CHAPTER 2 THE SCIENTIFIC BACKGROUND TO THE PROJECT WORK

The authors of this report believe that it is essential for the reader to be versed not only in the relevant legal and administrative frameworks (Chapter 1) but also to be aware of the development of and the techniques used to investigate archaeology and the palaeo-environment (section 2.1) as well as landslide investigation (section 2.2). As a result before the case study areas are provided in Chapters 4 and 5 the following Chapter presents an overview of the most widely used investigation and study techniques in these fields.

2.1
AN INTRODUCTION TO ARCHAEOLOGICAL AND PALAEO-ENVIRONMENTAL STUDIES AND THEIR APPLICATION TO THE STUDY OF COASTAL CHANGE

2.11
Introduction to archaeology

The project objectives for archaeological and palaeo-environmental studies

Task 1 of this LIFE project set out:

In pursuing this objective three aims were identified at the outset of the project. The first was to establish through the three national study areas in Great Britain, France and Eire the archaeological and palaeo-ecological potential of down-warped areas of the European coastline. The second aim sought means of linking the chronology of past environmental changes to current processes and future scenarios. The third aim sought to examine those procedures in the three national study areas where geo-archaeological research had been linked to coastal planning and coastal management decisions.

Since the inception of the project a fourth aim has sought to formulate best practice principles which should ensure the sustainable management of geo-archaeological resources and might improve their recognition in environmental assessment. In respect of the last aim the current operation of European Directive 85/337/EEC (05 No L17S, 5.7.85) was cited.

Three central questions which have been defined by the partners during their investigations.

  1. What archaeological and Quaternary palaeo-environmental contexts in the three study regions offer the best potential for resolving problems concerning the nature, scale and pace of coastal change and the history of coastal instability?

  2. How can coastal sites of archaeological and palaeo-environmental value be recognised and protected from loss, threat or oversight and how can they be ranked in terms of scientific or cultural importance?

  3. What methodologies in use in the study regions are best suited to the evaluation and investigation of coastal archaeological and palaeo-environmental sites?

The following text presents a background to the nature and purpose of archaeology and the ways in which this subject, as a science, has been applied to the study of the coastal zone. It also explains the development of related palaeo-environmental studies before setting out the findings of the LIFE investigation into these three basic questions. The application of these enquiries to the principal Task 1 objective is presented in Chapter 7.

The definition of archaeology

The term archaeology literally means the study of ancient things and it is essentially applied to the past works or artefacts of man. Before archaeology emerged as a science it was preceded by certain other allied interests in the fields of the arts. These pre-cursive interests were nurtured by those who collected portable antiquities such as sculpture, terracottas and coins. artefacts such as these were generally sought as a means of understanding early art history while knowledge of ancient or classical architecture was pursued in much the same manner. None of these activities can be described as archaeology although all of these topics contribute to archaeological knowledge. Early practitioners in the field of collecting include Nabonidus King of Babylon and his daughter Belshalti-Nanner. In the mid 6th century BC both dug in the ruins of the ancient Chaldean city of Ur and produced a personal museum of antiquities.

A vital distinction between the collection and study of portable antiquities and the study of archaeology is drawn at the investigation of context. While some archaeologists may be solely interested in the context in which archaeological material is found, or in the environment in which past communities were living, others may specialise in the study of the intrinsic character or typology of particular groups of artefacts such as pottery, stone tools, leatherwork or past decorative arts. Writing in 1967 Glyn Daniel rightly commented that there are probably as many different kinds of archaeologist as there are doctors and engineers while, given that there notably fewer archaeologists, the difference between them may seem more extreme.

Many and differing definitions have been devised for the raisonne d'Ítre of archaeology yet all seem agreed that a common aim of the archaeologist is to find explanation in the patterns, choices and experiences of past human communities while seeking to understand the changing interaction between these activities and the behaviour of the natural environment of the planet. Given that ninety-nine percent of human activity was carried out in the prehistoric period, which is literally before the development of written history, the evidence sought by the archaeologist is highly varied and frequently difficult to interpret.

The first steps towards the study of archaeology

During the 5th century BC a young Greek explorer, Herodotus, toured Asia Minor and Egypt where he examined ancient cities, temples and pyramids. Herodotus set out with the specific aim of improving and disseminating knowledge of the past. This he achieved not only by making measurements and drawings of ruins and monuments but also by recording gossip, stories and folk-memories of elderly or knowledgeable inhabitants at all the places he visited. All of this information he set out in a single book beginning with the words:

'This is the history of Herodotus of Harlicarnassus, published in order that what has happened may not be forgotten of men through the passing of time...'

Because Herodotus had resisted the desire to personally acquire antiquities and had, instead, been motivated by a desire to search for and disseminate new knowledge, he has been commonly claimed to be the father of archaeology.

A Chinese text attributed to 52AD has demonstrated that other early enquiries into the prehistoric past had also pursued patterns, meaning and explanation (Lowie 1937). This oriental document shows that Chinese scholars of the 1st century AD had already made a detailed study of ancient artefacts and had identified to their satisfaction a three age progression of past technical development. This led from a Stone Age to a Bronze Age and finally to an Age of Iron. This oriental achievement pre-dated by some 1800 years the European recognition of the three age system as propounded by the Scandinavian scholars P.F. Suhm and Christian Jurgensen Thomsen around the close of the 18th century.

The emergence of European archaeology

As a disciplined subject leading to sustained study in the western world, archaeology owes its development to European antiquarian interests of the 18th century. By this time antiquarianism had grown to be a popular and genteel interest which was largely focused on the collection and ordering of classical antiquities. In northern Europe these activities were inextricably mixed with the 'Grand Tour'; a protracted cultural indulgence which carried the privileged and the dilettante to view the crumbling grandeur of classical Greece and Rome.

Antiquarianism of this kind had little to offer the scientific world for its motives often lay no deeper than joy of the chase or in the questionable merits of competitive ownership and exhibition. At best, the acquisition of classical antiquities in northern Europe could be viewed as a cultural matter in which education and experience could be gained through physical contact with works of the past. Antiquities of this kind were first assembled in private collections but the philanthropic ideals of the 19th century later brought many of these assemblages into the public domain where they were commonly displayed in national and municipal museums and art galleries. The opening of the Royal French collections for public enjoyment in the Louvre was an early landmark in this process.

For the purpose of the present study it is particularly important that these earlier idiosyncratic interests are fully understood. This is because early misplacement amongst a fickle array of cultural interests has commonly denied recognition of the scientific role of archaeology. This early misapprehension concerning the purpose of archaeology has left a legacy of division and incomprehension which still impedes the science of today.

In Europe, where legislative and financial instruments have sought to ensure the care and sustainability of vulnerable archaeological remains, it is common to find that while an archaeological site may be viewed as an environmental asset subject to the protection of statutory legislation, the long-term security of the site archive (its artefacts, samples and records) may be far less secure. The latter task has often been relegated to a non-statutory function which may be adopted or discarded according to the disposition of individual local authorities. Where primary and irreplaceable archaeological evidence has been sought in the present study it has been found that some of this material has perished during episodes of decline or neglect in local museums. Such inequalities can easily arise as a result of loose and inconsistent cultural or 'arts' policies which are ill-prepared for the less prepossessing task of analysing and storing those scientific collections which may offer little aesthetic appeal. These policies can show marked variance across particular local or regional government boundaries. In the narrative presented here it will be seen that precarious and unsustainable positions for the safe-keeping of primary archaeological evidence can be born of a primary misconception concerning patronage of the arts. This impediment owes its origins to the differing cultural interests of the 18th century.

The dual development of archaeology and earth sciences during the 19th century

While patrons gathered classical antiquities in the 18th century, a small number of more scientifically-minded European antiquaries turned their attention to the collection and examination of local antiquities. This occurred at a time when the advance of the industrial revolution had brought widespread improvements in the form of new road, rail and canal networks to the landscape of 19th century Europe. These great engineering works soon led to the unearthing of a notable array of new archaeological and geological discoveries. These finds were destined to evoke a barrage of new and penetrating questions concerning both the natural and historic environment. Such questioning was to transform antiquarian curiosity into a disciplined enquiry which was soon to graduate to a publically-funded earth science.

The first questioning arose after newly-dug cuttings for railways, and gravel excavations for improved roads, had exposed fresh transects through the bedrock of the European countryside. In these Charles Lyell recognised evidence of a vast new timescale during which a processual history of natural forces had gradually shaped and sculpted the past and present landscape of the planet. Lyell identified these forces as those of erosion, transportation, deposition and volcanism and he was at pains to emphasise that the slow but relentless progress of these forces could still be witnessed today (Lyell 1830).

The change from antiquarianism to archaeology came in the wake of Lyell's discoveries when new and perceptive questions arose concerning the context or environment in which antiquities were being found (Lyell 1860). These questions were first posed when a new era of exploration and scientific enquiry began to enthral the drawing rooms and lecture theatres of many of Europe's major 19th century cities. The most ancient of these discoveries came from caves, rock-shelters and valley gravels where animal and human remains were being unearthed by miners and quarrymen who were disturbing deposits which were perceived to be of 'Ice Age' or 'antidiluvian' date (see Figure 2.1).

The discovery of the primitive human skull in the Cro-Magnon rock shelter at Les Eyzies in 1868 was an important landmark in this quest for new knowledge. Questioners were soon demanding an improved understanding of earth sciences while some raised new and profound doubts concerning other allied areas of accepted knowledge (see Figure 2.2). These ripples of disquiet soon unsettled the simple biblical explanation which had long been offered for both the genesis of the earth and origin of its human population. The conventional explanation was based on a theoretical timescale conjectured from a biblical genealogy. This had been devised by Irish theologian Dr James Ussher in 1597. This had since been appended to the authorised version of the bible and, for many, had gained the status of established fact.

Quarrying
Figure 2.1 New discoveries made during mining and quarrying in 19th century Europe evoked the development of the geological and archaeological sciences. This discovery in the Dream Cave, Wirksworth, England, raised questions concerning the nature, scale and pace of past natural processes (after Boyd Dawkins 1874)

The period 1850 to 1865 finally brought a loss of innocence to mankind for this was a time when the study of earth's history and the origins of human life were finally laid open to rational scientific examination. Lyell's revolutionary study of processual geology published in 1830 was followed in 1859 by the publication of Charles Darwin's Origin of Species. Darwin's new principles of evolution were promptly adopted by Edouard Lartet who, in 1860, was quick to point out that evidence of early man could be traced amongst the prehistoric animals of the Pleistocene period. His publication was promptly followed by the work of T.H. Huxley whose thesis on Man's place in Nature was published in 1863. By the time Darwin's Descent of Man had been published in 1871, Archbishop Ussher's crude calculation for a genesis of the earth in 4004 BC could be clearly seen to be naive and untenable although this hypothesis has still managed to retain a few exponents.

Discovery of Cro Magnon rock shelter at Les Eyzies
Figure 2.2 The construction of Europe's 19th century railway network brought fresh discoveries of early human remains and these evoked a new awareness of prehistoric time-scales and stratigraphy. This engraving shows the discovery of the Cro Magnon rock shelter at Les Eyzies in the Dordogne, France in 1868. The rock shelter at A is masked by colluvium B.

The recognition of archaeological evidence as a means of dating past geological and geomorphological processes

In the context of this present study we find that some of the most significant questions concerning the dating of natural erosion processes had been posed as early as the year 1800 when the English antiquary John Frere came to offer potential contradictions to the chronological scheme of Archbishop Ussher. In a gravel pit at Hoxne in Suffolk he had observed the accidental discovery of flint hand-axes or palaeoliths along with a variety of bones of extinct animals. All of these items had been sealed beneath a bed of marine sand and overlain by more than two metres of gravel.

In his report for the 13th volume of the new journal Archaeologia, Frere postulated that these stone tools had "been used by people who had not the use of metals and were active during a very remote period indeed; even beyond that of the present world. Having carefully examined the overlying deposits Frere questioned whether the different strata were formed by inundations happening at distant periods and bringing down in succession the different materials of which they consisted. " Given that the ground of the district was virtually flat it seemed to him that extremely powerful natural processes such as torrents must have produced these deposits and shaped the present configuration of the land at a very remote period in the past. By looking to torrents for his explanation Frere was still influenced by contemporary biblical explanations which sought to account for geological and fossil evidence in terms of a 'fluvial' event or divine flood. None of Frere's new stratigraphical evidence was effectively grasped until more than half a century later when the gravel pits at Hoxne were re-examined and his astute observations finally won full scientific recognition (Prestwich 1860).

Penned two hundred years ago, John Frere's report on Hoxne is a significant landmark in this present study for it first established a role for the use of archaeological evidence in the study and dating of natural processes. This alliance of archaeological and geomorphological studies was consolidated in the mid-19th century when the works and discoveries of the Reverend Boucher de Perthes, Georges Cuvier and John Prestwich provided tantalising new material which was to fuel new debate on the natural sciences. Alexander von Humbolt's Cosmos, published in 1845, presented a new scientific perception of the composition of the world and its place in the universe and this was an appropriate companion to Charles Lyell's 'Principles of Geology', first published in 1830. The works of these two men and the observations of Charles Darwin soon turned world scientific attention to a new and vastly extended timescale for the formation of the planet. Many of these revelations were now presented through the new science of geology where it was soon recognised that the recent segment of the earth's history encompassed a sequence of glacial and interglacial episodes which coincided with the evolution of human life. While geologists sought to refine their understanding of stratigraphy in cliffs, wells and mines; antiquaries, in their new role as archaeologists, now sought datable evidence of human activity in the cave-earths, valley gravels and brick-earths of the north-west European plain.

By 1880 a significant melding of these two lines of enquiry had been orchestrated by James Geike in his study of 'Prehistoric Europe'. Geike reviewed the typological periods which archaeologists had used to order their artefacts. He then applied this evidence to the dating of the major natural processes and events which could be identified during and after the Pleistocene period. The principles of processual geology established by Charles Lyell had demanded a new and greatly extended temporal vision of the earth's history and Geike now saw archaeological evidence and palaeo-environmental evidence as two virtually interchangeable means of calibrating the more recent episodes in the new geological timescale.

In Geike's view, both strands of evidence led to a single explanatory goal. Where submerged forests could be detected on the European coastline he found that decayed vegetable matter was indicative of past climatic conditions which differed from those of the present. The survival of a submerged dug-out canoe and the presence of flint tools in such a deposit provided a clear indication of its date. Where peat mires and lake-beds were disturbed in the process of agricultural improvement, Geike was prompt to point out that these deposits had accrued over substantial periods of time and that their dead aquatic fauna and their archaeological contents were potential indices of past climatic and environment change. This proposal was supported by specific evidence gathered in Scotland and Scandinavia where Geike was able to use tree and plant remains to demonstrate that a phase of 'Arctic' vegetation had been succeeded by increased dryness eventually leading to the Atlantic climate which dominates the north-western European seaboard today. In Scandinavia Blytt (1876) and Sernander (1908) were pursuing similar lines of past environmental reconstruction. All of these pioneer studies were eventually consolidated by F.J. Lewis and the Swedish botanist G. Samuelsson who were responsible for the final definition of the past climatic phases of the Holocene period.

2.12
The development of the link between archaeology and the study of coastal change

Early applications of archaeology and palaeo-ecology to the study of the European coastline.

By the close of the 19th century a number of the questions raised by Charles Lyell and James Geike had been taken up by other pioneers of the earth sciences.

Acting as an archaeologist, a botanist and geologist, as well as a banker and politician, John Lubbock (later Lord Avebury) was essentially a polymath who had been drawn to Ice Age studies by the Darwinian debate of the later 19th century. His interpretation of man's experience of the glaciations he had set out in his study of Prehistoric Times first published in 1865. Being particularly conscious of the power of ice, melt-water and fluvial forces he returned to the topic of natural processes and landscape evolution in 1896 when he published his explanation of 'The scenery of Switzerland and the causes to which it is due' (1896).

Eight years later, in 1902, Lubbock had turned his attention to 'The Scenery of England' where his knowledge of past natural processes drew him to examine raised beaches, shore-cut platforms, rates of cliff erosion, the behaviour of coastal sands, long-shore sediment movement, the accretion of spits and questions concerning a past rise in sea-level and the consequent drowning of river-mouths. Throughout his review of coastal processes, Lubbock was quick to seize upon historical, archaeological and palaeo-ecological evidence which could help to fix or calibrate past events of coastal change. At Great Yarmouth on the coast of Eastern England he was able to use records of AD 1008 to account for the formation of a sand-spit while at Orfordness, on the same coastline, he was able to allude to a chart of the late 16th century (see Figure 2.3). On the shore of the Solway Firth he weighed accounts of past fordings of the river against the evidence offered by the configuration of Hadrian's Wall. Here, he concluded, no significant change in sea-level or land elevation had taken place in post-Roman times.

The shingle ridges in the spit at Ofordness, Norfolk, England, as recorded by John Lubbock in 1902
Figure 2.3 The shingle ridges in the spit at Ofordness, Norfolk, England, as recorded by John Lubbock in 1902

On the bed of the North Sea Lubbock observed that the remains of bear, lion, elk, rhinoceros, hippopotamus and elephant had been found. He also noted that some 24 miles offshore from the Wash a deep channel was discernible on the Admiralty Chart. Lubbock realised that the present Fenland rivers were incapable of cutting such a channel below the level of the present sea unless, of course, that level had been lower in earlier times. If such a lowering had occurred, the seabed channel could be readily explained as an extended river valley. Drawing together these two strands of evidence Lubbock concluded that much of the North Sea had been a great plain accommodating the combined courses of the Rhine, the Thames and the Humber. Conjoined together, it seemed that these rivers had once flowed northwards to enter the Arctic Ocean. These observations, combined with those of Geike, accorded with Lyell's processual history of the Earth and they firmly established the concept of an ever-changing geography in Europe where man had seemingly become a recent temporal bystander (see Figure 2.4).

The early recognition of coastal changes in the Solent region of the United Kingdom

Part of James Geike's reconstruction of the submerged landscape and river system on the Atlantic seaboard of Europe
Figure 2.4 Part of James Geike's reconstruction of the submerged landscape and river system on the Atlantic seaboard of Europe, published 1881

Working as a geologist in southern England, Clement Reid examined the particular topographical configurations of the present hills, valleys, river systems and coastline. These he sought to reconcile with past processes of erosion, transportation and deposition. A notable element of Reid's work dealt with the Wessex chalklands and the Solent region where the evolution of the coastline was an exemplar which now begged explanation and debate. Both Charles Lyell and James Geike had raised the question of submerged forests on the English coastline and these now drew Reid's specific attention (Reid 1913) (see Plate 2a). Here he realised that the investigation of plant macro-fossils might establish the climate and environment in which these forests had flourished. Reid concluded that notable post-Glacial submergence was particularly discernible in southern England and that in some cases historical accounts gave credence to a proposal for a relatively recent date for the formation of some major geographical features. This included the detachment of the British Isles through the formation of the Dover Strait and the severance of the Isle of Wight (Reid 1905).

Unfortunately Reid's use of historic documentary sources was not always sufficiently critical to achieve objectivity between the weight of field evidence and the inferential evidence offered by the written word. In the case of his Solent study, Reid was too ready to accept that a classical description of a tied island called Ictis actually applied to the Roman island of Vectis (Isle of Wight). While this presumption had been popularly quoted since the 16th century (Camden 1586, 1637 etc.), the substance of the classical text presented clear evidence that the tied island in question served a Late Iron Age tin-mining community somewhere on the coast of Cornwall (Tomalin 2000).

Submerged prehistoric tree
Plate 2a Submerged prehistoric forests were first recognised as evidence of sea-level change in the mid 19th century. This tree exposed at extreme low tide on the shore of the Eastern Solent is an example, which has been sliced for tree ring dating. Isle of Wight , UK

2.13
The use of archaeological and palaeo-ecological evidence in the study of changing environments

The first investigations of palaeo-environmental evidence

While the pioneering studies of the late 19th century had defined at least one common objective which might be shared by the study of archaeology and earth-science, the recognition of a causal link between the activities of man and the character of his natural environment had yet to come. The first interests in such a link arose in the early 20th century. Some of these began when the barren setting of a number of ancient North African and Middle Eastern citadels prompted questions concerning their past environment and subsistence base. Expressed simply, the question was how could great cities thrive in the middle of a desert - unless, of course, there had been a time when such a landscape had not been desert.

Prior to the early years of the 20th century, comprehension of past Pleistocene and Holocene environments was largely dependent upon stratified sequences of differing geological deposits and their macro-fossil contents (see Figure 2.5). The first step towards a new analytic approach was achieved in 1888 when Tryborn identified sub-fossil pollen in the sediments of a Swedish lake. Contemporary investigations in Germany, by Carl Weber, achieved similar results while the Danish archaeologist G. Sarauw turned his attention to ancient pollen preserved within a submerged peat bed in the vicinity of Copenhagen (Sarauw 1897). Working in Sweden in the closing years of the 19th century Lagerheim was the first to draw upon these early experiments to form a recognised research discipline. This line of investigation was taken up by Alexander von Post (1916) who was the first researcher to construct diagrams to show different levels of pollen content within the same deposit. By 1928 the new analytical technique had been improved by Gunnar Erdtman (1928) who found that he was able to retrieve prehistoric pollen from an ancient buried channel of the Thames at Ebbsfleet in Kent.

During the 1930's palynologists working in Denmark and the United Kingdom seized upon a variety of peat deposits to extract ancient pollen grains in sufficient quantities to attempt the statistical reconstruction of past environments (eg. Godwin 1934). These deposits were considered to be especially valuable because von Post had realised that while peat mires usually accrete at a slow and steady annual rate, the arrival, settling and preservation of airborne pollen grains seemed to progress in tandem with this process. By 1941 evidence had been gathered from a broad geographic array of peat deposits covering a wide chronological range. The new study of pollen, palynology, was now sufficiently advanced to propose a standard zonation scheme which might characterise the major changes in the European climate since the close of the Pleistocene some 20,000 years ago.

Once palaeobotanists had refined the main characteristics of their Holocene pollen zones, the interpretation of pollen diagrams was viewed by some as a means of recognising a particular time zone and hence obtaining a relative date. Such chronological indicators were earnestly sought before the advent of radiocarbon dating in the 1950's. While this was a helpful way of dating the accretion of certain prehistoric peats and detecting the principal climatic changes of the past, this approach misread the full potential of pollen analytical studies. The new study was not only capable of revealing a detailed environmental picture of a particular site at a specific time but was also capable of revealing a sequence of environmental changes occurring over an extended period of time (see Figure 2.6). This gave pollen analytical studies a status of their own and it embued palaeo-environmental deposits with a scientific value which stood alongside the cultural remains which had been so highly regarded since the 18th century.

The particular palaeo-environmental potential of buried soils (palaeosols)

It was during the 1960's that attention specifically turned to archaeological sites as source of palynological information. Prior to this date, peat had been perceived to be the primary medium for the retention and preservation of ancient pollen but new extractive techniques developed by Professor G.W. Dimbleby at Oxford now identified buried soils and lake sediments as further sources of pollen evidence which might elucidate the environmental and climatic changes of the past (Dimbleby 1962). The burial of ancient soils usually took place where mounds or other earthworks had been deliberately built in prehistoric or later times. This meant that where suitable acidic soil conditions had prevailed, every upstanding earthwork could be viewed as a potential and hidden repository of a palynological record.

While pollen studies were enhancing the palaeo-ecological value of certain acidic peats and soils, the search was mounted for a means of extracting palaeo-environmental information from the wide array of alkaline soils in which the pollen grains were unable to survive. This investigation sough to overcome the problem posed by the sparsity of acidic pollen-bearing deposits in areas such as the south of England. This paucity of acid soil conditions had denied a record of past environmental events to very large areas of the European landscape.

An important result of the search for new palaeo-environmental indicators proved to be malacology, the study of terrestrial land-snails. Early experiments by archaeologists sought to characterise particular buried soils by identifying the number and nature of ancient snails preserved in a specific deposit. Recognising that many snails were naturally restricted to narrow climatic conditions and specific habitats, scientists realised that it was technically possible to identify the snails and then calculate the nature of the past or lost habitat (Evans 1971).

Unfortunately the information provided by the snails could not match the fine temporal sensitivity offered by the annual accretion of the pollen in the peat mires. Statistically, the requisite quantities of snail shells for analysis demands large soil samples. The ability of some species to burrow deeply into the ground could also impede or contaminate stratigraphical differentiation. Nevertheless, it was realised that colluvium, ie.'slope deposits', could be a highly rewarding source of steadily accrued palaeo-mollusca. Analyses of this kind have since assisted in understanding the environment and date of some of the major coastal changes and landslide events which have been examined in this study.

