1. Introduction
The arcuate planform of the East Sussex coastline, between Shoreham-By-Sea and Beachy Head, is dominated by erosional landforms east of Brighton Marina, except at Newhaven Harbour, Seaford Beach and Cuckmere Haven. Cliffs and shore platforms are developed on a variety of Chalk formations (Photo 1), with other rock types making a contribution locally. West of Brighton Marina, the Chalk outcrop strikes inland, and unresistant Eocene strata and overlying Quaternary drift sediments form a low elevation coastline that forms the eastern extremity of the West Sussex coastal plain.
The lithology and structural characteristics of the Chalk are spatially variable, with the disposition of parting planes - especially master joints - accounting for the nearly vertical form of the cliffs between Roedean and Beachy Head. The more complex morphology and higher elevation of the east-facing cliffs of Beachy Head (Photo 2) reflect the fact that the coastline cuts across the chalk cuesta. Much of the cliffed shoreline east to Seaford Head is protected by seawalls and groynes; defences have been progressively extended and upgraded since the early to mid-nineteenth century to give security to property and infrastructure. The coastal hinterland between Shoreham-By-Sea and Seaford forms a major element of the south coast conurbation, much of which has been developed over the past 150 years.
Wave energy is high in comparison to much of the rest of the coastline of south-east and central southern England. This is the result of its orientation with respect to both dominant and prevalent waves. Maximum wave energy is experienced along the shoreline between Seaford and Beachy Heads; for this sector, the influence of rock structure and resistance is probably subsidiary to wave forces in determining rates of recession and planform evolution.
Beaches are dominantly made up of a gravel backshore and a sandy foreshore, but the latter is often absent or suppressed where shoreline platforms occupy most of the inter-tidal zone. Littoral transport of coarse sediment follows a west to east pathway, but with local and short period reversals. Actual rates of longshore movement are almost everywhere below potential rates, primarily due to insufficient supply of gravel from input sources. Consequently, many beaches have a history of volume loss, with cross-profile steepening also diagnostic of supply deficit. Throughout the nineteenth and much of the twentieth century this problem has been addressed through the use of control structures, particularly sets of long and short groynes. Since the mid-1980s, emphasis has moved to replenishment and recycling as means of long-term beach management, exemplified by sites such as Seaford Beach and Rottingdean.
The mouths of the Adur, Ouse and Cuckmere rivers interrupt the continuity of this coastline. Well into the historical period each was a natural estuary, albeit gradually modified by mid to late Holocene sedimentation. Strong west to east longshore transport created shingle spits that caused the eastwards migration of each of these river exits. It is probable that they are barrier spits, fed in part from offshore, with a history of growth, breaching, retreat and regrowth over the past two or three millennia. Starting in early medieval times, the entrances to the Adur and Ouse have been utilised as major regional ports. Since the late nineteenth century, the ports of Shoreham and Newhaven have had their access and entrance channels protected by massive breakwaters (to the west) and piers/training walls (to the east). Their navigation channels have also been progressively deepened and routinely dredged.
The effect of breakwater and piers at both sites has been to impede the longshore transfer of coarse sediment, with very substantial updrift accretion. However, it is apparent that some by-passing occurs, thus they represent partial, rather than absolute, barriers to transport. Brighton Marina (Photo 3) is another artificial structure with a potentially similar effect, but the evidence suggests that it occupies the site of a natural transport discontinuity. Sand and finer sediments (silt, clay) move within an inshore zone wider than the seaward projection of any of these structures, and they are therefore less affected than coarse (gravel-sized) sediment. The principal transport sub-cells of the western and central section of this coastline are therefore delimited by these artificial constraints on littoral sediment drift, viz: (i) Shoreham-By-Sea to Brighton Marina (West); and (ii) Brighton Marina (East) to Newhaven Western Breakwater. Further east, Seaford Bay is a well-defined sub-cell, with its eastern boundary determined by its large terminal groyne and by Seaford Head (Photo 4). A further sub-cell boundary may exist immediately downdrift of the mouth of the Cuckmere River, but the evidence for this is uncertain.
During the last decade, knowledge and understanding of the coastline of East Sussex has been considerably advanced. The first regional Shoreline Management Plan (Gifford Associated Consultants, 1997) presented a comprehensive and detailed synthesis of previous, but hitherto fragmented, research. It also included original work on littoral sediment transport rates and provided elements of sub-cell sediment budgets. Beach morphodynamic behaviour was clarified by examining trends of beach volume, inter-tidal width and steepness over the past 100 - 120 years. This work has been followed up with the publication of four Coastal Defence Strategy Studies (Halcrow Maritime, 2001a and 2001b; Mouchel Consulting Ltd, 2001a; Posford Duvivier, 2001). Each contains a more specific analysis of coastal geomorphology in the particular context of defence history and future strategic approaches to shoreline management. Academic research based at the Universities of Brighton, Sussex, Bournemouth, Portsmouth and North London has contributed new insights into cliff and platform erosion processes, beach behaviour and shoreline evolution. Four European Union projects have examined (a) cliff geology and morphology in the context of risk analysis (ROCC, University of Brighton); (b) shore platform erosion (ESPED, University of Sussex) and (c) historic cliff and platform erosion, beach sedimentology and budgets (BERM and BAR, University of Sussex).
A major new source of coastal data is from the Defra-funded National Network of Regional Coastal Monitoring Programmes. The Programmes consist of topographic beach surveys, nearshore bathymetry, aerial photography, lidar, coastal hydrodynamics (waves and tides) and terrestrial habitat mapping. Specifications for data collection are consistent for all regional programmes and the data and analysis reports are made freely available under the Open Government Licence from www.channelcoast.org.
The Southeast Regional Coastal Monitoring Programme commenced in 2002. The Lead Authority is New Forest District Council, with data collection, analysis and reporting led by specialist teams at the Channel Coastal Observatory (CCO), Canterbury City Council and Adur and Worthing Councils. (See CCO Annual Survey Reports for further details). The coastline from West Wittering to Fairlight has been monitored since 1973 (Annual Beach Monitoring Survey).
This report attempts a fully up to date and comprehensive review of both historical and contemporary knowledge of sediment inputs, throughputs, stores and outputs. Its main objective is to collate and integrate analytical research previously published, concluding with suggestions of topics where further work is needed.
1.1 Coastline Evolution
The Open Coast
The ancestral coastline of East Sussex may be traced back to the interglacial period following the creation of the Strait of Dover by 'ponded' glacial meltwater. The current consensus is that this "catastrophic" geomorphological event occurred towards the end of the Anglian glacial of the Middle Pleistocene, or Oxygen Isotope Stage 12 (Keen, 1995). Ice sheet expansion during this phase is likely to have created a glacio-isostatic forebuldge in the present region of the English Channel, but subsequent subsidence cannot be determined from any surviving evidence. The earliest firm evidence of relative sea-level change along the present East Sussex coastline is the presence of an interglacial raised beach and adjacent buried cliffline west of Roedean. This is clearly exposed at Black Rock (Brighton Marina), but extends westwards to Portslade and Shoreham-By-Sea beneath more recent sediments (Redman, 1852; White, 1924; Young and Lake, 1988). The basal rock platform at the type locality has a maximum elevation of +8.5mOD, but descends to +6.0mOD at Durrington and coincides with mean sea-level at Lancing. Numerous authors have described the sequence of basal erratics; rounded chalk, flint and some sandstone clasts, and overlying solifluction/colluvial sediments resting against an obliquely-exposed fossil cliffline (Smith, 1936; White, 1924; Hutchinson and Millar, 1998 and ROCC, 2001). Although the mean elevation of this displaced shoreline has increased with progressive marine erosion, amino acid ratios obtained from raised beach molluscan shells give a last (Ipswichian or oxygen Isotope stage 7) interglacial age. This has been independently confirmed by Parfitt et al. (1998) using palaeoecological indicators. The soliflucted material is a product of cold climate conditions during the succeeding Devensian glacial stage. The presence of an interbedded palaeosol at the Black Rock section has been provisionally dated to Oxygen Isotope stage 5 (Keen, 1995).
The westwards decline in elevation of the Ipswichian raised beach might be interpreted as evidence of subsequent glacio-isostatic and/or tectonic downward displacement of this sector of coastline. However, there are complications arising from the probable creation of an isobaric forebulge during the late Devensian and its subsequent, and probably ongoing, relaxation. Although modern tide gauge records indicate relative sea-level change, it is currently difficult to quantify neotectonic movement from the fragmentary survival of evidence of earlier sea-levels (Keen, 1995). The tectonic setting of the adjacent West Sussex coastal plain and Solent terrace system is distinct from that of East Sussex and does not, therefore, provide a simple or direct analogy for longer-term coastline displacement in this region.
During the last glacial (Devensian) stage of the Pleistocene, sea-level fell by at least 100m. Robinson and Williams (1983) ascribe a distinct break of slope at this datum, some 14km south of Beachy Head, to cliffline formation during a period of low sea-level stability. Jennings and Smyth (1987; 1990) briefly describe a submerged cliff some 6km seawards of Beachy Head, which they ascribe to erosion during one or more stages of shelf emergence in the Devensian. Jones (1971; 1981) has identified the basal Chalk surface below the Ouse Valley to be -29.6mOD at Newhaven and -12.2mOD at Lewes. Smith (1985) has traced a buried channel extension of the River Cuckmere, tributary to his "Northern Palaeovalley", some 5 to 7km southwards of the modern East Sussex coastline. Unlike most other palaeovalley segments of the eastern English Channel, this is only partially sediment-filled and is therefore readily discernible from sea bed surveys. Bellamy (1995) reports evidence for a fluvially-eroded, but now infilled, buried channel that connects the mouth of the Adur with the same palaeovalley system. The confluence is inferred to have been some 30 km south of Shoreham-by-Sea, suggesting a contemporary mean sea-level 50 to 60m below its present position. By analogy with his detailed description and analysis of the sediment stratigraphy of the buried channel of the Arun further west, Bellamy (1995) proposes that much of the gravel infill was fluvially supplied under a periglacial regime. Castleden (1996) has recorded a bedrock floor at the mouth of the Cuckmere at about -25mOD, declining upstream to -13.1mOD at Exceat Bridge. Young and Lake (1988) state that a buried channel graded to -23mOD, now infilled with Holocene alluvial deposits is present beneath New Shoreham. Bell (1976) has determined a buried channel at -25.8mOD close to Newhaven, below an excavated Romano-British site. These various reports therefore suggest that, allowing for seaward extension of valleys at gradients of between 2 and 4m/km (Robinson and Williams, 1983), sea-level was at approximately -35mOD some 10 to 10.5ka B.P. Jennings and Smyth (1987; 1990) state that the extrapolation of rates of cliff recession for the past 200 years would locate the coastline at this time some 6-12km seawards of its modern position. White (1924) suggests it was approximately 3-4km seawards 4-5000 years B.P.
Devensian cold climate conditions promoted severe freeze-thaw weathering and active solifluctional mass movement over pre-existing Chalk valley slopes. The accumulation of soliflucted colluvial infill ("Coombe Rock") in pre-existing valleys is clearly visible along several coastline sections, where it has been exposed by cliffline retreat. All of these are now dry valleys, due to the loss of surface run-off on chalk slopes and absence of discharge in chalk streams, following the replacement of permafrost conditions by temperate climate during the modern Holocene interglacial. The permafrost table was within a few metres of the ground surface, thus preventing normally rapid rates of infiltration and percolation. Most dry valleys are also "perched" or "hanging", as their former seaward extensions have been removed by rapid coastline recession during the Holocene. Although several Coombe Rock sections between Roedean and Saltdean have been placed behind robust coastal defences, at locations such as Birling Gap this mechanically weak material is fully exposed to wave erosion. In addition, Devensian cold climate weathering has also reduced the coherence and stability of sub-surface Chalk and added other structural properties, which induce cliff failure along most parts of this coastline east of Brighton Marina (ROCC, 2001). A low angle soliflucted fan of Coombe Rock occupies the coastline west of Kemp Town, Brighton. It is difficult to reconstruct its former seaward extent, but it is likely to have been sufficiently extensive to provide a substantial source of flint debris for offshore to onshore shingle transport during the Holocene marine transgression (Jennings and Smyth, 1990; 2001). Release of flint may have been assisted by decalcification of this deposit during its exposure to earlier sub-aerial weathering.
Rapid recovery of sea-level rise during the early and mid-Holocene is recorded in complex sequences of minerogenic and organogenic sediments infilling the valleys of the Adur, Ouse and Cuckmere. Sea-level stood at approximately -27m at 9ka B.P., -15m at 7.5ka B.P. and -5m at 5.5ka B.P. It was during this period that brackish, and then marine, conditions penetrated into what are now the floodplain sections of the three main river valleys. Between about 5ka and 3.2 to 3.3ka B.P. inundation created a strongly embayed coastal plan, with well-developed ria-like inlets. Major headlands, especially Beachy Head, were probably well defined by about 5ka B.P., with consequent reduction of longshore sediment movement around them. Jones (1971) has dated the earliest entry of brackish conditions at the present day mouth of the Ouse to approximately 6.3ka B.P., using radiocarbon dating of organic sediments. Sediment stratigraphy indicates that marine conditions replaced brackish and that the tidal limit extended upstream of Lewes by approximately 3.2ka B.P., when sea-level was -2mOD. After 3ka B.P., terrestrial sedimentation became dominant, and steadily converted an estuarine embayment into a river floodplain. Human-induced acceleration of alluvial sedimentation due to forest clearance and land cultivation in Neolithic and later times is a significant cause of this relatively abrupt change in environmental conditions. Research undertaken at sites east of Beachy Head, such as Coombe Haven; Willingdon Levels (Jennings and Smyth, 1997; 1990; Jennings, 1990), Pevensey and Pett Levels and Romney Marsh has provided more detailed and securely-dated evidence for a complex suite of Holocene environment changes. However, these areas may not provide a precise analogy for the lower Ouse, as sedimentation of formerly open embayments was probably controlled by locally complex relationships between the growth and periodic breaching of gravel barriers, increases in river discharge due to reduced vegetation cover, and the hydraulic response of estuary shape to accommodate fluctuations in tidal prism.
The lower Cuckmere demonstrates local variations in the sequence, and timing, of Holocene events. Castleden (1996) argues that it was a narrow estuarine inlet by about 5ka B.P., with tidal conditions penetrating upstream to Alfriston. He points to perhaps ambivalent evidence of marine "trimming" of the 5m high cliffed slope on the eastern side of the valley below Exceat Barn. Estuarine sediments below modern alluvium suggest that tidal flat formation may have commenced as early as 4.5ka B.P., and that brackish conditions persisted until about 1.5ka B.P.
It is therefore probable that the East Sussex coastline was well embayed in Romano-British times, and that the lowermost sections of the Adur, Ouse and Cuckmere valleys persisted as natural tidal basins as late as the twelfth century AD (Bell, 1976; Robinson and Williams, 1983). Drainage and land claim does not appear to have commenced earlier than the tenth century in the case of the lower Adur, and it did not become an organised process in the Ouse Valley until the early sixteenth century (Brandon, 1971). Some minor river re-entrants, e.g. at the original sites of Brighton and Seaford, were not removed until the seventeenth and eighteenth centuries. Thus, the broad arcuate planform of the modern coastline between Shoreham-by-Sea and Beachy Head is largely a feature of the modern historical period.
Estuarine sedimentation was accelerated by longshore drift-fed growth of shingle spits eastwards across the entrances of each of the main inlets. This may have been accomplished during earlier stages of inlet definition following the deceleration of the rate of relative sea-level rise after approximately 5.5 to 5ka B.P. It is considered probable that spit elongation was accompanied by net accretion of gravel due to onshore barrier migration. Phases of rapid landwards barrier movement during recent centuries may be inferred from various historical sources (Gifford Associated Consultants, 1997). Breakdown, and backbarrier inundation, of the original (Norman) site of the fishing community of Brighthelmstone (Howell, 1870; White, 1924) is consistent with this interpretation. Nicholls (1991) considers that more or less continuous beach depletion occurred here between 1545 and 1713, with subsequent short episodes of erosion. Howell (1870) describes brackish estuarine deposits beneath the Pool Valley, which may therefore have been created by an earlier breach of a gravel barrier. The 'Old Town' of Brighthelmstone was accommodated beneath cliffs either side of a narrow alluvial valley, but retreated to higher ground between the thirteenth and early eighteenth centuries as a result of storm-induced erosion. The 'Great Storm' of 1705 destroyed what remained of the Old Town, and up until the mid-eighteenth century the 'New Town' was an open coast site exposed to erosion (Redman, 1852; Howell, 1870; White, 1924; Castleden, 1996). However, from approximately 1750 there was an abrupt change in morphodynamic behaviour, with net accumulation of beach material as part of the creation (or re-creation) of a barrier structure. This continued to move landwards across the site of the Pool Valley, until finally restrained by the urbanisation of the hinterland and the introduction of control structures after about 1790.
Indirect evidence in favour of barrier morphogenesis may be inferred from the evolutionary changes, especially eastward elongation, crest elevation and foreshore progradation, of the spits at the entrances of the Adur and Ouse. Barrrier emplacement is also a plausible explanation for the periodic closure of the mouth of the Cuckmere before the current training walls were introduced.
Jennings and Smyth (1987; 1990) provide detailed evidence from bio- and lithostratigraphic analysis of valley infill sediment at Coombe Haven, near Hastings, for coarse clastic barrier growth in the late Holocene. The original sediment feed may have been via eastwards longshore drift, with subsequent supply derived from offshore sources. Further west, Willingdon Levels and the Crumbles cuspate gravel complex provide further evidence of a complex sequence of events involving cyclic erosion and deposition linked to barrier growth and breakdown.
Willingdon Levels may provide a framework for understanding the late Holocene coastal evolution of the lower valleys of the Ouse and Cuckmere. During the period 5 to perhaps 3ka B.P. it was a low energy tidal embayment that interrupted what is now unidirectional longshore drift. Once sea-level stabilised, barrier spit growth extended across the entrance; this condition may have persisted for at least two millennia, and was not disrupted until the twelfth century (Jennings and Smyth, 1987; 1990). The relative importance of sources of supply of gravel to feed barrier growth is uncertain, but derived from either updrift sources, bypassing the then less prominent headlands of Seaford and Beachy Heads; or from offshore sources. The late Holocene growth of the Crumbles cuspate foreland at Eastbourne is strongly suggestive that the gravel supply, rather than sand, was predominant, though there may have been an earlier phase during which much of the sediment input was sand. It is thus conceivable that the first generation barrier here was composed of sand and that the structure broke down as this supply became exhausted, and/or the entrance of the tidal pass widened and deepened. Much of this offshore supply could have been derived from Pleistocene raised beach and Coombe Rock deposits, the latter rich in weathered flint. The Arun, Adur, Ouse and Cuckmere would have contributed additional coarse sediment when they drained a tundra environment. All of this material would have been reworked and redistributed during the complex sequence of Quaternary climate and sea-level changes. Substantial input would also derive from the earlier denudation of Chalk and Eocene bedrock, a long-term process which would have bevelled some 12 to 15km of Chalk outcrop, as well as the Portsdown to Littlehampton Chalk pericline further west. More speculatively, sand and gravel is likely to have been introduced into the eastern English Channel once the Strait of Dover had been created.
There was, then, an abundant supply of flint to build gravel barrier structures along the entire length of the shoreline of East Sussex at this time. Nicholls (1991) considers that the major period of barrier growth was 3 to 2,300 years B.P., known from sites elsewhere (e.g. Dungeness) to have been a period of exceptional storminess.
The analogy with better-developed barriers along the south and southwest coasts of England (e.g. Chesil Beach; Slapton Sands; Dungeness; Sandwich Bay) is close. However, all of these examples are now eroding, or are in a condition of chronic disequilibrium. A main cause of their degeneration is the exhaustion or substantial decline of offshore "reservoirs" of gravel and sand. In East Sussex, the erosion of beaches and accretion structures such as The Crumbles and Dungeness, since the eighteenth century is consistent with this. Barrier breakdown has also been accelerated by a switch from swash to drift alignment, which had the effect of promoting strong spit growth across the previously wider tidal passes of the Adur, Ouse and Cuckmere. This, in its turn, caused a previously more segmented shoreline, characterised by several discrete cells or sub-cells, to become dominated by unidirectional (eastwards) drift between major headlands (Nicholls, 1991).
Jennings and Smyth (1990), after documenting the history of the Crumbles cuspate foreland hypothesise that there have been several short period (up to 50 year) "pulses" of offshore to onshore gravel transport superimposed on a longer-term history of diminishing feed. This is considered to result from a change from a dissipative to a reflective wave climate domain. The last such "pulse" coincided with a major phase of growth some 300 years ago. It is tempting to adapt this concept to the coastline west of Beachy Head, but there is no specific evidence to support it. One important implication, however, is that the morphodynamic behaviour of regional beaches is subject to cyclical fluctuation, with rapid phases of sediment gain separating longer periods of loss. Seen in this context, coastal defences are not necessarily a fundamental cause of beach deterioration (e.g. Seaford Beach) as both current and historical losses are diagnostic of a tendency for natural erosion. However, in the case of the coastline between Shoreham and Newhaven, modern beach budgets have been greatly modified by the insertion of orthogonal structures such as massive breakwaters and groynes.
Nicholls (1991), in discussion of the hypotheses of Jennings and Smyth (1990) places strong emphasis on the fact that headlands (especially Beachy Head) were less prominent features of the coastline in the mid to late Holocene. They were therefore less effective barriers to longshore transport, thus linking West and East Sussex in a more unified littoral drift system. This view would allow potentially large quantities of gravel moving onshore at, and east of, Selsey Bill to feed beaches up to some 50km eastwards. Rates of long-shore transport during the late Holocene would have been a function of prevailing coastal planform, mean wave height, wave approach, the magnitude/frequency of large storms and offshore bathymetry. By-passing of estuary mouths would have been achieved under storm conditions, when remobilisation of offshore stores might also have occurred. This would create conditions for "pulsed" longshore transport, in contrast to episodic offshore to onshore gravel inputs preferred by Jennings and Smyth. Nicholls (1991) also calculates that an annual drift rate of approximately 80,000m³ per year along the West and East Sussex coasts would have been needed to create the gravel store of the Crumbles, Eastbourne over the 500 year period (800-300 years B.P.) during which he argues it was formed. However, it is possible that there were precursors to the modern Crumbles foreland, and that gravel from as far west as Selsey Bill might have moved longshore to supply Dungeness and the Camber ness. If this were the case, the drift rate along the East Sussex coast prior to approximately 2000 years B.P. must have been substantially more than 120,000m³ per year. This is some five to six times prevailing rates, and is considered by Jennings and Smyth (1991, in reply to Nicholls) to be unlikely. They therefore prefer a sediment budget dominated by offshore supply.
