1. Introduction
The West Solent comprises the tidal channel between Brambles Bank to the east and Hurst Narrows in the west. It has an average width of 4.5km, and is narrowest (1.48km) at its western entrance. It is strongly asymmetrical in cross-section, with a much wider and shallower inter-tidal shoreface (up to 2km in width) along the mainland between Hurst Spit and the mouth of the Beaulieu River than on the Isle of Wight side. The main channel is on average 10-15m deep, but reaches 20m in depth at its eastern end and 60m at Hurst Narrows. Included within this unit is the low-lying northern Solent shoreline extending from the fixed, but partial transport boundary of Hurst Spit (Photo 1) in the west to Calshot Spit (Photo 2) in the east. Both boundaries are by-passed by the transport of fine sediment.
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. Two extensive high resolution, 100% coverage swath bathymetry surveys of Christchurch Bay and the northern shore of the Isle of Wight have been completed in 2010 and 2011, respectively.
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).
1.1 Coastal Evolution
The West Solent represents a submerged and enlarged segment of a previous Pleistocene river system, the Solent River (West, 1980; Allen and Gibbard, 1993). This channel, its main tributaries, the Lymington (Photo 3) and Beaulieu Rivers (Photo 4), and their associated sediments have been strongly influenced by the Holocene sea-level transgression, which has caused submergence, widening, and deposition of sediments by tidal currents and wave action.
In the initial stages of its recent geomorphological development, a Chalk ridge between the Isle of Purbeck (Handfast Point) and the Isle of Wight (The Needles) was breached (possibly initially by fluvial incision) and removed in stages by marine erosion in the late Pleistocene (Wright, 1982; Nicholls, 1987) or early Holocene (Everard, 1954; Velegrakis, et al., 1999; 2000). This had two important effects, which probably developed contemporaneously: (i) rapid erosion of soft Tertiary (Eocene) sands and clays capped by Plateau Gravels, creating Christchurch Bay and (ii) inundation of the channel of the Solent River after the removal of a low elevation isthmus of land connecting the mainland and Isle of Wight shorelines west of the Lymington River (Velegrakis, et al., 1999, 2000; Tyhurst and Hinton, 2001). The general level of the ancestral Solent River channel is approximately -12 to -16mOD in Christchurch Bay and at Yarmouth, Isle of Wight. Nicholls (1987) recognised a former channel, at approximately -14mOD between Hurst Spit and Pennington, based on the recognition of fluvial gravel deposits beneath Pennington as being older than those now exposed at the shoreline. It is hypothesised that the western entrance to the Solent was not created until sea-level transgression reached approximately -12mOD between 8400-6500 years BP (Nicholls and Webber, 1987; Velegrakis, et al., 1999). A date of approximately 7,500 BP is now considered probable (Velegrakis et al., 2000). Linkage between the Western Solent and Christchurch Bay transformed the tidal currents in the area from weak to very strong; its channel was subject to tidal scour and rapidly deepened (Webber, 1980). In this way, the original fluvial valley of the lower Solent River was widened, principally along the north-west coast of what became the Isle of Wight, and converted into a quasi-estuarine channel.
Rapid cliff recession in Christchurch Bay released large volumes of sediments and it is probable that significant quantities of sand and gravel were swept into the Western Solent prior to the full development of Hurst Spit (Nicholls and Webber, 1987; Dyer, 1980). The floor of the West Solent is underlain by gravel-rich Pleistocene niveo-fluvial terrace deposits associated with earlier incision(s) of the Solent River. Such deposits have been recognised from borehole data at Stone Point (NCC, 1979; Green and Keen, 1987; Allen and Gibbard, 1993; Brown, et al., 1975), Pennington (Allen, et al., 1996; Nicholls, 1987) and Hurst Spit (Nicholls and Clark, 1986; Nicholls, 1987). Investigations suggest that Hurst Spit transgressed over low-lying Pleistocene gravel terraces in response to Holocene sea-level rise, but has been relatively stable in both size and position over the past 4000-5000 years since sea-level approached its present position (Nicholls and Webber, 1987). This is also the case for Calshot Spit (Hodson and West, 1972), which overlies saltmarsh deposits bevelled by advancing sea-level during the early Holocene, with little evidence of landward migration since approximately 6300 years BP. Further detail on the geomorphological history of the West Solent is provided in the unit on the Quaternary History of the Solent.
1.2 Hydrodynamic Regime
The Southeast Regional Coastal Monitoring Programme measures nearshore waves using a network of Datawell Directional Waverider buoys. The nearest measurement stations to this cell are at Milford-on-Sea and Hayling Island. The buoy deployed at Milford-on-Sea is in 10mCD water depth, and between 1996 and 2012, the prevailing wave direction was southwest-by-south, and with an average 10% significant wave height exceedance of 1.31m. [The buoy deployed at Hayling Island is in 10mCD water depth, and between 2003 and 2012, the prevailing wave direction was south-by-west, and with an average 10% significant wave height exceedance of 1.26m (CCO, 2012)].
Hurst Spit and the Isle of Wight provide substantial modification to incident waves generated in the English Channel and its Western Approaches. The West Solent wave climate is therefore low energy and at most points fetch limited with significant wave heights between 0.3 and 0.8m (Dyer, 1970b; 1980; Langhorne, Heathershaw and Read, 1982; Ke, 1995; Posford Duvivier, 1994; Halcrow, 1998; New Forest District Council, 2010). Extreme wave heights in excess of 1.6m are however experienced along the extreme east of this shoreline in association with south-east or south-south-east gale force winds operating over the deeper water of the east Solent (Posford Duvivier, 1994; Bradbury, 1995). Return frequencies of significant wave heights at Lymington, Lepe and Calshot are given in New Forest District Council, 2010. ERM (1998) indicate that the outer Lymington estuary has a wave climate dominated by this easterly fetch. The shoreline wave climate between Keyhaven and Pitts Deep is becoming increasingly energetic following erosion of protective mudflats and saltmarshes (Colenutt, 1999, 2001; Bray and Cottle, 2003; Baily and Pearson, 2007 and Cope et al., 2008). This process has led to construction of a series of rock breakwaters and wave screens in the Lymington River estuary to compensate for the loss of natural protection (Colenutt, 2002). Large waves, including swell, from Christchurch Bay can enter at Hurst Narrows, but their energy at the mainland backshore is much reduced by diffraction as well as by the presence of mudflats and (diminishing) saltmarsh.
Tidal range is small, but water movements are considerable because the range varies markedly over the short distance from Calshot (3.9m mean spring tidal range) to Lymington (2.3m) and westwards to Hurst Point (2.0m), with a double high water effect on spring tides (Halcrow, 1998). These differences set up significant hydraulic gradients that are further aperiodically influenced by tidal and surge conditions originating outside the Solent. These tidal factors combine with the morphology of the Solent channels, to influence flow configuration and resistance to flow. Volumetric assessment of tidal flows reveals substantial throughput and a mean east to west residual flow of 1,400m³ per year, although meteorological influences can reverse this direction (Webber, 1980; Sharples, 2000). Ebb and flood currents are also of differing magnitude and duration with the ebb shorter, and therefore faster, than the flood. These currents frequently flow in distinct pathways and not always in exactly opposite directions (Webber, 1980; Dyer, 1980). Current asymmetry increases progressively eastwards. In contrast to the Eastern Solent and Southampton Water, tidal currents are relatively rapid throughout the length of the Western Solent. Peak surface ebb current velocities of up to 3m per second (1.7 to 2.0m per second is more characteristic) have been recorded at Hurst Narrows (Heathershaw and Langhorne, 1988) and up to 1.8m per second at Solent Bank in mid-channel (Hydraulics Research, 1981). These decline to peak ebb velocities of 0.35m per second, and peak flood flows of 0.14m per second in the inner Lymington estuary (ERM, 1998; Black and Veatch, 2012a). The mobility of main channel sediments is therefore much greater and only coarse materials are stable on the bed for any length of time. The West Solent is thus a dynamic environment subject to sediment transport flux dominated by a complex pattern of tidal currents over all but its most easterly sector. Highest energies are achieved when very strong south-easterly winds and waves coincide with peak ebb tidal current velocities.
In contrast to the main channel, the north-west Solent shoreline is sheltered by Hurst Spit and the Isle of Wight. The gently shelving offshore and nearshore (shoreface) and inter-tidal gradients generally prevent rapid tidal flow close to the shore. Considerable clay and silt sedimentation has occurred in this low energy environment and extensive saltmarshes have developed (Ke and Collins, 1993). Although accretion has predominated, its spatial and temporal distribution has been variable. The overall sediment budget of the mudflats has been influenced by human activities such as land claim, port development, coast protection and dredging (Ke and Collins, 1993; Bray, et al., 1995).
2. Sediment Inputs
Analysis of Coastal Monitoring Programme 2005 and 2012 lidar and aerial photography datasets and 2003 and 2012 topographic baseline survey data, combined with other datasets, academic research and historical studies has enabled sediment budgets, transport rates and directions to be identified or verified.
2.1 Marine Input
» F6 · F7
In the 2004 version of the STS, it was postulated that coarse sediment may enter the West Solent through the Hurst Narrow during exceptional conditions. Analysis of Coastal Monitoring Programme swath bathymetry data provides evidence of a series of three 10-20m high, steep angled, sub-marine terraces that are located in 25-60m water depths between Fort Victoria and Hurst Point (Photo 11), associated with the formation of the Solent and the former position of the Solent River. These large-scale features restrict or inhibit bedload transport into the West Solent from Christchurch Bay. Examination of tidal curves for Lymington, Yarmouth (Isle of Wight) and Totland reveal marked asymmetry, because the ebb flow is concentrated into a shorter time period than the flood (Webber, 1980). The ebb flow is therefore considerably more rapid than the flood and transport of coarse bedload sediments (sand and gravel) is therefore likely to be in a net seaward direction, determined by peak current velocities. The nearshore bathymetry in the Hurst Narrows channel indicates a scoured bedrock bed. As a result, the 2004 F1 arrows have been removed. Furthermore, it was considered that the text for the F2 arrows did not refer to wave driven offshore to onshore transport, and therefore the F2 arrow has now been changed to a T0 ‘tidally driven’ arrow, to demonstrate that the suspended sediment is transported due to tidal transport [see T0]. The F5 arrows have also been changed from ‘wave driven offshore to onshore’ to ‘estuarine sediment transport’ as the text does not refer to wave driven transport. The transport of fine sediment into the Lymington and Beaulieu Estuaries is likely to occur on the flood tide, rather than waves carrying the sediment up the estuary. The F5 (2004) arrows are therefore now (EO2) [see EO2]. F3 and F4 can be found in the interrelating North West Isle of Wight Maps.
