Coastal contexts; Coastal habitats
Within at least the last 70,000 years of human history, coasts have been favored increasingly as places for people to settle, typically in order to take advantage of coastal foods, as well as places from which to interact readily with other human groups (Bailey and Parkington, 1988). Long-standing claims that coasts stimulated the development of agriculture by allowing Homo sapiens a unique opportunity to domesticate plants (Sauer, 1962; Binford, 1968) remain credible. Yet it is also apparent that “many of the world’s coastlines that have the most productive environmental conditions for heavy dependence on marine and intertidal resources … were only colonised by human populations relatively recently” (Bailey, 2004, 41). Coastal food exploitation was so important to some societies that it came to define them (Szabó and Amesbury, 2011; Jew et al., 2013), whereas for others it was less important, something that might feature in societal subsistence only during periods when preferred options were unavailable (Holdaway et al., 2002; Cleuziou and Tosi, 2007).
Food was only one reason why humans have been attracted to coasts throughout the past 100,000 years. The relative ease of movement along coastal strips, where intervening barriers like mountain ranges are less common than in inland areas, meant that the main movements of many early human groups were along continental coasts (Davidson, 2013). Later, as maritime networks became important for trade, coasts were increasingly settled, a major factor in explaining the subsequent concentration of people in coastal cities (von Glasow et al., 2013).
Owing to their location at the nexus of the lithosphere, hydrosphere, and atmosphere, coasts are among the most changeable of Earth’s landscapes and have long posed unique challenges to people who sought to locate their permanent settlements there. Foremost among these challenges are changes to coastal landforms and (food-producing) ecosystems resulting from changes, typically gradual, in the level of the land-sea interface caused by relative sea-level change. More localized changes may result from abrupt tectonic change, the impacts of large waves, and by changes in coastal topography as might be driven by the development of offshore reefs or by changes in process regime such as nearshore sedimentation.
It is important on one hand to understand that coastal settings occupied by humans were and still are invariably dynamic, yet it is also important to appreciate that the present is not necessarily a helpful guide to the past. The processes that control the form of a modern coastline are not always those that existed in the past. Further, many ancient coastal occupation sites are now often far from the modern shoreline, perhaps inland or underwater, emphasizing the susceptibility of coasts to change.
This entry looks first at the most common coastal settings occupied by humans, describing examples from different parts of the world, before discussing the coevolution of coastal environments and coastal societies, illustrating the complexities of this with various examples. Finally there is discussion of several key issues in the contemporary understanding of coastal settings for past human activities.
Common coastal settings
Coastal peoples of the past occupied the full range of coastal settings, but in most parts of the world, they favored a few that – at least at the time of their initial occupation – maximized livelihood opportunities, particularly for subsistence. This means that there are few cliff-top or hinterland locations that were intentionally selected in preference to adjacent coastal lowlands. Exceptions are found on high cliffed oceanic islands such as Rurutu (French Polynesia) and many of those in the Marianas Archipelago (northwest Pacific), either because of the difficulties of occupying the shoreline or because of the impoverished resources it contained compared to elsewhere (Weisler et al., 2010; Amesbury, 2013). Three of the most common coastal settings that became sites for permanent human settlement in almost every part of the world are coastal plains, river mouths, and islands, each of which is discussed separately below.
Varying in breadth from perhaps a few tens of meters, as often found around the fringes of high oceanic islands, to several hundred kilometers, such as were available for human use along many continental margins during the low sea levels of the last glacial period, coastal plains have long attracted humans much as they do today. It can be demonstrated that, at times when population densities were comparatively low, human interest in such coastal plains was proportional to the width of the plain (Lowery et al., 2012).