Evolution of the mouth of the Adur and adjacent coastline
Although it is likely that the river Adur had a wide estuary mouth in Romano-British times, its exit was deflected progressively eastwards by the sustained growth of a gravel barrier spit during the ensuing 1000 years. Although Morris (1931a and b) noted substantial erosion at the site of Old Shoreham in the fourteenth and fifteenth centuries, this should not be taken as evidence of the absence of a spit enclosing the harbour at that time. It is probable that Shoreham spit is a barrier structure that has experienced periodic breaching, a hypothesis originally proposed by Ward (1922). Residual salt marsh and mudflats at Lancing (reclaimed in the sixteenth century) and a chain of marshy lagoons between Goring and Hove surviving up to the mid-eighteenth century (Brookfield, 1952) have been interpreted as formerly contiguous barrier-confined features. Widewater lagoon is the most well defined surviving component of what may have been a nearly continuous back-barrier wetland. Brookfield (1949; 1952) made the indirect suggestion that Shoreham spit was substantially the result of onshore barrier migration, rather than simply linear growth fed by longshore drift. Breaching would have occurred on several occasions, almost certainly during the several "super" storms that occurred during the thirteenth and fourteenth centuries. If this hypothesis is accepted, Shoreham spit is thus one part of a regional coarse clastic barrier system that has experienced progressive landwards translation over the past several centuries. The erosion and loss of the former port of Pende, to the west (Ballard, 1910; Ward, 1922); and rapid storm-induced onshore beach migration at Brighton in the mid-seventeenth century and Worthing in the 1820s (Gifford Associated Consultants, 1997) are consistent with this explanation. The latter event involved almost 100m of shoreward movement in a few days, eliminating the pre-existing lagoon or marsh at East Worthing. Young and Lake (1988) report borehole evidence for well-rounded storm beach gravels overlying alluvium beneath the A259 at Shoreham, as further support for barrier translation.
Net eastwards longshore transport west of the modern entrance to Shoreham Harbour is a significant process, and was undoubtedly responsible for the eastwards migration of the mouth of the Adur prior to the construction of breakwaters and training walls (Martin, 1939). It was the likely cause of the infill of the 'new cut' of 1698, which temporarily provided New Shoreham with a direct outlet to the sea, possibly through a breach site. However, the effectiveness of longshore drift in closing this artificial entrance within 20 to 30 years was aided by contemporary land claim of the alluvial marshes of the lower Adur. This occurred over a relatively short period (Ward, 1922; Martin, 1939) and reduced river mouth current velocities due to a reduction in the estuarine tidal prism. Between 1700 and 1760 there was almost 2km of eastward spit migration, leaving New Shoreham - built in the eleventh century because of channel siltation up-estuary - some 1.3km from the entrance to the Adur. A new channel, opposite Kingston-by-Sea (Southwick), was cut in 1762, piers were constructed and the former river mouth closed off. This proved ineffective, requiring a new channel some 0.6km eastwards in 1783. Poor maintenance quickly rendered this entrance un-navigable and the spit migrated 30 to 40m per year eastwards during the subsequent 35 years. Prompted by concerns of economic decline, the 1762 channel was re-opened in 1818 and provided with substantial training walls (Brookfield, 1949). Dredging was thereafter undertaken regularly to avoid shoaling, and the former river channel east of Kingston was converted to a canalised harbour basin in the 1850s. The modern western breakwater and eastern training wall continue to maintain this entrance, with a routine dredging operation to avoid natural sedimentation in areas of weak tidal currents and limited wave penetration. The western breakwater, in particular, intercepts a substantial proportion of potential longshore by-passing of the harbour mouth by gravel. This has necessitated deliberate management of the local sediment budget, removing a proportion of the updrift surplus and transferring it to downdrift beaches, which have a long history of sediment deficit. This operation, and the complex flux of both onshore and offshore - directed sediment flows in the entrance channel is addressed in Section 2.4 and Section 4.2, respectively.
Historical evolution of the mouth of the Ouse
The medieval exit of the Ouse was southeast of the original port of Seaford, as a consequence of the progressive eastwards downdrift growth and onshore migration of a gravel barrier spit across an original shallow tidal embayment. This phase of growth presumably occupied at least several earlier centuries and promoted estuary mud flat accumulation. The Mill Creek represents a surviving portion of the early to mid-sixteenth century estuary, whose mouth was abandoned before 1540 by continuing mobility of the distal end of the spit and by siltation of the tortuous access channel to the by then moribund port of Seaford (Morris, 1931; Carr-Gregg, 1952). Alluvial sedimentation was promoted by the reduction of the tidal prism due to lower estuary land claim or "inning" (Brandon, 1971; Farrant, 1972). The latter process, however, increased flood hazard from high discharges of the Ouse (Robinson and Williams, 1983). It is uncertain if inundation of medieval properties, dating from the mid-fourteenth century at Seaford (Bell, 1976) was due to this cause or to storm overtopping of the barrier spit. Morris (1931) stated that a new exit was created in 1565, between Tide Mills and the site of the Buckle Inn, as a direct result of a storm surge induced breach. However, Brandon (1971) used documentary evidence to confirm that a channel was deliberately cut, and a port facility established, at New Haven in 1539. This was partially blocked by eastward longshore transport over the next 150 years, despite periodic attempts to clear and re-open it. This probably explains, in part, the progressive deterioration of the drainage of Lewes and Laughton Levels during the seventeenth century (Brandon, 1971). The Ouse is known to have overtopped, and broken through, the spit at Tide Mills, in 1676 and 1698. The latter event created a new exit, but the "cut" of 1539 was re-opened in 1731 with the addition of protective piers. However, the problem of accretion was apparently accelerated by their presence, and eastwards drift was not effectively constrained. In response, the first substantial rock breakwater was constructed, west of the harbour entrance, in 1791; the river channel was straightened and Newhaven re-established as a port. To ensure access, the approach channel was given additional protection through the extension of the length of this breakwater to 150 m (1847) and 800m (1878-1890). The latter, constructed of rock and concrete, is effectively the modern structure. The first breakwater, or pier, on the eastern side of the Ouse entrance dates to 1664, and has been periodically reconstructed, strengthened and relocated to widen the river mouth over subsequent centuries.
Although breakwaters, together with channel deepening and dredging, have removed the hazard of the accumulation of a bar at the outer harbour entrance since the mid-nineteenth century, other impacts have been experienced. These include the accumulation of gravel immediately updrift of the western breakwater; the accretion of finer sediment on the eastern sides of both breakwaters and the progressive erosion of Seaford Beach. The latter effect, however, may not be linked directly to the impedance of longshore transport by the presence of the harbour breakwaters. Other causes, including the gradual depletion of the finite sand and gravel store in Seaford Bay, may be implicated. Substantial losses of beach sediment can be correlated with individual storm surge events, e.g. 1825 and 1877 (Morris, 1931); 1897 (Robinson and Williams, 1983) and 1943 (Hydraulics Research, 1986c). These issues are examined in more detail in Sections 3 and 4.
Cuckmere Haven: Historical Change
The original ria-like estuarine inlet was progressively infilled by alluvial sediments during the later Holocene, a process that was accelerated by the eastwards growth of a coarse clastic barrier spit fed by both longshore and onshore gravel transport. This caused the lower channel and mouth of the Cuckmere River to be deflected eastwards, with periodic closure due to a combination of "pulses" of increased onshore or longshore sediment supply and reduced tidal prism due to land claim in the lower floodplain. Fluctuations in the position of the mouth between the late eighteenth and early twentieth centuries, in some of these cases related to breaching of the spit beach, are recorded by Gifford Associated Consultants (1997). Isaac (1915) has described short-term blockage of the river entrance by surge-induced beach sedimentation in July 1912, which caused extensive floodplain inundation. The river mouth (Photo 5) is now in its approximate late eighteenth century position (Castleden, 1996) and is confined by training walls originally introduced in the late nineteenth century, but subsequently upgraded following periodic failure (Clifton and Cecil, 1999). This has induced the formation of a small ebb delta with a continuously shifting pattern of distributary channels. Net accretion at this point is probably maintained by local wave refraction caused by the set back of the coastline, creating a shallow embayment.
1.2 Wave and Tidal Conditions
Tidal Regime
The mean (spring) tidal range for this coastline, increases from west to east, viz: Shoreham-By-Sea 5.6m; Brighton, 6.01m; Newhaven, 6.3m; Beachy Head, 6.4m. Tidal streams, derived from Associated British Ports Research and Consultancy Hydrodynamic Model of the English Channel, are given in the regional Shoreline Management Plan (Gifford Associated Consultants, 1997). Flow is eastwards on the flood tide, westwards on the ebb. A strong tidal current residual (headland vortex) exists off Beachy Head, characterised by rectilinear currents with peak (spring) velocities of 1.3m per second. Velocities are much lower elsewhere, e.g. 0.8m per second (springs) and 0.4m per second (neaps) in the entrance channels to the port of Shoreham and Newhaven Harbour. These decline to less than 0.3m per second within 200m of mean low water. Ebb tidal current speeds are faster than on the flood at the mouths of the Adur, Ouse and Cuckmere.
Wave Climate
The Southeast Regional Coastal Monitoring Programme measures nearshore waves using a Datawell Directional Waverider buoy deployed at a network of sites. Between 2003 and 2012 the buoy deployed at Rustington in 10mCD water depth, determined that the prevailing wave direction is from the southwest-by-south, with an average 10% significant wave height exceedance of 1.57m. Between 2008 and 2012 the buoy deployed at Seaford in 13mCD water depth, determined that the prevailing wave direction is from the southwest-by-west, with an average 10% significant wave height exceedance of 1.68m (CCO, 2012).
Otherwise, knowledge of the regional wave climate is limited to a number of relatively short term studies undertaken by consulting companies for specific projects. Most have involved hindcasting using regional data sets on wind speeds.
Both observational (instrumental) and hindcast data on significant wave heights (Hs) are available for Shoreham-By-Sea from several studies. Hydraulics Research (1984) recorded maximum Hs of 5.25m during a short-term monitoring period of offshore waves. A later study (Hydraulics Research, 1989; Jelliman, et al., 1991) used the MULTIGRID wave model to create an inshore wave climate, derived from a hindcast offshore wave climate, using data for a one year period. This approach utilised three-dimensional bathymetry to compute the effects of refraction and diffraction set up by both seabed relief and - more importantly – the Shoreham harbour breakwaters. The latter can increase wave height by a factor of 1.6 to 1.8. This study concluded that waves up to 6.0m in height can be generated during south and south west approaching storm surge waves with a recurrence of 1 in 20 years. The equivalent height for waves coming from the east or south-east is 4.7m. Halcrow (1990) measured Hs values between 1.45m and 4.3m at the outer entrance channel over a 6 month period between October and April. These values were re-calculated at 2.30m and 4.87m, respectively, for a 1 in 5 year storm recurrence. Hydraulics Research (1986b), specifically examining long period waves off Shoreham seawards of the approach channel, give Hs values of between 4.0m (annual recurrence) and 4.7m (1 in 10 year frequency). Posford Duvivier (1993) reported that Hs exceeded 3.0m for 1.6% of a three month summer season monitoring period, within the main entrance channel. This apparently related to diffracted waves "caught" between the western breakwater and eastern pier. The longest hindcast data series is that assembled by Duane (1998) covering the period 1971 to 1998. The wave climate is described in detail by means of "wave roses" giving the frequencies of occurrence of different combinations of wave height and direction. Reasonable correspondence was achieved between the hindcast waves and several months of direct field measurements. Refer also to Mitchell and Pope (2004).
Further east, Bailey (1999) notes a maximum wave height of 7.23m off the West Pier, Brighton; however, the basis for this calculation is not clear. Hydraulics Research (1986a) stated that there is a 1 in 10 year frequency of Hs of 6.24m off Brighton. Halcrow Maritime (2001b) state Hs to be 4.79m for a 1 in 5 year recurrence at Brighton, but this appears to have been derived directly from earlier analyses for Shoreham. Mouchel Consulting (2001) proposed a maximum Hs of 4.4m (annual recurrence) for waves offshore Newhaven. Hydraulics Research (1985) undertook wave monitoring, over a 12 month period, to construct a wave climate for Seaford Bay. Wave rider buoy data was fed into the OUTRAY wave refraction model to derive Hs of 1 m and 3m for 7.2% and 0.9% respectively, of the measurement period during which south-westerly waves were operating. A maximum Hs of 1.8m was achieved by waves approaching Newhaven Harbour from the south-east.
Posford Duvivier (1993), using wave data for Shoreham and Pevensey Bay calculate a once per year frequency of Hs of 4.56m at Birling Gap. This increases to 5.48m for a one in 20 year recurrence. There is no quantitative data for wave height or energy impacting the shoreline east and west of Birling Gap, but it is generally acknowledged that regional maxima are experienced between Seaford and Beachy Heads. Diffraction set up by the ebb delta offshore Cuckmere Haven would be expected to locally reduce Hs values. Conversely, they would be increased by refractive convergence of incident waves around the several salients of Beachy Head.
All the above studies are in agreement that waves from the west and south west are both dominant and prevalent (May, 2003) along most parts of this coastline. Hydraulics Research (1989a) calculated that 48% of all waves approaching outer Shoreham Harbour propagated from between south and west. Waves moving in from between east and south-east directions represented 28% of the total during the measurement period. The remaining 16% were represented by waves from the south, with Hs below 0.8m. These figures remove the complicating effect of re/diffraction set up by breakwaters and training walls, which causes changes in approach as waves move from offshore to inshore. Although only three studies (Hydraulics Research, 1985 and 1989a and Jelliman, et al., 1991) were conducted over most, or all, of a one year period, it is well established that wave height and energy is greatest during the winter months. Most surge events have occurred between October and February, though documented records are almost non-existent for some sectors of this shoreline. One small sector where waves from the south-east may be locally dominant is between Newhaven Harbour East Pier and Tide Mills (Seaford Bay). This is due to a combination of coastline orientation and protection afforded by the western breakwater (Hydraulics Research, 1985). Local diffraction, and reduced incident energy, has been noted for the sector between Brighton Marina and some 2-300m further east (Posford Duvivier, 1994b; 1995; 2001).
Seaford Bay was one of the locations for which wave modelling exercises were undertaken as part of the DEFRA Futurecoast Project (Halcrow, 2002). An offshore wave climate was synthesised based on 1991-2000 data from the Meteorological Office Wave Model and then transformed inshore to a prediction point in Bracklesham Bay at -5.5mOD. Potential sensitivities to likely climate change scenarios were then tested by examining the extent to which the total and net longshore energy for each scenario varied with respect to the present situation. Results suggested that a one to two degree variation in wave climate direction could result in a 6-12% variation in longshore energy and confirmed that the Bay was significantly more sensitive to this factor than most other south coast locations, as might be expected of a swash aligned coastline.
Sea-Level
Several studies (Gifford Associated Consultants, 1977; Posford Duvivier, 1995 and 2001; Mouchel Consulting, 2001; Halcrow Maritime, 2001; Hydraulics Research, 1985 and 1986a; HR Wallingford, 1997) calculated extreme water levels and their recurrence intervals. All approaches require an estimate of contemporary relative sea-level rise, but at present there is a wide range of estimates. This reflected, above all, uncertainty over prevailing rates of land movement, including possible co-adjacent seabed displacement. Shennan (1989) proposed isostatic/isobaric subsidence to be 0.1 to 1.4mm per year, whereas Bray, Carter and Hooke (1994) calculated the range to be 0.5 to 1.5mm per year. The latter authors therefore suggested a future relative sea-level rise of 5-8mm per year. Bailey (1999) gave the much lower figure of 1.70mm per year for Brighton, but this would seem to be a form of extrapolation of recent sea-level data for Dover. Thompson (unpublished, 1977) estimated 4 to 5mm per year for Newhaven (and the coastline west to Portsmouth), whilst Blackman and Graff (1978) deduced from a part of the tidal gauge data for Newhaven that sea level is currently rising at a rate of 4.1mm per year (+/- 1.65mm per year). Extreme sea-levels resulting from tide-surge interaction have been analysed for the eastern English Channel (Tomasin and Pirazolli, 2008). For the tide gauge record for Newhaven, 1983 to 2002, the maximum recorded level was 7.69mCD.
2. Sediment Inputs
2.1 Marine Inputs
Experiments conducted between 4km and 10km offshore of Shoreham-by-Sea by Crickmore et al., (1972) indicated a minimal landward drift of gravel. At the 9m water depth, shoreward transport was measured at 1,000-1,500m³ per year per km whilst at 12m depth it reduced to less than 500m³ per year per km. No transport was recorded in excess of 18m depth. This rate of transport will only occur where potentially mobile shingle exists on the seabed. There is a potential for this process to occur over the nearshore bed between Brighton and Peacehaven, but it is uncertain whether material can actually pass across exposed shore platforms to be supplied to beaches. Furthermore, it would be expected that the operation of the process over time could have already depleted potentially mobile inshore gravel reserves since re-supply would be unlikely from further offshore due to lack of gravel mobility in greater water depths. The knowledge of these processes was derived from experiments with radioactive tracers carried out over 2 years and correlated with wave data. HR Wallingford (1993) subsequently confirmed patchy distribution of inshore gravels together with the lack of shingle mobility beneath water depths greater than 15m to 18m. In the original SCOPAC (2004) study, this information was used for a F1 arrow, in the nearshore of Shoreham-By-Sea and also feeding the beaches at between Shoreham-By-Sea and Peacehaven. This is not mentioned by the study by Crickmore et al., (1972), and is only a theoretical assumption. The F1 arrows have been removed for the 2012 update, and replaced where appropriate with O1 arrows (see section 4.1) to represent evidence of transport occurring in the offshore zone.
Joliffe (1978) also undertook tracing of coarse particles at a number of experimental sites, overlying chalk surfaces and "patchy" gravels, 4.8 to 9.0km offshore Shoreham Harbour. Although observation was limited to three weeks, shoreward migration in water depths of less than 6 to 12m was observed, but none in depths greater than 15m. Smallest clasts moved furthest, with disc shaped particles the most mobile. Angular/sub-angular shapes were more readily entrained than rounded forms. Highest mobilities occurred across bedrock surfaces, with almost no movement recorded over silt and sand. There was no apparent correlation between clast transport and the velocity and direction of residual tidal streams, thus all movement was imparted by wave-induced stresses. In an earlier experiment, Joliffe (1972) introduced tagged gravel clasts into water depths up to 4m below maximum low water 90-180m south of Hove. Rapid shoreward dispersal occurred wherever bedrock or packed cobble surfaces were present, with some reaching the beach within 2-3 weeks. Tidal currents (<0.75m per second) were again considered an ineffective transportation agency. Wave heights during the experimental period varied between 0.5 and 1.0m. Sand ripples and fine gravel ribbons inhibited the movement of larger clasts.
Although patchy seaweed growth was observed, there was no sighting of kelp-rafting; Joliffe (1972; 1978) dismissed kelp-rafting as a possible auxiliary mechanism for moving gravel onshore in his other experiments. However, Gifford Associated Consultants (1997) incorporate assumptions of kelp-rafting of gravel sized clasts into the estimations of net onshore gravel transport. These vary from 2,500m³ per year, between Shoreham Harbour Entrance and Brighton Marina, to 6,000m³ per year along the nearshore sector between Cuckmere Haven and Beachy Head. As kelp-rafting has not been directly observed along any part of the East Sussex coast, its relative importance in relation to total onshore "creep" cannot be assessed. The volumes quoted by Gifford Associated Consultants (1997) are very small components of the total sediment budget (Section 6).
A high resolution, 100% coverage swath bathymetry survey between Dungeness and Newhaven, extending 1km offshore of MLW was commissioned by the Southeast Regional Coastal Monitoring Programme, and completed in August 2013. Analysis of this bathymetry data indicates that the nearshore seabed of Shoreham-by-Sea is shallow and gently sloping and consists mostly of sand and a small area of rock and coarse sediment found to the east of the harbour entrance. No bedforms are discernible to confirm sediment mobility or indicate direction of offshore or onshore sediment transport. Apart from the sandy seafloor offshore of Seaford, the nearshore seabed offshore between Newhaven and Beachy Head is dominated by an extensive, almost continuous rock platform. This exposed platform extends seawards from the inter-tidal zone some 600m, where it terminates. Seawards of this abrupt junction, the surficial sediments are comprised of mixed and coarse-grained sediments. The drainage of the River Cuckmere has eroded the rock platform. In Cuckmere Haven the sub-tidal deltaic deposits of sand and finer silty sediments are a relatively constrained pocket of sediment and of sufficient thickness to mask the underlying platform. Finer sediments are found in sheltered areas with less wave action, but no bedforms are discernible to confirm sediment mobility or indicate direction of offshore or onshore sediment transport. To the east of Beachy Head, bedforms are discernible at the seaward and steep boundary of the rock platform, indicating strong seabed currents past the headland (see section 4.1). The seabed offshore of Beachy Head is more complex with significant rock outcrop and shore parallel northeast-southwest oriented ridges extending offshore, indicating former shoreline positions.
Experimental work and diving observations by Joliffe (1972) off Newhaven Harbour indicated that gravel particles were mobile over bedrock and other smooth surfaces free of silt, sand and gravel. However, they lost mobility once they encountered areas of fine-grained deposition. Similar work carried out in Seaford Bay (Joliffe, 1964; 1972; 1978), using fluorescent-coated indigenous rounded gravel clasts, did demonstrate that material could move shoreward from modest (<5m) depths. As almost all of this material was originally derived from beach foreshores, it probably represents recycling rather than fresh input. Similar processes would be likely to operate within the Cuckmere Haven embayment, although they have yet to be studied directly. In the original (2004) SCOPAC Sediment Transport Study, this movement was represented with F2 arrows to represent wave powered offshore to onshore transport to Seaford Beach, Cuckmere Haven and towards Birling Gap. The swath bathymetry survey does not present any evidence of bedform features or a nearshore sediment feed. Therefore, the F2 arrows have been removed for the 2012 update, and replaced by O2 arrows (see section 4.1) where appropriate to represent evidence of sediment transport occurring offshore.
2.2 Fluvial Input
FL1 R. Adur, FL2 R. Ouse, FL3 R. Cuckmere
Rendel Geotechnics and the University of Portsmouth (1996) estimated suspended and bedload delivery at the mouths of all rivers discharging at the coast of central-southern England. Their figures are based on an assumption that more than 90% of load is diverted into storage before reaching the open sea, much of it in alluvial plains, estuarine channels and inter-tidal flats. On this basis, suspended sediment delivery by the Cuckmere is calculated at 1,320 tonnes per year, and by both the Ouse and Adur at 2,682 tonnes, per year. Gifford Associated Consultants (1997) give figures of 800m³ per year (Cuckmere); 3,700m³ per year (Ouse) and 2,800m³ per year (Adur) based on rating curves. In all three cases, coarse bedload is considered negligible, as only very limited exposures of river terrace gravels provide suitable input within catchments mostly underlain by Chalk and a variety of clays and fine grained sandstones. A figure of 131 tonnes per year of bedload is estimated for the Cuckmere (Gifford Associated Consultants, 1997), but no data is given for either the Ouse or Adur.
It is probable that input of river-transported coarse sediment into the nearshore zone occurs episodically, during high discharge events. Suspended load is removed seawards almost immediately, but the presence of a small ebb delta at Cuckmere Haven and dredging of navigation channels giving access to Shoreham and Newhaven Harbours indicate some potential retention of fluvially-derived coarse sand in their estuarine and nearshore environments.
Any further evaluation is hindered by the absence of site specific appropriate gauging and monitoring data, but it is clear that the quantitative significance of fluvial input into the regional sediment budget is very low. It has diminished to its present negligible amount as a result of the reduction of the tidal prisms of all three estuaries caused by progressive land claim since medieval times.
