Geomorphology 283 (2017) 48–60

Contents lists available at ScienceDirect

Geomorphology

journal homepage: www.elsevier.com/locate/geomorph

Storm impacts and shoreline recovery: Mechanisms and controls in the

southern North Sea
S.M. Brooks a,⁎, T. Spencer b, E.K. Christie b
a
Department of Geography, Environment and Development Studies, Birkbeck, University of London, Malet Street, London WC1E 7HX, UK
b
Cambridge Coastal Research Unit, Department of Geography, University of Cambridge, Downing Place, Cambridge CB2 3EN, UK

a r t i c l e i n f o a b s t r a c t

Article history: Storm impacts play a significant role in shoreline dynamics on barrier coastlines. Furthermore, inter-storm recov-
Received 14 July 2016 ery is a key parameter determining long-term coastal resilience to climate change, storminess variability and sea
Received in revised form 3 January 2017 level rise. Over the last decade, four extreme storms, with strong energetic waves and high still water levels
Accepted 4 January 2017
resulting from high spring tides and large skew surge residuals, have impacted the shoreline of the southern
Available online 14 January 2017
North Sea. The 5th December 2013 storm, with the highest run-up levels recorded in the last 60 years, resulted
Keywords:
in large sections of the frontline of the North Norfolk coast being translated inland by over 10 m. Storms in March
Barrier dynamics and November 2007 also generated barrier scarping and shoreline retreat, although not on the scale of 2013. Be-
Intertidal bar tween 2008 and 2013, a calm period, recovery dominated barrier position and elevation but was spatially differ-
Foredune dynamics entiated alongshore. For one study area, Scolt Head Island, no recovery was seen; this section of the coast is being
Alongshore sediment transport reset episodically landwards during storms. By contrast, the study area at Holkham Bay showed considerable re-
covery between 2008 and 2013, with barrier sections developing seaward through foredune recovery. The third
study area, Brancaster Bay, showed partial recovery in barrier location and elevation. Results suggest that recov-
ery is promoted by high sediment supply and onshore intertidal bar migration, at rates of 40 m a−1. These pro-
cesses bring sand to elevations where substrate drying enables aeolian processes to entrain and transport sand
from upper foreshores to foredunes. We identify three potential sediment transport pathways that create a re-
gion of positive diffusivity at Holkham Bay. During calm periods, a general westward movement of sediment
from the drift divide at Sheringham sources the intertidal bar and foredune development at Holkham Bay. How-
ever, during and following storms the drift switches to eastward, not only on the beach itself but also below the
–7 m isobath. Sediment from the eroding barrier at Brancaster Bay, and especially Scolt Head Island, also sources
the sediment sink of Holkham Bay. Knowledge of foredune growth and barrier recovery in natural systems are
vital aspects of future coastal management planning with accelerated sea-level rise and storminess variability.
© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).

1. Introduction recovery presents a further challenge. These issues are particularly

pressing on low-lying coasts because in such low gradient settings
Shoreline response to variation in environmental forcing, both both long-term and short-term water level increases can translate a
chronic (sea-level rise) and acute (storms), is a vitally important yet long distance inland (Michener et al., 1997). Not surprisingly, therefore,
highly complex issue, dependent on numerous inter-related factors. there has been considerable recent research on storm impacts on barri-
Over the latter half of the twentieth century the rate of global mean er shorelines (e.g. Stone et al., 2004; Sallenger et al., 2006; Houser et al.,
sea level rise was 1.7 ± 0.3 mm a−1 (Church and White, 2006) and a 2008). Most recently, research has focussed on the coastlines of North-
comparable 1.4 ± 0.2 mm a−1 around UK coasts (Woodworth et al., west Europe where the most extreme storm sequence for 60 years oc-
2009). More recently, this rate appears to have accelerated globally curred during winter 2013–14 (Dissanayake et al., 2014; Castelle et al.,
(Ablain et al., 2009) as well as regionally around the UK (Wahl et al., 2015; Masselink et al., 2015; Wadey et al., 2015), affecting wide areas
2013). At the same time, ongoing uncertainty characterises future vari- of Atlantic France, western England (especially on SW coasts) and the
ability in tropical and extra-tropical storminess (IPCC, 2013). For coastal southern North Sea. These studies have served to further reinforce the
communities, infrastructure and biodiversity, elucidating the resulting need for a better understanding of the spatio-temporal dynamics of
morphodynamic linkages between storm forcing and shoreline storms and their interaction with surges and extreme waves and how
these impact upon the considerable morphological variability that
⁎ Corresponding author. characterises the coastal receptor (e.g. Ciavola et al., 2007; Sabatier et
E-mail address: [email protected] (S.M. Brooks). al., 2009; Vousdoukas et al., 2012; Anthony, 2013; Spencer et al.,

http://dx.doi.org/10.1016/j.geomorph.2017.01.007
0169-555X/© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60 49

