Search

We use a sedimentological approach to examine the formation and deformation processes associated with the accumulation of shell deposits in two major clusters of shell mounds, the Weipa group in the monsoonal environment of Albatross Bay in the Cape York Peninsula of northern Queensland, Australia, and the Farasan Islands group in the semi-arid environment of the southern Red Sea sector of Saudi Arabia. The comparison of such disparate case studies is deliberate, intended to highlight generic issues of shell accumulation and degradation irrespective of the taxonomic composition of the shells or cultural and environmental histories. It also reflects recent fieldwork in both regions conducted in parallel with collaborative arrangements for sharing of ideas and approaches and exchange of personnel in order to establish a common baseline for comparison. Comparative analysis of shell composition, fragmentation, and accumulation highlights similarities despite the different cultural and environmental contexts of the two case studies. These similarities suggest that the size and form of shell deposits are altered by a combination of processes reflecting ongoing changes in deposit composition unrelated to human actions of shell discard. Even where large shell deposits are visible and available for sampling, what is preserved is neither a static reflection of initial deposition nor of undisturbed or “completed” form. We consider the influence of such processes on assessments of rates of deposition and the interpretation of variations in the shape and size of shell deposits.

This Special Issue brings together fourteen articles that present new methods, ideas, and approaches in the study of coastal prehistory with examples drawn from the Americas, Australia, Europe, Saudi Arabia, and South Africa. In this introductory overview, we set out the rationale for combining articles on shell middens and submerged landscapes and the underlying logic of the order in which we have chosen to present the articles. The sequence begins with studies of marine molluscs and moves progressively outwards from small-scale studies of midden composition to large-scale studies of submerged landscapes, and from land to sea. We summarise the contents of each article and highlight connections between them and similarities and contrasts. We conclude with some final comments about the relationship between on-land and underwater investigations and identify the taphonomic, formational and deformational histories of archaeological deposits, materials and landscapes, and the variable impact of sea-level change as unifying themes.

© Springer International Publishing AG 2017. We report on a preliminary exploration of the submerged landscapes in the Saudi Arabian sector of the southern Red Sea aboard the Hellenic Centre for Marine Research (HCMR) Research Vessel, AEGAEO, in May–June 2013. The survey sampled areas of the continental shelf down to the shelf margin at ~130 m depth in the vicinity of the Farasan Islands and combined high resolution acoustic techniques with sediment coring to reconstruct features of the now-submerged landscape of potential archaeological significance, including geological structure, topography, palaeoenvironment, and sea-level change. The region is currently of wide interest and significance: to archaeologists because it is currently regarded as one of the primary pathways of dispersal for early human populations expanding out of Africa during the Pleistocene, in which the extensive but now-submerged shelf region may have played a key role; and to marine geoscientists because the Red Sea offers unusual opportunities as a ‘laboratory’ for investigating Pleistocene sea-level change. Preliminary results indicate that the submerged landscape was characterised by a complex topography with fault-bounded valleys and deep basins, some of which may have hosted, at least intermittently, fresh water during periods of lowered sea level.

A large number of sites dated to the Late Middle Paleolithic and the Early Upper Paleolithic have been recorded in the Cantabrian region (northern Iberia), making this area a key location to investigate the lifeways of the last Neanderthals and the first anatomically modern humans. The stratigraphic sequence from El Cuco rock-shelter was originally attributed to the Early Upper Paleolithic based on radiocarbon dates measured on bone apatite. However, new radiocarbon dates on shell carbonates from the lower levels produced inconsistent dates with those previously published. In order to clarify this anomaly, a reassessment of the chronology of levels VI to XIV was undertaken. The review was based on new radiocarbon dates performed on bones and shells, and a re-evaluation of the lithic assemblages. Bone samples did not produce radiocarbon dates due to a lack of collagen preservation but radiocarbon dating of shell carbonates provided dates ranging from 42.3 to 46.4 ka BP. These dates are significantly older than that previously obtained for level XIII using biogenic apatite from bones (?30 ka uncal BP), suggesting that the bone apatite used for radiocarbon dating was rejuvenated due to contamination with secondary carbonate. Lithic assemblages, defined in the first place as Evolved Aurignacian, have now been confidently attributed to the Mousterian techno-complex. These results suggest a Middle Paleolithic chronology for this part of the sequence. The new chronology proposed for El Cuco rock-shelter has significant implications for the interpretation of Neanderthal subsistence strategies and settlement patterns, especially for coastal settlement and use of marine resources, not only in northern Iberia, but also in Atlantic Europe. Radiocarbon dates OxA-27196 and OxA-27730 were funded by the NERC Radiocarbon Facility (NF/2012/1/18) and they were part of the project NF100413 (Newton International Fellowship to IGZ). Radiocarbon dates OxA-30851, P-32155 (ORAU) and UGAMS-9076 were part of the project HAR2013-46802-P funded by the Spanish Ministry of Economy and Competitiveness. Two other samples that failed to be dated at the ORAU (P-35524 and P-35523) as well as the date Beta – 382682 and elemental analysis were obtained in the project HAR2010-22013, funded by the Spanish Ministry of Science and Innovation. Isotopic analysis was performed as part of the projects HAR2012-33956 funded by the Spanish Ministry of Economy and Competitiveness and EUROREFUGIA (Ref. 322112) funded by the European Union through FP7-PEOPLE-2012-CIG (https://eurorefugia.wordpress.com/), with additional support from the Max Planck Society. IGZ and ABMA were funded by the Juan de la Cierva (JCI-2012-12094) and the Ramón y Cajal (RYC-2011-00695) Research Programmes of the Spanish Ministry of Economy and Competitiveness, respectively