Davies (1973) located several large zones of freshwater discharges from the Chalk aquifer of the South Downs in the nearshore area between Hove and the Seven Sisters. Some are linked to higher rates of groundwater transmissivity within major joint and fissure systems of dry valleys. They discharge across inner shore platforms and beneath overlying beach sediments. The evidence was derived from airborne thermal line-scan imagery, which was insufficient to resolve either quantities of discharge or any suspended sediment output.
2.3 Coast Erosion
2.3.1 Cliff Erosion
There is a considerable literature providing estimations of rates of cliffline recession, both before and after the construction of coastal defences. Most data derives from analysis of the position of the cliff top on successive Ordnance Survey maps and plans, of which the most detailed and accurate is presented in Dornbusch (2002) and Dornbusch et al. (2006b and 2008a). These latter sources provide a review of previous work and calculate a mean retreat rate for each 50m sector of Chalk cliffline between Black Rock, Brighton and Beachy Head between 1873 and 2001 (using in addition to map editions orthophotography from a 2001 sortie). This research, and other sources (e.g. Halcrow Maritime, 2000a; Posford Duvivier, 1993; 1997; 1999; 2001, Mouchel Consulting, 2001a and b) also attempt estimations of the input of coarse (gravel sized) sediment from the release of flint from eroded Chalk and the infill of Coombe Rock in the several truncated dry valleys. With the exceptions of the work of Moses et al. (2001) and Dornbusch (2002; 2003), the additional contribution from shore platform abrasion are excluded from these calculations (see section 2.3.2 on shore platform erosion).
Contemporary erosion rates are less than those calculated for the past 100-150 years where basal protection provided by massive seawall construction, cliff re-profiling, drainage and other management measures have been undertaken. Most of these date to the early years of the twentieth century, with major works completed in the early 1930s and reconstruction and extensions in the periods 1975-1985 and 2003-2005. The BERM project (Moses et al., 2001) concluded that 36% of former (now potential) yield of gravel from cliff erosion between Rottingdean and Beachy Head has been lost due to shoreline defence and protection.
Most of the unprotected cliffed coastline consists of near-vertical Chalk slopes, whose morphology is controlled by toe erosion exerted by breaking waves and various sub-aerial weathering processes (Photo 1 and Photo 2). Spatial variations of details of morphology reflect complex inter-relations between rock lithology, structural planes, and coastline orientation with respect to wave climate (May, 2003; Mortimore et al., 2004). Dip angles are low, but there are numerous steeply inclined to near vertical joint, fault and other fracture surfaces and tension planes. Two distinctive local morphological variations occur at (i) Newhaven Heights, where Palaeogene sands overlie the Chalk; and (ii) eastern Beachy Head, where slope failure partially in underlying rocks has created an Undercliff and offshore reefs.
Details of Chalk stratigraphy, lithology and structure are given in Mortimore, et al. (2004). Most of the protected cliff frontage is developed in the Newhaven and Culver Formations of the Upper Chalk, which has an estimated flint content of 1.5 (+ 0.5) % of total volume. Between Seaford and Beachy Heads, the Seaford Formation, also Upper Chalk, dominates. In this case, the flint content is 2.5 (+ 0.5) %, but increases to 3.5 (+ 0.5) % at Seaford Head itself and 4.5% east of Bel Tout. These volume estimations, given by Dornbusch (2002), are lower than the average of 5% used in several previous studies.
Cliff top fissures parallel to the cliff face and both plane and wedge failures, are frequent and usually joint-controlled. They promote block detachment via shear failure, occasionally creating substantial fall-slides in excess of 50-70,000 tonnes (e.g. Seven Sisters, 1925; Seaford Head, 1986; Beachy Head, 1813 and 1999 - see Photo 6). Most falls and topples are small scale, involving the detachment of wedge-shaped units of less than 0.5m³ at a frequency of 8-10 years at any specific site. There are several weathering processes involved in reducing the coherence and stability of Chalk cliff slopes, notably salt crystallisation and repetitive freeze-thaw, wetting and drying and heating and cooling cycles. The most potent process is probably frost weathering (Cleeve and Williams, 1988), which is also highly effective in the denudation of chalk shore platforms (see following section). However, there is thought to be a complex link between frost disintegration and salt crystallisation induced by spray. Wetting and drying tends to have most effect on the frequent thin seams and bands of clay and silty clay within the Chalk. Groundwater saturation, affecting pore pressures and capillarity, are fundamental to the triggering of large falls. In places, e.g. eastern Beachy Head, groundwater seepage at the cliff base, especially where dissolution pipes are available, assists mechanical toe erosion. Most researchers (e.g. May, 1971; Posford Duvivier, 1993) have reported that the majority of losses due to both weathering and erosion occur during the harsher climatic conditions of winter. As an example, over 200,000 tonnes of chalk were detached in several failures behind the toe protected cliffs west of Peacehaven Steps during the winter of 2000/2001 (Mortimore, et al., 2004). However, heavy rainfall and storm surge events do occur in summer, when there may also be shrinkage of clays following drought conditions. All of these processes can open up fissures to critical widths and depths. Solution is also likely to be involved in joint and fissure opening, but its role has not been specifically investigated. Overall, there is little direct and convincing evidence for the present day effectiveness of direct basal erosion. Notches and occasional shallow caves may not be contemporary features, but exhumed from beneath former debris accumulations.
May and Heeps (1985) conclude that subaerial processes of weathering and mass movement operating on the Chalk cliffs of East Sussex are more effective than basal abrasion by waves. However, Dornbusch et al. (2008a) fail to identify any satisfactory statistical correlations between magnitudes of cliff retreat and climatic factors such as average precipitation and temperature. They also reject changes in wave climate over the past 175 years as an explanatory factor. Wave energy is partially absorbed by cliff foot debris created by falls, slides, topples, etc. but this is relatively rapidly broken down, with the exception of the boulder-sized material, and removed by wave action. In comparison to most other rock types outcropping on the south-east and central southern coastlines of England, Chalk talus has a relatively short residence time, especially much fine grained material which is removed in suspension, and is frequently replenished. For this reason amongst others, this cliffline is retreating approximately parallel to itself when viewed over longer timescales. May and Heeps (1985) observe that there is no positive correlation between rates of cliff top retreat and the presence/absence of persistent basal talus. The same is also the case for cliff retreat rates and beach width. Dornbusch et al. (2008a) consider that their detailed data sets on beach width changes and cliff top retreat 1873 to 2001 do not justify the acceptance of any direct cause and effect relationship. Even when beaches were at their widest (37m) during this period (they progressively narrowed during the twentieth century, and in some instances disappeared entirely) higher and more energetic waves would have accessed the cliff foot environment. The loss of beach material would reduce abrasion of platform surfaces and possibly the cliff toe, but this relationship is difficult to substantiate. It is perhaps more feasible that accelerations in rates of cliff recession are related positively to reduction of beach elevations. Cliff height also often fails to correlate with mean retreat rates but there is a strong positive statistical relationship between increasing platform width and diminishing cliff top rates of recession. This latter correlation is counter to what would be intuitively expected, i.e. highest retreat rates where platform widths are narrowest, thereby least modifying wave energy. (There are, however, exceptions - refer to the example of Birling Gap discussed by Dornbusch et al., (2008a)). These statements require further more detailed research before their validity can be fully accepted, but point towards the importance of climatic and geohydrological conditions. This and other conclusions are stated in a comprehensive and critical review of current knowledge of the dynamics of Chalk coast erosion by Moses and Robinson (2011). They feature several sites along the East Sussex shoreline and highlight deficiencies of understanding, in particular the debate on the relative efficacy of factors that promote cliff retreat. This has arisen because of the contrasting spatial and temporal scales for which data has been recorded. Most research has relied on measurements of cliff top recession, with a scarcity of evidence for process operation and rates of change at cliff toe: platform junctions. Linking Chalk lithology and cliff erosion rates has proved problematic [refer to opposed interpretation in Mortimore et al., (2004) and Dornbusch et al. (2008a)]. Debris accumulations are known to protect cliff bases from direct wave erosion, but there is insufficient objective data on the relationships between their magnitude, frequency and longevity. A fundamental problem is that of linking measurements of cliff retreat and shore platform denudation. Experimental data on platform downwearing is only available for short-term periods, and this proves difficult to retrospectively extrapolate to give insight into its quantitative contribution to cliff profile change.
Robinson and Williams (1983) calculate that the mean retreat rate for the 22km of Chalk cliffs between Black Rock and Beachy Head is 0.3 to 0.5m per year (1873-1975). For the period 1873 to 2001, Dornbusch et al. (2006; 2008a) derive an estimate of 0.35m per year for the period 1873 to 2001 (0.37m per year, 1873-1927; 0.27m per year, 1928-1973 and 0.28m per year, 1974-2001). These figures refer to cliff top recession (for successive 50m sectors in the case of the work by Dornbusch et al.) and use an average cliff height of 45m. Locally, rates may be at least (and probably greater than) 1.5m per year where single large volume and/or frequent smaller falls have occurred. Events such as these reduce the meaningfulness of average values. Posford Duvivier (1997) estimated a mean rate of retreat for the sector between Seaford and Beachy Head, over the last 100 years, to be 0.4m per year. A retreat rate of 0.5m per year over the past three to five millennia, previous to cliff protection, would yield approximately 25,000m³ per year of gravel (Jennings and Smyth, 1990). This, however, is based on the assumption of 5% flint content, now considered to be an over-estimate (Moses, et al., 2001; Dornbusch et al., 2003; Mortimore, et al., 2004). It also does not factor in the progressive increase in cliff height as the coastline has retreated. Nevertheless, 25,000m³ per year is close to the average potential littoral drift rate under contemporary conditions (refer to Section 3), suggesting that there would be a close correspondence between input and throughput without human modification of process regimes. Any difference might be accounted for by offshore to onshore input.
Several studies have made independent estimates of coarse sediment yield from calculations of retreat rates for specific sectors of this coastline. Most of these assume an average flint content of 5%. For the length of coastline between Brighton Marina and Peacehaven, Posford Duvivier (2001) calculated a yield of 2,200m³ per year for the period 1870-1990. This, however, has now reduced considerably to 640m³ per year, where there are defences. Posford Duvivier (1999) considered that the current yield for the relatively short lengths (total 1.5km) of unprotected cliffs (Photo 7) was a little more than 1,900m³ per year. For the cliffline between Seaford and Beachy Head, Posford Duvivier (1999) proposed that approximately 5,000m³ of flint gravel has been released annually over the past century. Of this total, 1,400m³ is contributed by the cliffline between Seaford Head and The Seven Sisters; 1,300m³ derives from Birling Gap (Posford Duvivier, 1993) and 2,000 to 2,500m³ is introduced by the denudation of the cliffs between Bel Tout and Low Gap, Beachy Head. Halcrow Maritime (2000a) suggest a yield of 5,400m³ per year for the dynamic and complex cliff environment between Beachy Head and Eastbourne. All of these figures are derived from basic formulae using cliff retreat rates (back to 1870) of 0.3 to 0.7m per year and an average cliff height of 60m.
Moses, et al., (2001), Dornbusch, (2002); Dornbusch et al., (2006b; 2008a) and Stavrou et al. (2011) have made detailed re-assessments of available data on the retreat rate of currently unprotected cliffline at the present time between Rottingdean and Beachy Head. This produces a cumulative flint gravel delivery of 4,610+ 890m³ per year, assuming a flint content of the Chalk of 3%. This figure increases to an annual input of 7,700m³ if all sectors exposed directly to processes of abrasion loss since 1870 (determined from map evidence) are considered. The volume proposed by Posford Duvivier (1999; 2001) is 5,640m³ per year though subject to a 7% error margin. It is clear that there has been a substantial reduction in gravel input from cliff erosion over the past two centuries, in comparison to earlier millennia (Jennings and Smyth, 1990). The protection of over one half of the cliffed coastline is clearly the major explanation. Supply of shingle from cliff and platform erosion at the rate calculated by BERM Project researchers for the Holocene period would be insufficient by almost an order of magnitude to account for the quantity of coarse clastic sediment currently in storage in, and in transit through, the contemporary beaches. It must therefore be the case that they hold sediment that has been supplied from other sources, presumably offshore and nearshore, but no longer available. Input from present day and recent shore platform abrasion is a relatively negligible quantity, between 5 and 10% of the quantity derived from cliff recession.
With the exception of the 1.5km length of frontage that remains unprotected between Saltdean and Peacehaven (Photo 7), the Chalk cliff toe has been removed from the influence of wave-induced erosion since the completion of seawall construction. Starting in 1906, this was extended between 1928 and 1936 along the shoreline linking Roedean to Saltdean, but the present bullnose seawall, and walkway, at Peacehaven (Photo 8) was built between 1975 and 1982 over a frontage length of 2.2km (Stammers, 1985). Reconstruction of the defences between the Marina and Saltdean was undertaken between early 1990s and 2004. The principle behind the new defences is to build a strong wall with (in the case of the last phase) groyne reconstruction and an attempt at damping wave energy at the foot of the wall using rock armour. This approach was derived from the view that the coast in this area had insufficient shingle feed for groynes to accrete beaches of any useful size (Winfield, 2004; DEFRA, 2007). The absence of protection at Telscombe Cliffs is interrupted by a headwall around the Portobello wastewater discharge facility.
Before protection, the 25-44m high cliffs, which truncate a sequence of dry valleys, were nearly vertical, and the cliff top was retreating at rates of between 0.76m per year at Roedean, 1826-1897 (May and Heeps, 1985) and 0.35m per year at Western Peacehaven, 1873-1962 (May, 1966). Various authors have calculated retreat rates for specific locations between the mid-nineteenth century and the 1940s; these average, for the cliff face and top, at about 0.40m per year (Thorburn, 1977; Mott MacDonald, 1997; Gifford Associated Consultants, 1977; Mouchel Consulting, 2001). For the sector between Brighton Marina and Portobello, Stavrou et al. (2011) calculate retreat rates for 1873 to 2005. The mean figure for this period is 0.22m per year, with progressively declining rates between 1931 and 1980 as a direct result of the building of cliff foot protection structures. For the unprotected cliffs east of Saltdean, an average retreat rate of 0.32m per year has prevailed over this 132 year period.
In addition to seawall construction, several sections of the former cliffline were regraded, by hand, to angles of between 70 degrees and 80 degrees in the early to mid-1930s (Palmer, et al., 2002). This was undertaken to reduce the hazard of rock falls and slides originating from continuing sub-aerial weathering and mass movement. The Chalk to the west of the Portobello outfall is mechanically weaker than it is at, and east of, Peacehaven (Posford Duvivier, 2001; Mortimore et al., 2004). It is also intensely fractured and fissured in places, especially close to the cliff top and adjacent to truncated shallow dry valleys infilled with Coombe Rock and the Black Rock Interglacial Raised Beach (High-Point Rendel, 1999; Palmer et al., 2002). East of Peacehaven, chalky loessic silt has infilled fissures, reducing mechanical strength. Planar and wedge failures are the most usual form of cliff face degradation that continue to operate; their size and shape is controlled by joint set orientation and their intersection by faults (Mouchel Consulting, 2001; Mortimore, et al., 2004). Several falls and topples directed by lithostratigraphical variation have occurred in recent years, probably induced by high pore pressures following heavy or prolonged rainfall. This process was especially active during the winter of 2000/2001, when fifteen failures took place over a 300m length of former, now protected, cliffline now landward of Brighton Marina. It has been estimated that weathering, especially due to freeze-thaw, becomes active 15-20 years after cliff trimming. An example, at Black Rock, Brighton Marina, occurred in 1971 (Corbett, 1990). Most failures involve relatively small quantities of material, but a few larger scale events have removed between 1 and 10m of the cliff top. Instability tends to originate in the upper part of the cliff face and propagate downwards (see Mortimore et al. (2004) and Lawrence et al. (2007) for further details, together with a discussion of the management practices of the consequent risks to public safety). There is evident potential for larger magnitude failures along joints and reactivated relict slip planes, especially where basal protection modifies groundwater movement (Corbett, 1990; Palmer et al., 2002; Mortimore et al., 2004). Moses and Robinson (2011) state that three quarters of a (undated) rockfall of 2,200m³ at Friars Bay was removed from the inter-tidal zone in less than four months. Current rates of cliff recession are difficult to quantify, as failure events are apparently spatially random and a temporally-integrated periodicity has not been established. Moses, et al. (2001) estimated an approximate rate between 1873 and 1975 of 0.26m per year, lowest east of Telscombe Cliffs. Stavrou et al. (2011) however calculated a much lower rate of 0.01m per year between 1970 and 1980, but accelerating to 0.05m per year 1980 to 2005 as a result of the several failures triggered by intensive rainfall events during this period, notably the exceptionally wet winter of 2000/2001.
Posford Duvivier (2001) calculated that cliff and platform erosion between Brighton Marina and Peacehaven yields 2,200m³ per year of gravel, derived from the release of flints from the Chalk. Input of coarse clastic material is thus no more than 0.5m³ per metre (Posford Duvivier, 2001).
Thorburn (1977) calculated that the 30-60m high cliffs of this sector, which are unprotected along the length of Friars' Bay, receded at a mean rate of 0.46m per year, 1925-1955. Castleden (1996) gives a similar figure, and Mouchel Consulting (2001) propose a maximum rate of 0.6m per year for the period 1875 to 1975. Gifford Associated Consultants (1997) state that cliff top retreat between 1875 and 1980 was 0.4 to 0.5m per year. Moses, et al. (2001) calculate a contemporary rate of 0.25m per year. Conjugate joint sets control cliff morphology, giving a near vertical slope and promoting occasional plane failures.
From Harbour Heights to Castle Hill, Newhaven, the Chalk is overlain by Palaeogene sands and clays, preserved in a shallow syncline. Although the cliff base at Castle Hill is fully protected by the wide gravel beach that has accumulated against the western breakwater, the Palaeogene sediments have produced a dynamic 15-20 degree hummocky upper slope. Shallow translational sliding, gullying and mudflows are all active processes, particularly in winter, with a major slip occurring in 1943 (Castleden, 1996). Debris aprons, partly vegetated, have accumulated at the cliff base, concealing parts of the Chalk outcrop. This has created a distinct undercliff feature, subject to advance when sub-aerial processes are especially active (as during the exceptionally wet winter of 2000/2001). Prior to the construction of the first breakwater and groynes in the early nineteenth century, Palaeogene sediment was rapidly removed by wave action. The wide beach now largely prevents this, and material delivered to the cliff base by mudflows, gullying, etc. is now artificially managed. Where Palaeogene sediment does reach the unprotected cliff base at Harbour Heights, it is very quickly dispersed offshore via suspension transport.
Solution pipes in the chalk below Palaeogene rocks direct the downward movement of groundwater, and may provide potential sites for cliff slope failure once exposed in cliff faces.
Defence structures are limited to tetrapod blocks and a revetment at Splash Point and a short sea wall below the east-facing cliffs at Cuckmere Haven. Thus cliff erosion and retreat is largely unrestrained, producing near vertical faces up to 85m in height. Palaeogene sediments overlie the Chalk to an average depth of 9m, thickening eastwards. They have collapsed into vertical solutionally-widened joint planes, or pipes, which in places descend to (and some below) the cliff base. Former pipes are evidenced by circular shallow dish-like features on the adjacent shore platform. Wedge and plane failures are a regular occurrence, with shear planes defined by inclined fractures; flow slides are more characteristic where the Palaeogene overburden is thickest. Solid Chalk is replaced by Coombe Rock at Hope Gap, an infilled truncated dry valley, thus locally enhancing erosion rates. Caves have developed along joints and other primary fractures, and may have contributed to occasional block falls.
This sector of cliffline is more exposed to wave energy than those to the west, a fact apparently confirmed by calculation of retreat rates since the 1870s. (Williams (2005) recounts that in 1850 this cliff was dynamited to create material that would function as a 120m long groyne designed to reduce the loss of shingle from Seaford beach to the west. Some 380,000 tons of material was dislodged, setting back the cliff face by 30m. In the event, much of the chalk debris was relatively quickly removed by wave action, though the flint content had greater longevity). Thornburn (1977) calculated maximum rates of 1.26m per year, 1925-1955, though both Castleden (1996) and Gifford Associated Consultants (1997) consider cliff top recession at Seaford Head (Photo 4) to be 0.3m per year, 1875-1980. Moses, et al. (2001) state that contemporary retreat is approximately 0.26m per year taking the cliffs of this unit as a whole and factoring in the protective role of basal debris from cliff falls. Bedwin (1986) considered that less than half of the original Iron Age hillfort at the top of Seaford Head now survives, due to cliff erosion. Making some allowance for the probable location of this defensive structure in relation to the position of the cliffline at the time of its construction, based on its surviving plan, he deduced a retreat rate of between 0.25 and 0.35m per year over the past 2,500 years.
Cliff height increases progressively eastwards to 70m, with greater morphological complexity in the same direction.
The cliffline of the Seven Sisters (Photo 1) is developed at right angles to the dip of the Chalk, and truncates six backslope dry valleys. The original coastal planform would have been more indented, but it now has a straight north-west to south-east orientation. This is due to a complex combination of the trend of primary rectilinear joint planes; a low elevation shoreline platform and full exposure to only moderately refracted south-west approaching waves (May and Heeps, 1985; May, 2003). This sector of the East Sussex cliffline may owe its character to the subservience of structural controls to wave energy at the cliff base (May, 2003). Small volume falls and topples are frequent, with large-scale events occurring once every 50 to 60 years. Basal debris derived from cliff falls has a short residence time as it is broken down and redistributed quickly, forming a flint and chalk clast dominated fringing beach. Cartographic and photographic evidence for the past 100 years indicates that the Seven Sisters have maintained a near vertical face over that period; this suggests equilibrium between cliff top and cliff base recession and local dominance of basal wave erosion. Although some authorities state that the retreat rate is lower than for the immediately adjacent cliffline sectors, May (1966) calculates it to be 0.51m per year for 1873-1962. Gifford Associated Consultants provide a figure of 0.6m per year for the 1km stretch of shoreline west of Birling Gap, 1875-1962; May (1971) considered that cliff top retreat may have been over 0.9m per year following failure events, and Thorburn (1977) reports 1.25m per year recession between 1973 and 1975. Moses, et al. (2001) calculated an average rate of 0.42m per year (range of between 0.11 and 0.57m per year) at the present time for the sector between Cuckmere Haven and Beachy Head; the only rate specific to the Seven Sisters is 0.70m per year for 1873 to 1997 (Dornbusch et al., 2008). The consensus view is that this is an actively eroding sector of cliffline, at rates above the regional average. There are, however, no specific calculations of coarse clastic sediment yield from the release of flints. Posford Duvivier (1999) calculated shoreface erosion to be a mean of 5mm per year. Rates of platform lowering along this sector are also considered to be above average for the East Sussex coastline (Dornbusch et al., 2007). Williams et al. (2004) describe and analyse a major cliff fall that affected the Seven Sisters cliffline in 1914, which had an estimated volume of 12,500m³. Debris from this singular event travelled as a narrow projection some 75m. Across the adjacent shore platform; its longevity was not recorded. No comparable event has occurred subsequently.
