Abstract

Using sea levels to find datable evidence of human dispersion along a route from Asia to the Americas. This paper examines some of the ideas about how people can disperse from eastern Siberia, northern Korea, China, and Japan to Alaska and the Pacific Northwest Coast of the Americas. It argues for a milder maritime climate along the Pacific Northwest coast regardless of ice-age conditions or not. A concern is for a progression of age of occupation among sites left by dispersion and which should be found along any path people actually took. Focusing on practical living requirements of the continental steppe of ice-age Beringia contrasted to those needs along it's southern coast, an argument is made that the coast can support dispersal during an ice-age, leaving a "standard" winter dwelling built from stone and set into earth. From ongoing research the initial idea of finding these has changed to implicitly dating sites by sea level without having to recover artifacts at depths of 30 to 60 fathoms. If successful, this would encourage more extensive underwater research. Finding the sites is the difficulty, but the recent finding of a tree stump near Queen Charlotte channel at over 120 meters implies important facts about local climate during an ice age along this coast. The odds of finding anything are not good, to quote Chris Goldfinger from Oregon State University's Marine Geology Active Tectonics Group: "..because the shoreline went through the surf zone, and anything that wasn't nailed down, i.e. bedrock, has been severely battered by about 500 years in or close to the surf zone."

UPDATE February 12, 2001 Click here for the latest arguments for underwater research and possible site locations.

Sea Level

From general texts on the ice ages, the common references to sea level changes use several related factors, with a study by Shakleton (1987) as a basic data set, for the most recent rise in sea level data from NOAA was used for the graph at the end of this paper. From the graphs of the past 150,000 years, there is an obvious relationship between major reversal stands near the 30 fathom contour line from about 42,000ybp until 10,000ybp. These were changes of direction where the level either rose to that height and subsided or the reverse. This can explain the obvious shoreline geomorphology at that depth. Consider this section of coastline along the west coast of Vancouver Island near present day Swiftsure Bank. (1 fathom=6 feet=1.8278m)

Adapted from Aharon and Chappell, 1986, Huon Peninsula, Papua New Guinea curves
in "Quaternary Environments", 1993 by Williams, et.al. from Arnold, London.

In other words, in this area, except for Beringia, there wasn't much change in width to the actual coastline, with the largest real changes from the inland waterway drying out and having a lot more ice in them, and certain areas surrounding Vancouver Island and the Queen Charlotte's. Offshore it's very similar to that era today. The differences will relate to cold continental air circulation at that time and that affect upon plants, and the creation of sea ice during the winter. To follow dispersion along the coast at any time means living with the current coastline. From Cape Flattery north, on the ocean side, things are not all that different than at any time in the past, very similar ecosystem. The drier air which occured during the last glacial maximum and created steppe leaves pollen sampling which would indeed signify such conditions. While many studies of inland changes have been made, much good work is available from the University of British Columbia, Vancouver, Canada. One would expect whatever changes in the interior to be more radical than on the shoreline next to the ocean.

For archeology this is important information about what is possible for humans to do to get from Alaska to Vancouver Island by living their lives and never exploring yet expanding southward. No matter that you can't really go south without crossing huge rivers or a lot of giant valley glaciers on determined treks with adequate food and clothing for a few days at a time. These obstacles overcome by chance, luck and population pressure, the coastline still does not encourage penetrating the interior from the Kodiak Islands to Vancouver Island. For these hundreds of miles the best food sources are right at the shoreline and it's immediate forest. Not much incentive to go inland during an ice-age or not. It's only a few miles or maybe ten to the ice of the Coast Range for hundreds of miles during an ice age. This would seem to imply more population density on the coast, there should be a good density of ancient campsites, but under 30 or so fathoms these are for all practical purposes hidden.