Studies of the episodic development of Warren Farm Spit, which is composed of coarse gravel, using map comparisons and field survey revealed episodic spit growth and eastward extension between 1898 and 1976 (Human, 1961; Dobson, 1964; Sawyer, 1976; Clark and Gurnell, 1987; Hooke and Riley, 1987; Hydraulics Research, 1987; Williams, 1988; Lobeck, 1995). Analysis of Coastal Monitoring Programme data provides evidence of continued episodic realignment and reorientation of the spit in response to occasional storm generated waves (e.g. 2008). However the eastward extension is limited due to tidal currents at the Beaulieu River mouth and low rates of longshore transport, resulting from a starvation of source material. The spit is therefore relatively stable in extent but morphologically variable in response to extreme wave and tidal conditions Analysis of aerial photography (2013), topographic baselines (2003 to 2012), lidar (2005 and 2012), and single beam bathymetry surveys of the intertidal morphology provide evidence that the inter-tidal and sub-tidal swash bars and other gravel features have migrated shoreward, but volumes of material have been deposited onshore (Photo 5).
Analysis of Coastal Monitoring Programme data, including aerial photography (2013), topographic baselines (2003 to 2012), lidar (2005 and 2012), and single beam bathymetry surveys of the extensive intertidal morphology, provides evidence that the swash bars and other gravel features (Photo 2) are relatively stable in terms of volume but vary in terms of morphology and position in response to extreme wave and tidal conditions. They have migrated shoreward by up to 70m 2001 to 2013 but limited volumes of material have been deposited onshore.
2.2 Fluvial Input
FL1 Lymington and Beaulieu Rivers
Two main rivers drain into the West Solent, the Lymington and Beaulieu Rivers, for which no long-term measured discharge and sediment input data are available. Rendel Geotechnics and the University of Portsmouth (1996) calculate that both rivers, together with Bartley Water, Dark Water and Avon Water, contribute approximately 785 tonnes per year of suspended load to the West Solent. Sedimentological studies of the Beaulieu River indicate that the upper estuary sediments are sandy muds, much coarser than the fine clays and silts closer to the estuary mouth (Codd, 1972). Both the major and minor rivers, notably the Lymington, possess major impediments to sediment transport, such as dams and weirs. The Lymington causeway was built in 1731, and has removed most potential fluvial sediment input. Codd (1972) concluded that the fine materials of the mudflats that flank the Beaulieu estuary are predominantly derived from the Solent, with only very small quantities of coarser sediments supplied by the river. Subjective and somewhat imaginative assessment of the morphometry of the Beaulieu estuary suggested that it is a tidal palaeomorph as opposed to being a submerged fluvial valley (Geyl, 1976a and b). If this interpretation is correct, the meandering channel might have developed in response to tidal currents during a time of fluctuating sea-levels. Although speculative, this hypothesis is partly corroborated by the contemporary sedimentological evidence of Codd (1972) and indicates that marine sediment input is likely to have far exceeded fluvial input throughout its Holocene history.
2.3 Coast Erosion
Much of the coast between Hurst Spit and Calshot is low lying and protected by seawalls, earth embankments and revetments (Hydraulics Research, 1987; Oranjewoud, 1988; Southern Water, 1989; Halcrow, 1998; Posford Duvivier, 1997; 1999; NFDC, 2005; 2010). Such areas are more liable to tidal flooding than coast erosion and are likely to act as sediment stores rather than sources. Despite this, coast erosion is active at some locations and low cliffs of restricted extent are capable of supplying sediment (Ke and Collins, 1993, Posford Duvivier, 1997). The wide inter-tidal foreshore, seawards of confining protection barriers, is subject to erosional scour and entrainment of fine grained sediment at rates that reflect its exposure to incident waves, its width and water depths (Posford Duvivier, 1999).
Analysis of Coastal Monitoring Programme 2005 and 2012 lidar and 2013 aerial photography, 2005 to 2012 topographic baseline survey data, combined with other datasets, academic research and historical studies has enabled sediment budgets, transport rates and directions to be identified or verified. Taking the eroding shoreline of the north-west Solent as a whole, Bray et al. (1998) suggest a total cliff sediment yield of some 7,000m³ per year, but only between 1,000 and 2,000m³ per year represents coarse clastic debris that is retained on local beaches. Suspended sediment yield resulting from shoreface erosion is in the range of 2,500 to 5,000m³ per year (Posford Duvivier, 1999). The fate of this material is unknown, but it is presumed that a large proportion is removed from the West Solent by tidal flows.
The shoreline between Lymington and Pitts Deep is sheltered from wave action by wide (but eroding) mudflats and saltmarshes so that there is little beach development or effective drift, and potential for sediment supply is limited. Exposure is greater along low-lying shores to the east of Pitts Deep where a transgressive upper gravel beach (Photo 7) is controlled by a variety of groynes and other coastal defence constructed piecemeal along numerous small private frontages (Venner, 1986a; 1993).
Analysis of Coastal Monitoring Programme data confirms the beach widths, heights and sediment thicknesses along this frontage are stable with low, minimal rates of west to east along-shore transport of mixed sediment. The continued linear rates of saltmarsh and mudflat erosion to the west increases exposure of the low lying shoreline between Pitts Deep and Sowley, which releases limited volume of sand and mixed sediment. Low cliffs along the Pitt’s Deep frontage have experienced accelerated rates of recession, estimated at 0.8m per year, since the early 1980s (New Forest District Council, 2010). The Sowley foreshore fronting the ephemeral marsh-lagoon feature south of Sowley Pond, continues to experience cyclic, episodic breaching and sealing in response to infrequent but extreme wave conditions, with cross-shore movement of material between the beach or spit bank and the lower inter-tidal foreshore.
Active erosion along a 600m long section of coastline between Inchmery House and Lepe has formed a distinctive series of retreating cliffs up to 6m high cut into Tertiary sands capped by Pleistocene river terrace gravels (Photo 6). Analysis of Coastal Monitoring Programme data confirms the continued reduction of beach cross-sectional area and simultaneous erosion of the cliffs at Inchmery, which releases sand and mixed sediment to the foreshore, a veneer of sand and mixed sediment overlying uncompacted mud. The coarsest fraction appears to be transported cross-shore onto the muddy inter-tidal foreshore, with the finer mixed and sand fractions being transported westwards, although rates and volumes are low. The fronting beach has narrowed as sediment has been transported westwards to a stable ness-like feature, which is incrementally extending west and south. Further east, however, beach widths, heights and sediment thicknesses along this frontage are stable with low, minimal rates of eastward along-shore transport of mixed sediment. To the east, between Lepe House and Stone Point, the toe of the cliffs are protected by a series of timber and rock revetments so that the cliffs are now inactive and vegetated (Hampshire County Council, 2004). Posford Duvivier (1994, 1997) note some acceleration of erosion, and falling beach levels, possibly due to the migration of the Beaulieu River channel towards this shoreline (some 30-50m since the early twentieth century). This may be in response to the elongation of Needs Ore Spit. Current erosion yield is calculated at 2,000m³ per year, approximately one half of which is coarse (gravel) sediment. Monitoring since 2000 indicates continuing reduction of beach cross-sectional area along the Inchmerry House frontage (New Forest District Council, 2010).
Marine and sub-aerial erosion along this segment has formed low, discontinuous cliffs at Stone Point and to the east of Stansore Point extending to Bourne Gap. Although they extend to no more than 9m above OD, (average of 5m) historical cliff toe recession and sub-aerial mass movement has produced a locally important sediment supply to beaches due to their dominant gravel composition. These cliffs are inactive and partially vegetated due to the effects of coast protection at Stone Point and the accretion of a wide protective gravel beach along the southern half of Stanswood Bay. Hooke and Riley (1987) calculated that 17m of net accretion of the foreshore between Stone and Stansore Point occurred between 1865 and 1985 thus isolating cliffs from marine erosion. Stratigraphic studies revealed that the cliffs are composed of four distinct deposits of mid to late Pleistocene age (West, 1987; Green and Keen, 1987; Allen and Gibbard, 1993). These comprise upper and lower gravel units, each of 2-3m thickness, with an intervening silty horizon. The sequence is overlain by a 1m thickness of Brickearth. Both gravel units are interpreted as terrace deposits of the former River Solent and contain considerable quantities of angular flint clasts. Map comparisons covering the period 1868-1994 reveal cliffline retreat of between 0.20 and 0.27m per year (Halcrow, 1998) although cliff recession would have ceased well before 1994.
Low timber revetments and groynes installed between Lepe and Bourne Gap (Hampshire County Council, 2003) were severely damaged during 2008 storms. Analysis of Coastal Monitoring Programme data however, confirms the continued erosion of the low cliffs east of Stone Point, which releases Pleistocene sand and mixed sediment to the foreshore, although rates and volumes are low, less than 1,000m³ per year; a reduction from the 2004 estimated volume of 3-10,000m³ per year. However, monitoring since 2000 has revealed a slight increase in erosion rates in comparison to the previous 40 years (New Forest District Council, 2010). Losses are characteristically episodic, associated with extreme wave and /or antecedent weather conditions.
Low eroding cliffs cut into sand, clay and silt of the Barton Sand (Eocene) and Headon Hill Formation (Oligocene) and Pleistocene gravel are present along this segment (Hinsley, 1990; Posford Duvivier, 1999), with an average height of 6m (maximum 10m). Map comparisons covering the period 1868-1971 indicate overall, long-term retreat (Hooke and Riley, 1987). Coastline recession of 1.5m per year over the period 1967-87 (Hydraulics Research, 1987) and 0.2 to 2.0m per year (Oranjewoud, 1988, 1990; Posford Duvivier, 1999) are reported. Posford Duvivier (1997) give a mean of 0.5m³ per year for the period 1898-1976. The late Pleistocene river terrace gravels exposed in the eroding cliffs have a significant potential to supply coarse material to the local beach in Stanswood Bay. Basal erosion and sub-aerial mass movement is active (discrete events) at specific sites, including as a specific feature localised re-entrants due to groundwater seepage. For the cliffline fronting Stanswood Bay, Lock (1999) calculates a potential yield of gravel of 1,420m³ per year, with over 500m³ available for beach storage. For Stanswood Bay frontage, Posford Duvivier (1997) propose a total yield of 5,000m³ per year; of which 1,000m³ per year is gravel that contributes directly to the local beach. Rates of erosion since 1989 have marginally declined in comparison to the previous thirty years, apparently in response to the increased height and width of the fronting beach (New Forest District Council, 2010).
