PART ONE – ARCHAEOLOGICAL DISCOVERY IN ORGANIC TERRAIN
Association of Archaeological Sites with Organic Terrain
Although reported artifacts and archaeological deposits from organic terrain are relatively rare, the association of archaeological sites with organic terrain is more frequently recorded in regional site databases, but one suspects that proximity is probably under-reported, particularly in regions where blanket peat, marshland, sloughs, or muskeg are ubiquitous. This apparent association – like proximity of habitation sites to potable water – is largely a tacit assumption and largely untested.
An area of muskeg where clear association with peat has been documented is the ‘hinterland’ 10 km east of Athabaska River near Fort McKay, Alberta, where an Alsands impact assessment project (Conaty 1980) determined that archaeological sites were readily detected and relatively abundant in an area of muskeg that previous studies had considered to have low potential. These interior sites illustrate previously unrecognized material culture related to a boreal forest subsistence pattern (LeBlanc and Ives 1986). The surficial deposits there are glaciofluvial outwash and gravel and landforms include (abandoned) braided stream beds, discontinuous terraces and bars. Although it appears flat at first, there is as much as 5 m relief. “In postglacial time, lows on this depositional surface have filled with peat deposits or bogs, so that roughly 90% of this part of the lease consists of muskeg or other organic terrain. This variability in edaphic conditions, coupled with the successional features of boreal forest vegetation, has created a mosaic of plant communities”… and… “All known sites occur on knolls and ridges, while Conaty’s (1980) sampling programme showed convincingly that archaeological sites are rare or absent in muskeg in this region.” (LeBlanc and Ives 1986: 60).
A similar pattern of archaeological site occurrence on prominent (but low relief) land forms in close proximity to wet lands and organic terrain was noted during an archaeological resource inventory of Elk Island National Park, in the Beaver Hills near Edmonton, Alberta. Here, the terrain is characterized by an abundance of small kettle lakes, in various stages of hydrosere vegetative succession, interspersed by a similar area hummocky moraine and kames (Swayze 1990). Typically, these sites were represented by a few utilized flakes, a bi-polar core, or a fragment of a bifacial or unifacial chipped stone tool, found on a convenient knoll or ridge overlooking a nearby kettle pond, often with a sunny aspect.
However, in terrain like the Athabaska hinterlands and the Beaver Hills (and in the peat-blanketed areas of eastern Canada and the central arctic) where peat is ubiquitous, a high incidence of association with archaeological sites can, perhaps, be expected. In the temperate zone of Ontario, however, where fen peat deposits are common but not dominant features, the association is not so obvious, but, several archaeologists who have studied Palaeo-Indian sites in the province seem to, tacitly, acknowledge that such an association exists. In one study though, archaeological site locations (and concentric areas around them termed settings and vicinities) from four areas along the north shore of Lake Ontario (spanning several millennia from the late Pleistocene to early Holocene) were studied in a locational analysis to examine the topographic and environmental character of the terrain selected by Palaeo-Indian and early Archaic hunter-gather people (Swayze 1987). Because the early Holocene lakeshores of the Ontario Basin were considerably lower than modern Lake Ontario the archaeological site sample is assumed to be skewed towards an inland hinterland economic pattern because, presumably, large seasonal villages were located in proximity to the (now drowned) Great Lake shore. The terrain variables, of the archaeological zones measured, included: stream density, marsh/swamp/wetland density, maximum relief, soil drainage, and soil texture. These archaeological data were compared to a representative control sample of non-archaeological locations, and both samples were stratified by broad surficial zones (i.e. till, clay, moraine, etc.). In each geographic subset analysed, the terrain around archaeological sites differed significantly from the control sample locations, showing consistent patterns thought to be related to economic subsistence strategies. Of particular interest to the subject of archaeological association with peat and organic terrain, it was observed that stream density of 1st through 4th order creeks and marsh/swamp/wetland density of hunter-gatherer were often significantly higher than a representative ‘random’ location. This suggests Palaeo-Indian people had a more diffuse economic strategy based upon an inland, or ‘hinterland’, littorial adaptation, based on a variety of fish, plants and animals. Wetland littorals are interface zones between terrestrial and lacustrine environments and they provide diverse wildlife habitats that hunter-gathers appear to have consciously selected.
An example of a Historic Euro-Canadian settlement pattern with a strong association to peat deposits (muskeg) has been recorded in the Long Beach segment of Pacific Rim National Park, on the west coast of Vancouver Island (Swayze 1989). At the turn of the of the 19th to 20th century, numbers of young British immigrants – “remittance men” – took up homestead allotments in the rain forest. In those few years (1890-1914) before a wave of enlistment for the Great War ended the experiment, these raw pioneers struggled to drain a few muskeg ‘peat fields’ where strawberries were their only ‘crop’; to make a tiny clearing in the timber nearby for a shack; and to slash corduroy roads to interconnect their clearings. Their dreams were based on strawberries and the unfulfilled promise of a railway link around the south coast of the island to Victoria.