land in the west by tidal flat progradation.

14 E.J. Anthony / Geomorphology 199 (2013) 8–21

limiting condition is further exacerbated by the large tidal range in thisarea, compared to themore storm-wave dominatedmeso- tomicrotidalcoasts bordering the rest of the North Sea.

3.2. Sand supply from the shoreface and barrier accretion

Based on the large-scale sand circulation patterns evoked inSection 3.1, Anthony (2000, 2002) suggested, for the storm and tide-dominated coasts of the eastern English Channel and the southernNorth Sea, a sequential relationship of coastal (dune, estuarine andbackbarrier) accretion involving shoreward attachment of tidal sandbanks, followed by backshore sand flat accretion (and/or estuarineand backbarrier infill), embryo dune development and foredune devel-opment. Subsequent studies involving determination of shorelinechanges at a secular scale (Aernouts and Héquette, 2006; Chaverot etal., 2008), bathymetric chart differencing at the same scale (Héquetteand Aernouts, 2010), monitoring of beach-dune budget changes(Anthony et al., 2006, 2007a, 2009), monitoring of recent coastal pro-gradation (Aubry et al., 2009), and stratigraphic analysis of an isolatedinland dune barrier (Anthony et al., 2010), and of backbarrier depositsnear Dunkerque (Mrani-Alaoui and Anthony, 2011), confirm this rela-tionship. In this setting, significant seaward shoreline translation is as-sociated with the formation of a wide sand flat, followed, with a timelag of the order of years, by the growth of aeolian dunes. Such transla-tion is essentially related to episodic wholesale onshore welding of alarge tidal sand bank driven from the shoreface by repeated activity ofstorm waves. The welded sand bank forms extensive temporary (atscales of decades to centuries) intertidal backshore sand flats thatserve as a basement and a sand source for the subsequent rapid accre-tion of aeolian dunes, essentially at the dry inner flanks of these flats(Anthony et al., 2006, 2007a). A fine recent example of this has beendocumented from the coast midway between Calais and Gravelines(Fig. 1), where, during the 20th century, localised accretion resulted ina shoreline sand-flat bulge up to 3 km-long and over 1 km-wide(Fig. 7), following the onshore welding of a large tidal sand bank(Garlan, 1990). Bathymetric and shoreline changes in this area showan extensive sand flat that underwent progradation of more than300 m between 1949 and 2000 (Héquette and Aernouts, 2010). Theseauthors calculated from the differences between bathymetric charts awelded bank volume that grew up to about 100×106 m3 in the course

of the 20th century. Héquette and Aernouts (2010) further showed,from SWAN wave propagation modeling based on sequential bathy-metric changes in this site, the increasing effect of bank accretion onnearshore dissipation of wave energy.

Significant sand supply to the backshore flat from a welded banksource is episodic, and results solely frommajor storms. These are capa-ble of driving large amounts of sand onto the upper beach and to thebackshore flat from the shore-attached sub-tidal sand bank source, asshown by correlative analysis of sequential digital elevation models of

15E.J. Anthony / Geomorphology 199 (2013) 8–21

the beach, the backshoreflat and embryo dune front, and correspondingstrongwind and storm episodes (Anthony et al., 2006). Swash washingover the backshore sand flat rapidly percolates over this surface, en-hancing its accretion. Wind action over the dry parts of the sand flat re-distributes sand towards embryo dunes that develop over time into alongitudinal dune complex (Anthony et al., 2007a).

Apart from this dominant, but relatively rare, mode of large-scalebank-sourced accretion, sand supply to the dune barrier may also beassured via two other secondary modes: (1) successions of largesand waves up to 500 m-long and more than 5 m-high, detachedfrom tidal banks and ridges close to the shore and driven ashore bystorms (Beck et al., 1991; Anthony et al., 2005), and (2) activestorm-generated migration of small (0.05–0.25 m) to medium(0.25–0.5 m) 2D to 3D subaqueous dunes over beach longshore barsduring tidal submergence (Sedrati and Anthony, 2007).

3.3. Barrier shoreline variability and erosion

A remarkable feature of the southern North Sea coast is its markedalongshore and temporal variability in terms of accretion, erosion andstability (Anthony and Héquette, 2005; Aernouts and Héquette, 2006;Anthony et al., 2006; Chaverot et al., 2008). Fig. 8 shows detailed andhighly variable patterns of shoreline changes determined by Chaverotet al. (2008) for Wissant Bay and for part of the dune barrier shorelinewest of Calais and covering the period 1949–2000.

TheWissant Bay dune barrier is in a particularly critical condition interms of erosion. Nearly 80% of the 8 km-long shoreline in this bay iseroding and parts of transects 13 to 33 (Fig. 8a) show some of thehighest rates of historical shoreline retreat in France. Parts of the centraland western sectors of the bay retreated by up to 250 m between 1949and 2000, following an early period of stability and even progradation(Aernouts and Héquette, 2006). This eroded part of the bay shows out-crops of peat on the beach that represent former backbarrier vegetation.In areas close to Cape Gris Nez, an upper beach frame of gravel has ac-cumulated as the foredune has retreated (Anthony and Dolique,2001). This retreat constitutes a threat in the coming years, because ofthe likelihood of a storm breaching of the narrowing dune barrier. Theprofile of the beach in the eroding sectors of the bay is significantlylower than in the shorter accreting sector in the east. In terms of theoverall shoreline dynamics, the western and central parts of the bayform an updrift erosional sector linked to a downdrift sand sink in theeast, the latter characterised by significant foredune growth and activeformation of embryo dunes (Anthony et al., 2006).