2014). Combinations of maximum still water levels, peak significant In this study we focus on three sub-sections of the North Norfolk
wave heights and direction of wave approach result in a complex and (Fig. 1A) coast that together comprise the UK Environment Agency's
varied relationship with the shoreface that they encounter (Betts et ‘Superfrontage 2’ of Cell 5 of the ‘Second Generation’ Shoreline Manage-
al., 2004; Haerens et al., 2012). ment Plan (SMP5; East Anglian Coastal Group, 2010; Figs. 1B, 2). The
Progressive technical advances over the past two decades in partic- most westerly study area is Brancaster Bay (Fig. 2A), comprising a
ular can be utilised to elucidate more clearly some of these complexities. dune barrier varying in elevation, reaching a maximum height of 15 m
Vertical aerial photography of shoreline position can now be better east of the Royal West Norfolk Golf Course. Part of this frontage has
ground-referenced by the use of high resolution (three dimensional been protected with groynes and revetments, a response to rapid shore-
data quality of b50 mm and often b 20 mm) instrumentation. More re- line erosion and retreat over the past two decades. The westward end of
cently, these methods have been supplemented with airborne and ter- Brancaster Bay is terminated by a tidal delta at Titchwell; here the beach
restrial LiDAR datasets to ensure wider spatial coverage and more seaward of the dunes is wide and gently sloping. Eastward from
accurate determination of volumetric change. Thus, for example, Brancaster Bay, Scolt Head Island (Fig. 2B and D) projects into the sea
Anthony et al. (2006) provide a comprehensive assessment of 3-D sed- and forms a barrier island that stretches 6 km alongshore. Scolt is a rel-
iment exchanges between the beaches, foredunes and established atively young feature, forming over the past 3000 years (Andrews et al.,
dunes of the North Sea coast of France using laser scanning techniques. 2000). The main gravel and sand spine of the island supports an irregu-
To link storm-induced shoreline changes to particular storm events or larly-spaced series of landward-trending gravel ridges, or ‘ laterals’,
sequences of storms where still water levels, peak significant wave which enclose well developed back barrier marshes (Steers, 1934).
height and direction of approach are so important, there now exist ar- There has been progressive setting back of the barrier over time (with
chival data sets from wave buoys and tide gauge networks that can be the effect of recent storms (since 2006) being documented in Brooks
time-matched to capture the extreme storm-driven sea states that can et al., 2016) but it also extended westwards at an average rate of
lead to large scale shoreline re-positioning (e.g. for UK coasts: 2.53 m a−1 over the period 1891–2013. The third setting is Holkham
Hackney et al., 2013; Earlie et al., 2015; Brooks et al., 2016). Adopting Bay (Fig. 2C) between the eastern end of Scolt Head Island and Wells-
a methodology that combines GIS-based assessment of shorelines next-the-Sea. Holkham Bay has extensive sand dunes comprising the
digitised from aerial photographs (e.g. use of the USGS Digital Shoreline barrier, with very little gravel present. Behind are young back barrier
Analysis System (Thieler et al., 2009)); cross-shore profile analysis; field marshes that have been accreting rapidly over the past decade, to
measurements for ground-referencing; and archival records of storm form 20 ha of embryonic saltmarsh, characterised by the common
occurrences strengthens our understanding of the way shorelines be- saltmarsh-grass Puccinellia maritima. The dunes, up to 12 m in elevation,
have during both storm impact and storm recovery phases. This ap- are colonised by marram grass (Ammophila sp.) and there are signs of
proach allows consideration as to whether or not such shorelines foredune and embryo dune development (Fig. 2E, F). The dunes were
exhibit some sort of longer term equilibrium, allows identification of cut back by up to 15 m in the storm of 5 December 2013 (Spencer et
the primary controls on shoreline recovery, and informs debate on fu- al., 2015). Landward of Holkham Bay is the largest area of reclaimed
ture shoreline response to rising sea levels and storm variability. freshwater grazing marsh along the North Norfolk coast, at 295 ha
An earlier study (Brooks et al., 2016) focussed on storm impacts (East Anglian Coastal Group, 2010).
along a 6 km-long shoreline barrier, Scolt Head Island on the North Nor- The mean significant wave height along the North Norfolk coast is
folk coast. In this paper we triple the lateral extent of frontage consid- 0.49–0.73 m as recorded in nearshore wave buoys at Scolt and Cley (5
ered, from Holkham Bay (to the east) to Brancaster Bay (in the west), and 7 m water depth respectively) between September 2006 and Sep-
and extend our analysis to consider morphodynamic processes in the tember 2009. Further offshore at the Blakeney Overfalls wave rider
intertidal and subtidal zones. Specifically, we: buoy (10 km offshore, 18 m water depth), monthly mean significant
wave height ranged from 0.8 to 1.0 m over the same period
1. Investigate both cross-shore changes in shoreline profiles and along- (Environment Agency, 2014). There is a macrotidal semi-diurnal tidal
shore changes in shoreline position to examine variations in decadal regime, with mean spring tidal range falling from 6.4 m at Hunstanton
(1992–2014) behaviour; to 4.7 m at Cromer. Mean High Water Springs and Highest Astronomical
tide are 2.35 m and 2.99 m ODN respectively (ODN = Ordnance Datum
2. Examine in detail the shoreline positional change at these locations
Newlyn where 0.0 m approximates to mean sea level) at Cromer
in response to four high magnitude storms that have been identified
(http://www.ntslf.org/tides/hilo?port=Cromer), 15 km to the east of
from the archival record contained within tide gauge and wave buoy
the study area's eastern boundary. However, under storm surge condi-
records; and
tions maximum water levels can considerably exceed predicted tidal
3. Use intertidal zone cross-shore profile analysis for years of no known
heights. Thus in both the 1953 and 2013 storm surges, maximum
storm occurrence (calm periods) to identify contrasting recovery be-
water levels of N6 m ODN were recorded at Blakeney Harbour Quay
haviour at these locations for recent (2008–2013 and 2014–2015)
(Spencer et al., 2015). The typical wind regime of the North Norfolk
post-storm periods and explore their alongshore interactions.
coast reflects the passage of low pressure systems from the west, with
a dominant south-south westerly component. From this direction
2. Location and setting winds rarely reach speeds in excess of 16 m s−1, but the occurrence of
windspeeds over 8 m s− 1 is common (Weybourne weather station,
The North Norfolk Coast stretches over 45 km from Hunstanton in 2008–2013). Around 5% of winds from a south-south westerly direction
the west to Kelling Hard in the east (Fig. 1A). The regional geology is exceed 8 m s−1. During storms when winds approach from the north
characterised by a shallow, gently-inclined offshore slope which, west, north and north east, wind speeds are almost always in excess
under the influence of a macro-tidal range and moderate wave energy of 8 m s−1, commonly exceed 12 m s−1 and can exceed 16 m s−1; in ad-
climate, has given rise to a wide Holocene sedimentary prism. This is dition, fetch lengths are at their maximum from the north and north
characterised by a wide beach or sandflat in front of sand and gravel east.
barriers with inter-barrier tidal channels and back-barrier salt marshes Thus the coast of North Norfolk is vulnerable to, and shaped by, high-
(both natural and reclaimed) (Andrews et al., 2000).The higher barriers magnitude storms which approach from a northerly direction. Such
support sand dunes (locally with pine woodlands). Brackish water events can generate large surge residuals and strong onshore waves,
reedbeds are found where freshwater streams and seepages reach the particularly when they coincide with high spring tides. Twenty such
coast. Small coastal settlements are restricted to the higher storm events have been identified since 1881 that have been sufficiently
(ca. N 10 m) landward hinterland. damaging to have been reported in the media (Brooks et al., 2016). The
50 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60

A B

Fig. 1. General location and environmental setting of Brancaster Bay, Scolt Head Island and Holkham Bay, North Norfolk Coast. A) Location within the general setting of the UK east coast
and (inset) North Norfolk coast. B) Superfrontage 2 with the three study site locations, and their main landforms and habitats. Wave recording locations are also shown at Cley, Scolt Head
and Blakeney Overfalls. C) North Norfolk Coast aerial photograph from 15th July 2013 with symbols showing Environment Agency cross-shore profile locations. © Environment Agency
copyright and/or database right 2015. All rights reserved.

most devastating storm in terms of coastal inundation, damage to infra- further below, but in summary they include: 1) digitising shorelines
structure and agriculture, and loss of life was the 31st January–1st Feb- from annual (typically August–September) UK Environment Agency
ruary 1953 storm which claimed N 2000 lives in The Netherlands (EA) aerial photography; 2) using EA bi-annual cross-shore profiles
(Gerritsen, 2005). The 5th December 2013 storm surge in places along (6-monthly intervals from 1992 to present, with a ‘summer’ (typically
the North Norfolk coast generated higher water levels than in 1953 August–September) and a ‘winter’ (typically January–February) survey
(Spencer et al., 2015). With strengthened post-1953 defences and bet- each year) to provide a 2-dimensional assessment of change at point lo-
ter early warning systems, no lives were lost from this event. Neverthe- cations; 3) obtaining field Real Time Kinematic (RTK) data to identify
less shoreline impacts were considerable in terms of retreat in the the shoreline position following the 5th December 2013 storm; and 4)
barriers, overtopping and breaching as well as flooding. using available records of waves (https://www.cefas.co.uk/cefas-data-
hub/wavenet/) and still water levels (https://www.bodc.ac.uk/data/
3. Methods online_delivery/historical_uk_tide_gauge_data/) to assess the sea states
experienced during periods of greatest shoreline change. We also
We used a variety of methods to develop a picture of positive (ad- consulted the Weybourne weather station for wind speed and duration
vance) or negative (retreat) shoreline change. The details are described (recorded every minute) on specific days in the recovery phases. These
 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60 51

A B C

E F

Fig. 2. Superfrontage 2 (East Anglian Coastal Group, 2010) on the North Norfolk coast. A) Brancaster Bay looking east towards Scolt Head Island in the far distance, showing barrier cliffing
following the 5th December 2013 storm (photo: SM Brooks 12.01.2016); B) Scolt Head Island following the 7th–8th November 2007 storm showing breaching and overwash of the barrier
(photo: D. Friess 11.12.07); C) the barrier at Holkham Bay looking east towards Wells-next-the-Sea shortly after the 5th–6th December 2013 storm (photo T. Spencer 16.1.2014); D)
general setting of Scolt Head Island showing the barrier and the back barrier marshes following the 5th December storm with Brancaster Bay in the far distance (photo: M. Page
9.12.13) E) Holkham Bay looking east showing steep seaward facing dunes of 12 m fronted by foredunes of 5 m (photo: S.M. Brooks 30.12.15); F) sand saltation on Holkham Beach in
winds of 12.07 m s−1 from 241.3° at Weybourne, with a wet beach section in the breaking zone associated with the landward face of the intertidal bar preventing saltation (photo
S.M. Brooks 30.12.16).