The cliffs at, and immediately adjacent to, Birling Gap (Photo 9) have been subject to detailed inspection by several authorities, not least because of the threat posed by their retreat to the survival of clifftop properties (McGlashen, et al., 2008). A public enquiry in 2001 generated several independent analyses, all of which observed that recession rates here are faster than to the east or west. This is partly the result of the presence of mechanically weak Coombe Rock in-filling a sectioned dry valley, with intensely shattered Chalk bedrock beneath that descends below normal beach crest level (Robinson and Williams, 1998). In addition, the Chalk to the immediate east of Birling Gap lacks marl seams and has a lower density of major joints. Erosion may also be facilitated by the lower elevation of the shore platform here, which extends 150 to 200m offshore. However, as observed by Robinson and Williams (1998), it might have been expected that the low mechanical strength of cliff material exposed here would have created a re-entrant in the shoreline plan. Coombe Rock infill fails readily when saturated, but only yields small amounts (1-10m³) per event. In contrast, joint set controlled toppling failures in the Chalk to either side create occasional detached rock masses of up to 10,000m³ in volume, such as the Baily’s Brow event in 1925 (Mortimore et al., 2004). The latter occur with much less frequency than the former, which may partly explain the nearly uniform retreat rate. Chalk debris and flints released from the cliffs by erosion create an intermittent beach.
Wealden District Council, and its predecessor authorities, have conducted annual monitoring of cliff top retreat since 1951. The average rate for the period 1951- 2001 is 0.56m per year (maximum of 2.55m per year for the winter of 1973/4). Robinson and Williams (1998) calculate recession to have been 0.68m per year 1873 to 1975 based on analysis of successive editions of Ordnance Survey maps. Posford Duvivier (1997) propose 0.77m per year and Dornbusch et al. (2006; 2008) 0.71m for the same period. Cleve and Williams (1987) give a similar figure of 0.7m per year for the period 1873-1976. May (1971), in a more detailed analysis, proposed a rate of 0.91m per year of clifftop retreat, 1875 to 1961, whilst May and Heeps (1985) noted considerable temporal variability - e.g. between 0.28 and 0.98m per year, 1951-1962. This is emphasised by Moore, Collins and King (2001), who integrated cliff foot and cliff top retreat, 1874-1975, to reveal a range between 0.64 and 1.41m per year. Dornbusch et al. (2008a) calculated an average rate of 0.61m per year, 1873-2001, with a slight decline since 1973. May (1971) was able to elucidate magnitude-frequency characteristics of slope instability events at Birling Gap. He reported considerable year on year variability, with 42% of cumulative losses from slips and falls, 1950-1962, occurring in just two winter seasons. 70% of recorded failure events each detached less than 11m³, which together were only 12% of total loss. Lower frequency, higher magnitude events, all of them wedge or planar failures, were therefore dominant. Most erosion took place during winter months due to a combination of frost weathering, prolonged rainfall, increased pore pressures and storm waves, a continuing feature dramatically emphasised by losses to erosion during the succession of high energy winter storms of 2013/14 that were seven times the norm recorded over the previous forty years. However, May (1971) recorded summer losses that were induced either by late summer storm surges effecting basal notching or by drought-induced shrinkage, followed by intense rainfall. This was especially effective in Coombe Rock and overlying silty drift sediments. Dornbusch et al. (2008a) discuss the linkages between cliff retreat and shore platform geometry at Birling Gap. Between 1873 and 2005 the platform almost doubled in width, by 80m, whilst the height of the cliff: platform junction with reference to mean sea level increased. These changes are considered to be directly related to a reduction of the rate of cliff recession over this period.
Beachy Head (Photo 6) is a morphologically complex cliffed headland reaching a maximum height of 165m whose south-facing slopes are cut into the backslope, and east and south-east-facing slopes are sectioned through the scarp slope, of the South Downs cuesta. May (2003), Castleden (1996) and Mortimore et al. (2004) distinguish several subdivisions based on cliff height, morphological properties, structural attributes and lithological variations. Between Birling Gap and Beachy Head Lighthouse, there are pronounced fault-bounded failures that generate major falls and slides. Each large-scale event yields 50-100,000m³ of Chalk, with consequent large debris fan accumulations running out over the inner part of a wide shoreline platform. These are relatively rapidly broken down to fine grained material that is removed in suspension by waves and produce only small quantities of gravel sized sediment stable on local beaches. Arcuate shaped ridges of large Chalk boulders mark the limit of earlier falls and slides, and increase the surface roughness of the platform. This, in turn, absorbs wave energy that might otherwise impact at the cliff base. A major flow slide and fall occurred in January 1999, probably triggered by critical pore water pressure in the controlling joint(s) and/or fault(s). This event released some 75 to 100,000 tonnes, burying the cliff base to a height of up to 10m and extending seawards to the base of the lighthouse as a 150m wide talus cone (Mortimore et al., 2004). Basal notching is active, creating comparatively short-lived caves. One example, Parson Darby's Hole, was artificially enlarged in the eighteenth century to provide sanctuary for shipwrecked crews and a base for smuggling; it had totally disappeared by 1930 (The Times, 8 November, 1958).
Gifford Associated Consultants (1997) calculated a cliff top retreat rate of 0.3 to 0.5m per year, 1875-1979 for this sector, with a mean of 0.35m per year. Some parts, however, are comparatively stable, receding at a rate of 0.1m per year (Moses, et al., 2000; Dornbusch, 2008a). Here and west to Cuckmere Haven the highest cliffs are eroding at the lowest rates. The relocation inland of the Belle Tout lighthouse in 2000 (McGlashen, 2003) was in response to a calculated site specific retreat rate of 0.19m per year. As cliff failure is episodic, controlled by the rate of opening up and downward penetration of tension fissures, mean rates of recession are somewhat notional. Larger events are separated by intervals of 5-30 years (Robinson and Williams, 1998).
Between Beachy Head lighthouse and Falling Sands, cliff height varies between 55 and 160m, with periodic large falls dominating the erosion system. These are controlled by near vertical primary joints. "Bottom up" failures; flow slides; and plane, wedge and slab failures also occur, creating both spatially and temporally discontinuous debris stores at slope angles up to 60 degrees (May, 2003; Mortimore et al., 2004). The fronting shore platform is of variable width, but provides significant protection from the full force of wave impact against the cliff base. May (2003) distinguishes four downslope sequential cliff slope segments, though not all are developed continuously around the headland. The upper free face is broken by tensional fissures, which created a set of pinnacles ("The Seven Charles") in the early eighteenth century. These fell, one by one, over some 150 years up to 1870. May (2003) states that basal wave erosion selectively picks out lines of weakness to create ephemeral caves, and has created a "fretted" outline. Basal debris stores have a characteristic threshold angle of 30 degrees; fine material is progressively flushed out, leaving boulders as the dominant particle size. These provide at least temporary protection from wave attack and can constitute minor headlands, confining small "pocket" beaches of chalk rubble and flint gravel. Some parts of the cliff base appear fresh and lack evidence of basal notching. These have probably been recently revealed following the natural removal of large amounts of debris, which may have a residence time of anything between 5 and 50 years. There are few estimates of rates of cliff erosion; Gifford Associated Consultants (1997) state 1.1m per year, 1875-1979.
Between Falling Sands and Holywell (Eastbourne), the east-southeast-facing cliffs have been developed in Chalk overlying Upper Greensand and Gault Clay. The geological strike runs approximately parallel to the coastline trend, with both older formations outcropping over the shore platform. The Upper Greensand forms a staircase-like set of inter-tidal ridges, or reefs, due to its outcrop being repeated by slip faulting (Castleden, 1996; May, 2003). The largest example of this is Head Ledge. The effect of their presence is to reduce the potential of wave erosion at the cliff base. The stratigraphical succession here has created conditions for rotational slope failure, with marine erosion of the Gault Clay undermining the overlying Upper Greensand and Marly Chalk. Cliff morphology is thus characterised by an upper backscar and a sequence of semi-rotated blocks making up an Undercliff; these indicate past relatively small scale failure events. An aquiclude at the lower cliff base probably assists hydraulic erosion by breaking waves (Castleden, 1996). Whitebread Hole and Cow Gap are both truncated chalk scarp face coombes infilled with relatively weak colluvial sediments. Cliff recession rates have not been calculated for this sector. The geomorphology of the shore platform is briefly described by Castleden (1996) and May (2003) who establish that it is partly cut across landslide blocks. Rates of vertical platform abrasion and horizontal expansion at this location are not reported in the literature.
Halcrow (2001) calculated an annual loss of 540,000m³ of material from erosion of the cliffs and platforms composing the entire Beachy Head promontory. However, almost all of this is fine sediment that is removed offshore. Only 5,400m³ is coarse clastic (gravel-sized) flint and chalk material that contributes to local beaches and can potentially move eastwards downdrift. This assumes that the average flint content of the various chalk divisions outcropping at Beachy Head is 1%. Mortimore, et al. (2004) indicates a content closer to 2%. Moses et al. (2001) do not provide data on flint yield for the entire length of Beachy Head, but a 1-2% content for the south-facing cliffs may be a realistic estimate. Posford Duvivier (1999) propose that 2,500m³ per year of flint is released from the erosion of the cliffs, and reworking of debris stores, between Birling Gap and Beachy Head Lighthouse. This quantity is based on the assumption that the flint content of the in situ chalk is 5% of volume, which is probably an over-estimate.
2.3.2 Shore Platform Erosion
The Chalk cliffline of East Sussex is extended seawards by a shore platform that varies in width, at mean low tide, of between 100 and 200m. Overall gradient is mostly below 5 degrees, (maximum of 7 degrees, minimum of 0.5 degree) but micromorphology is provided by sequences of near-horizontal benches separated by "risers" between 0.1 and 0.5m in height. The backing cliffs slopes are commonly between 60-80 degrees, but the cliff: platform junction is often obscured by the presence of a narrow gravel beach and/or groynes and seawalls (Rottingdean to Peacehaven), and rip-rap (east of Splash Point, Seaford). Gradient is usually steepest adjacent to the backing cliff, and declines seawards between High and Low Water marks. Seawards of the upper beach, platforms are characteristically cut into exposed Chalk bedrock. Scattered patches of fine to medium gravel clasts, forming a locally thin veneer, sometimes temporarily conceal the rock substrate. Common microtopographic features are flat topped flint capped microcuestas, gutters and runnels, some 0.5 to 4.0m apart, 0.25 to 0.5m in width, orientated perpendicular to the low tide shoreline. Their density of dissection increases seawards (down gradient), probably as a function of the progressively longer duration during tidal cycles that these features channel flowing water. At their terminal point, some runnels can be 0.5m deep and 1.0m in width; a few appear to extend below maximum low water (Robinson and Jerwood, 1987a and b). Ridges between gullies are frequently developed on bands of flints.
The lower part of many platforms are often partially, and occasionally wholly, covered by seaweed. Their gently sloping surfaces can be traced below low water, where in some cases they appear to be replaced by submergent reef-like forms.
Platform elevation with respect to mean sea-level is regulated either by position within the local tidal range or by inequalities in the relative erodibility of the substrate. Wright (1967) has noted that the cliff: platform junction between Cuckmere Haven and eastern Beachy Head coincides with the level of mean high water at spring tides, but descends to that of mean high water during neap cycles in the vicinity of Birling Gap. In the latter case, fractured, deeply weathered Chalk infilling the dry, truncated valley, exposed in the modern cliff face (and presumed to extend seawards) accounts for its low elevation close to mean low water springs (Robinson and Williams, 1998). A lower than average elevation is also apparent adjacent to Seaford Head, where there are several small but permanent pocket beaches. It has been suggested that locally enhanced sub-beach lowering of the upper platform surface may be taking place (May, 2003). This might be promoted by processes such as solution, though given that there is a higher than average rate of supply of flint debris at this site, abrasional lowering is a plausible mechanism. Wright (1970) has made the general point, in relation to various Chalk platform sites along the East Sussex and Kent coastlines, that platform elevations below the regional mean may be a direct function of relative exposure to wave energy. None of these competing hypotheses have been examined in sufficient detail to provide safe generalisations on cause: effect relationships.
A local variant of platform microtopography occurs near Hope Gap Beachy Head, where there are a set of circular depressions with raised rims, in addition to slightly sinuous gutters developed approximately along the principal joints. These have been interpreted as truncated solution pipes (Castleden, 1996) developing examples of which are visible in the face of the cliffs at Seaford Head.
Weathering and Erosional Processes
Researchers at the University of Sussex have made original contributions to both qualitative and quantitative understanding of the processes determining the macro and micro morphological characteristics of Chalk shore platforms, focusing on the East Sussex shoreline. Their salient observations are given below; details of field and analytical techniques are provided in several papers, e.g. Williams et al. (2000); Dornbusch et al. (2007) and Dornbusch and Robinson (2011).
Jones (1981) provisionally identified three "activity zones" on the Chalk platform in the vicinity of Brighton Marina, viz:
- Upper platform, adjacent to the beach, where abrasion is dominant as a result of wave-induced movement of gravel clasts;
- Middle platform, where Chalk removal is principally the result of solution and bioerosion, with abrasion confined to runnels and gutters;
- Lower platform (possibly extending to below mean low water) where erosion is mainly undertaken by wave quarrying (block detachment) and abrasion. This zone is dominated by the progressive breakdown of Chalk and flint debris, and is a potential source of fresh input of beach material.
Jones (1981) and some earlier observers, also suggested that freeze-thaw weathering may play a significant role in zone (b). Other physico-chemical weathering processes of in situ bedrock reduction that may be important include repetitive wetting and drying cycles, due to twice daily tidal submergence and emergence; salt crystallization during the period of low tide exposure, and diurnal insolation heating/evaporative cooling. All of these processes fluctuate, in terms of their relative significance, on a seasonal basis. Biological weathering and erosion is also subject to both diurnal and other longer timescale periodic variations in effectiveness, though processes such as grazing by algae and rasping by molluscs will take place in at least one zone during all states of the tide and throughout the year.
In recent years, several researchers have undertaken quantitative research on the efficacy and rates of operation of some of these processes. To test the widely held, but previously unsubstantiated view that there is little effective movement of gravel clasts across the middle platform zone Ellis (1986) monitored paint-sprayed flint particles at sites between Rottingdean and Friars' Bay over a seven-month period. He was able to detect small net eastwards movements of both gravel and sand, but the quantities involved were very small. Concentrations of gravel in runnels were apparent, with movements of up to "several" metres during single tidal cycles. Thin spreads of coarse sand and fine gravel were observed covering relatively small areas of middle and lower platforms following storms, possibly implying net offshore to onshore transport. These did not persist, with the exception of the runnels. Ellis (1986) therefore concluded from this evidence that wave-induced abrasional erosion is currently relatively insignificant. This is even more likely where there is a persistent cover of seaweed. However, he did concede that abrasion, particularly during winter months, might be more important when he evaluated the results of repetitive microerosion meter surveys of a total of 44 sites between Black Rock (Brighton) and Peacehaven. These cover a range of elevations and incorporate upper and lower, as well as middle, platform zones. Andrews and Williams (2000) observe that many of Ellis's sites have high limpet population densities, thus implying bioerosion as a potentially significant contributory process to platform reduction. Indeed, Ellis (1986) makes a similar point, but his research approach did not include the collection of appropriate data. Dornbusch and Robinson (2011) discuss and analyse high resolution measurements obtained by microerosion meters and laser scanning of block detachment and microstep backwearing undertaken at several sites between Roedean and Hope’s Gap. They are of a similar magnitude to rates of surface downwearing, with the two processes combined effecting a Chalk volume loss of between 0.0012 and 0.0100m³ per m².
Both Ellis (1983, 1986) and Andrews (2000) present micro-erosion data indicating that rates of platform reduction are highest in the upper zone, adjacent to beaches, as there is strong circumstantial evidence in favour of the dominance of abraisonal scour. It is also an area where limpet colonisation does not occur.
It is thus apparent that the concept of the 'wave-cut' or wave abrasion surface can only be applied to a small area of the Chalk shoreline platforms of East Sussex. This is reinforced by the lack of any statistical correlation between platform orientation and microtopography with respect to (i) wave fetch and mean wave height; and (ii) mean rates of downwearing (Ellis, 1986). The implication must be that other geomorphological processes are more significant. Three that have received some attention are frost weathering; salt weathering (haloclasty) and bioerosion.
Robinson and Jerwood (1987a and b) and Jerwood et al. (1990) have examined the evidence for frost weathering and salt crystallisation, using three groups of sites, viz: (i) Saltdean to Friars' Bay (west of Newhaven); (ii) Seaford Head, and (iii) Cuckmere Haven to Beachy Head. The impact of freeze-thaw cycles was evaluated using field data collected during periods of exceptionally cold temperatures in January 1985 and February 1986. They describe extensive cracking and spalling of coherent Chalk, especially where it is adjacent to thin, irregular intercalated bands of clay. Detached fragments were observed to fall, or be washed, into runnels, where they either broke up after a few tidal cycles or suffered volume loss (presumed to be due to abrasion). The percentage frequency of spalling per unit area decreased down-platform, from 36% close to the limit of the upper beach to less than 1% at the seaweed encrusted low tide platform edge. In addition to spalling, mostly of small projecting surfaces, intense sub-vertical cracking by repeated freezing and thawing was evident. This particularly affected detached rock masses, including those of boulder size. In comparison to the Chalk, very little spalling, flaking or cracking of flint nodules was observed. However, the long-term role of frost weathering in releasing flints (which occur in bands, and make up 2-5% of the Upper Chalk areas studied) may be significant. In this context, Dornbusch and Robinson (2011) report a lower rate of downwearing for the period 2001 to 2007, during which there was a lower incidence of severe frosts than in the preceding 30 years. (However, this might also be related to less storm wave impact during this same period.)
Disintegration of Chalk by repeated freeze-thaw cycles takes place once air temperatures fall below freezing point but seawater remains unfrozen. Because of its salt content and the dissipation of latent heat released by liquid: solid phase change, seawater has to cool to -2.5°C before freezing. In practice, as laboratory experimentation demonstrated, sub-surface temperatures in seawater-saturated Chalk may need to decline to between -4.5°C and -6.0°C before freezing commences. Thus, upper platform areas are more susceptible to freeze-thaw effects because of their longer exposure to cooling effects during each tidal cycle. The high thermal conductivity of seawater ensures rapid thawing during the rising tide, therefore minimising the cooling period over platform areas adjacent to mean low water. Mid and upper platforms exhibited substantially more evidence of frost weathering, especially the corners and edges of runnels and detached boulders with a larger ratio of surface area to mass and relatively greater exposure to wind chill. Robinson and Jerwood (1987a) considered that although their field data relate to two exceptionally cold periods, the processes described are arguably likely to occur during most winters. They observe that a single frost cycle may be sufficient to initiate Chalk surface spalling, thus its cumulative effects as a factor in the sculpting of shoreline platform relief must not be underestimated. Whether, or not, frost weathering modifies microrelief is an open question, though the evidence is in favour of its effect in widening runnels and wearing back micro-scarps.
It is probable that frost weathering does not operate alone, but combines in a complex way with salt crystallisation that occurs in the micropores and cracks of Chalk surfaces. The laboratory experiments undertaken by Robinson and Jerwood (1987b) did demonstrate that the two processes when working together are potentially highly destructive; however, the dynamics of their inter-relationship are not clear. Most damage is likely to occur in winter, when warming is least and near surface saturation will persist even over the highest parts of platforms. Water will also collect in pools, thus potentially promoting salt crystallisation during periods of tidal emergence.
The role of bio-erosion has recently been evaluated by Andrews (2000) and Andrews and Williams (2000), whose quantified approach provides some objective evidence for processes previously implied by others (e.g. Jones, 1981; Ellis, 1986). Their work focuses on the effects of the limpet Patella vulgata, which colonises large areas of entire platform surfaces except for sites close to the upper beach. They are most numerous around mid-tide levels, and co-habit with other fauna, e.g. the polychaete worm Polydora ciliata and piddocks such as Pholas dactylus, that also contribute to bioerosion through drilling and burrowing lower platform surfaces. Patella vulgaris removes Chalk through grazing (i.e. ingestion) and by excavating hollows ("homescars") for shelter after feeding. Through analysis of the calcium carbonate content of captive adult specimens, it was calculated that each consumes approximately 5.0g per year of Chalk. Based on estimations of limpet population densities, this converts to a mean rate of platform lowering of 0.15mm per year. Rates may be as high as 0.5mm per year in those areas of highest concentration. Other researchers (e.g. Ellis, 1986) have deduced mean rates of platform lowering, due to all processes combined, of approximately 2.3mm per year; however, this figure appears to decline to close to 1.3mm per year where limpets are present in significant numbers. Thus the question arising is whether they provide a protective role or simply colonise more stable areas where erosion and weathering processes are less active.
In an earlier study, restricted to one site (Hope Gap, Beachy Head), Day (1990) calculated a rate of platform lowering of 0.28mm per year based on the carbonate content of limpet faecal debris. Andrews and Williams (2000) comment that the results from their study are probably conservative. This is because they confined measurements to ingestion and radula rasping, omitting the effects of shell edge attrition and dissolution within homescars. Other questions remain, in particular the ecological relationships between limpets and barnacles. The latter also contribute bioerosion, but it is uncertain how or if their population densities and productivities are affected by the presence of limpets. Solutional weathering and erosion by algae is another source of biological reduction of platform surfaces that awaits further research.
Rates of platform lowering and sediment yield
Researchers at the University of Sussex have undertaken experimental field and laboratory based research into the downwearing of Chalk platform surfaces characterised by contrasting microtopographies and locations with respect to the wave climate and tidal frame of East Sussex. This work has also been linked to quantitative estimations of the release of flints as a component of the budgets of local beaches. Their work is summarised below; for measurement and analytical techniques, refer to Williams et al. (2000), and Dornbusch et al. (2006a; 2011).
Ellis (1983; 1986) reported rates of vertical erosion of between 2mm and 1cm per year, based on repetitive micro-erosion meter measurements of 44 sites between Brighton Marina and Peacehaven. These sites combined a wide range of platform elevations, microtopography and positions within the tidal frame. The majority produced rates between 1.2 and 8.5mm per year, with a mean of 2.32mm per year. Lowest rates, between 0.001 and 0.003mm per year, were associated with sites lacking abrasional material, where covered with seaweed or with little or no development of microrelief. Highest rates correlated with localised areas of active abrasion, such as upper platform surfaces adjacent to beaches or in gutters and runnels transporting debris (refer also to Moses and Robinson, 2011). For almost all sites, some 50-70% of erosional loss occurred between December and March/April, (Foote, et al., 2006) thus implying the significance of frost weathering, low evaporation rates and higher energy waves in areas of active abrasion. The role of geotechnical properties of the various lithostratigraphical horizons of the Chalk outcropping on the shore platforms of East Sussex as regulators of rates of downwearing are reviewed in Moses et al. (2006). Moses and Robinson (2011) provide a comprehensive review of research into platform erosion rates, much of it featuring the East Sussex coast. The merits and shortcomings of the various techniques used (principally microerosion meters; soft copy photogrammetry and laser scanning) are highlighted, and the authors emphasise that, used individually, they have given some significantly different data on mean rates of platform downwearing. This may be the result of either contrasts in the spatial scales at which they are deployed or the constraints of platform microtopography on resolution of detail. Mean rates of between 2.3 and 3.0mm per year are considered both representative and reliable, but the short timescales of field experiments undertaken thus far may introduce underestimation (Dornbusch et al., 2008a)
Assuming that the Upper Chalk underlying almost the entire shoreline platform of East Sussex has a 4 to 5% average volume content of flint, a platform reduction rate of 2.3mm per year would yield between 300 and 400m³ per year (Moses, et al., 2001; Dornbusch, 2002; Dornbusch et al., 2003; 2008a). This is in the order of only 5-10% of flint release from Chalk cliff erosion, averaged for the past 100-140 years. This is considerably less than the potential yield derived from a flint content of 5-10% assumed in some earlier studies, based on simplistic calculations (see below), and therefore represents a very small input to regional beaches. As coastal protection since the 1930s has excluded a calculated 36% of potential input of flint clastic material from cliff erosion (Moses, et al., 2001), the implications for the present and future littoral sediment budget are clear.