At the coastal site of Limpet Run in Sandown Bay, Isle of Wight, UK (Study Area P5), buried soil and a slope deposit of calcareous hill-wash was found exposed in the face of the cliff (Figure 2.7, see also Figure 7.43). These deposits had accrued to a depth of 2m and they comprised nine specific layers or contexts which were generally rich in land-snails (Allen 1994). Above the Greensand bedrock was a barren Late Glacial marl (context 9) which was apparently the product of melt-water or solifluxion. At this deep level negligible numbers of land snails could be found but their numbers eventually increased as the soil became more chalky in context 8. The snails at this level were representative of mature deciduous woodland at this coastal location. Their habitat was soon followed a distinct clearance horizon represented by context 7. Here the shade-loving snails gave way to Vallonia sp which favoured cleared or tilled ground.

The snails in the overlying colluvium in context 6 produced clear evidence of open countryside with the species Pomatias elegens and Trichia hispida suggesting the survival of some limited shade while arable activities progressed. Similar conditions were evident in context 5 where the appearance of the snail Helix aspersa occurred. This snail is a known Roman introduction and it appears here in a context which is also dated by the occurrence of Roman pottery. In the post-Roman slope deposits the evidence of an open landscape continues while the occurrence of Vitrea contracta suggest that the grass at this site is now being cropped at a very short level in the manner produced by grazing animals.

The analysis of snail shells at this cliff-top site has produced valuable evidence of coastal and humanly induced change while the archaeological material has secured the timescale which has been vital to the dating of these events. It is interesting to observe that colluvial deposits continued to accrue in well stratified malacological horizons until sometime during or after the Roman period. The upper land-surface represented by contexts 3, 2 and 1 showed little time-depth and this may be attributable to the retreat of the cliff-face which, by this time, had exposed all of these terrestrial deposits to collapse and loss. The environmental history conveyed by the snail analysis also suggests that the pastoral use of this coastal landscape in post-Roman times had been an significant inhibitor of erosion.

Limpet Run

While palynological and malacological studies of buried soils have been highly beneficial to this LIFE study it is important that the rarity and vulnerability of these deposits should be stressed. Effectively, continuous stratigraphical sequences capable of producing a long and finely calibrated record of environmental change are only to be found in peat mires and lake beds. Sometimes, concealed deposits of peat are to be found in the fills of palaeo-channels in rivers and creeks but they are notably absent from large sectors of the European coastline. Where calcareous slope deposits or colluvium survives undisturbed, this can provide a valuable yet coarser record of environmental change. Deposits such as these also deserve protection and investigation.

In general, single horizons of buried soil usually offer little evidence of sustained or incremental changes in the environment but they are capable of providing individual 'snapshots' of particular events in the past. One of the most productive sources of evidence of this type are ancient layers of humus which have been suddenly buried or overwhelmed during a rapid event. Such actions can instantly transfix and seal the current horizon of vegetation, preserving it from the future effects of erosion and degradation. A sudden marine inundation or an episode of landsliding can occasionally be a suitable mechanism but, for the main part, scientists are dependent upon past human activity to preserve soils of this type. Artificially buried soils, suitable for analysis, are also difficult to locate. Where these soils have survived in appropriate and potentially informative coastal locations, it has not been unusual to find that they have been thoughtlessly disturbed and scientifically ruined by recent agricultural processes.

With the exception of deep colluvial deposits in the floors of ancient dry valleys, buried soils of the prehistoric period do not normally survive unless they are securely covered by a thick protective blanket of raised earth. A particularly valuable resource of this kind was created during the European Neolithic and Bronze Age periods when burial mounds were commonly raised over small areas of buried soil. A further resource is presented by ancient banks thrown up for the purpose of land enclosure or defence. The exposure and destruction of these feature by coastal erosion is an issue which requires improved and consistent attention across the changing European seaboard. This is especially important where stratified sequences of land-snails or pollen are known or suspected to have survived.

Archaeology and the study of wind-blown sands

The examination of coastal sand-dunes and sand sheets has been a major feature of this LIFE study and it is exemplified by the collaborative archaeological and oceanographic research carried out on the Médoc coast of Western France (Study Area P17). The Médoc dunes and sand-sheets are of a massive nature and there can be no doubt that their accretionary processes have been driven by powerful natural events. The presence of archaeological horizons within these deposits are a vital means of dating the episodes of sand accretion and identifying the climatic conditions which were conducive to these processes. Where the accretionary phases of these dunes are interrupted by periods of stability, a number of these episodes are associated with human activity on the stabilised land-surfaces. See Plate 2b.

The Médoc dunes on the Aquitaine coast of France
Plate 2b The Médoc dunes on the Aquitaine coast of France contain buried soils and archaeological features which serve to fix the dates of past dune behaviour.

During the coastal archaeological audit of the Isle of Wight, UK, ancient stabilised dunes some 5m high were identified in a cliff-top location overlooking the south-west coast. Archaeological material recovered from an underlying palaeosol provided clear evidence that the accretion of this sand had commenced during the Early Bronze Age at a date which might be attributed to c.2000-1600 BC. This is an event which corresponds with the formation of the primary sand sheet at the base of the great dune of Pyla on the Aquitaine coast (Chapter 7, section 7.2.10).

At the cliff-top site at Redcliff in Sandown Bay of the Isle of Wight, UK (Study Area P5), a Neolithic land-surface has been overwhelmed by sand accretion commencing in Early Bronze Age times (Tomalin 1990). This event could be compared with a more sensitive record of contemporary local events presented by a land-snail diagram for an adjacent slope deposit some 500m to the east (Allen 1994). See Plate 2c.

Redcliff, Sandown, Isle of Wight.
Plate 2c At Redcliff, Sandown, Isle of Wight, UK, a Neolithic and Early Bronze Age land-surface was overwhelmed by wind-blown sands in the early 2nd millennium BC.

A particularly successful example of an archaeological and palaeo-ecological investigations into the chronology of sand accretion comes from the sandy coastal grassland or 'machair' environment on the Scottish island of Harris (Evans 1971, 54-55), see Figure 2.8. This is an island which is now totally devoid of trees. A land-snail diagram prepared for this site reveals that primary woodland was present on the island in Neolithic times while the local climate was transporting wind-borne sand which was helping the machair to grow in height. There follows a clear episode of woodland clearance which coincides with human occupation on the machair and a period of stability and negligible sand movement.

Further changes in the snail population show that this phase of woodland clearance leads to the development of open countryside and a long history of intermittent human occupation on the machair. Notable human influence is exerted by the arrival of a Beaker community in Early Bronze Age times. Bronze Age and Iron Age activities on the machair seem to coincide mostly with phases of stability when the snail shells indicate some limited return of woodland or shade. These are interrupted by periods of rapid sand accretion when there is no evidence of on-site human occupation. Meantime, the snail population reverts to those species which are suited to open country.

The final episode of human occupation occurs during the Iron Age when snails of the open country are present and all effective accretion of sand has ceased. Evans (ibid) comments that this palaeo-environmental examination of a coastal archaeological site indicates that natural processes of sand accretion were largely modelling the character of this island while human impact must also be suspected of imposing some specific effects on this changing environment. This impact is particularly apparent during second phase of the Beaker activity when the deposition of stained sands suggests that wind may have been stripping a nearby area of the machair. This action may be promoted by overgrazing for Evans observes that none of the artefacts recovered from the levels of prehistoric occupation show any indication of arable activity.

Figure 2.8 The sand (machair) deposits on the Scottish island of Harris are a palaeo-environmental archives of coastal and climatic change. This snail diagram reveals that the island was formerly wooded and that sand accretion is contemporary with human impacts which include woodland clearance and the grazing of animals. Archaeological features within the dune deposits provide relative dates for these events (after Evans 1971)

Archaeological and palaeo-ecological studies and the measurement of human environmental impact

As pollen analytical studies progressed during the early 20th century, evidence began to emerge of vast and irreversible human impacts upon the natural post-glacial environment of Europe. The most striking impact arose with the New Stone Age or Neolithic revolution. This event owed its origins to the first experiments in arable farming and agriculture which were under way in the Middle East some ten thousand years ago. The result was a fundamental change from subsistence based on hunting and gathering to the development of sedentary societies which could control and regulate their own food production. Seizing the advantage offered by the indigenous presence of natural wheat, barley, wild sheep and goats, these societies expanded throughout Asia Minor. By 5000 BC both the cereals and the animals had been carried across the Bosporus to establish new colonies in south-eastern Europe. The spread of the new Neolithic economy throughout Europe took another 1,000 years, reaching the distant Orkneys by 3500BC (Ritchie 1990).

The progression of the Neolithic throughout Europe was to bring fundamental and irreversible changes to the Middle Holocene landscape. The planting of cereals demanded fields and this meant that clearances were necessary within the great blanket of climax deciduous woodland, or 'wildwood', which now occupied much of the Northwestern European Plain. The pollen studies of the later 20th century have revealed the magnitude of this prehistoric transition. The nature of these human impacts was first recognised in Denmark by Iverson (1941) who described the process as landnam or 'clearance'. The individual impact of these early clearances was localised and relatively small but the cumulative effect was profound. Ignorance of crop-rotation forced Neolithic communities to progressively clear ever increasing tracts of woodland while in their wake there was left an altered or impoverished soil capable only regenerating secondary woodland of a different ecological character. On valley-sides and hill- tops the progressive impact of clearance could be particularly severe. Here, the removal of forest vegetation brought poor absorption, increased run-off and intense seasonal erosion. These processes created down-slope soil movements which could bury dry valleys and choke and alter the character of rivers (Scaife and Burrin 1992). The early mosaic or patchwork of Neolithic clearances was to lead to a more dynamic regime of widespread clearance, the date of which varies in different parts of Europe. Clearances of this latter kind have been dated to the Bronze Age and they are certainly evident in the regions covered by this present study.

A further outcome of widespread prehistoric clearance was increased exposure to wind erosion. In south-eastern Europe the effects were particularly dire. Deposits of loess and brickearth were particularly attractive to prehistoric farming communities yet once over-ploughed or over-grazed these soils were highly vulnerable to wind erosion. On the Médoc coast of France, at Le Gurp it is clear that, during the Bronze Age, agricultural activity was taking place on the stabilised surface of the sand sheet. This activity was terminated by a renewed episode of rapid sand accretion. This has posed the question as to whether increased or excessive human agricultural activity in the vicinity of the coastal dunes assisted the process of wind erosion and the regenerated movement of sand (see Plate 2d). On the south-west coast (Study Area P3) of the Isle of Wight, UK, a stabilised soil containing evidence of Early Bronze Age occupation was overwhelmed by a massive deposit of dune sand accreting to a height of 4.5m and this has posed a similar question. In all of these instances the dating of episodes of sand accretion rests on the recognition of archaeological horizons within the dunes (Plate 2e).

Evidence of coastal change offered by the archaeological investigations of middens

The past consumption of marine food and the discard of fish remains and edible mollusca can be a valuable indicator of past coastal conditions where sufficient samples of this waste can be obtained. Fortunately some prehistoric and historic coastal communities were in the habit of discarding large quantities marine food waste in dumps or mounds which have commonly been described as kitchen middens. These archaeological deposits are a valuable resource for the study of coastal change. In Ireland 180 Mesolithic habitation sites have been detected but only 13 sites, amounting to 7%, have produced faunal remains. With three exceptions these sites all lie in coastal locations and nine of these have yielded the remains of marine food (Wijngaarden- Bakker 1989).

Médoc coast of France at Le Gurp
Plate 2d Bronze Age agricultural plots buried within the great coastal dune system on the Médoc coast of France at Le Gurp show that human activity was curtailed by a major event of sand deposition in the 2nd millennium BC. These events are mirrored by modern sand accretion in the nearby town of Soulac-sur-Mer.
south-west coast of the Isle of Wight, UK
Plate 2e On the south-west coast of the Isle of Wight, UK, ancient cliff-top sand dunes attest an episode of coastal and climatic change around 2,000 BC. The dune deposits overlie a buried soil which has been dated by the presence of Early Bronze Age beaker pottery.

Studies of marine food waste in the United Kingdom include those at Northton on the island of Harris where the discard of edible molluscs had taken place over at least two millennia while the coastline was accumulating a growing blanket of dune-sand (Evans 1971), see Figure 2.9. While the molluscs had been gathered from the neighbouring shoreline during Neolithic, Early Bronze Age (Beaker) and Iron Age times they had been steadily buried by accreting sand which had eventually reached a height of 3 metres. Analysis of the terrestrial molluscs had already provided significant evidence of vegetation changes in the on-shore environment but it was the consumption of marine food by prehistoric people which was to reveal the history of sea-level change at this remote location.

Kitchen midden deposits, Harris, Scotland

Eight edible species comprising limpet, cockle, dog whelk, winkle, mussel, scallop, oyster and fresh-water mussel were retrieved and quantified. At the lowest level large quantities of the edible cockle Cardium edule were being consumed. These could be attributed to extensive tracts of intertidal sands which were accessible to the Neolithic inhabitants of the site. Their diet reflected an environment which was compatible with a low and stable sea-level.

In the succeeding period of Beaker occupation Cardium was virtually absent, implying that sea- level had risen to a higher level where there was an absence of intertidal sand. The character of this new habitat is reflected in the seafood diet of the community which is now heavily dependent on the gathering of limpets (Patella vulgare) from a rocky shore-line.

It appears that sometime during the 1st millennium BC these changes were reversed for it is evident that the Iron Age population at the Northton site had been able to regain access to an intertidal zone which was once again a successful habit for the cockle. The palaeo-environmental investigator at this site observes that these interpretations of sea-level, won from the archaeological evidence, indicate a scenario which shows good agreement with the land- snail evidence (Evans 1971, 62). He also observes that, for some, a contrary argument might be advanced in which changes in food consumption might be attributed to the idiosyncrasies of personal taste. This doubt he dispels by reference to later documentary sources which confirm that:

'on the north-western coasts of Scotland the greatest abundance of cockles occur..... and there they may form not a luxury but even a necessity of life...... The inhabitants of those rocky regions enjoy an unenviable notoriety for being habitually dependent on this mean diet.... Without this resource, I believe it not too much to say that many individuals must have died for want'.

During the progress of the LIFE project, middens of Iron Age and Roman date were identified in the south-west Isle of Wight coast study area (Study Area P3) and the Undercliff of the Isle of Wight (Study Area P4). Where preliminary analyses have been carried out at St Catherine's Point, the winkle species have been found to be indicative of changes in coastal morphology and sea temperatures in Roman times (Herbert and Trott pers comm).

A study of the location of prehistoric shell middens can be a further means of advancing an understanding of past coastal change. On the Atlantic coast of Portugal an array of Late Mesolithic middens have been examined in the upper reaches of the Sado river (Arnaud 1989), see Figure 7.20. Today, saline conditions normally penetrate this valley for a distance of some 30km yet the shell middens are to found some 50km from the mouth. The distance of 20km from the present estuarine shore to the middens appears to be uneconomic in terms of transport. Transportation is further impeded by the unhelpful siting of some of the middens which can be on the tops of steep slopes above the river. These middens have provided a valuable series of radiocarbon dates which places the discard of the shells around 5470-4480 BC. This places this human activity in the early Atlantic Period when a different coastline must be conjectured. The archaeologists have also observed extensive alluvium on the floor of the valley and this may be responsible for post-Mesolithic changes which have shortened the tidal section of the estuary.

When reviewing the cultural and environmental significance of Early Holocene shell middens, archaeologists have come to suspect that these features are part of a world-wide phenomenon in which an expanding human population broadened and improved its subsistence strategies during and after the close of the Ice Age (Bailey 1989). The exploitation of coastal food resources appears to be a natural step in this process and it is not surprising to find that shell mounds occur in much of the New World, in Australia, the Pacific; Africa; the European Atlantic seaboard and the Baltic (Bailey 1977,1978 and 1989; Biglake 1974; Cipriani 1955; Fairbridge 1976; Osborne 1977; Shackleton and van Andel 1980; Swadling 1976).

Why discarded shells should be specifically heaped rather than randomly scattered has not been explained although it seems likely that the practice may owe much to such issues as the celebration of food-winning success; the reinforcement of group endeavour and group-identity and an expression of territoriality or ownership over a particularly favoured fishing spot or a place of social gathering. In the Sado valley in Portugal and on the Spanish Cantabrian the distribution of middens has been interpreted as an indication of earlier shorelines (Bailey 1983). Where the sea has subsequently advanced, submerged middens can be an important palaeo- geographic indicator although their survival and identification underwater can present problems. The ability to date these shells by oxygen isotope (0-18) greatly enhances their value for the study and dating of coastal changes and this makes the identification and protection of shell middens a important objective in European coastal management.

2.14
Early Studies of archaeological evidence of sea-level and coastal changes

The submergence of archaeological and historic features has been one of the most obvious manifestations of sea-level rise and coastal change. Greek philosophers such as Pythagorus and Aristotle considered that disastrous events induced by of floods and volcanoes had modelled the geography of the past but it was the Roman philosopher Strabo who was to establish the principles which were to guide European earth science of the 19th century. Writing in the 1st century BC, Strabo rejected earlier theories commenting that:

'we should seek our explanations from things which are obvious and in some measure in daily occurrence, such as deluges earthquakes and volcanic eruptions and sudden swellings of land beneath the sea'.

In northern Europe an early observer to take up this topic was Tudor historian William Camden. Writing of the Cornish peninsular Lands End, UK, and the seas beyond, he commented that 'people assert that there was a land, Lyonnesse, so-called from some fable or other, covered over by an inrush of the sea' (Camden 1586). Sixteen years later Richard Carew was to add a little archaeological colour to this assertion when describing the Seven Stones reef lying between Lands End and the Isles of Scilly. Here, he observed, fishermen 'had drawn up peeces or doores and windows' (Carew 1602; Thomas 1985). Later, these discoveries were attributed to items of shipwreck but subsequent amendments to Camden's work gave further credence to the submergence theory by asserting that 'fishermen still see the tops of houses under the water' (Gough 1789; Thomas 1985).

Carew's evocative account of a lost landscape off the coast of Cornwall coincides with other contemporary enquires into the subject of submergence. The severance of the British Isles and the flooding of Dover Strait was a topic raised by the English historian Richard Rowlands writing under the pseudonym of Richard Verstegen in 1605. Acknowledging earlier speculations by others he proposed the existence of a lost isthmus or umbilical some 'six English miles wide'. Interestingly, he turned to palaeo-ecological arguments to support his theory by using the case of the wolf in Britain. No one, he observed, would willingly import these 'wicked beasts' therefore at some point in the past 'they did themselves pass over'. At the time that Rowlands was writing , the activities of British wolves were still a popular anecdote, the last having been killed in 1526.

It was not until the 19th century that these early speculations on submergence and coastal change were scientifically reviewed. In 1830 the initiative was seized by Charles Lyell who cited a broad array of archaeological and historic examples in which evidence of shoreline retreat and submergence could be found. Lyell's historical examples included substantial loss of land in Brittany during the 9th century AD and the loss of the Yorkshire coastal town of Ravensper. In AD 1332, the latter had been acknowledged as a successful medieval port at the mouth of the Humber estuary yet it now lay somewhere beneath the North Sea (see Figure 2.10). On the Suffolk coast, at the celebrated site of Dunwich, Lyell was able to use both historical and field observation to account the progressive erosion and loss of another medieval town.

At the Roman and medieval site of Reculver on the North Kent coast, similar processes of erosion and loss were evident. Here, illustrations of 1781 and 1834 were used to demonstrate the rapid nature of the sea's current advance (see Figure 2.11a and 2.11b). Later, a map of 1685 was to consolidate this evidence (see Figure 7.36). Lyell found that most of his British examples demonstrated encroachment by erosion rather than submergence but examples of some submerged prehistoric forests in the mouth of the river Tay clearly convinced him that physical evidence of sea-level rise and land subsidence was also to be found (Lyell 1840, 1, 46).

The historic loss of medieval towns and villages on the coast of the North Sea in Yorkshire, England,
Figure 2.10 The historic loss of medieval towns and villages on the coast of the North Sea in Yorkshire, England, aroused the interest of Charles Lyell in 1830. This diagram by Professor J.A. Steers summarises the known losses

Lyell carried his search for unambiguous archaeological evidence of submergence to the west coast Italy and the Bay of Baiae (see Figure 2.12). Here an inland cliff marked a past retreat of the sea leaving a margin of re-colonised land known as the Plain of Starza. Two documents of the 16th century helpfully alluded to this retreat where, in AD 1511, Ferdinand and Isabella gifted land to the University of Puzzuoli at the place 'where the sea is drying up' (Lyell 1840, 2, 397). Twenty-seven years later, in AD 1538, a further retreat took place in the bay when eye witnesses to a volcanic event observed that the sea had dried up about 200 paces leaving the fish to the mercy of the inhabitants.

Reculver, North Kent

The most striking observation was, however, yet to come. Situated on the newly emerged land at the edge of the bay were the magnificent ruins of the Temple of Serapis, a fine classical structure of the 2nd century AD. While a direct volcanic cause could account for the land movements which had taken place here, Lyell was quick recognise the broader European implication of this case study commenting that:

'If these appearances presented themselves in the eastern or southern coast of England a geologist would naturally endeavour to seek an explanation in some local depression of high water mark..... but there are scarcely tides in the Mediterranean and, to suppose that sea to have sunk generally from twenty to twenty-five feet since the shores of Campania were covered with sumptuous buildings is an hypothesis clearly untenable..... Thus we arrive, without the aid of the celebrated temple, at the conclusion that the recent deposit at Puzzuoli was upraised in modern times above the level of the sea.'

Plan of Starza in the Bay of Baiae, Naples, Italy

Turning to the extant archaeological remains at Puzzuoli, Lyell was able to find clear and measurable evidence of past changes in the relationship between land and sea-level. The temple of Serapis had been constructed with fine solid columns of marble and granite (see Figure 2.13). Three of the marble columns still remained standing at their origin height of 12.6m. The first 3.6m of each column comprised a smooth stone surface while the uppermost 6.3m was generally similar in character. The intervening 2.7m near the middle of each column was found to riddled with bore-holes of the marine mollusc Lithodomus Cuv. These proclaimed a period of submergence subsequent to their construction in the 2nd century AD while their extant positions demonstrated that this process had been sufficiently smooth to avert toppling.

Lyell went on to observe that other classical ruins were also present in the bay and that these still remained submerged. He was now able to employ archaeological arguments to determine a more detailed chronology of past coastal changes. The temple of Serapis was lined with a fine marble floor but archaeological excavations revealed the presence of a an earlier mosaic floor lying some 2m below. This mosaic was attributed to the early 1st century BC and Lyell concluded that it had been replaced by the upper floor when the lowering of the land surface had demanded a solution to water-logging. By the end of the 4th century it seemed that the temple was subsiding again. This process was to continue until a maximum depth of 5.7m was attained. By this time the floor and lowest portion of the columns were submerged in marine sediment while the central section of the columns had become subject to lithodumus attack.

During the eruption of Monte Nuovo in AD1538 the submerged land-surface and the temple at Puzzuoli were raised by some by 5.7m but this event did not conclude the history of the temple. Lyell commented that at the time of its recognition in 1838, the floor of the temple was normally dry and accessible in most weathers and these conditions persisted until at least 1807. By 1822 however, the Italian architect Antonio Niccolini found that the floor was awash with tidal water twice a day and by 1838 it had become a popular spot for fishing. Between 1822 and 1838 Niccolini used a hydrometer to measure the relative rate of sinking and this he calculated to be 7mm per year. These observation and measurements by Italian archaeologists, architects and geologists presented a major advance in European studies of past changes in sea-level during the timescale of human activity.

Temple of Serapis at Puzzuoli, near Naples, Italy
Figure 2.13 In the Temple of Serapis at Puzzuoli, near Naples, Italy, Charles Lyell & Antonio Niccolini (1840) found clear evidence of land/sea level changes during the historic period. Borings by the marine mollusc Lithodomus showed that the temple columns had once been submerged to a depth of at least 6.3m. Historic accounts witnessed the slow up-raising of the land surface after volcanic activity at Mount Nuovo in AD 1538. Note also the notch line where the columns were formerly subjected to wave attack.

In the wake of the observations at Puzzuoli other unambiguous archaeological examples of changed sea-level were identified on the European coastline. A striking prehistoric example was recognised in the Golfe du Morbihan, France, where, on the tiny island of Er Lannic, a stone circle could by seen rising from the intertidal zone (Lukis 1868; Crawford 1927; Atkinson 1976), (see Figure 2.14, and also Plate 7p). Other semi-submerged megaliths were identified elsewhere on the north-west European seaboard, notably in the creek at Kernic near Plouscat in Finistere and in the ria or submerged valley which now forms the large inlet known as Cork Harbour, Eire (Study Area P20) (see also section 7.42).

