About the Study

SCOPAC Committee

Chairperson Councillor Mrs M Penfold MBE, West Dorset District Council.

Vice-Chair Councillor Jackie Branson, Havant Borough Council.

Technical assistance provided to Councillors by Mr Lyall Cairns (Southern Coastal Group Chair) and Dr Samantha Cope (SCOPAC Research Chair).

The STS 2012 update

The 2012 update of the SCOPAC Sediment Transport Study (STS) was funded by the Environment Agency under FDGiA, grant number LDW 41230, with additional contributions from SCOPAC.  

It is referenced as: New Forest District Council (2017). 2012 Update of Carter, D., Bray, M., & Hooke, J., 2004 SCOPAC Sediment Transport Study, www.scopac.org.uk/sts.

Sediment Transport Study 2012

1. Introduction

The sequence of three almost landlocked harbours of Portsmouth (Photo 1), Langstone and Chichester are shallow tidal basins, which were created after sea levels approached present-day levels approximately 5-5,500 years BP. This marine transgression flooded a sequence of low relief valleys which drained the coastal plain and subsequently contributed up to 3.5m of sediments that largely conceal the late Pleistocene and Holocene sequences which overlay Eocene sands and clays (Sorour, 1999; Allen and Gardiner, 2000; Allen, 2000a, 2000b). The extensive mudflats and saltmarshes (Photo 2, Photo 3, Photo 4) indicate that sedimentation of fine-grained material has kept pace with the inundation of the harbours over recent millennia. This has been confirmed by research on recent historical rates (Cundy and Croudace, 1996). The detailed history of late Holocene sea-level fluctuation has not been fully researched, although a conceptual framework has been established for Langstone Harbour (Allen and Gardiner, 2000).

Compared to several other sites in the Solent system, the inundation of Langstone Harbour occurred comparatively late in the Holocene period. Deeply incised river channels occupied the main Langstone and Broom, Channels up until approximately 7-8,000 years BP, as evidenced by buried terrestrial peats at -10.7 to -11.9mOD (Mottershead, 1976). Thereafter, these channels were infilled and the present day Langstone basin was a wide alluvial valley, with limited and probably diminishing relief, up until at least the mid to late Iron Age (c. 2,500 BP). The stratigraphy of organic and mineralogic sediments suggests that sea-level rise did not affect the conversion of this floodplain into a shallow tidal basin until late Iron Age times or the first century AD.

The four main harbour islands are remnants of low interfluve divides, located on a locally slightly more resistant clay substrate. Archaeological evidence for the navigation of the harbour and the construction of quays in early/mid-Romano-British times (approximately 1,800 years BP) is well documented. By the sixth or seventh century AD, sea-level was between 2.7 and 3m higher than a millennium earlier. This history is suggestive of the original presence of a barrier structure across the present entrance to Langstone Harbour, which was finally breached as a result of the combination of storm surge events and ongoing sea level rise. This hypothesis is consistent with the strongly recurved spits that provide the present day definition of the harbour mouth. The availability of more stratigraphical detail from borehole logs in the southern part of Langstone Harbour, together with radiocarbon dates of organic horizons, would help to elucidate the precise sequence of events. Whether the idea of barrier emplacement and a comparable sequence of environmental changes can be applied with any confidence to Chichester and Portsmouth Harbours remains a challenging issue for future research. It may be feasible to conceive of an ancestral late Holocene "superbarrier" between Gilkicker (Gosport) and Selsey Bill, an idea implied by Wallace (1988, 1990), that subsequently segmented when it intercepted the sediment accumulations of the ebb tide deltas.

The estuary entrances are very narrow (Photo 1 and Photo 5), so that waves generated outside cannot penetrate far into the harbours. Internal sediment transport and sedimentation processes therefore differ markedly from that at the open coast because tidal currents are the major mechanism and wave action is relatively insignificant due to limited fetch. The intensity of transportation motive forces shows well-defined spatial variation, and sediments deposited range from gravels and sands at each harbour entrance (high energy conditions) to fine silts and clays at each estuary head, and adjacent to tidal channels (very low energy conditions). The presence of large ebb tidal deltas seawards of the mouths of each of the three harbours indicates the seawards flushing of coarse to medium grain sizes since approximately 5,500 years BP.

Because of the differing range of sediments and transport conditions the harbours require a different analysis compared to the open coast. A budgetary approach is therefore adopted, involving the identification of sediment inputs, stores, transfers or circulations of sediments and outputs. This approach is appropriate because each of the harbours behave as self-contained units (sediment sinks) despite being connected by shallow channels at their northern limits. The tidal pass at the entrance to Portsmouth Harbour (Photo 1) constitutes a stable fixed boundary between transport cells to the east and west. Littoral sediment movement at each harbour entrance converges from both directions, with only limited bypassing in the cases of Chichester and Langstone Harbours and apparently none across the mouth of Portsmouth Harbour.

Environmental baseline data which supports the determination of sediment transport, has been compiled by various independent studies (e.g. Collier, et al., 1995; Dunn, 1972; Budd, 1985; Fontana, et al., 2000; Baily, et al., 2000; Bray and Cottle, 2003; Baily and Pearson, 2007; Cope, et al., 2008) using remote sensing and other techniques. A detailed classification of vegetation types in Portsmouth Harbour (Baily, et al., 2000), and quantification of changes in the extent of saltmarsh in all three harbours (Baily and Pearson, 2007; Cope, et al., 2008; Bray, 2010) has been undertaken. In addition, Baily et al. (2000) mapped areas of bare (unvegetated) mud, sand and gravel derived from large scale, high quality false colour infra-red aerial photography e.g. Photo 6. Further details of the extent of widespread saltmarsh losses in each harbour are provided in section 5.4.

A major new source of coastal data is from the Defra-funded National Network of Regional Coastal Monitoring Programmes. The Programmes consist of topographic beach surveys, nearshore bathymetry, aerial photography, lidar, coastal hydrodynamics (waves and tides) and terrestrial habitat mapping.  Specifications for data collection are consistent for all regional programmes and the data and analysis reports are made freely available under the Open Government Licence from www.channelcoast.org. 

The Southeast Regional Coastal Monitoring Programme commenced in 2002. The Lead Authority is New Forest District Council, with data collection, analysis and reporting led by specialist teams at the Channel Coastal Observatory (CCO), Canterbury City Council and Adur and Worthing Councils. (See CCO Annual Survey Reports for further details).

The Southeast Regional Coastal Monitoring Programme measures nearshore waves using a network of Datawell Directional Waverider buoys. The nearest buoy to the harbours is the buoy deployed off Hayling Island in 10mCD water depth. Between 2003 and 2012 the prevailing wave direction was south-by-west, with an average 10% significant wave height exceedance of 1.26m (CCO, 2012).

2. Sediment Inputs

2.1 Marine Sources

» F7 · F8 · F9  

Sediment is transported into the harbours from the Eastern Solent. The narrow harbour entrances restrict the entry of waves, but increases tidal current, which are the most important transport mechanisms. Littoral transport is convergent at the mouths of all three harbours, with Portsmouth Harbour entrance functioning as the western cell boundary (Bray, et al., 1995). The tidal regime in each habour is complex, characterised by a double high water (young flood stand preceding slack high water stage) (Halcrow, 2010a and b). Analyses of tidal flow at the harbour entrances by Hydraulics Research Ltd (1959), Portsmouth Polytechnic (1976), Harlow (1980), Wallace (1988) and Halcrow (2002; 2010 a,b,c) indicate that the mean ebb current is of shorter duration but significantly greater velocity than the flood current. There is an elevation difference of 4.1m between Mean High Water Springs and Mean Low Water Neaps at the entrance to Portsmouth Harbour, 4.2m and 4.3m at the mouths of Langstone and Chichester Harbours respectively.  At the entrance of Chichester Harbour, maximum spring tide flood velocities are 1.4m per second, and maximum ebb velocities are 3.3m per second. The corresponding velocities for the other harbours are lower, at 0.9 and 1.8m per second for Langstone and 0.95 and 2.04m per second for Portsmouth Harbours. In the shallow water of the upper harbour creeks, tidal velocities do not exceed 0.4m per second. Thus, net bedload transport of coarse sands and gravels is dominantly seaward at each harbour entrance, forming extensive part-submerged coarse sand and gravel delta deposits (e.g. Spit Sand and Hamilton Bank) immediately offshore. Transport rates, however, are probably small (Gao and Collins, 1994). Input of coarse sediment is therefore only possible when spring flood tides coincide with southerly storms, so that a combination of tidal and wave driven currents transport this size range of material too far into the harbours for their re-entrainment by the ebb tidal stream (Hydraulics Research Ltd, 1984). Although this process has not been studied in detail, it may explain the flood banks of relatively coarse material, which exist immediately up-estuary from the entrance channels of all three Harbours. It may represent a periodic "pulsed" input, with now limited potential for progressive accretion. The accumulation of coarse to medium grain sandy sediments on Sword and Sinah Sands in Langstone Harbour, Ballast Bank in Portsmouth Harbour and Stocker's Sands in Chichester Harbour may exemplify this process.