In many places today, coastal plains represent features created only after sea level fell from its mid-Holocene maximum at 4–6 ka BP (typically 1–2 m above present; see Figure 1d); for this reason, they are found today along most parts of the world’s coastline after having become available for human settlement only in the later Holocene (i.e., after 4000 years BP). For example, it has been argued that the initial colonization of the tropical Pacific Islands was controlled by regionally variable sea-level fall in the late Holocene (Dickinson, 2003). More commonly, coastal plains, or at least their younger parts, were occupied progressively as sea level fell during the late Holocene. Haifa Bay and the Zevulun Valley Plain in the eastern Mediterranean exemplify this situation (Porat et al., 2008). As these coastal plains grew seaward with gradual sea-level decline, they were sometimes extended by the accumulation along the coast of sediments derived both from land and shallow offshore (underwater) shelves, particularly those on which (coral-) reef-fringed lagoons had developed. The interplay of erosion and sedimentation accounts for many of the variations in coastal landscapes found along coastal plains. Human manipulation of coastal landscapes sometimes produced unanticipated effects. Examples include the harbors constructed by trading peoples along the coasts of the Mediterranean and Red Sea that subsequently started to infill and, despite ingenious attempts to prevent this, invariably fell into disuse with often profound societal consequences (Marriner and Morhange, 2007; Hein et al., 2011).
Landscape elements of coastal plains are numerous, but those that appear key for most settlers in such settings include sheltered sites, on straight coasts typically behind dunes/berms, beach ridges, or around the fringes of coastal lagoons and wetlands. Examples of back-dune occupations may be found in Fiji and Oman, where they were established by many seasonally transient groups (Anderson et al., 2006; Cavulli and Scaruffi, 2013), and coastal-lagoon occupations were a critical step in the emergence of modern societies along the northeastern New Guinea coast (Terrell, 2002). Occupation sites in coastal wetlands were generally situated at higher locations where crops might be cultivated to supplement foods obtained from surrounding areas. Lakes were important components of many wetlands. One example is Šventoji (western Lithuania), where the role of late Holocene sea-level oscillations is clear in transforming coastal environments and influencing human lifeways, particularly through the successive creation and infilling of coastal lagoons (Stančikaitė et al., 2009).
Along embayed coasts, it was possible for humans to live directly on the shore in places sheltered from dominant winds and waves. The founder settlement in the Tonga archipelago (southwest Pacific) at Nukuleka was located in the lee of a sand spit at the entrance to the Fanga ‘Uta lagoon on the leeward side of Tongatapu Island (Burley and Dickinson, 2001). Spatial and temporal variations in human occupations along the coasts of the sinuous Limfjord (Denmark) were regulated by various physical factors, including salinity and sedimentation (Lewis et al., 2013).
Where a particular coast is fringed by an offshore reef, which reduces the size of waves reaching the shore, shelter was a less important concern, and early settlers sometimes even built stilt-house settlements over shallow water. For example, the use of such settlements is a defining feature of early Lapita colonizers in the southwest Pacific (Green, 2003).
River mouths are attractive to humans because of the access both to inland areas and freshwater resources, as well as many coastal resources. Well-documented examples of river-mouth cultures include those around the estuaries of the Irrawaddy (Myanmar), Sông Hồng or Red River (Vietnam), as well as many smaller ones (Stark, 2006). River mouths are often more dynamic environments than coastal plains or most parts of islands, requiring more frequent adjustments by their occupants. A study of archaeological sites at the mouth of the Santa River (Peru) was one of the first to show the dynamic nature of settlement patterns in such places (Wells, 1992: see Figure 4).
Highly valued for their fertile, well-watered soils as well as ready access to aquatic food resources, river deltas are highly dynamic river-mouth environments, and their coastal fringes especially so. Some of the most profound changes have involved abrupt shifts in river mouths that forced changes in the locations of human activities in the past, leading in some cases to migrations and cultural shifts. Examples come from the Huanghe (Yellow) and Yangtze River deltas in coastal China (Chen et al., 2008) and from the Danube Delta in southeastern Europe (Romanescu, 2013). Deltas are especially prone to flooding, yet there are several studies showing how people in the past continued to utilize these areas during wetter periods. An example comes from the Sông Hồng, or Red River, Delta (Vietnam) (Funabiki et al., 2012).
In the past few centuries or more, the dynamics of environmental changes in deltas elsewhere have been so marked that delta cities, well known in antiquity, have proved controversial to rediscover. A good example is Vineta, which was considered to be “the greatest of all cities in Europe” by the traveler Ibrahim Ibn Yaqub in AD 970; it was probably located on a marshy island in the Oder River Delta (Poland) (Brysac, 2003). The city of Herakleion in the Nile Delta (Egypt) likewise disappeared into the realm of myth for several centuries after its abandonment in the first century AD following river-channel shifts and subsidence (Stanley et al., 2004).
