Vincent May

The features of the Dorset coast between Highcliffe in the east and Lyme Regis in the west are the result of thousands of years of marine and sub-aerial processes acting upon a wide range of geological materials whilst climate, sea-level and the human use and modification of the coast have changed significantly. Coastal processes act on timescales that range from the few seconds of a wave breaking to the many millennia of sea-level change. Similarly, these processes also occur on spatial scales of a few millimetres to the scale of the English Channel and beyond. For example, a pebble falling from the cliffs at Budleigh Salterton in East Devon several thousand years ago and found today on Chesil beach has probably made a journey equivalent to the distance from Earth to the planet Neptune. During all of that time, the combined effects of waves, tides and currents have moved the pebble up, down and along the shore and buried within the beach as well.

Waves
Most waves which reach the Dorset coast have been generated by winds blowing over the Atlantic Ocean. Their size depends upon the strength of the wind, its duration (the length of time the wind has been blowing at that speed) and the fetch (the distance of sea over which the wind blows). Waves from the southwest have a fetch in excess of 4000 km, whereas waves from the south or east-south-east have very restricted fetch (120 km and 240 km respectively). As a result, waves generated from the southwest are able to attain greater size than those from other, less-exposed directions. As waves travel away from their source, they spread out and become lower in height. The wavelength or time between wave crests is the wave period (in seconds). Southwesterly waves reaching the Dorset coast often have periods of 10 seconds or more, but waves from the southeast are much shorter (with periods of around 5 seconds). The height of a wave as it approaches the shore is affected both by its wavelength and by the slope and shape of the seabed. As waves travel into shallower water, they are affected by the slowing effect of contact with the seabed. The distance between waves is reduced, the waves become steeper and 'break'. Wave steepness (height divided by length; H/L) is usually between 0.03 and 0.06. Long waves (swell) are usually very flat, whereas shorter waves are much steeper and cause more problems for small boats.

Waves with heights over 8 m have been recorded along the Dorset coast, but most waves are much smaller. The significant wave height (H 1/3 or Hs) is the average height of the highest one-third of all waves that occur in a given period. Oceanographers and coastal engineers use Hs to characterize the wave conditions which occur in particular weather conditions and locations. The maximum wave height (H max) for a given period of time is very important for the design of coast protection and flood defence schemes, harbours and oil and gas offshore platforms. On 14th October 1976, a wave recording buoy off Bournemouth recorded a wave with a height (H max) of 8.3 m. The significant wave height was 5.0 m and the mean wave period 7.6 s.

Table 1 shows the significant wave heights in a series of storms at Ringstead Bay during 1989 and 1990. The maximum water level is the highest level to which waves ran up the beach and does not relate directly to significant wave height because these waves approached the beach at different states of tide and from different directions. The return period is the probability (or chance) that a wave of a given height will occur in a given time period. Therefore, the 2.90 m high waves could be expected to occur once in 60 months, but that does not mean that they will only occur once in 60 months. Although the individual storms had a return period not more than 1 in 5 years, the probability of the combination of such storms (i.e. their joint return period) was estimated as greater than a 1 in 1000 to 1 in 1500 year event, according to an analysis by HR Wallingford (West Dorset District Council 1995).

Waves along the coast are commonly small in height even when breaking. As waves travel into shallower water, they become higher. This is to conserve the energy contained between each wave as the waves slow and become closer. This changed or transformed wave height has been estimated by Halcrow as part of the work for the Shoreline Management Plan. Their estimates show that mean transformed wave height off Canford Cliffs is 0.2 metres with a 1-in-50 year value of 2.7m, increasing to 0.6m and 4.0m respectively between Bournemouth and Boscombe Piers and to 0.9m and 4.9m at Solent Road. Thus, breaking waves along the shore from Canford Cliffs to Hengistbury Head are usually lower than 1 metre in height, but they become gradually larger as the shelter from waves approaching up the English Channel is reduced. Large waves can cause enormous damage, but are fortunately rare. Large waves in Lyme Bay usually occur when high tides, strong onshore winds and low pressure coincide. Such surges have produced waves in excess of 6.5m (return period = 1 in 5 years) and 9m (return period = 1 in 50 years).