The flood tidal stream is of lower velocity but longer duration than the ebb tidal stream (a mean of 0.7m per second for the Langstone entrance channel and 0.95m per second for Chichester Harbour mouth), thus net transport of suspended sediments (fine sands, silts, clays and organic particles) is into the harbours (Harlow, 1980; Wallace, 1988; HR Wallingford, 1997; Halcrow, 2010a and b). This, however, has not been substantiated, and there is some contrary evidence (HR Wallingford, 1994). These sediments can reach the innermost parts of each estuary, whereupon deposition is assisted by a tidal stand at high water and trapping by intertidal flora. All three harbours may be regarded as virtual sediment sinks (HR Wallingford, 1997), where the rate of sedimentation has kept pace with post mid- to late-Holocene sea-level rise. This applies most obviously to Portsmouth Harbour, where the seaward output of sediment appears to be negligible, and relative sea-level rise may be marginally faster (Webber and Walden, 1981; Bray, et. al., 1994).

F7 Suspended Sediment Input at Chichester Harbour Entrance (see introduction to Marine Sources)

Harlow (1980) suggested that the fine grained sediments of Chichester Harbour comprise material eroded from the open coast between Hayling and Selsey and placed in suspension by wave action. The harbour perimeter has not significantly changed shape in recent centuries, a feature taken to indicate that sedimentation has kept pace with sea-level rise (approximately 2-3mm per year). The contemporary rate of deposition was calculated from the area of the harbour and the above rate of relative sea-level rise, to give a figure of 40,000m³ per year sediment supplied by marine input. This calculation is of relatively low reliability because it is based on untested assumptions. Input from the episodic erosion of East Head and Black Point spits and associated nearshore sand banks provides an auxiliary, but unquantified, source (ABP Research and Consultancy Ltd, 2000, 2001). The only direct evidence collected to demonstrate the net budgetary significance of the input of suspended sediment through the harbour entrance suggests it may be negative (HR Wallingford, 1994). Estimations of future relative sea-level rise of 0.6cm to 1.0cm per year between 2000 and 2100 (Bray et al., 1994) will have an impact on this "steady state" budget, but it is impossible to make firm quantitative estimates (HR Wallingford, 1995, 1999; Halcrow, 2010a and b.)

Analysis of available data and literature provided no evidence for gravel input by weed rafting and the speculative 2004 arrows have therefore, been removed.

F8 Suspended Sediment Input at Langstone Harbour Entrance (see introduction to Marine Sources)

The contrast between the velocities of ebb and flood tidal streams at Langstone entrance is such that net transport of suspended sediment is into the harbour, Harlow (1980); Wallace (1988); Humby and Dunn (1975). The flood duration is 7.1 hours, that of the ebb 5.7 hours (Gao, 1993). This information has been supplemented by results from a radioactive tracer survey conducted on behalf of Portsmouth Polytechnic (1976). This study monitored the dispersion of sewage effluent (suspended solids) from the Fort Cumberland (outside Langstone Harbour) and Budds Farm (inside the harbour) outfalls during both spring and neap tides. It clearly demonstrates that up to 60% of the effluent released at Fort Cumberland can be transported into the harbour by the first flood tide. Some of the tracer was transported eastward during a spring tide and entered Chichester Harbour, whilst effluent released at Budds Farm tended to be retained within the estuary. It is therefore suggested that suspended sediments are readily transported into Langstone Harbour by the flood tidal streams, and once inside tend to accumulate there. Minerals associated with offshore deposits have been identified in suspended sediments sampled from Langstone, and also Portsmouth, Harbours (Portsmouth Polytechnic, 1976; Algan, 1994). This combination of direct and indirect evidence provides strong qualitative information of net suspended sediment input. However, the process has yet to be quantified to establish its significance. Gao (1993) calculates a potential transport capacity of 22x109kg per year at the Langstone entrance channel, but this figure refers to the total flux associated with both ebb and flood currents.

F9 Suspended Sediment Input at Portsmouth Harbour Entrance (see introduction to Marine Sources)

The relative velocities and durations of the ebb and flood tidal streams (Hydraulics Research Ltd, 1959) differ in a similar manner to those described for Chichester and Langstone entrances. This results in presumed net input, and retention, of suspended sediment (Wallace, 1988), but no quantitative estimate of this effect is available.

2.2 Fluvial Input

The volume of freshwater flowing into the harbours is, as explained below, small in comparison with tidal fluxes, carrying relatively little suspended sediment (Harlow, 1980).

FL1 Hermitage and Brockhampton Streams

Analysis of streams flowing into Langstone Harbour (Portsmouth Polytechnic, 1976) revealed that only the Hermitage Stream supplies freshwater in significant quantity, with a maximum daily discharge of 1.4 million m³. This is less than 0.25% of the spring tidal prism. It was stated in this report that little, if any suspended or bedload material is brought into the harbour by local streams and drainage outfalls, although there are no field measurements to corroborate this. Gao (1993) calculated a maximum daily discharge input of 40.6 x 104m³, and a minimum of 10.3 x 103m³, but these figures may include some allowance for freshwater ingress via seepage along the northern perimeter of the harbour. Based on various estimates, freshwater discharge is between 16 and 45m³ per second.

FL2 Wallington River

The only significant fluvial input into Portsmouth Harbour is the Wallington River with a flow of between 0.2m³ per second and 33m³ per second (Hydraulics Research Ltd, 1959; Rendel Geotechnics and University of Portsmouth, 1996). It is stated that sediment input from this source is negligible, but was not tested by field measurements.

FL3 River Ems, Chichester Harbour

A suspended load input of 1,450 tonnes per km² per year and bedload of between 268 and 541 tonnes per year is estimated by Rendel Geotechnics and the University of Portsmouth (1996).

FL4 River Lavant, Chichester Harbour

Unpublished Environment Agency data, and estimates using basic formulae (Rendel Geotechnics and University of Portsmouth, 1996), suggest an approximate input of suspended load of 2,100 to 2,400 tonnes per km² per year. Bedload discharge is probably negligible, below 100 tonnes per year.

2.3 Coastal Erosion

» E1 · E2 · E3  · E4 · E5 · E6  · E7 · E8

Sediment can be supplied to the harbours from erosion of both drift and substrate materials discontinuously exposed around their perimeters. Examination of maps indicates that much of the shoreline environment comprises reclaimed land at or below mean high water level, with erosion prevented by protective artificial bunds, earth banks or sea walls. Failure of these structures would arguably result in flooding rather than any significant erosion and sediment supply (HR Wallingford, 1994). Exceptions occur along some sections of the eastern and western shores of Hayling Island (Photo 8), the south eastern shore of Chichester Harbour (Photo 9) and the four major islands in Langstone Harbour, which are not protected and for which there is evidence of erosion of their outlines, and overall retreat, since the late eighteenth century (Allen, 2000a). No parts of the perimeter rise more than three metres above mean high water level so erosion forms low cliffs which can only supply limited sediment volumes even when retreating (Clare, 1996).

Clare (1996) suggests an erosion rate of 0.8m per year for the period 1967-1995 affecting the cliffs that comprise London Clay, in the south-east of Langstone Harbour (Photo 7). The only other quantitative study of harbour perimeter erosion is by Hooke and Riley (1987) who plotted the position of mean high water mark between 1870 and 1965 using successive Ordnance Survey maps.  Their major finding was that the harbour shape had changed very little since the sixteenth century, except in areas of land claim and oyster bed construction. This conclusion is supported by other analyses, e.g. Tubbs (1999), Gao and Collins (1994) and Allen and Gardiner (2000) for Langstone, and W.S. Atkins (2000b) for Chichester Harbour. Cliff erosion is therefore either too slow to be mapped at the cartographic scale of analysis (mostly 1:10,560) or is poorly represented by historic changes of mean high water position. Many qualitative statements relate to the erosion of low cliffs around the perimeters of the harbours (e.g. HR Wallingford, 1997). These reports do suggest that some sediment is supplied from this source and that a part of this contribution is from relict late Pleistocene and Holocene materials (Sorour, 1999; Allen, 2000b). Loss of protective vegetation due to riparian agriculture, and/or recreational activities, may be independent causes of instability and sediment yield. The lack of penetration of waves operating on the adjacent open coasts, together with limited fetch for internal wave propagation due to the historical extent of Spartina marshes, previously limited the potential for erosion of both seabed and cliff sediments. Fetches are now increasing as marshes retreat indicating that shoreline wave energy should also be increasing. Maximum significant wave heights measured in Langstone Harbour are 0.8m (Baffins) and 0.86m in Portsmouth Harbour (Whale Island), with an extreme (1 in 50 year recurrence) of 1.10m calculated for Anchorage Park, Langstone Harbour (HR Wallingford, 1995; Halcrow, 2010a and b). For south-east Chichester Harbour (W.S. Atkins, 2000b), comparable figures are 0.46 and 0.56 m. An extreme 1 in 100 year wave height of 1.9m for the north-west of Portsmouth Harbour has been calculated (HR Wallingford, 1995; Halcrow, 2010a).