Map comparisons and aerial photo analysis have revealed significant foreshore erosion and narrowing of the intertidal zone since the 1950s (Tubbs, 1980; Hydraulics Research, 1987; Hooke and Riley, 1987; New Forest District Council, 2010; CCO, 2012). This process probably results in significant release of fine sediments; however it is not regarded as sediment input but a redistribution of existing sediment.
In addition to the routine data collection and analysis undertaking by the Coastal Monitoring Programme, an additional site-specific study was undertaken following the removal of a concrete case pipeline from the Cadland foreshore between Bourne Gap and Hillhead, Calshot in September 2002. A regular schedule of monitoring of foreshore and cliff top positions at Calshot Cliffs between 2005 and 2009 showed localised erosion of the beach and cliff face, lowering of beach levels, and removal of cliff-derived sediment from the foreshore (CCO, 2009a). 3m of cliff top retreat was recorded during the survey period, which compared to 15m of recession between 1986 and 2006. This was directly related to crest cutback and reduction of cross-sectional area of the fronting beach.
E13 Keyhaven, Lymington, Beaulieu, Calshot mudflats and saltmarshes (see introduction to coast erosion)
Although most areas of mudflat and saltmarsh have eroded considerably during the 20th century, a substantial area of inter-tidal mudflat and saltmarsh is still present along the northwest shore of the Solent. It is most extensive between Keyhaven (Photo 1) and the Lymington River (Photo 3), extending east but becoming increasingly fragmented towards Pitt’s Deep. It is also highly fragmented in the mouth of the Beaulieu River (Photo 5), and between Lower Exbury and Inchmery House it extends as narrowing margins upstream. From Pitt's Deep east to the Beaulieu River, saltmarsh is absent, and there is only a relatively very narrow fringe of muddy lower foreshore. E13 arrows have been added to represent input of fine sediment (silt and clay) from saltmarsh erosion.
Photogrammetric studies of saltmarsh change between 1946 and 2001 revealed major losses in saltmarsh area, primarily as a result of the die-back and loss of Spartina marsh (Colenutt, 1999, 2001; Bray and Cottle, 2003; Baily and Pearson, 2002; 2007; Cope et al., 2008). Colenutt (2001) estimated that saltmarsh erosion losses were likely to continue in the future, at a continued linear rate for Keyhaven and Lymington. Bailey and Pearson (2002) estimated 54ha (total for both areas) would remain in 2050 and only 11 ha in 2100, which were similar to predictions arrived at by Colenutt (2001). On this basis, it is clear that fine sediments are likely to continue to be released into the estuary as marsh retreat proceeds. Refer to section 6 for full details of losses sustained in the West Solent determined by the researchers mentioned above.
The sediment store of fine-grained sediment currently "locked up" as saltmarsh and mudflat is being relatively rapidly transferred to suspended load throughput and loss from the Solent system. The implication is that the current hydrodynamic regime strongly favours store depletion, and that mudflats and outer marshes are in disequilibrium with forcing factors. As the causes promoting erosion are unlikely to alter in the foreseeable future, this situation will continue. Indeed, as relative sea-level rise accelerates, it will probably intensify in coming decades.
2.4 Channel Erosion
The bed of the West Solent is mantled by extensive Pleistocene river terrace deposits of a similar nature to those recognised at Hurst Spit, Pennington and at Stone Point (Nicholls and Clark, 1986; Green and Keen, 1987; Allen and Gibbard, 1993). Significant scour of the West Solent tidal channel has occurred in areas where tidal currents are rapid e.g. Hurst Narrows (Dyer, 1970b; Webber, 1980; Halcrow, 2002). This process involves mobilisation of considerable volumes of sand and gravel from these deposits, together with some fines from underlying Eocene bedrock. The coarser sediments have been deposited in a series of banks within the channel of which Solent Bank is the best known. It must be asked whether contemporary erosion of the channel bed continues to supply sediment to the West Solent system. An equilibrium may have developed whereby the bed has become armoured by mobile coarse sediments and further scour is now limited. If this is the case the gravel banks of the Western Solent represent a finite deposit.
Sand and gravel deposits at Solent Bank were dredged between 1950 (Hydraulics Research, 1981) and 1994. Detailed examination of bed levels in the vicinity of Solent Bank involving chart comparisons revealed significant changes, including patches of erosion that lowered the bed by up to 3m, over the period 1963-1973 (Hydraulics Research, 1977, 1981). It was uncertain whether this involved bed erosion of in situ Pleistocene gravel deposits or redistribution of existing mobile sediments, although the latter is assumed. It could be that dredging over Solent Bank removed the armoured surface gravel veneer exposing looser or softer materials to scour by tidal currents, so that renewed localised bed erosion became possible. The occurrence of extreme wave and tide conditions may also be important in eroding or re-mobilising the bed materials, but positive evidence is extremely limited. Rapid shoreline gravel accretion was recorded at Warren Farm Spit after a storm in 1952 and an offshore source of supply of this material was postulated (Hydraulics Research, 1987). Such supply may be related to intermittent scour of channel deposits, but this suggestion remains speculative.
Gravel transport within the West Solent Channel is addressed in further detail in section 4, and dredging of Solent Bank is covered in section 5.2.
3. Littoral Transport
Littoral drift is not a major process in the West Solent because of the shelter to wave action provided by Hurst Spit and the Isle of Wight. Waves are therefore fetch limited and being usually of low height and short period they do not have much influence on the main channel bed (Langhorne, Heathershaw and Read, 1982). Littoral drift of coarse sediments is therefore restricted to more exposed parts of the upper shoreline from Pitts Deep to the Beaulieu River and eastwards from Lepe, where it is supplied by erosion of gravel and gravel-sand formations exposed in the low cliffs.
Detailed studies of the historical pattern of littoral drift on Hurst Spit have indicated a long-term west to east supply from Christchurch Bay and transport to Hurst Point, with material also transported northwards from Castle Point to the distal end of the spit, North Point. At Hurst (Castle) Point, a divergence of transport was recognised with some material being supplied to the Hurst Narrows tidal channel and the remainder drifting northwards to recurves on the spit at Hurst Point (Nicholls and Webber, 1987). The continued drift from Hurst Point to North Point results in accretion at the distal end of Northpoint, dredged every five years with the material being recycled back on the eastern side of the spit (Colenutt, 2002). Analysis of Coastal Monitoring Programme lidar (2005 and 2012), aerial photography (2013), and topographic baseline (2005 to 2012) data, in conjunction with beach operations and recycling records, has confirmed the consistency of sediment transport pathways, volumes and rates of coarse sediment from Christchurch Bay, alongshore of Hurst Spit, and between Hurst Point to North Point. Distal recurvature is the product of wind waves generated over local fetches, and refracted swell waves. The calculated volume of northward drift between Hurst Point and North Point is 1-3,000m³ per year (New Forest District Council, 2010). Analysis and seabed mapping interpretation of a combination of extensive high resolution, 100% coverage swath bathymetry surveys of Christchurch Bay (2010) and the northern shore of the Isle of Wight (2011) collected through the Coastal Monitoring Programme, indicate that coarse and mixed sediments are evident on the immediate fringe of Hurst Spit, south of the Castle to Hurst Point.
Net transport in the Hurst Narrows tidal channel is offshore due to the dominant ebb current and this pathway is regarded as a long-term supply to the Shingles Bank. It is widely held that intermittent pulses of sediment can be transported into the western Solent from Hurst Point by wave action during storms (Dyer 1970a, 1971, 1980; Webber 1977; Hydraulics Research, 1987). This view is largely unsupported by direct evidence and no studies have evaluated the periodicity of formative event(s) or how it might relate to accretion of Hurst Point. The possible quantitative significance of this input is therefore impossible to evaluate. Detailed sampling of bed sediments has been undertaken from a limited area on the bed of the southern margin of the main channel between Yarmouth and Hampstead Ledge (Langhorne, Heathershaw and Read, 1982). Analysis revealed that roundness of clasts was insufficient for the main supply source to be from relatively well rounded contemporary beach gravels on Hurst Spit. Further details relating to the morphodynamics and sedimentology of Hurst Spit are given in the unit on Christchurch Bay.
Studies of the episodic development of Warren Farm Spit (Photo 5), which is composed of coarse gravel, using map comparisons and field survey revealed intermittent growth in an eastward direction between 1898 and 1976 (Human, 1961; Dobson, 1964; Sawyer, 1976; Clark and Gurnell, 1987; Hooke and Riley, 1987; Hydraulics Research, 1987; Williams, 1988; Lobeck, 1995; New Forest District Council, 2010). Occasional storms appear to have been very influential in providing sediment promoting rapid short-term growth, e.g. some 100m of distal extension across former mudflats in 1952/3, (Halcrow, 1998). This pattern of development strongly indicates supply and distribution of gravel by eastward drift. An unknown fraction of this supply may be supplied from nearshore banks of coarse to medium grade sediment, possibly deriving from the now largely relict Beaulieu River ebb tidal delta. Erosion of the saltmarsh between Pitt’s Deep and Sowley in the thirty years up to the early 1950s led to a breach in the barrier at Sowley in 1955. Coarse sediment supplied to the western spit thus created, derived from ongoing erosion of the adjacent intertidal foreshore, fed its eastwards extension until the breach was finally closed in 2008 (Cope, et al., 2008; New Forest District Council, 2010.) Net eastwards drift along this frontage was therefore clearly illustrated. In 1986, a tidal channel (Bulls Gap) separating Needs Ore Point and Warren Farm Spit from Gull Island, first opened in 1727, was closed by building a causeway and using 13,000 tonnes of gravel recharge (Clark and Gurnell, 1987; Venner, 1986b; Lobeck, 1995). After storms in 1987, it is reported that gravel was transported eastward from Warren Farm Spit to supply the Gull Island barrier feature (Williams, 1988), and its eastwards and northwards growth and realignment has continued since (Halcrow, 1998; New Forest District Council, 2010). Bradbury et al. (2005) applied a barrier inertia model to Gull Island and concluded that whilst it was sensitive to surge events, it would be unlikely to be overtopped in any combination of extreme wave and tidal conditions.
Analysis of Coastal Monitoring Programme lidar (2005 and 2012), aerial photography (2013), and topographic baseline (2005 to 2012) data, has confirmed the consistency of weak rates of less than 1,000m³ per year eastward alongshore transport of coarse/mixed sediment (CCO, 2009b, 2012).