Reichmüth and Anthony (2002) showed that the beach in the moststrongly eroding central sector (Dune d'Aval, Fig. 8a) of the barrierunderwent a net volumetric loss of over 10% between 1996 and 2000.Digital elevation models of this eroding sector highlight an upperbeach subject to much stronger budget fluctuations than the dunefront (Anthony et al., 2006). The role of storms in barrier retreat inthis sector has been demonstrated by joint monitoring of waves andcurrents, water levels and surges, and high-resolution beach profiling(Ruz and Meur-Ferec, 2004; Sedrati and Anthony, 2007, 2008). Sedratiand Anthony (2008) have shown that under strongwind conditions co-inciding with spring tides, even the upper beach is subject to strongwind-, wave- and tide-induced longshore currents that lead to activeeastward migration of subaqueous 2D and 3D dunes during tidal sub-mergence. Such strong bedform development on the upper beach inWissant is a source of significant day-to-day beach profile mobility. Be-cause of chronic erosion, the lowered upper beach surface is subject tostrong moisture control (Ruz and Meur-Ferec, 2004), which, togetherwith the complex surface topography imprinted by bedforms, tends tolimit mobilisation of aeolian sand towards the foredune in this sector.Ruz and Meur-Ferec (2004) further showed that aeolian transport andthe evolution of the upper beach and dune front were strongly con-trolled by the magnitude and frequency of occurrence of high waterlevels, and that aeolian deposition on the upper beach above mean

highwater spring tide level in this eroding sector occurred only in sum-mer, whereas upper beach and dune-scarp erosion were likely to occurthe rest of the year whenever storms coincided with high (spring tide)water levels. These authors also indicated that a single storm couldcompletely annul anymild gains because of summer aeolian accumula-tion. Sedrati and Anthony (2008) showed that the annual mean shore-line retreat rate of over 4 m calculated for this central sector byAernouts and Héquette (2006) may be attained in just 24 h duringstorms associated with high surge levels (up to 1 m) and high springtides.

Marked alongshore variability is also shown by the dune barrier be-tween Sangatte and Calais (Fig. 8b), east of Cape Blanc-Nez (Fig. 1). Inthis example, the relatively stable shoreline in Sangatte alternatesalongshore with an erosional sector over a few transects (ca. 0.5 km),and then with a sector of significant progradation (transects 33 to 41,Fig. 8b), associated with a very shallow accreted inner shoreface,succeeded alongshore to the northeast by a sector of variable retreat.Old maps of the Calais area suggest a welded sand bank in the area ofstrong progradation. Perhaps more significantly, Chaverot et al.(2008) also highlighted alternating phases of erosion and accretion atseveral locations, indicating relatively sharp reversals from accretionto erosion and vice versa.

East of Dunkerque (Fig. 1), the dune barrier was seriously dam-aged at the beginning of the 20th century by urban developmentand by the construction of the Atlantic wall and blockhouses ofWorldWar II. In the 1980s, the 10 to 20 m high foredune was affectedby breaches and blowouts, and by erosional scarps cut during storms.The foredune is presently in a state of decadal-scale stability, attribut-ed in part to human intervention (Ruz and Anthony, 2008). Active re-habilitation carried out in the early 1990s, based on a clear dunemanagement scheme involving plantations, construction of fences,and protection of the foredune, has even resulted in mild accretionand incipient foredune development in places. This managementscheme is still actively implemented. Mild dune scarping in winterin this area is often followed in spring and summer by limited forma-tion of embryo dunes. Beach surveys at an experimental site east ofDunkerque reported by Reichmüth and Anthony (2007, 2008) high-light a relatively balanced sediment budget indicated by negligibleprofile volumetric variations, notwithstanding significant fluctuationsin profile morphology. These surveys also bring out longshore varia-tions in the characteristics of the beach profile that have been attrib-uted to the effects of breakwaters and to offshore protection fromwaves by a nearshore bank, the Hills Bank, extending along thecoast over a distance of about 9 km. The crest of the bank is common-ly exposed at spring low tides, and forms a shoal at a distance of about1400 m from the beach near Dunkerque (Fig. 1). The bank is separat-ed from the beach by a 10 to 15 m deep sub-shore-parallel channelthat is constantly dredged for navigation. Stable sand budgets yieldedby digital elevation models of the beach-dune interface (Anthony etal., 2007b) tend to confirm the state of quasi-stability of this part ofthe foredune. Further east towards Belgium, the dune front showsmore pronounced erosion, and various World War II blockhousesnow lie on the beach.

Looking at the entire southern North Sea coast of France betweenCapeGris Nez and Belgium, large stretches showeither signs of stability,or mild but chronic erosion of foredunes (Chaverot et al., 2008), withthe western end of Wissant Bay epitomizing extreme erosion. Thegross stability of the beach-foredune system suggests conditions ofrather limited sand supply from offshore, with exceptions in bank-sourced areas, such as near Calais. This situation embodies a clear para-dox, given: (1) the abundance of fine to medium sand on the nearbyshallow shoreface, and (2) the frequency of storms in the southernNorth Sea. The reasons for this require further studies, but one ofthese reasonsmay be the prevalence of equilibrium conditions betweenthe bank field and the hydrodynamic forcing, other than in exceptionalsituations such as that of Calais.