Outgrowth of river deltas – a result of both sea-level fall and (accelerated) terrestrial sediment delivery to river mouths – has led to the creation of new areas for coastal settlement that grew in importance in many places. The progradation of the Grijalva Delta (Mexico) witnessed the embryonic emergence of maize cultivation in Mesoamerica, as the first farmers in the region occupied its fertile fringes (Pope et al., 2001).
Smaller islands may be effectively coastal in their entirety (or very nearly so), and they often exhibit the full range of coastal settings found along the fringes of larger landmasses. Yet on such islands, owing to a lack of large rivers, coasts may feature less dynamic, more resource-rich ecosystems of the kind that perhaps drove, or at least sustained, the successive eastward colonization of tropical Pacific islands (Kennett et al., 2006). At population densities below carrying capacity, islands provided attractive environments for potential settlers because of this resource richness, but also perhaps because their restricted areas made them easier to manage than more diverse landmasses. For example, critical to traditional food production in prehistoric Hawaii was the ahupua’a system that involved the vertical integrated management of food systems from the highland island centers downslope, across the coast, and into the deep ocean (Kagawa and Vitousek, 2012).
Yet precisely because of their circumscribed nature, such islands were also susceptible to abandonment by prehistoric populations when resources were no longer sufficient to sustain them. Illustrations of this include the Line Islands (central Pacific) and the Pitcairn Island group (southeast Pacific), both of which were abandoned several hundred years ago (Weisler, 1996; Di Piazza and Pearthree, 2001).
Coastal settings on islands vary largely according to the composition and form of the island in question. Older volcanic islands tend to have irregular coasts with numerous indentations marking drowned river valleys, while younger volcanic islands may have more regular coasts. There is no tradition of coastal settlement along the “iron-bound” (fringed by young, hard lava rock) coasts of young Savai’i Island (Samoa), for example. Most limestone islands are similar, and many of those that have emerged recently possess comparatively little coastal lowland. One example is Niue Island (central Pacific), where most settlement is along a 23-m-high cliff-top terrace (Nunn and Britton, 2004).
Atolls are types of low (wholly coastal) island that, in parts of the northwest Pacific, have been occupied continuously for around 2000 years. Their restricted terrestrial biota meant that subsistence was dominated by food acquisition within shallow-water reefs and lagoons, although this came to be supplemented by novel techniques of onshore production (Kayanne et al., 2011).
Changing coastal settings and their effects on coastal societies
As coastal environments were transformed by climate-driven (especially sea-level) and tectonic changes, their human occupants were often forced to adapt. Yet humans sometimes altered the coastal landscapes they inhabited, and these modifications often brought about unanticipated responses that also required adaptation.
In earlier societies, adaptation commonly required refocusing subsistence strategies, particularly as food-producing ecosystems changed. In more recent times, owing to the greater potential impacts of humans, it is often more difficult to separate the anthropogenic effects from environmental changes driven by natural causes (see below). The present section illustrates these points by looking at the parallel evolution of coastal environments and societies during specific periods of prehistory. Since the potential for tectonic disruption of coastal sites is not time dependent in the way that climate-driven changes have invariably been, coastal sites affected by tectonism are considered in a separate section.
Pleistocene times (>10,000 years ago)
Some of the first forays out of Africa by Homo sapiens may have intentionally followed coastal routes because of the availability of coastal resources (Bailey, 2009). A particular case in point is the southern dispersal route that early humans are thought to have taken from Africa to Southeast Asia (Field et al., 2007), which may have initially involved crossing the Red Sea at a time when sea level was lower and the Bab al-Mandab Straits were relatively easy to cross (Bailey et al., 2007). At such times, presently arid coasts like those of the Farasan Islands and mainland Arabia may have received more rainfall than today, permitting contemporary occupation of a number of coastal settings. There are indications that early humans in this area also targeted nearshore marine resources; some giant clam species show signs of having been overharvested here as much as 125,000 years ago (Richter et al., 2008).