E1 Warblington Castle (see introduction to coastal erosion)

Harlow (1980) reported that low cliffs of Brickearth were eroding near Warblington Castle. This cliff was 300m in length and yielded small quantities of clayey silt and flinty detritus. Chalky clay and Coombe Rock outcrop on the beach and are also subject to erosion. The cliff exposure is now largely protected by an embankment, upgraded in 1989 and subsequently maintained to prevent breaching.

E2 Tournerbury (see introduction to coastal erosion)

Hydraulics Research Ltd (1987) reported that the earth embankment between Tournerbury and Pound Marsh on Hayling Island was rapidly eroding; it has recently been partially upgraded.

E3 Chichester Harbour (see introduction to coastal erosion)

The perimeter of Chichester Harbour was field walked by Cartwright (1984) to locate archaeological remains exposed in low cliffs, or eroded from them. Although not a geomorphological study, this work identified the areas subject to erosion; the presence of archaeological remains, primarily worked and fire-cracked flints, amongst the sediments of the upper foreshore suggested a long history of erosion. A crude pattern of sediment supply was inferred from this information. The presence of eroding Tertiary cliffs and platforms along the more exposed southern margins of the Bosham, Chidham and Thorney peninsulas (Bone, 2006) has also been noted in various unpublished sources and by observation. This may provide a locally significant input of clay, sand and gravel derived from Eocene bedrock and overlying drift sediment.

Deterioration of sea defences west of Longmere Point, Thorney Island has allowed localised erosion of the earth banks and low cliffs behind, causing development of several scour holes (Calderwood, 1986). These 2-4m high cliffs are cut into the underlying Eocene substrate, and various overlying drift sediments. Cliffs of up to 5m in height along the south-facing coastlines of both the Bosham and Chidham peninsulas occur where there are no protection measures, or where the latter have temporarily failed (Carter, 2006). At both sites, there is clear evidence for small scale slumping, basal notching and platform abrasion in Eocene clays, overlying Quaternary drifts, particularly Brickearth, have been undermined and have left coarse materials on the adjacent beach.

E4 Creek Point (see introduction to coastal erosion)

Collapse of structures associated with oyster beds at Creek Point has introduced some erosion since 1932 (Hooke and Riley, 1987). This site has recently benefitted from reconstruction. Erosion of low cliffs, 1.5 to 3.0m in height, along the Mengham to Selsmore frontage is also encountered where former saltmarsh has been converted into mudflats (W.S. Atkins, 2000).

E5 Portsmouth Harbour (see introduction to coastal erosion)

A small outcrop of London Clay is subject to erosion at Hardway, Gosport

E6 Langstone Harbour (see introduction to coastal erosion)

The low cliffs south of Stoke, cut into London Clay and Reading Beds on the west coast of Hayling Island are subject to shallow slumping and provide a small sediment yield - Photo 7 and Photo 8 (Clare, 1996). Rates of recession may be accelerating due to narrowing and steepening of the adjacent inter-tidal slope. Allen and Gardiner (2000) report both surveyed and visual evidence of active erosion of low cliffs at the junction of saltmarsh and mudflats in the north-east part of the harbour. During the programme of archaeological survey, cliff line recession, block detachment and the rapid accumulation and destruction of small debris stores were observed on Baker and Long Islands.

Erosion monitoring conducted by the RSPB, 1988-1996, at six sites on North Binness and Long Islands revealed active wave abrasion of shore platforms; cliff retreat at a mean rate of 0.34m per year and the creation of "furrows" and "spurs" at the edge of low cliffs delimiting areas of saltmarsh. Evidence of rafting of pottery shards and two sarsen blocks during storms, between 1997 and 2000, is also reported (Allen and Gardiner, 2000). All of the artefact scatters are presumed to be the result of long-term vertical and lateral erosion. These observations would confirm the view of Tubbs (1999) that inter-tidal profile narrowing has been dominant since the early nineteenth century. Localised compaction of sediments may contribute to profile lowering, but this is not considered to be a significant factor.

E7 Harbour Entrances (see introduction to coastal erosion)

Study of Langstone and Chichester Harbour entrances by Hooke and Riley (1987) revealed significant erosion at Eastney outfall (0.48 m per year 1870-1932), together with a shortening and thickening of Eastney Spit (Photo 1) and erosion of the proximal point of Black Point, Hayling (0.32 m per year, 1910-1968). Webber (1974) also reported erosion between Eastney outfall and Eastney Spit. These spits and associated beaches are predominantly composed of gravel so that their erosion releases material to the tidal streams in the entrance channels. Net bedload transport of coarse material at each entrance is offshore due to the greater velocity of the ebb current so erosion at the harbour entrances cannot supply sediment to sites within the harbour except under very infrequent combinations of southerly waves and peak flood tide velocities. Further details are given in the units on (i) East Head to Pagham Harbour, and (ii) Chichester Harbour to Portsmouth Harbour Entrance "open" coastline.

E8 Saltmarsh erostion - new arrows

E8 arrows have been added to appropriate sites within Portsmouth, Langstone and Chichester harbours to represent input of fine sediment (silt and clay) from historical and ongoing saltmarsh erosion. The following table is taken from the Solent Dynamic Coast Project (Cope et al., 2008) which presents a quantified inventory of saltmarsh loss throughout the harbours, interpreted and mapped from historical (1946-2002) panchromatic and false colour infrared aerial photographic cover supported by ground control and map analysis. This research was informed by previous work using the same or similar methodology (Baily et al., 2000; Bray and Cottle, 2003; Baily and Pearson, 2007; Cope and Gorczynska, 2006; Cope, et al., 2007).

Table 1 presents the earliest (Year 1) to most recent (Year 2) saltmarsh extent mapping for the harbours. The % saltmarsh loss per year can be used to compare the differences in rates of contraction. Details for each harbour are given in section 5.4. Variability in % saltmarsh loss per year is attributed to historical land reclamation, exposure to wave attack, elevation of the marsh, the presence of salt pan formation, sea level rise leading to coastal squeeze, Spartina dieback and dredging (Baily and Pearson, 2007; Cope, et al., 2008).  

2.4 Sewage Input

Significant quantities of suspended solids are carried within sewage effluent and this must be considered as an input where outfalls empty into the harbours. Sewage discharges can increase nutrient concentrations which can in turn increase primary production, thereby facilitating sedimentation of organic matter (Portsmouth Polytechnic, 1976). Sewage outfalls exist at Budds Farm (discharging into Langstone Harbour at times of high storm water runoff) and at Fishbourne. Although details of discharges are recorded (Portsmouth Polytechnic, 1976; Thomas, 1987), sewage effluent has been examined as an impact on the biology rather than the sedimentology of the harbours. Despite this, studies with radioactive tracers (Portsmouth Polytechnic, 1976) indicated that once discharged into the harbours, sewage effluent was not easily removed and tended to accumulate. A net water volume input into Langstone Harbour from Chichester Harbour via Chichester Channel (3.5 x 106m³ spring tides; 0.97 x 106m³ neap tides) may have a limited impact on the accumulation of organic material. The net export of 0.73 x 106m³ (springs) and 0.20 x 106m³ (neaps) of water from Langstone to Portsmouth Harbours, via Ports Creek, is considered to be negligible in this respect. Major changes in sewage and flood water discharge into Langstone Harbour became effective from late 2001, thus effectively removing this input source.

3. Sediment Transport within the Harbours

Introduction

Although some beaches within the harbours have a maximum fetch of up to 5km, wave generation is inhibited by shallow water so only limited energy for sediment movement is available. The maximum recorded significant wave height in Portsmouth Harbour is 0.50m, with the 1 in 100 year signifcant wave height estimated to be 0.80m (Halcrow, 2010 a, b, c). Waves of this height would only occur along the northern shorelines in association with southerly winds (HR Wallingford, 1994, Halcrow, 2010 a, b, c). In sheltered areas, such as the Camber, wave heights are less than 0.20m (Halcrow Maritime, 1999; Halcrow, 2010 a and b).

The Coastal Monitoring Programme principally provides lidar and aerial photography data for the harbours, with topographic baseline survey data limited to the western shore of Hayling Island. Analysis indicates negligible changes in beach widths and levels, with unquantifiable low volumes of sediment deriving from erosion of low cliffs and banks. Sediment is transported by several mechanisms:

3.1 Littoral Transport

LT12 Thorney Island

Harlow (1980) suggested that littoral drift in all of the harbours was possible in some of the wider embayments, but does not give any specific locations. An example is the bay between Marker Point and Longmere Point on Thorney Island (Chichester Harbour) where Calderwood (1986) reported apparent north-westward littoral drift of gravel. Analysis of Coastal Monitoring Programme data indicates that this narrow fringing foreshore is stable in position and extent suggesting transport rates and volumes are negligible.

