- Management Summary
- Research Design & Time Line
- Environment & Native American Culture
- GIS Design
- Archaeological Database
- Archaeological & Environmental Variables
- Model Development & Evaluation
- Model Results & Interpretation
- Project Applications
- Model Enhancements
- Model Implementation
- Landscape Suitability Models
- Summary & Recommendations
- Archaeological Predictive Modeling: An Overview
- GIS Standards & Procedures
- Archaeology Field Survey Standards, Procedures & Rationale
- Archaeology Field Survey Results
- Geomorphology Survey Profiles, Sections, & Lists
- Building a Macrophysical Climate Model for the State of Minnesota
- Correspondence of Support for Mn/Model
- List of Figures
- List of Tables
Minnesota's Environment and Native American Culture History
By Guy E. Gibbon, Craig M. Johnson, and Elizabeth Hobbs
3 Table of Contents
3.2 The Land: Now and in the Past
3.2.1 Minnesota Biomes
3.2.2 Landscape Changes
3.3 Minnesota's Precontact and Contact Period Culture History: An Outline
3.3.1 Precontact Period
3.3.2 Fur Trade
3.3.3 American Period
3.4 Archaeological Regions of Minnesota
3.4.1 Southwest Riverine (Region 1)
3.4.2 Prairie Lakes (Region 2)
3.4.3 Southeast Riverine (Region 3)
3.4.4 Central Lakes Deciduous (Region 4)
3.4.5 Central Lakes Coniferous (Region 5)
3.4.6 Red River Valley (Region 6)
3.4.7 Northern Bog (Region 7)
3.4.8 Border Lakes (Region 8)
3.4.9 Lake Superior Shore (Region 9)
3.5 Ecological Classification System for Minnesota
3.6 Implications for Predicting Site Locations
Predictive models of archaeological site location vary widely in the assumptions and underlying principles on which they are based (Kohler and Parker 1986; Kvamme 1990). Two main reasons for variation among models are: (1) differences in the social organization of the people whose archaeological remains are being modeled; and (2) differences in the composition and history of the landscape that is the backdrop to site location. The basic assumptions underlying predictive models of site location in complex societies with a hierarchical settlement structure and a well-developed market system are quite different from those of models of site location in less complex hunter-gatherer, band-based societies whose members move from place to place throughout the year in small groups in an annual settlement-subsistence cycle. Similarly, models for dynamic landscapes must be different from models for static landscapes. In dynamic environments, the micro niche in which a settlement was once found may be markedly different from its location today, and many sites may be deeply buried by massive sediment transportation, such as alluvial fan or dune formation. A first-step in developing a predictive model of site location, then, is to assess the nature of the societies whose activity locations are being modeled and of the landscapes they lived in.
Chapter 3 reviews Minnesota’s environments, both in the present and in the past, and the history of the state’s precontact and contact period Native American people. The emphasis throughout is on those factors and processes that affect the development and use of an archaeological predictive model of precontact and contact period archaeological site location in Minnesota. Besides outlining the background features that constrain and channel the modeling process for this period in Minnesota, the chapter serves as an introduction to the state's early Native American culture history, climate, and environment for those unfamiliar with its cultural and environmental history.
At historic contact Minnesota was a mosaic of forests, lakes, wetlands, and prairies (Marschner 1974; Aaseng et al. 1993; Tester 1995; Minnesota Department of Natural Resources 1995). Biologists organize this mosaic in a variety of ways, depending upon their research interests. None of these arrangements is ideal for predictive modeling, for they suggest static environments with stable geological and biological characteristics. When viewed through the millennia the land was dynamic and constantly changing. To simplify this review, descriptions by the first government land surveyors in the nineteenth century are used to divide Minnesota into four biotic provinces: boreal forest (spruce and pine), mixed hardwood forest (conifer/deciduous forest), deciduous forest (including oak savanna), and prairie (Marschner 1974; Cleland 1966; Mason 1981:56-61). Since plant communities reflect regional climatic patterns, and since animal species are better adapted to some plant communities than to others, biotic provinces are associations of climate and plant and animal communities. Figure 3.1 shows the location of these biotic provinces at specific times in the past.
184.108.40.206 Boreal Forest
The boreal forest is the most northern of the four provinces (Pielou 1988; Larsen 1980). It is also called the Spruce-Fir-Moose-Caribou Biome, a more descriptive label from the perspective of Native American hunter-gatherer societies (Cleland 1966). Although it now covers only a small slice of northeastern Minnesota, it expands to the north, stretching from Alaska to Newfoundland and northward to the treeline. Referred to over much of its area as the subarctic, it is known for its long, cold winters, deep snowcover, and short, warm summers. It is a recently deglaciated land of thin soils, exposed bedrock, and immature drainage. An intricate maze of bogs, lakes, and rivers remains as a heritage of this withdrawal of glacial ice. Its forest cover is taiga, or boreal coniferous forest, a forest composed of often dense stands of black and white spruce, white birch, jack pine, balsam fir, and tamarack. Poplar, cedar, and willow are also present, as are white pine, red pine, and hemlock. The latter species are intrusive members from the mixed hardwood province to the south and occur in greatest densities along the southern edge of the boreal forest.
Game animals associated with the boreal forest include moose, caribou, black bear, wolf, lynx, wolverine, marten, fisher, red fox, porcupine, beaver, and snowshoe hare. Fish are abundant in lakes and rivers. The most important of these are bass, pike, pickerel, whitefish, lake trout, sturgeon, smelt, perch, bullhead, catfish, sucker, and freshwater drum. Turtles, frogs, and clams are also present in some lakes and rivers.
For hunter-gatherers with a limited technology, the boreal forest is a difficult environment within which to gather adequate food energy throughout the year. Its game animals are often dispersed and tend to have low population densities, and its wild plant foods are limited. In addition, its dense forests with their bogs, lakes, and rivers, and its severe winters with heavy snowfalls make travel difficult, even with canoes and snowshoes. As a result, human populations before historic contact tended to be low in numbers and dispersed throughout much of the year. They congregated only in seasons and usually at places where fish and game were particularly abundant.
220.127.116.11 Mixed Hardwood Forest
Most Minnesotans call the mixed hardwood forest the "north woods" (Beyers 1989; Daniel and Sullivan 1981). Although a distinctive biotic province in its own right, it is also a broad transition zone between the much larger boreal forest and deciduous forest provinces, with boreal forest plants and animals increasing in frequency to the north and deciduous forest plants and animals becoming more common to the south. While all of the boreal forest trees are present, this biome is dominated by still greater quantities of white, red, and Norway pine, cedar, alder, yellow birch, beech, elm, hemlock, aspen, basswood, and sugar maple, especially to the south. All of the boreal forest mammals are present, too, but caribou, wolverine, and lynx are rare, while moose, mountain lion, and bobcat are more common. White-tailed deer are common where deciduous trees are abundant.
The climate in the mixed hardwood forest is intermediate as well. While winters are still long and cold, summers are longer and warmer than in the north. Although the land was also heavily glaciated, its soils are richer in organic content and tend to be deeper than those in the north. It retains, however, extensive lakes and bogs, again a heritage from the passing of the last glacier.
Since it is a mixture of boreal forest and deciduous forest plants and animals, the mixed hardwood forest was less harsh for hunter-gatherers. A wider variety of plants and animals were present as a source of food and materials, and fish were equally abundant. Though the growing season was too short to allow southern domesticates such as maize (corn) and beans to mature, the harvesting of vast amounts of wild rice eventually supported sizeable Native American villages in some areas.
18.104.22.168 Deciduous Forest
An increasingly narrower northwestward extension of the deciduous forest province separates the mixed hardwood forest and prairie biotic provinces in Minnesota (Yahner 1995). Also called the Oak-Deer-Maple Biome, this province is characterized by broadleaf deciduous trees, like oak, hickory, maple, beech, walnut, butternut, elm, ash, basswood, and cottonwood, which drop their leaves in winter. Large numbers of a wide variety of animals were present. Although white-tailed deer were the primary game animal, black bear, elk, opossum, raccoon, cottontail rabbit, squirrels, gray fox, bobcat, mountain lion, wolf, mink, otter, beaver, muskrat, and woodchuck were hunted, too. Occasional buffalo and badgers inhabited areas of open grassland. The climate of the province was still more moderate, with shorter winters, less snowfall, and longer, hotter summers. Soils were deeper and richer, and drainage systems much more mature than in the northern provinces.
Of the four biotic provinces reviewed here, the densest populations of Native Americans in eastern North America were in the deciduous forest province. Besides its rich plant and animal resources, its longer growing season made the cultivation of domesticated food plants possible and even highly productive. In some areas, especially in the southeastern United States and along the central Mississippi River valley, complex farming societies developed. These developments eventually influenced Minnesota, too, as will be mentioned in the following section of this chapter.
The fourth biotic province is the prairie (Costello 1980). Also called the Grass-Oak-Bison Biome, it contained prairie vegetation and animal associations. Typical mammals included buffalo, elk, skunk, badger, jack rabbit, ground squirrel, gopher, and coyote. While generally thought of as a region of sweeping tall grass prairie with scattered copses of oak and hickory, forests were present along stream valleys, around lakes, and on some plateaus and low hills. The most common tree species were oak, sycamore, cottonwood, elm, hackberry, maple, basswood, and beech. Winters and summers were similar to those of the deciduous forest province, but with less precipitation.
Herds of buffalo on the prairies acted as magnets that attracted Native Americans. Hunter-gatherers from the northern forests of the state and farmers from the southern deciduous forest entered the prairies seasonally to hunt these herds. Small groups of people who lived year-round on the prairies had a successful buffalo-hunting economy that remained largely unchanged for thousands of years. Eventually, however, sizeable farming villages were established along a few of the major rivers flowing through the prairies.
22.214.171.124 Landscape Complexity
This description of Minnesota’s major biotic provinces at historic contact has glossed over complexities. A few are worth stressing here, for they are important in understanding the context in which precontact Native American cultures developed and some of the many difficulties involved in predicting the location of their settlements.
First, each biome was composed of a rich mosaic of still smaller biomes. Small lakes were abundant in some parts of the prairie but not in others, and lakes and bogs were equally scattered throughout the mixed hardwood forests. Furthermore, the major biomes merged into one another in such intricate patterns that it is often difficult to determine with precision where one began and the other ended.
Second, Native Americans did not regard these provinces as bounded regions within which they had to confine their pursuit for food energy. The biotic provinces were resource zones, and most Native American societies moved freely from one to another throughout the year. Their movement across biomes makes the task of predicting the location of sites more difficult, for the ability to predict site locations at a high level of probability depends in part on understanding why sites are where they are. Within a single biotic province the relative seasonal abundance of food sources can be used to model yearly subsistence-settlement patterns and site-specific activities. This modeling process becomes complex when the extent of the annual range is unknown, for subsistence options become multiple and uncertain when different combinations of the seasonal use of the four biomes are possible.
Finally, the four provinces have not been static biomes. There have been significant changes in their distribution and composition both before and after historic contact. For example, the southward bulge of deciduous forest in the south-central section of the state, the Big Woods, developed only in the late precontact and contact periods (Grimm 1981, 1983, 1984). Farmers and loggers have severely altered the face of the prairie and the biotic composition of the mixed hardwood forest. Even more drastic changes occurred long before historic contact, when massive sheets of ice pushed across the state. As these ice sheets inched southward, they compressed the biotic provinces. When they retreated northward at the end of the Ice Age, they left a raw, harsh, exposed environment that only slowly developed a soil base capable of supporting larger trees and grasslands. Because these dramatic changes in Minnesota’s past landscapes play an integral role in predicting archaeological site location in the state, they are briefly outlined in the remainder of this section.
126.96.36.199 Wisconsin Glacial Period
People apparently first entered the Americas near the end of the last glacial ice age, the Wisconsin (Bonnichsen and Turnmire 1991; Fagan 1987). At the time glaciers, sheets of ice so large and thick that they depressed the land they covered and raised land around their edges, covered most of Minnesota. Glaciers not only limited the range of early settlement in the state, they resculptured its surface as they retreated northward by depositing tons of rocks and sediments, and by creating many rushing rivers and extensive lakes. Minnesota's famous "10,000 lakes" are legacies of these late glacial events (Ojakangas and Match 1982; Zumberge 1952; Pielou 1991).
The Wisconsin glacial period began about 60,000 years ago and ended around 10,000 years ago. Much of Minnesota was covered with glacial ice throughout this period, although glaciation took place in a complex series of ice lobe advances and retreats (Wright 1971, 1972a; Dyke and Prest 1986). During the maximum southward extent of ice masses about 20,000 years ago, only an area around Wolf Creek in central Minnesota and the southern two corners of the state remained ice free (Figure 3.2). These unglaciated areas, however, were not well suited for human habitation, for they were sparsely covered by vegetation and contained few animals. Two principal ice lobes covered large areas of Minnesota near the end of this period. The Superior Lobe crept southwestward out of the Lake Superior Basin into east-central Minnesota, and the Des Moines Lobe first moved south through the Red River valley and then turned east into the Minnesota River valley. These huge masses of expanding ice gradually depressed the land they covered from a few feet near their edges to several hundred feet at their northern cores.