Implicit in the crossing of the Wallace Line from Sunda to Sahul (Southeast Asia to Australia) perhaps 60,000 years ago is a familiarity with tropical coasts that included a degree of maritime technology sufficient to permit the successful crossing of ocean gaps as much as 70 km wide. Most likely, the sites occupied by such humans were on coastal plains fringed by broad reefs or around river mouths where coastal inhabitants depended largely on intertidal and reef-flat food resources but also pelagic resources (Balme, 2013). It is possible that coastal environmental changes, perhaps driven by last glacial sea-level fluctuations, stimulated pioneer occupations of offshore islands in Sunda, leading eventually to the first colonization of Australia and New Guinea (Sahul).
A final example of Pleistocene human interaction with coasts is that involving initial human arrival in the Americas ca. 16,000 BP from the western Pacific Rim, perhaps along a shoreline that is now largely invisible due to submergence by later sea-level rise and fragmentation by differential tectonics (Erlandson and Braje, 2011). These authors note that, at the time, this former shoreline would have been associated with “rich and diverse resources from both marine and terrestrial ecosystems” (p. 28), perhaps a “kelp highway” that could have sustained migrants from Japan through the Kuril Islands to Kamchatka, the southern shores of Beringia, and thence along the ice-free coastal fringe of glacier-covered North America to the California coast from where migrants began to settle onshore and inland (Erlandson et al., 2007).
Early Holocene times (10,000–6000 years ago)
Postglacial sea-level rise drowned ice-age coasts in almost every part of the world, causing their inhabitants to move either inland and upslope or offshore. Along many coasts during the early Holocene, sea-level rise was accompanied by broadening of the diet range of their human inhabitants (Marín-Arroyo, 2013). This diet change may have been a response to changes in productive coastal environments and ecosystems associated with sea-level rise in addition to economic intensification linked to increases in coastal populations.
During the early Holocene, accompanied by climate changes, sea level rose about 60 m but this rise was neither monotonic nor regionally uniform (Smith et al., 2011). Sea-level transgression was instead oscillatory; there were periods of rapid rise and periods of temporary fall. For example, sea level rose about 3 m in 200 years during the 8200 BP Event (Hijma and Cohen, 2010). In coastal Portugal, this led to rapid submergence of the lower Tagus Valley that was accompanied by a massive decrease in the availability of coastal foods that forced an inland shift of coastal people who had previously occupied now-drowned coastal plains (Bicho et al., 2010). Similar events at the same time may have displaced coastal people occupying parts of the western Pacific Rim leading them to undertake deliberate voyages to settle lands beyond development of the horizon, something that may have led eventually to the maritime traditions that involved the earliest occupations of Pacific oceanic islands (Nunn, 2007b).
Middle Holocene times (6000–3000 years ago)
After global sea level stabilized along most coasts around 7000–5000 BP, the bioproductivity of nearshore coastal ecosystems increased sharply leading to the comparatively rapid occupation of many coasts by humans during the middle Holocene (Day et al., 2012). An example comes from the northern shoreline of New Guinea, today the largest Pacific “island” but until about 7500 years ago a promontory of Australia. Except in a few places, coastal settlement during the terminal Pleistocene and early Holocene was not possible in this part of New Guinea because of its fringe of steep cliffs plunging into deep water, but by about 6000 BP, when sea level had reached close to its present level, many of these coasts “started to evolve into rich floodplains, river deltas and lagoons” (Terrell, 2004, 605). People began to settle these areas and establish trade networks.
While coastal societies often respond to climate-driven sea-level changes, there are also instances where these societies have had to respond to climate changes alone. One example comes from coastal Syria where separate periods of increased aridity in the mid-late Holocene transformed productive coastal plains into hot desert (Kaniewski et al., 2008). A comparable study of the southwest coast of the Barents Sea (Norway) found that alternating warm and cool climates during the last 2000 years could be correlated with economies based mainly on cereal agriculture and fisheries, respectively (Sjögren, 2009).
Late Holocene times (<3000 years ago)
Along Mediterranean coasts, the sea-level rise of the earlier Holocene continued into the late Holocene, so that the coasts in this area show signs of continued transgression. In western Greece, for example, people at the start of the late Holocene settled close to the shore of the saltwater Messolonghi Lagoon and the adjoining freshwater Etoliko Lagoon, exploiting these diverse environments until about AD 1350 when continuing sea-level rise breached the isthmus joining both lagoons (Haenssler et al., 2013).