LT13 South-East Chichester Harbour

The presence of thin, narrow gravel spits, such as Ella Nore and Horse Pond in south-east Chichester Harbour indicate possible local northeastward pathways of movement, fed by local eroding cliffs. Analysis of Coastal Monitoring Programme data indicates that these increasingly vegetated spits are stable in position and extent suggesting transport rates and volumes are negligible, although there was some growth of the Ella Nore feature between 2003 and 2008.

LT14 Western Hayling Island

Analysis of Coastal Monitoring Programme data supports northwards shoreline drift at negligible to low rates indicated by the growth directions of small and historically impersistent shoreline gravel spits (Photo 11 and Photo 12) fed by local eroding cliffs and banks.

3.2 Tidal Sediment Transport

» T1

Sediment transport by tidal currents primarily involves a reversing motion of suspended material in response to ebb and flood tidal flow. Information reviewed in previous sections indicates net seaward transport (output) of coarse materials (bedload) and net shoreward transport (input) of mostly fine materials (suspended load). Net transport through the entrance channels is relatively small in quantity. Few studies of the pathways of transport have been conducted but the majority of transport is probably concentrated in tidal channels where current velocities are greatest (Gao and Collins, 1994; HR Wallingford, 1994, 1997). Detailed investigations of current velocities over a range of tidal conditions in Langstone Harbour indicated the likely sediment transport paths (Portsmouth Polytechnic, 1976). This study, and Gao (1993), showed that currents were principally westward at the northern connections between harbours so that a small quantity of net suspended sediment transport might be expected from Chichester to Langstone and from Langstone to Portsmouth Harbours. However, most research indicates that sediment transport is mostly confined to a closed circulation within the harbour system (Portsmouth Polytechnic, 1976). The release of fine sediment formerly stabilised by Spartina saltmarsh has increased sediment concentrations in flood tide streams, leading to net accretion in the upper part of tidal channels in Chichester Harbour (HR Wallingford, 1994a, 1999). A part of this load may derive from the abrasion of the expanding area of mudflats. Small, transitory low elevation banks in the north-east part of Langstone Harbour (Allen, 2000b) suggest that the coarser fraction of sediment suspended by both tidal currents and waves may be diverted into short term storage. They may be the product of very occasional exceptionally high levels of wave energy associated with storms.

T1 Sand Circulation on Sword Sands by Tidal Streams

A sidescan sonar survey and analysis of air photos covering 30 years revealed the presence of sand waves on Sword Sands (Langstone Harbour) which suggested bedload sand transport close to the harbour entrance (Humby and Dunn, 1975). Measurement of these sand waves indicated intermittent movement associated with the progression of tides from springs to neaps. Transport was northward in the centre of the bank and southward along the edges. Although the scope of study did not reveal the full extent of sediment circulation, it was concluded that dynamic stability of the bank was preserved despite rapid transport rates due to the existence of separate ebb and flow channels and associated transport pathways. It was inferred that bank instability could be caused by variations in sediment supply. Similar circulation systems probably operate in other parts of the harbours, but as yet they have not been detected.

3.3 Gravel Rafting

Attachment of buoyant weeds to gravel sized particles on beaches around the harbour perimeter was observed by Harlow (1980) and Wallace (1988). These were frequently larger than unattached clasts; some were well rounded whilst others were angular. Concrete and brick fragments were also present. It was concluded that more rounded pebbles are or had been transported into the harbours from the open shore whilst sub-angular to angular particles were moved from the harbour bed to supply the beaches (Harlow, 1980). The process has not been directly measured or studied so this information is of low reliability. The 2004 arrows indicating this process have been removed.

4. Sediment Outputs

4.1 Tidal Current Transport Output

Refer to Halcrow (2010 a, b and c) for a succinct account of the hydrography of Portsmouth and Langstone Harbours.

Sediments are transported out of the harbours by the dominant ebb tidal stream. However, the longer duration of the flood tidal stream causes net input of both coarse and fine suspended sediments. The greater velocity of the ebb tidal stream results in net output of sand and coarser materials which form semi-submerged deltas extending several kilometres offshore (Harlow, 1980; Wallace, 1988; HR Wallingford, 1996; Halcrow, 2010a, b and c). Thus, coarser materials should be a net output from the harbours. A detailed current metering programme conducted over a representative tidal range in Langstone Harbour by Portsmouth Polytechnic (1976) revealed that peak current velocities diminished significantly away from the entrance. Transport of coarse particles is therefore only possible within the main tidal channels adjacent to the entrance. As coarse sediments are not readily transported within the harbours output by this mechanism is probably small (Harlow, 1980). The majority of material flushed offshore by ebb currents therefore consists of coarse sediments transported into the entrance approaches due to wave action in the eastern Solent. This may amount to little more than 1,000m³ per year (Harlow, 1980). Although neither its volume nor provenance has been extensively studied, it appears unlikely that this process is a significant net output from the harbours (HR Wallingford, 1997). Further discussion will be found in the unit on the open coast between Portsmouth and Chichester Harbour entrances.

4.2 Land Claim

Significant areas of Portsmouth Harbour and, to a lesser extent, Langstone Harbour have been reclaimed (Hydraulics Research Ltd, 1959a and b and 1987; Portsmouth Polytechnic, 1976; Colebourn, 1984; Hooke and Riley, 1987; Privett, 1990; HR Wallingford, 1995; Universities of Portsmouth and Newcastle, 2000; Cope et al., 2008; Halcrow, 2010a). Land claim has impounded large volumes of previously potentially mobile harbour sediment so that this practice must be regarded as an output. Although the above studies record major areas and dates of land claim, few precise details of the volume and character of sediments are known. Land claim also affects harbour sediment budgets as the tidal prism is reduced, tidal currents diminish and sedimentation is stimulated. This effect was theoretically calculated for Portsmouth Harbour by Hydraulics Research Ltd (1959a and b) using a physical model. The potential impact of two proposed major reclamations was investigated at several points in the major tidal channel, and it was concluded that ebb and flood currents could be reduced by up to 25%. The effect of this at the entrance channel was modelled, using coloured sand tracers. Results showed that material would be flushed out of the harbour entrance but that it would tend to be re-deposited closer inshore. This study was not fully representative of hydraulic conditions because wave effects were not included and the scale of the model was not consistent.

The impact of land claim on the tidal prisms of both Langstone and Chichester Harbours is probably small (W.S. Atkins, 2000b), and its effect on the sediment budget of the harbours is therefore difficult to determine because even slightly diminished tidal flows reduce both sediment inputs and outputs. Circumstantial evidence suggests that it may have greatest effect on reducing marine inputs. Universities of Portsmouth and Newcastle (2000) calculate that just over 9% (circa 85 ha.) of the pre-1770 area of Langstone Harbour has been reclaimed, most of it by localised enclosures for grazing meadows, recreation space, wharves  and salterns. In the case of Portsmouth Harbour, the figure is much larger, approximately 35% (estimated to be some 1,850ha since the early sixteenth century) with the process continuing up to the present in response to various development pressures because of the confinement of Portsea Island on urban growth. Between 1970 and 1973, almost 27% of the extant intertidal area was polderised and converted to a mix of manufacturing and infrastructure related land uses. The effect on tidal currents was thus comparable to that modelled by Hydraulics Research Ltd (1959 a and b). Since the early 1980s most land claims in Langstone and Portsmouth Harbours have been connected with marina developments (e.g. Eastney, Haslar and Port Solent). Port Creek, connecting both harbours, has also been straightened and confined by 10ha of marginal land claim. Some sites have been created, and others extended, through the use of landfill. In the mid and late nineteenth centuries, dock excavation in Portsmouth Dockyard provided spoil used in the construction of Alexandra Park and some artificial islands, e.g. Burrow, Whale and Horsea (HR Wallingford, 1997; Nicholas Pearson Associates, 1997; Pitt, 1970; Portsmouth City Council, 2000; Privett, 1990a and b; Halcrow, 2010a). Pewit is the only natural island in the harbour, comparable to those in Langstone.

4.3 Dredging

Several main channels in Portsmouth, Langstone and Chichester harbours are dredged for navigational purposes and parts of their tidal deltas are also dredged for the same reason, and previously for aggregates. These practices represent permanent actual outputs from the harbour sediment systems, although aggregate removal has been discontinued since the early 1990s.

Portsmouth Harbour

The entrance (Photo 1) and approach channels are routinely maintained for navigational purposes, with less frequent capital dredging. The narrow entrance and consequent strong (ebb) tidal currents produce a "self-flushing" mechanism that reduces the need for frequent dredging. Examination of charts and dredging records for the period 1783-1954 by Hydraulics Research Ltd (1959a and b, 1985, 1993) HR Wallingford (1993; 1995; 1997) revealed that the entrance channel was deepened to 4.3m below Chart Datum in 1783, 10.4mCD in 1928, 11.7mCD in 1986 and 12.6mCD in 1996. Much capital dredging was carried out between 1881 and 1928 and involved extraction of an average of 25,000m³ per year. Dredging removed the Inner Bar, at the harbour entrance, in the late nineteenth century. This was an equivalent feature to the tidal deltas offshore the mouth of the other harbours, and has not redeveloped subsequently. Partial removal, and flattening, of the Ballast Bank, adjacent to the entrance to the former submarine base of HMS Dolphin of the harbour mouth took place in 1985 (Hydraulics Research Ltd, 1984, 1985, 1986). Preliminary analysis of bed shear strengths revealed that this feature is marginally mobile, but it is uncertain if it represents a product of bedload input (i.e. a flood tide delta) and if it contributes to output.