Submerged and semi submerged Neolithic stone circles on the shore of Er Lannic in the Golfe du Morbihan, France

While large prehistoric stone monuments in semi-submerged positions presented robust evidence of sea-level rise at a few specific locations on the European seaboard, the recognition of less overt evidence was slow to emerge. Ever since William Borlase had speculated on sea- level changes in the Isles of Scilly in 1753, the issue of the scale and pace of past submergence had been unresolved. Writing in 1927, Crawford drew attention to submerged archaeological evidence in these islands drawing particular attention to ancient stone walls or 'boulder hedges' which could be traced downhill and through the intertidal zone before being lost in the sub-tidal zone. He also noted that the Roman writer Solinus had referred to what appeared to be a single island, Siluram insulam, when he was apparently describing this place in the 3rd century AD (Crawford 1927).

2.15
Recent advances in archaeological studies of submerged landscapes and changing coastlines

Some recent investigations into submerged landscapes in Europe

In the early post-war years of the 20th century the military technology which had been used to produce the self-contained underwater breathing apparatus (SCUBA) was soon deployed for civil scientific and recreational use. In Europe a clear lead was held in France where Jacques-Yves Cousteau had already pioneered the use of self-containing diving equipment in the pre-war years. In his seminal study of the coastline of England and Wales, J.A Steers was prompt to emphasise the importance of submerged archaeological and geomorphological evidence, observing that:

'aqualung or skin diving can be usefully applied to many aspects of coastal work. It has been used successfully in the exploration of now submerged archaeological sites, and it is being applied to the investigation of erosion features, beaches caves etc.'

Since Steers made these preliminary observations in the 1960's the expansion of scuba diving and seabed prospection has won substantial advances in the recognition of archaeological evidence in the sub-tidal zone. This situation has been admirably summarised by American archaeologists who have observed that:

'it is no longer a question that man occupied the continental shelves of the world and the land bridges between the continents during the middle and latter Quaternary period of geologic time... On the other hand the identification of early man's sites on the continental shelves is another matter.. (Kraft et al1983).'

When considering the potential for the discovery of shelf occupation sites in 1983, the American view was that turbidity, poor visibility and deep sedimentation were factors which impeded the recognition of most evidence which was still tantalisingly elusive. To these impediments was added the destructive power of past inundation; an effect which was best mitigated by a rapid marine advance.

Despite all of these problems, good progress was made in archaeological prospection at a number of locations on the European shelf. On the sea floor adjacent to Capo Sidero on the island of Corfu submerged scatters of Middle/Upper Palaeolithic implements were found in a submerged beach deposit 10m below present sea-level (Sordinas 1983). These were first detected when washed-up artefacts were recovered by beach-walkers.

In the mouth of the Baltic, in the Strait of Oresund, navigational dredging produced a prehistoric knife made of reindeer antler and an array of flint axes, scrapers and flint waste and animal remains. On-shore topography was used to extrapolate a submerged landscape where suitable areas might be defined for an underwater search using a trawled scraper/sampler. These explorations were then followed up by an archaeological diving team. This work led to the identification of two submerged occupation sites situated 8m and 18m below sea-level. The scraper/sampler used by the prospection team also recovered prehistoric flintwork from a third site lying at -23m. This could not be inspected by the diving team. The deepest of these submerged settlements, site 3, could be attributed to an event prior to 6900 BC while site 2 could be attributed to an event no later than 6000 BC (Larsson 1983). It should be noted that investigation of site 1 was curtailed due to damage and loss caused by dredging activity.

A valuable appraisal of submerged coastal archaeological sites was that presented by Flemming in 1983. This study defined the environmental contexts in which such sites were to be found, noting in particular that rias, lagoons, estuaries and sheltered alluvial coasts offered the most favourable conditions for the preservation of archaeological material. One of the conclusions of this LIFE study is that rias and estuaries are also principle focii for modern European commercial maritime activity and for this reason the sustainability of the submerged archaeological and palaeo-environmental record is under particular stress at these locations.

In the United Kingdom the archaeological study of submerged coastlines was to remain largely unaddressed until the 1980's. In 1973 the very substantial remains of a Dutch East Indiaman of 1748 had been recklessly disturbed and damaged by treasure-hunters on Hastings sands (Marsden 1974 and 1985). This had led to the hasty passing of the Protection of Wreck Act, a Private Members Bill which served to focus Government attention solely upon the shipwreck element of the underwater archaeological resource. Contemporary academic energies were also being expended upon the underwater excavation and recovery of the Tudor warship Mary Rose and this channelled further attention towards the purely cultural and nautical aspects of coastal archaeological research.

An evocative series of papers published by the Society of Antiquaries in 1980 sought to restore a balanced understanding of coastal archaeology. These papers, entitled 'archaeology and coastal change' were able to show little contemporary coastal fieldwork but they were able to present a important statement of potential (Thompson 1980). This publication was promptly followed by a global appraisal of 'Quaternary coastlines and marine archaeology' in which a lack of coastal fieldwork seemed apparent within the United Kingdom (Masters and Flemming 1983). This lapse in British interest was soon amended by a new Scillonian study appropriately entitled 'Exploration of a drowned landscape' (Thomas 1985). This was followed by new archaeological and palaeo-environmental investigation of the submerged peat and Mesolithic occupation site in the intertidal zone of the English Channel coast at Westward Ho! near Plymouth (Balaam et al. 1987). Meanwhile, on the eastern coast of England, a series of new annual progress reports was being produced by Essex County Council. These announced the results of pioneering experiments in intertidal fieldwork in the Blackwater estuary (Murphy et al. 1982-87).

The Blackwater or 'Hullbridge' project was highly successful in identifying a broad array of prehistoric structures and submerged land-surfaces which were capable of revealing a detailed history of sea-level rise and coastal change. The Hullbridge project had adopted a multi-disciplinary approach and this, with the encouragement of English Heritage, was soon extended to the Solent region on the south coast of England where a somewhat similar history of Holocene submergence had been suspected (Tomalin 1992; 1995; Loader et al 1997). Further initiatives by English Heritage extended this approach to the North Sea coast of Lincolnshire and Humberside (Van de Noort 1997) by which time the labour-intensive nature of these new surveys had become clearly apparent. A statement on England's coastal heritage', published by English Heritage and the Royal Commission on Historic Monuments in 1996 was followed by a full review of 'England's coastal heritage' published in the following year (Fulford et al. 1997). Some of its conclusions are embodied in this LIFE report.

Coastal archaeological surveys, audits and Shoreline Management Plans in the United Kingdom

Contemporary with the British publication on coastal archaeology in the 1990's was a new policy of rapid coastal audit. This sought to appraise the full spectrum of national coastal archaeological sites before further intensive surveys were commissioned. The commissioning of both surveys and audits was general funded by Central Government through a number of grants awarded by English Heritage to individual local authorities (see Plate 2f). By this device support was won from those local communities who would be entering into long term Shoreline Management Plans. Unfortunately, due to an impediment in the enabling Act for English Heritage, the scope of the coastal archaeological audits could not be extended into the sub-tidal zone where fully submerged archaeological evidence of coastal change were generally known to lie.

The recent commissioning of coastal archaeological surveys in 			the United Kingdom has made fundamental changes in the comprehension of concealed palaeo-environmental resources.
Plate 2f The recent commissioning of coastal archaeological surveys in the United Kingdom has made fundamental changes in the comprehension of concealed palaeo-environmental resources.

A particularly important achievement of the present LIFE programme has been the application of original seabed archaeological survey to a selected study area at Bouldnor on the of the Isle of Wight, UK (Study Area P1). This study adopted a 'seamless' vision of the coastline, by first examining the active processes of slope movement and instability along the receding cliff-line (see Dix Volume 2, Study Area P1). It then examined the evidence of submergence in the intertidal zone; pursuing this to a depth of 12m on the sea-floor. The latter aspect of this work has been particularly important because at a depth of 11m significant traces of in-situ human habitation (Momber Volume 2, Study Area P1) have been located within the context of a deeply submerged early Holocene forest (Scaife Volume 2, Study Area P1). The LIFE project has been successful in locating precisely the type of submerged archaeological evidence which has been excluded from the remit of the governmental heritage agency in England. This survey has also been highly successful in identifying a key sequence of sediment archives overlying the horizon of human activity. The archaeological, stratigraphic, palaeo-ecological investigation and absolute dating of this archive is a key component of this concluding report (see Momber 2000; Scaife Volume 2, Study Area P1). It also complements other studies of which have focused upon the shoreline of the Eastern Solent at Wootton-Quarr (Study Area P2), (Tomalin et al. forthcoming).

Parallel to the development of new coastal archaeological initiatives in the United Kingdom were new policies for coastal protection and coastal management. These changes embodied the use of coastal cells as a basic framework for organising the future management of the coastline (MAFF1995). The first outcome of this initiative was the commissioning of Shoreline Management Plans, the purpose of which has been examined in Chapter 1. An important element of these plans was the collaboration of central government and local government in pursuing the common aim of achieving consistency in the overall management of the coast. The first round of these plans was generally complete by the year 2000. After this time, long-range policy prudently allowed review of the plans at five year intervals in order to encompass new contingencies. A number of management prescriptions in this first generation were inconsistent in their comprehension of long-term coastal changes and the evidence offered by the archaeological resources concealed within the coastal zone.

The relative and absolute dating of archaeological material in coastal contexts

Since Charles Lyell and Antonio Niccolini made their first observations on the unearthing of archaeological structures in the coastal zone, the study of stratigraphy has become the principal means of dating a succession of coastal sediments. Guided by a general law of superimposition, the stratigraphical approach recognised that the oldest sediment would rest at the bottom an accumulated sequence and the youngest would lie at the top. artefacts and biological remains recovered from individual layers or 'horizons' might help to place dates on a relative succession of events but none of this evidence could provide an independently calibrated timescale.

The search for an independent or absolute timescale developed in the early 20th century when Baron de Geer realised that lake-water sediments in Scandinavia represented annuals outflows of post-glacial meltwater. By counting the fine individual layers or varves, the years might be counted backwards to the close of the Ice Age (de Geer 1912) (see Plate 2g). This search for calibration on an absolute timescale was soon applied to ancient tree rings but both techniques encountered considerable problems when gaps and uncertainties arose in the collected samples.

Subsequently, other ingenious methods were devised including palaeo-magnetic dating. This relied on a reconstruction of past variations in the position of magnetic north. Molecules in fired clay remained fixed upon the position of the magnetic pole at the time of their firing. Provided that the clay remains in its origin position it is possible to examine the alignment of the fixed molecules and calculate the time of their firing. This technique was successfully used in the medieval tile kiln examined in this LIFE report. It has already been noted that pollen diagrams can sometimes provide a helpful indication of date when the result of the analysis accords with known examples of past climatic and vegetational conditions. This is not, however, the primary purpose of pollen analysis and the chronological information gained can only be placed on a relative timescale which has been divided according to climatic zonation.

The breakthrough in absolute dating came with the measurement of isotope decay during the post-war race for the mastery of nuclear physics in the mid 20th century (Libby 1946). The technique was first applied to carbon 14 which has since offered an acceptable accuracy of measurement for the past 50,000 years. It was found that this measurement could be applied to all former living tissue and was particularly suited to wood. Cross reference initially to the tree rings of the American bristlecone pine and later to European oak trees has shown that radiocarbon years cannot be viewed as calendrical years and for this reason varying degrees of error arise at different periods during the past. These differences are due to past fluctuations in solar radiation but they can be 'smoothed' by use of calibration table s which can convert and confine these dates within an acceptable margin of error expressed in calendrical years. Where calcium carbonate has been secreted in the shells of molluscs, isotope decay in oxygen 18 can be used as an absolute means of dating. The technique has a high range of error for the Holocene period and it is best applied to land and marine mollusca of the older Pleistocene.

Expense and limitations in the availability of appropriate samples means that archaeologists still rely heavily upon relative dating based upon the character and typology of artefacts. With the help of radiocarbon the frameworks for artifact-dating are now more refined and this means that absolute dating is not necessarily the best choice. Radiocarbon dating has also been used to promote and cross-check the building of a tree-ring chronology. Where tree-ring dates are sought, most samples will be need to offer a minimum of 100 rings but if these conditions are fulfilled then it should be possible to gain a match with a record of north-western European tree- rings which now reaches back 7,000 years.

The counting of annual layers of melt-water sediments or varves  in Scandinavian lake-beds
Baron Gerhardt de Geer
Plate 2g The counting of annual layers of melt-water sediments or varves (a) in Scandinavian lake-beds was pioneered by Baron Gerhardt de Geer (b) as an early means of obtaining an absolute date for the Ice Age.

The interrogation of sediment archives in the reconstruction of a long-term coastal- change chronology; a Solent case-study at Bouldnor

On the north-western shore of the Isle of Wight, UK, at Bouldnor, two geomorphological features were investigated by this LIFE project with a view to obtaining new information for the regional coastal-change chronology. Above mean high water mark a substantial cliff composed of Hampstead Clay was subject to marked recession and landsliding. The age of this cliff and the periodicity of its landslide events was considered to be highly pertinent to the age of the Western Solent seaway. The geomorphology of this cliff was investigated by the University of Southampton School of Ocean and Earth Science whose findings are presented in Volume 2 of this report (Study Area P1).

The second feature to be investigated lay in the sub-tidal zone. This comprised an underwater cliff lying some 200m offshore from the present cliff-line (see Figure 2.15). This feature had first been observed during an underwater archaeological survey conducted in 1986-1989 (Tomalin et al. 1986; Tomalin 1992 and 2000; Momber 2000). The submerged cliff was quite unlike its on-shore counterpart for it was found to be composed of soft Middle Holocene sediments exposed in a near-vertical face which was being actively eroded by sub-tidal processes. For the purposes of the LIFE study, this site was further examined by the Hampshire and Wight Trust for Maritime Archaeology and samples were removed for pollen analysis by Dr R.G. Scaife of the Quaternary Environmental Change Research Group of the University of Southampton.

On investigation, the sub-tidal cliff-face at Bouldnor was found to contain a sediment archive some 8m deep. The submerged cliff-face lay at the edge of a shallow coastal shelf which, at the point of sampling, could be traced to a depth of -3.72m. Here its surface was found to be composed of brown grass and sedge peat. The exposure presented by the underwater cliff allowed the underlying sediments to be traced and sampled to a depth of -12m OD. This revealed a history of sea-level rise accompanied by the accretion of marine and brackish sediments. This sequence had been interrupted by two episodes when terrestrial conditions had been established for periods sufficient to sustain the development of coastal fen or fen carr woodland. These episodes were represented by the middle and upper peats and they bore evidence of a respite in the prevailing processes of sea-level rise and marine advance.

At an early stage in the LIFE project it was recognised that the sediment archive preserved at Bouldnor held a key to the coastal change chronology of the entire Western Solent. The beginning of this coastal change record lay at a depth of -12m OD where submerged tree trunks, boles and root systems attested the presence of Middle Holocene woodland dominated by the growth of oak and alder trees. In this study the core from this basal peat deposit is denominated 'Bouldnor 3'. The context of this core was examined and sampled by the Hampshire and Wight Trust for Maritime Archaeology who found that it also contained flint artefacts left by a contemporary Mesolithic community (Momber 2000). A radiocarbon date obtained from this peat deposit places beginning of this forest around 6600-6400 cal BC (BETA- 140104). This is complemented by a date of 6440-6110 cal BC (GU-5420) obtained from an in-situ tree bole.

The nature of this coastal woodland is revealed in the pollen diagram assembled by Dr Scaife (Figure 2.16). This shows that oak and hazel trees were thriving while the warm-loving lime or linden tree was just beginning to make its appearance. This environment is characteristic of the Boreal/Atlantic transition when a warming climate and a marked rise in sea-level is well recognised throughout Europe. The pollen diagram also shows a strong marshland element in which the incoming growth of alder trees predominates. There are also reed swamp elements and aquatic types which include the pond weed Potomogeton. Dr Scaife interprets the Basal Peat as a product of rising sea-level, its first effects being the ponding-back of local freshwater river systems to create higher ground-water levels and waterlogging. The results would be anaerobic conditions suitable for the accumulation of peat and the proliferation of alder carr woodland. The presence of buckthorn (Rhamnus cathartica) willow (Salix), oak (Quercus) and yew (Taxus) and a ground flora of sedges are all typical of a fen wood environment.

Submerged forest sediment archive at Bouldnor, Isle of Wight

Above the basal peat of Bouldnor 3 there follows a long sequence of marine sediments amounting to 6.32m of clay and silt. This major stratigraphic unit signifies the continuing rise in sea-level during the Middle Holocene. The first interruption in this process is marked by the middle peat of 'Bouldnor 2'. This peat occurs at a level of -5.10m OD where it is preceded by a grey silty clay representing the transition from marine conditions. The pollen diagram for the middle peat bed (Bouldnor 2) suggests that this transition began with a saltmarsh in which chenopods, sea lavender, thrift and beaked tasselweed were present (Armeria'B' and Ruppia maritima). On nearby firmer ground, oak and hazel woodland predominated while some lime/linden trees may also have been present. By the time peat deposition had been established at this location, the surface of the old saltmarsh had become well colonised by grasses (Poaceae). The diagram shows that there is now a notable absence of halophytes and this suggests that the sea is now some distance away.

Bouldner Cliff (Summary)

In the uppermost section of the Bouldnor 2 deposit the pollen diagram shows a return of freshwater aquatic types (Ruppia maritima, Potomogeton and Typha/Sparganium) and as saltmarsh conditions reappear the accretion of marine silt is re-established. This leads to a further marine phase during which 0.66m of silt is deposited before the uppermost peat appears at - 4.10m OD.

The diagram for the upper peat (Bouldnor 1) begins with a low grass count (Poaceae) but a modest presence of pond weed and herbs characteristic of saltmarsh. The main phase is characterised by an increase in grasses and sedges as the saltmarsh gives way to a fen environment. The tree pollen is still dominated by oak and hazel and this probably includes trees and bushes colonising the drier margins of the coastal fen. In the uppermost level of the peat the grass pollen falls while the appearance of bog moss (Sphagnum) indicates that the mire had become more acidic. None of these developments show any hint of a return to marine conditions yet at a level of -3.72m OD this peat is truncated where it now becomes the present seabed.

Absolute dates obtained from the middle and upper peats at Bouldnor help to fix the chronology of sea level rise and coastal change in the Western Solent. A date of 4920-4530 cal BC (BETA- 140103) marks the onset of the grass/sedge fen in Bouldnor 2 and a date of 4530-4330 cal BC (BETA-140102) fixes the onset of the upper peat in Bouldnor 1. These absolute dates show that the valley of the western Solent was a wooded basin in the 7th millennium BC when climate change was favouring the development of deciduous trees including oak, hazel and lime/linden. Wetland conditions within the basin favoured the proliferation of alder while the presence of aquatic plants reveals that the environment could be readily susceptible to flooding. There is little indication of marine influence at this time although the sequence appears to have been abruptly terminated by a marine transgression which was responsible for the accretion of more than 6m of marine clay and silt.

The absolute dates provided by radiocarbon dating suggest that accretion of the marine sediments at Bouldnor probably commenced before the close of the 7th millennium BC. This would accord very well with the steep rise in world sea-level which is recognised during Flandrian 1. An important question is whether this event carried sufficient local impetus to secure the formation of an open seaway and the severance of the Isle of Wight. The dating of this event is particularly important for it is the point of origin from which the evolution of all subsequent coastal processes in the Solent region must be measured.

The occurrence of a further episode of coastal peat formation in 'Bouldnor 2' suggests that the severance may not have been achieved until after the inundation of the late 7th and 6th millennium BC (Tomalin 2000). The absolute date of 4920-4530 cal BC (BETA 140103) for the onset of further peat formation in Bouldnor 2 suggests that the primary advance of sea was achieved over a period of some 2,000 years yet the environment at Bouldnor remained sufficiently sheltered to allow the fine particles of marine silt to accrete in the low-energy environment of a coastal saltmarsh. A very similar history is evident in cores obtained from neighbouring sediment archives at Newtown and Yarmouth where accretion of the same marine silt continues into the 4th millennium BC (Tomalin ibid; Scaife Volume 2, Study Area P1).

A particularly interesting and unusual feature of the pollen diagram is the very high incidence of pre-Quaternary pollen in the beginning of pollen zone 1 in Bouldnor 2. Dr Scaife observes that this is the product of erosion of Cretaceous and Tertiary rocks and it seems to have been generated either by fluvial processes or by coastal erosion on an appropriate cliffline of the region. This event seems relatively short-lived because it is not found in the succeeding pollen zone 2 of Bouldnor 2. In the upper peat bed (Bouldnor 1) the pre-Quaternary pollen re- appears in modest quantities and this seems to suggest that the process may have continued since it was first detected in association with marine influences in the middle peat (Bouldnor 2 zone 1). If the implied association with marine influence holds true, this pollen could denote the first effective marine attack on the present cliff-line at Bouldnor Cliff and it would place this event in the early 5th millennium BC.

An important feature of this west Solent study (Study Area P1) is the complementary role played by other sites and sediment archives which have been examined on this coastline. At Bouldnor the sequence stopped at a depth of -3.72m OD where the peat was truncated on the sea floor. This truncation deprived the site of any record of events after c. 4,000BC. The closest analogy could be found some 5km east of Bouldnor at Newtown, where the salt marsh silts could be detected at a depth of c. -12.00m OD (unit 10/11). This silt accretion post-dated a peat horizon and a deposit of freshwater gyttja dated at 6570-6180 cal BC (GU-5427). At this site the silt sequence had continued to accrete until a present height 1.15m OD had been attained but there had been some notable interruptions during this history. These included the formation of two peat horizons (units 4 and 2) resting at -3.95 and -3.38m OD. After salt marsh accretion had resumed at Newtown a Neolithic trackway had been laid on a nearby surface at a height of -1.59m OD. This structure had been dated at 2920-2500 cal BC (GU–5341) and it helped to resolve an anomalous reversal in the dating of the two peat horizons.

At Yarmouth another Neolithic trackway had been found in a similar coastal position and this had produced a comparable date of 2920-2620 cal BC (GU-5260). The position of this structure could also be compared with a local stratigraphic sequence recovered in a deep sediment core in the mouth of the drowned river valley. This core indicated that marine influence had been become discernibly stronger after peat accretion had ceased at a height -5.51m OD. This marked an event when and the Yar river had been overwhelmed by marine silt and it could be dated at 3650-3360 cal BC (GU-5419). Prior to this date the river had accrued a long series of intercalated peats and riverine sediments and these could be traced down to a depth of - 11.21m OD. Stratigraphic unit 17 in the Yarmouth core showed that by 6380-5840 cal BC (GU- 5397) a peat-forming environment with oak woodland had developed on the floor of the Yar valley at a level of -9.51m OD. This could be compared with the basal peat at a depth of - 11.68m OD in Bouldnor 3 which was dated slightly earlier in the 7th millennium BC.

At Newtown the basal deposit of freshwater gyttja in unit 16 was identified at a height of - 12.22m OD and this was also dated to the late 7th millennium (6570-6180 cal BC (GU-5427). At Bouldnor it was evident that marine influence must have been well underway by the early 6th millennium BC but in the river Yar no such event could be traced prior to the mid 4th millennium BC. This difference seems to suggest that the river channels at Newtown, Bouldnor and Yarmouth all had their own courses to the sea and that the western Yar behaved in an independent manner. This may tell us something about the former catchment and flow rate of the Yar which, at that time, probably included a substantial tract of land in the area which is now occupied by Freshwater Bay.

When the marine phase finally commenced in the Yar valley it seems likely that this transition lay close to the event which was responsible for the severance of Wight. Dr Long has commented that while coastal peats were forming on the margins of the western Solent, high- energy coastal processes such as those responsible for a severance event are unlikely to have occurred. The Yar and Newtown cores present evidence from drowned and truncated river systems where the new erosive effects of severance could be naturally dissipated by narrow mouths and the buffering effects of river budgets. At Bouldnor, however, we have a sediment sequence which has accrued in the path of the developing Solent seaway and this shows that sheltered peat-forming environments were terminated before the close of the 4th millennium BC.

The examination of submerged sediment sequences in the western Solent has shown that these resources are best sampled at more than one location and that evidence offered by pollen, diatom and sedimentological studies should be all be fully pursued (Figure 2.17). This case study also found that single absolute dates from palaeo-environmental deposits could present anomalies and contradictions and that these uncertainties could be resolved by drawing

comparison with absolute dates offered by contemporary archaeological structures. In this case- study all of these approaches were employed for the purpose of establishing a coastal change chronology which could explain the natural processes which were now being confronted by the Shoreline Management Plan.