Understanding environmental conditions faced by hunter-fisher-gatherers during the Pleistocene and Holocene, and interpretation of subsistence strategies, social organisation and settlement patterns, are key topics for the study of past human societies. In this respect, oxygen isotope values (?18O) of mollusc shell calcium carbonate can provide important information on palaeoclimate and the seasonality of shell collection at archaeological sites. In this paper, we tested P. vulgata shells from northern Iberia as a paleoclimate archive through the study of shell oxygen isotope values and sclerochronology of modern samples. Results showed that limpets formed their shells close to isotopic equilibrium, with an average offset between measured and predicted values of 0.36?. This offset is significantly reduced with respect to those reported in previous studies, probably due to the use of highly resolved data on the isotopic composition of the water when calculating predicted values. Despite large intra-specific variability, shell growth patterns of P. vulgata revealed a common pattern of higher growth in spring and a growth cessation/slowdown in summer and winter. The seasonal growth cessation/slowdown did not exceed three months. Therefore, a correct interpretation of the season of shell collection is still possible. Reconstructed seawater temperature exhibited a high correlation with instrumental temperature (R2 =0.68 to 0.93; p b 0.0001). Despite periods of growth cessation/slowdown, mean seawater temperatures and annual ranges were reconstructed accurately. As demonstrated here, seawater temperature can be reconstructed with a maximum uncertainty of ±2.7 °C. Therefore, our study shows that oxygen isotope values from P. vulgate can be used for the reconstruction of paleoclimate and the season of shell collection This research was part of the projects NF100413 (Newton International Fellowship granted to IGZ) and HAR2013-46802-P (funded bythe Spanish Ministry of Economy and Competitiveness, MINECO).

Shell middens, sometimes in the form of mounds of great size, are a ubiquitous indicator of coastal settlement and exploitation of marine resources across the world. However, shell middens are relatively rare before the mid-Holocene because most palaeoshorelines before that time are now submerged by sea-level rise since the Last Glacial Maximum (LGM). Previously reported examples of underwater shell middens are almost unknown and of uncertain status, and it has generally been assumed that such deposits would not survive the destructive impact of sea-level rise or would be indistinguishable from natural shell deposits. Recently, two examples of underwater shell deposits have been independently discovered and verified as anthropogenic midden deposits – a Mesolithic shell midden on the island of Hjarno in the Straits of Denmark, and a Middle to Late Archaic shell midden in the Econfina Channel of the Gulf of Mexico, Florida, USA. We report the comparative geoarchaeological analysis of these deposits, using a sedimentological approach to unravel their formation history and post-depositional transformation. Despite the differences in coastal geomorphology and geology, cultural context, molluscan taxonomy and preservation conditions between these sites, the results demonstrate similar sedimentological profiles that are distinctive of anthropogenic deposits, demonstrate their origin as subaerial deposits at the shore edge before inundation by sea-level rise, and show that these properties can be identified in sediment samples recovered from coring. These findings support arguments that such sites likely exist in greater numbers than previously assumed, that they can be identified from minimally invasive techniques without the need for extensive underwater excavation, and that they should be sought to fill critical gaps in the temporal and geographical record concerning Late Quaternary human use of coastal zones and marine resources.