At Swanage, extreme wave heights have the following heights and return periods (Table 3).

The Shoreline Management Plan (SMP) provides detailed information about wave heights and energies which are very important for assessing both the movements of sediment alongshore and the wave energy to which beaches and coastal structures may be exposed. The longest wave periods for the period 1993 to 1997 (SMP) are associated with wave heights between 0.5 to 1.0 m. The 1 in 50 years extreme initial wave heights estimated by Hydraulics Research and Halcrow for the SMP increase towards the east of Poole Bay with the one in 50 year nearshore wave typically ranging from three to five metres.

Tides

Tides result from the effects of the gravitational forces of sun and moon as the earth and moon follow their individual orbits. The Dorset coast is affected by semi-diurnal tides, i.e. tides which occur roughly twice daily. Tidal heights change monthly and annually as well as over much longer periods depending mainly on the moon's phases and the Earth's elliptical orbit round the Sun. When the tide-generating forces of the Sun and Moon act in the same direction the solar and lunar tides coincide. Spring tides occur when the high tides are higher and the low tides lower than the mean. At New Moon the Sun and Moon are in conjunction (i.e. both aligned with the same line of longitude) and at Full Moon they are in opposition - in both situations the Moon is in 'syzygy'.

Tides along the Dorset coast vary from tides in the east with a small tidal range to tides further west which have a larger tidal range. East of St Alban's Head, there is a double high tide, but in Weymouth Bay and around Portland there is a double low tide. Poole Harbour has a small tidal range (1.8m at spring tides, 0.6 m at neaps) and a double high water, which means that the water is often above mean tide level for 16 out of 24 hours.

Outside the bar at Christchurch, according to Admiralty Tide Tables (1998), water level falls to about -0.7 m OD, and so the tidal range on Spring tides will be about 1.7 m.

Like the coast to the east, Weymouth Bay is micro tidal, with a range less than 2.0 m. In contrast, the mean tidal range at Bridport at springs is 3.5m with MHWS at +1.8m OD and MLWS at -1.7m OD (Nunny 1995). The differences in tidal peaks contribute to tidal currents, but in both Weymouth and Poole Bays, currents are generally weak. Tidal currents in Poole Bay have typical peak tidal current velocities less than 0.3 metres per second, although at Poole Harbour entrance flows are significant and increased by discharges from the harbour.

Water levels at the shoreline also depend upon the weather conditions. Onshore winds (especially from the south or south-east) can increase water levels by up to 0.2 m in Poole Bay. In contrast, offshore winds lower water levels by about 0.1 m. Surges (when water levels significantly exceed the predicted tide levels) are greatest when high spring tides coincide with low atmospheric pressure and onshore winds. For example, at a mean high water springs level of 0.6 metres OD, the frontage from Poole Harbour to Hengistbury Head has a 1 in 100 year chance of an extreme water level ranging from 1.82 to 2.4 OD metres. Within estuaries such as Christchurch and Poole harbours, the extreme water levels and the likelihood of widespread flooding are even greater if the rivers are also at high levels.

Sea level rise
Long-term changes in sea level result from changes in the volume of the oceans (eustatic sea level change) and changes in the height of land relative to the sea (isostatic change). Changes in ocean volume mainly result from:

• freezing and melting of the major ice bodies such as Greenland or Antarctica
  
• changes in ocean temperature.

Changes in land levels result from:

• uplift or lowering during major seismic events
  
• depression or uplift of the crust because of changes in the mass of ice or sediment on the land or seabed.

So, although sea levels are rising throughout the oceans due to changes in glacial ice and ocean warming, in some places sea levels are falling because the rate of crustal uplift is faster than the rate of eustatic effects.