Rates of lowering due to limpet grazing and burrowing, summarised in the previous section, are on average below those ascribed to the combined effects of non-biotic erosion. In general, bio-erosion contributes some 12-15% of total losses, rising to nearly 35% at sites of exceptionally high faunal densities (Andrews and Williams, 2000). Research into frost and salt weathering has been unable to quantify rates of platform denudation specific to these two processes.
Posford Duvivier and British Geological Survey (1999) have calculated mean rates of shoreface erosion for all sectors of the East Sussex coastline. For the sector between Rottingdean and Seaford, a rate of between 1.3 and 4.0mm per year is deduced from the application of a set of assumptions to limited detailed knowledge of nearshore/offshore conditions. The figures for sectors to the east (5mm per year, Seaford Head-Birling Gap; 15mm per year, Birling Gap to the eastern end of Beachy Head) are well in excess of those proposed by Ellis (1986) and others. Using the figures above, and an assumed but questionable flint content of 5% for the entire length of platform, annual input of coarse clastic debris would be approximately 5,500 to 5,700m³.
Jennings (1990) has stated that the level of the Chalk platform seawards of the eastern part of Seaford Beach fell some 3-5cm between 1900 and 1985 (4.1mm per year). Posford Duvivier (2001) report a rate of lowering of nearly 2cm per year along parts of the Brighton Marina to Saltdean frontage over a similar timescale. These very high rates were calculated close to sites of protection structures of known construction date. Mouchel Consulting (2001) suggest similarly high rates (1.0-1.5cm per year) for platform surfaces adjacent to the seawall between Saltdean and Peacehaven. Further confirmation is provided by Robinson and Williams (1998) who describe 'ribs' of Chalk beneath concrete groynes built in the early 1970s at Peacehaven standing almost 0.5m above the adjacent inter-groyne platform. See Dornbusch et al. (2007) for further similar observations and measurements. This gives a rate of lowering of 1.6cm per year. These high rates of platform reduction do not necessarily imply an acceleration in recent years, as all are located in areas of wave turbulence and therefore localised very high abrasion losses. They give an indication of maximum rates of lowering in the upper zones of platforms, where abrasion is the likely dominant process. Dornbusch et al. (2007b) provide further evidence that implicates heavy machinery used in the construction of seawalls and concrete groynes in the late 1970s as a significant factor accelerating rates of platform lowering in their immediate vicinities. Comparative analysis of air photos taken in 1973 (pre-construction) and 2005 of the shoreline between Brighton Marina and Peacehaven revealed platform denudation rates between four and twenty-five times greater at sites adjacent to seawalls and groynes than for “natural” sites. The average loss of elevation at the former was measured at 0.35m, with a maximum at one site of 2.2m. It is probable that Chalk weakened by the compressive and tractive forces of construction equipment is initially removed by wave action, creating linear channels and troughs. These in turn trap coarse beach material, which deepen them further by abrasional scour. Seawalls built by manual labour in the 1930s provided no evidence for an anthropogenic contribution to platform downwearing.
Platform Evolution
Several authors (Robinson and Williams, 1983; Ellis, 1986; Cleve and Williams, 1987; Andrews and Williams, 2000; Moses, et al., 2001 and Dornbusch, 2002) have stated that there has been no significant increase in width of platforms over the last 150 years of reliable map evidence. As cliff recession has been ongoing, except for those sectors where seawalls have been built since the 1920s and 1930s, it must be concluded that they have maintained a quasi- equilibrium profile. In broad terms, this might be the result of the reduction of wave energy at the cliff: platform junction as the platform widens and less debris is removed. However, the evidence suggests that Chalk debris has a short residence time and fresh inputs of flint material are small. Equilibrium, therefore, must be maintained by an approximate balance between cliff retreat and sea-level rise - a relationship that can be assumed to be nearly constant for the last 5,000 years. The shore platform of East Sussex is therefore a contemporary, not a relict, landform. Ellis (1986) has made the point that if assumed average historical rates of cliff recession of 0.5m per year have operated over the past 2,000 years, a platform 1km in width would have been created. However, to maintain a constant equilibrium gradient over this time period, a rate of lowering of 11mm per year would be required. As contemporary rates are much lower, perhaps platform evolution takes place episodically rather than progressively. The alternative is to assume that measurements of modern platform lowering rates are gross under-estimates, but this seems unlikely. Ellis (1986) has suggested a development model characterised by a regime where frost weathering is dominant over bioerosion and wave abrasion, with short periods of intense erosion, for example during very cold and/or stormy winters, separated by phases of modest geomorphological activity. This could give rise to a sequence of evolutionary changes, starting with a flat surface, which is thereafter diversified by runnel/gutter initiation, frost weathering and bioerosion. Storm-induced wave quarrying creates block removal and micro-scarping between discrete platform levels. This model accommodates both vertical erosion and horizontal retreat, and can progress at different rates at specific locations. If various "stages" were to be defined, the morphological expression of each would be expected at the present time - indeed, they might be co-adjacent. Field mapping supports this, but Ellis's model is far from definitive - there are several alternative sequences, as well as arresting factors.
Given (i) rising sea-level, possibly up to 4mm per year on this coastline by 2030-2040; and (ii) continuation of protection at the cliff: upper platform junction along more than 40% of the length of the platform-fronted shoreline, there are some evident implications for future management. Landward expansion will mostly be excluded, but present rates of lowering will continue - thus the period of submergence will increase and beaches will be more vulnerable to wave erosion. This will enhance the effects of abrasion, but increased water depth and a longer submergence time might suppress sub-aerial and biological weathering and erosion. Whatever the outcome, there will be a reduced input to sub-cell littoral sediment budgets.
2.4 Beach Management: Replenishment and Recycling
Several beach systems along this coastline have experienced long-term net loss of volume and narrowing of their inter-tidal widths. Indices of retreat (and, in a few locations, advance) are given in Gifford Associated Consultants (1997) and Halcrow (2004), and are discussed further in Section 5. Operating authorities have responded in several cases with programmes of beach management involving substantial inputs of recharge materials. Most involve further periodic renourishments and/or routine recycling of sediment from areas of recurring accretion to those of persistent deficit. Williams (2005) and Moses and Williams (2008) provide a critical evaluation of the role and future viability of the recharge of the gravel beaches of Sussex and east Kent, citing the locations listed below. For a specific perspective on nourishment, using a local example but widely applicable, (Dornbusch et al., 2008c) refer to N2 (below).
Since the completion of the breakwaters defining the entrance to Shoreham Harbour in the late 1880s, beaches immediately eastwards (downdrift) have suffered sediment starvation. This is confirmed by apparently 'one-off' replenishments in 1886 and 1902 at Hove (Joliffe, 1972). Sir William Halcrow (1988; 1990) calculated a net reduction of volume of almost 150,000m³ between 1962 and 1988 subject to considerable year on year variability due to changes in incident waves. This was approximately 25% of the "equilibrium capacity" of this beach, calculated theoretically on the basis of average wave climate, mean particle size and unimpeded longshore transport. Loss would have been greater but for periodic dumping of gravel and rubble (from urban waste), estimated at 35,000m³ between about 1970 and 1985. Sir William Halcrow (1967), in a report on the condition of Brighton beach, revealed that some 400,000m³ of spoil dredged from the entrance to Shoreham Harbour was placed on, or close to, the toe of East Beach between 1954 and 1957. Smaller quantities may have been introduced subsequently. As much of this material is likely to have been silt and fine sand, only a small proportion would have been retained. Ward (1922) reported that in the first decade of the twentieth century approximately 90,000 tons of gravel was removed annually from between and west of, the harbour breakwaters. Some was placed on East Beach, though only when seriously depleted. Most was exported as aggregate, much of it overseas.
When Shoreham Port Authority assumed responsibility for East Beach in 1987, it was in a depleted condition. The Beach Management Plan proposed by Sir William Halcrow (1990) proposed a biannual transfer of gravel from the beach immediately updrift of the western breakwater to East Beach. It was suggested that up to 15,000m³ per year might be artificially by-passed, without impacting on the geomorphological or ecological integrity of West Beach (i.e. an amount roughly equal to the calculated drift rate). In practice, the annual average transfer, 1993-2000 was 8,500m³ (Vaughan, 2001), though there were substantial inter-annual variations. For example, no material was moved in 1996, but over 22,000m³ was transferred in 2000. All recharge between 1988 and 1992 used spoil from reclamations for harbour development for groyne bay infilling. Analysis of Environment Agency ABMS profiles for East Beach, 1988-2000 revealed a substantial increase in volume up to 1993, but some marginal reduction between 1995 and 2000 (Vaughan, 2001). Between 2003 and 2007 approximately 60,000m³ were transferred (Halcrow, 2004; Worthing Borough Council, 2009), and 50,000m³ between 2006 and 2011 (Worthing Borough Council, 2012) - i.e. 13,600m³. per year during this eight year period during which East Beach lost a net quantity of 37,000m³. Southwick Beach recorded net accretion of 17,200m³ once input from recharge was subtracted (in 2010 a relatively small additional quantity of recharge material was taken from a site adjacent to the western perimeter of Brighton Marina). It is therefore evident that managed by-passing and recycling of gravel has stabilised beaches up to some 3km downdrift of the harbour entrance, but this condition can only be maintained if this practice is continued. Since 2003 the underlying trend at East Beach has been one of gradual but cumulative erosion, as 97,000m³ would have been lost had there been no recharging.
Analysis of Coastal Monitoring Programme baseline topographic (2008-12), lidar and aerial photography data indicates actual rates of artificial sediment by-passing at Shoreham of approximately 12,000m³ per year (see table below).
Replenishment of beaches has been undertaken at Rottingdean and Saltdean with maintenance of existing beaches in a scheme at Ovingdean and against the eastern arm of Brighton Marina. Sites in question are dominantly flint gravel beaches with a history of depletion long pre-dating the construction of Brighton Marina. Only where beaches have been seen as necessary for amenity purposes have they been artificially created, elsewhere sea walls and rock toe revetments have been constructed to dissipate wave action. In these latter cases, beaches that existed up to the early twentieth century have been virtually lost. A total of 213,000m³ of sand and gravel, dredged from the Owers bank south of Littlehampton, was added between 1994 and 1996 (103,000m³ at Saltdean and 110,000m³ at Rottingdean.) This is approximately 200m³ per metre frontage within the main schemes. Posford Duvivier (1997) state that the recharge material contains 20 to 30% sand, and is close to the mean particle size of indigenous gravel clasts, i.e. 13mm long axis. Beach profile modelling was carried out in conjunction with replenishment, with foreshore slope and crest height designed to maximise capacity to absorb and re-distribute wave energy. Rottingdean, Saltdean and Ovingdean beaches are designed to adjust their orientation, between 11 and 25 degrees, in response to changes in vectors of wave approach (Posford Duvivier, 1997).
Monitoring of the behaviour of Saltdean Beach over 4.5 years (2000 to 2004) revealed an absence of cross-shore sediment exchange between the upper and lower elements of the beach, and that the pattern of movement within the individual groyne bays was independent of groyne type ( whether rock or concrete) and spacing (Dornbusch et al., 2008c). Detailed comparative analysis of beach volumes within successive groyne compartments based on ground survey and photogrammetric measurements for both Saltdean and Rottingdean Beaches (Dornbusch, 2008) indicated that rock groynes allowed throughput transport of sand and gravel between 0.5 to 1.0m (maximum of 1.27m) per day. This was a function of differences in beach cross-sectional area either side of each groyne, with negligible contribution from changes in incident waves. By contrast, concrete groynes inhibited sediment transfers. At Saltdean, beach rotation proved to be significantly less than predicted in the design of the 1996 protection scheme. This work drew the critical observation that possibly up to 50% of the recharge material introduced in the mid-1990s was in excess of requirements for viable shoreline defence.
This now stabilised barrier beach is situated between the mouth of the River Ouse and the terminal groyne at Seaford Head, occupying a frontage of 4.2km. The immediate hinterland is low-lying, and at risk from overtopping and flooding. The built-up environment of Seaford town is especially vulnerable, thus there has been considerable investment in recharge, reprofiling and recycling during recent decades.
Joliffe (1972) provides documentation of the history of Seaford beach since the early nineteenth century, though details of specific levels and volumes are difficult to reconstruct. Erosion and depletion were the dominant features, indicating that gravel was difficult to retain and tended to move rapidly alongshore, or offshore (Williams, 2005). The position of mean Low Water moved 107m landwards, 1879-1961 (60m between 1910 and 1945). The first groynes were introduced at Tide Mills in 1836 (Large, 1989), and a seawall between the Buckle Inn and Tide Mills was built in the 1880s. Joliffe (1972) states that free gravel by-passing of Newhaven Harbour occurred up to an earlier phase of breakwater construction in 1844. However, the pathway of supply between Newhaven and Seaford Beach may have been indirectly via Seaford Bay. After breakwater and pier construction by-passing appears to have been reduced; especially following the lengthening of the western Breakwater, completed in 1890. Significant erosion of the central sector of Seaford Beach occurred within a few years, requiring the insertion of some 80 closely-spaced groynes, recharge of 41,000m³ of shingle from adjacent to the eastern side of Newhaven East Pier and the upgrading/extension of the seawall. The new breakwater also introduced a drift reversal west of Tide Mills because of localised change in wave climate (Large, 1981; Shave, 1989). Although there was some net periodic accretion at either end of the beach, between the Buckle Inn and Seaford Head, e.g. 160,000m³, 1898-1927; and 112,000m³, 1927-1961 near East Pier (Joliffe, 1972), progressive lowering and steepening was experienced up to the late 1970s. This was despite recharge with gravel, taken from Dungeness, in 1936 and 1958; the construction of a new set of alternating long and short groynes in the 1950s, and attempts at sediment recycling (e.g. 26,000m³ between 1936 and 1940.) The latter again utilised surplus accretion against the eastern pier of Newhaven Harbour (Shave, 1989).
The causes of depletion were the subject of considerable debate and controversy by the early 1960s (Hydraulics Research Station, 1963). The conventional view was that the prime cause was the effect of the western breakwater of Newhaven Harbour on substantially intercepting potential longshore drift. This argument was largely based on the historical evidence for eastwards barrier spit migration (see section 1.1), but ignored the fact that there had been several events of beach drawdown and breaching (the most recent of the latter in 1875). It was becoming apparent that the seawall and groynes promoted both drawdown and scour (May, 1966) whilst erosion of the Chalk platform providing the beach foundation was seen as one other possible cause. This was lowered by up to 3m between 1900 and 1950 (Joliffe, 1972).
Experimental studies carried out by the Hydraulics Research Station in 1961-62 (Joliffe, 1964) used physical models, fluorescent tracers and diving inspection to determine movement of gravel on and offshore Seaford Beach. This work demonstrated that gravel clasts had a tendency to move offshore when wave heights exceeded 4m. They did not return if moved into deep water or if they were incorporated into the silt-clay-fine sand layer occupying much of the seabed (see section 3 and section 4 for further detail). This evidence added a further dimension to the debate on the causes of beach depletion, emphasising that net onshore to offshore transport may be a significant reason why coarse clastic sediment is not inherently stable on Seaford Beach.
Subsequent research on regional shoreline evolution (Jennings, 1990; Jennings and Smyth, 1990; Nicholls, 1991) has contributed the concept that the gravel beaches of the East Sussex shoreline are, in part, relict barrier or barrier spit forms. This could be relevant to Seaford Beach, which partly originated from barrier elongation across the mouth of the Ouse. If its sources of supply from offshore have diminished, or have been exhausted, then it is a degenerate accretion structure that would be expected to suffer overtopping, breaching and net loss of volume. This effect would have been accentuated by west to east longshore drift impedance caused by the insertion of training walls, piers and breakwaters at the entrance to Newhaven Harbour. Thus, the latter factor is not seen as the fundamental cause of beach erosion along the Seaford frontage (Jennings, 1990). The idea of "pulses" of gravel supply from offshore may also be relevant, with erosion over the past 200 years progressively removing earlier inputs (Jennings and Smyth, 1990). However, Seaford Bay is deeper than the area offshore the Crumbles, Eastbourne to which this argument was specifically applied. The analogy may therefore be only superficially attractive. The work of the former Hydraulics Research Station (Joliffe, 1964) would appear to discount any significant net onshore transfer of gravel, despite the fact that peak wave energy for the East Sussex coast occurs close to Seaford Bay.
The Southern Water Authority took management responsibility for Seaford Beach in 1981, and commissioned Hydraulics Research to carry out a range of hydro- and morphodynamic studies (Hydraulics Research, 1985; 1986c) to determine appropriate measures to secure future beach integrity (Holmes, 1989). This work concluded with a recommendation to renourish with 1,450,000m³ of gravel, over a frontage length of 2,500m, together with a large terminal groyne to contain eastwards littoral drift. This work was carried out in 1986/7 and involved beach geometry specifications of a 1 in 7 seaward slope; crest height of 6.5mOD and crest width of 25m (Stone, 1990; HR Wallingford, 1996). The actual quantity of gravel added to this beach is stated to be 1,670,000m³ but see Williams (2005) for other figures quoted in various sources.
It was assumed that the previous eastwards littoral drift rate of between 20,000 to 38,000m³ per year would operate after renourishment and re-profiling were complete. In the absence of intermediate groynes, this would create an immediate net accretion of 50,000-70,000m³ per year updrift of the terminal groyne, but with a commitment to routine recycling within the beach system thereafter. Losses of some 15% of volume (approximately 200,000m³) were anticipated within 6 months of scheme completion, mostly due to removal of fines. Monitoring studies carried out after final remodelling of the beach profile in March 1988, using 58 transects (Hydraulics Research, 1988b; 1989b; 1991; Brampton and Millard, 1996; HR Wallingford, 1996) revealed close correspondence between theoretical expectations and actual performance. Losses were less than anticipated at first, but storms in 1989/90, and 1992 generated strong mid/backshore scarping following the consolidation of the matrix between gravel clasts. This assisted wave reflection, creating significant losses from foreshore scour.
Between 1987 and 2007, monitoring revealed a spatial pattern of net losses and gains over different sectors of the beach, with net accretion recorded at both eastern and western ends (Brampton and Millard, 1996; HR Wallingford, 1996; Worthing Borough Council, 2009). These fluctuations were equalised through periodic recycling, with a small net gain for the beach as a whole over this period. Since 1995, this practice of twice yearly recycling has been maintained, with quantities determined by analysis of beach condition for preceding years. Between 1995 and 2003 this input averaged 80,000m³ per year westwards from the Splash Point terminal groyne and 40,000m³ per year eastwards from the Newhaven end of the beach. This was reduced to an annual total of 42,000m³, obtained from a combined surplus from the extreme ends of this beach between 2003 and 2007. Recharge was further reduced in 2009 to 10,000m³ and 5,000m³ in 2010 after an input of 50,000m³ in 2008 (Environment Agency et al., 2012). This more recent trend suggests that an additional natural source of sediment feed is contributing, albeit temporarily. Earlier ABMS data, 1991-2000, indicated an average annual gain of 13,800m³ (maximum accretion, 1992, of 65,000m³, maximum depletion, 2000, of 75,000m³). Management has since sustained this restored and remodelled beach, but clearly Seaford Beach has experienced ongoing erosion during the past 25 years despite a possible auxiliary offshore to onshore feed of gravel from Seaford Bay. This is tentatively identified by HR Wallingford (1996), rather contradicting the conclusions of previous investigators that there was net offshore transport of coarse material. It has not been possible to date to confirm this postulated input, though the recent reduction in the annual quantity of shingle required for recharge (see above) may be indicative. The central section of the beach is illustrated by Photo 10.
Irregular recycling of gravel making up the beach face of the barrier spit across the southern part of the former mouth of the Cuckmere River is practised by the Environment Agency. The main objective is to counteract the effect of net west to east longshore drift of partially blocking the river mouth or reducing the effectiveness of the training walls (Clifton and Cecil, 1999). The volume recycled annually is roughly 5-7,000m³.
3. Littoral Transport (Beach Drift)
Analysis of Coastal Monitoring Programme baseline topography, lidar and aerial photography data, along with all research conducted on longshore littoral transport, from early qualitative/ observational studies to recent numerical modelling, concludes that the dominant pathway is west to east. Local net directional reversal, however, occurs between the western end of the replenished beach at Seaford and the East Pier of Newhaven Harbour. Short-term reversals have also been documented at several locations, most often in association with south-easterly waves during the winter. The dominant process inducing longshore movement is that of breaking waves, with gravel-sized clasts being the principal sediment type. Tidal currents alone are ineffective in moving sediment in the breaker/nearshore zone, but may combine with waves in transporting fine sediment, mostly sand, in the nearshore and offshore zones and across and into estuary/harbour entrances.
Gifford Associated Consultants (1997) propose a conceptual model with estimated rates of longshore transport derived from the LITPACK numerical deterministic process model. Rates are calculated for a sequence of cross-sections through the transport corridor, using EA ABMS profiles (1973-1995); local bathymetry; bottom sediment samples; tidal levels and hindcast wave climate. In addition, measures for the "spread" of bed material sizes; bed roughness and transport velocities for the size-range of sediments potentially involved in longshore transport are used to refine model predictions. The SANDS database provided an important input source for much of this information, and helped to produce adjustments to predictions in areas of more complex offshore bathymetry (e.g. bedrock outcrops) or where obstructions to movement occur such as marinas; breakwaters; long groynes and other local salients.
Calculated transport rates (quoted below) are for potential movement but are incompletely adjusted for groyne spacing and dimensions. Because the latter are impediments to uninterrupted transport, actual drift rates are likely to be significantly lower than predicted. Lack of sediment supply is however, the principal reason why potential rates are rarely attained.
Analysis of Coastal Monitoring Programme baseline topography 2007-12 and aerial photography data indicates an eastward drift rate of 10-20,000m³ per year between the eastern beach of Shoreham Harbour through to Hove, which takes into account groyne storage. This rate includes the 10-15,000m³ per year of sand and gravel that is artificially by-passing the breakwater and harbour channel to protect the western side of the harbour channel.
Halcrow (1988) calculated that some 24,000m³ of both sand (mostly) and gravel (small quantities) annually by-pass the breakwater, of which 14,000 to 20,000m³ per year is gravel. The rate of longshore transport of gravel declines markedly immediately eastwards of the eastern training wall, not only because of the presence of both structures impeding downdrift movement, but also as a result of a reduced angle of approach to the shoreline of shoaling and breaking waves (Chadwick, 1990). Gifford Associated Consultants (1997) calculated from LITPACK that the drift rate along East Beach to Portslade is 15,000 to 17,400m³ per year; Scott, Wilson, Kirkpatrick (1994) proposed the somewhat lower rate of approximately 14,000m³ per year. Almost all of this is gravel. Halcrow (1990) computed an actual eastwards transfer of 14,200m³ per year based on an adjustment factor taking into account the role of groyne storage. It is generally agreed that net transport rates are highest in winter, at an equivalent annual rate of a maximum of 50,000m³ (Halcrow, 2004), at least between East Beach and Hove. However, Morphett (1989) observed a period of westward longshore movement on West Hove beach during winter, when waves approaching from the east/south-east were prevalent. If this were experienced routinely, it would significantly reduce the net annual rate of eastward movement. There is uncertainty if higher winter season rates prevail along Brighton beach; this may again be due to drift reversals reducing the total amount of coarse material moved eastwards during this period of the year (principally November-February). Halcrow (2004) report on observations of “pockets” of net westward littoral transport between Shoreham East Beach and Brighton Marina, particularly during winter storms. These are described as “re-circulation eddies” that may have onshore and offshore directed components; thus, their net effect on overall longshore transport is uncertain.