When the glacial lobes began their retreat around 14,000 years ago, the resulting meltwater formed enormous rivers and lakes. Figure 3.3 shows the location of major glacial lakes in Minnesota. The largest of these, Lake Agassiz, with a basin of almost 600,000 square miles, covered all of northwestern Minnesota at one time and was the largest glacial lake in North America (Teller and Clayton 1983). It began forming in the southern Red River valley 11,700 years ago and finally disappeared from the state around 9000 years ago. During much of this period, its only outlet was the Minnesota River Lowland, for possible northern outlets were barricaded by ice. The size of its principal southern outlet, known as Glacial River Warren, is still visible today, for it is the broad Minnesota River valley (Matsch 1983). As the ice continued to retreat, previously blocked northern drainage outlets gradually opened and Lake Agassiz began to drain northward as the Red River does today.
Lake Superior also fluctuated greatly in size and form during the Wisconsin glacial period. During the height of the Ice Age, when enormous amounts of water were locked up in the continental glaciers, water levels in the lake were quite low. As the ice retreated, the level rose again, as did much of the surrounding land when the weight of the ice was removed (LaBerge 1994; Phillips 1993). Periodic eastward discharges from Glacial Lake Agassiz also had catastrophic effects on Lake Superior (Clayton 1983; Drexler et al. 1983; Teller and Thorleifson 1983). The result was an intricate shift in lake levels, locations of outlets, and beach positions that geomorphologists and archaeologists try to reconstruct to decide when particular parcels of land were above water and where they were in relation to the shore of the lake.
Depositions of till, channel cutting by meltwater rivers, the formation of lakes from gigantic blocks of melting ice, and other geological processes triggered by the glacial retreat created an actively changing landscape in the state between 12,000 and 10,000 years ago. These shifting landscapes were unlike any in Minnesota today. Only in the southern corners of the state did more ancient landscapes, which formed before the final glacial surge and retreat, survive. Nonetheless, although not covered with ice during the Wisconsin glaciation, they were still affected by late glacial geological processes. Strong winds along the front of the glacier masses deposited a deep mantle of fine-grained dust called loess in both areas. Fine, silty soils eventually developed within this loess deposit. Meltwater rivers also deposited outwash, and widened and scoured valleys in these "unglaciated" sections of the state.
During the greatest southward thrust of Wisconsin ice, either tundra or a 400-mile wide, spruce-dominated boreal forest bordered the ice edge from the Atlantic Ocean to the Rocky Mountains (Wright 1976a, 1976b; Amundson and Wright 1979; Webb et al. 1983; Webb 1981). Across most of late glacial Minnesota, spruce forest probably grew to the edge of the ice and might even have grown over stagnant ice as the glaciers receded. In fact, the presence of buried stagnant ice might have been the reason tundra was not present, for the presence of the ice blocks would have retarded the creation of the many lakes and bogs typical of a tundra until after active ice had retreated still further northward. The pollen of sedges (Cyperaceae), wormwoods (Artemisia), grasses (Graminae), willows (Salix), and of other plants commonly associated with tundra has been found only in the northeastern corner of the state. Tundra is a treeless area with permanently frozen subsoil that can only support lichens, mosses, and stunted shrubs. This limited zone of tundra was present from the beginning of the retreat of local ice to about 10,000 years ago, when a boreal spruce (Picea) forest spread across the area (Wright 1971).
188.8.131.52 Early Holocene
By 12,000 years ago, all of southern Minnesota and most of the central part of the state were free of glaciers and covered, for the most part, by open spruce parkland that contained coniferous trees and grasses (Figure 3.4). This early forest has no modern counterpart, for it contained small amounts of oak and other temperate climate deciduous trees, such as black ash, and lacked jack pine, a common tree in northern boreal forests today. This peculiar mix of trees was probably the result of the different rates of migration of particular tree species from distant refuge areas and of a more temperate, warmer climate than now exists in northern boreal forests (Wright 1971, 1976a, 1976b; Amundson and Wright 1979). Many low-lying areas within the forest were probably filled with open water and marsh or swamp. Openings on higher ground were most likely covered with shrubs and grasses. In some areas, such as the southeastern corner of the state, the spruce forest seems to have contained large, open, swards of grassland.
A deciduous forest dominated by birch, elm, and other trees that shed their leaves at the end of the growing season entered the southwestern corner of Minnesota as early as 12,000 years ago and spread outward about 11,000 years ago. At the same time, that is, about 11,000 years ago, a new oak-elm forest penetrated the southwestern corner of the state and spread rapidly northward and eastward. This oak-elm forest covered most of southwestern Minnesota throughout the early Holocene (post-glacial) period.
Red or jack pine or both was present along with birch and elm in a mixed forest that replaced the earlier spruce forest in the southeastern corner of the state by 10,200 years ago. Eventually, oak replaced birch and pine trees across southern Minnesota, while pine replaced spruce forest in central Minnesota about 10,000 years ago. The northward retreat of the spruce forest is a good marker for the end of the Wisconsin ice advance in different parts of the state. The end of the Ice Age was not an event that occurred at one point in time, but was a process that progressed slowly from south to north.
By 11,200 years ago, the last massive sheets of glacial ice had receded north of Minnesota, although ice remained in the Lake Superior Basin for at least another 1000 years. The pioneer spruce forest was short-lived in northern Minnesota. It was replaced by a rapid intrusion of pine trees about 11,000 years ago (Figure 3.5). With the recession of Lake Agassiz, open oak forests spread over large portions of the northwestern and north central sections of the state.
The composition of the animal communities that occupied these Early Holocene forests is not fully known, although fossilized bones from around the state provide some clues (Kurtén and Anderson 1980; Harington and Ashworth 1986; Agenbroad 1984; Dreimanis 1967; Guthrie 1980). The spruce forest was probably inhabited by now-extinct animal species, such as mastodon and giant beaver, and by many modern mammals that were common in the north woods in the early historic period. Mastodons preferred feeding on the coarse vegetation of pine and spruce forests, for they found it difficult to find food in the winter in deciduous forests that dropped their leaves. More open areas near the ice contained mammoth, barren-ground caribou, and musk ox, and animal communities in open prairie-like areas to the south included elk, mammoth, and now extinct forms of large buffalo.
The reason for the extinction in the Americas of many large mammals like mammoth and mastodon at the end of the Ice Age remains a puzzle. Their disappearance has been attributed to overkill by human predators and to climatic change (Martin 1967). A combination of these processes seems a more likely explanation (Martin and Klein 1984). Modern species of animals probably inhabited the deciduous and mixed deciduous-pine forests that followed the spruce forest northward. White-tailed deer, moose, porcupine, weasels, fisher, otter, coyote, bobcats, red fox, timber wolf, black bear, and beaver are familiar examples. There were also many birds, fish, and amphibians (Hocutt and Wiley 1986; Bailey and Smith 1981; Stewart and Lindsey 1983). In fact the composition of the plant and animal communities of these forests and of the expanding southwestern grasslands seems very similar to their modern counterparts at historic contact, although these communities are not well known.
184.108.40.206 Middle Holocene
The following Middle Holocene postglacial period in Minnesota roughly coincides with the Atlantic, the Sub-Boreal, and the early half of the Sub-Atlantic climatic episodes (Bryson, et. al. 1970) (Table 3.1). Environmentally, the Middle Holocene was a dynamic period during which the prairie-forest border migrated far to the northeast of its present position (Figure 3.6). These severe vegetation dislocations were a result of the Atlantic climatic episode (also called the Mid-Holocene dry period or Prairie period); during which dry and possibly warm westerly winds blew across the state (Wright 1976b; Forester et al. 1987). At its most severe, annual rainfall may have been 20 percent less than modern levels and the average temperature 5 degrees Fahrenheit (2.8 degrees Celsius) warmer than today. Eventually, even deciduous woods became restricted to major river valleys, the edges of large lakes, and other less xeric, fire-protected areas throughout much of the state. At its most eastward position about 5000 B.C. (7000 years ago), the prairie-forest border in north-central Minnesota was more than 100 miles northeast of its modern position. However, by 4000 B.C. (6000 years ago), the climate had become cooler and wetter, and the prairie gradually retreated, reaching its approximate modern borders about 1000 B.C. (3000 years ago) (Figure 3.6).
These climatic shifts are intricately recorded in an uninterrupted sequence of annual laminations or varves that extend from the historic period back to late glacial times at Elk Lake in north central Minnesota (Bradbury and Dean 1993a). The history of the lake is clearly divided by varves into three well-defined episodes: postglacial lake, prairie lake, and modern mesic-forest lake with the prairie lake corresponding to the Mid-Holocene dry period (Bartlein and Whitlock 1993; Bradbury and Dean 1993b; Whitlock et al. 1993).
Cool N. Hemisphere
Mild N. Hemisphere
("Little Ice Age")
("Little Ice Age")
|Warm N. Atlantic|
"Medieval Warm Period"
ca. A.D. 950
ca. 6000 B.C.
ca. 7200 B.C.
|ca. 8850 B.C.|
Besides producing dramatic shifts in the alignment of the prairie-forest border, the Atlantic climatic episode caused many other changes in the distribution, density, and composition of plant and animal communities. These changes profoundly affected human adaptations in the state. During much of this period, the vast ocean of prairie that covered all but the northeastern corner of the state presented a startling contrast to the boreal vegetation present at 9000 B.C. (11,000 years ago) and the mixed hardwood forest at 8000 B.C. (10,000 years ago). Few, if any, spruce trees that once covered the state, remained in northeastern Minnesota, and oak-shrub vegetation extended northwestward along the prairie-forest border. Undoubtedly, animal communities suited to these zones of vegetation were present, too, for Figure 3.7 shows the strong correlation between the distribution of bison bone beds and prairie. A new arrival at this time was white pine, which entered northeastern Minnesota by 2500 B.C. (4500 years ago) from its refuge in the Appalachian Highlands.
As cooler and especially moister conditions returned to the state after 4000 B.C. (6000 years ago), the forest gradually moved westward, often in jumps, with oak woodland advancing across morainic ridges and with oak-poplar-brush prairie or woodland filling prairie openings. When these fire hardy woodlands became sufficiently dense and extensive to serve as effective fire barriers, mesic forest began to develop in their lee. The result was a continually changing prairie-forest mosaic under the influence of broad-scale climatic and local conditions (Clark 1993; Grimm 1983, 1984). The prairie-forest ecotone remained more sharply defined throughout most of the mid-Holocene, however, than it was in more recent periods. To the northeast, as the mixed forest moved westward, white pine migrated with it, eventually bypassing other pine species after 2000 B.C. (4000 years ago) in its migration through northwestern Minnesota.
The modern diversity of the mixed forest, and perhaps the floristic tension zone along its southern edge, dates to this expansion. However, the composition of the developing deciduous and mixed forests in the western Great Lakes region was not the same everywhere. East-west gradients existed that affected the food resources available for human exploitation. Beech, eastern hemlock, maple, hickory, and ash trees were more abundant to the east and pine, birch, and oak to the west. The east-west gradient in the north was from a mixed forest of pine with some maple trees in the east to forests of pine with increased numbers of birch trees in northwestern Minnesota. Travelers crossing the western Great Lakes would have found beech becoming scarce first, then eastern hemlock, maple, pine, and finally birch before they walked through oak scrub and out onto the prairies of northwestern Minnesota.
220.127.116.11 Middle-Late Holocene Transition
Cooler, wetter conditions, which coincide with the Sub-Boreal and early Sub-Atlantic climatic episodes, continued throughout the late Middle and early Late Holocene periods. In northern Minnesota extensive peatlands developed largely after 2000 B.C. (4000 years ago), when the cooler, wetter conditions slowed the decomposition of plant materials on poorly drained lake plains, especially on the bed of glacial Lake Agassiz (Graham 1957; Wright and Glaser 1983). Today these peat beds are about 10 feet thick. The cooler conditions also resulted in the reappearance of spruce trees in the north.
These large-scale vegetation shifts, which were caused by continent-wide climatic episodes, are easy to trace in the pollen record. Reconstructing the composition of small-scale localities exploited seasonally by human foragers at the time is, however, much more difficult. For example, shallow prairie lakes and marshes periodically dried completely, while deeper lakes often shrank drastically. Mapping the distribution of surface water in a small region for a particular moment in the past, is not a simple task.
The position of the prairie-forest border and the composition of the forests and woodlands could also have changed rapidly and dramatically within short periods for other reasons. A trail through the mixed forests in the north would have passed through a variety of growth phases caused by fire and the successional redevelopment of climax forests. Fire, soil, and topography also controlled the local distribution and composition of prairie, oak scrub, and mesic forest. The almost annual late-summer fires were blown northeastward by dry, southwesterly winds until they encountered natural fire breaks, such as areas of high relief, lakes, and streams. All these factors contributed to the constantly shifting patterns of local vegetation along the prairie-forest border.