Cored sediment archives in the Eastern and Western Solent

2.16
Past and present coastal communities in Europe and the evolution of their coastal problems

The first defences against tidal flooding

Before the opening of the 19th century the power and the omnipotence of sea was unquestioned. On the north-west European seaboard some notable schemes for the reclamation of certain coastal wetlands had been pursued in Roman times but few if any of these achievements had been sustained into the 2nd millennium AD. Examples included the west bank of the river Severn where an inscribed stone at Goldcliff records works of the first cohort of the 2nd Legion Augusta which were apparently associated with some kind of sea-defence (Boon 1980). It is possible, however, that this may have been a drain to remove back-ponded water rather than a wall to resist the sea (see Figure 2.18). Later, in the 17th century, the same area was subject to disastrous flooding. On the east coast of the Severn estuary, coastal wetlands were recovered by Roman drainage works whilst very similar activity was also pursued on the coastal boundaries of the English Fenlands (Darby 1940; Godwin 1978).

Goldcliff inscription of 1606

In north-west Europe opportune improvements or reclamations of coastal farmland had been practised during medieval times but these activities had always grown from land drainage schemes and none had sought to establish settled urban populations in close proximity to the sea (see Figure 2.19). The medieval religious houses of northern Europe took a notable lead in the construction of modest waterways 'lodes' or 'rhynnes' and these, in low-lying coastal areas, would often be reinforced by earthen banks designed to resist overtopping during floods or high tides. Monastic works of this kind commenced on the east bank of the Gironde, France, (Study Area P16) in the 13th century.

In the English Fenlands contemporary schemes were of a modest nature until a major new waterway or drain, called Morton's Leam, was constructed between AD1478 and 1490. This was designed to relieve flooding by the river Nene (Godwin 1978). In AD1600 an English Act of Parliament was passed 'for the recovering of many hundreds of acres of marshes' but it was not until Dutch technology was introduced by Cornelius Vermuyden in 1631 that larger and more ambitious drainage schemes were promoted with the help of adventure capital. With these larger drainage schemes there came a need for the construction of wind-pumps, sluices and sea banks but these protective works were still largely aimed at controlling the outflow or debouchment of tidal rivers and they offered little direct challenge to the domain of the sea. It was not until the middle decades of the 19th century that a new temerity was to arise.

Early drainage of coastal wetlands of Aquitaine , France
Figure 2.19 Early drainage of coastal wetlands of Aquitaine , France, is shown in this contemporary view of the battle of La Rochelle in 1627. A portion of the Aquitaine dunes is also shown

In Eire, Aidan O'Sullivan reports that there has been little historical or geographical research on the history of coastal reclamation (but see Rowe and Wilson 1993; and Furlong 1996) for studies of the evolution and reclamation of Wexford Harbour). During this LIFE project the Discovery team of Eire conducted a specific study of the development of coastal defence and land reclamation in the Shannon Estuary (Study Area P19) where events seem to reflect the general history of Irish enterprise in securing and defending desirable coastal farmland. On the Co. Clare coastline (including the Fergus Estuary) there is some 5,900 hectares of identifiable reclaimed estuarine land. Large tracts also occur on the south bank of the estuary stretching inland around the River Maigue and surrounding Limerick city. This suggests that the Shannon is a highly modified estuary, and that in its original form the river channel probably meandered through large tracts of wetland and marsh.

The Civil Survey of the mid 1650's for the County of Limerick indicates that land reclamation had taken place along the Shannon Estuary and the River Maigue. At Newtown in Co. Limerick, at least two earthen banks lie up to 2 km inland of the present day sea defence, the innermost of these being approximately of late medieval date. Further evidence from the Down Survey Maps indicates that large parts of the Shannon Estuary had been reclaimed by 1655. Indeed, in a lease of land in the vicinity of Clarecastle, dated to 2nd October 1656, made by the Earl of Thomond's agent to Colonel Henry Ingoldesby, a directive is included to 'make up the seabanks and sluices' (Inchiquin Mss., 353-4). In a description of Clare in 1682 Hugh Brigdall, an attorney living in Ennis, stated that 'towards the brink of the Shannon the south east bounds of this county lieth a rich vain of land with many thousand acres of marsh defended with banks from the fury of that river' (Ó Dálaigh 1998, 68).

William Petty's Map of 1685 shows that much of the area around Islandmagrath and Islandavanna was at that time open to the sea. However, these banks required upkeeping to maintain the agricultural value of the lands. During the Williamite War, Terlagh [sic] O'Brien had promised to supply King James' administration with 300 tonnes of hay or corn from his land at Lattoon, beside the Fergus Estuary. In 1690 the Commissioners for Co. Clare subsequently reported that he had allowed his farmland to go derelict and 'had suffered the seabankes, draines and sluces thereon to goe to decay…'(Inchiquin Mss., 469) and that there was no way that the land was going to supply the amount of tillage required. It may be more than coincidence that Dutchmen introduced the growing of coleseed on the marshlands in the 17th century, the very nation with most expertise in land reclamation (Ó Dálaigh 1998, 69).

In Thomas Moland's survey of the Earl of Thomond's estates in Clare for the year 1703 there are references to corcas banks, sea banks and sluices along the Shannon in the parishes of Kilfentinane and Kilnasullagh (Ó Dálaigh 1998, 80). There were, however, problems at the time as the banks in the vicinity of Islandmagrath were exposed to much damage by the Fergus (Ó Dálaigh 1998, 90). In the late 18th century Arthur Young, the noted agriculturalist and travel writer recorded that 'there are 20,000 acres ….called the carcasses' and that 'when in tillage, they sometimes yield extraordinary crops; 50 stat barrel an acre of bere have been known, sixteen of barley, and from 20 to 24 of oats are common crops…' (Hutton 1970; 285, 291-2).

Throughout the 19th century it was popular for travellers to publish their experiences of peoples and places encountered through the course of their travels and visitors to Ireland were no exception. A number of individuals such as the architect and botanist Joseph Woods in 1809 and the celebrated Victorian author William Thackeray in 1842 admired the fertile quality of the lands bordering the Shannon and Fergus (Ó Dálaigh 1998, 144, 198). During the worst ravages of the famine, a Scottish agricultural expert, James Caird, carried out a survey on farming in the south and west of the country on behalf of the British government in the Autumn of 1849. He alluded to the tracts of rich alluvial land, the 'corcases', which provided very high rents before the onslaught of the famine (Ó Dálaigh 1998; 240).

Reclamation supported by government-backed schemes began in the first half of the nineteenth century, with the first known survey carried out in 1822. This was followed in 1834 by the passing of the First Shannon Navigation Act, the purpose of which was 'the improvement of the River Shannon from its source to its mouth' (Parliamentary Papers 1835). This Act appointed the Shannon Commissioners, among whose functions were to 'ascertain works necessary to be executed for the improvement of the navigation of the River Shannon ... to make estimates of the expense of such works [and] to determine the districts or counties which may be considered directly benefited by these improvements' (Parliamentary Papers 1835). Work around the estuary began in 1840, and by 1846, 90,000 people were employed on drainage and coastal improvement schemes. However, in 1852 work had stopped as a result of objections by landlords to the costs of the schemes. Further proposals for the drainage of 9,000 acres of land proximal to the estuary were considered in 1853 but failed for financial reasons.

On the western side of the Fergus Estuary, around the areas of Islandmagrath and Islandavanna, substantial areas of land were reclaimed from tidal marshes in a series of works in the nineteenth century. Landward of these additional areas of reclaimed tidal wetland occur, although works associated with these are as yet undated. In 1860 'An act for the improving of the navigation of the River Fergus, and the embanking and reclaim from the sea of wastelands subject to be overflowed by the tide on the eastern and western sides of the river in the County of Clare' was passed. Following this Act, which saw a switch of emphasis from navigation to drainage and land reclamation, activities related to reclamation in the Fergus Estuary accelerated.

In 1864, plans for the drainage of in excess of 7,000 acres of land between Islandmagrath and Islandavanna were proposed, including substantial areas of embankment at a cost of around twenty eight thousand pounds. Plans were proposed which would encompass the drainage and reclamation of land in many of the townlands and parishes surrounding the Fergus, including Ballygireen, Ing West, Ing East, Kilnasoolagh, Clenagh and Kilmaleery, as well as areas of marshland adjoining the townlands of Clenagh, Coney Island, Kiladysert, Lisheen, Killone, Buncraggy, Lissane and Islandmagrath among others. In 1879, the work on the Fergus Estuary began and although tidal pressures and later breaches caused the abandonment of some of the plans, large areas of the Fergus Estuary were drained and turned over to agriculture by the 1930s.

Most of these embankments were built and maintained by the local landed gentry until the passage of the Land Purchase Acts when responsibility passed to local trustees set up by the Land Commission. These embankments, however, were never maintained properly and breaches certainly occurred. Most of these embankments were transferred to the Commissioners of Public Works under the terms of the Arterial Drainage Act, 1945. During the 1950s, a number of important schemes were carried out by the Commissioners to enhance the flood protection along the Owenagarney River near Bunratty, from Bunratty to Rineanna on the north side of the Shannon and at the mouth of the Fergus.

In 1961 a severe storm resulted in the breaching of embankments along the Shannon Estuary (Study Area P19). At the worst affected areas farm land was flooded to a depth of 5 metres and at Coonagh on the north side houses were flooded to the eaves. This initiated a further number of schemes at Coonagh, the Ballinclough embankment, the Mellon-Ringmoylan embankment, the Ringmoylan-Foynes embankment and along the River Maigue in the area north of Kildine. The development of well constructed embankments in the second half of the 20th century has allowed the protection and drainage of existing agricultural land with some reclamation, and aided the industrial development of the region with a focus on the Shannon airport. These works were carried out before intertidal archaeological investigations in the Shannon Estuary had been able establish a long-term vision of the past behaviour of this coastline.

Problems concerning historic shorelines of cities and ports

In most of the Mediterranean states of Europe the construction of coastal works claims a long historic pedigree. The first communities to confront the sea had done so in a modest way when seeking to protect their mercantile interests in trade and shipping. Trade and naval interests in the classical world of Greece and Rome had demanded safe docking facilities. This nurtured ancient expertise in the construction of stone harbour walls, mooring basins and embanked and revetted waterfronts or quays. The Pax Romana carried these interests to most parts of the Mediterranean and Atlantic seaboards while a growth of riverine traffic also encouraged waterfront constructions on a number of Europe's principal rivers (Izarra 1993).

A heightening of maritime interests in medieval times coincided with Europe's new enthusiasm for global exploration; this brought renewed energies and refinements to the construction of waterside works. In coastal and esturine cities such as Venice, Genoa and Seville fine new buildings, piazzas and monuments were carried by architects and designers to the very brink of the lapping tide. To contemporary minds this was the threshold of a domain from whence endless wealth could be delivered by valiant crews arriving on boundless and benevolent tides. For the early builders and designers of such great shore-side cities, any prospect of coastal change was securely tended by divine providence.

Unlike the Atlantic the tidal range within the Mediterranean was reassuringly low and any thoughts of a future rise in sea-level were unconsidered. The fortunes invested in grand buildings in some of these seaside cities were truly enormous and they established, for succeeding centuries, some principal cultural, economic and social centres which could not, effectively, be moved. Unfortunately, a concealed future cost seems to have followed some of these decisions and it seems that it has fallen to the present generation confront some of these consequences. We can now see that certain decisions in civic planning, made in centuries past, have delivered the historic cores of some of Europe's finest maritime cities perilously close to Neptune's scaly grasp. This veritable brinkmanship can be readily witnessed by anyone who has viewed the tidal flooding of the fine Venetian piazza of S. Marco Square (Plate 2h).

S. Marco Square in Venice, Italy,
Plate 2h The regular flooding of S. Marco Square in Venice, Italy, is a reminder that most of Europe's historic city-ports were constructed without regard for the cumulative effects of sea-level rise.

A coastal history of a very different nature can be detected in some of the major historic cities and ports on the Atlantic seaboard of Europe. At locations such as Bordeaux, London and Southampton walled Roman settlements with attendant waterfronts gave rise to maritime communities in the 1st to 4th centuries AD (see Figure 2.20). Later, these sites were to accommodate highly successful trading communities in medieval and post-medieval times. On the bank of the Liffey at Dublin (Study Area P22), Viking pioneers of the 9th century founded a successful port and trading community which was to develop in much the same manner. All of these historic settlements were constructed on the shores of inlets or estuaries where a substantial tidal range immediately prompted a regard for the likelihood of flooding.

Roman city of Bordeaux

The earliest buildings on these sites were erected above the mean low water mark and, with prudence, they usually cleared the highest astronomical tide of the 1st millennium AD. Such a height has been described as the minimum occupation level or MOL (Thomas 1985), Figure 2.21(a) and 2.21(b). On the bank of the Thames in England much of the Roman town of Londinium was constructed on an old and well drained gravel terrace while at Southampton similar Roman development gave appropriate respect to the intertidal margins of the Itchen river (Study Area P9).

It is particularly unfortunate for modern inhabitants that early respect for the natural tidal margins in many of our city-ports was soon compromised. This occurred when the economic life of a coastal town became increasingly drawn towards its commercial threshold on the waterfront. At first, ships approaching these early maritime communities would often settle on 'hards' where they could be approached and unloaded at low tide. Where settlements flourished, these arrangements were soon replaced by raised wooden walkways or simple piers projecting across the intertidal mud. The widening and linking of piers prompted the collective construction of continuous wooden waterfronts set well out from the shore. Waterfront construction also permitted much of the intertidal zone to be backfilled with waste material (often a source of valuable archaeological information). In the eyes of the local inhabitants, such back- filling soon became a laudable project of reclamation which could permit the construction of new stores and warehouses closer to the point of delivery. Such perceived improvements are still sought by small riverine communities today.

Figure 2.21

A particular advantage of the embanked waterfront was its ability to receive ships and cargoes during all states of the tide. However, once such a waterfront had been maintained beyond memory of its construction, respect for the natural course of the river was very easily lost. While the economic prosperity of a town continued to grow, competition for docking space prompted a further search for more facilities for waterfront mooring. From at least the 13th century, demands such as these were resolved in some European ports by daily harbour charges which prompted each crew to make a fast turn-round. A less prudent solution was also pursued through the competitive construction of a second generation of piers and outworks capable of outflanking the first. Such works would further confine and override the natural course of the river. All of these historic developments are pertinent to a legacy of shoreline management problems which will challenge some of Europe's major coastal cities in the 21st century.

Historic problems in the development of seaside resorts

It was during the 19th century that technology employed in waterfront reclamation was applied to a new array of shoreline defences. The development of Europe's rail network and the growth of seaside leisure brought landlocked visitors in search of romantic coastal scenery and the sunniest and sandiest of strands. In hand with these new interests came coastal development which sought to accommodate demands for residence and high living standards in successful new resorts towns such as Nice, Cannes, Rimini and Pescara. On the north-western coast of Atlantic Europe a less clement climate and a higher tidal range did little to dampen this new enthusiasm. With the aid of good rail communications every Parisian could seek vacation in Biarritz. Similarly, the London steam train could deliver hundreds of enthusiastic visitors and potential new residents to resorts such as Brighton, while those who had gained hard cash in Lancashire cotton mills could disperse their wealth along the artificial seafront bounding the sands of the new town of Blackpool.

While wealth and entrepreneurial enthusiasm was directed into Europe's new resort towns of the mid 19th century, problems of shoreline maintenance were soon to arise. Sandy beaches had been a prerequisite of the new coastal leisure industry yet houses, shops and hotels were ill- suited to this type of soft and unstable terrain. Where sea and sand threatened to advance, the primary solution was seen to be the seawall, or promenade. Using a technology won from the construction of waterfronts, harbour defences and land-drainage schemes, engineers now sought to hold both sand and sea in its rightful place. A particularly apposite example, embraced by this LIFE study is Soulac-sur-Mer on the Médoc coast of France (Study Area P17).

Where attractive sandy embayments bordered soft undrained wetlands, the engineering confidence of the 19th century remained undaunted. With the power of the steam engine Brunel and De Lesseps had built tunnels and canals and with this technology it seemed that any inconvenient wetland could now be ditched and drained. While supreme confidence was held in these techniques, new street grids and building plots were promptly laid out to meet a burgeoning demand for residential properties by the sea. Before long, some of these new properties were occupying notably low-lying land which was highly susceptible to coastal change.

Occasionally, the desire of purchasers and developers to find sandy beaches in the United Kingdom gave way to an alternative appeal of a special maritime micro-climate resembling the warm and envied coastal environments of Mediterranean France. This LIFE project identifies two UK examples, at Selsey and Ventnor, where development followed this course. Both are cases where imprudence in 19th century town planning has bequeathed significant problems to the present community. Plate 2i shows similar development at Sandown on the Isle of Wight, UK.

The opening of the 20th century brought some salutary lessons to authorities responsible for new coastal towns. Large seawalls and promenade structures built to protect property and to prevent cliff erosion were found to cause erosion on the beaches below. Where some groyne systems had been installed, they were found to cut off the supply of sediment to beaches situated downdrift. As a consequence, a demand arose for an ever-lengthening series of sea defences. In the wake of these shoreline works, coastal 'ribbon development' followed, sometimes with poor quality chalet townships developing on low-lying and flood-threatened land (Steers 1946).

The emergence of coastal protection policies and recognition of the time/depth factor

As mistakes were slowly recognised in the planning of some 19th century coastal developments in England, there gradually emerged a tentative interest in lost wisdom. Some early questions addressed the former character of the British coastline as witnessed and managed by earlier communities. This wisdom was embodied in the Steers Report, commissioned by the UK Government in 1945, and it led to Coastal Protection Act of 1949. The Great North Sea Surge or flood of 1953 served to underline the imprudence of erecting buildings on flood plains and coastal flats. A new message, readily recognised by archaeologists, was that new development should be confined to those traditional areas of settlement where past communities had, over the centuries, come to terms with the natural constraints and vicissitudes of the coastal environment. Unfortunately, prior to 1992, this policy was not clearly articulated and links with archaeological, geomorphological and palaeo-environmental studies not effectively forged.

Building work at Shanklin in the I860's, Isle of Wight, UK.
Plate 2i The construction of 19th century seaside resorts on the Atlantic seaboard of Europe gave little thought to sustainable locations which might resist the future advance of coastal processes. Building work at Shanklin in the I860's, Isle of Wight, UK.

On other sectors of the European coastline, approaches to coastal planning and protection developed along differing lines. The current LIFE programme has examined case examples in France and Eire where the history of 19th century coastal development followed notably different courses. In France, the highly desirable climate of the Mediterranean coast diverted the interests of 19th developers from the prospect of constructing resorts on the more turbulent Atlantic seaboard. Exceptions were developments in Deauville, Trouville and Le Touquet, the latter being appropriately suffixed 'Paris Plage'. Where the principal demand for resort development occurred on the Cote d'Azur, the coastal geology was mostly well suited to the needs of the builder. The low tidal range of the Mediterranean also proffered a lesser degree of coastal threat.

Set within a more favourable natural environment, coastal planning in France required less central direction and it remained free to follow regional and local policies while initiatives for coastal protection could often be generated at local or commune level. Prior to new interests in climate change, interest in archaeological and palaeo-environmental data from the coast has also been less acute. An exception, examined in this LIFE study, is the Médoc coast where a powerful and highly unusual process of dune movement threatens an array of coastal properties and landscapes. Here, a multi-disciplinary approach embracing the energies of both the local community and the archaeological and oceanographic departments of the University of Bordeaux is unravelling a long history of human adaptation to the contrary movements of the advancing sand dunes.

Implications of sea-level rise for the city of Dublin

On a more isolated sector of the European coast, the development of coastal planning and protection has followed yet another course. In Eire no significant demand for the construction of coastal resorts arose during the 19th or 20th centuries. Where coastal development has since occurred it has often been ad hoc and loosely controlled. Such developments have largely sought to gain desirable coastal views in the vicinity of small existing settlements. Land-use on the coast of Eire is almost entirely rural and this has meant that a strong demand for protective works has not arisen in the manner experienced on the more populous coast of the United Kingdom. The Shannon Estuary (Study Area P19) is nevertheless bounded by large embanked defences of early 20th century date while earlier systems of earthen seawall can be traced back to the 1670's. It is particularly apposite that this estuary has been targeted by the Government of Eire for intertidal archaeological investigations by its Discovery Programme. The Government of Eire has also acted quickly to assess the coastal implications of climate change, its first report being published in 1992. An important exception to the general pattern of coastal population in Eire has been historic and recent development around the city-ports of Dublin and Cork (Study Areas P22 and P20), Figures 2.22 and 2.23. At both of these locations habitation has expanded on to low riverine margins where there are significant implications of sea-level rise.

Implications of sea-level rise for the city of Cork

After the 1953 flood in Southern England, the conditions seemed right for targeted archaeological studies of those coastlines where past human settlement had been adapted, threatened or destroyed by rising sea-levels. Today, this area is defined by the 5 metre contour and it embraces some 50% of England's coast; Purnell 1995, Figure 2.24. Important archaeological precedents had already been established by a palaeo-environmental team working on prehistoric settlement and fluvio-marine sediments on the margins of the Wash. Professor Steers, too, was alert to the historical and archaeological dimension, citing the special qualities of flat coasts where a physiographic evolution could be traced by means of historical and archaeological evidence (Steers 1946). As a result of the Coastal Protection Act, public funds began to flow into studies and schemes concerning coastal protection and flood prevention. Professor Steers was quick to appeal for an integrated approach to coastal studies where he observed 'the archaeologist can give substantial help'.

Areas of susceptibility to flooding in England and Wales
Figure 2.24 Areas of susceptibility to flooding in England and Wales. Indicative flood plain map (after the Institute of Hydrology, Wallingford; and the Environment Agency). Black = areas that may be affected by a 1 in 200 year coastal flood event. Hatched = areas that may be affected by a 1 in 100 year fluvial flood event. (Crown copyright 2000; after Purnell 1968).

Sadly, and perhaps due to the retrospective nature of their training, most British archaeologists were to disregard the use of their skills as a means for predicting or modelling future events, yet the early work of Warren (1936) and subsequent studies by Akeroyd (1966), Churchill and Wymer (1965) and Balaam et al (1987) were all individual responses to this challenge. The study of in-situ archaeological remains in the intertidal zone was not perceived to be a priority during the 1960s, yet in the following decade an outstanding contribution was presented by Professor Charles Thomas (1985) in his work on the Isles of Scilly. Moreover, the vital links with coastal planners and engineers were never effectively forged during this period. After its inauguration English Heritage found itself excluded from a role below low water mark, where archaeological sites surviving in situ, in the context of submerged landscapes, were left to the mercy of mineral extractors, navigational dredgers, coastal developers using parliamentary bills and other sectoral interests. The same low water boundary discouraged the Local Authorities of the coast from enquiring into, recording or conserving the submerged geo-archaeological contexts which might serve to elucidate the nature, scale and pace of changes to their coastlines.

In some respects, many of the in situ archaeological structures in the intertidal zone may have suffered on account of their seemingly unprepossessing nature. In some cases they were found to be simple traps, fish weirs and trackways serving communities which may have been subsisting on the borders of their contemporary society, Figure 2.25. It is, however, the choice of their location, and indeed the elevation of their location, which has become so important to 20th century society (see Plate 2j).

Plate 2j The LIFE study has found that ephemeral archaeological structures such as this prehistoric wooden trackway in the Shannon Estuary, Eire, will not survive once they have been uncovered by a change in coastal processes. This problem of unsustainable loss is common to the European seaboard.

Perhaps in no other context can in-situ archaeological remains be so pertinent to society's present concerns. The survival of such remains in the intertidal zone tells us of sustained or sustainable shorelines. Conversely, the natural exposure of such remains attests processes which are leading to unsustainable shorelines. In either event, much of the national coastline will require new and continued archaeological monitoring if we are to interpret, successfully, the message which the past has written into the dynamic environments which characterise each of our coastal cells. This is a task which should be integrated fully into the work of each local coastal protection authority and its relevant coastal study group.

Achaeological features in situ

In 1992 the Government of the United Kingdom invited new initiatives for an integrated approach to shoreline and coastal zone planning. For this purpose, new central and local government frameworks were currently being formulated. These included new roles for the Department of the Environment, the Ministry of Agriculture Fisheries and Food and the local Coastal Protection Authorities. A Green Paper on England's Heritage produced in 1996 proposed a statutory role for local Sites and Monuments Records (EH 1996). This seemed to offer new opportunities to improve archaeological monitoring of the coast and to promote the ongoing gathering of intertidal and near-shore archaeological and palaeo-environmental data.

Within the Solent study area, the Hampshire and Wight Trust for Maritime Archaeology was one of the products of this new and refreshing national review; its explicit objective being the assistance of local authorities in the recognition and protection of the sub-tidal archaeological resources of the region. Unfortunately, since the issue of the Green Paper in 1996, no further progress seems to have been made by the United Kingdom Government on this proposal. In 1998 an announcement was made by the UK Government that the principles of European Agenda 21 were to be applied to all aspects environmental management and planning but since that date no moves have been made to empower or resource local authorities in monitoring the historic environment either on land or within the sub-tidal zone.