Anthropogenic shell accumulations (shell middens), often of great size, occur in their tens of thousands around the world’s coastlines. They mostly date from the Mid-Holocene onwards and are frequently taken as symptomatic of a Postglacial ‘revolution’ involving world-wide population growth and intensification in exploitation of marine resources. However, the comparative rarity of earlier deposits may have as much to do with Postglacial sea-level rise and the loss of evidence from earlier palaeoshorelines as with genuine socio-economic trends. Here we investigate the underwater Mesolithic (Ertebolle) shell midden of Hjarno Vesterhoved in Denmark, one of the first underwater shell middens to be systematically verified as an anthropogenic shell deposit in a region world-famous for its many hundreds of Ertebolle shell mounds on the present shoreline. We show how a combination of geophysical survey, coring, excavation, stratigraphic interpretation and macroscopic analysis of midden contents can be used to identify underwater deposits, to unravel their taphonomic and post-depositional history in relation to surrounding sediments, and to distinguish between cultural and natural agencies of shell accumulation and deformation. We demonstrate the presence of an intact underwater shell-midden deposit dated at 5400–5100 cal BC, one of the earliest in Denmark. We demonstrate the usefulness of such material in giving new information about early coastal subsistence economies and greater precision to the measurement of palaeo-sea levels. We discuss the implications of our results for an improved understanding of the Mesolithic record in Denmark and of biases in the archaeological record of Late Pleistocene and Early-to-Mid Holocene coastal contexts. We emphasise the importance of researching more fully the geomorphological and taphonomic processes that affect the accumulation, destruction, burial, preservation and visibility of underwater archaeological deposits, the need to extend underwater investigations more widely and to more deeply submerged palaeoshorelines, and the combination of methods required to advance such investigations.

Radiocarbon dating of closely associated marine mollusk shells and terrestrial material (mammal bones or charred wood) collected from archaeological contexts in northern Atlantic Iberian coastal areas is used to quantify the marine 14C reservoir effect (ΔR) for the coastal waters off the Cantabrian coast of northern Iberia. For the first time, ΔR values were reliably determined for these coastal waters and, also for the first time, a ΔR was calculated for the Late Pleistocene in Atlantic Iberia. Pairs of coeval samples of different carbon reservoirs selected from Upper Paleolithic (Late Pleistocene) and Mesolithic (Early Holocene) contexts yielded ΔR weighted mean values of –117±70 14C yr and –105±21 14C yr, respectively. These values show oceanographic conditions characterized by a reduced offset between atmospheric and surface water 14C contents, suggesting a nonexistent or very weak upwelling and some stratification of the water column. Similar oceanographic conditions have been recorded in other areas of Atlantic Iberia during the Holocene, such as off Andalusian and northwestern Galician coasts. Results not only provide useful information on environmental conditions but also a framework to obtain more precise and reliable absolute chronologies for the Late Pleistocene and Early Holocene in northern Iberia.

Recent studies in Atlantic Europe on crab remains, goose barnacles and sea urchins have revealed that these species can be of great help in determining patterns of shellfish collection and providing new information on subsistence strategies of hunter-fisher-gatherers. Current excavations at the Mesolithic shell midden site of El Mazo cave (Asturias, northern Iberia) have produced a sizeable amount of crustacean and echinoderm remains from a long stratigraphic sequence that covers an important part of the Mesolithic chronological range, providing the opportunity to investigate long-term exploitation patterns. Results show that echinoderms (sea urchins) and crustaceans (goose barnacles and crabs) were present throughout all of the stratigraphic units (from 8.9 to 7.6 cal ka), suggesting that they were a persistently exploited food source. However, these resources were not intensively exploited, save perhaps sea urchins at the base of the sequence. From a quantitative perspective, these resources have been traditionally interpreted as minor resources exploited opportunistically to help group survival. However, given the pattern of continuous exploitation exhibited by these resources in northern Iberia and other areas of Atlantic Europe, we suggest that they can be interpreted from a qualitative perspective as stable resources with a significant social value.