By the end of the Middle-Late Holocene transition period (ca. 1500 B.C.), the major vegetation zones had, with a few exceptions, reached their presettlement positions. Climatic oscillations continued to occur throughout the Sub-Atlantic and Post-Sub-Atlantic, but they were minor compared with the impact of the warm, dry westerlies of the Atlantic climatic episode. The major difference between the distribution of vegetation then and now was the absence of the Big Woods, the bulge of elm-oak forest that covered a large area of south-central Minnesota in the late precontact and early European-American settlement periods (Grimm 1981, 1983, 1984). Although oaks, aspen, and willow had initiated the invasion of the prairie in this area at the end of the mid-Holocene dry period at ca. 2500 B.C., The expansion of Big Woods trees (elm, basswood, ironwood, sugar maple, ash, hickory, butternut) did not occur until 300 years ago. Grimm concluded that the expansion was a result of a reduction in fire frequency probably caused by increased precipitation and decreased temperature during the "Little Ice Age" (A.D. 1550-1915) (see Table 3.1). Fire had probably also reduced the width of the belt of deciduous trees that extended northwestward from the Big Woods until this time as well (Clark 1990) (Figure 3.8).
The life ways of the earliest people to live in Minnesota were an adjustment in part to these rapidly changing Late Glacial and Holocene landscapes. Since these landscapes differed from modern landscapes in distribution, and often in composition, predictive modelers cannot assume that the present environmental associations of a site mirror those when the site was created by Native Americans hundreds, if not thousands, of years before European-American contact.
The Native Americans encountered in the middle of the seventeenth century by Minnesota's first European explorers were heirs to a rich and varied cultural tradition that can be traced back at least twelve thousand years. A goal of Minnesota archaeologists is to identify and chart the sequence of archaeological cultures that makeup this long tradition (Dobbs 1990a; Dobbs 1990b; Anfinson 1987; Wilford 1941, 1955, 1960).
Progress has been made over the years in developing a meaningful precontact and contact period sequence for the state. Because of the Early Holocene's turbulent landscape and the masking of Middle Holocene settlement locations by thick deposits of sediment, the more recent portions of the sequence are known in greater detail than are earlier portions. This is reflected in the absence in the state of named archaeological cultures for early Paleoindian, Plano, and Archaic societies.
The first human inhabitants of Minnesota were most likely Paleoindians (Fagan 1987; Florin 1996; Harrison et al. 1995; Magner 1994; Mason 1981; Pettipas and Buchner 1983). These pioneers entered the state in small numbers as the lobes of the last major glacier, the Wisconsin, receded. Their sites are difficult to find since they are small, contain few artifacts, are few in number, and are often deeply buried beneath more recent sediments. In fact Paleoindian sites are known primarily from a scatter of large and distinctive lanceolate projectile points. Archaeologists know little about their daily life compared with the life ways of their descendants. In some areas they seem to have been highly mobile gatherers and hunters who pursued big game such as bison, woodland caribou, mastodon, and mammoth. However, in the Late Glacial and Early Holocene forests of Minnesota they probably relied more on gathering and the hunting of a variety of smaller animals (Gibbon 1996a).
The Paleoindians were followed by Archaic hunters and foragers (Bleed 1969; Dobbs 1978; Gibbon 1996a, 1996b; Michlovic 1983; Phillips and Brown 1983; Shay 1971; Steinbring 1974). Their customary ways of behaving emerged in part as adjustments to a rapidly changing postglacial environment and the extinction of mammoth and other large Ice Age mammals. New complexes of plants and animals appeared as the spruce forests of the Paleoindian period followed the retreating ice northward. The melting ice exposed new land surfaces with extensive lakes and large, swift rivers quite unlike any in present-day Minnesota. These people increasingly specialized in the exploitation of smaller game, fish, shellfish, plant foods, and other energy resources that were not very abundant in late glacial environments. Their more varied artifact assemblages reflect this adjustment in subsistence practices. Archaic hunters and foragers seem to have been less nomadic and more numerous than Paleoindian societies. As a result, their sites, which are easily identified by the presence of large notched and stemmed projectile points, are more frequently discovered and excavated by archaeologists than are the camps of earlier people. However, because of massive sedimentation during the earlier phases of the Archaic, which correspond with the Early and Middle Holocene periods, many Archaic sites are deeply buried in river valley deposits and are masked from surficial searches.
In Minnesota, Woodland cultures are separated into an earlier Initial Woodland time period (ca. 500 B.C. - AD 500) and a later Terminal Woodland period (ca. AD 500-1650). Although a hunter-gatherer way of life continued, the Initial Woodland is marked by the first appearance of pottery vessels and the construction of earthen burial mounds (Anfinson 1987; Benn 1979; Gibbon 1990; Hudak 1974; Lugenbeal 1976; Stoltman 1973). These innovations were not adopted in all areas of the state at the same time or necessarily together. The result was overlap in time between Late Archaic and early Initial Woodland cultures, just as there had been an earlier period of transition between Paleoindian and Archaic cultures. Because Initial Woodland sites are not as deeply buried as earlier sites and burial mounds frequently mark their presence, they are more easily found and more frequently excavated by archaeologists. They have also been grouped together into archaeological cultures. This greater degree of attention is reflected in the presence for the first time of the names of regional archaeological cultures, such as Howard Lake, Fox Lake, Malmo, and Laurel.
Terminal Woodland people were also hunters and gatherers (Gibbon and Caine 1980; Gibbon 1994; Lugenbeal 1976). However, their economy in the mixed hardwood forests to the north was increasingly supplemented by the harvesting of wild rice. The number of people in the region rose dramatically, and major abrupt changes occurred in ceramics and other artifact forms and in settlement patterns. Archaeologists have generally relied on the geographic distribution of the distinctive ceramics and burial practices of the period to identify archaeological cultures in this northern region of the state. Examples are Kathio, Blackduck, and Psinomani. In the southern part of the state covered by deciduous forests and prairies, some Terminal Woodland people gradually began growing maize, the domesticated plant we call corn. Some people also built distinctive effigy mounds having the shape of birds, bears, and other animals. Most of these southern Terminal Woodland societies abruptly adopted new life ways and artifact assemblages, as well. Archaeologists group these transformed societies together and regard them as a northern expression of a "Mississippian" way of life. Mississippian sites are easily distinguished from Woodland sites by their distinctive ceramics, by their larger size and greater artifact density, and by the presence of maize (corn) fragments. Three Mississippian complexes have been identified in Minnesota: Silvernale, Oneota, and Plains Village (Dobbs 1984; Gibbon and Dobbs 1991; Gibbon 1979, 1982, 1983, 1991, 1993; Wilford 1945).
The transformation in life way of Terminal Woodland people was similar in broad outline throughout the state and may have been a regional expression of changes occurring at the same time across eastern North America. Although connections with historic ethnic groups are still clouded, these newly emergent cultures mark the first appearance in Minnesota of the life ways of the Dakota, Iowa, and other Native American societies as they were before European-American contact. It was Mississippian people to the south and Terminal Woodland people to the north who met the first Europeans to visit the state in the middle of the seventeenth-century.
According to most accounts, the first Europeans to set foot in Minnesota were two Frenchmen, Sieur des Groseilliers and Sieur de Radisson (Folwell 1956:7-13; Nute 1978). Although the details are open to dispute, they apparently entered Minnesota between 1659 and 1660. Like other Europeans and Americans who followed them for the next 180 years, they came in small numbers in search of natural resources, such as furs, that could be exploited for their own market system (Gilman 1982, 1992). By 1678, merchants in Quebec and Montreal had formed a company to trade with the Dakota, the dominant Native American people in the state at the time. These fur trade-related activities initiated the French period of exploration and occupation in Minnesota, which lasted into the early 1760s (Folwell 1956:1-52). However, European trade goods, diseases, and influences had already reached north central North America years before Groseilliers and Radisson set foot in the state, and at least some Dakota had earlier contact with Europeans in their own travels eastward. This period, when European trade goods, diseases, and influences became available via native intermediaries is generally called the "protohistoric" period (Mason 1981:382).
In their hundred years of activity in Minnesota, the French explored much of the state and surrounding region and established an extensive network of forts and fur trading posts. These forts and posts, which were along major bodies of water, were the centers of their activity. Some of the earliest were built by Daniel Greysolon, sieur du Luth, in the St. Croix River region and at present day Fort William, in the early 1680s. Many more were built after the French decided to occupy and settle the Mississippi Valley from north to south in the final years of the seventeenth century. Among these were posts built by Pierre Charles le Sueur on Madeline Island in northwestern Wisconsin in 1693, on Prairie Island at the mouth of the St. Croix River near Red Wing in 1694, and at the mouth of the Blue Earth River near Mankato in 1700; Fort Beauharnois on Lake Pepin, which was established by Rene Boucher in the late 1720s; and two posts, Fort La Jonquiere on Lake Pepin and Fort Duquesne near the present city of Little Falls, which were built in the 1750s. For the most part, the exact location of these forts and posts has proven elusive (Birk 1978).
Following the Treaty of Paris in 1763, the British began their half-century of activity in Minnesota (Folwell 1956:53-72). Like the French, their primary interest was exploration and the fur trade. During this period, the British built many fur trade posts across the state. It was also the period during which major changes occurred in the distribution of Native American people in the region. By 1800, the Anishinabe (Ojibwa), who had been pushing into Minnesota from the northeast, took control of the lakes and forests of northern Minnesota, and the Dakota (Sioux) had moved south along the Minnesota River valley and westward into the Dakotas (Treur 1994). Legal possession of the state passed to the United States with the treaty of peace of 1783 between the British and Americans. However, the United States did not exert control of the state until Zebulon Pike’s expedition through the state in 1805-1807 and the establishment of Fort Snelling in 1819 at the junction of the Minnesota and Mississippi rivers (Pike 1965; Ziebarth and Ominsky 1970; Hall 1987).
The French and British presence in North America resulted in enormous changes in Native American life ways in Minnesota and surrounding regions. Among these were: (1) an influx of Native Americans from the east, who were drawn by trade opportunities or pushed westward by the Iroquois wars; (2) massive depopulation of native peoples in some areas because of warfare and newly introduced diseases; (3) a switch from hunting for subsistence to hunting for trade; (4) a gradual movement southward and westward by the Dakota (Sioux), and the movement into the northern part of the state by the Anishinabe (Ojibwa); and, (5) the replacement of many Native American manufacturing materials (e.g., stone, pottery, bone) by European materials (e.g., copper, brass, iron, glass, porcelain). However, activity in the state still focused primarily on hunting, gathering, and horticulture. Travel and settlement were still largely restricted to corridors along larger bodies of water.
With the dawn of the American period, Native American and European-American life ways in Minnesota began to change dramatically. The opening of a commercial sawmill in the village of Marine on St. Croix in 1839 marked the beginning of the lumbering business in the state. Minnesota became a Territory in 1849 and the thirty-second state in 1858. All these changes were accompanied by an ever increasing torrent of European-American settlement and the establishment of towns, cities, and non fur trade-related enterprises. The factors that determined the location of Native American settlements and daily activities became increasingly different from those that had determined the location of earlier settlements and activities. In addition, the location of these settlements and activities is increasingly documented in archival records. For these reasons, the Mn/Model project’s modeling effort was restricted to archaeological site locations that predate 1837 and the active beginning of the American period in the state.
An initial assumption in the research design phase of this project was that it was unlikely that a single model could adequately predict the location of archaeological resources in all regions of the state. The primary reason was the environmental and cultural diversity of the state throughout the precontact and contact periods described in sections 3.2 and 3.3 above. It seemed more likely that predictive models would have to be formulated for specific regions. Although the state has been divided into regions by archaeologists several times using different criteria (Johnson 1969; Minnesota Historical Society 1981; Wilford 1941), it was decided that the most promising regional framework for the initial model building phase was the framework suggested by Anfinson (1988, 1990) (Figure 3.10). Among the reasons for this decision were: (1) his nine regions seemed a pragmatic number for analysis (other frameworks ranged from three to twenty-seven areas); (2) it is based on features of the natural environment that have been stable throughout the precontact and contact periods, but especially during the Late Holocene when most sites in the archaeological database were probably formed; (3) the regions were thought to exhibit significant internal environmental and, to some degree, cultural consistency; and (4), although designed for the precontact period, the framework probably applies as well to the presettlement contact period.
Anfinson's nine archaeological regions are based on the assumption that the distribution of natural resources significantly affected the distribution of precontact Native Americans, which is an assumption that underlies nearly all contemporary predictive modeling projects (see Section 2.2). Given this assumption, according to Anfinson (1990:138), a regional framework should (1) reflect the distribution of significant resources, (2) contain relatively uniform internal environments; and (3) maintain internal environmental cohesiveness through time. Although the resulting regions would not bound settlement-subsistence systems, for precontact and contact period Native Americans were very mobile, it was assumed that the "types of sites, types of site locations, and even particular types of artifacts" should be consistent within a region (Anfinson 1990).
This regional framework was developed over the past 20 years by Scott Anfinson. It was begun while he was employed by the Minnesota Historical Society as Municipal-County Highway Archaeologist (Anfinson 1976). He first defined seven environmental or archaeological regions, based on a series of biophysical variables (i.e. hydrology, topography, geology, early historic vegetation). A minor modification of this system was made in 1983 by subdividing the northeastern coniferous region into northeast and central areas, yielding a total of eight regions (Anfinson 1984). The purpose of this scheme was to assess the archaeological potential and demonstrate differences in site distributions linked to resource exploitation within each region. Regional boundaries were based on previously defined physiographic areas and presettlement vegetation distributions, supplemented by information on lake basins (Anfinson 1984:172). He also described the physical characteristics of each archaeological or ecological region along with its archaeological record.