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71

60

40

20

0

-20

-40

-60

Sho

relin

e ev

olut

ion

(m)

Transects

1949 (reference year)196319771983

198919932000

Sangatte

Blériot-Plage

50°58'N

1°50'E

1 2 3 4 5 6 7 8 910

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

0 0.5 1km

b

Shoreline in 200022 Transect

Dune du ChâteletDunes d'Aval

Dune d'Amont

50°54'N50°53'N

1°37

'E

1°36

'E

1°40

'E

1°39

'E

1°38

'E

50°53'30"N

50°52'30"N

Wissant Bay

WISSANT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 2526 27282930 31 32 33 34

3536

37 38 39 4041

0 250 500

m

Transects

-10

-8

-6

-4

-2

0

2

4

Rat

e of

sho

relin

e ch

ange

(m

.yr-

1 )

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

1949-19631963-19771977-1983

1983-19891989-19931993-2000

a

Fig. 8. Rates of shoreline change calculated for various time slices between 1949 and 2000: (a) Wissant Bay, (b) between Sangatte and Calais. Note the strong fluctuations along-shore and over time.From Chaverot et al. (2008).

16 E.J. Anthony / Geomorphology 199 (2013) 8–21

Table1

Region

alco

ntrols

andsp

atiala

ndtempo

rals

calesof

morph

odyn

amic

interactioninvo

lved

intherelation

ship

betw

eenstorms,thesh

orefacean

dthedu

neba

rriersh

orelinein

thesouthe

rnNorth

Sea.

Region

alco

ntrols

Spatiala

ndtempo

rals

calesof

interaction

Macro

Meso

Micro

1.3D

shorefaceba

thym

etry

2.Storm

clim

ate(b

arom

etricpressu

re,w

indsp

eedan

ddirection,

storm

wav

een

ergy

,windstress)

3.Macrotida

lsetting

(tidal

stag

e,tida

lcurrents)

1.Hyd

rody

namic

mod

ulation

(wav

een

ergy

dissipation,

radiationstress

grad

ients,long

shorecu

rren

tge

neration

,setup

)2.

Tida

lban

kch

aracteristicsan

dmob

ility

(san

dsu

pply)

1.Inne

rsh

orefacesand

wav

emob

ility

2.Be

achmob

ility

(barsan

dtrou

ghs)

3.Fo

redu

nemob

ility

(erosion

/accretion

)

1.Flat

beds

and2D

–3D

dune

s(sub

aque

ousto

intertidal)

2.Be

ach-

to-dun

esw

ashan

dae

olian

tran

sport(b

ar-troug

hfetchsegm

entation

,ba

ck-bea

chsw

ashan

dae

oliantran

sport)

17E.J. Anthony / Geomorphology 199 (2013) 8–21

4. Discussion

Given the apparently intricate relationship between the shorefaceand the coastal sand barriers in the southern North Sea, themain ques-tion raised by the variability in barrier shoreline trends highlighted inthe foregoing section, and posed in this paper, relates to the way thisrelationship is mediated by storms. This relationship revolves aroundtwo related elements: (1) interactions between storm waves andshoreface sand banks in this macrotidal setting, and (2) sand supplyand dune accretion where banks are close to the shoreline, or sand re-moval and dune erosion where protection and sand-sourcing are notassured by a bank in proximity to the shoreline. Themain regional con-trols are, thus, the 3D shoreface bathymetry and the storm and tidalsetting. Embedded in these lower-order conditions are the meso-scale and micro-scale processes of sand delivery to, or removal from,the shoreline. These embedded relationships, which assume mor-phodynamic feedback loops, are summarised in Table 1. They providea fine example of the synchroneity in sand supply from shoreface todune highlighted by Aagaard et al. (2004) and Houser (2009). Macro-scale interactions basically concern large-scalemodulation ofwave ener-gy through dissipation, and the generation of strong longshore currentsby gradients in radiation stress, wind stress and tides. Meso-scale inter-actions revolve around the mobility of sand waves, bar-trough beachmobility and foredune mobility. Micro-scale processes concern upper-flow regime conditions and bedformmobility in the subaqueous domain(tidal banks, sand waves) and in the surf zone (beach bars and troughs),and, finally, swash-zone and aeolian beach-dune sand transport. Theseaspects are first discussed with reference to the contrasts in responseof the various sectors of the barrier shoreline in the southern NorthSea. The potential role of storm context is then evoked as a source ofvariability in shoreline response to storms. Following this, the mor-phodynamic regime prevailing in this area is briefly compared to thatof more classical wave-dominated coasts.

A situation of significant accretion is clearly expressed in parts ofthe barrier between Sangatte and Calais (transects 33 to 41, Fig. 8b),and especially between Calais and Gravelines (Fig. 7), where jointbeach progradation and active foredune growth match the sustainedsecular sand sourcing by sand banks that have welded onshore underthe influence of storms. Further east, the barrier in the Dunkerque sectorshows overall relative equilibrium associated with a relatively stableforedune and intertidal bar-trough beach system. These conditions aretypical of a site that has fluctuated over the past century between milderosion/accretion and relative stability. Clearly, no generalised accretionof the foredune occurs in this eastern sector of the North Sea coast ofFrance, as previous studies had already shown (Clabaut et al., 2000;Vasseur and Héquette, 2000).