The inner Camber basin is dredged to 1.8m below CD, except for the Wightlink terminal where the vessel berth is maintained to 3.5mCD (Halcrow Maritime, 1999). Significantly greater depths, up to 11.2mCD exist alongside the berths in the Naval Base and the Commercial Docks at Flathouse/Continental Ferry Port.

Licensed (FEPA) maintenance dredging of navigation channels between 1987 and 1997 removed 237,000 tonnes from within Portsmouth Harbour and 333,000 tonnes from Hamilton Bank and Spit Sand immediately south of the harbour entrance. The latter figure may, however, underestimate the actual total. (Universities of Portsmouth and Newcastle, 2000). Capital dredging for the same period was confined to the harbour, and accounted for 1,530,000 tonnes. Whilst maintenance dredging removed essentially mobile sediment, capital dredging involved increasing water depths at berths, quays, docks and marina basins in addition to the depth of approach channels. It therefore also removed previously immobile sediments. The figure for maintenance dredging gives a very crude approximation of sediment input and internal reworking for this ten year period. All dredged material ("spoil") was deposited at the Nab dumping site at the entrance to the East Solent, and therefore represented a significant output term in the harbour's sediment budget. It is possible that between 2 to 5% of the above totals are actually retained, in the form of banks marginal to the main channels created by the dredging process.

As navigation channels have been progressively deepened, both ebb and flood tidal current velocities have increased by small amounts. This may have either inhibited sedimentation or promoted accretion in upper harbour areas. Neither of these effects can be demonstrated from existing hydrographic data (Universities of Newcastle and Portsmouth, 2000). The effect of dredging on the overall harbour sediment budget is therefore impossible to evaluate.

Langstone Harbour

The harbour entrance channel, Langstone Channel (up to Bedhampton), Kendall’s Wharf and Broom Channel are currently dredged for navigational purposes (Universities of Portsmouth and Newcastle, 2000; Halcrow, 2010a). Langstone was a port until 1914 and approach channels within the harbour were dredged over the period 1882-1914 (Portsmouth Polytechnic, 1976). Aggregate dredging has largely removed the previously existing inner bar just seaward of the entrance (Wallace, 1988) but this practice continued on the west side of the East Winner bank (Hydraulics Research Ltd, 1987) until (as reported by STS 2002). Sinah Sands within the harbour has also been periodically dredged (Dunn, 1972; Portsmouth Polytechnic, 1976). No long-term reliable volumetric information is available, but dredging at the entrance could significantly increase suspended sediment concentrations, causing increased short-term input during the flood tide. Harlow (1980) estimated dredging activity over the previous 10-15 years to be in the order of 56,000 tonnes per year in the area of Langstone Bar, which, along with East Winner Bank, represents a store for sediment transported seaward from Gunner Point. Hydraulics Research Ltd (1987) cite a permitted extraction rate of 78,400 tonnes per year over Langstone Bar. A report by the Universities of Portsmouth and Newcastle (2000) suggest removal of "over" 120,000m³ per year from the Bar and the East Winner, 1987-1997 though dredging of the Bar (approximately 6,000m³ each year) ceased in 1994. Small quantities (less than 1,000m³ per year) have been periodically used to replenish Southsea Beach, west of Southsea Castle. However, by 1994 Langstone Bar had almost disappeared and by 2002 it was no longer visible from the aerial photography.

There are several references to dredging activity within the harbour which are not readily substantiated by quantitative data. Portsmouth Polytechnic (1976) state that Kendall Brothers, operating out of Kendall's Wharf, own the 'prescriptive right' to dredge Sinah Sands. This was continued (up until 1995) at the rate of 'several thousand tonnes' per annum. This sediment was destined for use on land, hence represented a loss to the system. Whilst the right to dredge still exists, future major dredging activity in this area is not foreseen. Two other areas subject to dredging are (i) the approach channel to Bedhampton Quay in the north of the harbour, which was dredged between 1970 and 1975 (Portsmouth Polytechnic, 1976) and 1991-91 (approximately 23,500 tonnes in the latter case); and (ii) the channel servicing Kendall's Wharf, where over 4,600 tonnes were removed in 1995-96. These operations continue, as and when required.

Quantitative information relating to dredging activity in Langstone Harbour reported by the Universities of Portsmouth and Newcastle (2000), based on FEPA licence data indicate that a total of 143,190 tonnes of sediment was dredged from Langstone Harbour and its approaches between 1987 and 1997. However, it is uncertain as to precisely how much material removed from the East Winner was utilised subsequently in local beach recharge operations, which therefore would not amount to a net loss from the regional sediment budget. It can be stated with certainty that the material periodically dredged from Bedhampton channel and dumped at the offshore Nab Tower site represents a loss from the system. This channel is currently maintained at a depth of 1.8m below Chart Datum, to give access to aggregate unloading at Bedhampton Quay. However, this dredging must be assessed in light of the fact that it is not continuous and hence any impact may be of limited temporal extent.

Chichester Harbour

Routine maintenance dredging is undertaken for navigational purposes on the ebb tidal delta outside the harbour, with part of the spoil used to maintain the replenished beach in south-east Hayling, according to availability (Havant Borough Council, 2000). Dredging of the Chichester Bar began, on a significant scale, in 1973 (Shaw, 1974). Some 600,000m³ was removed by this means between 1974 and 1982. Between 1988 and 2000 average annual quantity of dredging has not exceeded 20,000m³. Small scale maintenance dredging of the approach channels to harbour marinas (e.g. Chichester Yacht Basin and Eastoke Creek) occurs, but precise details of quantities removed - and their fate - are not available. HR Wallingford (1999) demonstrated the potential beneficial use of disposing of this material at or near the harbour entrance (and possibly elsewhere) to offset predicted net losses of fine grained sediments  from the overall sediment budget of the harbour due to mudflat erosion (refer to next section).

5. Sediment Stores

Mineralogic analysis of estuarine clay-size fraction sediments in all three harbours has revealed that they derive substantially from the previous erosion of Cretaceous and Eocene rocks (Algan, et al., 1994). Their provenance includes hinterland catchments and the substrate of the Solent and adjacent English Channel. However, some reworking of seabed sediments in the Solent is a probable auxiliary source. The distribution pattern of minerals characteristic of terrestrial and marine sources is fully consistent with mixing induced by estuarine sedimentation.

5.1 Beaches

Gravel and sand beaches exist around parts of each of the harbour perimeters, but description is incomplete and fragmentary. Harlow (1980) described a beach near Warblington Castle composed of granules and pebbles up to 75mm diameter. A significant quantity of the beach clasts is larger than the modal size of those in adjacent cliffs indicating transport, and sorting, from another source. Hydraulics Research Ltd (1987) mention appreciable gravel accumulations seawards of Portchester Castle, and Calderwood (1986) reported an upper gravel beach on Thorney Island between Marker Point and Longmere Point. Thomas (1987) also identified an "upper shore gravel" around the perimeter of several parts of Chichester Harbour. Sampling revealed that 15% of the material was less than 63 mm diameter and 25% to 60% was greater than 4 mm diameter.

Gravel beaches of significant width and height occur along much of the west facing shores of Thorney and Bosham peninsulas, and the west and south-west harbour coastline of Hayling Island (Photo 7, Photo 8 and Photo 13). Bellew (1990) analysed two small ephemeral gravel spits immediately northwest of Langstone Bridge. Sampling revealed very variable, poorly sorted, sediments including clasts up to 100mm in diameter. Gravel content was variously 50%-80%, sand 15%-30% and silt/clay 2%-30%. However, it is uncertain if these are natural landforms. Jenkins (1988) examined sediments at the Kench (Langstone Harbour) (Photo 3) and identified two probable relict shingle spits. W.S. Atkins (2000) report a "silty shingle" zone occurring at the foot of defences around the Selsmore peninsula, south-east Hayling (Photo 14). As at Langstone Bridge, the sediments were very poorly sorted, indicating either minimal transport by waves or currents or the possibility of the contamination of this deposit by former unloading of dredged marine gravels. Furness (1982) analysed the particle sizes of 18 samples from 8 sites on the Langstone Harbour perimeter. The highest organic matter content was for sediments sampled from the north shore, whilst the coarsest sediments were located near the harbour entrance. Investigations in Portsmouth Harbour at Tipner Lake, Stamshaw Lake and Portchester have also identified gravel foreshores, (Thomas, et. al, 1989a and b) a feature which appears characteristic of each estuary.

Gravel beaches and spits also occur around Pilsey Island (Photo 15) and along the south east margin of Chichester Harbour (Photo 6).