Four years later, Anfinson (1988) refined the system by adjusting regional boundaries and reorganizing the northeastern part of the state into two regions. Two years later, Anfinson (1990) formally published the resulting nine-region scheme in connection with the Woodland period, along with discussions of the physical and precontact cultural characteristics of each region. Although differences in physiography and mid-nineteenth century vegetation were discussed in this paper, a shift to emphasizing surface hydrology (i.e. lakes) in the construction of archaeological regions was made (Anfinson 1990:139, 143). The archaeological regions in this revision remained largely unchanged from those defined earlier. They retained qualifying terms based on location, vegetation, and hydrology (e.g. Central Lakes Deciduous, Southeast Riverine). Later, most regions were divided into subregions on the basis of river drainage basin or river edge for finer control of intra-regional comparisons and to create areas of comparable size (Scott Anfinson, personal communication 1998). The following is a description of the process used by Anfinson (1990:139-145) to divide the state into archaeological regions. It should be noted that the regions resulting from this most recent discussion are largely unchanged from his earlier results.
The framework Anfinson developed is a system of hydrologic regions based on the presence or absence of lakes, lake morphology, and lake depth. Anfinson began the process of regionalization by identifying four major areas of the state where there are almost no lakes. These were the Lake Agassiz plain in northwestern Minnesota, the Lake Superior shore highland in northeastern Minnesota, the inner Coteau des Prairies in extreme southwestern Minnesota, and the Rochester Till Plain in southeastern Minnesota (Figure 3.9). For this, he defined "no-lake" as an area where there are 0-1 lake basins per township. He then separated the Lake Agassiz plain into two regions because the history of the eastern arm was significantly different from the Red River valley area proper (Figure 3.10). Besides being vacated by the glacial lake earlier (ca. 11,300 B.P. for the eastern arm compared to ca. 9000 B.P. For the Red River valley area), extensive peatlands began to develop in the eastern arm of the Agassiz basin by ca. 4000 B.P. A few large, shallow remnant lakes (e.g., Red Lake, Thief Lake) are present in the eastern arm. The western division was mostly prairie following the final retreat of Lake Agassiz. The dividing line between the two regions follows the Campbell beach, a remnant of the last glacial advance between ca. 9900-9500 B.P., extending from central Polk County into central Roseau County, then straight north to the Canadian border (Figure 3.10).
This process of regionalization resulted in the identification of five "non lake" regions and one very large lake region, which was then divided into four smaller regions. A bedrock lakes region at the far northern end of the Minnesota lake district was identified that was clearly different from lake regions to the south. Besides being formed in glacially-eroded bedrock cavities in the Canadian Shield, these far northern lakes have different bottom profiles and water chemistries, which has resulted in different floral and faunal compositions (e.g., wild rice is less abundant and there are different fish species). A below 10 m/over 10 m lake depth criterion was used to separate the drift-floored lake area to the south into a shallow lake region in southwestern Minnesota and a deep lake zone in central Minnesota. Paleoecological studies in northwestern Iowa and northeastern South Dakota showed that lake levels in this part of north central North America can drop up to 10 m during periods of severe drought (Van Zant 1979; Watts and Bright 1968). A dry lake, besides losing its aquatic plants and animals, becomes an ineffective firebreak.
The deep lake zone of central Minnesota was divided into two by a watershed line that places the Mississippi Headwaters, St. Louis River, and Kettle River watersheds in a northern region and the Crow Wing, Mississippi-Sauk, Crow, Rum, Snake, Lower St. Croix, and Metropolitan watersheds in a southern region. For much of the precontact period and during the contact period, the two regions shared many environmental characteristics. Both were dominated by woodlands, white-tailed deer was the principal game animal in both, and together they comprised the principal wild rice area of Minnesota. However, they were separated by significant climatic differences. Coniferous forests covered the colder northern area. Its lakes were frozen longer throughout the year, and native corn could not be productively grown. The warmer, more southern region was covered with deciduous forest, had been overrun by prairie during the Prairie period, contained bison as an occasional intruder, and was warm enough to grow corn productively.
The resulting nine regions are: (1) Southwest Riverine; (2) Prairie Lake; (3) Southeast Riverine; (4) Central Lakes Deciduous; (5) Central Lakes Coniferous; (6) Red River Valley; (7) Northern Bog; (8) Border Lakes; and, (9) Lake Superior Shore (Figure 3.10). As Figure 3.10 illustrates, all regions but 1 and 8 are further subdivided into subregions identified by position within a region, such as 5s (5 south), 5e (5 east), and 5c (5 central). During the development of the framework, the names of regions 4 and 5 have appeared variously as Central Lakes, Central Deciduous Lakes, Central Lakes Deciduous, Central Coniferous Lakes, and Central Lakes Coniferous, and the name of Region 2 as Prairie Lake or Prairie Lakes. The terminology adopted in this project is that recorded on the SHPO archaeological site database guide.
Initial predictive models (Phases 1 and 2) were developed for each of these nine archaeological regions and, in some cases, their subregions. Because of the importance of environmental characteristics in the modeling process adopted in the Mn/Model project, the environment of each region is summarized in some detail in the remainder of this section. In turn, because of the compactness and completeness of Anfinson's (1990:145-151) descriptions, they are repeated here in an amended form with permission of the author and the editor of the series in which they appeared. Since the models are based on modern environments, characteristics of the Late Holocene and especially the historic environments of each region are emphasized.
The environmental descriptions are followed by a review of site location in each region by archaeological tradition and cultural complex. This review is based primarily on Anfinson (1988, 1990), who developed the concept of the regions and has written about the archaeological character of each region. Other pertinent references are included in the text. Since Anfinson divided precontact Minnesota into three periods in his discussions, that division is followed here. In his scheme, Minnesota was occupied in the Early Prehistoric period (ca. 9500-6000 B.C.) by Late Glacial and Early Holocene hunter-gatherers, whose archaeological cultures belong to the Paleoindian tradition. The Middle Prehistoric period (ca. 6000-3000 B.C.) corresponds with the Prairie period and was dominated by Early Archaic hunter-gatherers. The beginning of the Late Prehistoric period (ca. 3000 B.C. - A.D. 1650) is marked by the appearance of modern (Late Holocene) climatic and environmental conditions, and the presence of Late Archaic Woodland, and Mississippian archaeological cultures. Anfinson's three environmental divisions should not be confused with Johnson's (1988) Early, Middle, and Late prehistoric periods, which are based on cultural factors, such as the presence of ceramics, earthen burial mounds, and maize horticulture. This review serves the purpose of establishing a baseline of lore about the location of site types within a cultural historical framework for each archaeological region. This baseline can be used to generate hypotheses of settlement systems for Minnesota's past archaeological cultures that can be tested using the Mn/Model database and used to support interpretation of the statistical models.
Located in the extreme southwestern corner of Minnesota, Region 1 includes all of Rock County, large parts of Pipestone and Nobles counties, and small portions of Lincoln and Murray counties. It is the smallest archaeological region in Minnesota, although it is part of a larger out-state region that includes portions of northwestern Iowa and southeastern South Dakota.
Bedrock outcrops of Sioux Quartzite are common in the western part of the region. While there are no outcrops of lithic materials of good flaking quality for chipped stone tools, occasional exposures of Catlinite, a soft, clay-rich stone produced from chemically weathered quartzite, were mined by Native Americans to make pipes and plaques. The region was not glaciated in the Late Wisconsin. Extensive loess deposition during the Late Pleistocene covered pre-Late Wisconsin grey till and contributed to the formation of soils dominated by fine silty loams.
Rock River is the major drainage-way. It flows south to connect with the Missouri River system. Many small but deeply entrenched creeks (e.g., Split Rock, Pipestone) flow to the southwest. There are 140-160 average annual frost-free days and the average annual precipitation is 24-26 inches. The last spring frost occurs in early May and the first fall frost at the end of September. The daily temperature highs average about 85 degrees F in July and about 24 degrees F in January.
At the time of European-American settlement, Region 1 featured tallgrass prairie vegetation and a stream-dissected landscape without lakes. Trees were scarce due to regular prairie fires and occasional droughts, but some woody vegetation was present along the major streams and at favorable topographic breaks. The largest woods were along the Rock River in south central Rock County. These were typical river bottom forests dominated by elm, ash, and cottonwood.
Major game animals at the time of European-American settlement included bison, elk, and smaller upland mammals. The only common game bird in the Late Holocene may have been the sharp-tailed grouse; waterfowl were not abundant because of the lack of extensive wetlands. Fish were not a major resource in Region 1 due to the lack of lakes and multiple large rivers. Seasonally available prairie plants included the prairie turnip (Psoralea esculenta) and the ground plum (Astragalus caryocarpus).
During the Early Prehistoric period (9500-6000 B.C.), The Southwest Riverine Region was covered with an open boreal forest that was gradually replaced by a deciduous forest towards the end of the period. At present, no Early Prehistoric sites or isolated artifacts, such as fluted or Plano projectile points are recorded in the SHPO archaeological database for this area. Early Prehistoric sites are most likely located along streams and adjacent to glacial features, such as meltwater channels and pro-glacial lakeshores. If sites from these small numbers of mobile hunter-gatherers are present in the region, they may be deeply buried in river valley alluvium or loess.
Recorded Middle Prehistoric period (6000-3000 B.C.) sites are rare here. Plants of the short-grass prairie and large bison herds were present in this region during the Prairie period. The great majority of the sites recorded in the Southwest Riverine Region are probably associated with the Late Prehistoric period (3000 B.C. – A.D. 1650), when the climate was wetter and possibly cooler than during the Prairie period. Large base camps and winter camps were probably uncommon during this period because of a lack of large woods and a sufficiently diverse subsistence resource base. The habitation sites that were present seem clustered along the Rock River, where the region's greatest concentration of wood and water occurred. It is possible that Late Prehistoric period habitation sites are under represented in the archaeological database, for some could be buried under alluvium. For at least part of this period, temporary Plains Village and Oneota campsites associated with the Pipestone Quarries (21PP2) should be present along rivers and streams.
Because ceramics are rare in the Southwest Riverine Region, it is difficult to associate components with ceramic-producing Woodland, Oneota, and Plains Village complexes. Artifacts collected during most surveys are almost exclusively lithic debitage. Gibbon and Hruby (1983) used cluster and discriminate analysis to generate tentative associations between samples of lithic debitage and particular functions, spaces, and time periods. Their analysis indicated that Archaic sites are widely distributed throughout the Rock River drainage in Minnesota. While the largest Archaic sites occur on terraces, bluffs, and particularly hilltops with panoramic views, they also occur on many other kinds of landforms. Hunting and butchering seem to have been the primary activities taking place at Archaic sites. Woodland and Mississippian sites (identified by the presence of Sioux Quartzite among the raw materials) could not be differentiated because of the absence of ceramics. Sites of this category are located on bluffs and terraces along permanent water courses, especially the Rock River. Although these were temporary rather than permanent settlements, a variety of activities apparently occurred there. Special activity sites ('lithic scatters') seem located in all topographic settings during all time periods.
Region 2 covers most of southwestern and south central Minnesota. It includes all of Big Stone, Blue Earth, Brown, Carver, Chippewa, Cottonwood, Faribault, Freeborn, Jackson, Lac Qui Parle, Le Sueur, Lyon, McLeod, Martin, Nicollet, Redwood, Renville, Scott, Sibley, Stevens, Swift, Watonwan, and Yellow Medicine counties and portions of Douglas, Grant, Kandiyohi, Lincoln, Meeker, Nobles, Otter Tail, Pipestone, Pope, Rice, Steele, Traverse, and Waseca counties (Figure 3.10). The region extends into northeastern South Dakota and north central Iowa (Anfinson 1997).
The interior topography of Region 2 is the typical swell and swale topography of a ground moraine. Hilly end moraines are found along northern, eastern, and southern edges. The major topographic features are the Minnesota River trench, which bisects the region southwest to northeast, and the scarp of the Prairie des Coteau highland in the west. Lake basins are numerous and vary greatly in size. All of the lakes are shallow, with none exceeding 10 m in depth. Most of the major rivers of the Prairie Lake region empty into the Minnesota River from the north or south. Exceptions are the Shell Rock River in the southeast, which flows south to the Cedar River in Iowa, and the south fork of the Crow River in the northeast and the headwaters of the Cannon River in the east central area, which flow east into the Mississippi River.
Soils are medium to fine textured prairie soils in the central and western part of the region, and fine to medium textured prairie border soils in the east. Climatically, precipitation ranges from 28 inches per year in the southeast to 22 inches per year in the northwest. Temperature ranges and growing season are about the same as those in Region 1, although it is cooler in the northwest. Lakes freeze over by early December and are ice-free by early April. Bedrock outcrops are rare in the Prairie Lake Region, especially deposits of good-quality rock for chipped stone tools. In the eastern part of the region, occasional outcrops of Paleozoic rocks near the confluence of the Blue Earth and Minnesota Rivers include some deposits of high quality chert. In the western part, there are outcrops of Sioux Quartzite in Cottonwood County. Granites and poorly consolidated Cretaceous rocks, such as shales, are found in the Minnesota River valley.