Within the last millennium in many parts of the world’s coasts, minor perturbations of sea level occurred that in some cases led to significant responses from coastal-dwelling humans. Some of the most notable occurred as a result of rapid sea-level fall of as much as 80 cm during the AD 1300 Event (approximately AD 1250–1350) along many Pacific island coasts. This sea-level fall led to the rapid depletion of coastal foods, both onshore as a result of water-table lowering and offshore as a result of reef-surface exposure and increased lagoon turbidity. In turn, this led to conflict and, along many island coasts, the abandonment of coastal settlements in favor of others in fortifiable positions, typically upslope. Examples are known from most tropical Pacific island groups (Nunn, 2007a) with more recent work reported from Fiji and Timor (Nunn, 2012; O’Connor et al., 2012). In addition to having direct impacts on human settlement, the AD 1300 sea-level fall also caused changes to coastal landscapes that are implicated in coeval cultural changes. These include the infilling of coastal embayments and the emergence of offshore reef flats on which islands grew or could be built. Examples of the former include Tikopia (Solomon Islands) and Kawai Nui Marsh on O’ahu Island in Hawaii (USA), while the latter is exemplified by fortified Lelu Island off the coast of Kosrae (Federated States of Micronesia) and perhaps some of the artificial islands off the coast of Malaita (Solomon Islands) (Nunn, 2007a).
Tectonic disruption of coastal sites
Tectonic activity has significantly impacted the activities of coastal populations, particularly along coasts adjoining convergent lithospheric-plate boundaries. The most disruptive types of tectonic change are usually those that are abrupt and involve rapid uplift or subsidence of as much as several meters. The impacts of (associated) tsunamis often cause major problems.
The ancient Achaean city of Helike (Greece), located on a delta of the Gulf of Corinth coast, was abruptly submerged during an earthquake-tsunami in 373 BC but was subsequently uplifted, and its remains now lie buried beneath post-earthquake delta sediments (Soter and Katsonopoulou, 2011). Seismic subsidence accompanied by liquefaction caused major disruptions on at least two occasions to inhabitants of the port city of Ayla (now Aqaba, Jordan) (Al-Tarazi and Korjenkov, 2007). In the tectonically active island arcs of the southwest Pacific Ocean, many similar instances are known, the most extreme being ones where entire inhabited islands abruptly sank, in most cases probably due to an earthquake-triggered landslide along an adjoining ocean trench. Examples include the “vanished” islands of Teonimanu (Solomon Islands) and Malveveng and Tolamp (Vanuatu) (Nunn, 2009b).
Away from plate boundaries, some of the most marked coastal changes have occurred in places like Scandinavia where the land is rising as a result of isostatic rebound. The Viking-era shoreline (AD 800–1050) on the Estonian coast, for example, is now known to be 3–4 m above present sea level, emphasizing the importance in such places of understanding “the relation between a given site and the shoreline at the time when the site was used” (Ilves and Darmark, 2011, 147–148).
Conclusions and key issues
When considering the relationship between coastal settings and past human societies, there are a number of issues that remain insufficiently understood or acknowledged by many researchers. Three such issues are discussed below.
The importance of understanding coastal paleoenvironments
Despite calls to consider the effects of postglacial sea-level rise on coastal archaeological records, “many archaeologists working in coastal areas … have ignored such warnings” (Erlandson and Braje, 2011, 34) and continue to interpret past human-environment interactions in terms of the landscape configurations and landscape-forming processes they observe today. Such comments also extend to dynamic river deltas, such as the tendency of Egyptologists to plot the present-day rivers of the Nile Delta onto maps of the valley in the past, something that may have “misled interpretations of ancient monuments and settlements” (Hillier et al., 2007, 1011). This is hard to justify in an age when the understanding of (coastal) landscape change is so far advanced, and there are numerous examples of how this understanding can provide insights into the development of ancient societies.