datasets were used to: 1) establish the shoreline position for the entire November 2008 and 5th–6th December 2013. The nature and impacts
decadal-scale period of available records (1992 to 2014) and for shorter of the last storm in this sequence has been well documented (Spencer
periods between 2006 and 2007, 2007–2008, 2008–2013 and 2013– et al., 2015; Wadey et al., 2015). We were unable to consider storms be-
2014; 2) provide the 6-monthly change in the crest of the shoreline bar- fore summer 2006 as there were no datasets to provide nearshore wave
rier between summer 2006 and summer 2014; 3) assess cross-shore information in the locality before summer 2006. Thus the key time pe-
profile change for the same periods, but also including the period riods for the following analysis were: (a) 1992–2014 to assess decadal
2014–2015 (following the major storm of 2013). Wave and still water shoreline change; (b) 2006–2007, 2007–2008 and 2013–2104 to assess
level data sets were used to define thresholds for three large storms, storm impacts; (c) 2008–2013, a long period of no major storms; and
identified from the EA Sea State Reports for 2006–2007 and 2006– (d) 2014–2015 to assess recovery following the large surge event of
2009 (Environment Agency, 2014; Brooks et al., 2016) and from archi- 5th December 2013.
val records (Brooks et al., 2016). The threshold used to define these
storms was taken as the combination of significant wave heights ex- 3.1. Alongshore shoreline change for the entire North Norfolk barrier
ceeding 3.5 m at Blakeney Overfalls (offshore) or 2.2 m at Scolt Head Is- coastline
land (inshore); a northerly (337.5–22.5°) direction of wave approach;
and still water levels exceeding 3 m ODN in the Immingham tide A major issue for studies such as this one is that of defining the
gauge record. The dates of the storms thus identified occurred between shoreline (Moore, 2000; Stockdon et al., 2002). This is especially a prob-
31st October–3rd November 2006, 17th–20th March 2007, 7th–11th lem when there is no clearly defined point where topography and
52 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60

vegetation change abruptly over small distances. When a barrier has re- respectively. 15 of the 18 profiles were amenable to further analysis.
cently been scarped by storms this line is sharp and clearly identifiable One profile from Brancaster Bay crossed the heavily defended section
from aerial photography. However, as the barrier recovers, local pat- of shoreline at the clubhouse of the Royal West Norfolk Golf Club and
terns of erosion and deposition, and the accompanying re-vegetation a further profile from the eastern end of Brancaster Bay was not includ-
of the shoreline profile, makes it more difficult to define the shoreline ed because it crosses a beach with no landward dune or barrier. Similar-
position. To deal with this problem, EA ground survey cross-shore pro- ly one of the profiles from Holkham Bay (the furthest eastward) did not
file data were used to define the maximum break-of-slope. These points cross the dunefield and was discounted.
were then plotted in their exact x-y locations on the closest aerial pho- In order to assess the impact of the four identified major storms, the
tograph to the survey date. The image was inspected for degree of veg- summer cross-shore profiles were plotted for each year between 2006
etation cover and pixel colour change, with the point of maximum and 2014. The summer-to-summer plots were used to calculate change
change taken to be the shoreline position. Errors with this method in the crest location for each year. This was not possible for all profiles,
arise from difficulties in identifying correctly the colour change but are as defining crest location for washover deposits (e.g. EA profile refer-
estimated to be no N0.75 m horizontal distance for aerial photographs ence number N017) and for gently-sloping sand dunes (e.g. N021)
with a pixel size of 0.25 m (e.g. 2006 vertical aerial photography) and where the surface is gently undulating, is problematic. Hence the num-
0.60 m for a pixel size 0.20 m (e.g. 2014 imagery). Geo-rectified UK EA ber of cross-shore profiles included in this analysis was a total of 11
vertical aerial photographs for 1992 and 2014 were used to digitise from the 15 available profiles.
the shoreline position within ArcMap 10.1, using the above methodolo-
gy. The Digital Shoreline Analysis System (DSAS; Thieler et al., 2009) 3.3. Annual cross-shore profile analysis 2008–2015
was then applied, with shore-normal transects spaced at 10 m along-
shore intervals, to find the Net Shoreline Movement (NSM; m) over The assessment of shoreline recovery was carried out for two dis-
this 22-year time period. tinct phases of no storm activity. Between summer 2008 and summer
In order to investigate the impact of major storms on shoreline posi- 2013 there was recovery which we assessed using the EA cross-shore
tion in the latter part of the decadal record, shorelines were then profiles and by carrying out a DSAS assessment of change. A second
digitised from the summer 2006, 2007, 2008, 2013 and 2014 imagery. phase of recovery was also assessed following the 5th December 2013
Defining the shoreline as the crest of the barrier was straightforward storm, again using the EA cross-shore profiles. In these recovery assess-
and clear for the summer 2014 photograph because there was a sharp ments we not only looked at the shoreline position (as defined by the
break of slope resulting from barrier cutting during the December barrier crest) but also assessed in detail morphological changes taking
2013 storm, and Real Time Kinematic (RTK) surveys were undertaken place upon the beach in the intertidal region of the profile. The tidal
along the barrier immediately after the storm. The aerial photographs range is slightly different for each of the study locations, involving an
from summer 2006, summer 2007 and summer 2008 had many sec- east-west gradient in tidal range along the North Norfolk Coast
tions (covering around 60% of the total shoreline length) where the bar- (Admiralty Tide Tables, 2010). This gradient generates higher extreme
rier crest could be defined clearly with sharp changes in adjacent pixel water levels in the west of around + 1.5 m at Hunstanton compared
colour from well vegetated to non-vegetated surfaces. Using the Digital with Cromer (East Anglian Coastal Group, 2010). Data reported in the
Shoreline Analysis System (Thieler et al., 2009), shore-normal transects Shoreline Management Plan for Mean High Water Springs (MHWS)
were cast from a baseline at 10 m alongshore spacing, allowing assess- along the coast at Hunstanton, Burnham, Wells, Blakeney and Cromer
ment of shoreline change at a very high level of spatial densification. were also used to fit an Ordinary Least Squares Regression (r2 = 0.95)
Brancaster Bay, Scolt Head Island and Holkham Bay included a total of which was then used to find the level of MHWS at each EA cross-
518, 591 and 613 transects, respectively. Some transects were removed shore profile location. EA cross-shore profiles were used to plot the pro-
from the analysis, where stream outlets prevented clear digitising of the file section from the elevation between MHWS and 0.0 m ODN, and then
shoreline, where overwash deposits masked the shoreline position, from 0.0 m ODN extending seaward as far as the data were available.
where there were artificial structures defending the shoreline, or Finally, we assessed morphological changes in the EA cross-shore
where the shoreline was so sinuous that transects crossed the shoreline profiles below 0.0 m ODN by plotting summer survey data from sum-
at a highly oblique angle. For the remaining 299, 511 and 328 transects mer 2008 through to summer 2013 (just before the 5th December
at Brancaster Bay, Scolt Head Island and Holkham respectively, the 2013 storm) and then looked in greater detail at the region below
alongshore End Point Rate (EPR, m a− 1) for the years 2006–2007, 0.0 m ODN for the 6 monthly surveys (winter and summer) for the pe-
2007–2008 and 2013–2014 was then calculated for each of the three riod summer 2011 to summer 2015.
study areas. This statistical measure is generated from the distance be-
tween the two shorelines at each transect, divided by the time interval,
to return the average annual retreat over the period. It is therefore inde- 4. Results
pendent of the time period used for the assessment.
To assess shoreline recovery in the period without major storms, the 4.1. Decadal shoreline change, 1992–2014
digitised shorelines from summer 2008 and summer 2013 were used to
quantify shoreline change, again using the End Point Rate statistic from Over the longest timescale (1992–2014) changes in shoreline posi-
DSAS with an alongshore transect spacing of 10 m. tion are shown using the Net Shoreline Movement (m) statistic from
DSAS (Fig. 3). Also shown are the calculations of change using the 15
3.2. Decadal cross-shore (at-a-point) shoreline change, 1992–2014 EA cross-shore profiles; it is clear that the at-a-point results from the
cross-shore profiles verify the results from the alongshore DSAS analy-
The alongshore analysis was validated by the use of the EA cross- sis. For Brancaster Bay, the maximum shoreline retreat was 44.97
shore profile data (Fig. 1C). Positional accuracy with this method is to (±2.30) m. In places there was some moderate shoreline advance but
within ±20 mm horizontally and ±30 mm vertically (Lee, 2008). The the average Net Shoreline Movement along this frontage was 17.42
most seaward located sharp break-in-slope occurring at an elevation (±0.87) m of retreat. At Scolt Head Island, there was less alongshore
above 4 m ODN was defined as the crest of the barrier. The difference variability (Fig. 3), apart from transects from the advancing western
in location of this crest position between 1992 and 2014 was used as end of the barrier (0–80 m chainage; note the large advance between
the metric for barrier migration. EA cross-shore profile spacing is 1 km these transects which defines the considerable westward extension in
alongshore; thus a total of 6, 5 and 7 cross-shore profiles were available the barrier over this period). As the system is behaving differently at
for the Brancaster Bay, Scolt Head Island and Holkham Bay study areas this location, these transects were removed from the calculation of
 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60 53