Analysis of Coastal Monitoring Programme baseline topography 2007-12 and aerial photography data indicates that between Hove and Brighton Marina, the eastward drift rate is 3-10,000m³ per year. This is a reduction from the 2004 estimated rate of 10-20,000m³ per year. Sediment accumulates at Black Rock, west of the Marina, by approximately the same volume per year that the unit receives from by-passing at Shoreham. This accumulated material is periodically recycled between Hove and Brighton.
Halcrow Maritime (2001) estimated that the potential eastwards drift rate (for all grades of sediment) for this stretch was close to 50,000m³ per year. However, insufficient sediment feed, coupled with the storage effect imposed by an efficient but relatively widely spaced groyne system, reduced this to an actual rate of some 18,000m³ per year In the zone 400m updrift from the Brighton Marina breakwater (Photo 3) this declines to 4,000m³ per year, a feature of the longshore transport pathway that apparently pre-dates the building of the marina (Scott, Wilson, Kirkpatrick, 1994). Halcrow Maritime (2001) expressed the view that net longshore drift is lowest for this sector along central Brighton beach (Photo 11) as it has a better aligned swash orientation than adjacent beach lengths.
Analysis of Coastal Monitoring Programme baseline topography 2003-12 and aerial photography data indicates that between Brighton Marina and Newhaven, the drift rate increases slightly eastwards. Actual rates immediately downdrift from Brighton Marina are much reduced, resulting in beach starvation; the rate between the Marina and Rottingdean is less than 1,000m³ per year, which increases to 1-3,000m³ (closer to 1,000) per year towards Saltdean and Newhaven. These calculated rates are a reduction from the 2004 estimated rate of 10-20,000m³ per year. The reduced sediment transport rate is due to the lack of beach material.
Chadwick, Morfett and Rees (1986) derived a theoretical potential rate of longshore transport of gravel of 34,300m³ per year for the sector between the Marina and Saltdean. This was based on the application of a longshore power energy equation, and was acknowledged to be much higher than the actual rate because of the constraining role of groynes (on transport) and seawalls (on accretion). They proposed an adjusted figure of 28,530m³ per year. The rate computed by LITPACK (Gifford Associated Consultants, 1997) is 29,000m³ per year, which is also based on an allowance for groyne storage. The site of the Marina is Black Rock Ledge, a drift barrier and a site of near zero transport for several decades prior to marina creation, in 1973-5 (Posford Duvivier, 2001; Halcrow Maritime, 2001b; Halcrow, 2004). The presence of the Marina does now create a minor wave diffraction effect, but this is not thought to be sufficient to have any significant impact on downdrift transport except perhaps for very fine grained sediment. Ellis (1986) carried out a short experiment on a beach site 400m east of the Marina wall. He tracked the movement of several gravel sized clasts, and found that they moved a short distance westwards in January but remained static in July. This apparent evidence for counter drift was considered to be due to a "tidal vortex" set up by the Marina wall augmenting the wave diffraction effect. Posford Duvivier (1997; 2001) gave evidence of sustained drift reversal, during winter, over the Marina-Saltdean frontage, particularly in the vicinity of Ovingdean. Nonetheless, the overall, net component of movement is eastwards. Drift reversals are the most probable cause of fluctuations in beach volumes (Posford Duvivier, 2001). Ellis (1986) observed a site at Rottingdean that exemplifies this, and which also demonstrated more or less uninterrupted eastwards transport. Both here and at Bastion Steps, Peacehaven, Ellis (1986) reported almost no measurable movement of gravel clasts on the chalk shore platform surface. This is probably due to the effects of microrelief. It needs to be understood also that that beach volumes and profiles have changed radically in places over the last 20 to 40 years as a result of the local authority’s inability in the 1980s and 1990s to fund groyne maintenance in advance of the reconstruction schemes which in general have removed the old groyne field and left most of the coast temporarily open to the free movement of shingle. The present groyne field traps and stores most of the coarse sediment in transit.
Joliffe (1972) stated that 60,000m³ of shingle accumulated against the updrift side of the breakwater within three years of its reconstruction in 1890. This would suggest a drift rate of 20,000m³ per year if there was no bypassing, however this is not representative for such high accretion rates have not been sustained since. Joliffe (1964) considers coarse bedload transport around Newhaven breakwater to be probable, thus accounting for some supply of gravel deposits in outer Seaford Bay. There is no direct evidence for this, although Ridehalgh (1959) does mention that some "exotic pebbles" were placed on the beach between Friars Bay and the terminal stub of the breakwater. These were subsequently recovered on Seaford Beach. As there is no mention of quantities involved, length of experimental period and prevailing hydrodynamic conditions, this observation has low reliability and cannot be verified.
Unidirectional (eastwards) littoral transport has been assumed for the sector east of Peacehaven. Ellis (1986) undertook tracing of gravel-sized clasts at a site on Friars' Bay over a six month period. This revealed net movement westwards during the winter months, and the reverse during spring and summer. Overall, somewhat more sediment moved eastwards. As with other results reported by Ellis (1986), the representativeness of hydrodynamic conditions during the experimental period cannot be ascertained.
Between the East Pier and Tide Mills, the net direction of longshore transport is east to west; east of Tide Mills, it resumes the regional norm of west to east. A well-defined drift divergence within the highly managed Seaford beach therefore exists. Analysis of Coastal Monitoring Programme lidar 2007-12 data indicates that sediment transport rates are naturally low, in the order of 1-3,000m³ per year; however, if replenishment and extraction operations are taken into account, the sediment transport rates are increased substantially, in the order of 100,000m³ per year. The composite beach has an upper foreshore comprised of gravel or coarse-mixed sediment, the lower foreshore and nearshore sediments of sand. The drift rate has not been quantified and the 2004 estimated rates have not been revised. Material deposited within the central section of Seaford beach is rapidly transported to either end of the beach where it is recycled to the centre. These operations are frequent thereby maintaining an unnaturally enhanced sediment transport rate. Therefore, if recycling operations were not undertaken, the central section of Seaford beach would be depleted. For further details relating to the Seaford Beach Management Plan (2003-12) refer to Environment Agency (2012) and Worthing Borough Council (2012). A partial littoral drift boundary for transport of gravel is suggested at the eastern boundary of Seaford beach, where the rock platform the fringes Seaford Head dominates the foreshore and significantly reduces or inhibits alongshore drift rate.
Counter-drift west of Tide Mills apparently operated towards the end of the nineteenth century. Morris (1931b) located the divergence boundary at Bishopstone, and suggested that east to west movement of both sand and gravel towards Newhaven East Pier was due to the fact that the latter structure, and the western breakwater, attenuated waves approaching from the south west. Waves coming from the south-east were therefore dominant over this short sector. This has been partly confirmed by later studies (e.g. Hydraulics Research Ltd, 1985; 1986c), although the ineffectiveness of modified south-westerly waves is also related to wave refraction set up by the breakwaters, piers and training walls of Newhaven Harbour. East Pier may also create an anticlockwise tidal circulation, although this would only account for the transport of sand (Robinson and Williams, 1983). LITPACK (Gifford Associated Consultants, 1997) calculates a drift rate of 25,000m³ per year for this sector, a figure that is presumed to include both sand and gravel.
For Seaford Beach east of Tide Mills, Hydraulics Research (1986c) calculated the potential eastwards drift rate to be between 20-38,000m³ per year. For the period October 1988 to September 1989, Large (1989) computed a rate not in excess of 17,000m³ per year, following comprehensive renourishment and profile remodelling in 1987. LITPACK modelling (Gifford Associated Consultants, 1997) gave a rate of 21,500m³ per year, with a probable but unspecified acceleration downdrift. Joliffe (1964), reporting on the results of a fluorescent tracer study of gravel transport on the foreshore of Seaford beach, tentatively identified an impersistent transport divide near the Buckle Inn with net westwards movement of shingle towards the mouth of the Ouse when operative. Thirty years later, and following massive beach morphological modification in 1987, this was still apparent from detailed analysis of beach profiles recorded during intensive post-nourishment monitoring (HR Wallingford, 1996; Brampton and Millard, 1996). However, it was considered possible that the transport divergence at Tide Mills/Bishopstone might temporarily migrate eastwards under specific wave energy conditions.
Between Seaford Head and Cuckmere Haven, the potential longshore drift rate of gravel calculated using LITPACK is 25,000m³ per year, but shortage of input sources reduces this to a lower quantity. However, there may be some irregular input from wave driven offshore to onshore transport (HR Wallingford, 1996; Halcrow, 2004). Because of the change in coastline orientation, rates are probably lower east of Seaford Head.
Experimental work on onshore to offshore, and reverse, movement of shingle was undertaken by Hydraulics Research Station (Joliffe, 1964; 1972) for Seaford Beach. Results were derived from recovery of fluorescent tracers. The specific objective of this work was to assess the relative effectiveness of extant long and short groynes in reducing prevailing littoral drift rates. It was reported that gravel mobility in the surf zone was inhibited where the seabed substrate was overlain by silt, mud or muddy/silty sand. Where gravel sized particles by-passed groynes, there was a tendency for a proportion of clasts to move offshore and thereafter return to inter-groyne bays where the nearshore beach gradient was relatively shallow. Where the substrate was muddy or silty, there was no return onshore movement. This imbalance between net off and onshore cross-transport, in favour of offshore loss, was especially evident in winter. This was considered to be due to higher energy breaking waves moving some gravel sized particles into deeper water (>10m), in Seaford Bay. Although the beach envelope at Seaford Beach has since been significantly modified by renourishment/reprofiling, this tendency towards removal of coarse foreshore sediment offshore may still prevail, and may account for a proportion of the temporary loss of beach volume during certain years between 1987 and 2005 (see Section 5 for further discussion, including an alternative opinion).
Analysis of Coastal Monitoring Programme lidar 2007-12 data indicates that the inter-tidal and sub-tidal foreshore for the majority of this section is dominated by an extensive, almost continuous rock platform, with largely sediment starved fringing beaches. This exposed platform extends seawards from the inter-tidal zone some 600m, where it terminates. Seawards of this abrupt junction the surficial sediments are comprised of mixed and coarse-grained sediments. The fringing beaches comprising coarse-grained gravel and cobbles may be connected and have an eastward drift of approximately 1,000m³ per year. To the east of Beachy Head bedforms are discernible at the seaward and steep boundary of the rock platform, indicating strong seabed currents past the headland. The drainage of the River Cuckmere has eroded the rock platform, and Cuckmere Haven beach behaves as independent pocket beach and does not receive sufficient alongshore sediment supply to maintain beach levels, suggesting an onshore feed of material from the sub-tidal deltaic deposits of sand and finer silty sediments, which are a relatively constrained pocket of sediment and of sufficient thickness to mask the underlying platform. Due to the net eastward sediment transport the Environment Agency undertake recycling operations from the mouth of the River Cuckmere placing approximately 5-7,000m³ per year onto the sediment starved west beach of Cuckmere Haven. The beaches of this sector are well adjusted to the dominant wave approach from the south-west.
LITPACK (Gifford Associated Consultants, 1997) determined a potential net eastwards drift rate of 25,000m³ per year, but actual rates are less than one half of this as input of sediment is relatively small. Some, perhaps most, gravel in transit is subject to temporary storage in the embayment that defines the mouth of the Haven and in 'pocket' beaches between minor Chalk and landslide debris salients, especially those composed of residual boulders from earlier cliff falls. A majority of Chalk and flint clasts are sub-rounded or well rounded (May, 2003), suggesting their retention in storage for longer than is characteristic for much of this and the downdrift coastline. The latter breakdown, providing occasional "pulses" of longshore moving gravel. Both training walls at the mouth of the Cuckmere River act as groynes, trapping filets of shingle; this indicates periodic westwards longshore transport. The latter is acknowledged by Halcrow (2002) but considered insignificant in the local sediment budget. Posford Duvivier (1999) suggest that 5,400m³ per year arrives on Eastbourne beach from Beachy Head, but it is not clear if this figure is based on long-term averaging of episodic throughput. If so, it might equate with the prevailing drift rate, though it is a probable over-estimate. Offshore to onshore gravel input is also thought to be subject to "pulses" of supply, though Nicholls (1991) considers this component to have been inactive since at least the eighteenth century. The place name of Beachy Head is unrelated to any former presence of well-developed beaches. Most authorities accept that there is some eastwards moving by-passing of this headland by gravel, but that the quantity is difficult to estimate. Sand is transported more freely and in greater quantity in the nearshore zone, with a probable component of non-periodic reversing on to offshore movement. The strong tidal eddy (vortex) current system augments the role of wave-induced sand movement (Halcrow Maritime, 2001a), both on and offshore, but with a probable superimposed net eastwards movement. Halcrow Maritime (2001a) calculated that sub-tidal currents might potentially transport as much as 16,000m³ of both sand and gravel eastwards towards Eastbourne, but the actual quantity is more probably close to 6,000m³ when the complex components of the tidal gyre of Beachy Head is factored in.
4. Sediment Outputs
4.1 Nearshore and Offshore Transport
A high resolution, 100% coverage swath bathymetry survey between Dungeness and Newhaven, extending 1km offshore of MLW was commissioned by the Southeast Regional Coastal Monitoring Programme, and completed in August 2013. Analysis of this bathymetry data indicates that the seabed offshore of Shoreham-by-Sea to Brighton is shallow, broad and gently sloping and consists mostly of sand and a small area of rock and coarse sediment found to the east of the harbour entrance. Much of the seabed is morphologically nearly featureless, except for occasional breaks of slope up to 4m in height, with Quaternary palaeovalleys now infilled by subsequent sedimentation. No bedforms are discernible to confirm sediment mobility or indicate direction of offshore or onshore sediment transport. To the east of Beachy Head, bedforms are discernible at the seaward and steep boundary of the rock platform, indicating strong seabed currents past the headland. The seabed offshore of Beachy Head is more complex with significant rock outcrop and shore parallel northeast-southwest oriented ridges extending offshore, indicating former but now degraded Quaternary low sea-level cliffline positions. The 20mCD isobath is some 3,000m offshore of Shoreham. This reduces to 1,500m between Newhaven and Birling Gap, and less than 800m seawards of Beachy Head.
Based on the recent bathymetric survey and evidence given in section 2.1, the F1 and F2 arrows were removed for the 2012 update, as the information indicated sediment transport occurring offshore, they have been replaced by O1 and O2 arrows.
Experiments conducted between 4km and 10km offshore of Shoreham-by-Sea by Crickmore et al., (1972) indicated a minimal landward drift of gravel. At the 9m water depth, shoreward transport was measured at 1,000-1,500m³ per year per km whilst at 12m depth it reduced to less than 500m³ per year per km. No transport was recorded in excess of 18m depth. This rate of transport will only occur where potentially mobile shingle exists on the seabed. The knowledge of these processes was derived from experiments with radioactive tracers carried out over 2 years and correlated with wave data. HR Wallingford (1993) subsequently confirmed patchy distribution of inshore gravels together with the lack of shingle mobility beneath water depths greater than 15m to 18m.
O2 Sediment Transport occurring offshore Shoreham-on-Sea to Beachy Head (see introduction to sediment outputs)
Out to about 1,000m seawards of MLW, sediments are predominantly fine to medium sands; sandy silts and silty clays. There is also shell debris, occurring in discrete patches and mixed with minerogenic sediment. However, local diversity is provided by some reef-like outcrops of exposed bedrock (usually Chalk) and areas of well-packed rounded flint cobbles (Morgan et al., 2008). These have been described in some detail by Joliffe (1972), at a site adjacent to the seaward terminus of the western breakwater protecting Newhaven Harbour, and are interpreted as the result of exposure by combined wave and tidal current scour. It is probable that in several areas of sand waves and 'ribbons', the thin sand layer (less than 0.5m) overlies a veneer of consolidated gravels from which fine sediment has been largely removed by winnowing. Much of this gravel deposit is composed of flint and is probably relict (White, 1924; Bellamy, 1995; Posford Duvivier and British Geological Survey, 1999). Its ultimate provenance is the Chalk, but a significant proportion has been transported from the catchments of the Adur, Ouse and Cuckmere rivers under earlier morphogenetic conditions (refer to Section 1). Some flints may derive from Palaeogene outcrops further west. Most flints with a polygenetic history are rounded or sub-rounded; those more recently removed from submarine Chalk outcrops, or from coastal sites, are angular or sub-angular in shape (Joliffe, 1972; Posford Duvivier and British Geological Survey, 1999; May, 2000).
Beyond approximately 1,000m from the coastline, sediments become coarser, with sandy gravels dominating. This pattern is particularly evident in the extreme west of the region, and also in outer Seaford Bay. Most of inner Seaford Bay is occupied by a relatively thick layer of muddy and silty sand.
Offshore sediment mobility is indicated by ripples and megaripples, patches, waves and ribbons of sand and possibly thin sheets of fine gravel moving across apparently immobile, weed-covered cobbles (Joliffe, 1972; Hamblin and Harrison, 1989). At Jenny Ground, 700m seaward of Portslade, sand patches give some indication of net eastwards movement (Halcrow Maritime, 2001b). Indeed, eastward movement is suggested by the majority of the features observed, but there has been no monitoring of changes in bedform morphology and position, and some may be fossil features (Gifford Associated Consultants, 1997). An anti-clockwise circuit of sand movement may be a semi-continuous feature offshore and in the lee of Beachy Head, set up by the headland-induced tidal vortex there. Halcrow Maritime (2000b) inferred from beach foreshore change that both on and offshore pathways of sand transport operate. However, modelling using offshore wave climate statistics; wave refraction co-efficients and bathymetry did not provide a definitive picture of how and when sand is mobilised. It is probable that coarse sand moves around Beachy Head as bedload, as well as in suspension.
Although the mineralogy of the sediments of Seaford Bay is not well known, the predominant silty and muddy sands do not appear to be well sorted, thus implying only limited mobility. Joliffe (1964) described how gravel sized clasts, moving offshore from Seaford Beach, became trapped in a silt-dominated surficial layer. Between Cuckmere Haven and Beachy Head, sandy gravels in the near-shore zone are apparently immobile. They grade into well-sorted sands at a water depth of 30m, which possibly form a source for periodic onshore sand transport.
4.2 Estuarine Sediment Exchange
EO1 River Adur Inlet (Shoreham Harbour)
Following the stabilisation of the entrance channel to the port of Shoreham in 1820, dredging has been a continuous activity (see also section 4.3). However, there are no reliable records of quantities removed prior to 1960. Between 1954 and 1957, over 1 million m³ of material was removed as part of a programme of channel and berth widening and deepening to allow collier access to the newly constructed power station on East Beach. It is reported (Ridehalgh, 1958; 1959) that a "substantial" quantity of dredged spoil had been dumped offshore, close to maximum low water, in previous decades. Loss of detailed dredging records for the immediately following years makes it impossible to assess whether this material had an impact on the local sediment budget. Ridehalgh (1958) states that the tendency for a bar to form near the end of the eastern breakwater, prior to 1955, was reduced by breakwater extension and reconfiguration; this created a scouring effect that was reinforced by a local "breakaway" tidal eddy during the ebb. The entrance channel has been stable in position for many decades, as it has been cut into substrate iron-cemented Eocene sandstones and lignites (Mortimer, 1997).
Posford Pavry (1984) used a wide variety of secondary data sources, including daily wind speed and direction for 1980-1984; Admiralty and Port of Shoreham routine quarterly hydrographic surveys between 300m offshore and 250m inshore of the harbour mouth, December 1975-1984; dredging records, and seabed sediment samples in a study reporting on the feasibility of increasing the depth and extending the length of the entrance channel. To this was added primary data on tidal current velocities and monthly hydrographic surveys between October 1983 and October 1984. Analysis revealed a persistent tendency for two bars to form, one just inside the head of the western breakwater and another between 120 and 170m to the south. The inner bar, which was shallower, had a quasi-cyclic behaviour, accreting each winter and eroding during the summer. There was no convincing evidence of net loss or gain of sediment on either bar. The rate of accretion during the winter of 1983/84 proved to be exceptional, and was the product of a higher than average frequency of high wind speeds and storm waves from the south-east. During the period of close monitoring of hydrographic change, sediment was lost from the outer, and appeared to be gained by the inner, bar. Dredging records for 1960-1973 indicated annual removal of between 12 to 23,000m³ per year, but only 8-13,000m³ per year, 1974 to 1984, from the outer approach channel. Over the period 1974 to 1983, much larger quantities were dredged from the eastern and western arms of the entrance channel (means of 55,600m³ per year and 18,000m³ per year, respectively). This suggested that for silt, silty sand and fine sand at least, there was a positive sediment budget. Maximum ebb tide velocities of 0.8-1.0m per second in the channel between the breakwaters ensured net seaward transport of medium and coarse sand, which was then deposited over the seabed beyond the terminal points of both breakwaters in response to reduced bedload transport threshold velocities. Seabed samples indicated that fine gravel was present in this area, and occasionally also further up the entrance channel. The latter could not be transported by tidal currents, and must therefore be introduced by wave-induced currents, estimated at a mean of 1.0m per second (compare mean tidal velocities of 0.3-0.4m per second).
Port of Shoreham dredging volume data for spring 1995 to autumn 2001 indicates removal of an average of 19,670m³ per year from the entrance channel between the harbour breakwaters; and 13,100m³ per year from the outer approach channel. For the period 2005 to 2007 the combined total was 45,000m³ per year. These figures conceal considerable inter-annual variation. Without detailed analysis of hydrographic charts providing data on the volumetric equivalent of seabed topographic change, it is difficult to know if dredging removes much of the inferred net input of sediment from nearshore and offshore sources. Hydraulics Research (1984), working with dredging data for the period 1975-1983, and using theoretically calculated values for inshore wave climate and tidal current velocities, concluded that net accretion for the entrance channel as a whole was 1,630m³ per year. Gifford and Partners (1997), however, estimate a net input from marine sources of 8,000m³ per year, but it is uncertain if this figure is adjusted for long-term average losses induced by dredging.
Hydraulics Research (1984) also used data on (i) inshore wave heights (derived from wave refraction calculations); (ii) local bathymetry and current metering; (iii) seabed samples, and (iv) formulae relating to nearshore sediment transport to estimate that approximately 200,000m³ per year of suspended sand moves across the outer entrance channel. (This figure apparently refers to transport in both west to east and east to west directions, thus does not identify a net transport vector). There is uncertainty as to how much of this sediment may be moved up-channel by tidal currents, and/or might be diverted into short-term storage within the inner and outer bars, but it is within a range between 30 and 100,000m³ per year. It was suggested that significant quantities might be temporarily driven into the outer entrance channel under high energy wave conditions, occurring approximately once every 10 years. As much as 24-25,000m³ in a week of sustained south-easterly approaching waves might be a reasonable estimate; much of this would be dispersed offshore and longshore shortly thereafter. Sea bed sampling indicates some cross-entrance bedload transport of gravel, estimated to be a net volume of approximately 14-20,000m³ per year (Hydraulics Research, 1984; Sir William Halcrow, 1990), with an apparent net eastwards component. Most of this transfer occurs at or around high water irrespective of the type of incident waves. The presence of a stable, armoured sea bed surface seawards of the outer channel approach has been taken as evidence of the absence of gravel transport in the offshore zone (Joliffe, 1972; 1978; Posford Pavry, 1984). This, however, cannot be regarded as a reliable conclusion.