Although Reichmüth and Anthony (2008) have attributed the rel-ative stability of the Dunkerque shoreline sector to protection by theHills Bank, Héquette et al. (2009) did not identify a clear-cut effect ofthe bank on incident waves, the dissipation of which would appearto reflect the overall effect of the shoreface bank field (Fig. 3c), ex-cept during low tide stages, especially spring tides, when the HillsBank may be particularly effective. Banks, as suggested inSection 3.1, may, however, mitigate offshore sediment transport, es-pecially during storms, because of gradients in radiation stress gen-erated by the complex 3D bank structure. The overall denseshoreface bank field off the Dunkerque site clearly moderates im-pinging storm waves and the potential for coastal erosion, butbanks are probably not close enough to the shoreline in this sectorto promote noteworthy coastal accretion, or such accretion is imped-ed by strong longshore currents in the troughs or channels betweenbanks. Héquette et al. (2009) have suggested that the relatively deepchannel between the Hills Bank and the beach forms a barrier to on-shore sediment transport, while conversely favouring longshoretransport. This channel is dredged constantly to maintain adequatedepths for high-tonnage ships.

18 E.J. Anthony / Geomorphology 199 (2013) 8–21

The situation of the Wissant Bay barrier radically differs from thatof the Dunkerque barrier sector. The reasons for the onset of erosionand the now chronic nature of this erosion in Wissant Bay, whichwas progradational up to the early 20th century, are still not clear.They do seem to involve, however, interactions between the LineBank offshore, storms, longshore sand transport in the shoreface cor-ridor of which the bank is a part, and probably the activity of currentgyres (Anthony and Dolique, 2001) related to the projecting headlandof Cape Gris Nez (Fig. 9). The Line Bank underwent erosion during the20th century, in part because of now-prohibited aggregate extractionon a massive scale. It is possible that the dynamics of the Line Bank it-self are embedded in the process of larger-scale storm- and tide-controlled sand migration (see Section 3.1) from the eastern EnglishChannel towards the Dover Strait and the southern North Sea(Anthony, 2000). Aernouts and Héquette (2006) showed that theshoreline retreat in the southwestern part of the bay has been mat-ched by net sand loss by the Line Bank, with attendant bathymetriclowering. A SWAN wave propagation model simulation by these au-thors further showed that incident wave energy in this eroding sectorhad increased in 2002 relative to 1977 because of the lower bathym-etry of the bank. As lowering of the Line Bank has occurred, dissipa-tion of storm wave energy appears to have been largely transferredto the bar-trough beach and to the foredune (Sedrati and Anthony,2008). These have been rapidly retreating, releasing sand, which isthen transported alongshore by strong storm-induced currents viaactively migrating intertidal 2D–3D dunes that develop over the in-tertidal bars and in the troughs. Observations of erosion and accretionpatterns along the bay sediment cell and sand transport calculationsby Sedrati and Anthony (2007) show that sand released from shore-line erosion in the western sector is not lost offshore, but is trans-ported towards the accreting sink zone in the east (Figs. 8a, 9). Thislongshore transport is particularly strong during storms coincidingwith high spring tides, most likely because of currents due to tidesand wind stress, and because gradients in radiation stress, as wavespass or break over the Line Bank, result in the diversion of offshoreflows alongshore, thus preventing sand loss offshore. One of the twotrailing edges of the Line Bank is close to the shore in the accretingsector (Fig. 9), probably providing sand for the coastal dunes andsheltering the shore from the larger storm waves.

Two clear aspects that come out from knowledge acquired on barri-er dynamics inWissant Bay are, thus, the close relationship between thehistoric lowering of the bathymetry of the Line Bank and foredune re-treat (Aernouts and Héquette, 2006), and the highly rhythmic natureof this foredune retreat which depends on the right combination ofstorm waves, spring tidal range and storm surge conditions (Sedratiand Anthony, 2008). In this large tide-range setting, the impact ofstorms may only become significant when storm conditions coincidewith large spring tides (e.g., Cooper et al., 2004; Pirazzoli et al., 2007;Ruz et al., 2009).Whereas storm erosion is important in the recent dra-matic evolution of this bay, however, the role of storms appears to havebeen exacerbated by changes in shoreface bathymetry and the influ-ence on enhanced longshore currents and wave energy supply to theshoreline in this macrotidal setting, where highwater at spring tides fa-vours significant onshore wave impingement, while low tides at springcan strongly truncate impingement.

The contrasting longshore patterns in foredune accretion and ero-sion exhibited by coastal sand barriers in the southern North Sea are,thus, to a large degree, strongly dependent on the interaction betweenstorms and the shoreface morphology expressed by the jumble oftidal sand banks and ridges that characterise the southern North Sea.The response to high-energy events is strongly conditioned by theinner shoreface, the 3D structure of which, in turn, depends on thelonger-termpattern of shorefacemorphodynamics and sand redistribu-tion embracing both the eastern English Channel and the southernNorth Sea. Bank location far offshore constitutes a limiting condition,which, together, with the relative inertia of large-scale shoreward

bank movement, explains the commonality of foredune erosion in thesouthern North Sea. The important but localised foredune accretion inthe Calais sector over the last century reflects the fortuitous location,at these sites, of sand banks that had migrated close enough to theshoreline over the last centuries to be finally driven ashore and weldunder the impact of the regular storm regime that controls, togetherwith tides and wind stress, the hydrodynamics and shoreface sedimenttransport in the southern North Sea. In contrast to these sites, morpho-logical changes affecting the Line Bank are deemed to be responsible, atleast in part, for the significant erosion of the western and central partsofWissant Bay, the bank no longer acting in these sectors as a dissipatorof storm waves nor as a supplier of sand. This erosion can only be bal-anced by fresh supplies of sand from offshore, a condition that is notlikely to be met in the foreseeable future. The persistence of such ero-sion clearly suggests that a storm-cut and fairweather-fill regime,more typical of fully wave-dominated systems, does not operate onthis storm- and tide-dominated coast because of the propensity forstrong longshore sand transport, generally to the east, over the innershoreface and intertidal zone during high tide.