No study has examined the sedimentology of the harbour beaches in sufficient detail to determine the origins and stability of this material. Some may have been artificially introduced during previous centuries, but most probably derives from a combination of the erosion of Quaternary gravels and sands in harbourside bluffs, the reworking of harbour bed sediments and occasional input from external marine sources during very high energy conditions.

5.2 Sand Banks

It was reported by Harlow (1980) that the coarsest sediments in the harbours were located close to the entrances and that they became increasingly finer further up-estuary. This distribution pattern was largely determined from field observations and examination of sediments sampled in Langstone Harbour. Humby (1980) reported that sediment size range was found to be principally related to sorting by tidal currents. Wallace (1988) noted a relative scarcity of coarse sediments in Chichester Harbour and attributed this to the dominance of the ebb current promoting output. This tidal asymmetry tends to flush coarse sediments out of the harbours, possibly also resulting in a modest net loss over unknown timescales. However, the development of a small gravel "bulge" that has developed since the mid-1980s close to Fort Cumberland outfall in the Langstone entrance channel reveals the possibility of at least short-term reversals of this trend. A similar pulse of gravel moved towards the Hayling Ferry landing stage, on the opposite side, in the early and mid-1990s and after 2001 (HR Wallingford, 1995; Cope et al., 2005). Both may be fed from the Langstone ebb tidal delta, but this remains speculative.

Portsmouth Harbour

The sedimentology of Portsmouth Harbour is very poorly covered by the literature except in two ecological studies of Stamshaw and Tipner Lakes (Thomas et al., 1989a and b). These studies include maps of surface sediment distribution, showing extensive muds and an area of fine sand within Stamshaw Lake. Sampling revealed a median grain diameter of 0.16 mm with 16% <63µ m and 84% between 63µ m and 4mm.

Langstone Harbour

Three distinct sand banks have been identified at the junctions of the major channels and their positions are indicated by Dunn (1972), Portsmouth Polytechnic (1976), and Hydraulics Research Ltd (1987). Mallard Sands are the furthest from the entrance and comprise the finest sediments, being composed of sandy mud (Dunn, 1972 and Hydraulics Research Ltd, 1987). Sinah Sands are almost adjacent to the entrance but east of the main ebb and flood tidal channels, which may be mutually evasive. They are composed of fine sand (Dunn, 1972) with a mean diameter of 0.17-0.22mm and very good sorting (Rowlatt, 1985). It was reported that the sand bank was receding by Dunn (1972) but accreting by Hydraulics Research Ltd (1987). The most reliable information is that of Humby and Dunn (1975) based on air photo analysis covering a 30-year period. This study indicated little significant change in overall sandbank configuration despite evident mobility of surficial sandwaves. Sword Sands are comparatively less well sorted and slightly coarser (Dunn, 1972) than Sinah Sands. Sampling also revealed that sorting was best at the summit of the banks and declined towards the tidal channels. Mean sediment size was smallest at bank crests and increased towards channel margins. Sampling in Langstone Channel (Rowlatt, 1985) showed that the channel floor comprised fine to medium sand (0.16-0.33mm diameter) together with much shell debris and some patchy gravels. Visual observations of Sword Sand (Rowlatt, 1985) indicated a relatively greater proportion of coarse materials including boulders, pebbles and shell debris. However, these materials were not adequately recorded by the sampling procedure, which was designed primarily for finer sediments. Air photos have revealed sand waves on Sword Sand indicating active sediment transport (Humby and Dunn, 1975). The results described above were broadly confirmed by detailed sampling at 62 sites by Humby (1980) which showed that the coarsest sediments were located in the main tidal channels within 300m of the entrance. Although detailed particle size distributions were presented by Humby (1980) no overall analysis, description or mapping of data was undertaken. Isolated accumulations of gravel, up to 7m in thickness, occur near Langstone village and Brockhampton. They may, in part, represent the sites of ballast dumping rather than natural sediment deposition.

Chichester Harbour

Thomas (1987) identified sand flats as a distinctive habitat covering 164 hectares (8% of total intertidal area), mostly near the harbour entrance. This area was characterised by material within the 1 mm - 60 mm size range but no distribution maps were presented. Hydraulics Research Ltd (1987) identified the major sand banks as being Pilsey Sand (Photo 6) and Stockers Sand. At Pilsey Sand, winds blowing across the wide sandy foreshore (Photo 16) entrain quantities of sand leading to small scale backshore dune formation (Photo 17).

5.3 Mudflats

The mudflats are generally situated away from the entrances, in locations where tidal current velocities are weak and fine sediments can settle during tidal stands (Harlow, 1980). Mudflats are now of much greater extent than 20-30 years ago because large areas of Spartina anglica marsh have been eroded, leaving low-lying mudflats with occasional hummocks - Photo 18 and Photo 19 (Portsmouth Polytechnic, 1976; Haynes and Coulson, 1982; Hydraulics Research Ltd, 1987; Collier et al., 1995, Cope et al., 2004; Cope et al., 2008). Uncolonised mudflats now occupy between 24 and 28% of the total area of each harbour - some 27km² in the case of Langstone Harbour (Adams Henry and ABP, 1995). In most areas, they are subdivided by integrated systems of apparently largely stable tidal creeks. Headward creek extension appears to be an essentially cyclical process of enlargement followed by infilling. The phenomenon of "coastal squeeze" of saltmarsh may be causing some net vertical accretion of mudflat surfaces.

Portsmouth Harbour

Very little information is available regarding the sedimentology of Portsmouth Harbour. Colebourn (1984) stated that in some areas the harbour mud was overlain by a thin veneer of gravel, but evidence supporting this view was not presented. Detailed surveys of limited areas have revealed substantial areas of mud in Stamshaw and Tipner Lakes (Thomas et al., 1989 a and b). Sampling in Stamshaw Lake enabled identification of an upper shore fine mud (mean diameter size 0.02mm with 94% <63µ) and a lower shore mud (mean size 0.05 with 54% <63µ). The morphology of the main mudflats was studied by Hydraulics Research Ltd (1959) using 5 charts between 1783 and 1954 and a field survey in 1956. The low water and 3 fathom contours were compared and no significant changes were detected. This stability was at that time attributed to stabilisation of salt-marshes by Spartina grass; however, much die back and retreat has occurred in recent decades leading to extension of low lying mudflats (Photo 2). Various sources suggest medium to fine gravel armouring the boundaries of creeks dissecting mudflats in both Langstone and Portsmouth Harbours (Nicholas Pearson Associates et al., 1997; Allen and Gardiner, 2000). Its provenance is unknown, but might derive from reworking of late Quaternary drift sediments underlying recent estuarine muds and silts. Unpublished surveys carried out by the City of Portsmouth Engineers' Department of the approach channel to the Continental Ferry Port have revealed apparent mobility of muddy sands and gravels. This may imply some tidal scour of the harbour bed and internal sediment re-working.

Langstone Harbour

The mudflats are composed of a mixture of clay, silt, very fine sand and organic matter (Portsmouth Polytechnic, 1976; HR Wallingford, 1997) and represent a long-term net accretion of material that is deposited from suspension during slack water. Limited sediment sampling at three sites near the northern and western perimeter revealed a mean sediment type of fine silt with approximately 75% finer than 0.06mm diameter (Portsmouth Polytechnic, 1976). Mudflats have extended significantly over the past 40 years due to Spartina anglica decline and subsequent erosion of naturally and artificially elevated salt marsh - see Photo 18 (Portsmouth Polytechnic, 1976; Haynes and Coulson, 1982; Budd, 1985; Collier and Fontana, 1996; W.S. Atkins, 2000; Baily et al. 2000; Baily and Pearson, 2007 and Cope et al., 2008). Mudflats may become relatively stable due to colonisation by algae; their decomposition increases organic content and mud cohesion - see Photo 20 (Haynes, 1982; Barne et al., 1996). A detailed programme of sediment sampling and analyses (Humby, 1980) determined particle size distributions for 62 sites, mostly on mudflats and in tidal channels. Channel sediments are much coarser and less well sorted than adjacent mudflat samples, which in turn become finer and better sorted further north and away from the main tidal channels. Near Long Island the sediment size range was 2mm-0.002mm diameter (fine gravel to clay) with a modal size of 0.037mm; further north in Chalkdock Lake the size range was 0.06mm to 0.004mm (coarse silt to very fine silt) with a modal size of 0.013mm. The results of these distributions were not mapped or further analysed and therefore are essentially as raw data. Sampling of clay revealed a specific gravity of 2.85, which was attributed to dissolved ferrous minerals from substrata beneath the mudflats and also other impurities such as zircon and magnetite (Turnbull, 1980). Examination of mudbank morphology over 30 years (Humby and Dunn, 1975) using air photos indicated little significant change of the configuration of the main banks and channels. Similar results were reported by Boylan (1984) from examination of OS maps for the period 1930-72. The effects of Spartina anglica dieback were not indicated by either survey so contemporary study of mudflat morphology is required for comparison. Sorour (1999) estimates a net vertical accretion rate of between 1 and 2mm per year but does not indicate precisely how this was determined. If it is reliable, Langstone Harbour may be gaining between 17 and 34 x 103m³ of intertidal mudflat sediment annually.