There are sites dated in eastern asia in the 55,000 to 35,000 year range which add to the wonder why there isn't a progression of younger dates moving eastward toward Alaska. It would seem by the odds that such evidence does exist, if not on land then under about 30 fathoms of ocean, all that's needed is to find it. If these people "migrated" over a short period of time, then it was weather holding them back. If they were surviving that climate in Eastern Siberia, why wasn't that lifestyle good enough to allow expansion eastward? Was the weather more severe than assumed? The coastal environment wasn't a party, with sea ice that barely broke up seasonally, but it wasn't worse than inland and most would consider it better somewhat in extremes. The wildlife was there. All this combines to suggest that living along the coast provided a livable local climate for humans to deal with during an ice age along with the food resources and therefore that's where the people dispersed eastward from, not farther north within Beringia proper which must have been more severe winters and would force people into locales which were refugia from the extremes of expecially northeastern Siberia. While people did roam around Beringia, and did disperse to the east into Alaska, they did so by wintering in warmer locations and avoided the extremes. My point here is that the coastal climate was no worse than these warmer places of survival in the far north. It was equally survivable, equally was populated and equally had opportunity for people to disperse along the coast eastward to Alaska. In support of this argument is this chart of loess deposits as they relate to the oxygen-18 curves. Note especially that for the period beginning near 14,000 years ago there is a wide band of loess deposits, as one would expect from a warming trend and reversal of some large global air circulation patterns. Note also that similar patterns occur at around 25,000 years ago, 33,000 years ago, 41,000 years ago and the earliest of concern at this time, another very long lasting series near 56,000-49,000 years ago. These represent about 14,000 years worth of warm climate, similar to the post-glacial, yet more appropriately called interglacial periods. Comparing this to today's sea level, it can be seen that sea level was lower, yet near as high as today's for most of these periods. This implies that many very old sites can be relatively shallow, under only 10-20 fathoms. Without excavation, these are not likely to be identified and are not as useful to "proving" a date for a site by depth.

Copied from the graph of An Zhinsheng, Liu Tungsheng 1987 data in Williams' et.al., Quaternary Environments, page 154.

Personally, from experience as a high altitude climber, consider that even -70° during a winter is a challenge to humans which requires such exacting technology, skills and knowledge to survive that most modern humans couldn't do it given the equipment and trained in the techniques without a rifle and manufactured clothing on a year round basis. It's sounds like an intellectual argument to do so, winters were extremely severe, every year, during an ice age. If you must create shelter and supply enough food, a spatial extent is required by each person supported by the tundra. Not fun. Ask an Inuit to go back to hunting seals with harpoons and see what they say about giving up their rifle. It's a quatum leap backward in technology and a quantum leap down in your odds for survival if a long winter combined with poor game that year. People who hunt seal, walrus, and bear live with the sea ice much of the year, this was the coastal way of life. The inland cultures with big game would migrate to refugia of micro-climates within the larger scope of the northern forest-tundra. Coastal people would only keep dispersing if they survived conditions which were similar to those being survived elsewhere at the same period. There isn't any barrier between Chukchi Peninsula and the Aleutian Range.

While the weather was harsh, it was near sea ice and had a summer, and there was open water in the brief summer. I expect reality to have been a continuous dispersion of people eastward at a slower rate than without an ice-age. People would have been east of the Aleutian Range by 38,000-22,000 years ago before the onset of the last glacial. That's 16,000 years to go 2,000 miles from central Kamchatka to Kodiak Island, 8-yrs/mile dispersal rate. To settle Patagonia from Alaska is roughly 10,000 miles, if people left near 14,000ybp and arrived in 9,000ybp that's 5,000yrs/10,000miles = 1/2-yr/mile, 16 times faster. If they arrived c.28,000ybp in Kodiak then to settle Patagonia would have taken 19,000 years, almost two years a mile. If we take the 14,000ybp start date as an assumed speed of dispersal, a half year per mile, and apply it to the period around 33,000 years ago, those people had 2,000 miles to go, it would only take 1,000 years to get across in a warm period of almost 2,000 years. Therefore, if one accepts the speed of dispersal for an post ice-age peopling of the Americas, it guarantees that people would have dispersed to the Americas roughly 10,000 years prior to that, if not before.