Fig. 3. A) Shoreline positions in 1992 (blue in online version/dark grey in printed version) and 2014 (red/light grey) for A) Brancaster Bay; B) Scolt Head Island and C) Holkham Bay. The
recurved laterals first identified by Steers (1934) and providing evidence of the episodic nature of barrier development can clearly be seen in B). D) Net Shoreline Movement (m) for
Brancaster Bay (blue in online version/black in printed version), Scolt Head Island (red/light grey) and Holkham Bay (green/dashed) for the period 1992–2014 (negative values =
shoreline retreat, positive values = shoreline advance). Gaps in the alongshore plots are due to the occurrence of tidal inlets (Brancaster Bay); shoreline washover deposits (Scolt
Head Island); or to the ends of the dunes where high shoreline curvature leads to transects being cast at highly oblique angles to the main shoreline trend (Holkham Bay). Symbols
indicate changes in barrier crest position between 1992 and 2014 found using the EA cross-shore profile analysis.

average rates of shoreline change so we could gain a clearer picture of 4.2. Event scale shoreline change, 2008–2014
the retreat of the barrier. Maximum Net Shoreline Movement was
59.52 (±2.98) m landward at the eastern end of the barrier. Average Typical cross-shore profiles for Brancaster Bay (EA profile reference
Net Shoreline Movement was 18.48 (± 0.92) m landward, slightly number N008), Scolt Head Island (N015) and Holkham Bay (N024)
higher than the rate recorded for Brancaster Bay. For Holkham Bay, are shown in Fig. 4 for each summer from 2008 to 2014. Variations in re-
the average Net Shoreline Movement between 1992 and 2014 was sponse are evident. The crest of the barrier at Brancaster Bay showed a
0.69 (±0.03) m of retreat, far less than at either Scolt Head Island or progressive inland movement between 5th September 2006 and 30th
Brancaster Bay. This figure, however, disguises a high level of variability August 2007, between 30th August 2007 and 11th September 2008, as
in shoreline behaviour at this location. Both the western and the eastern well as between 4th September 2013 and 9th September 2014. Between
ends of the embayment show a movement landward, with the maxi- 11th September 2008 and 4th September 2013, the position of the bar-
mum retreat at the western end of 70.17 (± 3.51) m and at eastern rier crest changed very little with respect to the shoreline. However,
end of 96.23 (±4.81) m. A 4.5 km stretch in the central sector of this some foredune recovery to seaward of the crest was recorded in this
frontage showed a highly variable state of shoreline advance, reaching time period and the crest elevation increased from 5.67 to 6.89 m
95.51 (±4.78) m at the point of maximum seaward advance (Fig. 3). ODN. At Scolt Head Island, the barrier was progressively reset landward
54 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60

Fig. 4. Typical summer cross-shore profiles in 2006 (blue in online version/black in printed version), 2007 (green/black), 2008 (orange/black), 2013 (grey/black) and 2014 (red/black) at
A) Brancaster Bay; B) Scolt Head Island; and C) Holkham Bay. Note changes to barrier crest location and elevation changes. The years between 2008 and 2013 are shown as grey stippled
lines.

during the periods of major storm activity and, as at Brancaster Bay, the The changing position of the barrier crest for each analysed cross-
position of the barrier crest changed very little in the intervening years shore profile along the Superfrontage between summer 2006 and sum-
(September 2008 to September 2013). Finally for Holkham Bay, while mer 2014 (Fig. 5) shows that periods of retreat coincide with winters
shoreline position was also relocated inland during the stormy periods with high magnitude storms. Previous research has constrained the re-
this dynamic was accompanied by clear shoreline advance between treat to within 6-monthly intervals and shown for Scolt Head Island that
September 2008 and September 2013. There was also evidence of re- the 2006–2007 retreat happened under the 17th–19th March 2007
covery in the foredunes between 2008 and 2013 taking place at event (no storm surge but the coincidence of high tides and large
Brancaster Bay and, more evidently, at Holkham Bay but not at Scolt waves) and that the 2007–2008 retreat accompanied a significant
Head Island. storm surge on 7th–8th November 2007 (Brooks et al., 2016). This is

Fig. 5. Change in barrier crest position between summer 2006 and summer 2014 from analysis of 11 EA cross-shore profiles. Note high retreat rates at individual profiles during 2006–
2007, 2007–2008 and particularly 2013–2014 compared with limited retreat and advance in the period 2008–2013.
 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60 55

further supported here by cross-shore profiles from Brancaster Bay and A

Holkham Bay where the 6-monthly retreat in the barrier crest at each
alongshore location took place either in the period 24th January–30th
August 2007 (22nd January–11th September 2007 for Holkham) or in
the period 30th August 2007–8th February 2008 (11th September
2007–5th January 2008 for Holkham). Further, for 2013–2014, shore-
line displacement took place in the period 4th September 2013–12th
March 2014 (22nd August 2013–28th February 2014 for Holkham)
and not in the other half of the year. As previously, field RTK surveys
showed that the cliffed edge of the barrier was more or less coincident
with the 2014 shoreline from aerial photography and set back from
the 2013 shoreline. This fixes the shoreline movement in the same 6-
monthly period as the storm of 5th December 2013. B
Using the digitised shorelines from 2006, 2007, 2008, 2013 and 2014
enabled the End Point Rate (EPR; m a−1) to be calculated for each time
period (Fig. 6). Alongshore variability is evident for all storms. A consid-
erable impact is associated with the 5th December 2013 storm, embed-
ded within the 2013–2014 retreat rates, since our RTK field surveys
testify to the fact that the annually averaged shoreline change was
achieved by just this one large event. Brancaster Bay, Scolt Head Island
and Holkham Bay all had very large inland translations of the shoreline
during this storm, with averages of 4.29 (±0.22), 4.81 (±0.24) and 7.36
(±0.97) m, respectively. By contrast, the EPR for the period between
summer 2008 and summer 2013 shows a very limited change in shore-
line position. At Brancaster Bay there was an average net movement C
seaward of just 0.29 (± 0.02) m per year (1.45 (± 0.07) m total), for
Scolt Head Island this was even less at 0.12 (±0.01) m (0.60 (±0.03)
m total) whereas at Holkham the shoreline position advanced at an av-
erage EPR of 1.54 (±0.08) m. This implies a Net Shoreline Movement
seaward by 7.79 (±0.44) m at Holkham Bay, slightly more than the sub-
sequent average retreat during the storm of 5th December 2013. While
the barriers at Brancaster Bay and Scolt Head Island appear to be being
progressively set further inland (consistent with the long-term behav-
iour shown in Fig. 3), the behaviour at Holkham Bay is rather different,
demonstrating resilience and recovery between storms and even show-
ing recovery towards an ever further seaward location in this period.
D
4.3. Cross-shore profile analysis, 2008–2013