Chichester Harbour

Thomas (1987) calculated that mudflats covered 1298 hectares, 61% of the total intertidal area. These areas were characterised by a high proportion of silt and clay with veneers and possible marginal chernier accumulations of coarse clastic material.

5.4 Spartina and Saltmarsh

Substantial parts of the harbours became colonised by Spartina anglica beginning in the first decade of the twentieth century. The swards of grass interrupted water flow and increased sedimentation so that large areas of high level Spartina marsh accreted up to 1.5m above the level of the adjacent mudflats (Hall, 1979; Bray, 2010), ultimately occupying about 20% of the total (combined) harbour areas. Since the early 1950s dieback of Spartina anglica has occurred, a slow process initially, but accelerating in the mid-1960s and subsequently slowing from the late 1980s and early 1990s. With the death of the binding flora, the high level marsh is eroding at its edges forming receding cliffs up to 2m high (Allen and Gardiner, 2000; Cope et al., 2008), and is also subject to sward fragmentation through "dieback" (Photo 18 and Photo 19). Reasons for local dieback are uncertain and numerous causes have been postulated. These are reviewed by Portsmouth Polytechnic (1979); Hall (1979); Haynes and Coulson (1982); Haynes (1982); Hydraulics Research Ltd (1987) and HR Wallingford (1997), Raybould et al. (2000) Mant (1996); Bray and Cottle (2003); Baily and Pearson (2007); Cope et al., (2008); SCOPAC (2011) with specific reference to Langstone Harbour in several of these studies. The process of recession and erosion has been monitored using air photo and photogrammetric analysis for several periods between 1946 and 2007 (Portsmouth Polytechnic, 1976; Hall, 1989; Haynes and Coulson, 1982; Budd, 1985; Collier and Fontana, 1995; Baily et al., 2000; Baily and Pearson, 2007; Cope et al., 2008 and Bray, 2010), coupled with ground control, map analysis and field survey. The major conclusions of research published before 2007 were that mature Spartina marsh is now largely absent from Langstone Harbour and most remaining areas (approximately 50ha) are subject to decline. Baily and Pearson (2007) calculate a 72% loss of cover between 1956 and 2001. Marsh edge retreat around the harbour islands varied between 40 to 140m in the 45 years covered by their survey. Cope et al. (2008) state that the rate of loss declined after 1984, but by 2001 only isolated swards survived. Bray (2010) has deduced similar rates of loss from specific areas, mostly as a result of edge erosion, but also records a few sites of recent slight gain, e.g. Gutner Point. Little recolonisation by Spartina has been recorded, but in some areas such as the islands there has been invasion by other saltmarsh flora, including Halminione porticuloides, Salicornia ssp. and Plantago maritima. Dead or moribund high level marsh has been eroded and is now reverting to low-level mudflats with weakly concave profiles, thus entailing considerable release of sediments (HR Wallingford, 1994a). This was calculated at nearly 4.0 x 106m³ since 1960 (Universities of Portsmouth and Newcastle, 2000). Collier and Fontana (1996) provided estimates of percentages of annual change in vegetation cover, 1982-1994. These revealed losses of between 5 and 8% of original Spartina cover, with near simultaneous equivalent gains by Enteromorpha. The ultimate fate of released fine grained sediments is unknown as the mudflats are less sheltered than the Spartina marsh, so only the coarser silts are redeposited. Decline of Spartina marsh fringing the harbour perimeter was recorded by Haynes and Coulson (1982). This has been accompanied by the increased concavity and reduced width of inter-tidal mudflats (Tubbs, 1999). Increased exposure of the perimeter to wave attack and accelerated erosion of sea walls, embankments and natural cliffs (HR Wallingford, 1994b, 2000) is a probable outcome (Photo 7, Photo 8, Photo 9, Photo 18, Photo 21, Photo 22, Photo 23). The re-colonisation of bare mud by algae mats may serve to suppress erosion rates experienced directly after Spartina degeneration or extinction. Another factor that may account for more recent deceleration of erosion loss is the exposure of more consolidated mud.

The literature previous to the Solent Dynamic Coast Project (Cope et al., 2008) does not specify the condition of Spartina anglica salt marsh in Portsmouth Harbour other than to mention a general decline, with some areas surviving in the North West (Photo 24) and virtual elimination from central areas, Stamshaw and Tipner Lakes and the Portchester channel. Low 0.5m to 1m eroding bluffs now separate mudflats from residual areas of Spartina, e.g. Fareham Creek and Portchester. The latter occupy some 140ha in several isolated areas in the north-west of the harbour, with contemporary loss mostly due to marsh edge recession. Ulva and Enteromorpha have spread widely since the early 1980s across vacant areas of mudflat (Baily, et al., 2000). Cope et al. (2008) state that the highest rate of loss in Portsmouth Harbour was at 5.5% per year between 1971 and 1984, with a significant proportion of this due to the reclamation of the north-east sector. Between 1984 and 2002 decline continued, but at the reduced rate of 0.9% per year. Baily and Pearson (2007) concluded that the 62% reduction in Spartina cover between 1971 and 2001 had the effect of contracting saltmarsh in Portsmouth Harbour to residual “islands” due to dissection and fragmentation.

In Chichester Harbour accelerated dieback of Spartina began around 1967 and has until recently been comparatively more gradual and spatially variable. Exposed sites have generally suffered most from erosion e.g. the south and east facing shores of the Thorney and Chidham peninsulas (Photo 19) the south east shore, particularly the stretch from Itchenor to Rockwood (Hydraulics Research Ltd, 1987). Sheltered sites suffered less e.g. Bosham Channel (Photo 25). Thomas (1987) calculated that Spartina marsh covered 611 hectares (29% of the total intertidal area) in 1984-85, but no distinction was made between areas of healthy and degenerate Spartina. In some areas, Spartina remains unaffected, in others (such as the Mengham to Selsmore frontage) Halimione and Salicornia spp. are co-dominant with Spartina anglica (Photo 26). The latter tends to form small monospecific swards in the central-east of the harbour. Baily and Pearson (2007) calculated that Spartina reduced in extent from 552 to 396 ha between 1971 and 2001 a 28.2% loss. For the longer period 1946 to 2002 Cope et al. (2008) conclude that contraction was close to 53%, with the highest losses at 2.7% per year occurring between 1965 and 1971. All researchers note that the inroads into Spartina have been greatest along the eastern shoreline of Hayling Island, due to both edge recession and fragmentation.

Following laboratory experiments, Tsuzaki (2004, 2010) found anaerobic soil conditions with impeded drainage to be the most likely cause of the dwarf growth forms and lack of re-colonisation of pans and mudflats by Spartina anglica on the south coast of England.  The thesis concludes that the ultimate demise of the Spartina anglica marshes of the south coast of England is the result of frontal and creek erosion of the mature marsh and the failure of Spartina anglica to establish itself as a pioneer on the newly exposed sediments of the foreshore.

Several of the research papers cited above note that in all three estuaries there has been only minor changes to the spatial plan of the network of tidal channels (creeks) over the past two hundred years. From this it might be inferred that sediment release from Spartina loss derives principally from the reduction of elevation and recession of the leading edge of inter-tidal marsh, with lateral channel abrasion a relatively less important contributor. Headward creek extension has been frequently observed, but is characteristically short term.

5.5 Thickness of Harbour Sediments

Borehole investigations have yielded site-specific information on the thickness of unconsolidated sediments at several sites. Pitt (1970) described a series of 28 boreholes in Paulsgrove Lake and on Horsea Island, Portsmouth Harbour. The boreholes detected a surface layer of recently deposited "soft" silt up to 0.8m thick beneath which there was a stiff grey-brown clayey silt with some chalk fragments; this deposit was up to 3.5m thick. Elsewhere (e.g. Port Solent, Gunwharf and beneath the alignment of the proposed Rapid Light Transit tunnel) sediments form a layer between 2m and 4m thick above unweathered Chalk bedrock (ABP Research and Consultancy Ltd, 1994). Bond (1985) described a series of boreholes drilled to investigate the site of a gas pipeline crossing at the head of the Fishbourne channel in Chichester Harbour. Six boreholes were sunk in the channel and determined an upper layer of soft alluvium 0.2m-0.9m thick; below this lie soft organic clays 0.5m-1.0m thick, beneath which occurred a series of variable chalky gravels, probably head deposits/coombe rock. In situ Chalk was encountered between -4.2 and -7.8mOD. The upper harbour estuarine deposits are therefore a maximum of 2.0m to 2.5m thick. In Langstone Harbour, Turnbull (1980) quotes a maximum marine sediment thickness of 12m and Mottershead (1976) reproduces a set of borehole logs for central and southwestern parts of the harbour indicating a Holocene and modern sediment thickness of between 3m and 11m. Each of these estimates of thickness relate to infill of formers incised channels. Auger surveys reported by Allen and Gardiner (2000) for north-east Langstone Harbour indicated an average thickness of marine silts of approximately 3m. This information needs supplementary data, but suggests a possible storage volume of intertidal sediment of between 55 and 400 x 106m³. This range illustrates great uncertainty on details of average sediment thickness.