2.17
Wisdom of hindsight and the archaeological and palaeo-environmental perspective

The sources of time/depth evidence on the European coastline

This segment of the LIFE study has examined histories of coastal management and protection in France, the United Kingdom and Eire and it has also viewed some additional examples of coastal instability in the United Kingdom and Italy. Where coastal protection policies have developed in France, the United Kingdom and Eire, the LIFE project has traced a history of awareness that stems from 19th century coastal development. Where problems have arisen in coastal protection all share a common factor. This has been a failure by past, and some present, communities to recognise the time/depth elements which control the dynamic changes in the coastline.

Recent research on global warming and the prediction of climate change has revitalised interest in the nature, scale and pace of coastal change. The latter were terms employed by the UK Government in 1996 and they encapsulate a line of enquiry which is common not only to coastal managers but to archaeologists, oceanographers and palaeo-environmental researchers who have all come to examine the evolution of natural and humanly modified environments in the coastal zone. This LIFE project has sought to explore the common ground shared by these different groups of specialists and it concludes that a collective approach can be particularly rewarding in the following areas:

  1. Downwarped coastlines with submerged forests, rias, coastal wetlands and drowned occupation sites;

  2. Estuaries and debouchements;

  3. Spits and bars;

  4. Dune coasts;

  5. Sheltered embayments;

  6. Lakes;

    Examples of archaeological and palaeo-environmental evidence contained within these types of coastland are further discussed in Chapter 7. The project has also recognised that there are coastland types where the detection of datable archaeological and palaeo-environmental evidence of coastal change can be difficult to find. These include:

  7. Hard cliff-lines with wave-cut platforms and sea-caves;

  8. Coastline with soft cliff subject to landsliding and instability;

  9. Pro-grading shorelines.

Grasping the Holocene timetable of coastal change; a chronological study of sea- level change in the Solent study area, UK, during the past ten millennia

Earlier studies concerning the Solent region

The importance of changing sea-level has been recognised in Britain since the nineteenth century studies by Clement Reid (1892; 1893; 1896; 1899; 1913) on submerged archaeology and landscapes were at the forefront of world studies. Subsequently there have been global studies which have provided information on long-term Quaternary changes in sea-level brought about by oscillations between cold and warm stages (glacial and interglacial periods) during the last two million years.

Notable in this respect have been studies of Calabria in southern Italy and of Barbados in the Caribbean. Regional effects of tectonics, mainly brought about by movements along plate boundaries and fault lines, can have profound effects on local sea-level. Local changes can also be induced by glacio/isostatic movements or rebound. It is these broad global changes in which local relative sea- level movements have occurred which are responsible for the creation of the present coastline. This has, in the past, been viewed largely as a part of the geomorphological discipline aimed at understanding the development of the coastal landscape. The realisation of global warming, however, now means that established models of environmental and sea-level change have taken on a further and important role in providing standards against which future changes and measurement of rates of change can be measured.

The principles of sea-level change in the North Atlantic and North Sea region have, in general, been established. Relative sea-level rise (RSL) has been reconstructed for a number of areas of southern England forming the data base of post-glacial eustatic change for the English Channel region. These studies include the Thames Estuary (Devoy 1979,1980), East Kent (Long 1992), Romney Marsh and East Sussex (Jennings and Smyth 1990; Long and Innes 1993) and for the Solent region (Long and Tooley 1995; Long and Scaife forthcoming). From the south-west of England are the studies of Heyworth and Kidson (1987) and Healey (1995).

Until recently there has been only very limited systematic analysis of Holocene RSL and palaeogeography of the Solent and Southampton Water region, UK. Past studies have relied largely on data interpolated from other regions noted above, based on eustatic and isostatic changes caused globally by melting ice-caps and the effects of linear subsidence (Shennan 1989). This is perhaps surprising given the available sedimentary archive and the importance of this coastal region and the role in which past sea-level changes have had in shaping the Solent coastal zone.

Although the coastal sediments of mainland Hampshire have been a source of interest for over a century, the studies were speculative being based on little measured data. Furthermore, these early studies were carried out on stratigraphical sections exposed by chance, for example the King George Dock at Southampton (Godwin and Godwin 1940). Early studies go back to those of James (1847), Shore and Elwes (1889); Shore (1893); Andersen (1933) coupled with Clement Reid's' account of British submerged forests (Reid 1913). More recent studies have been based on more detailed recorded data and the advent of radiocarbon dating (Nichols and Clarke 1986). Only three studies have sought to use a combination of litho- bio- and chronostratigraphic data to reconstruct the environments of deposition.

The study at Fawley was made by Hodson and West during the construction of a large power station in the 1960's (Hodson and West 1972). Here, pioneer studies provided the only absolute chronology for the Solent until the last decade. Unconsolidated mineral sediments and peat were found at -5.00 to 0m OD and dated to 2130-1560 cal BC and 2410-1780 cal BC.

An incised palaeochannel at -7.63 to - 7.33m OD was also found and dated to 5500-4960 cal BC. Long and Tooley (1995) provided the first detailed RSL graph for the Solent from a sediment sequence at Stansore Point near Calshot Spit at the entrance to the Beaulieu River (Study Area P7). A well defined peat layer at c.-3.5m OD to -0.5m OD was dated to 4800 BP and 2470-1880 cal BC and 770-200 cal BC.

From these dates and altitudinal measurements, index points were established and a standard age/altitude graph of past sea-level was plotted. These data formed the basis for more detailed examinations undertaken during the evaluation of new sites in Southampton Water at Hythe Bay, Bury Farm and the Hamble Estuary relating to the proposed new container ship terminal (Long and Scaife 1996). At these sites, peats in the Hamble Estuary (Study Area P10) have been dated to 3380-2890 cal BP at -2.33m OD and at Bury Farm to 1450-1110 cal BP) at -0.11m OD. Older peat deposits at 4330-3980 cal BP were reported from Hythe Marshes (Long and Scaife 1996).

Data from these studies have been combined with similar work relating specifically to the north- east Isle of Wight. Work on the Wootton-Quarr archaeological project (Tomalin, Loader and Scaife forthcoming) identified numerous archaeological sites in the intertidal zone which necessitated a fuller understanding of how sea-level has changed in during the Holocene; especially the middle and later phases and how changes in palaeogeography have influenced human activity. Consequently, an in depth study of the litho- and biostratigraphy of peats and mineral sediments in the eastern Solent was undertaken. The availability of radiocarbon dates has allowed the construction of an independent RSL graph for the north-east Isle of Wight/Eastern Solent region (Long and Scaife in Tomalin et al. in press). These data have also been integrated into the existing Solent noted above.

Comparison between these Solent data and those from southern England as a whole has demonstrated significant differences with a notable east to west trend in date/relative sea-level. The integrated Solent RSL graph is based on radiocarbon dated index points taken from reliable dates (such as those from Hodson and West 1972 and Long and Tooley 1995) from Hampshire and from 9 dated index points from the Isle of Wight coastal sequences (this figure excludes the recently obtained Bouldnor radiocarbon series), Figure 2.26. The relationship of the western Solent Bouldnor Cliff and Yarmouth sequences to the generalised Solent and southern England model is discussed further in this text (Study Area P1).

The existing data demonstrate that there are significant regional and local variations in the age/altitude graph caused by tectonic movements at local and regional scales. Such variations might be regarded as merely of interest in assessing past coastlines and palaeogeography but, however, differences of even a few centimetres are likely to be extremely relevant to the effects of continued sea-level rise in specific (local) areas in the future. The data discussed above come largely from the eastern Solent and Southampton Water with little information from the western Solent. This fact, coupled with the gaps in the age/altitude graphs for the early and late Holocene and suggested importance of local linear tectonic changes, suggests the importance of study in the western Solent. The research objective of providing additional datum points to be integrated into the existing RSL graph and to produce a comparative graph from the western Solent were seen as important aspects for understanding past and future coastal development. The Wootton-Quarr project (Tomalin 1995 and Tomalin et al. forthcoming) initially highlighted the presence of sedimentary sequences along the north west coast of the Island providing a valuable sedimentary archive which could be used for achieving these objectives. The LIFE project has enabled sampling and detailed analyses of the sequences.

From these data (i.e. the RSL graph of altitude/date), there is now substantial detail available on the rates of sea-level change in the Solent. These dates allow estimates and palaeogeographical reconstructions of the coastline to be made by interpolation from the fitted curve. However, such estimates are complicated because of the relative absence of datable peat formation during the last two millennia. Consequently, there have been marked differences between long and short term estimates of RSL in southern England which are important to understanding future changes. To bridge this gap between the late prehistoric period and the present, other data points have been sought. These have included historical documentary records of structural features which have been affected by sea-level rise (eg. St Helen's Church, Isle of Wight; Tomalin et al. in press).

Integrated relative sea level graph  for the last 9,000 years, Solent, UK

The radioactive element Lead 210 has also provided an indication of recent sedimentation rates in Southampton Water and finally, more recent biostratigraphical markers occurring in near-shore and saltmarsh sediments have provided useful datum points from which sediment accretion rates can be gauged. Of particular note is the expansion of pine pollen which occurs in many terrestrial peat mires from 1750 onwards. This is clearly related to the planting/introduction of pine for ornamental reasons in private estates and later for forestry purposes (Haskins 1978; Scaife 1980). This expansion of pine pollen has also been identified in salt marsh sediments at Poole Harbour (Long et al. 1999) and Newtown Creek, Isle of Wight. This coupled with other known ecological changes such as the expansion/colonisation of Spartina anglica from the late nineteenth century has been used as proxy data for sea-level rise and is further discussed in Volume 2 of this report.

Between c.1510-1260 cal. BC a rise of 1.99m at a rate of 0.59mm per year can be detected. It is estimated that between AD 1750 and AD 1890 sediment accretion rates rose from 0.29mm to 1.14mm per annum and further to 7.17 mm after 1890. Whilst these data have relevance for sea-level change and saltmarsh ecology, with Spartina colonisation the increased sedimentation rates and thus higher saltmarsh levels may be particularly important to future coastal management. Without sedimentation and rapid sediment build up, rising sea-level will significantly encroach the middle and upper saltmarsh zone for longer periods, killing off the stabilising saltmarsh plant communities, with potential therefore for substantial erosion of exposed unconsolidated, saltmarsh sediments.

Holocene relative sea-level changes in the western Solent

Coasts evolve in response to changes in the height of the sea relative to the land (relative sea- level) as well as changes in sediment supply, wave and tidal energy and changes in coastal morphology such as the build up and breakdown of spits and bars. Over Holocene timescales, relative sea-level (RSL) has risen by at least 20 m in southern England, with most of the rapid rise recorded during the period between 10,000 and 6000 cal. yrs BC as water from the melting ice sheets of the northern and southern hemisphere continued to cause an expansion in ocean volume.

These millennial scale changes saw significant movements in shoreline position in the English Channel and associated changes in palaeoenvironments. They caused the formation of the Solent that we know today. Evidence for these changes in RSL is widespread throughout the western Solent and here we present and interpret new RSL data collected from Bouldnor, Yarmouth Spit and Newtown (Study Area P1), combined with as previously published material from Stansore Point (Long and Tooley 1995). The intercalated peat and mineral sediments of Bouldnor Cliff, Yarmouth and Newtown Creek from the western Solent and from Stansore Point at the mouth of Southampton Water have provided information on the changes vegetation and local environment during the middle and later Holocene and of relative sea-level change during the same period. The latter sea-level change data are discussed here whilst data from the pollen studies are addressed in Volume 2 of this report.

Methods of analysis

The material used in this analysis comprises sixteen radiocarbon dated samples collected from either the top or the bottom of organic peat deposits (see Momber, Volume 2). These dated horizons are described as transgressive contacts, where organic semi-terrestrial sediment passes upwards into predominantly minerogenic intertidal sediment and regressive contacts where the reverse occurs. Pollen and diatom data is described in Volume 2 of this LIFE study. These data provide important information on the changing palaeo-environments associated with these changes in sediment and are essential for the meaningful interpretation of the radiocarbon dates and trends in RSL derived from them.

For each dated horizon an estimate is made using the pollen, diatom and other sedimentological data of the altitudinal relationship between the horizon and a tide level at the time of its formation. Typically, in this analysis the majority of samples represent habitats formed at, or close to, the former elevation of mean high water of spring tides (MHWST). These palaeo- environments are characterised in the pollen data by transitional vegetation communities between the semi-terrestrial (often alder carr) environments and the more open saltmarsh communities.

In addition to the above, two of the deepest samples in this analysis come from within a deposit of peat located on the present sea-floor of the western Solent (from Bouldnor and Yarmouth Creek, Study Area P1). Pollen analysis of this peat demonstrates that it began forming in response to waterlogging of the landscape caused by ponding back of water due to RSL rise before 8,000 BP. The ensuing waterlogged conditions promoted organic accumulation and preservation. Dates obtained on samples of wood found within this peat may have formed slightly above or slightly below MHWST depending on coastal configuration at the time of formation. Although the altitudinal relationship of these samples is not as clear as those higher up the sequences, because RSL was rising rapidly at this time, they nevertheless provide a reasonably good indication of the age and altitude of RSL during the initial rapid phase of inundation at the end of the Boreal/early Atlantic period (Chronozone Fl. Ic/II).

Today the tidal range within the western Solent varies between the sample sites by a matter of decimeters. In this analysis it is assumed that the variations seen today have remained constant during the past, but, however, this hypothesis is yet to be tested. It is likely that the major changes in coastal configuration that have taken place since c. 7000 cal. yrs BC were associated with larger variations in tidal range than those observed at present. By way of example, a recent analysis Van der Molen (1997) reports that tidal distortion across large saltmarshes with embayments or narrow inlets can cause monthly mean high water to be up to 0.55 m lower near the upland border of the Great Massachusetts marshes compared with the simultaneous high water level on the tidal flat. It is probable that the extent of saltmarshes in the western Solent was greater in the past. This raises the probability that MHWST was more variable than at present and that, in particular, samples collected from the inner parts of large saltmarshes may have formed below similarly-aged samples forming close to the open coast.

Local and regional factors

As noted above, changes in the relative sea-level (RSL) record the interaction of a range of factors, including changes in ocean volume and tidal range. These are examples of regional (eustatic) and local (tidal range) factors. Other examples of regional factors that are controlling long-term trends in RSL include uplift or subsidence caused by isostatic movements (loading and unloading of the earth's crust by water, ice and sediment). In southern England, long-term rates of crustal subsidence are estimated to be between 1 and 2 mm per year during much of the Holocene (Shennan 1989), although limited data existed from the Solent in this analysis (Long and Tooley 1995). Essentially, this is a long term average which cannot, on its own, explain short-term (century to decadal) changes in the rate of RSL change. Other more local factors capable of causing short-term changes in RSL include alternations in the magnitude and frequency of storm events, as well as changes in coastal configuration.

Another important local factor which influences the apparent rate of RSL change is sediment compaction. This processes, which is dependent on the nature of the sediment, the rate and magnitude of sediment accumulation as well as pore-water history, may cause sediments to be displaced downwards from their original elevation by decimeters or meters. This natural compaction may be aggravated further during the process of sample collection -core material may be compacted by up to 50% during retrieval and extrusion from sampling chambers. These compaction factors result in an over-estimate of the rates of long-term trends in RSL. Generally speaking, the impact of compaction can be minimised by collecting samples from deposits which lie close to, or immediately above, an uncompressible surface such as bedrock. In the current investigation, only the deeper samples from Yarmouth and Bouldnor come from the so-called basal peats; the remaining samples will have all been lowered somewhat from their original elevation.

Together, the local factors referred to above have the effect of introducing unwelcome "noise" into the RSL record, obscuring the long-short term signal of change. One way to estimate the importance of these local factors is to collect data from several sites and, through their comparison, establish outliers from general trends which may record more site specific processes. In this analysis we combine data from four sites: Yarmouth, Bouldnor, Newtown and Stansore Point.

Holocene trends in mean tide level

Having corrected the sixteen dated samples for the west Solent region for variations in tidal range, it is possible to plot them as a graph depicting changes in mean tide level (MTL) during the last 7000 cal. yrs BC (Figure 2.27). Age and altitude errors are not shown but the size of the symbol approximates (an age error of 200 cal. yrs and an altitude error of 0.50m). Until further data are available regarding the effects of compaction as well as changes in tidal range, a more sophisticated analysis of the latter errors is not practical.

The resulting graph depicts MTL rising steeply from c.-15 m OD at c. 6500 cal. yrs BC to c. -6 m OD by 4000 cal. yrs BC. The rate of rise then appears to drop, with MTL rising to -2 m by c. 500 cal. yrs BC. No data exist in this analysis for the last 2000 cal. yrs, but MTL has risen by no more than c. 2m during this interval, at an average long-term rate of 1 mm per year.

MTL graph for the last 9,000 years for the Solent, south coast, UK

There is, as is to be expected, a degree of scatter in the age/altitude data presented in Figure 2.27. One particular outlier is a date collected from Newtown core A at -7.4 m OD (OXA 4778, 4570 (65 BP, 3500 to 3040 cal.yrs BC). This datum point appears to be 2 m or so below the long-term trend observed within the data. A possible explanation may be that this dated horizon has been lowered by significant sediment compaction but there is also a possibility of false stratigraphy in the core sample and this is discussed in section 2.15, fourth sub-section.

Coastal evolution in the western Solent

The deepest sediments of postglacial age recorded in the western Solent comprise freshwater organic sediments recorded immediately above bedrock. Samples collected from these environments in this study are recorded beneath Yarmouth Spit as well as from Bouldnor. The bio- and lithostratigraphy at these sites is consistent in indicating a progressive waterlogging of the landscape, no doubt caused by the rapid increase in RSL observed at this time. During the initial phases of this waterlogging the rate of organic accumulation was able to keep pace with RSL rise, but with time these freshwater communities were inundated and shoreline retreat occurred.

At Bouldnor and at Yarmouth the ensuing period is characterised by a dominance of minerogenic saltmarsh and mudflat sedimentation, as RSL continued its upward trend. There is a pronounced lack of radiocarbon dates during the interval between c. 6200 cal. yrs BC and 5000 cal. yrs BC. This may reflect inadequacies in the sampling framework to date; it may also reflect the real effects of a rapid change in shoreline position forced by RSL rise.

After about 5000 cal. yrs BP there is a renewed phase of organic accumulation, and at several sites, thin or more substantive accumulations of peat are recorded. There is no doubt that the long-term rate of RSL continued upwards at this time, albeit at a somewhat reduced rate. Phases of organic accumulation are recorded after this time at all four sites analysed here, although the duration of organic accumulation varies between sites. Despite this variability, the extent of salt- and freshwater marshes which experienced organic sedimentation probably reached its maximum in the western Solent at this time.

The widespread nature of the change in sedimentation recorded reflects the operation of a broad change in the rate of RSL, but more local factors controlled the duration of peat formation at any single site. Given the likely importance of these local factors, one should be cautious not to over-interpret the wider significance of short-lived phases of organic sedimentation in terms of fluctuations in regional RSL. For example, changes in the availability of sediment supply to the coastal lowlands will have been an important influence on the rate of sediment accretion and hence the ability of coastal marshes to accrete in the face of a prevailing upwards trend in long-term RSL. Other local factors, such as the migration of tidal channels or the breaching and closure of barriers (as at Yarmouth and also at Stansore) may also have controlled short-lived phases of shoreline development.

After about 2500 cal. yrs BC, the Isle of Wight coastal sites show a return to coastal minerogenic sedimentation with no further significant organic sediments recorded during the late Holocene. On the mainland, at Stansore Point, the freshwater peats in this protected site are finally inundated by about 0 cal. yrs BC. Long et al. (in press) have argued that this shift to minerogenic sedimentation is recorded throughout the Solent and, though time transgressive, saw the demise of coastal wetland sedimentation throughout the region by c. 0 cal. yrs BC. The cause of this decline is not certain; it may reflect an acceleration in the rate of RSL, a reduction in sediment supply, or perhaps an increase in sediment reworking. Whichever the cause, during the late Holocene the coastal environments of the western Solent were radically different from their mid Holocene counterparts, with significantly reduced expanses of coastal wetland and saltmarsh. One can envisage an associated shift in the floral and faunal diversity of the region from the mid to the late Holocene.

Implications for future sea-level change in the western Solent

Future sea-level rise, as predicted using global ocean volume estimates (IPCC 1995) suggest a globally averaged increase of 0.49 m by 2100 AD using best-guess scenarios of change. To this global estimate must be added the long term rate of isostatic submergence in the Solent region, of about 1 mm per year. Together, these factors suggest that regional changes in RSL within the Solent will approximate 0.59 m by the end of this century. This may not sound much, but set against a late Holocene increase of not more than 2 m in the last 2,000 years this is indeed a significant increase. These estimates are, it must be stressed, based on a global model of ocean volume expansion which has uncertainties when applied to a regional context. These uncertainties include the extent to which regional perturbations to these global trends caused by changes in ocean heating and salinity will modify globally averaged values. At present regional scale estimates of future ocean volume change are in their infancy.

Future research

This analysis has demonstrated the long-term pattern of RSL in the western Solent during the Holocene. There are several limitations to our present knowledge which would benefit from further investigations and these are now highlighted:

The early Holocene development of the Solent

This analysis has benefited from the collection of deep cores from Yarmouth and also from the extraction of samples from the submerged cliff at Bouldnor (Study Area P1). The deep basal deposits provide a tantalising insight into the plant communities and environments which existed in the western Solent during the initial expansion of the English Channel. Further evidence for this phase of inundation is preserved within the sediment contained within buried palaeovalleys within the Solent (Bellamy 1995). Future research should concentrate on the collection and analysis of these and other equivalent deposits in the Solent and elsewhere along the eastern English Channel coasts. Such data are of importance, providing information regarding coastal evolution during the initial transgression of the Holocene, as mainland Britain became isolated from Europe at the end of the last glaciation. Information from this period is also of importance in constraining geophysical models of relative sea-level change and crustal movements, as well as variations in palaeotidal range.

High resolution studies of short-term changes in RSL

This analysis has highlighted the variable response of the coast to changes in regional RSL. However, at present we are not able to resolve what processes have been responsible for these changes – do they record changes in sediment supply and site exposure for example, or do some record short-lived but important variations in the rate of RSL driven by changes in climate or other forcing factors? The sampling completed to date has established the broad framework of change. What is now required is a more targeted analysis aimed at assessing the controls on site sensitivity to RSL and other factors. This information will not only illuminate investigations of Holocene coastal change, but will also provide an indication of how variable future coastal responses to RSL change may be.

The role of organic communities in coastal evolution

This analysis demonstrates clearly that freshwater and saltmarsh environments were more expansive during the early and mid Holocene and these supported a more diverse fauna and flora than at present. Moreover, the more expansive saltmarshes were clearly able to protect the coast against the continuing upwards trend in RSL for hundreds if not thousands of years. There is much interest in so-called soft-management techniques at present and we should be exploring the ways in which the self-sustaining communities of the early and mid Holocene interacted with the coast, with the objective of exploiting their potential for future coastal management initiatives. The Solent is an ideal laboratory within which this type of analysis can proceed, supporting a wealth of deposits which accumulated in a range of depositional environments throughout the Holocene.

2.18
The question of sustainability: potential, constraints and irreplaceable losses to the common European palaeo-environmental resource

The issue of sustainability

Unlike the natural environment the historic environment of Europe cannot be regenerated nor measured in its totality. Important elements of this environment can be totally concealed and this makes quantification and protection particularly difficult. The concept of achieving the sustainable use and management of irreplaceable environmental assets is an objective which stems from the Rio Convention on the sustainable management of the Earth's resources. The philosophy of sustainable management can be summarised by the view of the Brundtland Commission which has proposed that the activities and choices made by the present generation should not prevent future generations from making their own choices. In short, this has been interpreted as 'not cheating on your children'. The principles of sustainability have since been adopted as European Agenda 21 and it has since become incumbent upon each European member state to prepare and implement an Agenda 21 policy which will achieve a common shared objective.

In the United Kingdom the policy been largely re-cast as Local Agenda 21. This is a move which has suggested to some that responsibility and emphasis may be in the process of a shift from Central Government to the lesser resources and variable capabilities local government. Government guidance to local authorities on this topic invites the formulation of Local Agenda 21 policies, recognising that the natural environment and the 'cultural heritage' are important elements to be addressed.

In the context of the present LIFE study the issue of sustainability has been thrown into sharp relief. The new intertidal surveys at Wootton-Quarr (Study Area P2) and Langstone Harbour (Study Area P12) have identified high intensities of formerly concealed archaeological sites set within Holocene sediment archives which offer particular scientific value for the study of long- term environmental fluctuation and coastal change (Tomalin et al. in press; Allen and Gardiner in press). When extrapolated, these quantified samples suggest that the number of nationally important sites on the southern shore of the Solent may readily approach one thousand. This striking number of exposed sites raises perplexing questions concerning the management and sustainability of this resource. There is also a need to appraise the particular value of these sites in the construction of a local coastal change chronology.