Ice-age atmospheric circulations created a scrub desert on the southern side of Beringia and a rich grassland in the north where mammoth were found. If you are coastal, immediately inland is a desert where there usually is plenty of small game, from pikas to hares, so, one has a reliable way to take game without having to take big game, the toolkit can be small and minimal. Being desert, it gets less snow and more blue sky. If you are protected from the wind, there are plenty of shrubs to burn for fuel to roast the rabbit; the desert is altogether able to supply winter food and shelter in many ways. I find this lifesytle repeated on the eastern side of the Cascades and Sierras into the deserts of Mexico. Like most deserts, there are a lot of plants and animals living there, just not the variety as in the maritime climate which came to the same area of Beringia much later.

The population pressure along the coast is very high once habituated to the coastal resources, and, dispersion would stay along the coast which is of course directional and limited in extent inland. Coastal is not necessarily littoral, but does include it. The change to continental resources is to a desert food supply from the coastal environment, and to colder air (even with less oxygen as well in places), does this sound encouraging? I think it shows that most people would stay on the coast during an ice-age if that's where they grew up. If this is true, their numbers as they multiplied would expand more along the coast than inland, almost to the exclusion of expansion inland.

The idea here is that this well established ancient shoreline along the 30 fathom line and below is a good place to look for a progression of dates for a gradual expansion down the coast. While the actual shoreline varies, at the base of the headlands would have been a good place to camp. The terraces below the headlands offer very little protection from prevailing storm winds. I note the natives of this area built very sturdy and weatherproof houses if on the water. You don't hang out for a suntan on the beach all year; the climate requires good clothing and a lot of calories. This all points to an environment actually very similar to today's, except the valley glaciers and their rivers were more robust and the forest line was at a lower latitude, yet it was still basically a maritime province. Pack ice during the ice age created a lot of areas within this coastal route where the snow-bound land was surrounded by sea and valley ice for much of the year. The land had permafrost, yet the snow melted off every summer and plants were available, a spruce forest existed. Coastal people could live here more comfortably than within the continent during an ice age, and they lived farther north, so why didn't they live farther south? Was there a real separation somehow which doesn't seem to present itself anywhere else?

On the western side of Beringia in southern Primorye are groupings of sites c. 54,000-40,000ybp. At the end of this is an interglacial, c. 35,000ybp. These sites are quite far inland. Who was on the coast at this time? If people were on the coast, they would continue to disperse. It's absurd to think people didn't live on the coast as well as inland at any time in this area, which is a peninsula. That stages people at the beginning of a warming period on the western side of the north Pacific with a fairly low sea level and rising. Apparently it took 20,000 years to get to Alaska counting one interglacial and one glacial period to the beginning of the next interglacial; if people didn't get there until 14,000 years ago. Yet, we are to believe that it only took a few thousand years to populate the Americas after that. Something is wrong about that argument. These same people had an equal opportunity in an interglacial to disperse along the coast for something like 5,000 years before more extreme ice age conditions came again 22,000 years ago.

Adaptation from West's "The Archeology of Beringia" from original by
Hopkins in "Quarternary Research", 1973, 3:4:520-540.

To prove dates without having to recover artifacts, with seismic sounding to find artifacts by patterns along depth can supply evidence, isostatics and tectonics considered, a reasonable estimate then can be made. With enough of this type of find available as seismic evidence of where to look, suction dredging may supply artifacts. If for example a stone circle is found in sediments along the 30 fathom line, it's reasonable to bracket an error factor around 10,000 years as the minimum age and that place would be a good one to dredge. Here is a sample stone structure found, the main components of it are a hearth, a lowered floor separating the hearth from a raised seating/sleeping area.

Inspired by the photography of Robert Park in "The Archeology of Childhood:
Toys Tell Tales of Growiing Up in the Artic", Discovering Archeology, March-April, 1999

Floor plan of an idealized winter home for living in a coastal artic environment.
This is what the seismic reflection waves see.

An added aspect of this is complex, the land was dry and eroding as a land surface for some 20,000 years before being submerged since the last rise. The submersion, if where sedimentation is going on, will have deposited a thick layer of sediments over any artifacts on the previous land surface. These two have different seismic properties and the interface is typically visible in the seismic data, so the problem is creating an plot using this data which can be viewed and analysed for these non-geologic patterns, circles and curves in linear rockscapes caused by living spaces built into near shore areas. Focus is then on this type of reflectance patterns, it would fit the archtype above of hearth, floor and seating in stonework surrounded by more or less natural soils all buried in a more or less homogeneous sediment rain since being submerged. That's the methodology.