Tubbs (1999) has stated that the mouth of the former tidal inlet of Dark Water and the Mopley Stream were closed before 1600, with both valleys rapidly accumulating alluvial sediment. It is uncertain if this is indicative of spit growth fed by longshore drift, as closure might have been effected by onshore barrier migration. Both streams now discharge via seepage through their gravel "dams". Maps produced by Hydraulics Research (1987) indicate a net north-eastward drift within Stanswood Bay, but this information is of low reliability because sources of evidence were not indicated. Eastward drift is indicated at Hillhead and at the proximal sector of Calshot Spit by the distribution of gravel against groynes and other structures. A rate of approximately 1,000m³ per year has been calculated using BEACHPLAN model analysis (New Forest District Council, 2010). This study indicated that most sediment movement here is cross-shore.
Analysis of Coastal Monitoring Programme data confirms that the beach fronting the eroding Inchmery cliffs has narrowed as sediment has been transported westwards to a small stable ness-like gravel feature on the shore in front of Inchmery House, which is incrementally extending west and south (60m westwards and 35m southwards, 1954 to 2001, determined from air photo interpretation - New Forest District Council, 2010). The development of this feature also confirms the local drift reversal between Lepe and Inchmery (Photo 6); this is explained by the local shelter from south-westerly waves produced by the eastward and southwards growth of Gull Island and Warren Farm spit (which continued between 2003 and 2009) across the mouth of the Beaulieu estuary leaving the shoreline exposed only to waves approaching from the southeast. Beach volume loss at Lepe over the same period may be accounted for by the growth in storage of Warren Farm spit.
Erosion of the cliffs at Inchmery releases sand and mixed sediment to the foreshore. The coarsest fraction appears to be transported cross-shore onto the muddy inter-tidal foreshore, with the finer mixed and sand fractions being transported westwards, although rates and volumes are consistent and less than 1,000m³ per year. Posford Duvivier (1994) estimated a potential drift rate of 1,900m³ per year in the vicinity of Lepe, with net transfer from east to west, with half of the material assumed to be coarse-grained.
Between Lepe and Calshot beach widths, heights and sediment thicknesses along this frontage are stable with low rates of north-eastward along-shore transport of mixed sediment, with low rates of cross-shore transport evident. Rates and volumes are consistent and less than 1,000m³ per year between Stone Point and Stansore Point, and is probably the product of a locally complex interaction of refracted waves from different fetch directions. A consistent rate of less than 1,000m³ per year between Stansore Point and Hillhead has been calculated.
The alignment of Calshot Spit is generally taken to indicate drift towards the north-east (Dyer, 1980; Hydraulics Research, 1987; Hinsley, 1990). Long-term spit stability over the past 6300 years (Hodson and West, 1972), including the past 140 years of documentary records (Hooke and Riley, 1987; Oranjewoud, 1990), may indicate constancy of this transport pathway. The stability of shape and position of Calshot Spit since the construction of the Castle in the 1530’s suggests that gains and losses balance in the long-term. Between Hillhead and the proximal attachment of Calshot Spit the north-eastward drift rate increases to 1-3,000m³ (closer to 1,000) per year, which is lower than the 2004 estimated rate of 3-10,000m³ per year. This increased rate may indicate the influence of onshore feed to the spit (see F7) although no conclusive evidence is indicated from analysis of the nearshore bathymetry data.
A modest renourishment of Calshot beach adjacent to the proximal part of the spit was undertaken in 1994/5, however further information is not available. The existing groynes were installed in 1991 extending from the proximal end of the spit toward Houston House. The groyne bays did not receive replenishment in the period between installation, and 2012. Lock (1999) has calculated that the gravel foreshore of the axis of Calshot spit has prograded between 0.09 and 0.15m per year since 1890. Oranjewoud (1990) ascribe the absence of foreshore erosion to the steep slope into the central channel and the width of the gravel-dominated fronting inter-tidal zone. If the erosion yield of the updrift cliffline between Lepe and Hillhead to the littoral drift pathway is in the order of 2,000m³ per year (Lock, 1999; Bray, et al., 1998), some 220,000m³ of gravel has been input into Calshot spit since approximately 1870 (date of earliest large scale Ordnance Survey plan). This would indicate a modest net growth of the volume of material stored by Calshot spit over the past 120-130 years, but a more precise figure can only be calculated once losses to the near and offshore environments are quantified (Halcrow, 1998). This information is currently unavailable. The well-defined distal point may be due to erosion by tidal scour, but most studies suggest that Calshot Spit is, in a historical context, a quasi-equilibrium form. Recognised accretion phases, e.g. 37m of seaward growth near its proximal point 1867-1910 (Oranjewoud, 1990) must be balanced by periods of depletion. The latter, however, have not been specifically identified from map evidence.
Lobeck (1995) was able to demonstrate that sets of groynes constructed during the period 1868-1971, between Stansore Point and Calshot, were the direct cause of HWM recession, especially the proximal sector of Calshot Spit between 1910 and 1931. Stability returned between 1935 and 1971 and thereafter, when most groynes had virtually ceased to function due to neglect. Since 1993, backshore "chevron" groynes have been installed, and have helped to maintain stability.
4. Tidal-Current Sediment Transport
Low wave energy within the Western Solent results in transport of fine-grained sediments almost exclusively by dominant ebb tidal flow. The main pathways are in the central channel and close to the edge of eroding saltmarsh (Ke and Collins, 1993), as revealed by measurements of suspended sediment concentrations. Predominant movement may be east to west, towards the Hurst Narrows exit. Highest rates of transport occur when strong easterly winds coincide with ebb tidal currents. Channel bed sediment mobility induced by tidal currents has been the subject of much previous research, undertaken in three main phases:
- Broad scale descriptive and analytical survey in the late 1960s by Dr K.R. Dyer, then of Southampton University
- Dredging of Solent Bank, studied by Hydraulics Research in the late 1970s and early 1980s.
- Fundamental entrainment processes studied within a range of field and laboratory experiments by the Institute of Oceanographic Sciences (now National Oceanography Centre) in the 1980s.
Results from these studies have been assessed for reliability and compiled to create the sediment transport pathways indicated on the accompanying map.
(A) Sediment Distribution and Mobility by Dr K. R. Dyer
Investigations involved spatially extensive survey by echo-sounder and oblique asdic, and with diver observations to confirm the accuracy of the survey techniques. Tidal currents were measured at a variety of depths using meters (instantaneous measurements) and float tracking (averaged measurements). Sea bed sediments were studied by 500 Shipek grab samples taken from throughout the West Solent. Survey was therefore aimed at obtaining a broad view of sediment distribution, mobility and the influence of tidal streams. These techniques were less successful in providing intensive coverage at precise sites, so analysis of fundamental entrainment and sedimentation processes was limited.
These surveys revealed that much of the West Solent floor was covered by sedimentary bedforms, mostly sand and gravel waves. Crest orientation was generally at right angles to the trend of the main channel and the larger waves had pronounced asymmetry (Dyer, 1971). This proved a reliable guide to sediment transport direction and in conjunction with sedimentological indicators (e.g. fining in the direction of transport) it was possible to identify net sediment transport pathways. As bedform distribution is generally stable in the short term, it was postulated that sediment transport pathways are also relatively steady. Analysis revealed that sediment moved in different directions on opposite sides of the channel and there was often rapid change in the direction and mode of transport over short distances. Mapping of the transport pathways revealed a general west to east transport in the north part of the channel, with a series of local reversals caused by recirculating current eddies on the opposite side (Dyer, 1971). Measurement of tidal currents revealed small but important differences in both the strength and direction of ebb and flood currents. These often meandered, with small differences between ebb and flood currents resulting in slight net sediment transport. Repeated over many tidal cycles these net movements combine to produce three recirculating eddies of sediment off Egypt Point, Newtown Harbour and Hamstead Ledge, Isle of Wight (Dyer, 1971). Zones of accretion (stores) occur where a recirculating eddy meets an eastward moving transport pathway, and it was suggested that these positions are occupied by Solent Bank and Prince Consort Shoal (Dyer, 1971, 1972). Sedimentological analysis indicated that net eastward transport continued to Cowes where diminution of tidal current velocities causes deposition of coarse sand on Prince Consort Shoal and medium sand on Brambles Bank. It is suggested that these shoals and banks represent sediment sinks for the respective grain sizes.
Analysis of sediment samples revealed a series of grain size zones primarily based upon the relative proportions of the sand and gravel modes (Dyer, 1971). Gravel, with variable proportions of coarse sand and silty clays, was the main bed material. Calculation of peak bed shear stress from tidal current measurements indicated gravel was potentially mobile over most of the West Solent and that sand was unstable on areas of the bed subject to net sustained sediment transport (Dyer, 1970a, 1971, 1972). Further experiments were conducted into the spatial variations of sand distribution, and the ratio of trough to crest shear stress was found to be an important control on sandwave morphometry. Conclusive results could not be achieved due to difficulties of taking measurements over specified parts of bedforms. Limited studies were undertaken by divers to evaluate rates of transport, but data collected could not be converted to general mass transport rates as it was measured over restricted timescales and limited spatial areas (Dyer, 1980).
(B) Detailed Studies of Solent Bank by Hydraulics Research
The first phase of investigations comprised a literature review followed by comparison of eight hydrographic surveys and charts covering the period 1847-1973 (Hydraulics Research, 1977). This demonstrated the long-term historic stability of Solent Bank characterised by cyclic growth and recession. After 1960, and the commencement of large-scale aggregate dredging, Solent Bank was subject to continuous recession (see Section 5.2 for details of dredging history). However, comparison of dredged volumes with the measured lowering of the bank revealed replenishment of the licensed dredging area by 502,000m³ per year. The significance of this figure is difficult to evaluate because it is uncertain whether it represents a natural supply of new material, or simply a redistribution of sediment from neighbouring areas to fill dredged holes. Transport pathways facilitating replenishment could not be established, so this supply must simply be regarded as an estimate of potential gross transport occurring within the disturbed area.