Annual cross-shore profiles (2008–2013) between 0.0 m ODN and

MHWS for a representative cross-shore profile location in each of the
study areas are shown in Fig. 7. The frontage at Scolt Head Island,
where the barrier is retreating in its central/eastern section, shows no
inter-storm recovery. The shoreface gradient is steep (gradient =
2.44°) and narrow (60–100 m) between MHW and MSL (= 0.0 m
ODN). By contrast, the shoreface is of lower slope at Brancaster Bay
and Holkham Bay (0.91°–0.82°), with narrow beaches (190 m) at
Brancaster Bay compared to Holkham Bay (320 m) (Fig. 8A). Within
the intertidal zone, the cross-shore profiles reveal the existence of inter-
tidal bars at Holkham Bay, not so clearly evident in cross-shore profiles
Fig. 6. End Point Rate (m a−1) for the barrier coast for the storm periods 2006–2007 (blue
from Brancaster Bay or Scolt Head Island (Fig. 8B).
in online version/black in printed version), 2007–2008 (pink/light grey), and 2013–2014
The North Norfolk bars closely resemble those described from Faro, (orange/dashed) for A) Brancaster Bay; B) Scolt Head Island; and C) Holkham Bay. Also
southern Portugal (Vousdoukas et al., 2012) and particularly the bars shown are the cross shore profile locations (used to constrain the digitising of the
seaward of the Skallingen barrier, Danish North Sea coast (Aagaard et shoreline for earlier periods) and alongshore extents of reliable RTK field survey data
al., 2004). Here we follow the Danish terminology of an offshore inner (used to constrain the digitising for the 2014 shoreline). D) End Point Rate (m a−1) for
the period summer 2008 to summer 2013, a period of no storm activity, along the
bar and an inshore intertidal bar (Fig. 9A). The onshore movement of barrier for Brancaster Bay (blue/black), Scolt Head Island (red/light grey) and Holkham
the inner bar to become the intertidal bar between September 2006 Bay (green/dashed).
and September 2013 involved 500 m of inland translation, at a migra-
tion rate of ca. 70 m a−1. The intertidal bar migrated 200 m onshore be-
tween 2008 and 2013, occupying progressively higher elevations within The most recent movement of the intertidal bar at 6-monthly inter-
the tidal frame over this period (Fig. 9B). This intertidal bar migration vals is shown in Fig. 10 for Holkham Bay. Between 2012 and 2015, the
took place at a rate of 40 m a−1, a similar value to that found for inter- onshore migration was of the order of 150 m (i.e. 50 m a−1) and eleva-
tidal bars at Skallingen (Aagaard et al., 2004). Thus, it appear to take tion gain was of the order of 1 m. These bars migrate, gain elevation and
around 5 years for the inner bar to migrate from below MLWS to an in- become progressively exposed to a wider range of complex sediment
tertidal position and then a further 5 years for this new intertidal bar to exchange processes (see Jackson et al., 2016 for intertidal bar behaviour
move to an elevation of +1 m ODN. on a shorter timescale). Interestingly, the severe 5th December 2013
56 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60

Fig. 7. Cross-shore profiles between 0 m ODN and MHWS for 2008 and 2013 for A) Brancaster Bay (at EA cross-shore profile N008); B) Scolt Head Island (N015); and C) Holkham Bay
(N024).

storm surge event (the highest magnitude event of its type in the last still water levels (whether through surge mechanisms or because they
60 years) appears to have played only a limited role in onshore intertid- coincide with high spring tides or both) combined with large onshore
al bar migration. The year to year onshore movement seems very con- waves. These storm impacts are, however, separated by spatially-vari-
sistent with no discernible difference in the intertidal bar migration able barrier responses during non-storm phases. These responses can
rate between summer 2013 and winter 2014 than that indicated for be classed as either stasis (Scolt Head Island), partial recovery
any other 6-month interval. The 2015 summer bar also appears to (Brancaster Bay) or full recovery with shoreline advance (Holkham
have developed a trough offshore from the main bar crest, while the Bay). Alongshore variability in response is the result of the interaction
continued upward direction of travel has taken this bar to above 0 m between shoreface bathymetry, sediment availability and inshore hy-
ODN. drodynamics (including wave energy dissipation, radiation stress gradi-
ents, longshore currents and set-up). These controls operate at a series
5. Discussion of spatial and temporal scales.
At the macro-scale, List et al. (2006) found strong ‘mirroring’ behav-
Shoreline change along the entire 15 km Superfrontage 2 of the iour in post-storm retreat and recovery along the uninterrupted, sandy
North Norfolk coast involves periods of clear landward shoreline trans- outer barriers of the United States east coast, with storm erosion pockets
lation under individual high magnitude storms characterised by high being rapidly countered by accretion in the same locations to restore the
 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60 57

Fig. 8. Typical cross-shore profiles for the three study areas between A) Supra-tidal environments to ca. MLWS; and B) MHWS to MLWS in detail. Note: these cross-shore profiles are the
same as those reported earlier for the three study sites. Dashed line is Brancaster Bay, solid line is Scolt Head Island and irregular dashed line is Holkham Bay.

pre-storm shoreline. Comparison with this study is not straightforward along the entire beach front between Blakeney Point and Brancaster.
in that our study documents changes in barrier margins (i.e. upper fore- At Scolt Head Island drift rates are of the order of 190,000 m3 a−1. How-
shore) whereas List et al. (2006) took Mean High Water as their mea- ever, results from offshore sediment type mapping (Evans et al., 1998)
sure of shoreline position. Even so, there is no clear evidence that this have suggested a contrasting west-east direction for sediment transport
section of the North Norfolk coast exhibits similar erosion/deposition, below the −7 m isobath, offshore from the steep beach face onto the
‘reversing hotspot’ behaviour. This is due to three factors. First, the Burnham Flats (reported in HR Wallingford, 2002 appendix 11; East
long NE USA shorelines (45 km at Cape Cod and 113 km on the Outer Anglian Coastal Group, 2010). There is no reason to discount an alterna-
Banks, North Carolina) allow the capture of multiple hotspots along- tive model for the beach face that suggests that, at Scolt Head Island,
shore which is not possible along the restricted 17 km North Norfolk sediment movement is also west to east, involving easterly-migrating
frontage. Secondly, it is clear from List et al. (2006) that the erosion/ac- sand waves moving over an ebb tide delta in the Brancaster Harbour
cretion cycle is completed within 5–12 days of storm impact, a much Channel to then weld onto the western end of the barrier. It has also
finer temporal resolution than the longer-term shoreline change been suggested that late nineteenth century land reclamation might
discussed here. And thirdly, it is likely that the greater spatial complex- be responsible for an easier passage of the sand waves across the har-
ity and sedimentary variability of the North Norfolk coastline - from the bour channel due to a reduction in the tidal prism after this time
presence of mixed sand and gravel barriers, inter-barrier tidal inlets (Royal Haskoning and Pethick, 2003; Brooks et al., 2016). Thus in both
(with associated flood and ebb tide deltas) and fine sediment exchanges the beach face and below the −7 m isobath there are likely to be along-
between mudflats and vegetated marsh platforms - forces a more com- shore sediment exchanges which potentially result in sediment being
plex retreat-recovery signal. supplied to the more easterly locations along this frontage.
These observations feed into a wider discussion of nearshore sedi- On way of reconciling these different models is through the possibil-
ment exchange along this barrier coastline and engage with wider de- ity of differing barrier dynamics between ‘calm’ and ‘storm’ periods. This
bates on how shoreline orientation relative to the incident wave is particularly apparent at Scolt Head Island where the framework of the
climate is a major factor in storm retreat-recovery interactions island consists of both a main E-W spine and a series of irregularly
(Ashton and Murray, 2006a, 2006b). Since the earliest mapping and in- spaced, broadly N-S trending ‘laterals’. It was J.A. Steers (1960) who
terpretations of Steers (1934, 1960), there have been many conceptual, first argued that the dominant behaviour during calm, inter-storm pe-
sedimentological and numerical modelling studies of sediment trans- riods is a westward extension of the island, while during storm events
port that suggest a dominant westward transport between Sheringham this westward growth is halted and the barrier driven landward. In
(30 km east of Holkham Bay) towards Hunstanton in the west (e.g. the periods of island extension, the laterals, generated by sediment
Steers, 1927; McCave, 1978; Vincent, 1979; Oynett and Simmonds, transport under strong wave refraction, show a characteristic smooth
1983). The Shoreline Management Plan (East Anglian Coastal Group, ‘recurve’ geometry. However, the high angle of intersection between
2010) shows a consistent net sediment transport direction (270°; W) the older, more easterly laterals and the main island spine provides
58 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60