Examples are illustrated above. The Lapita-age settlement at Bourewa (Fiji) was reconstructed using observations of settlement character (particularly the extent of postholes that once supported over-reef stilt houses), reef configuration, and an understanding of sea-level change to demonstrate that this settlement had extended along a submerged sand spit which emerged subsequently (see Figure 2), changing the possibilities for coastal subsistence (Nunn, 2009a). Morphological changes at the mouth of the Santa River (Peru) linked to sea-level changes explain the changes in settlement pattern shown in Figure 4 (Wells, 1992).
The locations of ancient shell middens can plot out coeval shorelines, assuming that shellfish gatherers processed their harvest just beyond the reach of high tide rather than carrying it whole back to their communities. There is no single answer. In some places, the former appears true – most shell middens (Køkkenmødding) in Denmark are situated along former shorelines which have been reconstructed using midden locations (Gutiérrez-Zugasti et al., 2011). In contrast, marine shell concentrations that differ little from those in coastal middens have been found 23 km inland in the Norte Chico region (coastal Peru) and 10 km inland on northern Viti Levu Island (Fiji) (Creamer et al., 2011; Robb and Nunn, 2014).
The identification of ancient shorelines and the signs of their human usage (such as boat landings) is particularly difficult in places where the Earth’s crust is rebounding isostatically because this movement tends to be monotonic (only one direction: upward) rather than episodic. One approach in such places has been to use phosphate mapping to identify former shorelines, the assumption being that the human occupation of these localities resulted in an increase in surficial phosphate concentrations (from the butchering of game, the gutting of fish, human waste, burning); such a study resulted in the mapping of Österby Harbor (northwest Estonia) (Ilves and Darmark, 2011).
Imperatives for underwater archaeology
Understanding how shoreline emergence might affect coastal settings occupied by ancient humans can be obtained by investigating the sites on dry land, but the effects of shoreline submergence generally represent greater challenges. It is not simply an issue of former settlement sites being underwater but also that the material evidence associated with these sites may have been fragmented and dispersed as a result of submergence, first by the encroaching swash and backwash of wave action as the sea slowly enveloped the sites and then by further deterioration as the sites slowly sank into deeper water. For this reason, it is easier to reconstruct those sites that have been submerged only a few meters relatively recently rather than those that may be lying at water depths of 100 m or more over a lengthy interval, as is the case for most Last Glacial Maximum sites (dating from 22 to 18 ka BP) that now lie in offshore continental shelf locations. It is also important to appreciate that, once submerged, coastal settlements may become buried by sediments and even overgrown by reefs, which makes these settlements difficult to identify in many places. The danger is that they will be assumed to have been absent at a particular time, and erroneous chronologies of human history constructed as a result.
Since so much of the evidence is underwater, there are understandably few data-rich case studies of coastal societies affected by early Holocene sea-level rise. Some of the most compelling of these depend on data gathered through techniques of underwater archaeology. Underwater archaeological investigations have the potential not only to provide contexts for inferences from on-land sites (Bailey and King, 2011), but they can also demonstrate the existence of unsuspected coastal settings at particular times which in turn inform regional settlement models. An example comes from the Gulf of Maine (USA) where a slowing of postglacial sea-level rise 11,500–7500 years ago allowed development of coastal barriers and wetlands that may have attracted settlers, a finding contrary to earlier assumptions about the habitability of such formerly ice-covered coasts (Kelley et al., 2010). Perhaps the most comprehensive survey of now-submerged coastal settings occupied by a maritime society during the early Holocene comes from Doggerland (North Sea, between Britain and continental Europe) where a low-relief landscape of marshes, lakes, and wetlands dissected by rivers, now buried under marine sands, can been traced (Gaffney et al., 2009).
Debating the relative roles of natural and human processes in prehistoric coastal change
The inherent natural dynamism of coastal settings has not deterred them from being, through much of human existence, favored places for people to live, and as a result, it is not always easy to retrospectively distinguish the effects of natural and anthropogenic actions. There are many well-documented examples of ways in which humans modified coastal environments as well as examples of how coastal environmental change, linked to extraneous climate-driven changes, forced coastal dwellers to change the ways in which they lived. To judge which cause (natural or human) of an observed change in coastal societies was dominant, it is necessary to compare chronologies of both. If societal change was clearly not synchronous with possible (natural) forcing variables, then the latter is unlikely to have played a role in causing the former. But if there is demonstrable synchronicity, then a role for natural forcing should be considered possible.