Whereas the foregoing considerations mainly concern embedded-scale process-response interactions, another important point concernsstorm context variability. This includes variability in storm barometricpressure, wind speed and direction relative to the shoreline, setup andtidal stage during storms, all of which have a bearing on the outcomeof storm–shoreline interaction. The bewildering variability inherent inthis context will still continue to render unpredictable the response ofshores to many storms. One particular point concerns the relationshipbetween stormwind direction relative to the southern North Sea shore-line, barometric pressure, and tidal stage. A favourable combination ofstrong onshore winds, storm waves, large spring tides, and low baro-metric pressure must be considered as a potential generator of barrierbreaching that poses a permanent background threat to the southernNorth Sea coastal plain, large areas of which lie below present meansea level. Ruz et al. (2009) documented foredune changes near Dun-kerque caused by a storm characterised bymoderate but direct onshorewinds blowing more than 48 h in March 2007, associated with springtides. Notwithstanding the moderate wind speeds, this storm resultedin major foredune retreat, following a decade marked by larger stormsbut with little impact on the coast. The results show that on this mac-rotidal coast, erosive events are not necessarily associated with strongwinds, whereas wind direction and duration combined with a springtide appear to hold one of the keys tomedium-term foredune evolution.This example also illustrates a situation where a combination of highwater levels because of the spring-tide influence and storm-relatedwater setup in the coastal zone rendered the sand banks inefficient aswave dissipators, notwithstanding the moderate storm wave energy.

The foregoing elements of analysis of the storm–shoreline rela-tionship in the southern North Sea may also provide a template fora better understanding of temporal trends in this relationship, suchas those identified by Chaverot et al. (2008). These authors haveshown that the highly variable evolution of the shoreline of northernFrance during the second half of the 20th century, elements of whichare depicted in Fig. 8, displayed no clear relationship with storminessin most cases. They showed, for instance, that from 1963 to 1977, theshoreline significantly advanced seaward at several sites whereas theperiod 1972–1977 corresponded to that of maximum storm activityin this area. Conversely, shoreline retreat occurred at most sites dur-ing the 1990s, despite an identified significant decrease in stormfrequency.

The storm–shoreface–shoreline interaction pattern identified forthe southern North Sea coast is different from that of more classicalwave-dominated coasts such as those documented by Hequette et al.(2001) and Backstrom et al. (2008, 2009). It involves macro- to micro-scalemorphodynamic feedback between storms and the numerous dis-sipative sand banks, and onshore sand transfer via dominantly massbank migration. These sand bank movements most likely involve

Fig. 9. Sketch of Wissant Bay, showing the barrier shoreline status, the Line Bank offshore and aspects of the hydrodynamic and sand circulations offshore and on the beach. Bathy-metric contours are from Augris et al. (1995).Adapted from Anthony and Dolique (2001).

19E.J. Anthony / Geomorphology 199 (2013) 8–21

timescales of decades to centuries and include embedded smaller-scalesandmovements involving sandwaves in the subtidal zone close to theshoreline, beach longshore bar-trough formation and shorewardmigra-tion, and at increasingly smaller scales, bar bedformmigration and bothswash and aeolian transport in the intertidal zone (Table 1). Insediment-budget terms, this pattern also differs from that of thewave-dominated low-lying barrier island systems characterising theAtlantic and Gulf of Mexico seaboards, for instance. In many areas onthese coasts, the shoreface, nearshore multiple bar complexes, andbeaches evolve in a sand-limited context characterised by sand transferlandward into multi-decadal to century-scale storage in coastal dune,barrier, and flood-tidal delta sinks, in part via washovers during storms(e.g., Forbes et al., 2004).

5. Conclusions

1. Sand barriers in the southern North Sea exhibit contrasting long-shore and time-varying (multi-annual) patterns of foredune accre-tion and erosion that are, to a large degree, strongly dependent onthe interaction between shoreface morphology, dominated by sub-aqueous tidal sand banks that characterise the shallow southernNorth Sea, and storms.

2. The rather irregular pattern of shoreline accretion, stability or ero-sion shown by barriers appears to be largely from: (i) longshorevariations in the onshore supply of sand from the shore-parallelto sub-shore-parallel banks, and (ii) incident storm wave energyvariations related to shoreface bathymetry.

3. Foredune accretion is associated with areas wheremajor sand bankshave migrated close to the shoreline over the last centuries to be fi-nally driven ashore and weld under the impact of the storm regimethat controls, together with tides and wind stress, the hydrodynam-ics and shoreface sediment transport in the southern North Sea. Incontrast, morphological changes in nearshore banks, such as bankvolume depletion through longshore stretching or aggregate extrac-tion, can favour significant foredune erosion, such banks no longeracting as dissipators of storm waves.

4. These macro-scale interactions underline the synchroneity of sandtransport from shoreface to dune, and include embedded meso-scale interactions relating to the mobility of sand waves, bar-trough beach mobility and foredune mobility. These are, in turn,translated by micro-scale processes involved in 2–3D dunebedform mobility in the subaqueous domain (tidal banks, sandwaves) and in the intertidal domain (beach bars and troughs), aswell as in swash-zone and aeolian beach-dune sand transport.

5. In addition to the foregoing conditions, foredune accretion anderosion also depend on storm context variability that includesstorm characteristics and wind incidence relative to the shoreline,and tidal stage during storms. These parameters introduce furthercomplexity and unpredictability of barrier response to individualstorms. A likely illustration of this may be the evidenced lack of aclear relationship between periods of storminess and coastalresponse.