Fig. 9. Representative cross shore profile (EA reference N024) for Holkham Bay showing the locations of A) the Inner and Intertidal Bars in September 2006 (dashed line) and September
2013 (solid line) and B) the Intertidal Bar position each summer 2008–2013 (Note appearance of Inner Bar in profiles for 2010/11, 2012 and 2013).

evidence for the truncation of these fixed laterals from the long-term This regime has the following components: under calm conditions,
landward retreat of the barrier. These arguments are confirmed by Fig. westward sediment transport from the drift divide located to the east
6D which shows that in calm periods, the barrier is characterised by ero- at Sheringham; eastward transport in the beach zone by migrating
sion at its eastern end but deposition at its western terminus. While sand bars when the dominant wave direction is from the west; and east-
storm impacts at Scolt generate Steers' (1960) record of landward ward movement of sediment below the − 7 m isobath across the
translation (Fig. 6B), the long-term behaviour of the barrier reflects Burnham Flats and emplaced during, and immediately after, storms
the fact that these episodes are short-lived perturbations in a general from barrier retreat at Brancaster Bay and Scolt Head Island, the latter
history of eastern erosion, along-barrier sediment transport (which being especially important as a sediment source to maintain such
smooths out the fine detail of storm retreat (Fig. 6B) and westward ex- transports.
tension (Fig. 3B). At the meso-scale, Masselink and Short (1993) categorise the fore-
Finally at this scale, these observations and models suggest that the shore into four zones: 1) the subtidal zone influenced by shoaling
recovery taking place at Holkham Bay is sourced by the three-fold sed- wave processes; 2) the lower intertidal zone characterised by surf and
iment transport regime that makes Holkham Bay a dynamic sediment shoaling wave processes; 3) the mid-intertidal zone with swash and
sink (or zone of ‘positive diffusivity’ (Ashton and Murray, 2006b)). surf; and 4) the upper intertidal zone with swash and aeolian processes.
The width of each zone varies with tidal range and gradient of the
shoreface. Sediment exchanges take place between these zones, ulti-
mately leading to sediment being made available for aeolian transport
to foredunes and established dunes. Intertidal bars driven onshore
(Aagaard et al., 2006; Masselink et al., 2006) ultimately reach elevations
that permit drying times conducive to consequential aeolian transport.
Moisture content (precipitation and tidally influenced) has been
shown to be the primary control on aeolian transport as, through varia-
tions in shear strength, it directly affects the fetch length over which the
wind can entrain sediment (Sherman et al., 1998; Yang and
Davidson-Arnott, 2005; Delgado-Fernandez, 2010, 2011). Drying op-
portunity is strongly related to beach gradient through the extent and
persistence of the seepage face associated with fluctuations in beach
water-table (Jackson and Nordstrom, 1998). Hence, even allowing for
Fig. 10. Vertical and lateral intertidal bar migration at Holkham Bay between six-monthly
winter (w) and summer (s) surveys, 2012–2015 for a representative cross-shore profile
sediment size differences, there is far less drying potential on steep, nar-
(EA reference number N024). The time period affected by the 5th December 2013 storm row reflective beaches than on wide, dissipative beaches. At Scolt Head
surge is shown by the uppermost horizontal arrow. Island, the offshore gradients are steep and there is no evidence in the
 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60 59

cross-shore profiles to suggest the existence of onshore migrating inner Holkham Bay where large coastal tracts show clear long-term net shore-
bars. As a result there is limited sand supply and limited potential for de- line advance. As well as changes to the actual shoreline position, the
velopment of an upper beach flat that can generate a drying sand source area seaward of the barrier crest also shows the way in which the shore-
for aeolian entrainment and transport. Here sand dune recovery in calm line is evolving. While at Brancaster Bay foredunes developed between
periods is largely non-existent. At Brancaster Bay, there is little evidence 2008 and 2013, no such change was observed at Scolt Head Island. Even
for the presence of significant inner or intertidal bars. In addition, due to larger foredunes than those at Brancaster Bay developed in this period
higher tidal levels (0.54 m higher at MHWS than further east), the upper at Holkham Bay. Within a context of regional sediment transport, vari-
beach at Brancaster Bay inundates more frequently than at Holkham; ously westwards and eastwards, explanations for these differences can
the rising tide here covers an additional 28.65 m in the cross-shore ori- be found in the beach and shoreface gradient in the different study
entation. Between 2008 and 2013, foredune development seaward of areas (steeper at Scolt Head Island) and differences in the tidal range
the main dune barrier at Brancaster Bay resulted in vertical elevation (smallest at Holkham), resulting in differing inundation regimes along
gains of ca. 1 m on shore-normal profiles and lateral seaward extension this coastline. In addition, migratory subtidal and intertidal bars play a
of the upper foreshore by ca. 5 m. It can be concluded, therefore, that key role in shoreline recovery as these dynamics provide a mechanism
foredune recovery in calm periods does take place at Brancaster Bay for the onshore movement of sediments that ultimately provide source
but not at a rate that keeps pace with the long-term (1992–2014) land- materials for foredune construction.
ward shoreline displacement that results from storm impacts. Finally, at Temporally, barrier recovery potential from storm impacts under-
Holkham Bay, the combination of progressive onshore migration of pins the future survival of coastal landscapes, ecosystems, human com-
inner and intertidal bars, leading to extensive areas of drying sand in munities and infrastructure on high-energy coasts. The concern over
the upper intertidal zone, and exposure of these areas to onshore and future sea-level rise and changes to the tracks and magnitudes of
alongshore winds provides an ample sand supply to re-build foredunes extra-tropical storms and storm sequences (Vousdoukas et al., 2012;
and allow their seaward expansion in calm inter-storm periods. Here Castelle et al., 2015; Masselink et al., 2015), makes the understanding
between 2008 and 2013, foredunes showed vertical growth of 1.5 m of shoreline recovery potential and resilience between storms an impor-
with seaward advance of ca. 10 m. Furthermore, since the December tant part of both research and coastal management agendas
2013 storm, and unlike at the other two study areas, the shoreline at (Woodworth et al., 2009; Horsburgh and Lowe, 2013; Wahl et al.,
Holkham Bay has undergone significant foredune development linked 2013). Spatially, the differential recovery capacity recorded here at
to onshore intertidal bar migration. three locations found over a total distance of 15 km shows the impor-
The primary controls on shoreline recovery operating along this tance of topographic and bathymetric setting – and their impact on
17 km frontage are, therefore, the presence of migratory nearshore coastal hydrodynamics and sediment transport - in determining system
inner and intertidal bars, low shoreface gradients, sufficient drying response and differential shoreline recovery. Taken together, a better
times for sand entrainment on the upper beach flat and aeolian trans- appreciation of the time-space dynamics of barred shorelines, their
port in winds above thresholds of around 8 m s− 1. Similar controls beach faces and dunefields, provides important input into the shoreline
have been observed elsewhere around the southern North Sea management planning process, particularly where the use of scarce re-
(Aagaard et al., 2004; Anthony et al., 2006; Anthony, 2013). Finally, fol- sources for management needs to be effectively prioritized.
lowing Houser (2009), it is vital to understand how transport and sup-
ply of sediment are synchronised, involving the spatial and temporal Acknowledgements
coupling of nearshore and aeolian processes. Strong recovery is only
possible with synchroneity between transport and supply. Between This research was undertaken while SMB held a Research Fellowship
storms when drying is at a maximum (large supply), winds tend to be awarded by The Leverhulme Trust. Information on alongshore varia-
weak (transport-limited), while during storms, winds are strong tions in water level was collected under EU FP7 Collaborative Project
(large transport potential) but the backshore readily becomes wet, lim- (grant agreement no: 603458) ‘Resilience-increasing Strategies for
iting the sediment supply (supply-limited). In the presence of a strong Coasts – toolkit’ (http://www.risckit.eu). The research is also a contribu-
sand supply, the building of dunes supplied by sand from the upper tion to UK NERC BLUECoast Project (NE/N015924/1). The authors grate-
beach flat is highly dependent on this synchronisation. When strong fully acknowledge the UK Environment Agency (in particular David
winds above sand entrainment thresholds coincide with falling tides, Welsh of the Anglian Shoreline Management Group) for the supply of
and low still water levels produce prolonged beach exposure and effi- cross-shore profile data and aerial photography. Weybourne weather
cient drying, there is potential for dune building. More data are required data were supplied by the UK Meteorological Office through the British
on the interactions between wind strength and direction and fluctuat- Atmospheric Data Centre. We would like to thank Dr Grant Forster
ing moisture content in the beach on diurnal, monthly and annual time- (NCAS Research Scientist and Co-ordinator of the Weybourne Atmo-
scales and the implications for beach-dune sediment exchanges. There spheric Observatory Centre for Ocean and Atmospheric Sciences
is also a need to explore these relationships on inter-annual timescales (COAS)) for very prompt supply of data. Sarah Henderson of Holkham
and, ultimately, how they relate to decadal-scale fluctuations in storm Estate facilitated access to the Holkham dunes. We are grateful for the
impacts and variable inter-storm periods. careful review of this paper by Professor Eli Lazarus of the University
of Southampton and to an anonymous referee, both of whom greatly
6. Conclusions improved this manuscript.