Relative sea-level change is a major mechanism that could have forced adaptive change in coastal societies, and several examples were described above (see also Figure 3). Over the past decade, there has been an increasing number of case studies in which relative sea-level migration is cited as a major cause of change among coastal societies; these include tropical Pacific islands (Nunn, 2007a) and the coasts of Italy (Romano et al., 2013) and Portugal (Bicho and Haws, 2008). In many cases, the separation of natural and human causes of societal change appears almost impossible to achieve (e.g., Marín-Arroyo, 2013) given the paucity of available data.
- Amesbury, J. R., 2013. Pelagic fishing in the Mariana Archipelago: from the prehistoric period to the present. In Ono, R., Morrison, A., and Addison, D. J. (eds.), Prehistoric Marine Resource Use in the Indo-Pacific Regions. Canberra: Australian National University E Press, pp. 33–57. Terra Australis 39.Google Scholar
- Bailey, G., 2004. World prehistory from the margins: the role of coastlines in human evolution. Journal of Interdisciplinary Studies in History and Archaeology, 1(1), 39–50.Google Scholar
- Bailey, G. N., 2009. The Red Sea, coastal landscapes, and hominin dispersals. In Petraglia, M. D., and Rose, J. I. (eds.), The Evolution of Human Populations in Arabia. Amsterdam: Springer, pp. 15–37.Google Scholar
- Bailey, G. N., and Parkington, J. (eds.), 1988. The Archaeology of Prehistoric Coastlines. Cambridge: Cambridge University Press.Google Scholar
- Bailey, G. N., Flemming, N. C., King, G. C. P., Lambeck, K., Momber, G., Moran, L. J., Al-Sharekh, A., and Vita-Finzi, C., 2007. Coastlines, submerged landscapes, and human evolution: the Red Sea Basin and the Farasan Islands. Journal of Island and Coastal Archaeology, 2(2), 127–160.CrossRefGoogle Scholar
- Binford, L. R., 1968. Post-pleistocene adaptations. In Binford, S. R., and Binford, L. R. (eds.), New Perspectives in Archaeology. Chicago: Aldine, pp. 313–341.Google Scholar
- Brysac, S. B., 2003. Letter from Germany: Atlantis of the Baltic. Archaeology, 56(4), 62–66.Google Scholar
- Cleuziou, S., and Tosi, M., 2007. In the Shadow of the Ancestors: The Prehistoric Foundations of the Early Arabian Civilization in Oman. Muscat: Ministry of Heritage and Culture.Google Scholar
- Di Piazza, A., and Pearthree, E., 2001. An island for gardens, an island for birds and voyaging: a settlement pattern for Kiritimati and Tabuaeran, two “mystery islands” in the northern Lines, Republic of Kiribati. Journal of the Polynesian Society, 110(2), 149–170.Google Scholar
- Dickinson, W. R., 2003. Impact of mid-Holocene hydro-isostatic highstand in regional sea level on habitability of islands in Pacific Oceania. Journal of Coastal Research, 19(2), 489–502.Google Scholar
- Gaffney, V. L., Fitch, S., and Smith, D. N., 2009. Europe’s Lost World: The Rediscovery of Doggerland. York: Council for British Archaeology.Google Scholar
- Green, R. C., 2003. The Lapita horizon and traditions – signature for one set of Oceanic migrations. In Sand, C. (ed.), Pacific Archaeology: Assessments and Prospects (Proceedings of the International Conference for the 50th Anniversary of the First Lapita Excavation, Koné-Nouméa 2002). Les cahiers de l'archéologie en Nouvelle-Calédonie 15. Nouméa, Nouvelle-Calédonie: Département Archéologie, Service des Musées et du Patrimoine de Nouvelle-Calédonie, pp. 95–120.Google Scholar
- Kennett, D. J., Anderson, A., and Winterhalder, B., 2006. The ideal free distribution, food production, and the colonization of Oceania. In Kennett, D. J., and Winterhalder, B. (eds.), Behavioral Ecology and the Transition to Agriculture. Berkeley: University of California Press, pp. 265–288.Google Scholar