6. Large stretches of the southern North Sea foredunes of France showeither signs of stability, or mild but chronic erosion that may sug-gests conditions of rather limited sand supply fromoffshore, with ex-ceptions in areas where a nearshore storm-driven tidal sand bankhas become shore-attached. This paradoxical situation, given theabundance of sand on the nearby shallow shoreface, may resultfrom large-scale shoreface equilibrium with the hydrodynamiccontext.

Acknowledgements

Special guest editors Nancy Jackson and Karl Nordstrom, twoanonymous reviewers, and Jack Vitek, are thanked for their construc-tive review comments. Denis Marin and Patrick Pentsch prepared theillustrations.

References

Aagaard, T., Davidson-Arnott, R., Greenwood, B., Nielsen, J., 2004. Sediment supplyfrom shoreface to dunes: linking sediment transport measurements and long-term morphological evolution. Geomorphology 60, 205–224.

20 E.J. Anthony / Geomorphology 199 (2013) 8–21

Aernouts, D., Héquette, A., 2006. L'évolution du rivage et des petits fonds en Baie deWissant pendant le XXè siècle, Pas-de-Calais, France. Géomorphologie: Relief,Processus et Environnement 1, 49–64.

Anthony, E.J., 2000. Marine sand supply and Holocene coastal sedimentation in north-ern France between the Seine estuary and Belgium. In: Pye, K., Allen, J.R.L. (Eds.),Coastal and Estuarine Environments: Geological Society Special Publication, vol.175, pp. 87–97. London.

Anthony, E.J., 2002. Long-termmarine bedload segregation and sandy versus gravellyHolocene shorelines in the eastern English Channel. Marine Geology 187,221–234.

Anthony, E.J., Dolique, F., 2001. Natural and human influences on the contemporaryevolution of gravel shorelines between the Seine estuary and Belgium. In:Packham, J.R., Randall, R.E., Barnes, R.S.K., Neal, A. (Eds.), The Ecology and Geo-morphology of Coastal Shingle. Westbury Academic and Scientific Publishers,Westbury, pp. 132–148.

Anthony, E.J., Héquette, A., 2005. The role of shoreface sand supply in aeolian dune de-velopment on the North Sea coast of France. In: Baeteman, C. (Ed.), Joint INQUA-IGCP project 495 Conference, Dunkerque, pp. 12–13.

Anthony, E.J., Orford, J., 2002. Between wave- and tide-dominated coasts: the middleground revisited. Journal of Coastal Research (Special Issue 36), 8–15.

Anthony, E.J., Levoy, F., Monfort, O., Degryse-Kulkarni, C., 2005. Short-term intertidalbar mobility on a ridge-and-runnel beach, Merlimont, Northern France. Earth Sur-face Processes and Landforms 30, 81–93.

Anthony, E.J., Vanhée, S., Ruz, M.H., 2006. Short-term beach-dune sand budgets on theNorth Sea coast of France: sand supply from shoreface to dunes and the role ofwind and fetch. Geomorphology 81, 316–329.

Anthony, E.J., Vanhée, S., Ruz, M.H., 2007a. Embryo dune development on a large ac-tively accreting macrotidal beach: Calais, North Sea coast of France. Earth SurfaceProcesses and Landforms 32, 631–636.

Anthony, E.J., Vanhée, S., Ruz, M.H., 2007b. An assessment of the impact of experimen-tal brushwood fences on foredune sand accumulation based on digital elevationmodels. Ecological Engineering 31, 41–46.

Anthony, E.J., Ruz, M.H., Vanhée, S., 2009. Aeolian sand transport over complex inter-tidal bar-trough beach topography. Geomorphology 105, 95–105.

Anthony, E.J., Mrani-Alaoui, M., Héquette, A., 2010. Shoreface sand supply andmid- tolate Holocene aeolian dune formation on the storm-dominated macrotidal coastof the southern North Sea. Marine Geology 276, 100–104.

Aubry, A., Lesourd, S., Gardel, A., Dubuisson, P., Jeanson, M., 2009. Sediment textur-al variability and mud storage on a large accreting sand flat in a macrotidal,storm-wave setting: the North Sea coast of France. Journal of Coastal Research(Special Issue 56), 163–167.

Augris, C., Clabaut, P., Tessier, B., 1995. Le domaine marin côtier du Nord-Pas de Calais:Carte des formations superficielles au 1:100 000. IFREMER/Région Nord-Pas de Ca-lais/Université de Lille I.

Backstrom, J.T., Jackson, D.W.T., Cooper, J.A.G., Malvarez, G.C., 2008. Storm-drivenshoreface morphodynamics on a low-wave energy delta: the role of nearshoretopography and shoreline orientation. Journal of Coastal Research 24,1379–1387.

Backstrom, J.T., Jackson, D.W.T., Cooper, J.A.G., 2009. Shoreface morphodynamics of ahigh-energy, steep and geologically-constrained shoreline segment in NorthernIreland. Marine Geology 257, 94–106.

Beck, C., Clabaut, P., Dewez, S., Vicaire, O., Chamley, H., Augris, C., Hoslin, R., Caillot, A.,1991. Sand bodies and sand transport paths at the English Channel–North Sea border:morphology, dynamics and radioactive tracing. Oceanologica Acta 11, 111–121.

Betts, N.L., Orford, J.D., White, D., Graham, C.J., 2004. Storminess and surges in theSouth-Western approaches of the eastern North Atlantic: the synoptic climatologyof recent extreme coastal storms. Marine Geology 210, 227–246.