Until the last decade, the rate of sea level rise along the Dorset coast was taken to be about 3 mm per year and then was revised upwards to 5 mm per year by MAFF in the mid-1990s. DEFRA has advised (2001) that 6mm per year should be used for future coast protection and flood defence proposals. This value refers to still water levels in the open sea. Water levels are affected in the short term by tides, wave conditions, atmospheric pressure and the effects of wind direction with onshore winds raising the level. Such pressure-induced effects are usually referred to as storm surges. If climate change brings about, for example, a greater frequency of low-pressure systems crossing southern England, then water levels would exceed current levels more often. This would not be attributed to sea level rise which relates to the wider water mass of the oceans.

If the 6mm per year value were used, then in 25 years time still water levels would be 150 mm (about six inches) higher. Any effects of pressure changes or wind would be superimposed on this.

Coastal erosion
The combined effects of tides, currents and waves as well as the sub-aerial effects of weathering and erosion working on the different rocks and structures along the Dorset coast have produced a complex and intricate coastline. This is well illustrated by the earliest charts (for example Lieutenant Mackenzie's 1787 chart of the coast and seabed between St Alban's Head and Abbotsbury). The changes in the bays and headlands reflect the strength of the rocks which occur at the coast as shown in Table 3.

Section of Admiralty chart showing the coastline in 1787 (DCC).

Cliff and platform erosion take place as a result of

• marine processes, (direct hydraulic impact on the materials, abrasion by sand and gravels carried by the waves, and solution),
  
• biological processes (by boring organisms which carve burrows in the platforms and weaken the rocks, burrowing animals on the cliffs, and the effects of plants in the soil and subsoil of the coastal slopes) and
  
• sub-aerial processes (weathering by freezing and thawing, wetting and drying, by the impact of wind and rain, by gullying and by landsliding in all its many forms).

Some cliffs collapse as large rock falls affect the whole or part of the cliff face. Others particularly in the weaker sands and clays in the eastern part of the county are affected by gullying and small surface slides of sand or clay.

The more recent aerial photographs (for example the aerial photograph of Worbarrow Bay) not only show the bay and headland pattern clearly but also display the different colours of the strata and the detailed features of the beaches, cliffs and land behind the coast.

Aerial photograph of Worbarrow Bay and Headland (DCC).

Aerial photograph of Old Harry Rocks (BUL).

The effects of the sea upon the detailed structures of the coast are especially well illustrated at Ballard Down (see the photograph of Old Harry). Here, well-developed stacks, arches and caves have formed where joints and faults cross the headland at Handfast Point. As the Chalk dips northwards, harder bands dip below sea level and weaker materials form more and more of the base of the cliffs. The harder pedestal band provides greater strength at the base of rock columns and channels water through the joints. The sea opens up well-developed, near-vertical joints at the foot of the cliffs, and small blocks in the upper parts of the cliffs fail because of very close jointing in the upper part of the cliffs. The joints are harder than the surrounding bedrock, but individual blocks within the joints are vertical and are more likely to drop out than the horizontal and better-supported blocks that surround them. The stacks are a result of the narrowing of the headland, erosion along the larger joints and the relative resistance of a pedestal band to erosion.

As the cliffs retreat, sediment is supplied to the beaches, although the clay cliffs contribute little as the small sized material was carried in suspension by currents along and offshore. In harder rocks, a platform is often formed, its width depending on the slope of the strata. At Kimmeridge for example, the platforms are mainly structural, i.e. their surfaces are formed by the upper surfaces of strata dipping gently seawards. In contrast, the very hard and steeply dipping Portland Stone has very narrow benches along the Purbeck coast but wider ones on the south side of Portland where the dip is much gentler. The cliffed coasts may form steep vertical cliffs such as those in the Portland Stone or much of the Chalk. In contrast, in the clays of the Kimmeridge and the Lias, there are extensive landslides. Although the sea removes both bedrock and debris from these cliffs, the most active landslides depend upon the effects of groundwater on their strength. Both physical and chemical changes within the clays make them less resistant to the forces from overlying rock and rainfall accumulating within the rock. Theme 3 describes in more detail the landslides and the processes which produce them.