- Kirch, P. V. (ed.), 2001. Lapita and Its Transformations in Near Oceania: Archaeological Investigations in the Mussau Islands, Papua New Guinea, 1985–88. Berkeley: Archaeological Research Facility, University of California, Vol. 1.Google Scholar
- Lewis, J. P., Ryves, D. B., Rasmussen, P., Knudsen, K. L., Petersen, K. S., Olsen, J., Leng, M. J., Kristensen, P., McGowan, S., and Philippsen, B., 2013. Environmental change in the Limfjord, Denmark (ca 7500–1500 cal yrs BP): a multiproxy study. Quaternary Science Reviews, 78, 126–140.CrossRefGoogle Scholar
- Miotk-Szpiganowicz, G., Zachowicz, J., and Uscinowicz, S., 2010. Palynological evidence of human activity on the Gulf of Gdansk coast during the late Holocene. Brazilian Journal of Oceanography, 58(spe1), 1–13.Google Scholar
- Nunn, P. D., 1999. Environmental Change in the Pacific Basin: Chronologies, Causes, Consequences. New York: Wiley.Google Scholar
- Nunn, P. D., 2009b. Vanished Islands and Hidden Continents of the Pacific. Honolulu: University of Hawai’i Press.Google Scholar
- Nunn, P. D., and Britton, J. M. R., 2004. The long-term evolution of Niue Island. In Terry, J. P., and Murray, W. E. (eds.), Niue Island: Geographical Perspectives on the Rock of Polynesia. Paris: INSULA, pp. 31–74.Google Scholar
- Robb, K. F., and Nunn, P. D., 2014. Changing role of nearshore-marine foods in the subsistence economy of inland upland communities during the last millennium in the tropical Pacific Islands: insights from the Bā River Valley, Northern Viti Levu Island, Fiji. Environmental Archaeology, 19(1), 1–11.CrossRefGoogle Scholar
- Romano, P., Di Vito, M. A., Giampaola, D., Cinque, A., Bartoli, C., Boenzi, G., Detta, F., Di Marco, M., Giglio, M., Iodice, S., Liuzza, V., Ruello, M. R., and Schiano di Cola, C., 2013. Intersection of exogenous, endogenous and anthropogenic factors in the Holocene landscape: a study of the Naples coastline during the last 6000 years. Quaternary International, 303, 107–119.CrossRefGoogle Scholar
- Sauer, C. O., 1962. Seashore – primitive home of man? Proceedings of the American Philosophical Society, 106(1), 41–47.Google Scholar
- Specht, J., 2007. Small islands in the big picture: the formative period of Lapita in the Bismarck Archipelago. In Bedford, S., Sand, C., and Connaughton, S. P. (eds.), Oceanic Explorations: Lapita and Western Pacific Settlement. Canberra: Australian National University ePress, pp. 51–70.Google Scholar
- Terrell, J. E., 2002. Tropical agroforestry, coastal lagoons, and Holocene prehistory in Greater Near Oceania. In Yoshida, S., and Matthews, P. J. (eds.), Vegeculture in Eastern Asia and Oceania. Osaka: Japan Center for Area Studies, National Museum of Ethnology, pp. 195–216.Google Scholar
- Von Glasow, R., Jickells, T. D., Baklanov, A., Carmichael, G. R., Church, T. M., Gallardo, L., Hughes, C., Kanakidou, M., Liss, P. S., Mee, L., Raine, R., Ramachandran, P., Ramesh, R., Sundseth, K., Tsunogai, U., Uematsu, M., and Zhu, T., 2013. Megacities and large urban agglomerations in the coastal zone: interactions between atmosphere, land, and marine ecosystems. Ambio, 42(1), 13–28.CrossRefGoogle Scholar
- Weisler, M. I., 1996. Taking the mystery out of the Polynesian “mystery” islands: a case study from Mangareva and the Pitcairn Group. In Davidson, J. M., Irwin, G., Leach, B. F., Pawley, A., and Brown, D. (eds.), Oceanic Culture History: Essays in Honour of Roger Green. Dunedin North: New Zealand Journal of Archaeology Special Publication, pp. 615–629.Google Scholar