Chaverot, S., Héquette, A., Cohen, O., 2008. Changes in storminess and shoreline evo-lution along the northern coast of France during the second half of the 20th cen-tury. Zeitschrift für Geomorphologie 52 (Suppl. 3), 1–20.

Ciavola, P., Armaroli, C., Chiggiato, J., Valentini, A., Deserti, M., Perini, L., Luciani, P.,2007. Impact of storms along the coastline of Emilia-Romagna: themorphologicalsignature on the Ravenna coastline (Italy). Journal of Coastal Research (SpecialIssue 50), 540–544.

Clabaut, P., Chamley, H., Marteel, H., 2000. Evolution récente des dunes littorales àl'est de Dunkerque (Nord de la France). Géomorphologie: Relief, Processus etEnvironnement 2, 125–136.

Claudino-Sales, V.,Wang, P., Horwitz,M.H., 2008. Factors controlling the survival of coastaldunes during multiple hurricane impacts in 2004, 2005: Santa Rosa barrier island,Florida. Geomorphology 95, 295–315.

Cooper, J.A.G., Jackson, D.W.T., Navas, F., McKenna, J., Malvarez, G., 2004. Identifyingstorm impacts on an embayed, high-energy coastline: examples from western Ire-land. Marine Geology 210, 261–280.

Corbau, C., Tessier, B., Chamley, H., 1999. Seasonal evolution of shoreface and beachsystem morphology in a macrotidal environment, Dunkerque area, NorthernFrance. Journal of Coastal Research 15, 97–110.

Ferreira, O., 2006. The role of storm groups in the erosion of sandy coasts. Earth SurfaceProcesses and Landforms 31, 1058–1060.

Field, M.E., Roy, P.S., 1984. Offshore transport and sand body formation: evidencefrom a steep, high-energy shoreface, Southeastern Australia. Journal of Sedimen-tary Petrology 54, 1292–1302.

Forbes, D.L., Parkes, G.S., Manson, G.K., Ketch, L.A., 2004. Storms and shoreline retreat inthe southern Gulf of St. Lawrence. Marine Geology 210, 169–204.

Garlan, T., 1990. L'apport des levées bathymétriques pour la connaissance de ladynamique seédimentaire. L'exemple des Ridens de la Rade aux abords de Calais.Littoral '90, Eurocoast Symposium, pp. 71–75.

Gervais, M., Balouin, Y., Belon, R., 2012. Morphological response and coastal dynamicsassociated with major storm events along the Gulf of Lions coastline, France. Geo-morphology 143–144, 69–80.

Giardino, A., Van Den Eynde, D., Monbaliu, J., 2010. Wave effects on the mor-phodynamic evolution of an offshore sand bank. Journal of Coastal Research (Spe-cial Issue 51), 127–140.

Grochowski, N.T.L., Collins, M.B., Boxall, S.R., Salomon, J.C., Breton, M., Lafite, R., 1993.Sediment transport pathways in the eastern English Channel. Oceanologica Acta16, 531–537.

Haerens, P., Bolle, A., Trouw, K., Houthuys, R., 2012. Definition of storm thresholds forsignificant morphological change of the sandy beaches along the Belgian coastline.Geomorphology 143–144, 107–117.

Héquette, A., Aernouts, D., 2010. The influence of nearshore sand bank dynamics onshoreline evolution in a macrotidal coastal environment, Calais, Northern France.Continental Shelf Research 30, 1349–1361.

Hequette, A., Desrosiers, M., Hill, P.R., Forbes, D.L., 2001. The influence of coastal mor-phology on shoreface sediment transport under storm-combined flows, CanadianBeaufort Sea. Journal of Coastal Research 17, 507–516.

Héquette, A., Hemdane, Y., Anthony, E.J., 2008a. Sediment transport under wave andcurrent combined flows on a tide-dominated shoreface, northern coast of France.Marine Geology 249, 226–242.

Héquette, A., Hemdane, Y., Anthony, E.J., 2008b. Determination of sediment transportpaths in macrotidal shoreface environments: a comparison of grain-size trendanalysis with near-bed current measurements. Journal of Coastal Research 24,695–707.

Héquette, A., Ruz, M.H., Maspataud, A., Sipka, V., 2009. Effects of nearshore sand bankand associated channel on beach hydrodynamics: implications for beach andshoreline evolution. Journal of Coastal Research (Special Issue 56), 59–63.

Hill, H.H., Kelley, J.T., Belknap, D.F., Dickson, S.M., 2004. The effects of storms andstorm-generated currents on the sand beaches in southern Maine. Marine Geology210, 149–168.

Houser, C., 2009. Synchronization of transport and supply in beach–dune interaction.Progress in Physical Geography 33, 733–746.

Houser, C., Hapke, C., Hamilton, S., 2008. Controls on coastal dune morphology, shore-line erosion and barrier island response to extreme storms. Geomorphology 100,223–240.

Houthuys, R., Trentesaux, A., De Wolf, P., 1994. Storm influences on a tidal sandbank'ssurface (Middelkerke Bank, southern North Sea). Marine Geology 121, 23–41.

Lamb, H.H., 1991. Historic storms of the North Sea, British Isles, and Northwest Europe.Cambridge University Press, Cambridge. 204 pp.

Mendoza, E.T., Jiménez, J.A., 2006. Storm-induced beach erosion potential on the Cata-lonian coast. Journal of Coastal Research 48, 81–88.

Morton, R.A., 2002. Factors controlling storm impacts on coastal barriers and beaches—a preliminary basis for near real-time forecasting. Journal of Coastal Research 18,486–501.