Sea-level change has influenced human populations globally since prehistoric times. Even in early phases of cultural development, human populations were faced with marine regression and transgression associated with the glacial–interglacial climatic cycle, amplified by glacio-isostatic adjustments in some regions. Global marine regression during the last glaciation changed the palaeogeography of the continental shelf, converting former marine environments to attractive terrestrial habitats for prehistoric human occupation, and adding an extensive new increment of land, in the case of Europe amounting to an additional 40% of the existing land area, and on a global scale to some 20 million km of additional territory that is now submerged. These areas of the shelf were used as hunting and gathering areas, and as pathways of dispersal between regions and between continents, until they were resubmerged by the post-glacial marine transgression. They also most probably witnessed the earliest developments in seafaring, marine exploitation and permanent settlements. Based on modern marine research technologies and the integration of large databases, proxy data are now becoming available for the reconstruction of these submerged Quaternary landscapes. Concerted efforts are also now being devoted to the search for prehistoric archaeological sites and artefacts on the seabed, often in collaboration with marine scientists. This search has been stimulated by the increasing amount of material that has demonstrably survived inundation, often with excellent preservation of organic remains, by closer collaboration with offshore industry, and by the growing realization of the importance of these submerged data for understanding human prehistoric developments during periods of rapidly changing climate and environment. Moreover, these new research trends are not simply being driven by an archaeological need for scientific and technological input from other disciplines, but by collaborations involving genuine mutual benefit, in which all partners have something to gain. Archaeological problems often pose new questions about geological change, stimulating new techniques of observation, new technologies and new investigations, which in their turn can offer new data, often at higher resolution and with better dates, in relation to geological and environmental issues such as sea-level change and palaeoclimatic variability on the continental shelf. The expansion of early human populations to occupy new territory and new continents is one of the great narratives of human evolution. It is currently a theme of wide interest and topicality, and has received significant impetus and publicity from the new science of palaeogenetics. Together with ongoing discoveries of new fossil and archaeological material, and new studies of palaeoclimate and palaeoenvironment, the field itself is rapidly expanding, leading to new discoveries and new controversies (Grine et al. 2009; Gamble 2013; Smith & Ahern 2013). All the current indications are that the human species originated in Africa, with at least two major episodes of dispersal. The earliest, some time after about 2 myr ago, led to the expansion of Homo ergaster/erectus populations across southern Europe and Asia, extending from Britain in the west to China and Indonesia in the east and south. The second involved our own species, Homo sapiens (Anatomically Modern Humans or AMH), and took place some time after about 200 ka. This resulted in the replacement of earlier hominin populations in Europe and Asia, and expansion further afield: into New Guinea and Australia, certainly involving sea crossings over distances of at least 50 km, by about 50 ka; to the higher latitudes of northern Eurasia, with entry into the Americas by at least 15 ka; and into new territory exposed by the melting of the northern hemisphere ice sheets less than 10 kyr ago. Anthropological and archaeological investigations of these processes are directly connected

We report the results of 212 radiocarbon determinations from the archaeological excavation of 70 shell mound deposits in the Wathayn region of Albatross Bay, Australia. This is an intensive study of a closely co-located group of mounds within a geographically restricted area in a wider region where many more shell mounds have been reported. Valves from the bivalve Tegillarca granosa (Linnaeus, 1758) were dated. The dates obtained are used to calculate rates of accumulation for the shell mound deposits. These demonstrate highly variable rates of accumulation both within and between mounds. We assess these results in relation to likely mechanisms of shell deposition and show that rates of deposition are affected by time-dependent processes both during the accumulation of shell deposits and during their subsequent deformation. This complicates the interpretation of the rates at which shell mound deposits appear to have accumulated. At Wathayn, there is little temporal or spatial consistency in the rates at which mounds accumulated. Comparisons between the Wathayn results and those obtained from shell deposits elsewhere, both in the wider Albatross Bay region and worldwide, suggest the need for caution when deriving behavioural inferences from shell mound deposition rates, and the need for more comprehensive sampling of individual mounds and groups of mounds.

Over 3,000 shell-midden sites have been located in the southern Red Sea using digital-imaging techniques, in a combination of palaeo-landscape reconstruction and remote survey. The primary methods include digital-imaging techniques – high-resolution satellite images, false colour images and radar data. Surveying and recording these sites during excavation has also been enhanced using digital photogrammetry – allowing high-resolution site-level data to be incorporated into wider landscape reconstructions. The resulting data are combined to construct site location models that have been proved and tested in other areas of the southern Red Sea. We also show how satellite imagery can be modified and exploited for seabed mapping and the search for underwater sites.

The past decade has seen a rapidly widening interest in archaeological and scientific exploration of the submerged landscapes that were flooded by sea-level rise at the end of the Last Glacial period. That interest is shared by many different disciplines and constituencies, including archaeologists who recognise the potential significance of these underwater archives to improved understandings of world prehistory, palaeoclimatologists interested in modelling sea-level change, and government agencies charged by national and international legislation with managing the underwater cultural heritage in the light of expanding industrial exploitation of the seabed. This introductory chapter sets out the background to these developments and the role of the European network – SPLASHCOS – in promoting awareness of this new agenda to the many scientific disciplines involved in underwater research, government agencies, commercial and industrial interests, and a wider public. Here we set out the major themes that we have used to structure the chapters in this volume, ranging across techniques and strategies of underwater investigation, examples of underwater archaeological excavations, reconstructions of underwater landscapes, the role of the continental shelf in shaping patterns of early human dispersal and geographical expansion, and issues of training, outreach and management. Examples are drawn widely from Europe and other parts of the world. We summarise the individual chapters, identify their inter-relationships with each other and with the themes of which they form a part, and highlight their wider significance.