Brighton Marina
It is known that siltation occurs within the Marina basin and that periodic dredging operations periodically remove approximately 50,000m³. There is currently no publicly available data that allows an average annual rate of output to be calculated. All of this sediment must be derived from marine sources. Dredged spoil, which is predominantly silt, is deposited in the offshore zone well to the south-east of the main breakwater and is presumed not to re-enter the littoral sediment budget in any significant quantity.
EO2 R. Ouse (Newhaven Harbour)
The present configuration of the harbour dates from the completion of the Western Breakwater in 1890 and the extension of the East Pier in the early twentieth century. The approach channel is approximately 125m wide and 500m in length with a nominal depth of -6.0m LAT maintained by routine dredging. Between 1878 and 1974 depth above the outer bar was -3.7m LAT, but deepening was undertaken in the late 1970s and mid-1980s to provide access for larger draught vessels. Mean spring tidal range is 6.1m, and 3.0m during neaps, with peak (ebb) current velocities of 0.8 m per second occurring some 200m offshore the western breakwater. On the flood tide, water enters the main channel from the west and creates a tidal eddy in the lee of the Western Breakwater. This stream then takes a nearly circular route across the shallow water area to the east of the dredged channel, which it re-enters nearer to the channel head. The ebb tide current is confined to the dredged channel seawards of the breakwater, but then flows westwards in a more diffuse current (Hydraulics Research, 1988a). Its flushing effect has diminished in stages over the past four centuries as a result of estuary reclamation (Brandon, 1971).
The local wave climate is complex, and has been modelled (Hydraulics Research, 1988a) using data from the Seaford wave study (Hydraulics Research, 1985). The latter employed the OUTRAY wave refraction model, calibrated using hydrographic knowledge of Seaford Bay. The area to the immediate east of the dredged channel is exposed to wave approach from the south-east, but is protected from south-westerly waves by the 900m length of the Western Breakwater. Complex local refraction and reflection causes an increase in wave height to the east of the channel. Waves moving directly into the channel, from the south or south-south-east are also refracted by its well defined (dredged) boundaries. As both Beachy Head and Selsey Bill provide a protective influence, Newhaven Harbour is principally exposed to waves approaching between 90° and 270° (Posford Duvivier Environment, 1997).
Hydraulics Research (1988a) observe that south-westerly waves have the capability of entraining and suspending sand in Seaford Bay, which is then carried into the dredged channel on the flood tide. In an earlier study on Seaford beach, Hydraulics Research (1985) concluded, from wave energy and refraction modelling, that south-westerly approaching waves moving at peak velocities were able to disturb sand in water depths of up to at least 6.0m. Deposition of suspended sediment occurs due to the rapid reduction of wave energy in the landward sector of the approach channel, induced by depth shallowing and refraction. This effect is greatest when peak tidal current velocities operate in the area to the immediate east of the dredged channel, apparent from careful and detailed analysis of hydrographic chart data. Net westwards littoral drift between Tide Mills (Seaford) and the East Pier may also contribute a net transfer of fine sand into the inner approach channel. Hydraulics Research (1988a) quote an example of approximately 14,000m³ of sand and silt being deposited over a one month period in early 1988 during which high energy south-westerly waves were dominant. This accretion took the form of a shoal spreading across the channel, in a north-westerly direction from the end of the East Pier. Simultaneously, a presumed scour hole developed seawards of the eastern end of the dredged channel, possibly caused by a strong tidal eddy set up by the Western Breakwater. This event was probably characteristic of a long-sustained pattern of net winter siltation (Posford Duvivier Environment, 1997). Optimum conditions are the coincidence of maximum ebb tidal current velocities and large waves immediately following low water. Accretion rates will have progressively increased over the past 150 to 200 years as a direct consequence of reclamation reducing the tidal prism, and therefore the flushing effect of the ebb current.
It is thus apparent that the dredged channel is effective in trapping fine sediment, only a relatively small proportion of which can be re-distributed or removed seawards by tidal currents. Once deposited, sediment is less likely to be remobilised by diminished wave power, except under specific conditions involving incident waves from the south-east. However, some re-mobilisation may be induced by the deepest draught vessels that can use the navigation channel.
The mouth of the Ouse, in particular Newhaven Harbour, is therefore a natural sediment sink, though specific input does depend on how much material is moved out by the ebb current, thereby feeding shoreward moving waves and the incoming flood stream. Sediment sampling from the inner entrance (Hydraulics Research, 1988a) indicated a well sorted sand with a median diameter of 0.12mm, increasing to a coarser deposit further offshore. It would appear from this evidence that very little, if any, gravel-sized material is retained in the approach channel, although it is probable that material of this grade moves eastwards from the end of the Western Breakwater. Presumably, it is transported seawards into and across Seaford Bay (see sections 3 and 4.1 for further discussion). Longshore transport is therefore subordinate to onshore movement in determining the local sediment budget.
Gifford Associated Consultants (1997) and Halcrow (2004) suggest that Newhaven Harbour experiences a net annual input of about 7,000m³. Data provided on dredged quantities for both the inner and outer harbours, 1982-2002 (Newhaven Port and Properties Ltd, 2002) reveal an annual average of 55,400m³ and 137,100m³ respectively, and approximately 150,000m³ for both areas of the harbour for each year between 2002 and 2008. There have been substantial inter-annual variations, reflecting changes in vessel sizes, port operating conditions and specific infrastructure projects.
EO3 River Cuckmere (Cuckmere Haven) inlet
Reclamation of the lower flood plain, and channel straightening, has not only lessened the scouring effect of river discharge at the mouth of the Cuckmere, but the tidal prism has also been drastically reduced. These are the principal factors that have accelerated the growth of the tidal delta over recent centuries, which functions as both a store and sink for sediment. An unquantified proportion of the coarse sediment held in the delta is considered to be supplied by offshore to onshore transport (Halcrow, 2004). The contemporary input of fluvial sediment is probably insignificant. The position of the mouth of the river has migrated eastwards as the spit beach to the south has extended in the same direction (documentary evidence is only available from the late nineteenth century). Several storm induced breaches dating from 1910 were responsible for the realignment of this spit until the entrance channel was stabilised by training banks.
4.3 Dredging
Dredging for navigational access occurs at Shoreham and Newhaven Harbours, and involves removal of mostly fine sediment from both the approach channels and inner harbours. Excavation for larger berths has been periodically undertaken in recent years in response to increasing vessel sizes and traffic expansion. Details of quantities removed over the past 30 years are given in section 4.2.
Sediment is also dredged from the outer basin of Brighton Marina, crudely estimated at 25 to 35,000m³ per year (detailed figures have not been made available). As there is no longshore supply, this quantity must represent input from marine (nearshore and offshore) sources. Most is silt, and is therefore introduced as suspended load by both waves and tidal currents. Dredged sediment is dumped offshore to the southeast of the Marina site; it is not known if it makes a contribution to the downdrift sediment budget, but if so it is likely to be very small.
4.4 Beach Mining
Past sites include:
- Shoreham West Beach up to 90,000m³ per year
- Portobello Beach, Saltdean in "1920s" (Joliffe, 1972).
- Cuckmere Haven beach: there was periodic removal of gravel "up to 1952" (Joliffe, 1972), apparently under licence from the local authority. The main source was the foreshore but also some material was dredged from close to the river mouth. There are no detailed records available.
- Inside Newhaven Harbour - removal of residual beach deposit in the 1950s and early 1960s (Joliffe, 1972), used locally for aggregate.
The practice of beach mining has been discontinued at all these sites, and does not occur elsewhere.
4.5 Beach Shingle Abrasion Loss
A component of the University of Sussex-led BERM project (Dornbusch, 2002; Dornbusch et al., 2003; 2006b; Moses, et al., 2001) has been experimental research on the durability of flint gravel sized clasts on East Sussex beaches. Data has been obtained from (i) laboratory mill tumbling of flint clasts taken from sites between Telscombe and Friars' Bay; and (ii) a two year programme of field monitoring of abrasion loss of 'exotic' quartzite and limestone clasts introduced onto two East Sussex beaches composed of contrasting types of flint particles. The exotics had size, shape and other characteristics that were closely comparable to indigenous flints and were used to facilitate ease of recovery and re-measurement (Dornbusch, et al., 2002 a and b; 2003). Offshore wave climate and beach profiles were systematically measured during this experiment.
Laboratory experiments used both well rounded and freshly-broken flints, the latter derived from recent cliff falls. Abrasion rates were observed to increase in a linear form with increasing clast size and weight, at approximately 0.06% of weight per hour. However, precise abrasion rates were a partial function of water: mass particle ratios in tumbling mills, and it was noted that under these controlled conditions chatter marks, characteristic of abrasion, were not created. Abrasion rates declined over time, but losses were consistently greater for newly-fractured, broken, angular flints than for those which had a previous history of natural wear. Abrasion debris was predominantly silt sized, but some sand was also produced. It is the provisional conclusion that experimental tumbling cannot achieve the rates that prevail under natural hydrodynamic conditions.
Field monitoring of marked exotic tracers was carried out over a one year period (approximately 700 tidal cycles). Liassic limestone clasts suffered an 18.6% weight loss whilst for Budleigh Salterton metaquartzite the weight reduction was lower, at 5.7%. These figures indicate the important control of lithology, but mean wave height also provided some correlation with spatial variations of abrasion loss. Losses were, not surprisingly, greatest after storm events, but less expected was the observation that well rounded clasts continued to experience abrasion at a "significant" rate under normal conditions following storms. Initially angular particles became sub rounded or rounded within a few months. No direct correlation between weight loss and shape change with distance travelled was determined.
Tracer re-weighing suggests an average weight loss of 0.0078% of initial volume per tidal cycle (maximum of 1.23%). At this rate, destruction (to sand) would occur within about 20 years. Relating these results to flint, via tumbling mill experiments, suggests that it is some ten times more durable than quartzite, giving an estimated weight loss of some 2% per annum. Thus, existing beach clasts might be removed by abrasion over the next 200 years (a 25% loss every 50 years). Moses, et al. (2001) and Dornbusch et al. (2003) apply this conclusion to Telscombe Beach, which has a surface area of 1500m² and an estimated volume of 31,000m³. On the assumption that 50% of the beach area is subject to wave action during an average tidal cycle, and that the mean depth of disturbance is 0.5m, some 12% of the total clast population is exposed to active abrasion. If the mean annual weight loss is 2%, then 75 to 90m³ of volume is lost annually as a result of this process (0.3% of volume.) This needs to be compensated by fresh input, principally from eroding cliffs. In this case, this is achieved, as annual natural erosion yield is calculated to be 96m³. For data from Saltdean Beach, Dornbusch et al. (2003) calculate a loss of 15m³ (0.05% of original volume). Whereas Telscombe Beach is naturally supplied with flint clasts from adjacent cliff and platform erosion, Saltdean is a mixed gravel and sand beach that was recharged three years prior to this experiment. The lower rate of loss at the latter site may also be due its greater compartmentalisation (i.e. it is a less “open” beach) with more restricted longshore displacement of material. There are also contrasts in the characteristic shapes of clasts on the two beaches (more rounded at Telscombe), suggesting that the active layer reworked by wave action is shallower at Telscombe; thus surface clasts are more frequently exposed to abrasion.
Notwithstanding that some of these figures are rather crude- or at least qualified- estimates, the wider implications for beach management are clear. Although a positive budget can be established for Telscombe Beach, it is one of only a few localities between Brighton and Seaford Head which are unprotected. At all other sites on this sector, abrasion loss cannot be naturally counterbalanced by sediment yield from cliff erosion. Recharge is therefore a longer term inevitability, but replenishment supplies will also suffer in situ reduction due to abrasional wear.
5. Beach Morphodynamics
Gravel beaches are the dominant form along this shoreline, although many are built on coarse to fine sand foundations. Sandy foreshores are a feature of some sectors, but tend to be absent or of limited development where shoreline platforms front a narrow coarse clastic backshore beach. Most gravel beaches are composed of sub angular to sub rounded flint clasts, and characterised by a narrow, relatively steep, cross-sectional profile often with one or more berm ridges. An exception is the wide backshore of the beach at Tide Mills, Bishopstoke west of Seaford. Seasonal changes in elevation, width and profile morphology are in response to variations in incident wave energy and approach.
There are no systematic and comprehensive reviews of gravel beach sedimentology. Halcrow Maritime (2001b), describing beach form along the Southwick to Kemp Town (Brighton) frontage, state that medium to coarse gravel dominates the backshore, but is replaced by fine gravel where its toe is adjacent to the sandy foreshore. A good example of this also occurs between Saltdean and Telscombe Cliffs (Mott McDonald, 1999). In places, for example between Southwick and Hove, well-sorted medium to fine gravel extends below mean low water (Young and Lake, 1988). Here, and at beach replenishment locations such as Rottingdean and Seaford, the introduction of large quantities of gravel taken from offshore sources has modified indigenous beach sedimentology. Following a few weeks of varied wave conditions, the mid-profiles of wider beaches, e.g. Brighton and Seaford, often exhibit a mix of gravel and sand. Thereafter, sand particles move down through the voids between larger particles and add to a well compacted sandy substratum. (Compaction can also be the product of vehicle movements at recharge, recycling and reprofiling sites). This process is characteristic of the winter period, when storms may temporarily remove some of the gravel fraction into the nearshore zone. Under these conditions, when more of the sand basement is exposed it proves relatively resistant to wave-induced scour due to its low porosity. However, under extreme high energy conditions most sand is removed, revealing the underlying Chalk substrate.
Gifford Associated Consultants (1997) state that the mean grain size of gravel on regional beaches increases progressively eastwards. This is particularly evident on the lower part of the backshore, where the mean for the principal particle axis increases from 12 to 15mm between Hove and Seaford. This is taken to be in response to increasing wave energy in the same direction.
The majority of gravel clasts experience progressive "smoothing" and rounding of corners as a function of abrasional wear. This increases with residence time (Moses et al., 2001), but it is also related to provenance (Sections 1, 2 and 4 discuss the origins of beach gravel in more detail). Sand is introduced from nearshore and offshore sources, examined further in section 4.1. At a few locations where the longshore transport of gravel is impeded, clasts exhibit a higher than average degree of rounding. May (2003) notes that there is a larger proportion of "rolled" flint particles on the beach at Cuckmere Haven, where past fluvial input routed via the contemporary ebb delta is a possibility.
The presence of control structures, mostly groynes, introduces local variation to beach morphology by promoting updrift accretion and inducing downdrift sediment depletion. Beach profiles are sometimes exceptionally steep where long or widely-spaced groynes are effective in trapping a high proportion of gravel moving within groyne compartments, especially where drift rates are low (Cleeve and Williams, 1988). In some examples of recent groyne re-construction, beaches have been designed to re-orientate or "rotate" in response to changes in wave approach, height and period. An example occurs between Brighton Marina and Saltdean (Posford Duvivier, 1994b; 1995 and 1997). Where beaches are backed by seawalls, reflective scour is also a widely acknowledged factor influencing beach morphodynamic behaviour, especially in the immediate vicinity of breakwaters and long groynes. Several examples, closely tied to the history of sea defence construction, are documented by Joliffe (1972).
The first edition regional Shoreline Management Plan (Gifford Associated Consultants, 1997) undertook detailed analysis of historical change, between 1870-1992, of beach profile form and inter-tidal width. Successive Ordnance Survey map editions, supplemented by post-1973 ABMS and other aerial photographic cover, were used in the LITPACK numerical model to quantify net movements of the relative positions of mean High and Low Water for eleven discrete sections of shoreline. This was then converted to a Retreat/Advance classification. The LITPACK model was also able to calculate net volume changes 1972-1992 between adjacent profiles to a depth of 10 to 12m below ODN. As ABMS survey data only extends to mean low water, extrapolation seawards was based on bathymetric information. The salient points of this, and independent calculation of beach volume change, 1973-1999 (Halcrow, 2001b) and subsequent Regional Shoreline Monitoring (RSM) combining ground survey, photogrammetric/lidar imagery analysis and digital terrain modelling for 2003 to 2008 (Worthing Borough Council, 2009) are summarised below. Reservations about the accuracy of both ABMS (see above) and RMS measurements (see Worthing Borough Council, 2009) should be noted with respect to the precision of figures of volume losses and gains.
1. Shoreham Harbour entrance to West Hove
Historical long-term retreat of Mean Low Water (0 to 1.5m per year) and steady volume reduction at least since the early 1960s-estimated at 150,000m³ (Halcrow, 2004) is evident at East or Southwick Beach but with recent (2003 to 2011) modest net accretion due principally to beach recharge. (A loss of 37,000m³ occurred between 2003 and 2006) Worthing Borough Council (2012) reported that the position of Mean High Water migrated approximately 12m Seawards between 2003 and 2011. Overall long-term trend is for profile steepening with morphodynamic behaviour linked to recharge periodicity and wave reflection/refraction caused by the training wall. Rapid drawdown occurred following western breakwater extension in 1874, leading to "emergency" recharge at Hove in 1886 (Ward, 1922). The sector between Southwick and Aldrington has been maintained subsequently through a combination of groyning (dating from c.1870), deposition of dredged spoil, rubble dumping and recharge using excess gravel accretion updrift of the western breakwater.
2. West Hove to Brighton Marina
The long-term trend between Portslade and West Hove has been Low Water retreat (up to 1.25m per year), but seawalls have fixed the position of mean High Water since the mid -1850s. However, since at least the late 1960s, net accretion and mean low water advance has prevailed east of Hove at a mean rate of 18,200m³ per year, 1973-99 (Halcrow, 2001b). This beach has exhibited a progressive increase in width between the Palace Pier and east Kemp Town over several decades (Photo 11), and is efficiently managed by widely spaced groynes. The volume of shingle contained in this beach was calculated by Halcrow (2004) to be 1,820,000m³. The role of the west breakwater of the Marina in inhibiting longshore drift (Photo 3), (i.e. acting as a terminal groyne) and thus being a cause of steady beach volume increase, since its construction in 1973, is strongly implicated (Jezard, 2004; Halcrow, 2001b). Worthing Borough Council (2009) report accretion of approximately 25,000m³, 2003 to 2007. However, this has not been proven definitively. Thornburn (1977) implied that equilibrium has characterised the eastern sector of this beach since approximately 1930. The first generation of groynes were built in 1790, and have been maintained/upgraded continuously since. Halcrow Maritime (2001b) regard Kemp Town beach as a self-regulating, equilibrium dissipative system. This conclusion is based on detailed analysis of volume changes over six groyne-limited sectors for a 12 year period, using ABMS data. Jezard (2004) has confirmed net volume gain and advance of the toe of the upper gravel beach using ABMS profiles for 1990-1995. However, there was no discernible increase in beach volume in the beach sector immediately updrift of the Marina wall. No significant changes in profile gradient and morphology are apparent, with only modest losses - rapidly recovered - during storms. Halcrow Maritime (2001b) present photographic and anecdotal evidence of inter-annual changes in beach volumes, but state that there were no discernible effects on longer-term condition. A tentative conclusion is that the Kemp Town beach was accreting for many years prior to the construction of Brighton Marina, possibly due to the changing bathymetry in the Black Rock area and the local trapping effect of the Black Rock ledges. Crest flattening and some artificial reprofiling undertaken during the spring of each year modify the beach between the piers along the central Brighton frontage. This is undertaken for tourist amenity purposes.
3. Brighton Marina (Black Rock)
The site of the Marina previous to its construction between 1973 and 1975 was characterised by sediment dispersal and no net beach accumulation. It was ascribed by Joliffe (1972) to the effects of outflanking induced by the large terminal ("Banjo") groyne at Kemp Town. He was able to correlate loss of beach volume and width with the history of seawall construction, commencing in 1928. Other authorities have ascribed the absence of beach accumulation to a zero littoral transport rate (Posford Duvivier, 1997; 2001). Mean Low water retreated at a rate between 0.5 and 1.8m per year between 1960 and 1998 (Halcrow, 2004).
4. Brighton Marina to Saltdean
Both the long and short-term (post 1980) trends are those of profile steepening and Low Water Mark retreat at variable rates of 0.6 to 3.0m per year. This is confirmed by Dornbusch (2002), who identifies both strong control structures and apparent loss of offshore to inshore input as probable causes. Groyning commenced in 1887, but the seawall was completed in stages between the early 1930s and the mid-1980s. Chadwick, Moffett and Rees (1986) used ABMS data for 1972-1979 to demonstrate that this beach system, as a whole, was slowly reducing in volume. Ellis (1986), however, regarded the upper gravel beach at Rottingdean as stable. Posford Duvivier (1997; 2001) state that the beaches of this sector have a history of depletion, although there is comparative stability at Ovingdean. This was ascribed to a local, subtle change in shoreline orientation. There is uncertainty over the morphodynamic behaviour of the beach immediately downdrift of the Marina. Posford Duvivier (2001) state that wave diffraction set up by the breakwater creates reduced wave energy and provides conditions for modest sediment accretion. Jezard (2004), concludes that ABMS data for 1990-1995 indicates loss of volume, landward migration of the shingle toe of the beach and Mean Low Water retreat but beach stability apparently prevailed between 1974 and 1985. It is uncertain if the Marina has had any specific impact on beach drawdown, though Jezard (2004) considers trends for 1990-1995 demonstrate its effect on interrupting updrift supply, though this may not exceed 2,200m³ per year
Two substantial replenishment schemes in the 1990s at Rottingdean and Saltdean (section 2.3) and subsequent recycling have been carried out in response to the history of depletion of this gravel beach. Wide platform development has inhibited the extension of the narrow sandy foreshore. The long-term history of beach management by intensive groyne control has been relaxed since the mid-1990s to reduce the effects of differential accretion/depletion. This has facilitated more flexible response to seasonal profile changes, especially where enhanced by reflective scour from massive backing seawalls (Posford Duvivier, 1997; 2001). Short-term littoral drift reversals are frequent, but beach recovery has been checked, in the past, by closely-spaced groynes. Beach management since 1995 has involved the substitution of rock for timber groynes, the design of optimum crest heights and profile gradients to minimise volume fluctuations (Posford Duvivier, 1997), thus creating relatively high foreshore levels. Nonetheless, ongoing erosion is revealed by a net loss of 42,000m³ between 2003 and 2007 (Worthing Borough Council, 2009).
5. Telscombe Cliffs to Peacehaven Steps
The historical trend is for Low Water Mark retreat (0.6 - 1.4m per year, accelerating to 3.0m per year in recent decades), and profile steepening. The western section of this length of shoreline remains unprotected except around the headworks of the Portobello outfall. At Peacehaven, the seawall dates from the 1930s, though built in sections through to the 1970s (see Photo 8) (Stammers, 1982). Ellis (1986) estimated some 5,000m³ of volume loss from the beach at Bastion Steps, Peacehaven, 1979-1984. Ellis (1986) also describes substantial profile fluctuations affecting the backshore beach at a site opposite Bayview Road. He ascribed losses to the role of inefficient updrift groynes trapping an excess of potential sediment input. Worthing Borough Council (2009) calculated a loss of 40,000m³ between 2005 and 2007.