It is also possible to judge the changes that have taken place at the resorts from the contrasts between the old photographs and prints and present-day photographs. The mainly sandy cliffs fed the beaches and there were beaches wide enough for bathing machines. However, as the resort grew the construction of promenades cut off the supply of sand from the eroding cliffs and Bournemouth, like many other resorts began the fight to retain its beaches. In the early 20th century the Bournemouth cliffs provided about 115 000 cubic metres annually (of which 80% was coarse enough to stay on the beaches). By the mid 1980s, this had fallen to only 4000 cubic metres per year, mostly from the cliffs between Solent Road and Hengistbury Head. More rapid erosion in the 1990s may have raised this to about 12 000 cubic metres per year (of which 55% may stay on the beaches). Given this reduction, it is not surprising that the coastal towns have taken such extensive steps to maintain their beaches (See Managing the Coast for more details).

Beaches
The beaches are formed mainly of sand and gravel (usually called 'shingle' in Britain). Some beaches, however, have much coarser material, including cobbles and even boulders. Beaches have three zones of wave action which affect their morphology and the beach transport processes:

• breaker zone, often quite wide because waves are different heights and break at different positions
  
• surf zone, produced by the breaking waves and
  
• swash zone, where waves travel up the beach

Shallow sloping beaches may have a series of low sand ridges or bars, for example at Bournemouth.

The action of waves constantly moves and sorts the different sized material. On some beaches, the sediment is sorted alongshore so that the coarsest material is located at the position with the greatest wave energy and the smallest in the lowest energy environment. For example, between Studland and Hengistbury Head the beach sediment changes from sand to shingle. The beaches also become steeper where the material is larger. Many of the small beaches between Weymouth and Kimmeridge lie between headlands and offshore reefs which restrict waves. In these semi-enclosed beaches, it is common to find the largest material close to the more exposed centre of the bay and the sediment to become finer towards each end of the bay. However, these patterns depend on the direction of wave approach and the coarser material can be shifted within the bays.

Sediment also moves up and down the beaches. Many Dorset beaches face the dominant waves so that waves approach the shore with their crest almost aligned to the line of the beach. In these circumstances, there is little movement of water alongshore and most waves wash up and run back down the beach slope (as 'backwash'). There is little longshore sediment movement. Cuspate forms develop which locally may be up to 35 m across, but are usually smaller. (These are well illustrated on many of the aerial photographs held in the database).

When waves approach the shore at an angle, they move sediment alongshore (longshore sediment transport). This depends on wave energy and the angle of approach to the shore. It is possible to estimate the potential longshore transport based on wave data. From this, coastal engineers can estimate the quantities of sand or shingle needed to maintain the existing beaches. Along most of the Dorset coast, beach sediment moves from west to east, but prolonged periods of east or southeast winds can reverse this trend. Surveys during such periods can mislead surveyors into believing that movement is from east to west more generally. Sediment moved alongshore builds up on the 'updrift' side of obstructions such as headlands, the toes of landslides, jetties and groynes. Groyne fields (e.g. at Bournemouth) or long single groynes such as at Hengistbury Head retain the sediment. Transport of sediment continues on the down drift side of the structures, but without a continuing supply of sediment, these beaches lose sediment, become smaller, and the cliffs behind them are eroded. Erosion downdrift of the last groyne in a groyne field is known as 'terminal groyne erosion'.