This LIFE study has already observed that in-situ structures in the intertidal zone are, by virtue of their character and their elevation above Ordnance Datum, a vital measure of the sustainability of local coastlines. The study of such sites and the interpretation of their archaeological or earth science characteristics should fulfil a key role in each Shoreline Management Plan. At Yarmouth in the western Solent, in-situ remains have been located in the sub-tidal zone at an approximate depth of -3m OD. Here underwater survey has located Neolithic wooden structures which can be traced on the seabed where they appear to approach the more accessible but less stable context of the intertidal zone. On the eastern end of the Island, at Bembridge, a raised beach, now lying inland, serves to extend the chronological range of local and regional coastal change history (Preece and Scourse 1987).

On the Solent coast at Quarr and Newtown, a submerged coastal reed marsh of early Neolithic date accommodates an array of fish traps and wooden trackways. These are overlain by peat and oak forest belonging to the 4th and 3rd millennia BC. In-situ wooden structures in these intertidal contexts are in an excellent state of preservation with tree bark and axe signatures surviving in pristine condition. At Wootton-Quarr, further wooden structures erected during Later Bronze Age, Iron Age, Roman and post-Roman times survive in a similar state.

Now that these structures have been recognised, the question raised by this LIFE study is can their survival be sustained in-situ? Their presence tells us that their survival has been sustained to the present day, yet even when viewed over a future period of one or two decades, the answer must be that those structures which are now exposed will not survive (see Figure 2.28). If remedies are sought to counter the loss of these types of archaeological site then the issue of accommodating the intertidal and sub-tidal heritage will need to explicitly expressed. The regular review of Shoreline Management Plans offers one practical means of promoting this objective but it is also clear that all Member States will need to provide new and clear instructions on the needs of the historic environment within European Agenda 21. Archaeological and palaeo-environmental resources are a dual element which must be fully addressed within this agenda.

Some threats to in-situ archaeological and palaeo-environmental resources in the coastal zone

Whilst monitoring and field recording may mitigate the natural loss of exposed archaeological and palaeo-environmental deposits in the intertidal and sub-tidal zones, it does not address the problem of preserving these resources in-situ. There is, moreover, a wide range of humanly induced threats many of which are poorly restrained. Some of these operations are subject to a degree of control or mitigation arising out of EC 85/377 EEC, a European Directive passed in 1985 with the intention of assessing and regulating detrimental environmental effects. These regulations are guided by two lists or 'schedules' in which potentially detrimental activities are identified.

In the United Kingdom this directive is embodied in Statutory Instrument 1988 no. 1199 , Town and Country Planning (assessment of environmental effects) Regulations. These regulations were drafted before the adoption of European Agenda 21 and a need for revision is now clearly evident. The schedules of 1988 tried to 'second guess' the processes or operations which would incur detrimental environmental effects. Today the term 'environmental effect' would probably be substituted by the term 'sustainability' yet the principle remains essentially the same. A number of specific processes and operations pose unsustainable loss to those non-renewable resources which are of particular value to the study of coastal and climatic change and a cultural heritage of the coastal zone. To clarify these problems of sustainability an array of detrimental processes and practices noted during this LIFE project has been set out below.

Submerged peats and other sediment archives(shaded) exposed to damage and diminution on the floor of the Western Solent , UK
Figure 2.28 Submerged peats and other sediment archives(shaded) exposed to damage and diminution on the floor of the Western Solent , UK

Losses sustained from coastal processes

Natural processes may be seen as a substantial single threat to preservation in situ but we should remember that whilst nature's erosive functions can be destructive, her depositional processes can also be benign. An example might be the daily transport and accretion of fine sediment which the river Selune discharges into the Baie de Mont St Michel. This sediment provides the means of enveloping and preserving an array of materials on the foreshore of the medieval citadel (Plate 2k).In our Isle of Wight case study, we have observed that the Neolithic land surface and its overlying peat often survives in a state of equilibrium below the mean low water mark. Here, it seems, is an archaeological resource which is sustainable provided that it is protected from human interference. The human agencies posing threat in this potentially benign environment are those responsible for dredging, trawling and interventions in the local pattern of sediment transport and deposition. To these may be added polluting activities which are responsible for habitat depletion and induced biological change. These threats can only be effectively constrained by amended legislation or procedures which will support improved local curatorial controls.

barrel containing post-medieval archaeological material
Plate 2k This barrel containing post-medieval archaeological material demonstrates the persistence of a benign intertidal depositional environment in the vicinity of the historic citadel of Mont St Michel, France.

Losses induced by marine aggregate dredging

Dredging for commercial and navigational purposes can be particularly damaging to in situ archaeological remains in the intertidal zone as well as those which lie in the direct path of the dredger as it moves across the seabed or riverbed. In the United Kingdom flaws can be identified in the consultation procedures which currently permit loss of archaeological and palaeo-environmental resources in these concealed contexts. The Department of the Environment, Transport and the Regions (DETR) does not have in-house archaeological expertise to participate in the consultation procedure for the licensing of offshore mineral extraction. Moreover, English Heritage has not been empowered to deal with the historic environment contained within the sub-tidal zone. The impact posed by the extraction of minerals is recognised by European Directive EC 85/337 EU which seeks to assess and regulate environmental effects. Within these regulations the extraction of minerals is identified in schedule 2 (2c) but its inclusion in this schedule rather than Schedule 1 gives any environmental statement no more than optional status. This can be contrasted with the obligatory status accorded to operations cited in Schedule 1. Nevertheless, in the United Kingdom it is the policy of the DETR to require environmental statements for all applications before considering the licensing of off-shore mineral extraction. The off-shore operators also have a voluntary code of practice which acknowledges the importance of sea-bed archaeological sites.

In the United Kingdom local planning authorities, with ownership or access to local Sites and Monuments Records, have the potential to recognise and monitor archaeological resources in the sub-tidal zone. As Statutory Consultees they also have the opportunity, but not necessarily the resources, to advise the DETR on the potential effects of an off-shore mineral extraction proposal. Unfortunately, while enjoying a statutory right or option of response in this matter, these local authorities have no statutory requirement to do so. Many local authorities on the English coastline have Sites and Monuments Records (SMRs) which include extensive and 'live' inventories of the marine or sub-tidal archaeological resource. Some also maintain environmental records which can embrace the coastal palaeo-environmental resource. Should they wish to do so, these authorities can gather live information from the local maritime community on an on-going basis. The 'live' nature of these local SMRs makes their archaeological input into the mineral licensing procedure a crucial one. Unfortunately the coverage offered by these coastal SMRs is currently far from consistent, for some local authorities lack the motivation or resources to extend their data-gathering capacities into the sub-tidal zone (see Figure 2.29). These inconsistencies admit a serious lacunae in the gathering of evidence by which unsustainable loss might be recognised, measured and ameliorated.

An example of the need for the statutory extension of coastal Sites and Monuments Records can be drawn from the LIFE study area in the western Solent. At the mouth of Newtown Creek, Neolithic wooden trackways (Plate 2l) and other prehistoric structures became exposed and eroded in the 1980s after off-shore sediment paths had delivered a limited annual flow of sand and shingle to the western spit (Bray et al. 1991). The source of the path lies in the vicinity of Solent Bank, an area where mineral aggregate on the seabed was extracted by dredging under a license issued before1991. Active gathering of data in this off-shore zone should have alerted the mineral licensing body to the possibility of heritage-loss prior to this date (Figure 2.30).

Losses induced by navigational dredging

The extraction of seabed gravel or 'minerals' at the mouths of Europe's major and minor ports is a practice which is destined to increase in intensity. Commercial needs for the improvement of European deep sea ports has naturally increased the demand for deeper and more regular dredging. This LIFE project has drawn attention to the fact that those areas of the coastline which were most intensely favoured by Europe's prehistoric and historic communities are often the very same locations which are under greatest demand for commercial development today. Moreover, modern harbours are frequently sited at the heads of drowned valleys or rias where sediment archives, rich in palaeo-environmental evidence, have been preserved on the floor of the inlet.

Offshore archaeological  sites on a section of the east coast of Kent

The review embodied in this LIFE study has shown that the sea-floor at the mouth of prehistoric and historic harbours and ports can be strewn with important archaeological evidence which is highly vulnerable to the destructive powers of the dredging process. Examples of these seabed strews or artifact aprons include the mouth of Rotterdam Harbour, The Roman anchorage known as the Magnus Portus within the Solent (Figure 2.31), the mouth of Poole Harbour (Figure 2.32) and the El Cabo anchorage on the Cantabrian coast of Spain (Adams et al. 1990; Tomalin 1997; Velegrakis . 1992; Bueno and Salis, 1975). The phenomenon must be suspected to be widespread but where navigational dredging is carried out the opportunity for archaeologists to observe, monitor, retrieve or mitigate is currently remote.

The protective shingle banks at Newtown, Isle of Wight
Plate 2l The protective shingle banks at Newtown, Isle of Wight, UK, have been starved of shingle from their offshore source. The saltmarsh and old header channels inside the natural harbour are now degrading. Offshore (right) Neolithic trackways are now being exposed and destroyed.

Further problems are evident elsewhere in the monitoring and evaluation of the unsustainable losses which occur in the process of sea-bed and river-bed dredging. In the estuarine environs of the Gironde, France (Study Areas P15 and P16), archaeologists have been shown, in rare instances, archaeological material recovered during the navigational dredging of the river-bed. This information can be gleaned from manual workers but it is not readily available from the management of these operations. An identical situation is reported by Adams where dredging operations have been conducted in the mouth of the Thames. Fear of regulation or the confiscation of portable antiquities are important factors in non-disclosure. These factors apply to both the dredging and fishing industries.

Sediment transport pathways of the Solent Bank

The problem of damage limitation is exacerbated by the fact that the submerged action of the dredger is unseen, its impact being concealed from observation on the seabed. In the United Kingdom the operation of navigation dredging is self-regulated by the harbour authorities whose remit does not normally include responsibility for ensuring that their activities do not impose detrimental or unsustainable impact upon the concealed historic environment. The dredging process can also impose translational or collateral damage to adjacent areas of the coastal zone. Where the depth of dredging is increased within a regularly dredged navigational channel a broadened angle of repose will be formed in the channel side. This can disinter in-situ archaeological or palaeo-environmental material in the adjacent seabed. Deepened navigational channels can also alter the tidal regime which can impose detrimental impact on the nearby shore.

Plan of Solent seaway, UK

lan of approach to Poole Harbour

This collateral damage is very well demonstrated at the mouth of Wootton Haven where the loss of in-situ prehistoric and Roman intertidal deposits could be equated with an accelerated process of draw-down. This change was subsequent to a major improvement of the navigational channel for a car ferry (Plate 2m). This was dredged to within 50 metres of the archaeological features in the intertidal zone (Figure 2.33). Dredging of this channel was essentially a self-regulatory function approved by the Harbour Authority which required no curatorial input to protect archaeological or palaeo-environmental resources. The result in this instance was unsustainable loss of an archaeological and palaeo-environmental resource mitigated by rescue recording funded by English Heritage, the local authority and also the ferry company. The enabling legislation in this case dated from the late 19th century and it demonstrated the thoroughness which will be required if the reforms required by European Agenda 21 are to be made truly effective.

The archaeological and palaeo-environmental deposits at the 			mouth of Wootton Haven, Isle of Wight, UK
Plate 2m The archaeological and palaeo-environmental deposits at the mouth of Wootton Haven, Isle of Wight, UK, are a stressed intertidal resource.

Losses induced by bait-digging

Until recent times, bait digging was a sustainable occupation modestly pursued by local fishermen. Recently, these activities have been expanded by some entrepreneurs to become a semi-commercial process in which bait is exhaustively sought on a highly organised scale. On the Isle of Wight, estuarine beaches are extensively dug and the recovered worms can be shipped in bulk to mainland markets. On the Isle of Wight coast bait-digging holes were observed to be up to a metre in depth and diameter. The enthusiasm with which they are dug can soon produce a landscape which looks like a miniature lunar surface. In the archaeological study area on the Wootton-Quarr coast (Study Area P2), the holes were found to be widespread and they were undeniably destructive of archaeological and palaeo-environmental deposits otherwise surviving in situ at these particular locations. In the United Kingdom bait-digging is an operation which lies outside the control of the present Town and Country Planning Acts. In some parts of the United Kingdom local bye-laws have been used to control this practice but the issuing of such bye-laws can also evoke local objections to new restrictions which seek to curb that which has been seen to be a long established local custom.

Losses incurred by harbour works and developments

Harbour developments can be potentially destructive of in-situ archaeological and palaeo-environmental deposits in the intertidal and near-shore zone. This threat is also partially recognised in Directive EC 85/337 EEC where works concerning a trading port are cited in Schedule 1 (8) while harbours and yacht marinas are cited in schedule 2 (10d and 10j). This directive of 1985 places differential emphasis on those essential environmental assessments which are required for projects listed in Schedule 1 while the case for an environmental assessment is no more than optional for projects which are cited in Schedule 2. Drafted before the adoption of European Agenda 21, this directive and its accompanying schedules attempts to second guess those developments which will impose the greatest danger of what would now be considered an unsustainable loss to a non-renewable resource. The optional nature of Schedule 2 poses considerable threat to elements of the historic environment or palaeo- environment especially where they are concealed within the coastal zone. This threat applies in particular to situations where palaeo-environmental material is contained within a sediment archive (see Plate 2n).

This LIFE project has demonstrated that sediment archives are a vital 'document' of long-term coastal and environment change, yet the present EC directive provides no instructions which will provide for the identification, protection or 'reading' of this record where its survival has become threatened by a development proposal which is classified under Schedule 2. The optional nature of Schedule 2 can also invite an element of complicity where both the planning regulator and the developer are anxious to see a development proceed as quickly as possible. The open option offered by Schedule 2 allows a means by which the need for an environmental assessment can be discounted and this can leave the concealed archaeological or palaeo-environmental resource abandoned to unrecorded loss.

The harbour at Ancona on the Adriatic coast of Italy
Plate 2n The harbour at Ancona on the Adriatic coast of Italy. Because of geographic necessity the demand for modern harbour development is often focused upon precisely the same locations as Europe's historic harbours. Rich cultural and palaeo-environmental resources are readily threatened and destroyed at these locations, especially where they are concealed from sight, below ground or underwater. 1 = Cultural remains of medieval buildings on an ancient harbour frontage. 2 = Cordoned area covering undisturbed cultural and palaeo-environmental resources. 3 = Developed modern harbour destroying the historic environment of the former inter tidal zone. 4 = The sea bed cultural and palaeo-environmental resource of ancient Ancona, now stressed by the navigational requirements of the modern deep-water port.

In the United Kingdom, threat to the archaeological resource above mean low water mark may be mitigated by powers exercised under the Town and Country Planning Acts. These powers are guided by specific archaeological advice embodied in Government's advisory note PPG16. A particularly important instruction in PPG16 is the advice to the Planning Authority that 'prospective developers should in all cases include as part of their research... before making a planning application, an initial assessment of whether the site is known or likely to contain archaeological remains'. The advice goes on to emphasise that 'Local planning authorities can expect developers to provide the results of such assessments and evaluations as part of their application for sites where there is good reason to believe there are remains of archaeological importance'. While this advice applies specifically to archaeological resources rather than palaeo-environmental ones, its principles seem highly appropriate to any reform or up-date of Directive EC 85/337 EEC. This directive requires 'specified information' for the preparation of environmental statements (Schedule 3). While impacts upon flora, fauna, soil and the cultural heritage are all specified, adverse impacts upon any palaeo-environmental resource are not identified. This omission is detrimental to the interpretation and sustainability of these resources, especially in the vulnerable and changing environment of the European coast.

Losses induced by coastal protection works

Coastal protection works are neither cited in schedule 1 nor schedule 2 of Directive EC 85/337 EEC. This omission may arise from a presumption that these works might only be beneficial to all elements of the environment. In England and Wales, as in France, such a view is refuted where the more complex environmental sensitivities of coastal defence schemes have been clearly recognised during the preparation of Shoreline Management Plans. Where potential conflicts have emerged the guidelines to these plans have presumed that solutions should be achievable through 'continued consultation with all parties'. Unfortunately these consultations can be of a non-statutory and voluntary nature and they have presumed that all of the 'parties' concerned have been sufficiently equipped to allocate time and resources to resolving the problems raised by the plans. Where the externalisation of archaeological services and Sites and Monuments Records has been promoted in some local authorities, a further need has arisen to reinforce curatorial vigilance over the threat of unsustainable loss to the historic environment in the coastal zone. Where no statutory requirements exist this can be a ready target for 'efficiency savings', the loss in this case being borne by the concealed and susceptible archaeological resources which are contained in-situ in the intertidal and offshore zones.

Where coastal defence strategies have developed through the completion of Shoreline Management Plans, there has often remained an ambivalence between a traditional view led by 'tangible benefits' or property values and a more enlightened and flexible policy in which the inevitability of sea-level rise and coastal change has been acknowledged. The preparation of Shoreline Management Plans in England and Wales has introduced a new and welcome rationale to a formerly fragmented national policy but it has also introduced expectations that an immediate prescription or solution can be found for every management unit of the coast. An observation of this LIFE study is that some of these prescriptions and perceived solutions can lack sufficient depth of hindsight. In these cases there remains a need to gather sufficient palaeo-environmental information on the long-term behaviour of the particular process unit or sub-cell.

In the United kingdom the fora of regional consultation, based on coastal cells and sub-cells, undoubtedly mark a significant advance towards an informed consensus on coastal management but the concept of wisdom of hindsight has yet to be firmly grasped in the formulation of Shoreline Management Plans. A recent review of the archaeological content of these plans by Firth (1999) has shown substantial weaknesses in the comprehension of coastal archaeological and palaeo-environmental data. These weaknesses are specifically discussed in a later section of this text.

Practical experience of coastal protection works in the LIFE study areas has been instructive. In the Isle of Wight it has been noted that coastal defence works on the Ventnor coast (Study Area P4) had cut through a highly significant archaeological context which had formerly yielded evidence pertinent to the overall dating of landslide events in the entire process unit. Where more recent coastal defence works had been commissioned in east Wight, a full appraisal of concealed archaeological resources was fully embodied within the scheme.

On the Médoc coast (Study Area P17) the instigation of coastal defence works was largely left to the initiative of individual communes. At L'Amelie new hard defences had been constructed without any regard for the historic information imparted by the 2nd World War gun emplacements which were now submerged some 250m off-shore. These historic features attested rapid and irrepressible marine advance yet new defences had nevertheless been installed. These had met with rapid threat (Plate 2o). On the east bank of the Gironde (Study Area P16) earthen levees were maintained on a communal basis and were still successful in retaining the course of the river. Here, there was a need to assess the long-term viability of these defences within the scenario of sea-level rise. The drained farmland defended by the levees was very considerable. The reclamation of land from the sea is cited in Schedule 2 (1f) of Directive 85/377 EEC but no instances of this operation were contained within the specific LIFE study areas. Where land had been reclaimed from the sea it was of an historic nature, the oldest probably being that on the east bank of the Gironde.

recent rock armour defences
Plate 2o Local demands for coastal defence can be strongly expressed yet strategically unsound. These recent rock armour defences (A and B) on the Médoc coast of France fulfilled local aspirations to see action, yet seen in the time/depth context of the local coastal change chronology their life-span is limited. The coastal archaeological record and toppled WWII defences (C, D and E) demonstrate that this shoreline is rapidly retreating and poorly suited to hard defences.

Losses from induced biological change

Induced biological changes represent a topic which lies outside the scope of this present LIFE study yet the inter-action between the living environment and the palaeo-environment is an issue which has clearly emerged in this present study. The loss of living habitats has been the major driving force behind European Agenda 21 yet a balanced view of natural depletion versus humanly induced loss can only be achieved when it is measured against the palaeo-environmental record of the long-term trends which are emerging from the past. The needs to conserve the living environment and the sources of palaeo-environment data are equally profound and this LIFE study has found that both of these vulnerable resources are inextricably linked. On the open coast of the Solent, UK, at Quarr loss of in-situ archaeological features is occurring. Historic evidence for shoreline change shows rapid and increased erosion of this coast during the present century. Contemporary with this erosion is the expansion and subsequent diminution of the cord-grass (Spartina) which serves to cover and bond the ancient mud-flats.

The phenomenon of cord-grass change began in the 1860s when the North American species Spartina alterniflora was accidentally introduced into the indigenous communities of Spartina which was then growing in the Solent. The result was a new hybrid male species which was to lead, in turn, to a new and highly active species named Spartina anglica. The new species supplanted the indigenous cord grass and continued to expand during the 1930s and 1940s. Like its indigenous predecessor, it served to promote the settling and the bonding of fine silt particles which are daily transported by the Solent currents. Its coverage of the saltmarsh was, however, far more profuse. Since 1950, the new hybrid cord-grass has died-back at an alarming rate and it has left something of a ecological vacuum. With its decline has come the denudation of the mud-flats and the exposure and erosion of their in-situ archaeological remains. See Plate 2p.

cord grass
cord grass
Plate 2p The diminution of cord grass (a) to tiny residual patches (b) is symptomatic of greater biological change. The exposure of prehistoric archaeological features in this inter tidal mud demonstrates that after some five thousand years of stability,this inter-tidal zone has been changed to a degrading shoreline within the space of a few decades.

Losses induced by outfalls, effluent and pollution

Surprisingly, the discharge of waste water, flood water and effluent into the sea has not been specifically cited in Directive EC 85/337 EEC although Schedule 2 (10d) makes optional provision for a waste water treatment plant. It is interesting to observe that under Schedule 2 (10h), oil or gas pipeline installations are cited but no mention is made of water pipe-lines, the construction of which is equally capable of causing unsustainable damage to the historic and natural environment. In the United Kingdom the utilities agencies which are active in the intertidal and sub-tidal zones for the purpose of managing water and sewage were privatised under the Water Act of 1989. This Act requires these agencies to take regard of the environmental impact of their activities. If necessary their actions can be challenged under judicial review although this would require resources to be committed by the challenger. Control and discharge of flood-water are not responsibilities of these agencies but fall within the remit of the Environment Agency. This Agency is responsible for flood control on rivers but not coastal defence.

Both the construction and operation of pipe outfalls can be detrimental to non-renewable archaeological and palaeo-environmental resources within the coastal zone. Both of these operations pose the threat of unsustainable loss. Some aspects of this threat is recognised in Directive 85/337/EEC where schedule 2 (10c and 11d) identifies flood relief works and waste water plants as operations in which an environmental statement is advised.

In the Wootton-Quarr study area of the Isle of Wight (Study Area P2), spreads of green algal growth have been observed on the mud-flats since the 1960s and these have been attributed to increased discharges of sewage-effluent into the Solent tidestream (Tubbs 1991 and 1999). This algae has also served to alter the bio-diversity of the mud-flat by reducing the oxygen content (Plates 2q(a) and 2q(b)). It is clear in this instance that the loss of important intertidal archaeological remains on this shoreline is associated with the concurrent loss of a formerly stable benthic community. The antiquity and vulnerability of the archaeological remains demonstrates that no significant loss or disturbance of this kind has previously taken place over the past 5,000 years. The loss which is now taking place is concurrent with the effluent discharges and this prompts and justifies research into a solution for this seemingly unsustainable phenomenon.

Damage incurred from activities with 'all-terrain-vehicles'

Opportune sorties at low tide with ATVs can be highly damaging to in-situ archaeological remains and these vehicles seem to be a new type of threat. Their impact has been observed in the intertidal zone at Newtown, Isle of Wight, where enormous tyre tracks have rolled over areas containing the Neolithic wooden structures which have been noted previously in this text. Damage of this kind seems come from three sources. Some damage is the work of `night operations' conducted for military training purposes. Others are more covert activities in which similar vehicles are used to rake up shell fish. A further source of damage, observed elsewhere, is the use of recreational vehicles and 'joy riding'. Given that the most of the national coast offers uncontrolled access, these activities should be considered a direct threat to preservation in-situ.

Losses induced by mechanical beach cleaning

The operation of mechanical beach cleaning machines can be highly destructive to delicate wooden and ceramic archaeological materials which are exposed within the intertidal zone. As recreational demands for clean sandy beaches grow, demand for these machines increases. On the Médoc coast this problem as been identified as a significant one imposing destruction on an array of prehistoric intertidal features which are afforded very little archaeological monitoring or protection. These machines are also popular in the United Kingdom where their impact upon intertidal archaeological remains does not seem to have been documented. The chance recovery of the Soulac boar (Médoc coast, France), a masterpiece of European prehistoric metalwork, is an example of a beach artifact which happens to have escaped the destructive impact of a beach-cleaning machine.