Mouchel Consulting (2001a) note that beach sediments tend to be "patchy" in distribution and profiles variable. Groynes are decisive controls on beach width, with progressive downdrift widening within each groyne bay. Where beaches are widest their profiles are steepest, possibly a function of groyne orientation with respect to dominant wave approach. Groyne reconstruction/redesign addressed these problems, providing more opportunity for natural beach morphodynamic response to forcing conditions.
6. Peacehaven Heights to Harbour Heights, Newhaven and Western Breakwater
Between Friars' Bay and Harbour Heights, longer-term, but modest, early to mid-twentieth century depletion has been reversed by net accretion since the late 1960s. Profile steepening has characterised both phases, especially at Harbour Heights, but the upper beach frequently exhibits a well-defined storm ridge and multiple berms. The wide (1800-2000m) beach trapped by the western breakwater protecting Newhaven Harbour shows net accretion (approximately 5,000m³ per year since the 1840s), and crest elevation. The accumulation zone commences some 800m to the west of the breakwater, having migrated slowly in this direction since the 1940s (Robinson and Williams, 1983). Jezard (2004) reports a strong statistical relationship between distance updrift from the western breakwater and beach volume change; as distance increases, the accretion gain diminishes. This conclusion is based on analysis of EA ABMS profiles, 1973-2000. It is confirmed as a continuing trend for the period 2003 to 2007 (Worthing Borough Council, 2009), though there was a marginal loss of 12,000m³ during this period.
7. Newhaven (East Pier) to Tide Mills
Profile form has steepened, but this short sector exhibits seawards advance of both Mean High and Low Water of between 1.0 to 3.0m per year over the last 100 years (Jezard, 2004; Halcrow, 2004). This is ascribed by several authorities to net east to west littoral drift and protection from the direct impact of south-westerly approaching waves. Local wave refraction due to the pier may also contribute to low erosion potential. The net loss of volume of 45,000m³, 2003 -2007, reported by Worthing Borough Council (2009) is not readily explained; monitoring data in this case may not be fully reliable, but is unlikely to be misleading.
8. Tide Mills to Seaford Head
This frontage was renourished and its profile reconstructed in 1987, with subsequent monitoring and maintenance. Further details, together with the previous history of severe depletion, retreat of mean High and Low Water and profile steepening are given in Section 2.4. Since 1987, recycling and renourishments using surplus sediment accumulating at the western end and against the terminal rock groyne at the eastern end has maintained quasi- equilibrium and achieved a small net accretion gain. Between 2003 and 2007 this amounted to 27,000m³ (Worthing Borough Council, 2009) some offshore to onshore gravel feed is a possible contributing factor, but has not been demonstrated satisfactorily.
9. Seaford Head to Cuckmere Haven
Most of the beaches of this sector have been retreating (a mean of 0.5m per year since the 1880s), although the barrier spit beach that has grown across the mouth of the former Cuckmere tidal inlet (Photo 12) has advanced over the longer-term, at 5.0 to 6.0m per year. This trend was accompanied by net accretion and profile flattening. Since the late 1970s, however, beach behaviour has been fluctuating, perhaps because of the insertion of a revetment, and the artificial stabilisation of the mouth of the Cuckmere. More recently, since 2001, beach retreat has been the dominant trend at a rate of 1.0 to 1.5m per year (South Downs Coastal Group, 2006). Worthing Borough Council (2009) state that monitoring data revealed a net loss of volume of 143,000m³ from the foreshore, but with a small increment of accretion on the backshore, between 2003 and 2007. The spit beach to the south of the Haven mouth lost 36,000m³ over the same period, once gains from recharge had been discounted.
10. Cuckmere Haven to Birling Gap
This sector receives high wave energy and is characterised by a narrow, discontinuous, upper gravel beach mostly derived from material released by local cliff erosion that has retreated and steepened since the late nineteenth century. The beach at Birling Gap (Photo 9) occupies a small embayment and is up to 15m in width. Robinson and Williams (1998) observe that volume losses are greatest when waves approach from the south-east. They state that the entire beach has been removed, for brief periods, at least three times during severe winter storms between 1930 and 1995; the most recent occasion was in 2013. Posford Duvivier (1993) describe a residual beach confined to “pockets” on the shore platform in the winter of 1992, when wave notching of the cliff base took place; there was no record of this having occurred during the previous 35 years.
Moore, Collins and King (2001) state that ABMS data, 1973-2000, indicates ongoing beach retreat and narrowing; Robinson and Williams (1998) calculate that this beach was some 100m wider in the 1890s, giving an annual retreat rate of about 1.0m during the twentieth century. Halcrow (2001a) calculate a rate between 1.0 and 1.5m per year for the period 1970 to 2000. However, there was no direct evidence that narrowing has been accompanied by any significant loss of storage volume, as it has also experienced crest elevation and has steepened. This trend apparently changed around the year 2000 with a calculated loss of 65,000m³ from Birling Gap Beach between 2003 and 2008, most of it located in the backshore (Worthing Borough Council, 2009). For the whole sector west to Cuckmere Haven, volume reduction 2003 to 2007 was 386,000m³, concentrated on the foreshore to the immediate east of the Cuckmere inlet. [The source of this data (Worthing Borough Council, 2009) expresses reservations over the reliability of this figure]. Detailed surveys indicate that short sections of this beach have not exhibited clear evidence for significant change of profile geometry since the late 1970s.
Taylor et al. (2004) analysed the evidence for beach profile change over an average period of 113 years for several locations on the English and Welsh coasts, which included three from East Sussex. Their conclusions were based on the analysis of changes in shoreline positions (Mean High Water (MHW) and Mean Low Water (MLW)) from selected profile lines derived from a sequence of editions of large scale Ordnance Survey maps and plans, the earliest dated 1843. Steepening and narrowing was revealed to be a dominant trend, due to the rate of retreat of MLW exceeding that of MHW. This is largely accounted for by the fact that in many locations -including those in East Sussex - MHW has become fixed in position by the construction of seawalls and other “hard” defence structures. However, in a detailed independent re-analysis of the evidence for foreshore contraction along the shorelines of Sussex and Kent, Dornbusch et al. (2008c) consider that there is no clear case for progressive beach steepening over the period 1870 to 1990. Their research is based on measurements of changes in inter-tidal width recorded on Ordnance Survey maps and air photos for successive 50m profiles. Mapping errors alone weaken the evidence inconclusive, and there are several examples where seawalls have been built seawards of the pre-existing position of MHW. Additionally, changes along several profiles can be related to either shore platform erosion or sea-level rise.
11. Birling Gap to Beachy Head (Holywell)
Halcrow Maritime (2001a) have examined foreshore changes for a selection of pocket beaches around Beachy Head, using post -1980 ABMS data. This work confirms Gifford Associated Consultants' (1997) conclusion that beach profile steepening and Low Water Mark retreat has been ongoing since at least the mid-twentieth century at a rate of 1.0 to 1.5m per year.
6. Summary: Sediment Transport Sub-Cells and Budgets
Based on current knowledge of regional pathways of transport of gravel and sand, four sub-cells are delimited. From west to east these are:
- Shoreham Harbour entrance to Brighton Marina (west);
- Brighton Marina (east) to Western Breakwater, Newhaven Harbour.
- Newhaven Harbour to Cuckmere Haven (with a subsidiary sub-division at Seaford Head).
- Cuckmere Haven (east) to Beachy Head.
The boundaries of (i) and (ii) are provided by breakwaters, which have only been fully effective as barriers to longshore transport since the late nineteenth century. Neither are absolute boundaries, as there is strong evidence for their partial by-passing by gravel and especially sand. It has been suggested that Seaford Bay may be a virtually closed sub-cell, with little or no longshore movement of coarse sediment around Seaford Head and beyond Newhaven East Pier; this, however, has yet to be definitively demonstrated. The sub-cell boundary to the east of Cuckmere Haven also lacks definitive proof, but there is sufficient circumstantial evidence in its favour. A transient partial drift divergence boundary has been tentatively recognised within the central sector of Seaford Beach, but it would only appear to operate during the winter period after a sequence of changes of prevalent wave approach.
Gifford Associated Consultants (1997) identified a similar set of sub-cells from application of their regional conceptual sediment transport model (RCTM), but did not recognise the Shoreham Harbour breakwaters as a subdivision within a cell extending from Selsey Bill to Brighton Marina.
There is inadequate quantitative data to calculate full sediment budgets for any of the above sub-cells, but sufficient knowledge is available to present a conceptual understanding of each. One component that is shared by all is that contemporary stores of gravel are considered to be substantially 'fossil'. They represent now partially redistributed and eroded former coarse clastic barrier beaches created during the late Holocene. The offshore supplies of gravel from which they were naturally replenished have either diminished or are no longer moved shoreward by waves in sufficient quantities to offset continuing losses due to net offshore transfers, in situ gravel clast abrasion and concealment beneath protection structures. This issue has been discussed in further detail in Section 1, drawing in particular on the arguments of Jennings and Smyth (1990) and Nicholls (1991). As sediment yield from eroding Chalk cliffs resulting from basal protection structures has declined over the past 100 years, coarse sediment to compensate for losses has become insufficient, resulting in progressive reduction of beach volumes (detailed in Section 5). However, it is important to emphasise that budget deficits prevailed at a number of locations before the construction of shoreline defences. The latter have served to accentuate an existing problem, necessitating measures such as large scale renourishment along frontages experiencing otherwise unrecoverable natural losses.
A notional budgetary statement for each sub-cell is presented below, with explanatory comments. (Note that all beach sectors sustain a small but cumulative loss from abrasional wear of gravel clasts).
Brighton Marina is used as the eastern sub-cell boundary as there is no evidence for any by-passing by coarse sediment. Several studies have concluded that the platform site on which the Marina was constructed historically received negligible input from updrift beaches. Halcrow Maritime (2001b) conclude that the gravel budget for this sub-cell is strongly positive, as there are quantifiable inputs, but no clear observational, experimental and objective evidence for losses. The actual amount of natural bypassing of the Shoreham breakwaters is uncertain. As accretion immediately up-drift of the Marina breakwater is slow, most storage occurs on the beach fronting Kemp Town (see Section 5). It is impossible at present to provide any reliable quantification of outputs - for example, the proportion of gravel in dredge spoil removed from Shoreham Harbour approaches has not been recorded. Gravel input via bedload "creep" from offshore; onshore kelp-rafting and delivery of coarse sediment from the river Adur are also unquantified, but are thought to be collectively a small quantity. It therefore appears reasonable to conclude that exchanges of gravel between the beaches of this sector and offshore are of minor importance. (Halcrow, 2004).
A proportion of the gravel stored in the beach system between Aldrington/West Hove and Kemp Town, Brighton is assumed to be 'fossil', inherited from: (i) updrift littoral input pre-dating the construction of breakwaters at Shoreham Harbour, feeding eastwards spit migration; (ii) onshore barrier migration.
The sand budget is difficult to assess, and cannot presently be quantified. Halcrow Maritime (2001b) consider that it is dominated by longshore and both nearshore and offshore suspended transport and gains and losses may be in balance. Halcrow (2004) presume that there is potential for net offshore to onshore transport, but cannot be quantified. The Marina breakwater is not considered to significantly impede downdrift movement of sand and finer sediment grades. Sand and finer sediments deposited in the marina basin are removed by dredging.
In summary, this is a nearly closed transport cell for gravel, but its boundaries are less restricting on sand movement.
With the exception of cliff inputs, most of the above budgetary elements have been only approximately quantified. The imbalance between losses of beach volume between Rottingdean and Saltdean (see Section 5) and storage induced by the Western Breakwater of Newhaven Harbour indicate that the overall gravel budget is strongly negative. This is demonstrated by the requirements for renourishment of beaches in the 1990s. Further east, however, there may be an approximate balance between gains and losses, with a net positive balance between Friars' Bay and Newhaven Harbour Heights.
The sand budget is wholly unquantified, with no proof that dredged sand and silt removed periodically from the Marina is actually retained in the littoral transport zone. It is more probable that much of this material is dispersed offshore as suspended load.
Newhaven Harbour to Cuckmere Haven
Joliffe (1972) considered that Seaford Bay (East Pier, Newhaven Harbour to Seaford Head) functioned as a virtually closed transport cell even before intensive beach management commenced. This view has been endorsed by Posford Duvivier (1993). Work carried out using fluorescent tracers, diving observation and profiling (Joliffe, 1964; 1972) indicated that wave-induced net on and offshore movements of gravel were imbalanced, producing slow but progressive loss to deep water areas of Seaford Bay. However, his and other later work implied that inputs of both gravel and sand from sources updrift and seawards of Newhaven Western Breakwater, across Seaford Bay, may occur. Despite the robust construction of the Seaford Beach terminal groyne, it is by-passed under high wave energy conditions (Joliffe, 1978, Posford Duvivier, 1993). It is therefore more appropriate to locate the eastern limit of this sub-cell at the beginning of the Seven Sisters cliffs that form an eastern margin to the Cuckmere Haven embayment. Its role as a partial transport boundary has not been directly demonstrated, but May (2003) notes that the significantly higher proportion of rolled flint clasts on Cuckmere Haven spit beach compared to the regional average could indicate their retention there. There is evidence in favour of net eastwards growth of this spit during the twentieth century, punctuated by breaches that resulted in the repositioning of the exit of the Cuckmere River. There may also be a more complex circulation, whereby a proportion of gravel from the beach enters the Cuckmere inlet and is flushed seaward to be temporarily stored in the ebb-tidal delta before being driven back to the beach by wave action. A proportion of this sediment is likely to be moved eastwards by nearshore littoral drift. The lower foreshore morphology is strongly indicative of such processes, but specific investigative studies have yet to be undertaken.
The massive replenishment of Seaford Beach, south-east of Tide Mills in 1987 (section 2.4) is a clear indication that its previous sediment budget was strongly negative. Since then, however, various techniques of beach management have been applied to maintain volume stability, though annual estimated losses and gains have varied.
Progressive loss of sediment from re-entrant beaches between Seaford Head and Beachy Head (Gifford Associated Consultants, 1997) is probably the direct outcome of the conservation of Seaford Beach long before, as well as since, its systematic replenishment. Periodic recycling of coarse sediment on both Seaford and Cuckmere Haven beaches has no impact on the overall net budget. Recharge at Seaford does not appear to have had any downdrift benefits beyond Seaford Head.
Cuckmere Haven to Beachy Head (Holywell)
Although Beachy Head is a substantial, well defined headland that occupies nearly 5km of frontage, it is not an absolute boundary to longshore sediment transport of either gravel or sand. Posford Duvivier (1999) suggest that there is a potential supply of gravel to Eastbourne beaches from east of Cuckmere Haven, but it is unclear how much of this derives from cliff falls and how much is in transit from updrift beaches. Storage is likely within the several re-entrant beaches and/or when large cliff falls create temporary barriers to the lateral movement of flint debris. However, net storage is small, indicating that most erosion products must be removed seaward in suspension (fines) or laterally by drift. Indeed, Nicholls (1991) states that there is no evidence for any net accumulation of gravel on the west facing beaches below Beachy Head over the past 300 years. Halcrow Maritime (2001a) consider that much more sand than gravel by-passes Beachy Head, moving within a wide nearshore and offshore zone. This observation is based on numerical modelling of wave climate, using refraction co-efficients, and analysis of 1990s ABMS data. As there has been little fluctuation in the extent and volume of sandy foreshores of headland beaches since 1985, net onshore and offshore movements appear to be approximately balanced. Rectilinear complex tidal currents probably assist the net eastwards longshore flux of sand.
Posford Duvivier (1993) use unspecified documentary evidence to support their conclusion that the longshore pathway of gravel transport around Beachy Head has become much weaker since the late eighteenth century. The progressive erosion of the Crumbles gravel ness, Eastbourne, since the mid-eighteenth century may be indirect evidence that helps to corroborate this argument, although groynes at Eastbourne may have intercepted its gravel supply. However, Jennings and Smyth (1990) interpret the erosion and retreat of the Crumbles to be a response to diminished feed from offshore sources. This factor may also apply to Beachy Head and the coastline further west.
A significant gravel store comprises the pocket beach occupying the Birling Gap embayment. Moore, Collins and King (2001) attempted to calculate a gravel budget for this beach and argued that longshore transport of gravel had declined over the past 200 years due to the building of breakwaters and sea defence works along the updrift coast as far west as Brighton. The longshore extent of Birling Gap beach has reduced by some 800m since 1874, involving a volume loss of 48,600m³ (386m³ per year). A figure close to this is also calculated by Posford Duvivier (1996). Reasons for this include reduced input of flint gravel from the erosion of the dry valley sediment infill (Coombe Rock) exposed in the cliff line behind the beach. This in turn is due to a reduction in the exposed width of this section, from 230m in 1909 to 170m in 1999. Some 315m³ per year of fresh flints are derived from current cliff degradation. To this can be added approximately 1300m³ per year from the erosion of updrift cliffs, platforms and beaches along the Seven Sisters shoreline. Moore et al. (2001) estimate that losses are at or about 2,000m³ per year at this site, thus giving a net deficit approaching 400m³ per year. As the beach at Birling Gap is lower than those to the immediate east and west, losses may be at a maximum here. A budget deficit is likely to characterise all beaches on this sector.
Note on Cliff Erosion Input:
Estimates of the yield of flint debris from cliff and platform erosion between Rottingdean Head and Beachy Head depend, in particular, on assumptions of percentage flint content of in situ chalk. This is itself spatially variable (Mortimore et al., 2004). Most calculations use a minimum approximation of 5% (e.g. Posford Duvivier and British Geological Survey, 1999; Posford Duvivier, 2001), but 2% is a more justifiable estimate for locations between Cuckmere Haven and Holywell, Beachy Head. The BERM project (Dornbusch, et al., 2000; 2003) calculates a contemporary yield of 4610+ 890m³ per year for all undefended sectors combined between Rottingdean and Beachy Head based on an average flint content of 3-5%. This is lower than other published estimations (and substantially below the 25,000m³ per year calculated by Jennings and Smyth (1990) for the mid to late Holocene period).
7. Opportunities for Calculation and Testing of Littoral Drift Volumes
Data collected by the Defra-funded National Network of Regional Coastal Monitoring Programmes is pivotal for future improvement in estimating beach change. The Programmes consist of topographic beach surveys, nearshore bathymetry, aerial photography, lidar, coastal hydrodynamics (waves and tides) and terrestrial habitat mapping. Specifications for data collection are consistent for all regional programmes and the data and analysis reports are made freely available under the Open Government Licence from www.channelcoast.org.
The Southeast Regional Coastal Monitoring Programme commenced in 2002. The Lead Authority is New Forest District Council, with data collection, analysis and reporting led by specialist teams at the Channel Coastal Observatory (CCO), Canterbury City Council and Adur and Worthing Councils. Longer term Coastal Monitoring Programme data, when combined with other data sets, academic research and historical studies may enable sediment budgets, transport rates and directions to be identified and/or validated in the future.
Estimates of gross and net littoral drift derived from numerical modelling based upon wave hindcasting are available at several points along the shoreline due to previous studies in support of Coastal Defence Strategy Plans (Halcrow Maritime, 2001a and 2001b; Mouchel Consulting Ltd, 2001a; Posford Duvivier, 2001) and the Beachy Head to Selsey Bill SMP2 (Halcrow, 2006). The overall patterns and volumes of drift have been established with medium reliability with variable correspondence between modelled and observed processes (moderate to good agreement between R. Adur and Brighton and on Seaford Beach). Uncertainties encountered in applying these models included: (i) the problem of selecting a representative sediment gain size on the mixed sand and gravel beaches (sediment mobility is highly sensitive to grain size), (ii) the need to estimate (or ignore) the extent to which groynes on the upper beaches intercepted any potential drift, (iii) uncertainty relating to bypassing of the Shoreham and Newhaven breakwaters, (iv) limitations upon transport imposed by poor sediment availability, (v) uncertainties in the manner in which shore platforms could affect transport processes and (vi) high uncertainty in estimates of exchanges of gravel between beaches and the nearshore.
Some opportunities are available for calculation and testing of littoral drift volumes at Brighton (Kemp Town) Beach, West Newhaven (to W of breakwater) and Seaford Beach where beaches are open to inputs from the west, but drift is intercepted by substantial artificial structures or headlands. At such sites it may be possible to infer minimum rates of net drift based on analyses of beach volume change determined from Environment ABMS data. Such estimates could then be compared with corresponding data derived from transport modelling and beach plan shape modelling studies. Seaford may be the most problematic site for beach management affects it most, and there are complexities of nearshore bathymetry and sediment budget introduced by the R. Ouse entrance and breakwaters. A difficulty at all sites except Brighton is that the terminal structures are thought to be partially bypassed by sediment so that drift estimations based upon observed beach volume changes could significantly underestimate the amount of material actually in transit. An analysis is required of drift within the Cuckmere Haven embayment, but there is presently insufficient information assembled for this to be feasible. Further studies as outlined below would be required.
A potentially useful approach with which to investigate some of the uncertainties might be to undertake detailed sediment budget analysis of (i) Seaford Beach (Newhaven west breakwater to Seaford Head) and (ii) Cuckmere Haven (Seaford Head to Seven Sisters). In each case, a gravel barrier type beach occupies a partially enclosed embayment. It may be that by testing alternative input and output scenarios the recorded patterns of historical beach behaviour can be accounted for. Provision of realistic estimates of sediment inputs and losses would assist interpretations of drift using volume changes and aid the setting up of numerical models of drift. It could reduce reliance upon non-measured assumptions of offshore loss/gain previously employed as a means of explaining the lack of fit between modelled and observed changes in beach morphology. With such studies in mind, it is important that beach volume changes continue to be monitored with the Environment Agency ABMS and good records should be maintained of all beach management activities undertaken.
8. Research and Monitoring Requirements
Notwithstanding results from the Southeast Regional Coastal Monitoring Programme, and the summarised information and assessments from Coastal Defence Strategy studies and the Beachy Head to Selsey Bill SMP2 (Halcrow, 2006), recommendations for future research and monitoring that might be required to inform management include:
- The effective application of numerical modelling studies of beach behaviour and sediment transport processes requires the input of high quality nearshore bathymetric survey data, ideally with seabed sediment sampling. This is especially important for those sectors of the near and offshore environments with complex landform and sediment associations, e.g. offshore of R. Adur inlet, around Brighton Marina, shore platform morphology between Brighton and Newhaven, within Seaford Bay and the Cuckmere Haven embayment.
- To understand beach profile changes it is important to have knowledge of the beach sedimentology (gain size and sorting). Sediment size and sorting can alter significantly along this frontage due to beach management, especially the practices of recharge and recycling. Ideally, a one-off field-sampling programme is required to provide baseline quantitative information along this shoreline together with a provision for a more limited periodic re-sampling to determine longer-term variability. Such data would also be of great value for future modelling of sediment transport, for uncertainty relating to grain size is often a key constraint in undertaking modelling.
- Considerable improvement in quantitative estimation of cliff erosion yield has occurred in recent years, although only a few unstable sites are routinely monitored. Further basic research on rates and scales of rock platform downwearing would be appropriate, including some estimation of losses due to solution weathering.
MAP 27aPHOTOSMAP 27bLITERATURE REVIEW