At the time of European-American settlement, the region was covered with tallgrass prairie. Trees were uncommon in the western part of the region, but there were narrow river-bottom forests and oak woods along the major river valleys and small patches of woodland in fire-protected areas (peninsulas, islands, isthmuses) at major lakes. The Minnesota River valley contained the principal wood resources for the western part of the region. The eastern part contained extensive Big Woods vegetation in the north and oak parkland in the south. The oak woodland was very patchy with large prairie openings becoming more numerous to the west. While the modern prairie-forest border was established by the beginning of Initial Woodland times, extensive development of Big Woods vegetation (elm, maple, basswood) did not take place until the contact period; oak woodlands were present all along the eastern edge of the region until about 300 years ago.
Bison was the dominant upland fauna for the Prairie Lake region in the Late Holocene period, with occasional large elk herds also present. White-tailed deer were found along the Minnesota River valley and in the eastern woodland fringe. The major faunal differences between the two southwestern prairie regions are due to the many shallow lakes in Region 2. These lakes promote extensive populations of aquatic mammals (e.g., muskrats), waterfowl, and fish. They also contain a rich floral assemblage that includes such edible plants as water lilies and cattails. Wild rice was present in Region 2, but it was not extensive; it was primarily limited to the Minnesota River valley and a few northern and eastern lakes. Upland floral resources include the prairie turnip, the ground plum, and acorns in the oak woods.
18.104.22.168 Prairie Lakes Site Locations by Archaeological Tradition and Cultural Complex
While Early Prehistoric period sites have not been professionally excavated in the Prairie Lakes Region, the remains of fluted (Clovis, Folsom) and Plano (Browns Valley, Agate Basin, Hell Gate) projectile points are relatively common. It is possible that sites associated with the Early and Middle Prehistoric periods are deeply buried in colluvium and river valley alluvium, especially in the Minnesota River valley. Middle Prehistoric period sites are not common, but some, such as 21BW5/7, 21CO2, 21LN2, and 21YM35, have been excavated. During the Prairie period, subsistence seems to have focused on pedestrian bison hunting. Anfinson (1987) has identified two phases in this period, the Cherokee (ca. 7000-5500 B.C.) And the Itasca (ca. 5500-3000 B.C.). Some early Middle Prehistoric period sites may be on lake bottoms, since most lakes shrank drastically during the Prairie period. In the Late Archaic, with the onset of wetter climatic conditions, subsistence diversified and the aquatic resources of the prairie lakes became an important subsistence focus. Anfinson has assigned Late Archaic components in this region to the Mountain Lake phase (ca. 3000-200 B.C.). These later Middle Prehistoric period sites, like Woodland sites, are located for the most part on islands and peninsulas on moderate to large-sized lakes; some villages are also located along the major rivers. As with later Woodland sites, winter villages are most likely located in wooded and sheltered areas in large river valleys.
The first widespread and readily visible evidence for Native American occupation of the Prairie Lakes Region occurs late in the Middle Prehistoric period with the appearance of the Woodland tradition. Initial Woodland Fox Lake phase and Terminal Woodland Lake Benton phase ceramics seem indigenous to the region, while external Havanoid, St. Croix, Onamia, Kathio, and Sandy Lake ceramics are also present. Woodland base camps (identified by the presence of cord-marked pottery) are common and occur, for the most part, on islands or peninsulas on moderate to large-sized lakes throughout the region. Deeply buried Woodland sites, especially winter villages, may be deeply buried in the Minnesota River valley flood plain and along its major tributaries. Temporary campsites and special activity sites ('lithic scatters') occur in fairly large numbers along the rivers and around the lakes of the region.
The most visible Late Prehistoric period site type in the Prairie Lakes Region is the large agricultural village site, most of which are located on intermediate terraces of the Minnesota and Blue Earth rivers. Late Prehistoric period phases in the region include the Plains Village (Cambria, Great Oasis) and Oneota (Blue Earth). Large Cambria villages are largely confined to the Minnesota River valley and Oneota villages to the Blue Earth River valley (Anfinson 1987). Small campsites and special activity sites from this period are scattered throughout the region, but especially on former Woodland sites on the islands and peninsulas of moderate to large-size lakes. Some deeply buried Late Prehistoric period sites could be present in the Minnesota River valley.
Contact period sites in the Prairie Lakes Region are primarily associated with the Yankton Dakota at the time of contact (ca. 1700), with the Wahpeton and Sisseton Dakota by the early 1800s, and with French, English, and American fur and wintering posts. Major Dakota villages were concentrated along the Minnesota River. Posts were concentrated for the most part along the upper Minnesota River between 1750-1800. By the early 1800s they were established by American traders at wooded locations in the interior.
Most of southeastern Minnesota is in this region, which includes Dodge, Fillmore, Goodhue, Houston, Mower, Olmsted, Wabasha, and Winona counties, and portions of Dakota, Freeborn, Rice, and Waseca counties. The region continues into the adjacent corners of Wisconsin and Iowa.
Region 3, which was not glaciated during the Late Wisconsin Ice Age, is characterized by stream-dissected terrain. No natural lakes are found in the region's interior, although valley bottom lakes, some quite large (e.g., Lake Pepin), are found along the Mississippi River. Three major river systems extend westward from the Mississippi into the region's interior: the Cannon, the Zumbro, and the Root.
The climate of Region 3 is the mildest in the state with a growing season of at least 160 days per year. The mean January high temperature is about 23 degrees F and the mean July high temperature is about 85 degrees F. Average annual precipitation varies between 28 and 30 inches. The western part of the region has medium textured prairie and prairie border soils, while the eastern part has fine textured forest and prairie soils formed on loess deposits over Paleozoic bedrock.
The Southeast Riverine region has extensive rock outcrops containing occasional primary and secondary lag deposits of high quality flaking materials. Chert concentrations are found not only along the Mississippi River valley, but also just below the surface in less-dissected areas in the western part of the region.
In Late Holocene times, forests of elm, ash, and cottonwood lined the river lowlands. Big Woods forests of maple, elm, and basswood occupied the uplands near the Mississippi River. Patches of oak groves in the prairie, often described as "oak barrens," were scattered across the western part of the region. The middle of the region was a more open prairie.
Late Holocene subsistence resources in this region included deer, elk, and scattered bison in the uplands and mussels, fish, and waterfowl in the rich bottom lands. Edible plants would have included water lilies and other aquatic flora, and upland plants like the prairie turnip. Extensive oak woods would have been a rich source of acorns. The region's favorable climate and extensive bottom lands fostered Terminal Woodland period horticulture.
Only a few Early Prehistoric period components are recorded in the Southeast Riverine Region archaeological database. As in other regions, they are primarily from surface collections of Paleoindian projectile points. A few fluted points have been reported from the extreme southeastern corner of the region and narrow, leaf-shaped Plano points (e.g., Agate Basin) have been located in those counties bordering the Mississippi River Valley. Patterns of site location for this period are unknown, but Plano points seem to be concentrated in areas away from the main valley. Many sites of all periods may be deeply buried in floodplain alluvium in this region. Some Archaic Middle Prehistoric period sites have been excavated (e.g., 21WB56, 21WN15, 21WN1). In general, Archaic complexes in at least the eastern portion of the region have their closest similarities with complexes in Wisconsin, as reflected in projectile point styles, such as Durst Stemmed, Osceola, and Raddatz Side Notched. An example of a site of this period whose components are deeply buried in an alluvial fan is King Coulee, where deposits extend at least 14 feet below the surface and cucurbit seeds have been dated to 1750 B.C. (Perkl 1996).
The appearance of Woodland sites in the latter portion of the Middle Prehistoric period is associated with the first archaeological evidence of intensive human occupation of the region. In contrast to Early Prehistoric period sites, Archaic and Woodland habitation sites seem concentrated along the Mississippi River in a variety of settings, including bluff tops, rock shelters, caves, terraces, and on stream deltas and knolls in the flood plain. Large Woodland village sites are rare in the western section of the region and in the interior in general. As in the Archaic, Woodland cultural connections are generally to the south and east, as is evident in the presence of Marion Thick-like, Havanoid, and Effigy Mound ceramics (Perkl 1996).
The most visible Late Prehistoric period sites are large, probably palisaded, horticultural villages and thousands of earthen burial mounds in the Red Wing area at the juncture of the Mississippi and Cannon rivers. This locality was a focus of Effigy Mound (ca. A.D. 550-1200) and Mississippian (ca. A.D. 1100-1300) activity. Both Woodland Effigy Mound and Mississippian tradition (the Silvernale phase) are concentrated in the Southeast Riverine Region. Large Silvernale phase horticultural villages are located on terraces above the floodplain in the Red Wing area. Some later Blue Earth-related Oneota components are also present in the Red Wing area. Late Prehistoric period sites are uncommon in the interior, except for later Orr phase Oneota village sites, which are concentrated along a tributary of the Root River.
The Orr phase villages of the southeastern corner of the region were occupied into the Contact period, as is evident from the presence of trade goods. At this time, the Ioway and possibly the Oto occupied the southern part of the Southeast Riverine Region, while the Santee Dakota had gained control of the northern part. After about 1750, historic Santee villages were established in the region at Red Wing, Winona, and elsewhere, generally in the same locations in which Euro Americans settlements were later established, that is, along the Mississippi River and its major tributaries. By the late 1600s, French explorers, missionaries, and traders were building posts along the Mississippi River. By the late 1700s, Anglo-American trading posts were established in the interior, particularly in the northern part of the region.
This region includes most of central and east central Minnesota. It contains Anoka, Benton, Cass, Chisago, Crow Wing, Hennepin, Isanti, Mille Lacs, Morrison, Ramsey, Sherburne, Stearns, Todd, Wadena, Washington, and Wright counties and portions of Becker, Dakota, Douglas, Kandiyohi, Kanabec, Meeker, Otter Tail, Pine, Pope, and Swift counties. The region could be extended into west central Wisconsin.
The topography of Region 4 is a patchwork of moraines, till plains, and outwash plains. Many lakes are found throughout the region, some reaching depths of more than 30 m (100 feet). The Mississippi River flows through the eastern and central parts of the region, with the St. Croix River forming the eastern boundary. Streams flowing west into the Red River drain the western part.
The climate varies throughout the region with greater precipitation in the east and higher temperatures in the south. Average annual precipitation ranges from 22 to 28 inches. Average January highs range from 12 to 24 degrees F, while average July highs range from 78 to 82 degrees F. The frost-free season lasts up to 160 days in the south and up to 140 days in the north. The northern part of the region may have been unsuitable for growing most precontact varieties of maize.
The soils of Region 4 reflect a diverse glacial and vegetal history. Most have medium to coarse textures, with prairie soils in the south and west and forest soil in the north and east. Bedrock outcrops are limited to occasional granite rock exposures in the region's center and eastern edge.
In the contact period, the vegetation in the southern and western parts of the region was dominated by Big Woods species with many large inclusions of prairie and oak woods. As in Region 2, the development of extensive Big Woods forests was a recent event and much of the Big Woods area was probably oak forest during the Late Holocene. Oak forest was still dominant in the east at the time of White settlement. The northern part of the region was a mixed deciduous-coniferous forest dominated by pine.
Late Holocene period subsistence resources of Region 4 would have included white-tailed deer throughout the region, small herds of bison and elk in the south and west, and beaver, bear, and even moose in the north and east. Fish and waterfowl would have been plentiful. Wild rice beds were extensive throughout most of the region. Acorns would have been an abundant food resource.
Early Prehistoric period settlement patterns are poorly known in the Central Lakes Deciduous region, because the artifactual evidence is limited and consists for the most part of projectile points in surface collections. However, lakes and major rivers were apparently a focus of activity. A concentration of fluted points has been found at the A. H. Anderson site (21AN8) on Howard Lake in Anoka County, and Plano points have been found throughout the region. More western stemmed forms (e.g., Hell Gap) seem concentrated in the western part of the region, and more eastern narrow leaf and broad concave base forms (e.g., Eden, Angostura) in the east.
Early Middle Prehistoric period site location patterns are also poorly known, but again sites seem associated with lakes and major rivers. Examples are Petaga Point (21ML11), which is near the Lake Mille Lacs - Rum River juncture, and the Pine City sites along the Snake River in Pine County (Bleed 1969; Caine 1974). Some Archaic burials (21GR4, 21PO3, 21PO13) may occur in the western part of the region.
A major shift in subsistence-settlement pattern and technology occurred in the region during the late Middle Prehistoric period. Ceramics and mound burial were adopted by ca. 200 B.C., The bow and arrow by ca. A.D. 500, and wild rice harvesting began to be intensified. As the broad-based hunting and gathering focus of the early Late Holocene was replaced by a more focal concentration on wild rice, habitation sites became larger, the human population may have increased dramatically in size, and people became less nomadic. Larger village sites are now concentrated on major lakes in contrast to the more diffuse pattern of the early Middle Prehistoric period. Smaller campsites and special activity sites are also found along major rivers and larger lakes. The most common Woodland ceramic complexes in the region are Malmo, Howard Lake, Brainerd, St. Croix, Onamia, Blackduck, Kathio, and Clam River, with Brainerd and Blackduck concentrated in the northwestern portion and the rest in the eastern.