Palynology of fossil faeces is still extremely rare and its contribution to the interpretation of the environment undervalued. In this paper we present the results of pollen analysis performed on cave hyena coprolites [Crocuta crocuta spelaea (Goldfuss, 1832)] from Marine Isotope Stage 3 (MIS 3) sites of Cava Muracci (Cisterna di Latina, central Italy) and Tana delle Iene (Ceglie Messapica, southern Italy). This study provides new insights into the vegetation of the Late Pleistocene of peninsular Italy, until now known only through long pollen records. It also shows how the content of coprolites, combined with faunal data and geochronological investigations, can represent a potential integrative source of palaeoclimatic proxy data. Our results indicate that the surroundings of both of the hyena dens were characterised by a patchy landscape with open lowlands dominated by steppe and grassland vegetation, while a few mesophilous and thermophilous trees were present in more humid areas, probably along the marine coast and inland. The harsh glacial climate appears to have been milder at Cava Muracci than at Tana delle Iene. This could be due either to the different environmental features of the two sites or to the high climatic variability that occurred during MIS3.

Our aim in this paper is to create a palaeoenvironmental and spatio-temporal framework for interpreting\ud human land use and exploitation of large mammals in the Central Kenya Rift over the past 2 million years,\ud with particular reference to the Nakuru-Elmenteita-Naivasha basin and its adjacent rift flanks on the\ud Kinangop Plateau and Mau escarpment. We pay particular attention to the tectonic and volcanic history\ud of the region, and to the system of lakes that have undergone periodic expansion and contraction\ud during\ud the Pleistocene in response to climatic and tectonic controls. We use this information to reconstruct\ud topographic features as they would have existed at different periods of the past and their likely influence\ud on patterns of large-mammal movements. In addition we present a systematic mapping of variations in\ud the mineral nutrients of soils – soil edaphics – based on 150 analyses of trace elements in modern soil and\ud vegetation samples. Soil edaphics play an important role in animal and human health today, with serious\ud deficiencies in some areas, and would have exerted an important constraint on animal distributions and\ud movements in the past. We show how soil-edaphic properties vary across the region and demonstrate a\ud close relationship between them and the underlying geology and sediments, which allows us to project\ud patterns back into the past and extrapolate them over larger areas. Using this combination of variables,\ud we construct maps that help to place archaeological sites into their wider regional setting. We focus in\ud particular on the Acheulean site of Kariandusi and show that it occupied a unique window of opportunity\ud in place and time for trapping mammals constrained to move through a narrowly defined topographic\ud bottleneck between edaphically-rich areas. We infer that the site was used for a simple form of ambush\ud hunting and briefly explore the implications for differential site use, formation, preservation and visibility.

The history of sea level within the Red Sea basin impinges on several areas of research. For archaeology and prehistory, past sea levels of the southern sector define possible pathways of human dispersal out of Africa. For tectonics, the interglacial sea levels provide estimates of rates for vertical tectonics. For global sea level studies, the Red Sea sediments contain a significant record of changing water chemistry with implications on the mass exchange between oceans and ice sheets during glacial cycles. And, because of its geometry and location, the Red Sea provides a test laboratory for models of glacio-hydro-isostasy. The Red Sea margins contain incomplete records of sea level for the Late Holocene, for the Last Glacial Maximum, for the Last Interglacial and for earlier interglacials. These are usually interpreted in terms of tectonics and ocean volume changes but it is shown here that the glacio-hydro-isostatic process is an additional important component with characteristic spatial variability. Through an iterative analysis of the Holocene and interglacial evidence a separation of the tectonic, isostatic and eustatic contributions is possible and we present a predictive model for palaeo-shorelines and water depths for a time interval encompassing the period proposed for migrations of modern humans out of Africa. Principal conclusions include the following. (i) Late Holocene sea level signals evolve along the length of the Red Sea, with characteristic mid-Holocene highstands not developing in the central part. (ii) Last Interglacial sea level signals are also location dependent and, in the absence of tectonics, are not predicted to occur more than 1–2 m above present sea level. (iii) For both periods, Red Sea levels at ‘expected far-field’ elevations are not necessarily indicative of tectonic stability and the evidence points to a long-wavelength tectonic uplift component along both the African and Arabian northern and central sides of the Red Sea. (iv) The observational evidence is consistent with tectonic and isostatic processes both operating over the past 300,000 years without requiring changes in the time averaged (over a few thousand years) tectonic rates. (v) Recent bathymetric data for the Bab al Mandab region have been compiled to confirm the location and depth of the sill controlling flow in and out of the Red Sea. Throughout the last 400,000 years the Red Sea has remained open to the Gulf of Aden with cross sectional areas at times of glacial maxima about 2% of that today. (vi) The minimum channel widths connecting the Red Sea to the Gulf of Aden at times of lowstand occur south of the Hanish Sill. The channels are less than 4 km wide and remain narrow for as long as local sea levels are below −50 m. This occurs for a number of sustained periods during the last two glacial cycles and earlier. (vii) Periods suitable for crossing between Africa and Arabia without requiring seaworthy boats or seafaring skills occurred periodically throughout the Pleistocene, particularly at times of favourable environmental climatic conditions that occurred during times of sea level lowstand.