Algal growth
Algal growth
(a) (b) Plate 2q Algal growth (a) induced by effluent discharge and pollution can cause shoreline reversals which are destructive of both habitat (b), intertidal cultural remains and sediment archives. The exposure of archaeological remains in the intertidal zone can be an early warning of this type of unsustainable change.

Gains and losses from the fishing industry

Trawling with the aid of a beam or oyster blade can be damaging to both the natural and archaeological heritage of the seabed. Trawlers can disturb and retrieve large quantities of archaeological and ecological material but an important mitigating aspect is a general willingness of fishermen to report their finds. At high tide some trawlers are prepared to operate over the intertidal zone were the trawl can very effectively behead and destroy in-situ wooden structures (Plate 2r).

Modern GPS systems will permit most trawlers to fix the position of trawled artefacts with acceptable accuracy and many fishermen are prepared to do this provided that reasonable and tactful liaison is maintained by the local SMR archaeologist. Fishermen will, in particular, need to be convinced that they are not losing confidentiality over the fishing areas favoured by their particular crew. This type of liaison and data-gathering is a cumulative process which can only be developed by steadily building upon trust at local level. It also requires a sustained local archaeological presence linked to an on-going database.

A frame trawler
Plate 2r The operation of an A frame trawler easily disturbs and lifts archaeological and palaeo-environmental material from the seabed of the Solent Estuary, UK. Good monitoring and liaison could make this practice a valuable sampling and recording tool mitigating an unsustainable level of loss.

Problems of scoping and specification in the commissioning of environmental assessments and statements

This LIFE study has found that in the United Kingdom significant advances have been made in securing agreement and consistency in the formulation of a national coastal protection strategy. This had been achieved firstly by the formation of 'regional coastal groups' based upon natural coastal cells. The forming of these specialist groups has been followed by the commissioning and adoption of Shoreline Management Plans. Within the Solent study areas three of these plans were studied in detail and their coverage of the historic environment was assessed. Discussion was also held with Dr. Anthony Firth who has been responsible for a national archaeological review of these plans. All of these plans had been prepared by environmental consultants commissioned by individual coastal protection authorities. Coastal protection grant aid for these plans was provided by the Ministry of Agriculture Fisheries and Food (MAFF) which was responsible for the overall national initiative.

Most of the commissioned plans saw archaeological and palaeo-environmental resources as no more than a number of 'sites' which might present a 'conservation problem'. Some consultants had asked for no more than a map or inventory of known sites to demonstrate that the topic had been considered in their plan. Some even restricted this appraisal to nationally important or 'scheduled' ancient monuments. Most of these short-comings could be traced to inadequate briefing at the tendering stage. In most cases the need to appraise the archaeology and palaeo-environmental history of the coastline had been completely overlooked when the individual plans had been commissioned from the environmental consultants. As a consequence virtually no knowledge of long-term environmental trends had influenced the conclusions of the plans. Where the tender of a general environmental consultant had been appointed, their expertise was normally centred upon the natural environment.

In the Solent study area chosen for this LIFE project, differences in coverage and comprehension became clearly apparent when comparison was made between the individual plans prepared for the eastern and western Solent. These inconsistencies largely arose from loose specification and scoping before these plans had been commissioned. Here it seemed that an environmental appraisal had been requested without any distinction being made between the natural and the historic environment. Consequently, environmental consultancies tendering for the contract had prepared the reports according to a contracted price. Where the client had omitted to specify particular appraisal of the archaeological and palaeo-environmental evidence for past and impending coastal behaviour, this omission was generally followed to the letter in order to stay within the contract price.

Promoting sustainability through the curation of historic environmental resources

The LIFE review within the Solent and the conclusions of the national survey conducted by Wessex Archaeology in 1998 showed that much remained to be learnt and improved before sustainable management could be claimed for the historic environment of the coast. Discussion was also held with the Association of Local Government Archaeological Officers which is the peer body responsible for the professional standards of those curatorial archaeologists who are responsible for advising local authorities on the sustainability of archaeological sites which are known or concealed within the local landscape (Plate 2s). These discussions concluded that there remained a need to improve the curation of the coastal archaeological resource and that sustainability could only be achieved when this function was a statutory requirement of the local authorities responsible for the management of the coast.

Elsewhere amongst the LIFE partners, the present prospects of achieving sustainable management of coastal archaeological and palaeo-environmental resources seemed even more remote. In Eire, however, loss seemed almost entirely due to natural processes in the Shannon, Cork and Waterford study areas (Study Areas P19, P20 and P21). Here, the concept of rescue excavation or 'preservation by record' would probably suffice provided that sufficient specialist resources could be assigned to a long-term timetable . In Dublin human-induced loss could be recognised along the developed waterfront and the reclaimed margins of the river but this could not be quantified. The approach of the Irish Government was clearly enlightened, the interventive abilities of its Discovery team being enthusiastically supported by direct on-going funding facilitated by the Irish Lottery. An early assessment of climate change (McWilliams 1991) had also left Government well prepared for changing coastal scenarios. Nevertheless, the intensity of intertidal archaeological and palaeo-environmental sites revealed by the Shannon archaeological survey demonstrated the enormity of the task faced by a relatively small population endowed with a notably rich segment of the European coastal heritage.

In France the Médoc coast showed severe loss of both archaeological and palaeo-environmental resources. These were subject to the powerful aeolian processes which were responsible for the accretion and movement of the dune systems. Stabilisation of the dunes by forestry was recognised to be detrimental to concealed archaeological resources but in view of the necessity to secure the dunes by this method, the only practical solution was perceived to be preservation by record. In the intertidal and beachhead zones of the Médoc coast, wave action was imposing unsustainable damage to an exceptionally rich prehistoric cultural heritage. Here intervention and recovery was recognised to be the only practical solution but there remained a particular problem of under-resourcing for the vital tasks of on-going monitoring and recording. This French study showed a high degree of public and community awareness of archaeological problems. The centralised regulating system controlling for the reporting of portable antiquities and the authorisation of interventive rescue excavation appeared to be less successful being somewhat bureaucratic and discouraging of public participation.

The collective evidence gathered amongst the LIFE partners showed that the rationale of European Agenda 21 had not yet permeated most of the Government agencies who were responsible for management decisions which impacted on the long term sustainability of the historic environment. This was reflected in the resourcing of the curating bodies and a lack of European Agenda 21 guidance on the historic environment. This is particularly regrettable in the context of the European coastline where the combined effects of natural erosion, sea-level rise and human intervention are particularly unforgiving.

Bait digging
Bait digging can impose direct threat to concealed archaeological and palaeo-environmental resources by creating deep craters in the inter tidal zone, Solent Estuary, UK.

2.19
Investigation techniques used in the palaeo-environmental aspects of the life project

The investigation of Holocene sedimentary deposits and the dating of archaeological and palaeo-environmental archives has required a number of different field recovery techniques. These have been followed by laboratory analyses which have sought to establish new data for palaeo-environmental reconstructions and the study of the history of coastal changes.

Sedimentological studies

Coring

The organisations participating in this LIFE project adopted differing techniques for the recovery of samples for their requisite methods of palaeontological analysis. These sampling strategies and techniques were adapted to the methodological difficulties of obtaining continuous sediment profiles. In the United Kingdom these included the 10m high face of a submerged underwater cliff at Bouldnor in the Western Solent (Momber, Study Area P1, Volume 2) On-shore, sediment cores were sought at various depths up to 15 metres in Holocene estuarine deposits at Newtown, Yarmouth and Ranelagh, Isle of Wight, in the Shannon Estuary (Study Area P19) and on the Aquitaine coast (Study Area P17).

The range of equipment used in this diverse range of sediments and habitats has provided a useful insight into the shared problems of recovering field data in intertidal and sub-tidal zones. In some circumstances, heavy commercial civil engineering coring equipment is not viable for reasons of access to difficult intertidal locations. In some instances there was also a limitation imposed by cost. In areas like the Isle of Wight, where the tidal window was tightly restricted, standard hand coring devices such as the Russian/Jowsey corer were found to be satisfactory for obtaining cores for pollen and diatom analysis. Where time was available and firmer substrate present, larger samples for plant macrofossil analysis and radiocarbon dating were obtained using an engine powered (Cobra) corer with 1 metre sleeved tube or Stitz corer.

Along with the underwater sampling strategy and the diving procedures used at Bouldnor by the Hampshire and Wight Trust (Study Area P1) these techniques and equipment provided more than adequate core samples from a very substantial number of sites within the coastal zone. In the environment of the dune coastline of Aquitaine (Study Areas P17 and P18), heavier equipment of the commercial type was found to be more appropriate where there was a need to penetrate the depths of dune sand and sediments within the lagoonal basins. Here standard U100 percussion corers were used to recover 1.0 metre long cores. These provided very satisfactory samples for a range meeting all the requirements for pollen, diatom, foraminifera, ostracod and geochemical analyses as well radiocarbon dating. This equipment is described in more detail in section below.

The sampling and post-excavation storage of the material has been carefully addressed by participating research bodies. Typically, samples have been stored at low temperature (cold store) to prevent biological degradation and drying out of the samples. Wood from submerged archaeological features has been stored in holding tanks or, in the case of some artefacts from Wootton-Quarr, freeze dried for long term preservation and exhibition. Laboratory description of the sequences and extraction techniques for palaeo-environmental and micro-palaeontological material in general follows universally adopted practices. The use of this equipment and the application of these analytical techniques is further described in part 2 of this report.

Non-destructive analysis of core samples by radiography

Radiography allows observation of the internal structure of the cores without disturbing or dismantling them. This technique is based upon differential absorption of the power of X- Rays and on the nature of the sediment crossed. X-Ray radiography reveals low variations in the densities in the sediments which permits detailed observation of the primary (dynamic) and the secondary (biological, diagenetic, etc.) structures. In fact, the technique involves recording the shadow of an object with an X-Ray source on a fluorescent screen or on a film.

The University of Bordeaux uses a radioscopic chain called SCOPIX to study the sedimentary cores. This analysis chain uses a radioscopy / radiography unit. Inside a lead security cabin an X-Ray is emitted from a pipe where it's power and intensity can be regulated. It strikes a sample laid on a motorised trolley, which moves to allow radiography of the total core. The beam is controlled using a pull camera which can amplify brightness. The picture of the sample is revealed on a screen. These pictures are then transferred and recorded in a computer in numerical form using special software This can then be used to draw curves showing the variations of density along the core from the values of density on a known pixel line.

Lithological analysis

Water content

Samples are weighed before and after drying in an oven at a temperature of 60 degrees Celsius. The differences in weight of the samples permits deduction of their water content. This data contributes to the palaeo-environmental studies by revealing the variations in water content (i.e. environmental conditions) along the core.

Organic content

Calcination identifies the concentrations of organic matter in sediments by recording the differences in the weight of samples before and after calcination. The samples are placed in an oven with a temperature of 500 degrees Celsius for 8 hours.

CaCO3 content

The Calcimetry technique determines the amount of carbonate in sediment samples. In Geology, this kind of information is useful in the reconstruction of depositional environments (favourable environments for carbonate precipitation or not, environments which allow living organisms or not, etc.). Dry samples are crushed and weighed. The volume of CO2 coming from the sediment during its attack by chlorhydric acid is measured. This corresponds to the weight of reactive CaCO3. The ratio between the initial weight and the CO2 volume gives the carbonate rate.

Particle-size or grain-size analysis

The technique of granulometry aims to measure the size of the elementary particles of which the deposit is composed (sands, silts, etc.) and to define statistical frequencies of different sizes of grains in the sediment. This defines the conditions, environment and dynamics of past sedimentation or soil formation. The methods of granulometric analysis depend on the nature of the grains (friable, soluble, insoluble, floculated) and on their size. The technique of sieving is frequently used for quartz sands and the technique of laser granulometry for silts and clays.

In the case of sands and gravels sieving is used to divide each sediment sample into a series of dimensional classes using increasingly fine sieves. Sieving is commonly executed by agitating a dry sediment. The column of the sieves is vibrated for a quarter to half an hour, dependant on the fragility of the sample. Each of the resulting categories of sediment is weighed.

In the case of fine particles (silts) the system most frequently used by the University of Bordeaux for this kind of measurement is a "Malvern" laser-granulometer. The particles are firstly diluted in water then put in circulation with a pump system. Particles lighted by a laser deflect the beam. The quantity of deflected light and importantly the deviation angle allow measurement of the size of the particles. Inside the system, the luminous intensities are captured by silicium photodiodes. They are then numerised and analysed. Results are transmitted to a computer and can be presented in different forms (histograms, curves, etc.).

Palaeontological analysis

The identification and analysis of microfossils in Holocene sediments is necessary to reconstruct past environments and to detect changes which have occurred over time. The two principal microfossils examined during this LIFE study are described here.

Ostracods

Ostracoda are a sharply defined order of the Crustacea class, phyllum, Arthropoda. They possess a bivalved carapace of calcareous material which can be readily fossilized. Ostracod carapaces (or valves) are known from strata of upper Cambrian age onwards, and today they still are among the commonest organisms found in practically every aquatic environment. Because of their generally small size (about 0.5 to 1 microns), they may be classed as micro-organisms. The study of fossil ostracods fall within the field of micropalaeontology.

Ostracods occur frequently in many environments and in large quantities in non-marine sediments, where Foraminifera are lacking. Most fossilised Ostracods are benthic and their lives are strictly governed by their aquatic environment. Food, dynamics, biogeochemical conditions such as oxidation, ventilation, salinity, water stratification, phytal communities, chemical equilibria and predation all influence their regeneration and success. The studies of Ostracoda are commonly used to obtain palaeo-environmental information. They are studied at 5 levels : quantity (number of individuals), diversity (number of species), species (fresh, brackish, marine littoral, deep-marine species, etc.), types of populations, intraspecies variations.

The global quantity of the Ostracod fauna indicates the trophic level of the biotope (except in the case of eutrophication, in which the fauna disappears because of the dissoxygenation. The diversity of Ostrocods reflects the chemical stability of the biotope. The maximum number of species live in certain marine and freshwater environments. In brackish or hyperhaline conditions, the number of species decreases to one or two, but a large quantity of individuals can still occur because few species are capable of regulating their cellular permeability.

The type of species and the associations which are present indicate the nature of the environment (fresh, brackish, marine) and the distribution in comparison with depth (infralittoral, bathyal species), substratum (phytal species), life (swimmers, benthics, endobenthics, microfeeders, suspensivores, plant suckers, etc.).

The type or structure of the population (juveniles versus adults or sex ratio) are a good index of dynamic level. For example, a complete population with adults and juveniles probably lived in place with a low energy level.

Some species of Ostracods show a large morphological variability in size and in ornamentation. The mixing of diverse morphologies indicates successive occurrence of differing environmental conditions. Variations in the size of several species reflect changes in quantity of food such as that which can be experienced by spring or autumn generations. Variations in ornamentation also reflect quick changes in the carbonate equilibria at the water-sediment interface in relation with seasonal changes. Occurrence of nodes indicates inputs of organosilicated complexes and oligohaline waters (< 5 0/00). The examination of Ostracod fauna has provided information on the parameters of the biotopes in past marine and non-marine environment of the Holocene. It has also characterised palaeo- environments which can be directly compared with biogeochemical variations at the water- sediment interface. This, consequently, can by related to climatic events in the past.

Foraminifera

These unicellular organisms belong to a branch of the Rhizopods. Their size is generally between 10 mm and 600 microns. They have a calcareous or agglutinated test which can be fossilised. These enrolled tests present various shapes and ornamentations which may change within a species from one environment to the next. For palaeontologists, the characteristics of the test are primary features which can be used to distinguish one species from another and hence these distinctions form interpretations of time or environment. Foraminifera live exclusively in salt water. Although water depth used to be considered the major factor that influenced the distribution of benthic Foraminifera, it is now thought that there is combination of variables including temperature, salinity, hydrostatic pressure, water turbidity and light intensity.

Historically, Foraminifera have been considered predominantly marine organisms, with primitive or aberrant types inhabiting fresh water ecosystems. There is a group of Foraminifera that occurs in brackish environments and this brackish water fauna is geographically very uniform.

In conclusion, Foraminifera are found in marine environments from the deepest tidal channels to shallow ephemeral tide pools in marsh grasses. Benthic Foraminifera are more numerous and diversified than planktonic ones.

Benthic Foraminifera colonize the whole marine environment. The planktonic Foraminifera are more numerous in the open sea and live in water with normal salinities (35 0/00). They have a globular shape, a small size (150-600 microns) and a light test which allows them to float. They are good stratigraphic markers of both recent and ancient environmental conditions because of their rapid evolution and their large geographical distribution in the oceans.

Palynological analysis

Pollen, spores and other botanical microfossils may be found in sediments deposited as long ago as the history of plants, that is, some hundreds of millions of years. The study of pollen and spores in peat, mineral soils and lake sediments has a history going back to the beginning of the last century when pioneer work was carried out in Scandinavia by Lennart Von Post and Gunnar Erdtman.

Before the development of methods of absolute dating pollen analysis was often regarded as a technique for dating sediments as well as reconstructing past environments. This is still true of geological sediments where fossil pollen can provide stratigraphical bio-marker horizons but it is no longer the case for the Holocene period. This dating role of pollen analysis has been largely superceded by radiocarbon measurements from which absolute (calendar) dates can be obtained in conjunction with correlation obtained from dendrochronology.

With the advent of radiocarbon dating, there has developed an increased awareness of the degree of asynchroneity in vegetation and environmental changes. This been particularly apparent in the geographical spread of principal vegetation types since the final close of the last cold stage (glacial) period some 10,000 years ago. In studies of this period pollen analysis has been particularly effective in identifying the plant communities which were growing in the past. The basic premise of these studies is that pollen which becomes incorporated in the peats sediments and soils of past times generally reflects those plants and plant communities which produced the pollen at the time of the sediment deposition. Of course, account has to be taken of the possibility of contamination by reworked older or younger pollen and spores. It is from the identification and analysis of the sub-fossil pollen that basic data and interpretations can be made. These data can then be used as proxy for study of broader environmental aspects such as climatic change, human impact and coastal change.

In order to acquire pollen evidence, sediments are sampled sequentially throughout a stratified pollen profile which will represent a period of accumulation. Each sample at a specific depth should contain pollen which represents a particular age or period. Any changes which have occurred in vegetation and its associated pollen should be reflected in the different types and quantities deposited at each level in the profile. To portray these changes, pollen counts which have statistical validity are made for each level. These counts usually include some hundreds of pollen grains and spores. Relative percentages of the different types identified are calculated for each level. The percentages are then plotted as a histogram to produce a standard pollen diagram. Changes in each plant type through time can be seen through the changing percentages of the 'pollen curve'.

Since the stratigraphy of the sediments is important to the interpretation of the pollen, it is usual to show this on the left side of the pollen diagram. Figure 2.16 is greatly simplified and shows the entire depth of the submerged Middle Holocene land and estuarine sediments at Bouldnor, Isle of Wight (Study Area P1). The notation on the right side shows how the record left by the pollen reveals the changing character of the coastline over a period of some 2000 years. In this instance the recovery of the pollen was restricted to the three peat-bearing horizons within the stratigraphic column. This particular diagram has excluded these intervals to show the overall train of vegetational changes over the entire period of sediment and peat accretion.

Where changes in pollen types occur in phases, it is often desirable to divide the pollen diagram into broad pollen assemblage zones from which environmental interpretations can be made. Such pollen assemblage zones are usually defined without discussion of their cause or effects. This standard approach has been used to describe the pollen sequence of the various polliniferous sediments analysed by the different groups involved with this LIFE project. These techniques have been used with considerable success in studies which have sought to establish the earlier nature of the Holocene coastland in the United Kingdom and France.

Geochemical analysis

The study of metallic atmospheric fallout

Fallout of anthropogenic atmospheric metals (concentrations and isotopic compositions) in peat bogs and in lake sediments can allow the establishment of a geochemical chronologic tool drawing a parallel with theoretical anthropogenic metals emission curves. Where peat deposits have been investigated it has been found that they can proved to be a valuable archive of atmospheric change which can be particularly revealing where human activity has impacted upon the environment during the later Holocene.

In this study metal content and lead isotope ratios have been measured with an Elan 5 000 Perkin-Elmer-Sciex ICP-MS (Inductively Coupled Plasma-Mass Spectrometer). This instrument operates under standard settings in clean room conditions. Lead isotope ratios have been corrected for mass fractionation with the NIST 981 standard (206Pb / 207Pb @ 1.0933). Mass 204 has been counted for 15 times longer than the other masses (206,207 and 208) due to its very low abundance. 204Pb isotope content has been corrected for contribution of 204Hg.

Absolute Dating

Methods of dating past events on an independent or absolute chronological scale have long been sought by earth and environmental scientists and archaeologists. The application of absolute dating methods has been described in section 2.15. The theoretic base of these techniques is also briefly described here.

Dendrochronology

This technique uses the counting, measuring and comparison of the growth rings in trees to reconstruct a continuous and absolute record of annual variations in tree-growth and environment in the past. It is actually the most accurate known method of dating and it has been used effectively in this LIFE study. Evidence for tree ring chronology can be found where depositional conditions have been favourable to the preservation of branches and tree trunks. This has given certain parts of the European coastline as special value as a repository of past environmental data where sustained waterlogging has ensured the survival of this evidence.

The growth rings represent annual events and their thicknesses can give information concerning palaeoclimates. This information is contained in the wood and concerns both abiotic and biotic factors. The abiotic factors refer to constant parameters affecting the tree including the nature of the soil, altitude and exposure as well as variable parameters such as climate , temperatures and precipitation. The biotic factors are the intraspecific and interspecific concurrences, ages, fungal impacts, insects, human interventions etc. While growing for a shared period of time in the same area, trees will acquire a similar series of growth rings. Trees of different but overlapping ages show the same series of growth rings for the time when their lives cross. Starting with a trunk whose age is known, a series of increasingly ancient woods, each of partially overlapping age, can be used to establish a scale of past climatic behaviour.

In the study of the North Médoc coast (Study Area P17), Sanguinet Lake (Study Area P18) and the archaeological site of Lapartens good preservational conditions were found for the recovery of ancient waterlogged wood. Similarly, in the Solent a substantial overlapping series of tree ring samples was recovered from the submerged peats on the intertidal shelf at Wootton-Quarr (Study Area P2). All of these resources presented a wide range of samples which has greatly assisted the task of dating past environmental changes as well as fixing episodes of human activity. In the Solent study area the dendrochronological results extended from the time of contemporary Neolithic activity in the late 3rd millennium BC while in in the French study are the sites extended from the Early Bronze Age to the Roman period. The particular value in these studies was the use of combined methods of dating. Where a tree ring chronology was established the same growth-ring samples could be used for radiocarbon dating and this assisted the calibration of the radiocarbon curve. Moreover, palaeo-environmental studies could be tied to the dating offered by the tree rings and these could also be used to define a reference curve of past climatic variation.

Dating by Radiocarbon (14C)

The accuracy of radioactive measurements permits construction of an "absolute " chronology until the natural radiation levels of the world were altered by nuclear activity at the end of the Second World War. The measurement of the natural radioactivity of elements such as carbon contained in sediments allows absolute dates to be obtained with good accuracy. Sampling must be carried out with particular care because the risk of contamination from later organic material can be quite high although well-disciplined handling procedures can usually exclude this problem. Where waterlogged wood is recovered from intertidal contexts, contamination can arise from the borings of marine organisms and this is a further factor which requires special attention when samples are collected and prepared.

Where radioelements are used for dating they offer a set period over which their residual strength can be measured. In the case of radiocarbon this period has been determined at 5,600 years. This means that is impossible to use this radioelement to measure back more than 40,000 years because at this point the 14C quantity has been reduced beyond the point of detection.

The principle of the radiocarbon dating recognises that all living organism, both animal and vegetable , contain a set amount of 14C. This remains in equilibrium while it is 'exchanged' with the contemporary atmosphere while each organism is alive. The radioactive strength of the absorbed 14C is set by the effects of solar radiation in the upper atmosphere. At death, the exchange between the organism and the atmosphere ceases and after this time the residual content of 14C decays as a known rate. In consequence, the amount of residual 14C within a sample can be measured to determine the length of time which has passed since death.