A major concentration of population occurred in this region in the Late Prehistoric period (Gibbon 1994). Late Middle Prehistoric period pottery is found in numerous small sites in lakeshore settings throughout the region. In contrast, Late Prehistoric period villages, which are associated with Sandy Lake pottery in the northern half of the region, Oneota ceramics in the southern half, and Plains Village ceramics in the southwest corner, indicate large agglomerations of people. Large areas of the Central Lakes Deciduous Region were probably now used only for periodic resource procurement forays. In wild rice harvesting areas, villages are located near wild rice beds, such as stream inlets/outlets to lakes.
At contact, Santee Dakota groups controlled the eastern part of the Central Lakes Deciduous Region and the Yankton, Yanktonai, and other Dakota groups controlled the western part. By the mid-1700s, the Ojibwa had begun to move into the northern parts of the region and controlled this section of the Central Lakes Deciduous Region by the early 1800s. During this period of strife, much of the southern portion of the region remained unoccupied. In general, however, historic Indian village locations followed the Late Prehistoric period pattern and are often located near wild rice beds. By the late 1600s, French traders had entered the region and established posts on some major lakes and rivers, a pattern generally followed by later Anglo-American traders. The contact period as defined in this review ends with the establishment of the American settlement at Fort Snelling in 1821.
Located in central and northeastern Minnesota, Region 5 contains portions of Aitkin, Beltrami, Carlton, Cass, Clearwater, Crow Wing, Hubbard, Itasca, Kanabec, Koochiching, Lake, and St. Louis counties. The region incorporates much of what has been called the Headwaters Lakes region.
Region 5 is very similar to Region 4, especially regarding lake morphology. It can, however, be differentiated from Region 4 not only by vegetation and watersheds, but by topography. Hilly terminal moraines extend through the region's center and a variety of less rugged terrain of glacial origin cover the rest of the region, including ground moraines, outwash plains, and lake plains.
The Mississippi River traverses much of Region 5, flowing through or near several large lakes as it leaves its source in southeastern Clearwater County. The path of the Mississippi has changed significantly over the last 12,000 years, although by the beginning of the Late Holocene period it followed what is essentially its modern route. The western part of the region is drained by rivers flowing into the Red River, the northeast part is drained by the St. Louis River, which flows into Lake Superior, and the southeast part is drained by rivers flowing into the St. Croix River. Lake distribution in most of the region is very dense, with only the Glacial Lakes Upham-Aitkin plain lacking lakes (Figure 3.3). Many lakes are quite deep.
Average annual precipitation varies from 22 inches in the west to 28 inches in the east. Average July high temperatures are 74 to 78 degrees F, and average January high temperatures are 16 to 20 degrees F. The frost-free growing season ranges from 140 to less than 120 days. None of the region is suitable for intensive corn production. Lakes freeze over in mid-November and the ice is off the lakes by late April.
Soil types vary greatly but are generally coarse to medium textured forest soils. Fine textured soils and peat deposits are present in the northeastern part of the region on the Glacial Lakes Upham-Aitkin lakebed. Bedrock exposures are rare. Precambrian outcrops are found in the northeast, along with taconite deposits that contain some high quality flaking materials (e.g., chert, jasper, taconite).
Pine trees (white, jack, and red) dominated the Late Holocene period vegetation, although there were significant inclusions of deciduous trees (elm, maple, basswood, ash, oak, aspen, birch). Peat bog vegetation covers the glacial lake plains in the southeast. As in Region 4, Region 5 also contains extensive beds of wild rice.
Late Holocene faunal resources included deer, beaver, moose, and black bear. Fish and waterfowl were abundant in the many lakes and rivers. Wild rice was an important part of the economy in Terminal Woodland times.
As in the Central Lakes Deciduous Region, Early Prehistoric period settlement patterns are poorly understood, but sites near lakes seem to have been a focus of activity. A few fluted points have been found in the region, with the state's only excavated fluted point (a Folsom point) recovered at the Williams Narrows site (21IC23). Plano points are more common and appear to be concentrated in the northeast section of the region. An especially large concentration was found in the Reservoir Lakes area northwest of Duluth (Harrison et al. 1995). Late Paleoindian components may also be present at the Patrow site (21IC31) and at White Oak Point (21IC1).
Middle Prehistoric period Archaic and Woodland sites follow the pattern of the Central Lakes Deciduous Region in that village sites are located on major lakes and smaller sites on major rivers and around lakeshores. Very few sites occur in the interior. Archaic sites in this region have also been found along abandoned channels of the Mississippi River. An example is the large Eckstrom site (21BL40) on the south side of Lake Andrusia. The Itasca Bison site (21CE1) is an example of a buried bison kill site that dates to this period (ca. 7000 B.P.). The site is in a short creek that connects Lakes Elk and Itasca in Itasca State Park. Late Middle Prehistoric period subsistence and settlement pattern trends were similar to those of the Central Lakes Deciduous Region. Brainerd is the most common early ceramic. Burial mounds may not have been constructed in the region until ca. A.D. 700.
Late Prehistoric period Blackduck and Sandy Lake Woodland sites are numerous throughout the region, with a concentration in the Headwaters region in the central part of the Central Lakes Coniferous Region. Site location patterns are similar to those in the Central Lakes Deciduous Region. That is, with an increasing focus of wild rice harvesting, people become concentrated in larger villages often located on major lakes near wild rice beds.
At contact, the Santee Dakota were in the eastern part of the region and the Yanktonai in the west. By 1800, the Ojibwa controlled the entire region. Although French traders apparently did not establish posts in this region, Anglo-American fur posts were common throughout the region by the late 1700s, with a concentration along the Mississippi River.
Region 6 is in northwestern Minnesota. It contains Clay, Kittson, Norman, and Wilkin counties, and portions of Marshall, Pennington, Polk, Red Lake, Roseau, and Traverse counties. The region's eastern boundary in the south is defined by the Herman beach ridge and in the north by the Campbell beach ridge. The region extends into eastern North Dakota and southeastern Manitoba.
The flat plain of Glacial Lake Agassiz is dominant in Region 6. Bands of beach ridges are the only features of topographic relief. The soils tend to be fine in texture near the Red River, with coarser soils to the east on the beach ridges. While there are no bedrock outcrops in the Red River Valley region, lithic resources are available in beach ridge cobble deposits throughout the eastern part of the region. These are most accessible where rivers have cut through the beaches.
The entire region is within the Red River drainage basin. Many west-flowing rivers intersect the north-flowing Red. While lake basins are absent, at the time of European-American settlement there were many shallow marshes many of which dried up by late summer. Most of these marshes are now drained. Lake Traverse at the southern end of the region, while technically in the Red River trench, is excluded from the region as it has a better fit with the Prairie Lake Region.
Average annual precipitation ranges from 20 to 22 inches. The frost-free season is about 140 days in the south and less than 120 days in the north. July high temperatures average between 82 and 86 degrees F. January high temperatures are more regionally inconsistent, ranging from 24 degrees in the south to less than 12 degrees F in the north.
The vegetation at the time of European-American settlement was largely tallgrass prairie with river bottom forests (elm, ash, cottonwood) along the Red River and its major tributaries, and an irregular band of aspen-oak forest or aspen parkland along the northeastern edge.
The major faunal resource during the Late Holocene period was bison, which were present along the Red River in very large herds. Large elk herds were also reported in the contact period. Near the eastern edge of the region, deer were available and in the northeast, moose and even woodland caribou were present. Some fish and mussels were available in the Red River and its major tributaries. Waterfowl would have been seasonally abundant on the shallow marshes. Vegetal foods would have included various prairie species (e.g., prairie turnip), marsh plants (e.g., cattails), and berries and nuts from the riparian forests.
Since the Red River Valley Region was covered with water during the early Early Prehistoric period, no fluted points have been found in this area. However, Plano points have been recovered from the beach ridges in the eastern portion and may be present in the Agassiz Plain. An example of a beach ridge site is Greenbush (21RO11), which may have been a Late Paleoindian lithic workshop. Middle Prehistoric period Archaic sites are the first evidence of widespread occupation of the Red River Valley. Like late Early Prehistoric period components, many Archaic components are presumably buried by over a meter of alluvium in the river valley. Several such sites have been excavated (21NR9, 21NR29) (Michlovich 1986, 1987). The primary subsistence activity was bison hunting. Archaic habitation sites are concentrated on beach ridges.
Scattered evidence of Woodland habitation sites has been found at higher elevations along the Red River. Ceramics associated with these sites include Laurel, St. Croix, and Blackduck, with Laurel and Blackduck concentrated in the north and east parts of the region. These occupations may have been associated with seasonal bison hunting by otherwise woodland peoples. Earlier Woodland habitation sites, such as Lake Bronson (21KT1), may occur on beach ridges through the region. A similar pattern is present in the Late Prehistoric period in association with Sandy Lake ceramics, although some horticulture seems to have been practiced as well. There is a tendency for Late Woodland sites to be located on meander loops in the Red River Valley.
At contact, the western Dakota Teton and Yanktonai controlled the entire region except the far north, which was occupied by the Assiniboine. By 1800, the Yanktonai controlled the southern area and the Ojibwa the northern. British and American fur posts were located along the Red River as far south as Lake Traverse. Some posts were located inland along major tributaries.
Region 7 is in north central Minnesota. It contains Lake of the Woods and portions of Beltrami, Clearwater, Koochiching, Marshall, Pennington, Polk, Red Lake, and Roseau counties.
The region encompasses what is known as the eastern or Beltrami arm of Glacial Lake Agassiz. Peatlands cover most of the eastern two-thirds of the region. Poorly defined, discontinuous beach ridges occasionally interrupt the generally level topography. Remains of low recessional moraines are also present and reworked into boulder-strewn beach deposits by Lake Agassiz.
Outcrops of Precambrian bedrock occur around Lake of the Woods, along the Rainy River, and in eastern Koochiching County. Several meters of lake sediments and up to 30 m of glacial drift cover most of the bedrock. All but the western edge of the region contains peaty soils. The western soils are fine to coarse prairie soils.
Lakes and transecting rivers are generally absent from the region's core. Several large, shallow lakes that are remnants of Lake Agassiz (Red, Lake of the Woods, Thief, Mud) are present in the west. North-flowing rivers, which intersect the west-flowing Rainy River at the northern edge of the region, include the Roseau, Warroad, Rapid, Little Fork, and Big Fork. The southwestern part of the region is drained by the headwaters of rivers flowing west to the Red River.
Average annual precipitation ranges from 20 inches in the west to 26 inches in the east. Average July high temperatures are 74 to 78 degrees F and average January highs range from less than 12 degrees F in the west to 16 degrees F in the east. The annual frost-free season is less than 120 days. Lake of the Woods freezes over in late November and the ice is out by late April.
About 4000 years ago in response to a cooler, wetter climate, large areas of prairie vegetation turned into marshes and peatlands. The peatlands have a developmental progression from cattail marshes, to sedge meadows and heath bogs, and eventually to sphagnum moss. Bog conifers, such as spruce, tamarack, cedar, and balsam, are found on the uplands. The western third of the region could be characterized, in contact period times, as a wet prairie with patches of aspen and oak woodlands.
Late Holocene period game animals included deer, moose, caribou, beaver, and black bear. Some bison were present in the west. Fish were abundant in Red Lake, Lake of the Woods, and the major rivers. Waterfowl were seasonally abundant. Wild rice was present in the region, though not as abundant as in the lake regions to the south.
As in other regions, the Early Prehistoric period is the least well-known. Some Plano points have been found on the region's northern edge along the Rainy River, but fluted points have not been reported for the region. Sites dating to this period could be associated with glacial lake beaches. Middle Prehistoric period Archaic sites have different subsistence-settlement orientations, with those in the east (e.g., 21KC6) having a forest-oriented economy and those in the west (e.g., 21RO7) a bison-hunting focus. Like earlier sites, many Middle Prehistoric period Archaic sites could be buried in the interior peatlands.
Later Woodland sites were located at good fishing and hunting locations along major rivers and around the few lakes of the region. By at least 200 B.C., spring sturgeon fishing may have become a particularly important regional activity. The confluences of rivers, and rivers and lakes, at the edges of the region seem to have been preferred site locations, with the eastern Rainy River an especially important focus of activity. Many Archaic and Woodland components are probably deeply buried in alluvial strata at confluences along the Rainy River, as has been demonstrated by excavations at the McKinstry and Hannaford sites. Temporary campsites and special activity sites should be present in the interior peatlands of the region, but large habitation sites should be absent. Summer bison hunts may have occurred in the western part of the region.
A similar subsistence-settlement concentration continued into the Late Prehistoric period, although wild rice harvesting probably became important at the few wild rice lakes in the region. At contact, the Dakota, Cree, and Assiniboine were using the Northern Bog Region, but by 1750 it was controlled by the Ojibwa. French posts were on the Lake of the Woods and along the Rainy River in the early 1700s. Anglo-American posts were located in similar areas, but also along the Roseau River and at Red Lakes. Throughout the Contact period, activity was concentrated along the major lakes and rivers of the region.