My hope is to find many of these in support of people having used the coast for dispersal into the Americas before the last and coldest stage, prior to about 22,000 years ago; as opposed to 14,000 years ago for earliest dispersal. The ends of this particular home, an entryway and the other a wall, were on the same side as the hearth. The sitting and sleeping area was walled to the outside level. This is the type of dwelling being looked for in sediments using seismic data. The classic semi-circular wall formed by the edge of the seats to the floor is a target, being 2-3 meters long by 1-2 meters wide, a high reflectance layer of arranged stones with a 0.3-0.6 meter vertical difference to another highly reflective plane of flat stones, with another step up to a previous ground level having fewer arranged stones than the dwelling so reflectence should be different than the living area. These should show up as brighter "ears", concentric crescents, if plotted in 2-D with plane surfaces almost horizontal and surrounded by a softer surface. They also seem to be big enough features to resolve.

In relation to dispersion along the coast, then, the odds are best to find evidence of ancient campsites near the headland features to be found near the 30 fathom line (57 meters) because they would be re-used, or continuously used until being submerged 10,000 years ago. To follow up on this I have examined some online sources and coastal charts and find promising areas along the west coast of North America where other coastal features are obvious at this same range of sea level similar to the map above. Essentially, the sea was in this range of height, and changing direction at that level over eight times in the past 125,000 years (I count ten but these are graphs and I don't have the source data); hence a defined coast was eroded into many areas at this range of heights. While I expect this to correlate to global changes, the strategy is to search locations which look like they would have been good shelter and food supply areas along this depth or deeper.

Next step is acquiring seismic data which can be viewed and analyzed. This is supposed to discover camp circles for fires or from shelter poles or windbreak walls in addition to looking for winter homes. If these can be found in available data, hopefully that would be enough evidence to get something going on the idea to allow a successful grant to be made to hire a boat that would take specific suction dredge voyages to search for artifacts to establish dating from the Kodiaks to Puget Sound. If the coastal dispersion idea is true, there must be evidence of it in the narrow coastal zone from a current depth of 60 fathoms up. Seismic reflection can penetrate soft bottom sediments to a fathom or so, this would allow better odds at finding then dating winter dwellings. If sites can be found at depths relating to minimum ages greater than 17,000 years along this coast, most archeologists might be comfortable with the idea that this area was the conduit for a majority of the people who ended up in the America's before other avenues were open. It would essentially prove the Asia-America connection by focusing on dating the end of the route where it's narrow and long. This time period has a correlated sea level depth to search along, however, not that seaches haven't been made, none have been persistant and comprehensive enough so that the basic question of who peopled the Americas and how is still a topic of debate at this time.

This paper proposes methods and arguments to search for primary evidence with remote sensing means to keep costs low while being able to distinguish, rather definitively, dates of occupation by depth in consideration of isostatic events as well as sea level in a way which can be convincing to establish the fact of dispersion along the Pacific Northwest coast of North America. From existing evidence, this could have occurred in conjunction with interglacial warmings during the Malaya Kheta through the Piovsko-Novoselovsky in eastern Asia. The latter warming lasting 4,000-5,000 years, certainly long enough to traverse the distance eastward to Alaska along a shoreline which is below the 35 fathom contour. This leaves most of the route to Alaska dry from Primorye below the submerged headlands. This is the focus of attention, to find and note the depth of as many winter homes as possible by examining seismic datasets. Maximum lowest ice-age sea levels generally happened about 18,500-17,500ybp. Significantly, sea level drops below the lowest it had reached in the previous 140,000 years around 24,000-26,000 years ago. This is the first chance for humans to touch the land exposed, a depth near 45-47 fathoms. It remained dry until about 16,000-17,000 years ago, taking 8,500 years to turn around. The red line in the graph below shows some aspects of the changes. The step-like nature and thickness of the line is from it being composed of data points from sea level stations around the world, thus the line is the range of values at that time. They move together, so the implication is that the sea level change is global, there is a time delay however, and this is indicated by the curves for each station changing slightly later. The steep curve from right to left indicates sea level changing slowly to more quickly.