When this technique was first developed an assumption was made that no past changes had occurred in the level of solar radiation in the upper atmosphere but since that time past fluxes of radiation have been recognised. Use of specific tree rings of known date have helped to identify and track 'wiggles' in the radiocarbon curve. To overcome these uncertainties scientists have emphasised that their measurements mast be recognised as radiocarbon years and not conventional calendrical years. Radiocarbon years before present (BP) are cited in various sections of this report and these refer to retrospect measurements taken from the year 1950. The use of tree-ring chronology has enabled most dates to be converted to calendrical years and these are conventionally expressed as cal. BC or cal. AD. Either measurement will carry a varying margin of error and this is represented by an upper and lower date to which accommodates the perceived error. For dates more than 40,000 years old certain elements of the series of Uranium and Thorium can be used. Unfortunately, these elements are rare and few Holocene sediments contain enough to allow measurement.

Infra-red Stimulated Luminescence

Luminescence technique used by the University of Bordeaux dates the last exposure to light quartz and feldspar grains in a sediment body. Exposure to daylight zeroes the luminescence signal. Once quartz or feldspar grains are buried, the luminescence signal builds up over time due to the radioactive decay of naturally-occurring uranium, thorium and potassium isotopes and the interaction of cosmic rays. The age of the sediment is derived from the natural luminescence accrued in the grain divided by the dose rate from the natural radioactive decay in the surrounding environment. Luminescence techniques have traditionally been applied to aeolian sediments where transport processes have allowed sufficient exposure to light to fully zero the luminescence signal of the wind-blown sediment before deposition and burial. Infra-Red Stimulated Luminescence (ISRL) techniques have been applied to feldspar separated from the aeolian sands showing good comparison with independent age control.

The samples are taken in the field from clean exposures excavated into the dune forms. The freshly exposed face is sampled by hammering lengths of black plastic drainpipe and transferring the contents of the tube directly into opaque black plastic bags. In situ measurements of the gamma dose rate is then undertaken (NE Technology PSR8 ratemeter). All sample preparation and measurements for luminescence dating are undertaken in the laboratory under subdued orange light to prevent bleaching of the natural luminescence signal from the samples. Luminescence dating procedures are subdivided into the determination of the equivalent dose (ED) and the measurement of the dose rate to the sample.

Given the confidence in this technique provided by comparison with previous age control and its applicability to coastal sands, ISRL dating was chosen to date the coastal dune system of the Aquitaine coast, near Bordeaux, France (Study Area P17).

Geographical Information Systems

In the framework of the LIFE project the University of Bordeaux has developed a Geographical Information System (GIS) using Arcview software. This system allows the compilation of data including maps, drawings, commentaries, photos, etc. It is a powerful tool which allows the visualisation and analysis of all types of data from a geographical (distribution) point of view. For example, it allows visualisation of the landscape in a 3D form. This can emphasise the geographical relationship between different morphosedimentological areas. It is also a very accurate in recording the positions and relative positions of cores and samples taken in the study areas. This can be achieved because all the maps are georeferenced in a Lambert II extended form. The system is also capable of calculating surfaces areas, relief and volumes. The University of Bordeaux used scans of each study area with a 25,000 scale to compile the sedimentological and archaeological data. The altimetric data of Sanguinet Lake (Study Area P18) and Oleron Island (Study Area P14) were added to further morphosedimentological studies in dune system near Sanguinet Lake and in the marshes of Oleron Island.

In Eire and the Solent region Arc-Info GIS systems were used by the Discovery Team and the Hampshire and Wight Trust for Maritime Archaeology to assemble their contributions to this LIFE project.

Bibliography

Adams J. van Holk A.L.K. and Maarlevelde Th.J., 1990. Dredgers and archaeology; ship finds from the Slufter. Aalphen van der Rijn.

Akeroyd A.V. 1966. Changes in relative land and sea-level during the post-glacial in southern Britain with particular reference to post-Mesolithic period. Unpub. MA thesis. University of London.

Akeroyd A.V. 1972. 'Archaeological and historical evidence for subsidence in Southern Britain', Phil. Trans. Roy. Soc. London A272, 95-133.

Allen M.J. 1994. The land-use history of the southern English chalklands with evaluation of the Beaker period using environmental data. University of Southampton, Dept of Archaeology. Unpub. PhD. thesis.

Allen M.J. and Gardiner J. (in press). Our changing coast; an interdisciplinary survey of Langstone Harbour, Hampshire. Council for British Archaeology, London..

Anderson F.W. 1933. 'The New Dock excavations, Southampton', Proceeding Hampshire Field Club and Archaeological Society 12, 169-176.

Arnaud J.E.M. 1989. 'The Mesolithic communities of the Sado valley, Portugal in their ecological setting', in C. Bonsall (ed.), The Mesolithic in Europe. John Donald. Edinburgh. 614-631.

Atkinson R.J.C. 1976. 'Lukis, Dryden and the Carnac megaliths' in J.V.S.Megaw (ed.), To illustrate the monuments; essays on archaeology presented to Stuart Piggott. Thames and Hudson. London.112-124.

Bailey G.N. 1977. 'Shell mounds and raised beaches in the Cape York Peninsular', Mankind 11, 132-143.

Bailey G.N. 1978. 'Shell middens as indicators of post-glacial economies; a territorial approach' in P.A.Mellars (ed.), Hunter-gather economy in Late Pleistocene Europe. London. Duckworth.

Bailey G.N. 1983. 'Problems of site formation and the interpretation of spatial and temporal discontinuities in the distribution of coastal middens', in P.M. Masters and N.C.C.Flemming (eds.), Quaternary coastlines and marine archaeology; towards the prehistory of land bridges and continental shelves. Academic Press. London etc.

Balaam N.D. Bell M.G. David A., Girling M., MacPhail R.I., Scaife R.G. and Vaughan D., 1987. 'Prehistoric and Romano-British sites at Westward Ho!, Devon', in V. Straker and N..D. Balaam (eds). Studies in the palaeoecology and environment of South West England. British Archaeological Reports 181 (British series). Oxford. 163-264.

Bellamy A.G. 1995. 'Extension of the British landmass; evidence from shelf sediment bodies in the English Channel', in R Preece (ed)., Island Britain, a Quaternary perspective. Geological Society of London Special pub. no. 96. 47-62.

Biglake E.H. 1974. The exploitation of shellfish by the coastal tribesmen of the Transkei'. Annals of the Cape provincial museums, natural history 9, 159-175.

Blytt A. 1876. Essay on the immigration of Norwegian flora during alternating rainy and dry periods. Cammermeyer, Christiana.

Boon G.C. 1980. 'Caerleon and Gwent levels in early historic times', in F.H.Thompson (ed)., Archaeology and coastal change. Society of Antiquaries of London, occasional paper, new series no 1., 24-36.

Bray M.J., Carter D.J., and Hooke J.M., 1991. Coastal Sediment transport study; report to SCOPAC, 3, the Solent and Isle of Wight. University of Portsmouth.

Daniel G. 1967. The origins and growth of archaeology. Penguin, Harmondsworth.

Camden W. 1586. Britannia. London.

Carew R. 1602. The survey of Cornwall.

Churchill D.M. and Wymer J.J., 1965. 'The kitchen midden site at Westward Ho!, Devon, England: ecology, age and relation to changes in sea-level'. Proc. Prehistoric Society 31, 74-84.

Cipriani L. 1955. 'Excavations in Andamanese kitchen middens' Actes du IV congres international scientifique, anthropologique et ethnologique 2., 250-253.

Crawford O.G.S. 1927. 'Lyonesse'. Antiquity 1, 5-14.

Darby H.C. 1940. The draining of the Fens.

Darwin C. 1859. The origin of species by natural selection or the preservation of favouraed races in the struggle for life. Murray. London.

Darwin C.,1871., The descent of man and selection in relation to sex. Murray. London.

Devoy R.J.N. (1979) 'Flandrian sea-level changes and vegetational history of the Lower Thames estuary'. Phil. Trans. R. Soc. B 285, 355-407.

Devoy R.J.N. (1980) 'Post-glacial environmental change and man in the Thames estuary: a synopsis'. in Thompson, F.H. (ed.) Archaeology and Coastal Change. Society of Antiquaries of London. Occasional paper Soc. Antiquariess. New ser. 1,.134-148

Devoy R.J.N. 1987. 'The estuary of the Western Yar, Isle of Wight: sea-level changes in the Solent region', in K.E.Barber (ed). Wessex and the Isle of Wight. 115-122. Quaternary Research Association Field Guide. Cambridge.

Dimbleby G.W. 1962. 'Soil pollen analysis', Journal of Archaeological Science 12, 1-11.

EH 1996. Protecting our Heritage. A consultation document

EH and RCHME 1996. 'England's coastal heritage; a statement on the management of coastal archaeology'. Leaflet by English Heritage and the Royal Commission on Historical Monuments of England.

Erdtman G.1928. 'Studies in the Postarctic history of the forest of northwestern Europe. 1 Investigations in the British Isles'. Geol Foren Stockholm Forh. 50. 123-195.

Evans J.G. 1971. 'Notes on the environment of early farming communities in Britain', in D.D.A.Simpson (ed)., Economy and settlement in Neolithic and Early Bronze Britain and Europe. Leicester Univ. Press.

Evans J.G. 1972. Land snails in archaeology. Seminar Press. London.

Fairbridge R.W. 1976. 'Shellfish-eating pre-ceramic Indians in coastal Brazil'. Science 191, 1131-1144.

Firth A. 1999. Shoreline Management Plans and the historic environment; final report for English Heritage.

Wessex Archaeology. Client report prepared for English Heritage. London.

Frere J. 1800. 'Account of flint weapons discovered at Hoxne in Suffolk', Archaeologia 13, 100-102.

Fulford, M.G. Champion T.C. and Long A., 1997. 'England's coastal heritage; a survey for English Heritage and the RCHME'. English Heritage. London..

Furlong N. 1996. The history of land reclamation in Wexford Harbour. In D. Rowe and C.J. Wilson (eds.), An anthology of the Wexford Stobs and Harbour, 83-90. Enniscorthy.

Geer E.H. de. 1912. 'A geochronology of the last 12,000 years' C.R XI. Congr. geol. int. Stockholm. 241-253.

Geike J. 1881. Prehistoric Europe; a geological sketch. London.

Godwin H. 1934. 'Pollen analysis; an outline of the problems and potentialities of the method', New Phytologist 33, 278-305.

Godwin H. 1940. 'Pollen analysis and the forest history of England and Wales', New Phytologist 39, 370-400.

Godwin H. 1941. 'Pollen analysis and Quaternary geology', Proc Geol. Assoc., London. 52, 328-361.

Godwin H. 1943. 'Coastal peat beds of the British Isles and North Sea'. Journal of Ecology 31, 199-247Godwin, H. 1945a 'Coastal peat-beds of the North Sea region,as indices of land and sea-level changes'. New Phytologist 44 (1) 29-69.

Godwin H. 1945b. 'A submerged peat in Portsmouth Harbour'. New phytologist 44, 152- 155.

Godwin H. 1978. Fenland; its ancient past and its uncertain future. Cambridge.

Gough W. 1789. A revised edition of William Camden's Britannia. London.

Haskins L.E. 1978. The vegetational history of south-east Dorset. University of Southampton. Unpub. PhD. thesis.

HCC/WA. 1993. Langstone Harbour archaeological survey project; the assessment. Hampshire County Council and Wessex Archaeology.

HCC/WA. 1994. Langstone Harbour archaeological survey project; second interim report. Hampshire county Council and Wessex Archaeology.

Healey M.G. 1995. 'The lithostratigraphy and biostratigraphy of a Holocene coastal sediment in Marazion Marsh, West Cornwall, U.K. with reference to relative sea-level movements', Marine Geology 24, 237-252.

Heyworth A. and Kidson C. 1987. Sea-level changes in southwest England and Wales', Proceedings of the Geologists Association 93 (1) 91-113.

Hodson F. and West I.M. 1972. 'Holocene deposits at Fawley Hampshire and the development of Southampton Water'. Proceeding of the Geological Association 83, 421-444.

Hutton H.W. 1970. 'Arthur Young, a tour in Ireland; with present observations on the state of that kingdom; made in the years 1776, 1777 and 1778'. (4th ed.) (2 vols). Shannon.

Inchiquin Ms. Ainsworth J. (ed). The Inchiquin manuscripts. Dublin 1961.

IPCC. 1995. On the prediction of sea-level rise.

Iversen J. 1940. `The influence of prehistoric man on vegetation', Danm geol Unders Ser 1V 3 (6), 1-25.

Izarra. 1993. 'Lom et Fleuves en Gaule Romaire.'- Editions Errance. Paris.

James H. 1847. 'On a section exposed by the excavation of the new steam basin in Portsmouth Dockyard'. Quarterly Journal Geological Society of London 3, 249-251

Jennings S. and Smyth C. 1990. ' Holocene evolution of the gravel coastline of East Sussex', Proceedings of the Geologists Association 101, 213-224

Kraft J.C. Belknap D.F and Kayan I. 1983. 'Potential of discovery of human occupation sites on the continental shelves and near-shore coastal zone; a review of case histories', in P.M. Masters and N.C.C. Flemming (eds.), Quaternary coastlines and marine archaeology; towards the prehistory of land bridges and contiental shelves. Academic Press. London etc. 87-120.

Larsson L. 1983. 'Mesolithic settlement on the sea floor in the Strait of Oresund'. In P.M. Masters and N.C.C. Flemming (eds) 283-301.

Lartet E. 1860. 'On the co-existence of Man with certain extinct quadrupeds proved by fossil bones from various Pleistocene deposits bearing incisions made by sharp instruments', Quarterly Journal of the Geological Society 16, 471.

Lewis F.J. 1905-1911. 'The plant remains in the Scottish peat mosses. Part 1. The Scottish Southern Uplands', Transactions Royal Society, Edinburgh 31, 699-723; Part 2 (1906) ibid. 45, 335-60; Part 3 (1907) ibid. 46, 33-70; Part 4 (1911) ibid. 47, 793-833.

LLibby W.F.1946. 'Atmospheric helium three and radiocarbon from cosmic radiation' Phys. Rev. 69, 671-672.

Loader R., Westmore I. and Tomalin D.J. 1997. Time and tide; a archaeological survey of the Wootton-Quarr coast. Isle of Wight Council.

Long, A.J. 1992 'Coastal responses to changes in the East Kent Fens and southeast England, UK over the last 7500 years'. Proceedings of the Geologists Association 103,187-199.

Long A.J. and Innes J.B. 1993. 'Holocene sea-level changes and coastal sedimentation in Romney marsh, southeast England, UK. Proceedings of the Geologists Association 104, 223-237.

Long A.J and Tooley M.J. 1995 'Holocene sea-level and crustal movements in Hampshire and Southeast England, United Kingdom'. Journal of Coastal Research Special Issue No.17,299-310.

Long R.G. and Scaife R.G. 1996. 'Pleistocene and Holocene Evolution of Southampton Water and its Tributaries'. Environmental Research Centre, Department of Geography, University of Durham.

Long A.J., Scaife R.G and Edwards R.J. 1999. 'Pine pollen in intertidal sediments from Poole Harbour, UK; implications for late-Holocene sediment accretion rates and sea-level rise'. Quaternary International 55,3-16.

Long and Scaife forthcoming. On sea-level change in the Solent.

Lowie R.H. 1937. The history of ethnological theory. London

Lubbock J. (Rt. Hon. Lord Avebury). 1865. Prehistoric times as illustrated by ancient remains and the manners and customs of modern savages. Williams and Norgate.

Lubbock J. (Rt. Hon. Lord Avebury). 1896. The scenery of Switzerland and the causes to which it is due. Macmillan. London

Lubbock J. (Rt. Hon. Lord Avebury) 1902. The scenery of England and the causes to which it is due. Macmillan. London.

Lukis W.C. 1868. On the various forms of monuments commonly called dolmens in Brittany. Trans Internat. Congr. Prehist. Arch. third session. Norwich. 218-223.

Lyell C.P. 1830 and 1840. Principles of geology or the modern changes to the earth and its inhabitants considered as illustrative of geology. 1st ed. Murray, London and 6th ed.

Lyell C. 1863. The geological evidences for the antiquity of man. London.

MAFF.1995. Shoreline Management Plans; a guide for operating authorities. MAFF Publications PB 1471.

Marsden P. 1974 and 1985. The wreck of the Amsterdam. Hutchinson. London. 1st and 2nd ed.Masters, P.M and Flemming N.C., 1983. Quaternary coastlines and marine archaeology; towards the prehistory of land bridges and continental shelves. Academic Press. London, Toronto etc.

McWilliams B.E. (ed). 1991. Climate change; studies on the implications for Ireland. Environment Action programme. Government of Ireland, Department of the Environment. Dublin.

Momber G. 2000. 'Drowned and deserted: a submerged prehistoric landscape in the Solent, England. International Journal of Nautical Archaeology 29, (1), 86-99.

Murphy P. (ed). Annual reports on the Hullbridge Project. 1982-1987. Essex County Council.

Niccolini A. et al. 1854-1896. La case ed i monumenti di Pompeii disegnati e descritti. 4 vols. Napoli.

Nicholls R.J. and Clarke M.J. 1986. 'Flandrian peat deposits at Hurst Castle Spit' , Proceedings Hampshire Field Club and Archaeological Society 42, 15-21.

Noort R. van de. and Ellis S. 1997. Wetland heritage of the Humberside Levels; an archaeological survey. University of Hull.

Osborn A..J. 1977. 'Strandloopers, mermaids and other fairy tales: ecological determinants of marine resource utilisation - the Peruvian case' in L.R. Binford (ed.), For theory building in archaeology. Academic Press,. London etc.

O'Dalaigh B. (ed). 1998. The strangers gaze; travels in County Clare 1534-1950. Ennis.

Pirrazoli P.A. 1996. Sea-level changes; the last 20,000 years. John Wiley. Chichester and New York etc.

Preece R.C. Scourse J.D., Houghton, S.C., Knudsen, K.L., and Penney, D.N., 1990: `The Pleistocene sea-level and neo-tectonic history of the Eastern Solent, Southern England', Philosophical Transactions of the Royal Society of London, B, Biological Sciences, 328, (no 1249) 425-477.

Preece R.C and Scourse J.D. 1987. 'Pleistocene sea-level history of the Bembridge area of the Isle of Wight', in J.E Barber (ed.), Wessex and the Isle of Wight: field guide. Quaternary Research Association. Cambridge.

Prestwich J. 1860. 'On the occurrence of flint implements associated with the remains of animals of extinct species in beds of a late geological period at Amiens and Abbeville and in England at Hoxne', Proceedings of the Royal Society of London 10, 50.

PPG16. Planning policy guidance 16; Archaeology and planning. DOE 1990.

Purnell R.G. 1995. `Shoreline Management Plans - national objectives and implementation', in ICE 1995.

RCHME. 1992. Recording England's past; a review of national and local sites and monuments records, Royal Commission on historical Monuments of England, London.

Reid C. 1892 'Pleistocene deposits of the Sussex coastand their equivalents in other districts' Quarterly Journal Geological Society 48, 344-361.

Reid C. 1893. 'A fossiliferous Pleistocene deposit at Stone, on the Hampshire coast'. Quarterly Journal Geological Society 49, 325-9.

Reid C. 1896. 'An early Neolithic kitchen-midden andtufaceous deposit at Blashenwell near Corfe Castle'. Proceedings Dorset Natural History and Archaeological Society 17, 67-75.

Reid C. 1899. Origin of the British Flora London, Dulace and Co.

Reid C. 1905. 'The island of Ictis' Archaeologia 59, 218-288.

Reid C. 1913. Submerged forests; Cambridge manuals of science and literature.

Ritchie A. 1990. 'The first settlers' in C. Renfrew (ed.), The prehistory of Orkney; BC 4000- 1000AD. Edinburgh University Press. 36-53.

Rowe D. and Wilson C.J. (eds.). 1996. An anthology of the Wexford Stobs and Harbour. Enniscorthy.

Scaife R.G. 1980. Palaeo-ecological studies in the Isle of Wight. University of London unpub. PhD thesis.

Sernander R. 1908. 'On the evidence of post-glacial changes of climate furnished by the peat mosses of Northern Europe, Geol. Foren. Forh. 30, 465-478.

Shackleton N.J. and van Anden T.H., 1980. 'Prehistoric shell assemblages in the Francthi Cave and evolution of the adjacent coastal zone'. Nature 288, 357-359.

Shennan I. 1989. 'Holocene crustal movements and sea-level changes in Great Britain', Journal of Quaternary Science 4, 77-89.

Shore T.W. and Elwes J.W. 1889. 'The New Dock excavations at Southampton', Proceedings Hampshire Field Club and Archaeological Society 1, 43-56.

Simmonds B. 1980. 'Iron Age and Roman coasts around the Wash' in F.H.Thompson (ed)., Archaeology and coastal change, Society of Antiquaries of London. Occasional paper, new series no 1. 56-73.

Sordinas A. 1983. 'Quaternary shorelines in the region of Corfu and adjacent islets, Western Greece', in P.M. Masters and N.C.C. Flemming (eds.), Quaternary coastlines and marine Archaeology; towards the prehistory of land bridges and continental shelves. Academic press. London etc. 335-344.

Steers J.A. 1946 and 1964. The coastline of England and Wales. Cambridge. (2nd edition 1964).

Suhm P.F. 1776. History of Denmark, Norway and Holstein.

Swadling P. 1976. 'Changes induced by human exploitation of prehistoric shellfish populations. Mankind 11, 11-18.

Thomas C. 1985. Exploration of a drowned landscape; archaeology and history of the Isle of Scilly. London.

Thompson F.H. (ed.) 1980. Archaeology and coastal change. Society of Antiquaries of London. occasional paper, new series 1.

Tomalin D.J. 1990. 'An early Roman cliff-top salt-working site at Redcliff Battery, Sandown, Isle of Wight'. Proceedings Isle of Wight Natural History and Archaeological Society 9, 91-120.

Tomalin D.J. 1992. `Maritime archaeology as a coastal management issue; a Solent case study from the SCOPAC coast'. Proceedings of the 1992 SCOPAC Seminar on the role of regional coastal groups. Isle of Wight County Council.

Tomalin D.J. 1995. `Wootton Haven; An assessment and updated project design for the Wootton-Quarr intertidal survey'. Unpub. report to English Heritage.

Tomalin 1997. 'Bargaining with nature; considering the sustainability of archaeological sites in the dynamic environment of the intertidal zone'. Preserving archaeological remains in-situ; proceedings of the conference of 1st-3rd April 1996 at the Museum of London. Museum of London/University of Bradford..

Tomalin D.J. 2000a. 'The Geomorphological evolution of the Solent seaway and the severance of Wight; a review'. Proceedings of the Solent Science Conference at the Southampton School of Ocean Science. Elsevier

Tomalin D.J. 2000b. 'Wisdom of hindsight; palaeo-environmental and archaeological evidence of long term processual changes and the coastline sustainability'. Proceedings of the Solent Science Conference at the Southampton School of Ocean Science. Elsevier.

Tomalin D.J., Cross J. and Motkin D.L. 1988. 'An Alberghetti bronze minion and carriage from Yarmouth Roads, Isle of Wight, England. International Journal of Nautical Archaeology 17,75-86.

Tomalin D.J., Loader R.D. and Scaife R.G., forthcoming. Coastal archaeology in a dynamic environment, a Solent case study. English Heritage monograph series. London

Thomsen C.J. Ledetraal til Nordisk Oldkyndighed. Copenhagen 1836. (See also English edition entitled Guide to Northern Antiquities 1848).

Tubbs C.R.1981. The Solent, a changing wild-life habitat. Hampshire and Isle of Wight Wild- life Trust. Romsey.

Tubbs C.R. 1999. The ecology, conservation and history of the Solent. Packard. Chichester.

Van der Molen J. 1997. 'Tides in as saltmarsh (Great Marshes, Barnstable , Cape Cod, USA). Academisch Proefschrift. Febodruk BV, Enschede, Nederlands. 104.

Verstegen R. (pseud) Rowlands R. 1605. A restitution of decayed intelligence. Antiquities concerning the English nation. R. Brusney. Antwerp.

Von Post L. 1916. 'On skog stradstollen i sydsvenska torfmosselager foljder (foredragersreferat)'. Geol. for. Stock. Forh. 38, 384-394.

Warren, S. H, Piggott S, Clarke J.G.D, Burkitt M.C, Godwin H, and Godwin M.E., 1936. 'Archaeology of the submerged land surface of the Essex coast', Proc Prehistoric Society 2, 178-210.

Wilkinson T.J. and Murphy P.L. 1995. The archaeology of the Essex coast, vol 1: The Hullbridge Survey. East Anglian Archaeology report no 71.

Wilson J.G., Brennan M and Brennan B. 1993. 'Horizontal and vertical gradients in sediment nutrients in the mud-flats of the Shannon Estuary , Irelend', Netherlands Journal of Aquatic Ecology, 27, 173-180.

Wijngaarden-Bakker L.H. van 1989. 'Faunal remains in the Irish Mesolithic' in C. Bonsall (ed.), The Mesolithic in Europe. John Donald. Edinburgh. 125-133.