Modeling Upland/Lowland Adaptations & Oregon Coastal Prehistory (1996)
By Leland Gilsen
Slow Tectonic Change
Fast Techtonic Change
Sea Level Change
Littoral or Maritime?
In order to understand human use of the physical, biotic and cultural landscapes, it is necessary to explore the limits to the samples available. A key problem that arises in the archaeological record on the coast relates to changing sea levels, the impact of coastal erosion, the relative percentage of habitat types along changing sea level margins, seismic uplift and subsidence (earthquakes and tsunamis) and the growth of small estuarine systems.
Slow Tectonic Change
There is evidence for both slow and fast tectonic changes. The overall slow pattern "indicates that the smallest uplift has occurred along the north-central coast between Newport and Tillamook, with progressively higher uplift further south and along the very northernmost portion of the coast toward Astoria and the Columbia River" (Komar, 1992:4-5). Komar concluded that the central region is being submerged by the sea at a rate of about 1-2 mm/year (4-8 inches per century). The rise is also very slow, on the order of 0.7 MM/year. This change is significant today, but would have been insignificant during the sea level increases from Pleistocene ice-melt prior to the stabilization roughly around 5,000 years ago.
Also as Komar has noted, the Oregon coast is one of the most dynamic in the world. Waves and current reshape the shoreline in a complex way relating to many factors. Oregon has a unique ocean wave monitoring system using micro-seismometers in place since 1971. This system has the longest continuous record of waves on the west coast. Storm waves during December of 1972 often averaged 7 meters (23 feet). Given that the largest wave is usually 1.8 times higher than the average, waves on the order of 12.6 meters (>41 feet) probably hit the coast that year. A wave monitoring program off the coast in the 60's measured waves at 29 meters (95 feet) which is close to the world record of 112 feet. "All of the measurements on the Oregon coast confirm that it has one of the highest wave-energy climates in the world" (Komar, 1992:7). Typical spring tides run +9 feet above mean lower low water (the average of lowest daily tides). Perigean tides reach +10 feet. Storm driven perigean tides have reached +10.2 feet. This strongly suggests that any submerged sites would have been destroyed by wave action during sea level changes at the end of the Pleistocene.
The average erosional rates for Lincoln County were calculated by Smith (1978) at 6 meters (20 feet) from 1939-1973 (34 years)... 18 cm/year. Komar & Shih (1993:756) note that her data include the erosion of landslides, so this data is skewed upwards. Stembridge (1975:22) calculates the following average erosional rate ranges:
|Igneous||3-9 cm year||Metamorphic||3 cm year||Sedimentary||15-60 cm year||Terrace deposits||60 cm-6 m year||Sand deposits||3-30 m year|
The implications for preservation of archaeological deposits in an open environment subject to wave action are dismal. Sedimentation is rapid below the active surf zone, but erosion is just as fast during sea level drops and rises.
Fast Tectonic Change
There is, however, evidence for tectonic events related to subduction earthquakes. Based on layers found at Netarts Bay in Oregon and Willapa Bay in Washington, there have been six major quakes in the last 4,000 years at intervals of 300-1000 years. The last event was 300 years ago. Some areas dropped 1-2 meters during the quake. This creates the possibility that coastal sites in relatively protected and low energy environments could have been dropped below the most active wave zone and perhaps have been buried under tsunami debris as well. Such burial conditions were probably very rare. if they even occurred. There is great difficulty in generating general rates of erosion because of these events. At Bandon, for example, there is evidence for six major quakes in the last 4,000 years. Each of these quakes dropped Bandon 1-2 meters. There is evidence for strong coastal erosion following these events. Then, because the land uplifts 2.2 MM/year, erosion slows, stops, and land accretes, followed by another 300-500 year earthquake event that drops the land and starts the cycle over again. The Whisky Run terrace near Bandon has been dated to about 83,000 B.P. Based on that date, it should be 160-170 meters high (at 2.2 mm/year) but it is only at an elevation 30 meters... strongly suggesting that the earthquake process has removed 130-140 meters of upthrust over this time period. An average of a one meter drop every 600 years would account for this difference.
