Buried and Submerged Landscapes

This important argument has put the process of marine transgression into a whole new light. In the context of landscape archaeology, it has now emerged as a central mechanism, if not a primary trigger, for the early concentration of settlements in littoral regions - as well as for the formation of the first urban centers in the near East. This new evidence has unequivocally raised the issue of environmental change in general, and the role of sea-level rise in the formation of estuary, wetlands, and marsh habitats, in particular, to be the central causal factor or impetus for the formation of sedentary settlements. In tandem with the release of Kennett’s and Kennett’s paper on the role of seal-level rise in the development of early Near Eastern civilizations, two archaeologists, Erlandson and Fitzpatrick released a paper in 2006 (see additional reading) calling for a recognition of the archaeological potential of submerged archaeological sites and called for regional studies that deal with local variation.

.. ..careful reconstructions of coastal paleography and predictive modeling....to help determine the most likely locations of ancient coastal settlements as well as the places where settlements are most likely to be preserved...

This call to arms dovetails nicely with new insights into late Holocene sea-level rise and the ability of 3D geospatial techniques (3D GIS and terrain modeling) to capture the topographic subtlety and changing configuration of formally exposed cultural habitats (Figure 3).

Holocene Fluctuations and Exposure

Recently, the availability of tightly dated and micro-sampled pollen cores have revealed far-reaching new insights into the Late Holocene trends in sea-level rise that appear to contradict widely held assumptions about climate change over the past 3000 years. Just as these new high-resolution palaeoenvironmental sequences are showing correlations with archaeological events over the past 6000 years in the Near East, new high-resolution sediment and pollen cores from coastal marshes in both Europe and the Americas are now documenting equally significant fluctuations in sea-level rise and sedimentation rates over the last 2000 years as well.

Instead of static marsh habitats at the same water level over the last several millennia, it now appears that many of these sediment-filled drainages were exposed throughout the late prehistoric and early historic periods. The recorded high rates - orders of magnitude above the ‘official’ IPCC estimates of c. 11-21 cm per century - and extreme fluctuations in the rates of marine transgression over the past two millennia, are also strongly suggesting that much more of the late prehistoric landscape of now inundated drainages were exposed than we had previously suspected, when prehistoric settlements began to abound along the coastal drainages of North America and elsewhere.

For example, recent high-resolution core sample fractions by Dorothy Peteet and her team at the Lamont-Doherty Earth Observatory and the NASA/ Goddard Institute for Space Studies, taken at c. 4 cm (20-50 year) intervals from a Hudson River estuary (Piedmont marsh), upriver from New York City, documented extreme fluctuation is sedimentation rates - and by extension as a proxy for pollen-based indices of past disturbance - over the last 1500 years. The earliest and deepest core segment, dating from c. AD 500-1000, showed a rate of 18 cm per century, consistent with the nineteenth-century tide gage records for the region. However, it dropped to 3 cm per century between c. AD 800 and 1250, but jumped to from c. a 30-60 cm between AD 1250 and 1300. It then dropped down to 30 cm per century in the Colonial period, half the rate of the previous century, but nearly two to three times the currently generalized estimate of 11-21 cm per century.

Several similar core sequences from the coasts of Nova Scotia and Maine have used the relative composition of foraminifera (microscopic shell species variously adapted to either fresh or saline habitats) to extend the earliest nineteenth-century tide gage

Figure 3 Historic geospatial composite illustrating the process of 3D palaeo-environmental reconstruction of buried cultural landscapes, in this case the original topography and ground cover seventeenth century Colonial Manhattan. Relative topographic variability is extruded from historic map coverage, the actual elevations and depths from archaeological excavation records, or ‘ground truth’. The historic plant and tree cover is derived from excavated and dated archaeological plant remains and ethnobotanical records. Modern digital air photo coverage is used with computer-based GIS (Geographic Information Systems) software to georeference, or ‘rubber-sheet’, the historic landscape projection to real-world coordinates (By Joel W. Grossman, Ph. D., © Joel W. Grossman, Ph. D. 2008, Published by Elsevier Inc. All rights reserved.)