This region is in northeastern Minnesota and occupies northern Cook, Lake, and St. Louis counties; a small portion extends into extreme eastern Koochiching County to include Nett Lake and the western arm of Rainy Lake. It also extends into the immediately adjacent portion of Ontario where Quetico Provincial Park is today.
Much of the terrain is rugged due to abundant outcrops of ice-scoured Precambrian bedrock. Lake distribution is dense. The lakes are formed in bedrock basins scooped out of the bedrock by glacial action. High quality chert flaking materials outcrop at various locations in the region associated with the Gunflint and Vermilion iron formations. Patches of thin, coarse to finely textured forest soils are found in most of the region with peaty soils in the far west. The Rainy River drains the west and central parts of the region and the Pigeon River drains the eastern part.
Average annual precipitation varies from 26 inches in the west to 30 inches in the northeastern corner. The frost-free season is less than 120 days. July high temperatures range from 80 degrees F in the west to less than 74 degrees F in the east. January high temperatures range from 12 degrees F in the west to 20 degrees F in the east. Lakes freeze over in mid-November and are free of ice by the beginning of May. Winter snowfalls average more than 70 inches in much of the region with snow covering the ground for more than 140 days.
In the Late Holocene, a coniferous forest dominated by spruce and pine covered the region, but there were many inclusions of birch and aspen. Animals would have included deer, moose, caribou, beaver, and bear. Many fish were available in the lakes and rivers, and waterfowl were seasonally abundant. Large stands of wild rice are not common, although a few are currently present in the southwest near Nett Lake and along the eastern edge of the region.
22.214.171.124 Border Lakes Site Locations by Archaeological Tradition and Cultural Complex
Plano points in this region appear to be associated with lithic quarries. No fluted points are recorded in the SHPO database. Late Paleoindian components in the region, which are considered part of the Lakehead complex in adjacent Ontario, would be part of a caribou hunting focus. This was one of the few areas in the state covered with an open tundra environment. The few recorded Archaic habitation sites in the region are associated with lakes and extensive copper working. Houska Point (21KC6), at the outlet of Rainy River on Rainy Lake, and the Fowl Lake site (21CK1), in the northeast corner of the region, contain Late Paleoindian lanceolate points as well as extensive Archaic deposits. Another site in which copper artifacts were manufactured is Pickerel Lake, just across the U.S.-Canadian border in Quetico Provincial Park (Steinbring 1970, 1975; Gibbon 1996). Like Late Paleoindian sites, Archaic sites tend to occur on peninsulas and islands in lakes, often near quarries.
As in most other regions of Minnesota, the first widespread occupation of the Border Lakes Region seems to have been associated with the Woodland tradition. The earliest ceramics are associated with the Laurel complex and the more recent with Blackduck, Selkirk, and Sandy Lake. In general, subsistence in this region remained a broad-based hunting-fishing-gathering adaptation until contact, with the more diffuse settlement locations associated with this kind of adaptation. Most Woodland sites are small and temporary. They increase in number from east to west across the Superior National Forest. This density cline may be related to the increasing availability of wild rice to the west. In the Late Prehistoric period, wild rice harvesting became important in some areas in the western part of the region, with larger sites concentrating in these areas. Concentrations of Woodland habitation sites are present at Nett Lake in the southwest and Lake Vermilion in the south central part of the region. In the northwest section of the region, most Woodland habitation sites, and especially Late Prehistoric period Woodland sites, appear to be located near major bodies of water. In general, Woodland base camps are located near resource concentrations, such as wild rice beds and good fishing spots.
During the Contact period, the Ojibwa controlled the Border Lakes Region, although the Assiniboine and Cree occupied the region at contact. French traders canoed down the major water routes along the northern edge of the region and constructed posts at larger lakes, such as Rainy Lake, Little Vermilion Lake, and Basswood Lake. By the early 1800s, the British had posts at these and other lakes in the region, such as Vermilion Lake and Moose Lake.
This region borders Lake Superior in extreme northeastern Minnesota. It contains the eastern edges of Carlton, Cook, Lake, and St. Louis counties. The region extends along Lake Superior into Ontario as far as Thunder Bay.
The Lake Superior shore has rocky cliffs with many small bays and points. Abundant Precambrian bedrock exposures are present, making waterfalls common as short streams and rivers descend the 900-1500 foot drop down the eastern slope of the highland to Lake Superior. In contrast, the southern tip of the region is the flat plain of Glacial Lake Duluth, which is drained by the St. Louis and Nemadji rivers. Fine to coarse textured forest soils are scattered between the rock outcrops, with fine silty and clayey soils on the glacial lake plain in the southwest. Copper nuggets can be found throughout the region.
The climate of the Lake Superior Shore is more moderate than adjacent regions to the west due to the "lake effect". The frost-free season averages between 120 and 140 days. July high temperatures are generally 75 degrees F or less. January highs range from 20 to 24 degrees F. Annual precipitation averages between 29 and 30 inches. Lake Superior occasionally freezes in very cold winters.
Late Holocene period vegetation in Region 9 was dominated by white pine with inclusions of birch and aspen. Populations of game animals were not dense during Late Holocene times. Some deer, moose, caribou, bear, and beaver were present. Brown trout lived in the major streams below cataracts and large fish populations were present in Lake Superior. Waterfowl were seasonally found along the shore. Wild rice was not abundant.
126.96.36.199 Lake Superior Shore Site Locations by Archaeological Tradition and Cultural Complex
Recorded archaeological remains are particularly scarce in the Lake Superior Shore Region, although representatives of early periods have been found along the Superior shore between the U.S.-Canadian border and Thunder Bay, Ontario. While no fluted or Plano points are recorded for the region in Minnesota, a number of Late Paleoindian sites, such as Cummings, have been excavated in the Thunder Bay area. Dated to ca. 8500 B.P., Cummings is a habitation and quarry site located on a glacial beach ridge. Early Prehistoric period sites should be associated with Lake Duluth and Lake Minong beaches, but these beaches are submerged in Minnesota and the sites may be buried or destroyed. Early Prehistoric period and early Archaic sites between the border and Thunder Bay generally occur on post-Minong beaches. Archaic remains in Minnesota consist mostly of a few copper tools. Early Middle Prehistoric period Old Copper sites could be associated with Houghton stage beaches. Archaic sites are generally associated with major waterways.
Since Woodland ceramics are rare and mounds are absent along the Superior shore from Duluth to Thunder Bay, it is likely that Woodland use of the shoreline was limited. The few sherds that have been found along the southern Canadian Lake Superior shore were on the southwest side of peninsulas or small bays near the mouths of large rivers. Woodland settlement was most likely along inland lakes and waterways. However, one would expect to find warm weather base camps at good fishing spots at river deltas along the Lake Superior shore. If they are present, Woodland sites in the Lake Superior Shore Region could be concentrated in the estuary of the St. Louis River at Duluth.
As in the Northern Bog and Border Lakes Regions, the Assiniboine and Cree probably controlled much of this region before being replaced by the Ojibwa in ca. 1700. French missionaries had contact with this region in the mid-1600s and French trading posts were established soon after, with Grand Portage as a major regional center. English and American traders re-used many of the French posts and travel routes, and constructed new posts, including one near Duluth.
In Phase 3, the Ecological Classification System (ECS) sections and subsections became available for Minnesota (Figure 3.11). ECS is a method to identify, characterize and delineate units of land with similar climatic, geological, physical and biological features that are significant for natural resource management (Hanson and Hargrave 1996). ECS replaced the archaeological regions (Section 3.4) as the regionalization scheme in Phase 3. The reasons for this change are discussed in Chapter 4. This section discusses the development and characteristics of the Minnesota ECS.
ECS is a continent-wide, nested hierarchical classification system, with six levels organized by decreasing orders of scale (Minnesota Department of Natural Resources [MN DNR] Website, http://www.dnr.state.mn.us/ecs/index.html). Mn/Model makes use of only the top three levels (Table 3.2). Each ECS Province is divided into several Sections, each Section is divided into several Subsections. Ecological units for each level are defined by the dominant environmental factors that affect ecosystem processes and functions at that particular scale.
An ad hoc interdisciplinary work group representing the MN DNR, US Forest Service (USFS), and the University of Minnesota collectively developed the Province, Section and Subsection maps, beginning in 1991. It was completed to the subsection level in 1997. Delineation of lower levels of the hierarchy continues. Development of the Minnesota system is being coordinated with a similar effort on a national scale within the USFS to assure consistency between agencies.
Continental climate patterns, regional physiography, pre-European settlement vegetation
Surficial geology, bedrock geology, pre-European settlement vegetation, regional climate patterns
Surficial geology, bedrock geology, pre-European settlement vegetation, general soils, hydrology and subregional climate patterns
There are three ECS provinces (Figure 3.11) in Minnesota, named after the dominant biomes. These are the tall-grass prairie (Prairie Parkland Province), the central hardwoods (Eastern Broadleaf Forest Province), and the boreal forest (Laurentian Mixed Forest Province). Variables used to distinguish between the provinces are presented in Table 3.3. Additional variables used to distinguish between Eastern Broadleaf Forest Province and Laurentian Mixed Forest Province were winter temperatures and length of growing season. The Eastern Broadleaf Forest Province has warmer winters and a longer growing season than the Laurentian Mixed Forest Province.
Total annual precipitation
% precipitation in winter
Average annual rainfall deficit
50 to 75 cm
17 to 25 cm
Eastern Broadleaf Forest
Deciduous woodlands and savannas, oak forests, and maple-basswood forests
55 to 85 cm
14 to 29%
2.5 to 6 cm
Laurentian Mixed Forest
Aspen, white spruce, black spruce and balsam fir, along with red and white pine
55 to 85 cm
14 to 29%
2.5 to 6 cm
Each of the Provinces is divided into two to five Sections (Figure 3.11). An example of the criteria for defining separate sections of the Laurentian Mixed Forest Province is provided in Table 3.4. The 10 Minnesota Sections were subdivided into 24 Subsections. An example of the criteria for defining separate Subsections of the Northern Minnesota Drift & Lake Plains Section is provided in Table 3.5.
Northern Minnesota & Ontario Peatlands
Post glacial lake basins
98 to 111 days
Northern Minnesota Drift & Lake Plains
Complex of ice contact, glaciofluvial and lacustrine deposits
111 to 131 days
Northern Superior Uplands
Precambrian igneous and metamorphic bedrock near or at the surface
100 to 123 days
Southern Superior Uplands
Post glacial lake basins; 30 to 90 m of ice contact and glaciofluvial material at surface, deeply dissected by rivers
121 to 135 days
Western Superior Uplands
30 to 90 m of ice contact and glaciofluvial material
97 to 140 days
Total annual precipitation
level to gently rolling lake plains, outwash plains, stagnation moraines and ground moraines
110 to 120 days
57.5 to 67.5 cm
Pine Moraines and Outwash Plains
end moraines, ground moraines and drumlin fields separated by outwash plains and channels
110 to 120 days
57.5 to 67.5 cm
St. Louis Moraines
rolling to very steep end moraines, stagnation moraines, ground moraines and pitted outwash plains
111 to 131 days
60 to 67.5 cm
post glacial lake basin and a small ground moraine of reworked lake deposits
92 to 115 days
60 to 67.5 cm
Detailed descriptions of the ECS subsections are provided in Chapter 8, along with the descriptions of the Phase 3 models for the subsections. Additional descriptions of the subsections can be found on the MN DNR Website, http://www.dnr.state.mn.us/ecs/index.html.
Against the backdrop of Minnesota’s environmental and cultural diversity, Anfinson’s (1988, 1990) nine archaeological regions (Figure 3.10) were considered a useful initial focus for constructing models of archaeological site location in the state for several reasons. First, they are hydrological regions based on the distribution of different types of surface water, in particular the presence or absence of lakes, lake morphology (bedrock or drift), and lake depth. Since water was a focus of precontact and contact period settlement, various types of water may be associated with different patterns of site location. Second, compared with vegetation, the position of lakes and rivers has been stable since the Early Holocene. Third, although the vegetation cover of the regions did shift radically before about 3500 years ago, most archaeological sites in the project database probably date to this more recent period. While special attention had to be paid to the location of Early Archaic and Paleoindian components, it was thought that predictive models based in part on the distribution of contact period vegetation would perform well for recorded sites.
ECS subsections, adopted in Phase 3, had several additional advantages. First, the ECS is an integrated multi-factor classification scheme, based on climate, geology, geomorphology, and historic vegetation. As such, it was developed using a comprehensive interdisciplinary approach. The subsections defined by ECS are mostly smaller and more environmentally homogeneous than the archaeological regions and subregions. Moreover, ECS is conducive to modeling at multiple scales. In the future, modeling can be extended to lower levels of the hierarchy that are currently being developed. Moreover, boundaries of the ECS units are based on natural features, not straight lines. Because it is being developed nationwide, its use provides the potential for consistency across states and between agencies. Finally, it was designed to facilitate the understanding of relationships between single environmental components within homogeneous environmental regions. This kind of analysis is similar in many respects to attempting to understand relationships between archaeological site locations and environmental factors within the same regions.