Mrani-Alaoui, M., Anthony, E.J., 2011. New data and a morphodynamic perspective onmid- to late Holocene palaeoenvironmental changes in the French Flanders coastalplain, southern North Sea. The Holocene 21, 445–453.

Pirazzoli, P., Costa, S., Dornbusch, U., 2007. Flood threat anomaly for the low coastalareas of the English Channel based on analysis of recent characteristic flood occur-rences. Ocean Dynamics 57, 501–510.

Regnauld, H., Pirazzoli, P.A., Morvan, G., Ruz, M.H., 2004. Impacts of storms and evolu-tion of the coastline in western France. Marine Geology 210, 325–337.

Reichmüth, B., Anthony, E.J., 2002. The variability of ridge and runnel beach morphol-ogy: examples from northern France. Journal of Coastal Research (Special Issue36), 612–621.

Reichmüth, B., Anthony, E.J., 2007. Tidal influence on the intertidal bar morphology oftwo contrasting macrotidal beaches. Geomorphology 90, 101–114.

Reichmüth, B., Anthony, E.J., 2008. Seasonal-scale morphological and dynamic char-acteristics of multiple intertidal bars. Zeitschrift für Geomorphologie 52 (Suppl.3), 79–90.

Ruz, M.H., Anthony, E.J., 2008. Sand trapping by brushwood fences on a beach-foredunecontact: the primacy of the local sediment budget. Zeitschrift für Geomorphologie52 (Suppl. 3), 179–194.

Ruz, M.H., Meur-Ferec, C., 2004. Influence of high water levels on aeolian sand trans-port and upper-beach/dune evolution on a macrotidal coast, Wissant bay, North-ern France. Geomorphology 60, 73–87.

Ruz, M.H., Héquette, A., Maspataud, A., 2009. Identifying forcing conditions responsiblefor foredune erosion on the northern coast of France. Journal of Coastal Research(Special Issue 56), 356–360.

Sabatier, F., Anthony, E.J., Héquette, A., Suanez, S., Musereau, J., Ruz, M.H., Régnauld,H., 2009. Morphodynamics of beach/dune systems: examples from the coast ofFrance. Géomorphologie: Relief, Processus et Environnement 2009/1, 1–13.

Sallenger Jr., A.H., Stockdon, H.F., Fauver, L., Hansen, M., Thompson, D., Wright, C.W.,Lillycrop, J., 2006. Hurricanes 2004: an overview of their characteristics and coastalchange. Estuaries and Coasts 29, 880–888.

Sedrati, M., Anthony, E.J., 2007. Storm-generated morphological change and longshoresand transport in the intertidal zone of a multi-barred macrotidal beach. MarineGeology 244, 209–229.

Sedrati, M., Anthony, E.J., 2008. Sediment dynamics and morphological change onthe upper beach of a multi-barred macrotidal foreshore, and implications formesoscale shoreline retreat: Wissant Bay, northern France. Zeitschrift fürGeomorphologie 52 (Suppl. 3), 91–106.

Service Hydrographique et Océanographique de la Marine, 1968. Les courants demarée dans la mer de la Manche et sur les côtes françaises de l'Atlantique. No.550, Paris, 176 pp.

21E.J. Anthony / Geomorphology 199 (2013) 8–21

Stone, G.W., Orford, J.D., 2004. Editorial: Storms and their significance in coastalmorpho-sedimentary dynamics. Marine Geology 210, 1–5.

Stone, G.W., Liu, B., Pepper, D.A., Wang, P., 2004. The importance of extratropical andtropical cyclones on the short-term evolution of barrier islands along the northernGulf of Mexico, USA. Marine Geology 210, 63–78.

Suanez, S., Cariolet, J.M., Cancouet, R., Ardhuin, F., Delacourt, C., 2012. Dune recoveryafter storm erosion on a high-energy beach: Vougot Beach, Brittany (France). Geo-morphology 139, 16–33.

Tessier, B., Corbau, C., Chamley, H., Auffret, J.-P., 1999. Internal structure of shorefacebanks revealed by high resolution seismic reflection in a macrotidal environment(Dunkerque Area, Northern, France). Journal of Coastal Research 15, 593–606.

Van Lancker, V., 1999. Sediment and morphodynamics of a siliciclastic near coastalarea, in relation to hydrodynamical and meteorological conditions: Belgian conti-nental shelf. Unpublished Ph.D. thesis, University of Ghent, 194 pp.

Vasseur, B., Héquette, A., 2000. Storm surges and erosion of coastal dunes between1957 and 1988 near Dunkerque (France), southwestern North Sea. In: Pye, K.,Allen, J.R.L. (Eds.), Coastal and Estuarine Environments: Geological Society SpecialPublication, vol. 175, pp. 99–107. London.

Wang, P., Kirby, J.H., Haber, J.D., Horwitz, M.H., Knorr, P.O., Krock, J.R., 2006. Morpho-logical and sedimentological impacts of Hurricane Ivan and immediate poststormbeach recovery along the northwest Florida barrier-island coast. Journal of CoastalResearch 22, 1382–1402.

Wright, L.D., Xu, J.P., Madsen, O.S., 1994. Across-shelf benthic transports on the innershelf of the Middle Atlantic Bight during the Halloween storm of 1991. Marine Ge-ology 118, 61–77.

Zhang, K., Whitman, D., Leatherman, S.P., Robertson, W., 2005. Quantification of beachchanges caused by Hurricane Floyd along Florida's Atlantic coast using airbornelaser surveys. Journal of Coastal Research 21, 123–134.