A second phase of investigation involved further chart comparisons, repeated echo sounding surveys, side-scan sonar survey, current metering and sediment sampling using grab and vibrocore techniques (Hydraulics Research, 1981). Comparison of Admiralty hydrographic charts covering the period 1965-1975 revealed net accretion on the western margin of Solent Bank and erosion to the east. This pattern of change supported evidence of supply by the dominant eastward sediment transport pathway recognised by Dyer (1971). If this interpretation is correct, bed lowering during this period by dredging in the eastern sector of Solent Bank must have intercepted a significant proportion of eastward moving sediment. (See Section 5.2 for details of dredged volumes). Echo sounding surveys repeated on seven occasions between 1979 and 1981 showed a similar pattern of bed lowering but, with lower dredged volumes, replenishment reduced to 45,000m³ per year. This supports a mechanism of partial replenishment, primarily by redistribution from neighbouring areas at rates controlled by the size of dredged holes. Alternatively, it can be argued that hydrographic surveys are difficult to repeat precisely, so detailed analysis of volumetric change covering short periods can be inaccurate. It must be concluded that comparison of bathymetric charts and hydrographic surveys can yield valuable information on long-term trends and bed changes, but gives information of a less reliable nature relating to short-term changes. The directions and magnitudes of sediment transport pathways involved in bed changes are extremely difficult to determine without resort to more sophisticated survey and experimental techniques.
Sediment sampling over a range of different tidal states between 1978 and 1981 revealed that surface sediments were mostly mixed sand and gravel, with some patches of sand and silty clays. Vibrocore survey identified an upper mobile sand and gravel layer and a lower, immobile gravel-rich layer composed of Pleistocene fluvial terrace deposits. Side-scan sonar survey showed both deposits to be extensively scoured by dredging, frequently outside licensed areas. Removal of in situ deposits inevitably leads to lowering of the bank and may result in changed hydraulic conditions and alterations to sediment transport pathways. Tidal current measurements at a variety of depths indicated no significant velocity changes over a three-year period, but that direction of flow was critical to net transport (Dyer, 1971). It was concluded that not only are sediment transport pathways poorly understood in the vicinity of Solent Bank, but it remains uncertain whether lowering of the bank by dredging could have changed hydraulic conditions and thus altered sediment transport pathways in this part of the Solent. [Dredging ceased in 1994, but there have been no subsequent surveys that reveal if this has had a subsequent impact on hydrodynamic conditions and transport pathways].
(C) Fundamental Gravel Transport Studies by the Institute of Oceanographic Science
Preliminary surveys were undertaken in the West Solent using side-scan sonar, echo sounding, bed sampling and underwater TV. These surveys identified an area of bedforms developed in well sorted mobile gravel on the south margin of the main channel between Yarmouth and Hamstead Ledge (Langhorne, Heathershaw and Read, 1982). Further investigations employing more intensive coverage by similar techniques identified areas of large asymmetrical gravel waves (10-20m in wavelength), large and small symmetrical gravel waves between 1 and 2m in height and an area of nearly flat bed gravel. Gravel wave asymmetry indicated net north-eastwards or eastward sediment transport in the main channel. Detailed observations in the flat gravel bed zone found widespread mobility of all sediment sizes during spring tides when peak bottom current velocities exceeded 1.4m per second. Larger particles moved by sliding and rolling, whilst smaller particles were transported in microscale vortices which developed in the lee of larger particles (Langhorne, Heathershaw and Read, 1982). Long-term stability of these bedforms was uncertain because re-survey two years later was inaccurate, so individual gravel wave crests could not be relocated (Langhorne, Heathershaw and Read, 1982). Further work was undertaken relating sediment transport, observed by underwater TV (Williams, 1990; Williams and Tawn, 1991) to current velocity measured by an array of meters. Thresholds of movement were examined for different sediment sizes in a variety of tidal flows. A current velocity of 1.3m per second at 1m above the bed was found to be critical for gravel movement in the 20mm-60mm size range (Hammond, Heathershaw and Langhorne, 1984).
Much effort was devoted to acoustic detection of gravel transport. Acoustic energy generated by sediment transport was monitored by hydrophones and calibrated by comparison with observations of sediment transport by underwater TV and simultaneous tidal current measurements. Bursts of acoustic energy were found to be related to large instantaneous shear stresses and rapid transport events. These comparisons demonstrated the viability of this technique, which had the advantage of good temporal resolution of transport events, i.e. instantaneous monitoring (Thorne, Heathershaw and Troiano, 1984). The acoustic detection technique was then employed in a number of fundamental studies into bedload transport in unidirectional flow.
Simultaneous measurement of sediment transport using acoustic detection and current velocities revealed the existence of a "bursting" phenomenon created by currents flowing over gravel beds. Bursts of peak instantaneous shear stress over ten times greater than mean shear stress (measured over 20 minutes) were recorded in conjunction with surges of gravel bedload transport. It was concluded that this was achieved principally by sweep-like motions (dominant horizontal velocity component) in the bottom boundary layer (Heathershaw and Thorne, 1985). Detailed underwater TV recordings were analysed to produce quantitative estimates of sediment transport (Williams, 1990). A mean transport rate of 0.00045kg per meter per second was recorded for a 30 minute period (Thorne, 1986). Such data was of use for calibration of acoustic detection techniques but was of far too limited a spatial and temporal extent for a mass transport calculation for the Western Solent as a whole. The acoustic detection technique was used in conjunction with current measurements to compare the predictive capability of five bedload sediment transport equations. It was found that three of the equations gave results agreeing well with measured transport even though sediment size at the site exceeded the original calibration limits of the formulae (Williams, Thorne and Heathershaw, 1989a). Sediment transport was measured over a series of ebb and flood tides, and rates were calculated over time periods up to 147 minutes of up to 0.00032kg per meter per second. Results were again of too limited spatial and temporal representativeness to be applied to the West Solent as a whole but fundamental aspects of bedload transport were more clearly defined. It was concluded that bedload transport is associated with intermittent rushes ("sweeps") of high velocity fluid towards the bed (Williams, Thorne and Heathershaw, 1989b; Williams and Tawn, 1991).
It can be concluded that the observations of gravel waves confirmed earlier work by Dyer (1971) and indicated that the entire size-range of bed sediments were highly mobile. As this information was corroborated by a variety of survey techniques over several years, it must be regarded as highly reliable. Evidence indicated that net bedload transport was eastward and thus supported the major transport pathway recognised by Dyer (1971, 1972, 1980). Much of the detailed analysis of sediment transport was conducted over very limited timescales and restricted to a single site. The objective firstly was to calibrate the acoustic detection technique and secondly to examine fundamental bed stress: bedload transport relationships. A product of this work was the development of an acoustic measurement technique and calibration of bedload transport equations which could be more widely applied. It is only through the use of such techniques at representative spatial and temporal scales that the complex qualitative sediment transport pathways identified by Dyer (1971) can be verified and quantified. It would appear that net pathways of movement are different on either side of the main channel. Although transport is fundamentally due to tidal currents, both coarse sand and gravel can occasionally be entrained by waves of exceptional height generated by easterly fetch, and moved shoreward. These waves are the product of east or south-easterly winds blowing across the eastern English Channel, and then entering the Eastern Solent.
Aerial photography collected in 2013 through the Coastal Monitoring Programme and observations confirm the transport of suspended fine-grained sediments from Christchurch Bay, into the West Solent especially during the flood tide. Thus, it is likely that fine marine sediments and suspended clay sediments derived from cliff erosion become drawn into the West Solent. Remote sensing studies of suspended sediments within Christchurch Bay and the Western Solent also support these conclusions (Strisaenthong, 1982; McFarlane 1984). Please note that this was previously represented as F2 in the 2004 version but has been updated and categorised as a T0 arrow.
A general eastward transport of sand and gravel was recognised in the central and northern parts of the main channel (Dyer, 1971, 1972, 1980). This transport pathway extends eastwards from Hurst Spit and terminates at Solent Bank where much material has been removed by aggregate dredging since the late 1950s (Hydraulics Research, 1977, 1981). There is some evidence for channel deepening in the late 1980s (Ke and Collins, 1993). Prior to this, the transport pathway probably extended further east. It has been identified from the asymmetry of bedforms (Dyer, 1971) and its long-term existence was confirmed (Langhorne, Heathershaw and Read, 1982). Direct underwater TV observations of the sea bed indicated a very high degree of sediment mobility along this pathway (Langhorne, Heathershaw and Read, 1982; Williams, 1990). Transport was predominantly by bedload involving material ranging in size from coarse sand to cobbles. Suspended sediment transport was not observed, but might be inferred from the longer duration of the flood tide. With the suspension of aggregate dredging in 1994, it is probable that this pathway now extends further eastwards, and may feed transport moving away from Solent Bank.
Evidence of gravel wave asymmetry indicated continued eastward transport towards Egypt Point (Dyer, 1970a, 1971). Much eastward moving material must have been intercepted at Solent Bank by dredging, but studies by Webber (1977) and Hydraulics Research (1977, 1981) concluded that eastward transport continued beyond this location. Tidal currents become weaker east of Egypt Point and sedimentological evidence indicates that coarse sand is deposited on Prince Consort Shoal and medium sand on Brambles Bank (Dyer, 1971, 1980, Hydraulics Research, 1977). The fate of eastward moving gravel is less certain; deposition is possible as currents reduce towards Egypt Point, but no gravel banks have been recognised to indicate such a response. Alternatively, gravel may be transported westward back towards Solent Bank in a complex, perhaps unstable recirculating eddy (T9) along the southern margin of the main channel (Dyer, 1971, 1980; Halcrow, 2002).
Sand waves in the vicinity of Prince Consort Shoal indicate a node of temporary deposition where an anticlockwise movement of material in recirculation around Egypt Point (T2) meets a clockwise movement set up by a tidal current meander. These pathways were identified by Dyer (1971) and were corroborated by Webber (1977).
This pathway, only indicated indirectly by bedform asymmetry, reveals sand supply to Prince Consort Shoal from Osborne Bay and the East Solent (Dyer, 1971, 1980).
An anticlockwise eddy of sediment movement occurs on Bramble Bank, set up by the meandering of tidal currents (Dyer, 1971, 1980). This eddy supplies a westward moving sediment transport pathway which may feed Stanswood Bay or possibly link with the T7 pathway.
The dominant ebb tidal current has significant potential to sweep sediment out of Southampton Water (Dyer, 1970a, 1971, 1980; Sharples, 2000). Actual supply from this pathway is difficult to assess because there is relatively little information on coarse bedload sediment transport in Southampton Water. Gravel may be moved from the distal end of Calshot Spit and the adjoining gravel foreshore but this possibility has not been evaluated. The transport pathway probably joins with T5 in the western approaches to Southampton Water, and circulates around Brambles Bank (See units on East and Central Solent and Southampton Water).