Studies of the impact of physical environment on human evolution usually focus on climate as the main external forcing agent of evolutionary and cultural change. In this paper we focus on changes in the physical character of the landscape driven by geophysical processes as an equally potent factor. Most of the landscapes where finds of early human fossils and artefacts are concentrated are ones that have been subjected to high levels of geological instability, either because of especially active tectonic processes associated with faulting and volcanic activity or because of proximity to coastlines subject to dramatic changes of geographical position and physical character by changes of relative sea level. These processes can have both beneficial effects, creating ecologically attractive conditions for human settlement, and deleterious or disruptive ones, creating barriers to movement, disruption of ecological conditions, or hazards to survival. Both positive and negative factors can have powerful selective effects on human behaviour and patterns of settlement and dispersal. We consider both these aspects of the interaction, develop a framework for the reconstruction and comparison of landscapes and landscape change at a variety of scales, and illustrate this with selected examples drawn from Africa and Arabia.

We provide a critical review of the evidence for the long-term use of marine resources and coastal environments in human evolution and later development. We emphasise the importance of the coastal archaeological record in understanding patterns of human settlement and dispersal and draw attention to the large potential biases introduced by the destructive or obscuring effects of Pleistocene sea-level change. We note that lowered sea levels have been the norm for most of the Pleistocene and that periods of high sea level have been too short-lived to provide other than a fragmentary coastal record and one that is beset with ambiguities and uncertainties. We examine the geological processes of coastal uplift and underwater preservation that may help to mitigate these biases. Coastlines elevated by isostatic and tectonic processes, or with very steep offshore drop-offs at plate boundaries, are important in providing a potential window into coastal landscapes and archaeology formed during periods of lowered sea level. However, we conclude that the opportunities afforded by these geological factors are too limited to obviate the need for underwater exploration. We review the evidence now available from submerged landscapes across the Africa–Eurasia interface from the Atlantic coastlines of Northwest Europe to the southern Red Sea. We show that geomorphological conditions for the preservation of archaeological and palaeoenvironmental data are commonly present, that much material has already been discovered, and that new techniques, technologies and projects are providing the momentum for a rapidly expanding field of investigation. The results do not simply add to what we already know from sites on land, but are likely to produce qualitatively different evidence for coastal adaptations and landscapes that have no analogue on present-day coastlines. We note the strong probability that many coastal landscapes exposed at lowered sea level provided relatively fertile and productive refugia for plants, land mammals and humans at a time when increased aridity would have reduced or deterred hinterland occupation. We conclude that underwater investigation is essential if hypotheses of early human adaptation and dispersal are to be fully tested.

We explore the relationship between the edaphic potential of soils and the mineral properties of the underlying geology as a means of mapping the differential productivity of different areas of the Pleistocene landscape for large herbivores. These factors strongly control the health of grazing animals irrespective of the particular types of vegetation growing on them, but they have generally been neglected in palaeoanthropological studies in favour of a more general emphasis on water and vegetation, which provide an incomplete picture. Taking the Carmel–Galilee–Golan region as an example, we show how an understanding of edaphic potential provides insight into how animals might have exploited the environment. In order to simplify the analysis, we concentrate on the Lower Palaeolithic period and the very large animals that dominate the archaeofaunal assemblages of this period. Topography and the ability of soils to retain water also contribute to the differential productivity and accessibility of different regions and to patterns of seasonal movements of the animals, which are essential to ensure a supply of healthy fodder throughout the year, especially for large animals such as elephants, which require substantial regions of good grazing and browsing. Other animals migrating in groups have similar needs. The complex topography of the Southern Levant with frequent sudden and severe changes in gradient, and a wide variety of landforms including rocky outcrops, cliffs, gorges, and ridges, places major limits on these patterns of seasonal movements. We develop methods of mapping these variables, based on the geology and our substantial field experience, in order to create a framework of landscape variation that can be compared with the locations and contents of archaeological sites to suggest ways in which early hominins used the variable features of the landscape to target animal prey, and extend the analysis to the consideration of smaller mammals that were exploited more intensively after the disappearance of the elephants. We consider some of the ways in which this regional-scale approach can be further tested and refined, and advocate the development of such studies as an essential contribution to understanding the wider pattern of hominin dispersal.