5.6 Suspended Sediments

Sampling of suspended sediments was undertaken by the Admiralty at seven sites in Portsmouth Harbour, collected from various depths at hourly intervals on a spring tide. Maximum concentrations of up to 100ppm by weight were recorded in Tipner Lake and Fareham Creek. These compared with a maximum concentration outside the harbour of 10ppm and indicated that suspended sediments accumulate within the harbour (Hydraulics Research Ltd, 1959a), where concentrations in the north-west part are 10-15 times greater than near the entrance. Suspended sediments in Langstone Harbour were measured at 200mg/litre but concentrations were sensitive to variations in weather and wave conditions (Humby and Dunn, 1975).

Microscopic analysis of these suspended sediments revealed them to be mostly clay and silt size, medium sand, including flocculated particles up to the size of coarse silt. They are much larger than comparable sediments previously studied in other regional estuaries, a feature attributed to biological (mucus) binding. These flocs are relatively unstable, aggregating and disaggregating according to the hydraulic forces acting upon them (Humby and Dunn, 1975). Although the clay minerals within flocs matched local geological materials (using X-ray diffraction analysis), the size of the flocs were similar to those derived from sewage effluent suggesting this may also be a factor in their settlement. Bradbury (1989) examined spatial variations of concentrations of suspended sediments using enhanced multispectral Landsat satellite images. The presence of suspended sediments were clearly indicated. Analysis revealed that suspended sediment concentrations were greatest in the south and west parts of Langstone Harbour and the central and western parts of Portsmouth Harbour. Temporal co-representativeness was limited, for analysis was only conducted close to high tide. If the value reported by Humby and Dunn (1975) is representative, and assumptions on the mean tidal prism and tidal range are factored in (Portsmouth Polytechnic, 1976), quantities of suspended sediments are close to 10,000 tonnes on spring tides and 4,600 tonnes on neaps. This equates to a potential flux of approximately 3,490m³ and 1,620m³, respectively. HR Wallingford (1994a) reported, from a limited number of grab samples in south-east Chichester Harbour, that unconsolidated silty and silty-sandy mud overlies more consolidated, denser silty clays; the depth of the interface averages 0.05m below the bed surface. The latter have shear strengths well above shearing velocities induced by tidal currents. It was observed that ebb tide suspended sediment concentrations (25-35ppm) are higher than those for flood currents (8-12ppm). Concentrations increased up creek channels, but peak shear stresses are associated with spring tide ebb flows in the deepest parts of harbour channels, immediately south of the harbour entrance.

Suspended sediment concentration values on spring tide currents suggest that a layer some 20mm thick of silty clay is entrained on a single tidal cycle, with a proportion of this load entering the Itchenor and Bosham channels on the flood tide. Perhaps 75% of suspended load introduced by flood tides is removed seawards by succeeding ebbs. An unknown proportion of fine sediment thus removed from the harbour re-enters on the next flood, but this will diminish over several succeeding spring and neap cycles. Eventually, little of the fine suspended sediment load flushed seawards will return, having been diverted to either upper creek or outer harbour storage. The overall impact of suspended sediment flux on the main channels is therefore to induce some shallowing of the highest (innermost) creeks. There is probably net deposition on channel margins during weaker neap currents, but this is probably re-entrained during spring tides. HR Wallingford (1994a) concluded that some 55% of potentially mobile silt and clay is moved as suspension load. Tubbs (1999) has described several areas of slight elevation in upper creek areas, between 1850 and 1980 which would also indicate net accretion. Further information may be found at http://www.scopac.org.uk/sediment-sinks.html.

6. Sediment Budget

Analysis of sediment inputs and outputs based on existing information indicates significant, but largely unquantified input of suspended sediments through the harbour entrances. Natural outputs are limited to some immediate removal or later remobilisation of this input so the harbours have been subject to net accretion over recent millennia. A small quantity of coarse sediment input, by wave-induced bedload transport is balanced by ebb tide output of the same size-range of material. Net input of fines has been partially offset by losses through land claim and navigational dredging. Rates of accretion have not been precisely determined, although Harlow (1980) postulated that the apparent stability of mudflat and creek patterns indicated that sedimentation was keeping pace with sea-level rise with net accretion raising the elevation of mudflats. The evidence is not conclusive, although Wallace (1990) suggests that apparent stability results from the narrow harbour entrances, which might slightly reduce the effects of relative sea-level rise within the harbours by restricting the tidal range. Overall, the harbours are quasi-closed sediment circulation systems (sinks) maintained in a state of dynamic stability by their narrow entrances with a modest but cumulative input of suspended sediment. There is considerable internal re-circulation and some mixing of sediment, accentuated by the remobilisation of fine-grained materials due to Spartina "dieback". Some of this contributes to the flux of ebb output and flood input at the harbour entrances.

7. Knowledge Limitations and Monitoring Requirements

Data collected by the Defra-funded National Network of Regional Coastal Monitoring Programmes is pivotal for future improvement in estimating beach change.  The Programmes consist of topographic beach surveys, nearshore bathymetry, aerial photography, lidar, coastal hydrodynamics (waves and tides) and terrestrial habitat mapping.  Specifications for data collection are consistent for all regional programmes and the data and analysis reports are made freely available under the Open Government Licence from www.channelcoast.org.

The Southeast Regional Coastal Monitoring Programme commenced in 2002. The Lead Authority is New Forest District Council, with data collection, analysis and reporting led by specialist teams at the Channel Coastal Observatory (CCO), Canterbury City Council and Adur and Worthing Councils. Longer term Coastal Monitoring Programme data, when combined with other data sets, academic research and historical studies may enable sediment budgets, transport rates and directions to be identified and/or validated in the future, although the lack of significant wave energy and the poor development of beaches means that shorelines in this unit are not suited for definitive studies of drift.

Notwithstanding results from the Southeast Regional Coastal Monitoring Programme recommendations for future research and monitoring that might be required to inform management include:

Input of suspended sediment through the harbour entrances is clearly an important contribution to harbour budgets, but has not been adequately quantified. It could be determined from measurements of suspended sediment concentrations at each harbour entrance, coupled with simulated numerical modelling of the ebb depth-average and flood tidal stream velocities and directions.

Erosion of the harbour perimeters is a current process but sediment supply from this source has not been systematically quantified.

The input of organic material by primary production and the fate of this material upon decomposition have not been assessed using budgetary concepts. Similarly the sediment input of sewage effluent has not been evaluated nor has a precise relationship been determined between sewage discharges and increased primary productivity, which stimulates the production (input) of organogenic sediment.

Detailed, systematic surveys of harbour surface (seabed) sediment distribution are required. Existing maps only cover limited parts of Langstone Harbour and sediments previously sampled have not been fully analysed or mapped. Existing information should be used to plan field surveys of areas not previously studied, e.g. much of Portsmouth Harbour, to provide basic surface sediment data. Some limited sampling should be undertaken to characterise the sediments identified and provide quantitative information against which future variations can be measured.

Beaches around the harbour perimeter should be examined to determine their variable extent and sediment composition.

Spartina anglica dieback has been recorded and mapped by Haynes and Coulson (1982); Budd (1985) and Collier and Fontana (1996) for Langstone Harbour. However, the fate of sediments eroded from moribund Spartina marsh has not been established. Photogrammetric analysis of mudflat levels and tidal channel morphology (Collier and Fontana, 1996) indicates that some channels in the northwest and southeast have shallowed since the early 1980s, suggesting that one cause might be sediment redistribution resulting from the release of material from areas of Spartina mortality. Some areas of saltmarsh cliff erosion are balanced by co-adjacent accretion on mudflats whereas elsewhere (e.g. South Binness Island) the cliffline has been stable. Care must therefore be exercised in drawing conclusions as to the effect of Spartina "dieback" on the overall sediment budget of the harbour, particularly where there is evidence of erosion induced by human activities rather than vegetation change. Tentatively, Collier and Fontana (1996) conclude that - despite ongoing Spartina retreat, intertidal areas have exhibited an overall geomorphological or morphometric stability over the most recent 20 years. This research needs to be extended, in the context of the considerable recent progress in developing a GIS system to support extensive archaeological research carried out in Langstone Harbour since the early 1990s (Fontana et al., 2000).

The apparent closed sediment circulation established by Humby and Dunn (1975) for sand banks near the entrance to Langstone Harbour is an important observation, for it showed how transport could be rapid yet net sediment accumulation could remain stable. Assessment of the effects of any interference of the harbours, e.g. land claim and dredging is only possible when such circulations are understood. It is therefore suggested that other sand banks within the harbours should be similarly studied to determine whether closed or balanced circulation systems operate.

The main references for fluvial discharges typically date from the 1990’s and further fieldwork would be needed to assess whether discharges have increased with and if so the potential effect on sediment pulses.

Portsmouth, Langstone and Chichester Harbours