For the barrier coast of North Norfolk, storms have a differential im- References
pact, even over relatively short alongshore distances, depending on
Aagaard, T., Davidson-Arnott, R., Greenwood, B., Nielsen, J., 2004. Sediment supply from
shoreline setting. Landward retreat during storms can be up to 15 m shoreface to dunes: linking sediment transport measurements and longterm mor-
in a single event, depending on the morphological and sedimentological phological evolution. Geomorphology 60, 205–224.
mobility of the barrier. Between large storms, this barrier coastline is Aagaard, T., Hughes, M., Møller-Sørensen, R., Andersen, S., 2006. Hydrodynamics and sed-
iment fluxes across an onshore migrating intertidal bar. J. Coast. Res. 22, 247–259.
characterised predominantly by stasis at Scolt Head Island, partial re-
Ablain, M., Cazenave, A., Guinehut, S., Valladeau, G., 2009. A new assessment of global
covery at Brancaster Bay and full recovery at Holkham Bay. At the latter mean sea level from altimeters highlights a reduction of global slope from 2005 to
site there was considerable shoreline advance in the period 2008–2013. 2008 in agreement with in-situ measurements. Ocean Sci. 5, 193–201.
This combination of response and recovery has resulted in long-term Admiralty Tide Tables, 2010. United Kingdom and Ireland. NP201-01. Hydrographer of
the Navy vol. 1. UK Hydrographic Office, Taunton.
shoreline retreat (1992–2013) of 17.42 (± 0.87) m at Brancaster Bay Andrews, J.E., Boomer, I., Bailiff, I., Balson, P., Bristow, C., Chroston, P.N., Funnell, B.M.,
and 18.48 (±0.92) m at Scolt Head Island but only 0.69 (±0.04) m at Harwood, G.M., Jones, R.W., Maher, B.A., Shimmield, G., 2000. Sedimentary evolution
60 S.M. Brooks et al. / Geomorphology 283 (2017) 48–60