We report on the analysis of fish bones of the African catfish, Synodontis schall (schall), recovered from the site of Tell el-Amarna in Middle Egypt. A systematic programme of sampling and sieving at this site has resulted in a large and well-preserved faunal collection from Pharaonic (14th century bc ) and Late Roman (5th–6th centuries ad ) deposits. Bones of schall are well represented, and in particular the pectoral spine, a robust and well-preserved anatomical element, which contains a record of incremental growth structures. We present here our methods of analysis, using a sample of modern schall collected at monthly intervals over a 12-month period to validate the interpretation of the incremental growth structures. We use these methods to identify variations in size and growth characteristics between Pharaonic, Roman and modern material and to throw light on changes in the ancient Egyptian economy and the ecology of the Nile. Our results demonstrate that the schall of the Late Roman period grew more slowly and were subject to more intensive levels of fishing pressure than those of the Pharaonic and modern periods. We interpret this as evidence of climatic deterioration and a concomitant increased reliance on fish for food during the 5th–6th centuries ad .

Research on the archaeology of the continental shelf has made slow and intermittent progress over the past 40 years amidst an ongoing mood of scepticism in the wider scientific and archaeological community, but with a marked acceleration of interest and investment of new resources during the past decade. The chapters in this volume demonstrate that relevant evidence is now being retrieved and systematically examined across the world in all the major continents and in deeper as well as shallower water, and that new ideas and new data are now providing the momentum for future development. This chapter draws on examples from the other chapters in the volume to reflect on the history of developments in this field and the shifts in the climate of opinion that have led to increasing acceptance of its importance; the research questions that are now coming more clearly into focus and the directions they suggest for the future; and the challenges of developing purposeful strategies of exploration and discovery of new material.

Atholl Anderson's comment (above) on the recent finds from Jerimalai draws attention to the dangers of over-interpreting the wider significance of marine resources present in Pleistocene coastal sites without careful evaluation of at least three variables: (1) the accurate identification of the species represented and hence of their behaviour and accessibility to capture; (2) the actual quantities and rates of accumulation of the marine food remains; and (3) local ecological and oceanographic conditions. To these I would add a fourth variable: the bathymetry and submerged topography of the marine environment adjacent to the sites in question—the physical structure of what one might call the 'offshore catchment'—and changes resulting from relative sea level variation (including eustatic and isostatic/tectonic effects).

Evidence of the AD 1755 tsunami consisting of the same type of accretions produced by the re-deposition of earlier sediments, has been recorded at three different height along the coast of Gibraltar: Along a shallow sandy shore, the tsunami wave reached a run-up of 2–3 m, whereas along steep, cliff–lined shores (Rosia Bay) it surpassed 5 m. An overwash deposit was also identified at the bottom of a lagoon (The Inundation), at 0.5 m b.s.l., on the isthmus that joins the Rock with the mainland. Southern submerged platforms (Vladi’s Reef) were also affected by the erosional backwash to a depth of 22 m. The tsunamigenic sediments exhibit a bimodal granulometry, mainly composed of sands with a coarser fraction composed of marine faunal shells remains, together with larger clasts derived from the rocky substrate. All remobilized sediments were dated by historical methods and radiocarbon dating. Se han registrado evidencias del tsunami de 1755 localizadas a tres alturas diferentes a lo largo de la costa de Gibraltar, todas ellas como resultado del mismo tipo de acreción producida por el re-depósito de sedimentos previos. A lo largo de las costas bajas y arenosas estudiadas, la ola del tsunami alcanzó una altura de 2 a 3 m, mientras que en litorales acantilados y con fuertes pendientes sobrepasó los 5 m. Se ha identificado un posible registro sedimentario en el fondo de una laguna costera (Inundation), a 0.5 m b.n.m., en el istmo que une la Roca con el continente. Por otro lado, las plataformas sumergidas, como la meridional de Vladi’s Reef, sufrieron erosión por la retirada de las olas a una profundidad de hasta 22 m. Los sedimentos tsunamigénicos poseen una granulometría bimodal, compuesta principalmente de arenas con una fracción más gruesa de restos de conchas marinas, junto con cantos mayores de rocas provenientes del sustrato. Todos los sedimentos removilizados por el tsunami han sido datados por métodos históricos y de radiocarbono. Research supported by the Spanish-FEDER Project CGL2010-15810/BTE, AHRC Research Grants (AH/E009409/1), and COST-action TD0902 SPLASHCOS (Submerged Prehistoric Archeology and Landscapes of the Continental Shelf). We thank reviewers and guest-editors for helpful comments. It is a contribution to IGCPs 526 (Risks, resources, and record of the past on the continental shelf), 567 (Earthquake Archaeology and Palaeoseismology), and 588 (Preparing for coastal change).