The volumes of material moved alongshore are considerable. For example according to the SMP, the Bournemouth to Southbourne frontage has net eastwards drift, the most reliable estimate being 40,000 cubic metres per year, based on a nine-year record. In contrast, between Durley Chine and Branksome Chine, westward drift of up to 51, 000 cubic metres per year has been estimated. Because of the uncertainties involved in measuring the volumes of sediment in beaches, this value may vary by plus or minus 12,000 cubic metres per year. The potential drift has been estimated as greater than this for some beaches. For example just west of Flaghead Chine at Poole HR Wallingford estimated that up to 1995 potential drift to the southwest was 70,000 cubic metres annually. To maintain this quantity of sediment transport, sediment must be supplied either from the east, i.e. from Branksome, from the beaches, or from cliff erosion (now stopped by the sea walls) or be nourished artificially. To put this into context, 50,000 cubic metres is the equivalent of a beach 25 m wide, 2 m deep and 1 km in length.

In west Dorset, longshore transport between West Bay and Cogden Beach was estimated at 8000 cubic metres per year between 1974 and 1984 by HR Wallingford. After 1982, this increased to 14 000 cubic metres per year. West of Seatown the annual input of gravel from landslides has been estimated at 5000 cubic metres per year.

To build the immense barrier beach at Chesil beach required very large quantities of sand and shingle. Its volume is estimated at between 15 to 60 million cubic metres (Carr, 1980). The beach extends from Chesilton in the east, where it ends against a seawall and the cliffs of the Isle of Portland, to West Bay. Opposite the Fleet, Chesil Beach is between 150 m and 200 m wide, but it is narrower both near the cliffs in the west (e.g. 35 to 60m at Burton Bradstock) and at Chesilton (between 40 and 54m). The height of the ridge increases progressively from about 7m at Abbotsbury to a maximum, about 14 m above mean sea level, at Chesilton. The beach continues below the sea to depths of about -18 m OD at Wyke Regis and -11 m off West Bexington about 270 m offshore.

About 98% of the material is flint and chert probably from a range of local sources, but the remaining 2%, for example Triassic quartzites, probably come from the south-west, with 95% of quartzites from the Budleigh Salterton Pebble Beds (Carr and Blackley 1969). Not all the material is rounded, the flint and chert pebbles becoming more angular with depth (Carr and Blackley 1969). The massive pebble and cobble deposits are concentrated in the exposed, i.e. sub-aerial, part of Chesil Beach. A former coastal platform, largely planed-off bedrock, continues underneath the Chesil shingle barrier. Its junction with the slopes inland and an associated ancient pebble and cobble storm beach (Carr & Blakely 1973) occur at about -15 m OD opposite East Fleet, and at comparable depths as far west as Abbotsbury. The Fleet is infilled with silt, sand, pebbles and peat, some of which have been dated.

Many people have commented on the different sizes along Chesil beach. At Chesilton, the mean long axis is about 50 mm, 35 mm opposite Portland Harbour and rather under 25 mm opposite Herbury Point (Carr 1969). The common explanations for the graded sediment of Chesil are that

• there is a continuous size change along the beach
  
• sorting is by size and shape
  
• rates of pebble movement depend on pebble thickness
  
• different wave energy of storms from the south west and the east cause the sorting and
  
• different depths of water offshore and the available energy are a fundamental cause of the pebble distribution and sorting.

However, many local variations in the patterns probably result from the ways in which pebbles are moved by waves approaching the beach at different angles and the burial and re-emergence of pebbles depending on wave run-up and overtopping of the beach ridge.