Records, back a century, to AD 1750. And like the new Hudson River core data, they also recorded historic period rates of 30-50 cm per century in Maine and 60 cm per century from Nova Scotia, with most of increases occurring in the eighteenth century.

The dated column or core samples identified evidence of significant large climate signal, or fluctuation, in the region’s wetness, indicating the sudden onset of drought in the northeast, during what had been called the Medieval Warm Period in both Europe and the Americas, between approximately

AD 800-1350. This c. 500-year warm spell was followed by what climatologists have called the ‘Little Ice Age,’ between fourteenth and nineteenth centuries. Similar palaeoclimatic records of drought at this time from the Chesapeake estuary, and from the Jamestown Colony to the south, suggest that the entire northeastern US region experienced the sudden onset of a 4-5-century-long period of drought, parallel with the shift to warmer conditions.

These shifts are thus beginning to document ‘order of magnitude’ variances in contrast to the ‘straight-line’

Mean regression plots of the commonly ascribed to the multimillennium time span between the post-1850 tide gage data and the c. 3000-year-old chrono-metric dates for basal estuary or marsh sedimentation in the northeast. The new data instead suggests that rates of marine transgression appear to have (1) left significantly more dry land exposed than thought (the higher the rate of sea-level rise, the more land was formally exposed in the historic and late prehistoric periods), and (2) subject to sudden and significant shifts in regional climate trends. If these episodes of sudden climate change in the late prehistoric and historic (Medieval-Colonial) era patterns can be documented for other regions, then current discussions of ‘global warming’ may need to add, or factor in, the variable of ‘widely fluctuating’ Late-Holocene rates if sea-level rise and sudden climate change to the widely held straight-line ‘handle’ of the modern ‘Hockey Stick’ of contemporary climate modeling.

In terms of archaeological associations with the landscape, these episodes of higher late-prehistoric rates of sea-level rise also suggest that what were presumed to be low-lying or submerged landscapes within coastal drainages, may in fact have been significantly higher, dryer, ecologically diverse, and more available for human exploitation and settlement than previously presumed. These findings also strongly suggest that future palaeoclimatic and palaeoenvironmen-tal studies must now come to grips with the need to sample and analyse equivalent high-resolution (decade level) samples in order to ‘see’ comparable short-term climate shifts and changes in sea-level rates in other coastal regions.

Submerged Archaeological Survivals

Recently, some astounding off-shore discoveries in the British North Sea, and off the coasts of Denmark and Sweden, have documented the unexpected survival of well-preserved archaeological sites in now submerged off-shore ocean settings. These discoveries show high levels of stratigraphic integrity, the survival of a broad-range of prehistoric food and ecological data, as well as a wide assortment of diagnostic chipped stone, wood, and bone artifacts.

In 2003, British archaeologists and divers working in the North Sea discovered two prehistoric settlements, one tentatively dated to the Early Mesolithic (8500-10 000BP) and another to the Late Mesolithic (5000-8500 BP). While a prehistoric antler harpoon had been earlier recovered in North Sea fishing net, and a deep-sea core sample yielded a chance find of a Stone Age chipped stone artifact, these submerged sites are the first to be confirmed to have survived the often severe wave action of the North and Baltic seas. Marine archaeologists from the Estonian Maritime Museum also recorded a 1720 Russian or Swedish fort on the bottom of the Bay of Tallinn. The stone and timber fort was found at a depth of 8-11.5 m (24-35 ft), 900 m (2700 ft) off shore.

In 1999, similar cases of submerged archaeological sites, that had clearly survived the forces of tidal action and sea-level rise, were discovered by Harold Lubke along the Baltic coast in the Bay of Wismar. Fifteen sites, belonging to at least two and possibly three prehistoric culture groups were recorded along now inundated shore areas, one group at c. 0-5 m and oldest at —7 to —8 meters (21-24 ft) below mean sea level (msl). These seafloor sites revealed rich assemblages of stone and bone implements, most prismatic blades consistent with late Neolithic types, preserved plant remains, small-boned vertebrates, as well as numerous wooden artifacts including prongs and fishing equipment, an elm wood bow, parts of log boats, bone and antler points, a bone knife, harpoon fragments, antler strikers (to make fires), and an antler pendent and boar tusk knives. The zoo-archaeological remains included deer, boar, and large amounts of minute fish bone, including a predominance of eel and cod. The best preserved of these delicate artifacts were recovered from a band of submerged ‘peat’ reflecting its transition from a dry land site, to a temporary matrix of a tidal marsh before being buried under off shore sands of the rising sea.