Given the right conditions, preservation of archaeological deposits in estuarine environments dropped below the more limited estuary wave zone by earthquake subsidence is possible. Multiple events could drop the site below more active wave zones as the estuary was invaded by the rising sea levels. Such sites are probably very rare, and will be difficult to find.
Sea Level Change
The data indicates that during the sea level rise at the end of the Pleistocene, the coastal plains of Oregon were submerged much faster than current average slow tectonic processes. The geology supports this process. There is the possibility for sudden earthquake burial events that could have preserved sites in local low energy areas. But the overall high energy systems suggests such locations would not have survived long, if at all. Thus, most of the coastal archaeological record prior to 5,000 years ago (i.e.- stabilization) have probably been destroyed. The analysis of radio-carbon dates from SHPO supported coastal studies supports this possibility. But what was the coast like through time?
In general, sea level was between 120-130 meters below the current mean sea level during the last period of the Pleistocene. "The Holocene Transgression consists of two periods of different rates of sea level rise. A relatively rapid rise in sea level of 10 m/1,000 years occurred from 18,000 to 10,000-7,500 B.P., followed by a slower rate of advancement averaging 1.5 m/1,000 years. Eustatic sea level has remained relatively stable for the past 3,000 years" (Anderson, Marcus & Wilson 1990: 10). "The shoreline position was located near the modern shelf break, and consequently very little continental shelf existed" (Anderson, Marcus & Wilson 1990: 40).
The shelf off Oregon averages 40 km in width with a maximum near Heceta Head at about 70 km. A large submarine canyon (Astoria) is located about 16 km off the Columbia River. There are three large submarine banks off the coast from north to south: Nehalem, Stonewall and Heceta-Perpetua. These have a maximum relief of about 75 meters.
A Minerals Management Service report suggests that rocky habitat is a high-stand phenomenon: "The exposed early Holocene coastal plain (the continental shelf of the previous high stand) was likely characterized by a smooth, gently sloping surface dissected by meandering, low-gradient streams. Shoreface retreat during previous marine transgressions apparently eroded and redistributed a considerable amount of soil from the subaerial surface that once covered the uplands. The streams would have flowed into large coastal margin lakes and estuarine systems" (Anderson, Marcus & Wilson 1990: 40).
"From as early as approximately 7,500-5,000 B.P., coastal erosion began to shape the modern shoreline. Actually, there is abundant evidence suggesting much of the present coastline morphology was inherited from the previous high stand. For example, narrow terrace deposits, submerged wave-cut platforms, and notched headlands all suggest the modern level position has re-occupied the previous high-stand position of many coastal sites" (Anderson, Marcus & Wilson 1990: 42).
The study of beach sand and its origin through trace element studies indicates that "About 5,000 to 7,000 years ago, the rate of rise in sea level decreased as the water approached its present level. Just about at that same time, the beaches of Oregon came under the influence of headlands that segmented the formerly continuous shoreline. At some stage several thousand years ago, the headlands extended into sufficiently deep water to hinder further along-coast transport of the beach sands. This is shown by a study of the mineralogy of sand found on the present-day beaches (Clemens and Komar, 1988 a, b). The pattern of along-coast mixing of sand from the various sources, established during lowered sea levels, is still partly preserved within the series of pocket beaches now separated by headlands" (Komar, 1992:10).
Lee Lyman (1991:7) noted that at the present time "Of the approximately 480 km of coastline, 40% consists of rocky sea cliffs and headlands." Other data suggests that the current pocket beaches separated by rocky headlands formed around 7,000 to 5,000 years ago. The rocky headland habitats and patterns of off-shore rocky habitats fall into this time period. At some time prior to this date, the coastal plain may have been mostly sand beach separated by down-cutting riverine valleys. This kind of habitat would not have been optimum for shell-fishing or fishing. It would have been good plant gathering and land mammal hunting habitat. The large flat coastal plain was probably reflective of the remnant coastal plains of today... "... during the maximum of glaciers, the sea level was considerably lower, the shoreline was then on what is now the continental shelf, many miles to the west of its present position, and the beaches were backed by a smooth coastal plain." (Komar 1992:10).