This graph is from the dataset developed by: Tushingham, A.M. and W.R.Peltier, 1993, Relative Sea Level Database IGBP PAGES/World Data Center-A for Paleoclimatology Data Contribution Series # 93-016 NOAA/NGDC Paleoclimatology Program, Boulder CO, USA

This reveals an almost perfect parabola. While sea levels do change fast, when they change direction it's a parabolic curve. This increases the dry time between submersion near maximum. This feature means the odds of people establishing sites at this level is much better. My research of sonar and seismic reflectivity has to this time (sept/16/99) revealed a dearth of data close enough inshore to be of much use for the area from the Straits of Juan de Fuca to Kodiak Island, the Aleutian chain has the best sonar coverage but isn't as likely to have been on the dispersal route during low sea levels, as that would be farther north on dry coastal land now submerged (with ice of course).This would receive people from the direction of Chukchi Peninsula. If so, the locale then requires estimates of sedimentation rates and other data, time for the larger research vessel.

The data available does look possible to resolve the return signal of a winter home in the conglomerate "noise" as a background. It appears that I can find some data with return signal strength as well as timing. Since the timing and depth are actually secondary, building a 3-D grid with the power data seems to be the best method (again at this time in the project development). With the 3-D grid, winter homes should appear as columns of peaks separated by a highly textured surface from the random scatter of the wave-sorted stones and sand under some sediments since immersion. Looks like I need to buy a specific computer for this, my dev machine doesn't have the fancy graphics card and monitor, and it's only a 200MHz...not the 3-D whiz. Schedule for that is my tax return in March. The in-between is being spent with GLORIA and snippets of data to develop some discrimination filters for this type of thing, and to find other sources of sonar and seismic data with digital interpretation of what's usually analog. I'm also thinking of trying some analog raw signal tapes in the near future as it may be the only way to discriminate the power of the return at a discrete time. The GEODAS data seemed mainly concerned with strata and absolute depth.

Without a way to discover convincing evidence cheaply, research of this type would seem too expensive to attract consideration. Hopefully the idea of using smaller offshore boats with standard in-water seismic and digital recording equipment to discover sites by depth will carry a persuasion based upon the conflict of wanting a definite answer to the cost of getting it. With focused searching which can uncover sites no younger than 10,000 years ago, and establishing if older sites exist, once these are discovered the theory passes onto the next stage of dispersion, and the complex reality of a world 14,000 years ago when sea faring cultures could have landed in many parts of the world on longer open ocean voyages, offshore yet coastal, not always in sight of land. This would add a possibility less likely than land dispersion to the Pacific Northwest, but possible, and if so would have supplied a means to quickly settle a very long west coast to the tip of South America in areas quite difficult to people overland by dispersion theory.

Seismic data exploration of South America can establish this as well. Since seismic data can't tell between the site being put there on the lowering or rising sea, artifact retrieval and other means will have to define this. The seismic data can well establish a pattern of travel by studying important areas from Alaska to Patagonia. This body of knowledge is the intent of this paper, which is an outline of why and how to do it in a manner which can be successful in stages without large capital expenses in the reconnaisance of important finds to establish a definitive answer to dispersion from Asia to the Americas.

Send comments and questions to the author via email.

Copyright T.Mallard ©2005 USA
Last Modified: January 7, 2005

Related References (available at amazon.com):
Dawson, Alastair G., "Ice Age Earth: Late Quatenary Geology", Routledge, New York, 1992.
West, Frederick H., Editor, "American Beginnings: The Prehistory and Palæoecology of Beringia", Univ. of Chicago Press, 1996.
Wenke, Robert J., "Patterns in Preheistory: Humankind's first three million years", 1990, Oxford University Press, 3rd edition.
M.A.J. Williams, D.L. Dunkerley, P. Dedeckker, A.P. Kershaw, T.J. Stokes, "Quaternary Environmnets", 1993, Edward Arnold publisher, New York
J.J. Lowe, Mike Walker (Contributor), "Reconstructing Quternary Environments", 1997, Addison Wesley Publishing.