Bedform asymmetry and divers' observations indicate net offshore movement and south-west transport along this pathway (Dyer, 1971, 1980). Sediments may derive from beaches in the vicinity of Stone Point (E3), and may supply Lepe Middle Bank. Sand patches on the seabed between Stansore Point and the proximal part of Calshot Spit, seawards of the foreshore step, are reported by Posford Duvivier (1999). Similar fine sand accumulations occur immediately south of the mouth of Beaulieu River (Codd, 1972) but there is no information on whether they are persistent features.
Although indicated by bedform asymmetry (Dyer, 1971), this pathway appears to contradict other evidence of rapid accretion of the nearby Warren Farm Spit due to onshore gravel transport (Human, 1961; Hydraulics Research, 1987; Clark and Gurnell, 1987). Transport direction of this pathway may therefore be variable and subject to local and intermittent reversals, possibly in response to incident waves of exceptional height generated by east or south-easterly winds.
Stationary meanders are located at different positions on the ebb and flood tidal streams, resulting in variations in strength and direction of flow. A major ebb current meander in Thorness Bay causes net westward transport and results in a recirculating eddy of sediment, which transports material to Solent Bank (Dyer, 1971). At this location the recirculation meets the dominant eastward transport pathway (T1) and sediment appears to be deposited. This recirculating pathway was identified from analysis of bedform asymmetry and tidal flow (Dyer, 1971).
A divergence of the dominant eastward transport pathway (T1) is indicated by analysis of bedforms (Dyer, 1971). This pathway appears to supply sediment to Newtown Gravel Bank, located south of Solent Bank. From here sediment may be entrained by the recirculating eddy (T9) and re-supplied to Solent Bank. Chart comparisons covering the period 1963 to 1973 indicated significant accretion of Newtown Gravel Bank (Hydraulics Research, 1977), thus possibly confirming this process.
Westward transport due to an ebb current meander is indicated by bedform asymmetry. Sediment sampling revealed a westward deflection of a "tongue" of fine material extending offshore from Bouldnor. This may represent sediment supplied by cliff erosion being entrained by the westward flowing recirculating eddy (Dyer, 1971). The pathway meets the dominant eastward flowing transport stream.
Hydraulics Research (1991) undertook a sediment-size analysis in the outer Lymington estuary, revealing gravel and sand in the main channel (in contrast to the silty-clay composition of the banks of the Lymington River). Net seaward transport of fine sediment was inferred and it was demonstrated that shear stresses imparted by waves on recently settled mud was sufficient to re-entrain that material during the low tide period (shallow water depths). Net offshore transport of fines from the outer estuary may also be indicated by some 0.5m increase in depth of the main navigation channel between 1981 and 1991 (ERM, 1998). This is not accounted for by capital dredging, which amounted to only 15,000m³ during this period, or by maintenance dredging off Town Quay. Hydraulics Research (1991) and Posford Duvivier (1994) provide the tentative conclusion that silt and mud brought into the estuary as suspended load is flushed seawards by the ebb tidal current, though a proportion of input is trapped in marinas and mooring areas. Suspended sediment input to the Beaulieu estuary is probably removed by ebb tide currents, as there are few opportunities here for settlement and storage. The T12 arrow sediment type has been changed from sand/gravel (2004 version) to silt/sand as the literature indicates that it is the fine material that is transported tidally in suspension, rather than the sand/gravel.
5. Sediment Outputs
5.1 Tidal (Estuarine) Outputs
Tidally-driven sediment outputs are possible at each end of the Western Solent. Studies of tidal currents and bedforms at its eastern limit have revealed that currents diminish towards Egypt Point, gravel ceases to be mobile and sands are deposited from eastward moving pathways onto Brambles Bank and Prince Consort Shoal. These locations are believed to represent sediment sinks (Dyer, 1971, 1980). It is therefore concluded that sediment is not directly transported eastward out of the West Solent. In fact, some sediment may be recirculated back from Brambles Bank by north-westward moving pathways (T5).
A series of three 10-20m high, steep angled, sub-marine terraces are located in 25-60m water depths between Fort Victoria and Hurst Point (Photo 11), associated with the formation of the Solent and the former position of the Solent River. These large-scale features restrict or inhibit bedload transport into the West Solent from Christchurch Bay, even under extreme hydrodynamic and tidal conditions. Examination of tidal curves for Lymington, Yarmouth (Isle of Wight) and Totland reveal marked asymmetry, because the ebb flow is concentrated into a shorter time period than the flood (Webber, 1980). The ebb flow is therefore considerably more rapid than the flood and transport of coarse bedload sediments (sand and gravel) is therefore likely to be in a net seaward direction, determined by peak current velocities. The nearshore bathymetry in the Hurst Narrows channel indicates a scoured bedrock bed.
Surficial seabed sediment within the West Solent is transported in separate ebb and flood pathways (Dyer, 1980) so these inputs are not immediately moved back offshore by subsequent ebb tidal currents.
Ebb currents are shorter-lived but more rapid at Hurst Narrows (Webber, 1980) so potential exists for both suspended and bedload transport westwards out of the Western Solent. It is widely recognised that material moving into Hurst Narrows from Hurst Spit undergoes net offshore transport towards the Shingles Bank in Christchurch Bay (Dyer, 1970b, 1971, 1972, 1980, and Nicholls and Webber, 1987). It can be postulated that a similar fate might apply to sediment reaching the main channel from the opposing, Isle of Wight, shoreline. All previous research suggests a net eastward transport pathway in this area (Dyer, 1980), but this does not fit with the known tidal regime that favours output controlled by dominant ebb tidal currents. The uncertainty relates to the location of the "switchover" from dominant ebb current flushing out of Hurst Narrows to the prevailing eastward transport pathway (T1) that characterises much of the West Solent channel to the east. It is therefore postulated that sediment output driven by tidal current is possible also from the Isle of Wight shore, and may be linked with sediment transported westward by a recirculating eddy (T11). This portion of the pathway remains speculative, and subject to verification.
It is probable that some of the suspended sediments circulating within the West Solent are deposited on the north-west shore in sheltered areas such as the lower Beaulieu (Photo 5) and Lymington River (Photo 3) estuary mouths, although pathways and rates have not been verified through site-specific studies. In these areas, major intertidal mudflats and saltmarshes have developed although substantial net losses are now occurring owing to both Spartina erosion and "die back" and the migration of the Lymington and Beaulieu River channels (Ke and Collins, 1993; 2002; Colenutt, 2002; Baily and Pearson, 2007; Cope et al., 2008). Analysis of fine sediments in the Beaulieu Estuary revealed that the majority were derived from marine rather than fluvial sources (Codd, 1972), via tidal currents (Posford Duvivier, 1994). Erosional scour of the inter-tidal shoreface creates suspended sediment input that may be deposited upon the marsh surface, but is part of the complex sediment budget of the mudflat/saltmarsh system (Ke and Collins, 1993). Quantitative volumetric estimates of eroded mudflat and saltmarsh material (Posford Duvivier, 1999; Colenutt, 1999, 2002) indicate a significant net export of fine-grained sediments from the estuaries. Evidence from capital and maintenance dredging records suggest only limited volumes are deposited within the estuary systems, either in marinas or directly on the marsh surface. There is also a probable, but currently unquantified, input of fine sediment as a consequence of ebb tide scour of dredged channel margins near the entrance to Southampton Water (and possibly updrift of Calshot Spit). The residence time of this material is unknown.
5.2 Dredging
Dredging has been practised in the Solent for the past century, but it is only since the early 1950s that large-scale aggregate extraction has been undertaken. In the late 1930s dredging within the whole Solent was at a rate of approximately 150,000m³ per year (Shears, 1986). From the 1950s, when dredging effort increased, figures are available relating specifically to Solent Bank. Mean rates of extraction by dredging were 81,000m³ per year for 1950-59, 218,000m³ per year (1960-62), 920,000m³ per year (1963-71) and reducing to 681,000m³ per year for the period 1972-75 (Hydraulics Research, 1977). Thereafter, dredging was further reduced because of uncertainty as to its effects on adjacent sub-cell budgets, finally being suspended in 1994. Dredging records also include extraction at Pot Bank and Prince Consort Shoal but there are no figures specific to the last site. (Pot Bank is located west of the Needles and outside the West Solent).
Steady reduction and final cessation of dredging from Solent Bank is mostly attributable to fears over effects on adjacent shorelines, but it is also possible that the bank was steadily becoming worked out. These concerns resulted in studies of Solent Bank by Webber (1977) and Hydraulics Research (1979, 1981). Chart comparisons covering the period 1847-1973 revealed that the volume of Solent Bank fluctuated periodically, but remained within consistent limits until 1960, when rapid reduction in volume was recorded (Hydraulics Research, 1977). Chart comparisons covering the period 1968-73 indicated lowering of the bank by 3.2m (0.64m per year) followed by accretion of net 0.4m between 1974 and 1976 (Webber, 1977). Admiralty hydrographic chart comparisons covering the period 1965-1979 revealed surface lowering at 0.13m per year. A series of seven hydrographic surveys over the period 1978-81 revealed lowering at 0.20m per year. This suggests diminution of the rate of replenishment, as dredged output was reduced significantly over this period (Hydraulics Research, 1981). The general trend was for significant reduction in the level of the bank following intensification of dredging in the late 1950s, despite some significant natural replenishment. In 1950, Solent Bank was a shoal with a marked plateau within the -13mOD contour. Surveys between 1978 and 1981 showed that only a fragment of this former crest remained, with a mean overall summit/crest depth of -16.5mOD. Side-scan sonar and sediment sampling surveys demonstrated that much of a previous core of Pleistocene terrace deposits had been removed by dredging so the contemporary bank was composed of a relatively thin veneer of mobile sand and gravel overlying in situ Eocene strata (Hydraulics Research, 1981). This suggests that it has been permanently lowered because it is only a temporary resting place for eastward moving and recirculating sediment (Dyer, 1971) (see Section 4b). Replenishment is therefore unlikely to rebuild the bank to previous levels as sand and gravel are highly mobile over the whole bank and losses by throughput must be assumed to be constant.
The transport and sediment budget implications of this well documented volume loss are difficult to determine, but two major possibilities exist: (i) interception of sediment transport pathways and consequent reduction of supply to areas further down the transport pathway, and (ii) alteration of hydraulic conditions over the bank and associated changes to the pattern of sediment transport in its vicinity. These changes are of most immediate concern to the eroding northwest coast of the Isle of Wight. Examination of beach profiles and OS maps indicated that the greatest shoreline changes coinciding with the commencement of intensive dredging were either side of Newtown Harbour entrance. It was concluded that the contribution of Solent Bank dredging to these shoreline changes could not be discounted and it was this consideration that resulted in the withdrawal of dredging licences in 1994. Subsequent bathymetric and sedimentological changes have not been monitored.