Understanding of the origins and development of Chesil Beach has been helped by description of the complex environmental history of the Fleet. Within the Fleet, clays, silts and sands occur from -15m OD upwards to about 3.8m OD. where there are often thick Phragmites peat layers (dated at c.5000 years BP). All this deposition took place in a lagoon or estuary behind an earlier barrier beach. Mean grain particle size decreases westwards, indicating that energy lessened westwards along the Fleet. At West Bexington, foramifera and ostracods show that a tidal, near-marine, water body existed well to the west of the modern Fleet around 4000-5000 yr BP (Whitaker pers. comm.). The western Fleet was thus behaving more like an estuary than a lagoon. At the eastern end of the Narrows, peat deposits, with a high pine pollen content, dated at c.6200 years BP lie beneath the landward side of Chesil Beach at a depth of c. -5.3m. Layers of peat at depths of -3.00m OD, -3.60m OD and -4.32m OD have been dated between 4540 plus or minus 70 years and 4840 plus or minus 70 yrs BP (Coombe 1996). Two samples in the East Fleet (Coombe's cores 25 and 29) at -3.00m OD and -3.15m OD were dated at 3820 plus or minus 70 years and 4110 plus or minus 60 yrs BP respectively. The pollen in a sand sample beneath the lagoonal sequence included a high frequency of Pinus, together with Quercus, Corylus and Betula. All the cores indicate salt marsh above the sands. Throughout the Fleet, narrow shell beds rest on top of each peat bed. This brackish-marine lagoonal phase is less than 5000 years old and continues to today.

The generally accepted model for the longer-term evolution of Chesil Beach suggests that a forerunner probably existed as a bank well offshore of the present beach about 120,000 years ago. During the last glacial, when sea-level was about 130m lower than today, gravel deposits, probably comprising material from the Portland raised beach, solifluction deposits, river gravels and fluvio-glacial deposits, were dumped on the floor of Lyme Bay by late Devensian meltwaters. As these gravel deposits were eroded by rapidly rising sea levels at the end of the Devensian (20000-14000 years ago), waves transported the coarse materials landwards as a barrier beach. Closer to the modern coastline, the transgressing Chesil beach overrode existing sediments. A shallow lagoon which became the Fleet was rapidly filled with silt, sand, pebbles and peat from about 7,000 yr BP to 5,000 yr B.P. Between 4,000 to 5,000 B.P. the modern high shingle and cobble Chesil Beach formed a short distance seaward of its present position (Bray, 1990), probably on an existing sand-shingle barrier beach (May, 2003). Fossil landslide deposits (boulder aprons) identified some 2-3km offshore from Golden Cap (Brunsden & Chandler 1996) may indicate the extent of past cliff erosion. Relict cliffs to the west, abandoned by falling early Devensian sea levels, were reactivated by marine erosion and large quantities of gravel were transported eastwards, to feed and enlarge the new Chesil Beach (Bray, 1990). Bray's model assumes that intermittent pulses of gravel bypassed headlands such as Golden Cap, at intervals of 30-50 years, most recently at Golden Cap between 1949 and 1962.

Dunes and estuaries
When sand is eroded from cliffs or beaches are wide enough to allow sand to dry, the wind can transport this sand to the backshore or on to the top of cliffs. Much of the cliff top at Bournemouth for example is covered with the sand of old small dunes and patches of new dunes colonised by Marram and Lyme grass still appear here. Where there are wide sandy beaches, wind ('aeolian') transport carries sand inland where it forms small 'embryo' dunes and then gradually builds into larger dunes. The Dorset coast has only one large area of dunes, at Studland where accretion has taken place in cliff-foot dunes, spit and ridge development since before the 16th century. For the earliest detailed map of this area, see Treswell's map of Studland Parish dated 1585-6.

Ridges of sand dunes have built up along the coast to produce a landscape of dunes separated by lower brackish or fresh water environments.

In contrast, many of Dorset's rivers flow into estuaries. Some are very small and mostly reclaimed: for example the Brit at West Bay. Poole Harbour in contrast is large (over 3000ha) with extensive mudflats and salt marsh. Much of the salt marsh and mudflat area has been reclaimed around Poole and in the lower valleys of the Frome and Piddle rivers. The greatest growth of the salt marshes took place in the early 20th century when Spartina anglica (a vigorous hybrid from S. maritima and S. alterniflora first identified in Southampton Water) colonised Poole Harbour. Many of the charts (and the more recent aerial photographs) of Poole Harbour show the changes in the salt marsh and the channels particularly well. For more details, see Dorset's Marine and Coastal Habitats.