George Priest (Department of Geology and Mineral Industries - personal communication) indicates that he obtained a radiocarbon date of 9500 years ago on a peaty sample from a drill hole into the base of a Holocene channel at Siletz Bay (see DOGAMI Open-File Report O-95-5). This peaty sample resembles current salt marsh deposits in the area but is about 150 feet below sea level. This indicates that sea level had reached the coast at the bottoms of Pleistocene valleys by this time. How far seaward headlands reached and how much headland habitat existed before or after is not known. He goes on to caution that in his opinion although the amount of lithification decreases to the west, the relative amount of lithification may not drop significantly enough to make a difference until it was close to the lowest sea levels in the Pleistocene. Thus, the amount of rocky habitat may have been only slightly less through much of the Holocene record.
If the amount of rocky habitat increases to the east, the stabilization of rocky habitat may have caused a shellfish populations increase along the Oregon coast. Before the establishment of large scale rocky headlands and sea cliffs, there may not have been sufficient habitat to support a cultural economy that emphasized littoral resource exploitation. The key is the relative amount of lithification exposed at various sea levels.
With the submergence of the down-cutting riverine valleys came the emergence of estuarine zones as the outflow energy dropped and sand began to be deposited by the rivers. "If the size of estuaries is causally related to their biological productivity, then both the ecological evolution and size of an estuary may exert selective pressures on human groups living on the margins of estuaries. Certainly large estuaries may have more sites located on their terrestrial margins, but that may be simply a function of amount of habitable area. But will site density per unit of habitable area, site size, or permanence of occupation of sites covary with estuary size in some manner?" (Lyman 1991:7)
It is possible, indeed probable, that the basic patterns in ecology were quite different prior to 7,000 - 5,000 years ago for the Oregon coast. It is therefore not surprising that the archaeological record shows differences in the lowest deposits found in coastal archaeological sites. The changing location of, changing relative size, and evolutionary development of rocky habitat, sea cliffs, estuaries and dune lakes needs to be studied throughout the sea level changes over the last 15,000 years.
The concept of pocket beaches has some interesting implications for understanding Oregon prehistory after 7,000 - 5,000 years ago. Each pocket beach is described as a "littoral cell". Some of these cells also contain estuary openings for the larger coastal river systems. Pocket beaches are defined from headland to headland. The two primary productive systems along the coast are the estuaries and the rocky headland habitats. There is the possibility that the archaeological record could reflect patterns found in the pocket beach model. Perhaps the archaeological record should be examined to see if a pattern of pocket beaches (headland-to-headland) centered on the larger estuaries might reflect centers of cultural continuity and shared patterns in material culture.
Each cell could be a model for an economic exploitation zone based between headlands and using the estuarine resources between. The cells without estuaries might be looked at as possible borders between larger economic and political groupings.
Littoral or Maritime?
The issue found in Oregon models for coastal adaptation of littoral (tidal) vs maritime (open ocean) adaptations is fuzzy just as the transition between these zones is fuzzy. People tend to exploit resources that are relatively predictable, relatively sessile and relatively abundant. They can, however, choose to exploit less predictable, relatively active and relatively rare resources as well or any admixture of these three variables.
Human groups on the coast currently have the opportunity to exploit riverine, estuarine, littoral and neritic resources. Their choices are quite diverse. Most of these resources can be exploited with relatively simple technologies combined with the information obtained from observation and accumulated through cultural transmission. There are no changes in technology (as to be a change in kind) needed to exploit the common elements of the zones. There is enough fuzziness in technology and in the resource as to make distinctions between littoral or maritime adaptive models of questionable value at this time. Because of sampling problems both in sample sizes from excavated sites, and the sample that are simply missing from the archaeological record, conclusions regarding the relative strategies involved are speculative at best.