Dredging for navigation access and berth maintenance is undertaken at Lymington River. A study specific to the proposed dredging of Horn Reach (ERM, 1998), which reviewed the local hydrodynamic regime, revealed that the main navigation channel remained constant in width, but deepened by 0.5m, 1981-1992. Much of this erosion was presumed to be natural, though capital dredging removed the modest quantity of 15,000m³ during this period. Net accretion at Horn Reach, close to the causeway, took place between 1992 and 1997. Dredging of the area adjacent to the Town Quay, as well as the marina basins, was not less than about 60,000m³ during the period 1985 to 1995 (ERM, 1998). Colenutt (2001) quotes a mean of 40,000m³ per year for the Lymington estuary as a whole attributing siltation to material originating from the eroding mudflats and saltmarshes of the area. Historically, maintenance dredged sediments from the estuary have been disposed of during an ebb tide at a dumping site seaward of Hurst Narrows. It has been suggested that alternative placement sites on mudflats within the estuary could assist mudflat conservation although further appraisals of the feasibility of this option would be required (Colenutt, 2001)
There are no available records of dredging of the Beaulieu and Keyhaven channels.
5.3 Reclamation
Much of the low-lying belt of grazing land between Keyhaven and Lymington (Photo 8) was reclaimed in the 18th (used as salterns) and 19th centuries (conversion to grazing meadows) totalling around 260ha. This frontage has been protected by several generations of embankments and seawalls. The corner of land forming some 290 hectares on the western margin of the Beaulieu River mouth around Warren Farm and Needs Ore Point (Photo 5) was reclaimed in the 15th Century and is protected by earth embankments. Prior to this latter reclamation, the plan form of the Beaulieu outer estuary would have been similar to that of the Lymington River.
6. Sediment Sinks and Stores: Mudflats and Saltmarshes
In the West Solent, rapid growth in area and elevation of creek dissected mudflat and saltmarsh occurred between the 1880s and late 1920s or early 1930s. Tubbs (1980, 1999) has recorded in detail the invasion and spread of the hybrid cord grass, Spartina anglica, the prime factor responsible for the trapping and stabilisation of a very substantial volume of fine silt and clay. He suggests that the peak of Spartina colonisation was between 1925 and 1929. Colenutt (2001) provides an estimate of 734 ha for Keyhaven to Pitts Deep in 1921 based on analysis of hydrographic charts, reducing to 297 has in 1994 derived from aerial photography analysis. The reproductive vigour and rapid vegetative spread of Spartina anglica has been described and analysed by various authorities (e.g. Gray and Benham, 1990; Gray et al., 1991; Raybould et al., 2000). Tubbs (1980) has suggested that the creation of large areas of vacant mudflat may have commenced in the late eighteenth century, as a result of an unexplained loss of species diversity and/or reduction of colonising ability by the native Spartina maritima.
Since the late 1920s, there has been a steady, and cumulatively substantial, decline in the area of Spartina anglica-dominated saltmarsh fronting the shoreline of the north-west Solent (SCOPAC, 2011) (Photo 9 and (Photo 10). Oranjewoud (1990) calculate an average retreat of the leading edge of both marshes and mudflats of 4m per year, 1867-1968, with acceleration apparent from the early 1950s and Colenutt (2002) calculated mean recession of 3m per year for the period 1781-1994. LRDC International Ltd (1993) in a study of the saltmarsh either side of the Lymington River, seawards of the town marina, calculate that there has been close to 300m of retreat of the mudflat edge since approximately 1940 (6m per year). However, this research, which was based on detailed analyses of Ordnance Survey maps and aerial photography, indicated a slower recession rate, approximately 1m per year, for the upper saltmarsh since 1950. In the Pennington Marshes, to the West, some 640m of onshore movement of MLW has occurred since 1870. Spatial and temporal variation of the rate of loss is thus apparent from these studies; not less than 35% of the original extent of Spartina anglica colonised marsh had disappeared by 1980.
There is convincing evidence that erosion and recession rates have accelerated since the late 1970s. Bradbury (1995), for example, has noted rates as high as 8m per year for two monitored sites at the mouth of the Lymington River, with a mean rate of 3m per year (1992-1994) for this area as a whole. Baily and Pearson (2007) calculate 165m of retreat of the front edge of the saltmarsh in the outer Lymington estuary, 1971- 2001. Halcrow (1998) report that MLW retreated at 7m per year at Pylewell Point east of the Lymington River, since 1975, an acceleration by almost 2.6m per year compared to the period 1870 to 1975. However, shoreward movement of the MLW mark of Keyhaven Marshes appears to have been significantly slower, although 300m of recession occurred here in the century between 1867 and 1968 and 161m between 1971 and 2001 (Baily and Pearson, 2007). Accretion has occurred along the margins of, and within, Keyhaven Lake, which necessitates occasional dredging of this minor navigation channel. Recent net accretion along a part of the east bank of the Lymington River has also been recorded (ERM, 1998; Black and Veatch, 2012b). Spatial variations in the patterns of past and contemporary erosion and accretion reflect complex interactions between antecedent shoreface width, tidal current velocities and the supply of suspended sediments, as well as the locally very variable patterns of change of saltmarsh and mudflat ecosystem structure. An example of the latter is the loss of Zostera (eelgrass) from areas not taken over by Spartina, as a result of disease, in the 1930s and in the 1990s. In some areas, particularly the mouth of Beaulieu River, the area of Spartina anglica today may be partly controlled by artificial modification of the hydraulic regime, as well as in response to deliberate small scale planting (Lobeck, 1995).
Subsequent to the research summarised in the preceding paragraphs, analysis of the temporal and spatial scales of saltmarsh decline, retreat and- in some areas- virtual extinction has been completed by Bray and Cottle (2003), the Solent Dynamic Coast Project (Cope et al., 2008), Baily and Pearson (2002; 2007) and Bray (2010). These authors used interpretation of rectified and digitised panchromatic and near infra-red sequential aerial photography, supported by large scale maps, for various dates between 1946 and 2008. Despite some acknowledged sources of inaccuracy and marginal error, this new work provides a detailed and reliable inventory of the loss of intertidal saltmarsh, and its replacement by mudflats, along the north-west shoreline of the Solent.
All researchers are agreed that a primary cause of loss has been edge erosion, accompanied by lateral erosion (widening) of the network of dissecting creeks, leading to fragmentation. An exception occurs in the case of the saltmarsh behind Hurst spit, where rollback has been the main cause of loss (35% between 1971 and 2001, with much of this occurring as a result of extreme storm conditions in 1989 and 1991).
Rates and patterns of saltmarsh contraction are spatially variable, but increase progressively eastwards of Hurst spit and Keyhaven due to greater exposure to wave energy in that direction. Thus, at Keyhaven Cope et al. (2008) calculate a 50% loss (1.7% per year), 1971 to 2001. Baily and Pearson (2007) conclude a figure of 46.5% for the same period, and Bray (2010) proposes 46% (for 1971 to 2007). At Lymington, a 63% loss of area- almost all of it due to edge retreat in the outer estuary-occurred between 1946 and 2001; the rate accelerated from 0.8% per year, 1946-1954 to 2.3% per year in the period 1984 -2001 (Cope et al., 2008). The most recent published survey data for the Lymington estuary confirms ongoing erosional loss between 2008 and 2012 (Black and Veatch, 2012, a, b and c). At Pennington, to the south-west of the Lymington river estuary, one area fronting sea defences recorded a 98% loss, i.e. virtual extinction, between 1971 and 2007, with the rate of loss steadily increasing throughout this period (Bray, 2010). 16.3% of the initial area disappeared between 2001 and 2007. Cope et al. (2008) remark on a comparable loss for the Pitt’s Deep/Sowley intertidal marsh east of Lymington, where 38.7ha in 1946 had reduced to just 6.7ha in 2001 (an 83% loss, at an accelerating rate during this fifty five year period). Losses in the Beaulieu estuary have been comparatively less, with some uncertainty over the scale and rate. Baily and Pearson (2007) calculate a reduction of 73% of initial saltmarsh habitat between 1971 and 2001 (2.43% per year), whilst Cope et al. (2008) quantify the loss, between 1954 and 2001, to have been 53% (1.1% per year). Here, as at all locations along the northern coastline of the western Solent, the rate of Spartina mortality, and the consequent conversion of former saltmarsh to lower elevation mudflats, has steadily increased over the most recent six decades; in several locations it has apparently accelerated to peak rates since the mid-1990s.
It has been observed elsewhere in the Solent region that saltmarsh decline is often associated with mudflat accretion (refer to unit on Southampton Water, for example). Black and Veatch (2010a, b) using graduated reference stakes and repeated visual inspections note an average vertical accretion of 3cm between September 2010 and August 2012 both outside and leeward of a breakwater in the Lymington estuary.
A detailed analysis of the sediment composition and dynamics of the saltmarsh and inter-tidal mudflats of the north-west Solent is given by Ke and Collins (1993; 2002) and Ke (1995). These studies recorded that inner saltmarshes are elevated above adjacent mudflats, with weakly concave-upward transverse slopes of between 0.2 and 6 degrees. Extensively dissected by creek systems into a virtual pattern of saltmarsh islands, many are abruptly terminated by a bluff or "clifflet" between 0.7 and 1.6m in height fronted by a narrow abrasion platform. Mudflats are characteristically narrower than saltmarshes (the former with a maximum width of approximately 200m), and have higher transverse gradients. Bluff retreat involves the further subdivision of inter-creek blocks, including the temporary creation of small stacks (Photo 10). Low relief chenier ridges, composed of a mixture of sand, gravel and shell debris, occur at the edges of several saltmarshes, some at the top of and landwards of clifflets as a consequence of occasional overtopping by higher energy waves. (Photo 3). They are characteristically up to 0.8m in height and 10-20m in width, and are considered to mark a transition from tidally-dominated deposition on mudflats to wave erosion of the saltmarsh edge. Many chéniers rest on erosional surfaces cut into mudflat sediments. They may provide an element of protection (armouring) of the otherwise low erosional resistance of the seaward margins of saltmarsh. Where there is evidence of chernier steepening, an increase in wave erosion (or abrasion) may, perhaps, be inferred (Bradbury, 1995).