For most of human existence on this planet over the past 2 million years, sea level has been substantially lower than the present and has swung through changes of more than 100 m in response to the glacial–interglacial climatic cycle. At a time when modern society is increasingly concerned about the potentially destructive impact over the coming decades of a sea-level rise of 3 m or so, it is sobering to realize that prehistoric societies across the world faced a sea-level rise between about 16,000 and 6000 years ago of 130 m (Fig. 1). That change of course was spread over many human generations and many millennia, so that the full effects would not have been experienced within a single human lifetime. Nevertheless, the long-term cumulative effect of sea level rise and loss of territory would have been dramatic. On a world scale, substantial areas of continental shelf were successively exposed, creating potentially attractive territories for human settlement and migration and land connections between major land masses, and then removed again by sea level rise (Fig. 2). In Europe, during the last glacial period, the total land mass of the continent was extended by as much as 40% at the maximum marine regression (Fig. 3), with a corresponding loss of land when sea levels rose as the continental glaciers melted into the oceans. In some parts of the sea-level cycle, and especially in regions where the slope of the continental shelf is shallow, the effects would have been noticeable and sometimes dramatic within the lifetimes and memories of the people affected. Moreover, these changes have taken place repeatedly over the long Pleistocene history of human existence.

This paper examines the relationship between complex and tectonically active landscapes and patterns of human evolution. We show how active tectonics can produce dynamic landscapes with geomorphological and topographic features that may be critical to long-term patterns of hominin land use, but which are not typically addressed in landscape reconstructions based on existing geological and paleoenvironmental principles. We describe methods of representing topography at a range of scales using measures of roughness based on digital elevation data, and combine the resulting maps with satellite imagery and ground observations to reconstruct features of the wider landscape as they existed at the time of hominin occupation and activity. We apply these methods to sites in South Africa, where relatively stable topography facilitates reconstruction. We demonstrate the presence of previously unrecognized tectonic effects and their implications for the interpretation of hominin habitats and land use. In parts of the East African Rift, reconstruction is more difficult because of dramatic changes since the time of hominin occupation, while fossils are often found in places where activity has now almost ceased. However, we show that original, dynamic landscape features can be assessed by analogy with parts of the Rift that are currently active and indicate how this approach can complement other sources of information to add new insights and pose new questions for future investigation of hominin land use and habitats.

Your articles on human dispersal in the late Pleistocene epoch (Nature 485, 23; 2012) overlook the significance of now-submerged archaeological sites on the continental shelf during this period (126,000–11,000 years ago). It is wrong to assume that these were completely destroyed by the sea and that the interpretation of human movements must rely on proxy data, such as DNA or evidence from islands. More than 3,000 prehistoric sites on the seabed have been found and mapped, and in some cases excavated. They range in age from 500,000 to 5,000 years old, and at locations from the present-day shoreline out to a depth of 130 metres. These sites were extensive, often located on key travel routes and more attractive than arid hinterlands to human settlers. Marine archaeologists have recovered in-context stone artefacts, animal remains and human fossils from such sites. Some materials, including food remains, organics, bone, DNA and plants, are better preserved underwater than on land. Questions of early human dispersal will not be resolved until continental shelves are fully investigated — spurred by advances in modern oceanographic technology (see http://splashcos.org).

This paper presents a critical assessment of Killingley's approach to the determination of month of collection of marine molluscs by prehistoric people. The basis of the method is careful oxygen isotope measurements made in successive growth increments in the shells. We have analyzed specimens of Monodonta and Patella collected live on the coast of northern Spain, in conjunction with seasonality studies on molluscs from prehistoric sites in the neighborhood. These studies confirm the necessity of making careful analyses of each species under consideration. Given the significance both of interspecies differences and of climatic variability on timescales from a year upwards (particularly important in Killingley's area), we conclude that his apparent accuracy of ± a month is illusory.

Shell middens are one of the most important and widespread indicators for human exploitation of marine resources and occupation of coastal environments. Establishing an accurate and reliable chronology for these deposits has fundamental implications for understanding the patterns of human evolution and dispersal. This paper explores the potential application of a new methodology of amino acid racemization (AAR) dating of shell middens and describes a simple protocol to test the suitability of different molluscan species. This protocol provides a preliminary test for the presence of an intracrystalline fraction of proteins (by bleaching experiments and subsequent heating at high temperature), checking the closed system behaviour of this fraction during diagenesis. Only species which pass both tests can be considered suitable for further studies to obtain reliable age information. This amino acid geochronological technique is also applied to midden deposits at two latitudinal extremes: Northern Scotland and the Southern Red Sea. Results obtained in this study indicate that the application of this new method of AAR dating of shells has the potential to aid the geochronological investigation of shell mounds in different areas of the world.

International audience; The study area mainly includes the continental shelves of two European states (France and Spain), three North African countries (Algeria, Morocco and Tunisia), and Gibraltar and Monaco. Overall, the total shelf area in the western Mediterranean down to the 150 m isobath is 208,000 km2 within a total basin area of 1,343,620 km2. There is a major difference in the degree of available information from EU member states and North African countries.