As more research is done on the coast, dates for exploitation of shellfish and fish and pennipeds and other food resources are being pushed further and further back. This supports the concept that people were exploiting anything within the range of their technology, and that the range of technology allowed exploitation of essentially all of the zones noted above except for the extremes at both ends. Certainly the modern riverine, estuarine, littoral and neritic zones could have been exploited with the technologies at hand. The mix of strategies in relation to predictability, abundance and active/sessile could vary from place to place and season to season and culture to culture.
Just how much rocky habitat was available at lower sea levels is unknown ... but the evidence suggests that there were not "pocket" beaches prior to 7,000 - 5,000 years ago. Thus, rocky habitat probably became a major factor for resource exploitation at about that time. That sea resources were exploited prior to these dates is clear in the archaeological record. Because so much of the early coastal record was submerged (and probably destroyed), we may never know what the early patterns were except by extrapolation from other areas of North American that may have been pushed up faster than sea levels rose.
The archaeological evidence does suggest that coastal plain plant and animal resources probably dominated resource exploitation strategies until rocky habitats and estuaries formed with their rich habitats. Anadromous fisheries may have been important throughout human occupation, but may be hard to find in the archaeological record.
There is a simple (and probably biased) way to look at these issues. If the distribution of recorded archaeological sites is plotted in relationship to local habitats (counts per township), a pattern does seem to emerge. Sandy beaches simply contain few recorded sites. Rocky headlands and estuaries dominate the currently recorded site distributions. This is exactly where shellfish and fish habitats reach their greatest densities and diversities. It also corresponds to sea mammal breeding and rookery areas as well as bird nesting habitat.
Jones (1991) demonstrated that marine foods are rich and not "second-rate" resources. he also emphasized the importance of estuaries in coast adaptation. Hildebrandt and Levelett (In press) emphasize the estuarine aspect as well, but for the large systems, where early sequences are more likely. While huge estuaries may have longer life spans, many of the smaller ones are relatively recent structures.
Peterson, Scheidegger and Schrader (1984) studied the Holocene development of a small estuary along the Oregon coast. They did C14 cores and seismic reflections to study the development of Alsea Bay, a Holocene estuary that started as a river valley that was drowned by rising ocean levels.
"The first stage of deposition, 10 X 10 ± 500-7.5 X 103 yrs B.P., is characterized by dominantly fresh water, cross-bedded sands and a moderate sedimentation rate between 0.4-0.7 cm yr-1. The second stage of deposition, 7.5-5 X 103 yrs B.P., is associated with brackish water, laminated sands and silts and a high sedimentation rate, 1.1 cm yr-1. The third stage of deposition, 5 X 103-0 yrs B.P., is characterized by strong tidal flow, cross-bedded sands and a low sedimentation rate, 0.21 cm yr-1. The three stages of deposition reflect periods of transition from: (1) a shallow fluvial environment; to (2) a deep-water estuarine environment; to (3) a shallow-water estuarine environment" (Peterson, Scheidegger and Schrader (1984: 80).
So models based on estuaries must take into consideration the age and development of the system. Just because there is a fully developed system in place today, does not mean the past was the same.
There is increasing evidence for huge dunefields along the Oregon coast instead of a coastal plain covered with forests and grasslands. If this is true, then the coastal habitat was even less attractive prior to 7000-5000 years ago. The large areas of open exposed beach with its scarce resources backed by miles of massive dune fields would be a resource poor environment.
I feel that the relative percentage of rocky habitat did not rise above some critical value along the Oregon coast that encouraged an adaptive strategy for marine resource exploitation until about 7000-5000 years ago. It is my theory that the habitat was simply missing or of such a low quantity and quality, that id did not favor human use and adaptation. I believe archaeologists must not only look for sites, but for data about the changing landforms in relationship to ocean levels if good models for coastal adaption are to be developed.