The presence of significant quantities of coarse sand, gravel and, particularly, shelly debris indicates wave transport of sediment from nearshore and offshore sources. Much of the sediment composing the saltmarsh is silty clay, whereas the modal composition of the mudflats is sandy silts and mixed sand, silt and clay. This coarser texture is considered to be due to the mixing of sediment transported by waves and tidal currents from the sub-tidal zone with that removed by landward marsh erosion. At some 2-3m below saltmarsh surfaces, in situ gravel-sized clasts of wood are encountered. These are presumed (Ke and Collins, 1993) to derive from abrasion of offshore (now submerged) peat, probably late Holocene in age. The coarsest silts and sands occur in front of actively retreating clifflets, the result of the selective removal of fines in suspended transport. Sediment texture fines gradually from the main river mouths to inner estuaries, but sand, gravel and shell debris occurs in both the main and many tributary creek channels (Hydraulics Research, 1991). This material is at least partially derived from wave and tidal scour, revealing and releasing underlying substrate sediments.
Ke and Collins (1993) undertook detailed analysis of changes in the position of MHW, and the extent of both saltmarshes and mudflats, using successive editions of Ordnance Survey maps back to the early twentieth century. However, they focused on site specific changes since the early 1950s, noting that there has been a landward movement of the 0m isobath of 3m per year, 1959-1979, but accelerating to as much as 30m per year since 1980 at a few critical locations. Between 4 and 5m per year of retreat of the outer saltmarsh edge bounding the Lymington River edge occurred between 1959 and 1990; this, however, is only a mean rate, with annual losses of up to 12m having been recorded. Between Oxey and Pennington Lakes, saltmarsh was virtually eliminated between 1900 and approximately 1970, and was replaced by a fringing gravel beach. Lymington Spit was also removed during this period, but rates of edge retreat over Keyhaven Marshes were lower than further east, at approximately 3.5m per year. The lowest rates were in the immediate lee of Hurst Spit, at 0.8-1.1m per year, 1930-1980. These figures are generally consistent with those calculated by other research work, summarised in preceding paragraphs. However, it was apparent that there had been expansion landwards of the rear boundary of inner saltmarsh in some areas, thus partly compensating for losses at the eroding seaward front. Transverse profiles have also been more or less constant, suggesting that they have accreted vertically whilst retreat has been ongoing. Although more difficult to measure, using mean LWM, Ke and Collins (1993) were able to demonstrate retreat of the seaward margin of sub-tidal mudflats, at an average rate of 4.3m per year (1974-199) for much of the area between Pennington and Pitts Deep. Overall, this evidence of erosion and retreat suggests that the present day inter-tidal morphology is not adjusted to prevailing coastal hydrodynamics.
Analysis of Coastal Monitoring Programme lidar data demonstrated no net vertical accretion of saltmarsh, contrary to Ke and Collins (1993) conclusions, inferred from sediment sample concentrations. These were in the order of 45-60ppm, with values on spring tides higher than on neaps. They are highest at the edge of saltmarshes, where wave abrasion would be active. This is confirmed by significantly higher suspended sediment concentrations in winter, when waves of over 1.6m in height can penetrate into the western Solent. The greater availability of suspended sediment for potential accretion at and adjacent to marsh edges may be a partial explanation as to why erosional losses here in areas of Spartina "dieback" are no faster than they are in adjacent areas where vegetation cover is unaffected (given similar exposure to tidal current velocities and breaking wave heights).
Ke and Collins (1993) suggest several reasons why there has been substantial, and apparently accelerating, lateral erosion of saltmarsh and mudflats in recent decades. These include: (i) More exposure to winds and waves from the east and north-east; although the potential fetch is limited, waves from these directions can move unobstructed, as demonstrated by the orientation of the widest abrasion platforms; (ii) An increased frequency of storm surge events and significant wave heights since the early 1950s; (iii) Mean sea-level rise; (iv) Progressive northerly shoreward migration of meanders in the main West Solent Channel; and (v) Several independent, but cumulative, human impacts. These include shoreline protection, promoting "coastal squeeze"; ship-generated waves; marina/berth construction; channel dredging; changes in land use management in the New Forest catchments affecting delivery of fluvial sediment input, and the growth and decline of the salt working industry since the early nineteenth century.
Using data on marsh and mudflat recession and landward migration; inferred vertical marsh accretion, and changes in creek morphology, Ke and Collins (1993; 2002) have attempted to calculate a budget for fine sediment for Keyhaven-Lymington saltmarshes. Total eroded volume (including losses from dredging) is calculated at 15.4 x 104m³ per year. Saltmarsh surface accretion, representing approximately 30% of the loss from edge erosion, plus deposition at creek and channel boundaries, amounts to 3.2 x 104m³ per year. Thus, it is apparent that the overall budget is strongly negative, and represents a net current loss of around 12 x 104m³ per year. Sediment yield from inter and sub-tidal mudflat erosion averages as 11.6 x 104m³ per year, of which an estimated 70% is lost entirely as suspended sediment input into the West Solent. The remaining 30% is thought to contribute to channel and marsh accretion. Further information may be found at http://www.scopac.org.uk/sediment-sinks.html.
7. Summary
- The Western Solent tidal channel is characterised by former marine inundation and erosion of the Solent River valley formed in soft Tertiary and Pleistocene materials. Its shoreline is soft, low-lying and inherently sensitive to inundation and erosion, even when the eroding forces are relatively weak.
- It has experienced mixed behavioural trends since its inundation in the mid- to late-Holocene. Initially, it was depositional, but this altered swiftly to erosion following the breakthrough of the land/beach connection at Hurst Narrows. This opened up the Western Solent to tidal flows occurring between the East Solent and Christchurch Bay and established its present form of hydraulic regime. Since that time, the channel bed and eastern mainland shoreline east of Lymington and the north Isle of Wight shoreline have eroded releasing significant quantities of sediment, however, western parts have accreted and prograded significantly in the lee of Hurst Spit.
- Suspended sediments enter Hurst Narrows from Christchurch Bay. Eroding cliffs along the northern Isle of Wight shore contribute significant quantities of mainly fine fresh sediments to the Western Solent, but it is uncertain how much crosses the channel to contribute to local saltmarshes as opposed to becoming transported north-eastwards by the residual tidal flow. Several large re-circulating tidal eddies have been identified in the channel that could assist in the wider distribution of fine sediments in suspension.
- Since the 1930s a portion of the large sediment store of fine-grained sediment currently "locked up" in the Lymington, Keyhaven and Beaulieu saltmarshes and mudflats has been relatively rapidly eroded and transferred to suspended load. Some sediments are re-deposited onto marsh and mudflat surfaces, (where they have been subject to recycling) but the majority become distributed throughout the Solent system, or lost to the English Channel.
- It is probable that the channel of the Western Solent has yet to achieve an equilibrium adjustment of form to the imposed hydraulic conditions. The variable historic trends recorded along these shorelines, including the erosion of its mainland shore mudflats and saltmarshes, may therefore be products of this ongoing adjustment.
- Shoreline drift of sands and gravels is generally from west to east, although discontinuities occur at the Beaulieu River and also at Stansore Point where a weak drift reversal transports sediments westwards along the north east bank of the outer Beaulieu estuary. Drift is not an effective process along the mudflat-dominated Hurst to Pitts Deep frontage. Discrete shoreline units can thus be identified from: (i) Hurst Spit to Pitts Deep; (ii) Pitts Deep to Beaulieu River; (ii) Beaulieu River to Stansore Point and (iii) Stansore Point to Calshot Spit.
- Between Hurst Spit and Pitts Deep, a low energy fine sediment accretional environment is developed, although mixed accretion/erosion trends have occurred over the past 150 years with rapid erosion in recent decades.
- Between Pits Deep and the Beaulieu River the low-lying gravel fringed shoreline is transitional between accretion and erosion, but would appear to have altered from former accretion dominance towards becoming erosional and transgressive.
- Moderate erosion, cliff retreat and beach development occur to the east of the Beaulieu River, with a north-eastward trending shore drift pathway delivering small quantities of gravel towards Calshot Spit.
- The model outlined above is based on the assumption that the stability of Hurst Spit is maintained into the future. Major shoreline change and severe erosion of mudflats, saltmarshes and shorelines in its lee would be accelerated further should Hurst Spit suffer breakdown or permanent breaching.
8. 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. Although at present a 12 year time series of data has been collected, 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, although the lack of significant wave energy, modest development of natural linear beaches and prevalence of groynes means that shorelines of this frontage are not well suited for definitive studies of drift.
9. Research and Monitoring Requirements
Analysis of the data between 2006 and 2012/13, combined with other datasets, academic research and historical studies has enabled sediment budgets, transport rates and directions to be identified or verified. Littoral drift and cross-shore exchange rates and volumes have been calculated using details of cliff, and shoreface erosion inputs and beach volume changes. However, at certain sites either due to a lack of long-term data, data coverage or sedimentological information (e.g. composition and proportion of beach grade material arising from cliff erosion), quantification of sediment transport rates of gravel and sand has not been possible.
Notwithstanding results from the Southeast Regional Coastal Monitoring Programme, and the summarised information and assessments from the North Solent SMP2 (New Forest District Council, 2010), recommendations for future research and monitoring that might be required to inform management include:
- To understand beach profile changes it is important to have knowledge of the beach sedimentology (grain size and sorting). Sediment size and sorting can alter significantly along this frontage due to natural processes and beach management. 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. Examination of longshore and onshore-offshore grading of the various sediment parameters can be employed to indicate or confirm directions of transport, sources of sediment and possible residence (storage) timescales.
- Understanding of inputs of beach forming sediment from coast erosion would be enhanced by cliff section mapping and sampling of deposits to reveal the detailed thickness, composition and variability of Plateau and Valley Gravels. This material is of particular importance as it is a major local source of beach material.
- Further work to create a tidal model of the Western Solent would be beneficial, including work on quantification of bedload and suspended sediment transport volumes. This could in turn clarify the unquantified interaction of the Western Solent with Christchurch Bay which requires further work.
- Further work to establish the significance of the stability of Hurst Spit for the future protection of the Western Solent is recommended. If the protection of Hurst Spit was lost due to breakdown or permanent breaching, it would be of interest to understand/quantify the impact of increased wave penetration into The Solent, increased wave height and exposure, bedload and suspended sediment transport, flood risk, saltmarsh erosion and impact to the Keyhaven to Lymington sea wall.
- Further use of Coastal Monitoring data to further quantify sediment transport rates both above and below MHWS. Of interest is the interaction between Christchurch Bay and the West Solent, and the quantification of inputs, stores and sinks of sediment.