An internally consistent series of 32 radiocarbon determinations returned dates between 4100 and 6200 BP, which together with the artifacts distinctions suggested three phases of prehistoric occupation within this 2000-year time span. In 2002 Lubke published an important paper entitled ‘Submarine Stone Age Settlements as Indicators of Sea-Level Changes and the Coastal Evolution of the Wismar Bay Area’. Lubke combined his depth and bathymetric data with the radiocarbon age determinations to recompute earlier widely accepted estimates of sea-level rise that had been based on standardized morphological and analytical procedures from dated sediment cores. The archaeologically anchored calculations showed that the sea had reached the elevation of the prehistoric sites 1000-2000 years earlier than thought, reaching the —7 m mark of the earliest and deepest site around 5100 BC instead of 4000 BC, and the higher —3 meter mark of the more recent site by 4100 BC instead of 2000 BC.

In addition, like air-borne radar, a series of sideband-sonar scans of the bottom captured a broad range of topographic variability and defined specific landscape features that were found in association with the submerged prehistoric sites and could be used, in the future, to target the topographic characteristics of other similar underwater archaeological survivals. These included low beach gradients, original deposits imbedded in and protected by peat, where-the local topography contains indentations, ‘fossil’ estuaries and river valleys, submerged gullies, depressions with former wetland deposits, on the flanks of submerged islands, bays, estuaries, near-shore lagoons, and other forms of partial shelter.

Similarly from the Pacific Coast of North America Fedje and Jasenhans from Parks Canada and the Geological Survey of Canada documented well-preserved shell-fish-rich beaches and stands of submerged tree-stumps from drowned forests at a depth of 150 m (c. 450 ft) in Werner Bay of British Columbia. In addition, the recovery of a barnacle-encrusted chipped stone tool from a depth of 55 m (c. 150 ft) suggests human occupation predating 10 200 BP. As was the case for the Baltic, this unique find suggested rates of marine transgression that were significantly higher than prior estimates derived from traditional geomorphological coring and dating procedures.

Like the European studies, high-resolution marine side-band sonar and remote sensing surveys reveled that the Holocene artifact was found in association with drowned landscape features (palaeobeaches, former stream confluences, lakes, terraces, bogs, and estuaries) that suggested the potential for the discovery of other drowned sites in zones of now buried and submerged drainages and coastal formations. In addition, this evidence suggests that now submerged landscape of the Pacific Coast continental shelf was both exposed and occupied by humans in the Holocene. This new marine archaeological evidence, in conjunction with new palaeoclimatic evidence that the long-adhered - to ‘Ice-free corridor’ route for Early Man migrations into the Americas may not have been either open, or available, for some 2000 years after the earliest evidence of human occupations in south America - at Mont Verde in southern Chile - has added credence to the alternative theory in support of a coastal migration route.

These important new insights and alternate estimates for marine transgression suggest strongly that the discovery of future buried and submerged archeological sites will depend on the availability of high-resolution 3D control of the original preinundation and prelandfill topography (Figures 3 and 4). These deep-sea survivals also suggest that if fragile

Figure 4 A 3D paleo-environmental reconstruction of prehistoric topography and habitat of the formally exposed New Jersey Meadow-lands as it was before being inundated by rising sea levels ca. 2000-3000 BP. The digital terrain model, or 3D surface-mesh model, is derived from georeferenced 1865 US Army survey of bathymetry and mud-depth readings in the Hackensack Basin. The environmentally accurate 3D prehistoric tree and plant cover is identified from radiocarbon-dated pollen cores. (By Joel W Grossman Ph. D. © Joel W Grossman 2008. Published by Elsevier Inc. All rights reserved.)

Archaeological remains (with preserved organic artifacts and intact deposits) can withstand the rigors of coastal tides and wave action, they would have survived in Holocene tidal marshes as well____and that

We have to be prepared to look for them.