of the north Norfolk barrier coastline in the context of Holocene sea-level change. In: Jackson, N.L., Nordstrom, K.L., 1998. Aeolian transport of sediment on a beach during and
Shennan, I., Andrews, J. (Eds.), Holocene Land–Ocean Interaction and Environmental after rainfall, Wildwood, NJ, USA. Geomorphology 22, 151–157.
Change around the North SeaSpecial Publication 166. Geological Society, London: Jackson, D.W.T., Cooper, J.A.G., O'Connor, M., Guisado-Pintado, E., Loureiro, C., Anfuso, G.,
pp. 219–251 http://dx.doi.org/10.1144/GSL.SP.2000.166.01.12. 2016. Field measurements of intertidal bar evolution on a high-energy beach system.
Anthony, E.J., 2013. Storms, shoreface morphodynamics, sand supply, and the accretion Earth Surf. Process. Landf. http://dx.doi.org/10.1002/esp.3920.
and erosion of coastal dune barriers in the southern North Sea. Geomorphology Lee, E.M., 2008. Coastal cliff behaviour: observations on the relationship between beach
199, 8–21. levels and recession rates. Geomorphology 101:558–571. http://dx.doi.org/10.1016/
Anthony, E.J., Vanhée, S., Ruz, M.H., 2006. Short-term beach-dune sand budgets on the j.geomorph.2008.02.010.
North Sea coast of France: sand supply from shoreface to dunes and the role of List, J.H., Farris, A.S., Sullivan, C., 2006. Reversing storm hotspots on sandy beaches: spatial
wind and fetch. Geomorphology 81, 316–329. and temporal characteristics. Mar. Geol. 226 (3), 261–279.
Ashton, A.D., Murray, A.B., 2006a. High-angle wave instability and emergent shoreline Masselink, G., Short, A.D., 1993. The effect of tide range on beach morphodynamics and
shapes: 1. Modeling of sand waves, flying spits, and capes. J. Geophys. Res. Earth morphology: a conceptual beach model. J. Coast. Res. 9, 785–800.
Surf. 111 (F4). http://dx.doi.org/10.1029/2005JF000422. Masselink, G., Kroon, A., Davidson-Arnott, R.G.D., 2006. Morphodynamics of intertidal
Ashton, A.D., Murray, A.B., 2006b. High-angle wave instability and emergent shoreline bars in wave-dominated coastal settings – a review. Geomorphology 73, 33–49.
shapes: 2. Wave climate analysis and comparisons to nature. J. Geophys. Res. Earth Masselink, G., Scott, T., Poate, T., Russell, P., Davidson, M., Conley, D., 2015. The extreme
Surf. 111 (F4). http://dx.doi.org/10.1029/2005JF000423. 2013/2014 winter storms: hydrodynamic forcing and coastal response along the
Betts, N.L., Orford, J.D., White, D., Graham, C.J., 2004. Storminess and surges in the South- southwest coast of England. Earth Surf. Process. Landf. 41, 378–391.
Western approaches of the eastern North Atlantic: the synoptic climatology of recent McCave, I.N., 1978. Grain-size trends and transport along beaches: example from eastern
extreme coastal storms. Mar. Geol. 210, 227–246. England. Mar. Geol. 28, M43–M51.
Brooks, S.M., Spencer, T., McIvor, A., Möller, I., 2016. Reconstructing and understanding Michener, W.K., Blood, E.R., Bildstein, K.L., Brinson, M.M., Gardner, L.R., 1997. Climate
the impacts of storms and surges, southern North Sea. Earth Surf. Process. Landf. change, hurricanes and tropical storms, and rising sea level in coastal wetlands.
41:855–864. http://dx.doi.org/10.1002/esp.3905. Ecol. Appl. 7 (3), 770–801.
Castelle, B., Marieu, V., Bujan, S., Splinter, K.D., Robinet, A., Senechal, N., Ferreira, S., 2015. Moore, L.J., 2000. Shoreline mapping techniques. J. Coast. Res. 16, 111–124.
Impact of the winter 2013–2014 series of severe Western Europe storms on a double- Onyett, D., Simmonds, A., 1983. Beach transport and longshore transport. East Anglian
barred sandy coast: beach and dune erosion and megacusp embayments. Geomor- Coastal Research Programme Report No. 8. University of East Anglia, Norwich, U.K.
phology 238, 135–148. Sabatier, F., Anthony, E.J., Héquette, A., Suanez, S., Musereau, J., Ruz, M.H., Régnauld, H.,
Church, J.A., White, N.J., 2006. A 20th century acceleration in global sea-level rise. 2009. Morphodynamics of beach/dune systems: examples from the coast of France.
Geophys. Res. Lett. 33, L01602. Géomorphologie 2009 (1), 1–13.
Ciavola, P., Armaroli, C., Chiggiato, J., Valentini, A., Deserti, M., Perini, L., Luciani, P., 2007. Sallenger Jr., A.H., Stockdon, H.F., Fauver, L., Hansen, M., Thompson, D., Wright, C.W.,
Impact of storms along the coastline of Emilia-Romagna: the morphological signature Lillycrop, J., 2006. Hurricanes 2004: an overview of their characteristics and coastal
on the Ravenna coastline (Italy). J. Coast. Res. Spec. Issue 50, 540–544. change. Estuar. Coasts 29, 880–888.
Delgado-Fernandez, I., 2010. A review of the application of the fetch effect to modelling Sherman, D.J., Jackson, D.W.T., Namikas, S.L., Jinkang, W., 1998. Wind-blown sand on
sand supply to coastal foredunes. Aeolian Res. 2, 61–70. beaches: an evaluation of models. Geomorphology 22, 113–133.
Delgado-Fernandez, I., 2011. Meso-scale modelling of aeolian sediment input to coastal Spencer, T., Brooks, S.M., Möller, I., Evans, B.R., 2014. Where local matters: impacts of a
dunes. Geomorphology 130, 230–243. major North Sea storm surge. EOS Trans. Am. Geophys. Union 95:269–270. http://
Dissanayake, P., Brown, J., Karunarathna, H., 2014. Modelling storm-induced beach/dune dx.doi.org/10.1002/2014EO300002.
evolution: Sefton coast, Liverpool Bay, UK. Mar. Geol. 357, 225–242. Spencer, T., Brooks, S.M., Tempest, J.A., Moller, I., 2015. Southern North Sea storm surge
Earlie, C.S., Young, A.P., Masselink, G., Russell, P.E., 2015. Coastal cliff ground motions and event of 5 December 2013: water levels, waves and coastal impacts. Earth Sci. Rev.
response to extreme storm waves. Geophys. Res. Lett. 42 (3), 847–854. 146, 120–145.
East Anglian Coastal Group, 2010. North Norfolk shoreline management plan final plan Steers, J.A., 1927. The East Anglian coast. Geogr. J. 69, 24–43.
2010. http://www.eacg.org.uk/docs/smp5/the%20smp%20main%20report.pdf. Steers, J.A., 1934. Scolt Head Island. Geogr. J. 83, 479–494.
Environment Agency, 2014. Sea State Report Norfolk Year 3 and Summary for October Steers, J.A., 1960. Physiography and evolution: the physiography and evolution of Scolt
2006–September 2009 (RP039/N/2014). Shoreline Monitoring Group. Environment Head island. In: Steers, J.A. (Ed.), Scolt Head Island, second ed. Heffers, Cambridge,
Agency, Peterborough. pp. 12–66.
Evans, C.D.R., Crosby, A., Wingfield, R.T.R., James, J.W.C., Slater, M.P., Newsham, R., 1998. Stockdon, H.F., Sallenger Jr., A.H., List, J.H., Holman, R.A., 2002. Estimation of shoreline po-
Inshore seabed characterisation of selected sectors of the English coast. British Geo- sition and change using airborne topographic lidar data. J. Coastal Res. 18, 502–513.
logical Survey Technical Report WB 98/45. NERC, BGS, Keyworth, Nottingham. Stone, G.W., Liu, B., Pepper, D.A., Wang, P., 2004. The importance of extratropical and
Gerritsen, H., 2005. What happened in 1953? The Big Flood in the Netherlands in retro- tropical cyclones on the short-term evolution of barrier islands along the northern
spect. Philos. Trans. R. Soc. Lond. 363A:1271–1291. http://dx.doi.org/10.1098/rsta. Gulf of Mexico, USA. Mar. Geol. 210, 63–78.
2005.1568. Thieler, E.R., Himmelstoss, E.A., Zichichi, J.L., Ergul, A., 2009. Digital Shoreline Analysis Sys-
Hackney, C., Darby, S.E., Leyland, J., 2013. Modelling the response of soft cliffs to climate tem (DSAS) version 4.0 – an ArcGIS extension for calculating shoreline change. US
change: a statistical, process-response model using accumulated excess energy. Geo- Geological Survey Open-file Report, 2008–1278. US Geological Survey, Reston, VA.
morphology 187:108–121. http://dx.doi.org/10.1016/j.geomorph.2013.01.005. Vincent, C.E., 1979. Longshore sand transport rates — a simple model for the East Anglian
Haerens, P., Bolle, A., Trouw, K., Houthuys, R., 2012. Definition of storm thresholds for sig- coastline. Coast. Eng. 3, 113–136.
nificant morphological change of the sandy beaches along the Belgian coastline. Geo- Vousdoukas, M.I., Pedro, L., Almeida, M., Ferreira, O., 2012. Beach erosion and recovery
morphology 143–144, 107–117. during consecutive storms at a steep-sloping, meso-tidal beach. Earth Surf. Process.
Royal Haskoning, Pethick, J., 2003. North Norfolk Coast Coastal Habitat Management Plan Landf. 37, 583–593.
Final Report January 2003. Royal Haskoning, Peterborough. Wadey, M.P., Brown, J.M., Haigh, I.D., Dolphin, T., Wisse, P., 2015. Assessment and compar-
Horsburgh, K., Lowe, J., 2013. Impacts of climate change in sea level. Mar. Clim. Chang Im- ison of extreme sea levels and waves during the 2013/14 storm season in two UK
pacts Partnership: Sci. Rev.:27–33 http://dx.doi.org/10.14465/2013.arc04.027-033. coastal regions. Nat. Hazards Earth Syst. Sci. 15:2209–2225. http://dx.doi.org/10.
Houser, C., 2009. Synchronization of transport and supply in beach–dune interaction. 5194/nhess-15-2209-2015.
Prog. Phys. Geogr. 33, 733–746. Wahl, T., Haigh, I.D., Woodworth, P.L., Albrecht, F., Dillingh, D., Jensen, J., Nicholls, R.,
Houser, C., Hapke, C., Hamilton, S., 2008. Controls on coastal dune morphology, shoreline Weisse, R., Wöppelmann, G., 2013. Observed mean sea level changes around the
erosion and barrier island response to extreme storms. Geomorphology 100, North Sea coastline from 1800 to the present. Earth Sci. Rev. 124:51–67. http://dx.
223–240. doi.org/10.1016/j.earscirev.2013.05.003.
HR Wallingford, 2002. Southern North Sea Sediment Transport Study, Phase 2. Sediment Woodworth, P.L., Teferle, F.N., Bingley, R.M., Shennan, I., Williams, S.D.P., 2009. Trends in
Transport Report. HR Wallingford Report EX4526, Howbery Park, Wallingford, Ox- UK mean sea level revisited. Geophys. J. Int. 176:19–30. http://dx.doi.org/10.1111/j.
ford, UK. 1365-246X.2008.
Intergovernmental Panel on Climate Change (IPCC), 2013. Summary for policymakers. In: Yang, Y., Davidson-Arnott, R.G.D., 2005. Rapid measurement of surface moisture content
Stocker, T.F., et al. (Eds.), Climate Change 2013: The Physical Science Basis, Contribu- on a beach. J. Coast. Res. 21, 447–452.
tion of Working Group I to the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change. Cambridge University Press, Cambridge.