Irrespective of the regionalization scheme used, many aspects of the natural environment of Minnesota introduced difficulties into the modeling process. Among these are the large size of the state (80,000 sq miles) and the diversity of its present and past environments. At historic contact, vegetation ranged from coniferous forest in the northeast to a prairie in the west. In the past, the state's climate and environment were very dynamic. At first, the state was blanketed with a spruce forest, then by prairie grasslands, and eventually by a mosaic of mixed hardwoods, deciduous forest, and prairie. Changes in air mass, effective moisture, and amount of annual sunlight accompanied these shifts. Patterns of river down-cutting changed the ecology of river valleys, buried archaeological sites beneath many feet of sediment, and shifted lake levels. Lake in-filling altered the attractiveness of lakes as settlement areas, and the location of preferred habitation sites shifted. During the historic period, dam construction, lake and wetland drainage, mining, agriculture, and logging were among the many factors that severely altered the state's natural environment.
This dynamic aspect of Minnesota's natural environment through time meant that a decision had to be made about the modeling aims of the project during the formulation of its research design. In the predictive modeling literature in archaeology, two basic modeling aims have been identified and their differing goals clearly explicated. This dichotomy has been most explicitly discussed in Tomlin (1990:167-225), Warren (1990:90, 94-95), and Leusen (1996:181-185), but also see Kvamme (1992:22-23). These two aims are often called the Cultural Resource Management (CRM) approach and the Academic approach.
In the CRM approach, the governing aim is to describe regularities in the locations and patterns of known archaeological remains to predict where similar remains may be found. This approach produces reports and maps that describe the archaeological potential of various regions, so that an assessment of archaeological sensitivity can be made for planning purposes. The result of this kind of modeling should be more efficient labor-intensive surveying (Phase I survey) and, through better planning, a reduction in potential impacts to archaeological sites. A primary aim is to make an accurate prediction of archaeological sensitivity. As Leusen (1996:184) emphasizes, no "attempt need be made to explain correlations between site locations and aspects of the environment."
In the Academic approach, the governing aim is to reconstruct and understand the motives and causes that underlie such locations and patterns - in what archaeological remains can tell us about the people who produced them. The path to understanding past settlement choice involves explaining the variation found in the locations of known sites in terms of cultural and environmental conditions, site formation processes, and site circumstances.
These two modeling aims lead to different approaches to modeling past settlement distributions and to different procedures for interpreting, testing, and enhancing the results. The CRM approach is interested in location factors that have a significant statistical correlation to some set of sites. The models produced by this approach are based for the most part on the relationship of known surface sites with present environmental variables. While this non-historical approach is well suited for modeling the present distribution of sites, its models are usually unsuited to "making sense" of past human behavior, since the present environment often bears little relation to past environments (Brandt et al. 1992). The Academic approach is interested in locational choice factors involved in the original process of site selection. This approach is best suited for modeling the past distribution of communities and, consequently, entails a huge effort in reconstructing past cultural and environmental conditions and processes. Here the choice of independent variables is explicitly based on understanding the settlement location factors important to the prehistoric population studied - and it is an aim of the modeling attempt to increase the profession’s understanding of those factors.
In cases where the environment was stable for the period being modeled (e.g., Kvamme 1985:215), the distinction between the two aims can be blurred. This is obviously not the case in Minnesota. If the modeling aim of Mn/Model had been to increase our understanding of past locational choice factors, then a major and necessary goal of the project would have been the reconstruction of those factors. This would have involved a major effort, for, as stressed above, in many cases more recent and often dramatic changes in the landscape mask the reasons for the preferred settlement locations of earlier periods. Since Mn/Model has cultural resource management objectives, and because of the very large effort involved in establishing the project databases, the decision was made to base it on a CRM approach. Nonetheless, geomorphologic and paleoclimatic databases were incorporated into the GIS that have the potential, as future enhancements, to provide a base for understanding past settlement choice.
Another aspect of Minnesota's natural environment that affected the modeling process is its areal diversity. Because of this diversity, settlement locations differed among biomes, such as the prairie and northern hardwood forests. It was considered unlikely that one master model would effectively predict site locations across the entire state. This was the underlying rationale for the construction of regional and sub-regional models. Despite this diversity, the analysis in Chapter 8 points to underlying factors in site location such as elevated areas near large lakes or perennial streams or rivers that transcend these regions.
The nature of Minnesota's precontact cultures and of their archaeological record also provided potential problems in the modeling process. For example, during the 12,000-year history of the state's precontact Native Americans, populations were sparse compared with Midwestern states to the south, such as Illinois, Indiana, and Ohio. This meant that not all preferred locations were occupied in the past. These populations were organized, too, in a variety of subsistence-settlement patterns, all of which included some degree of seasonal mobility. As a result, preferred site locations were diverse at any one time, and their nature shifted through time. In addition, many of their activities, particularly before about 1500 B.C., were not the kind that resulted in the formation of the types of sites usually recorded in site databases, and a variety of natural processes, as mentioned above, buried many sites. The Mn/Model archaeological database was based, then, on a potentially difficult archaeological record for the precontact period. Because of these difficulties, sites from the last 3000 years of record dominate the archaeological database. Consequently, the models developed do a better job predicting these sites than predicting the older, less numerous sites.
It is this environmental and archaeological background that set the parameters for model development. Chapters 4 through 8 outline the development and assessment of the predictive models.
Aaseng, N.E., J.C. Almendinger,
R.P. Dana, B.C. Delaney, H.L. Dunevitz, K.A. Rusterholz, N.P. Sather, and D.S.
1993 Minnesota’s Native Vegetation: A Key to Natural Communities. Minnesota Department of
Natural Resources, Natural Heritage Program, Biological Report 20:1-111. St. Paul.
1984 New World Mammoth Distribution. In Quaternary Extinctions: A Prehistoric Revolution,
edited by P. S. Martin and R. G. Klein, pp. 90-108. University of Arizona Press, Tucson.
Amundson, D.C., and H.E. Wright,
1979 Forest Changes in Minnesota at the End of the Pleistocene. Ecological Monographs 1979:
1976 Minnesota Municipal and County Highway Archaeological Reconnaissance Study, 1975
Annual Report. Minnesota Historical Society, St Paul.
1984 Minnesota Municipal and County Highway Archaeological Reconnaissance Study, 1983
Annual Report. Minnesota Historical Society, St Paul.
1987 The Prehistory of the Prairie Lake Region in the Northeastern Plains. Unpublished Ph.D.
dissertation, Department of Anthropology, University of Minnesota, Minneapolis.
1988 Annual Report Minnesota Municipal County Highway Archaeological Reconnaissance
Study. Appendix 3. Minnesota Historical Society, St. Paul.
1990 Archaeological Regions in Minnesota and the Woodland Period. In The Woodland Tradition
in the Western Great Lakes: Papers Presented to Elden Johnson, edited by G. E. Gibbon, pp.
135-166. University of Minnesota Publications in Anthropology No. 4. Department of
Anthropology, University of Minnesota, Minneapolis.
1997 Southwestern Minnesota Archaeology: 12,000 Years in the Prairie Lake Region.
Minnesota Prehistoric Archaeology Series No. 14, Minnesota Historical Society, St. Paul.
n.d. Maps for Minnesota Context Studies. On file at the Minnesota Historic Preservation Office,
Bailey, R.M., And G.R. Smith
1981 Origin and Geography of the Fish Fauna of the Laurentian Great Lakes Basin. Canadian
Journal of Fisheries and Aquatic Sciences 38:1539-1561.
Bartlein, P.J., And C. Whitlock
1993 Paleoclimatic Interpretation of the Elk Lake Pollen Record. In Elk Lake, Minnesota: Evidence
for Rapid Climate Change in the North-Central United States, edited by J.P. Bradbury and W.
E. Dean, pp. 275-293. Special Paper 276, Geological Society of America, U.S. Geological
1979 Some Trends and Traditions in Woodland Cultures of the Quad-State Region in the Upper
Mississippi River Basin. The Wisconsin Archeologist 60:47-82.
1989 Northwoods Wildlife: A Watcher’s Guide to Habitats. Northwoods Press, Minocqua.
1978 The French at Lake Pepin: An Archaeological Survey for Fort Beauharnois, Goodhue
County, Minnesota. Minnesota Historical Society, St. Paul.
1969 The Archaeology of Petaga Point: The Preceramic Component. Minnesota Historical
Society, St. Paul.
Bonnichsen, R., and K.L. Turnmire
1991 Clovis: Origins and Adaptations. Center for the Study of the First Americans, Oregon State
Bradbury, JP, and W.E. Dean (editors)
1993a Elk Lake, Minnesota: Evidence for Rapid Climate Change in the North-Central United
States. Special Paper 276, Geological Society of America, US Geological Survey, Denver.
Bradbury, JP, and W.E. Dean
1993b Holocene Climate and Limnologic History of the North-Central United States as Recorded in
the Varved Sediments of Elk Lake, Minnesota: A Synthesis. In Elk Lake, Minnesota: Evidence
for Rapid Climate Change in the North-Central United States, edited by J. P. Bradbury and
W. E. Dean, pp. 309-328. Special Paper 276, Geological Society of America, US Geological
Brandt, R. W., B. J. Groenewoudt,
and K. Kvamme
1992 An Experiment in Archaeological Site Location: Modeling in the Netherlands Using GIS
Techniques. World Archaeology 24(2):268-282.
1996 Climatic Episodes (Holocene). In An Encyclopedia of North American Prehistory, edited by
G. Gibbon. Garland Press, New York. In press.
Bryson, R.A., DA Baerreis, and W.M.
1970 The Character of Late-Glacial and Postglacial Climatic Changes. In Pleistocene and Recent
Environments of the Central Great Plains, edited by W. Dort, Jr. and J. K. Jones, Jr., pp. 53-
72. Special Publication 3, University of Kansas Press, Lawrence.Caine, C.A.H.
1974 The Archaeology of the Snake River Region. In Aspects of Upper Great Lakes
Anthropology, edited by E. Johnson, pp. 55-63. Minnesota Prehistoric Archaeology Series, No.
11. Minnesota Historical Society, St. Paul.
1990 Fire and Climate Change During the Last 750 Years in Northwestern Minnesota. Ecological
1993 Fire, Climate Change, and Forest Processes During the Past 2000 Years. In Elk Lake,
Minnesota: Evidence for Rapid Climate Change in the North-Central United States, edited
by J. P. Bradbury and W. E. Dean, pp. 295-308. Special Paper 276, Geological Society of
America, US Geological Survey, Denver.
1983 Chronology of Lake Agassiz Drainage to Lake Superior. In Glacial Lake Agassiz, edited by
J.T. Teller and L. Clayton, pp. 291-307. Geological Association of Canada Special Paper 26.
Department of Geology, Memorial University of Newfoundland, St. John's.
1966 The Prehistoric Animal Ecology and Ethnozoology of the Upper Great Lakes.
Anthropological Papers No. 29, Museum of Anthropology, University of Michigan, Ann Arbor.
1980 The Prairie World. University of Minnesota Press, Minneapolis.Daniel, G., and J. Sullivan.
1981 The Northwoods of Michigan, Wisconsin, Minnesota, and Southern Ontario. Sierra Club
Books, San Francisco.
1978 Archaic Subsistence in Southwestern Minnesota: The View from Granite Falls.
Unpublished Masters thesis, Department of Anthropology, University of Minnesota, Minneapolis.
1984 Oneota Settlement Patterns in the Blue Earth River Valley. Unpublished Ph.D. dissertation,
Department of Anthropology, University of Minnesota, Minneapolis.
1990a Outline of HIstoric Contexts for the Prehistoric Period (ca. 12,000 BP - A.D. 1700)..
Institute for Minnesota Archaeology Reports of Investigations 37. Prepared for the State Historic Preservation Office, St. Paul.
1990b Historic Context Outlines: The Contact Period (ca. 1630 A.D. - 1820 A.D.). Institute for Minnesota Archaeology Reports of
Investigations 39. Prepared for the State Historic Preservation Office, St. Paul.
1967 Mastodons, Their Geologic Age and Extinction in Ontario, Canada. Canadian Journal of
Earth Sciences 4:663-675.
Drexler, C.W., W.R. Farrand, and
1983 Correlation of Glacial Lakes in the Superior Basin with Eastward Discharge Events from Lake
Agassiz. In Glacial Lake Agassiz, edited by J. T.Teller and Lee Clayton, pp. 309-330. Geological
Association of Canada Special Paper 26. Department of Geology, Memorial University of
Newfoundland, St. John’s.
Dyke, A.S., and V.K. Prest.
1986 Late Wisconsinan and Holocene History of the Laurentide Ice Sheet. Géographie Physique et
Fagan, B. F.
1987 The Great Journey: The Peopling of Ancient America. Thames and Hudson, New York.
1996 Late Paleo-Indians of Minnesota and Vegetation Changes from 10,5000-8000 BP.
Unpublished M.A. thesis, Interdisciplinary Archaeological Studies Program, University of
Folwell, W. W.
1956 A History of Minnesota, Vol. 1. Minnesota Historical Society, St. Paul.
Forester, R.M., L.D. DeLorme, and
1987 Mid-Holocene Climate in Northern Minnesota. Quaternary Research 28:263-272.
1979 The Mississippian Occupation of the Red Wing Area. Minnesota Historical Society, St. Paul.
1983 The Blue Earth Phase of Southern Minnesota. Journal of the Iowa Archeological Society
1991 The Middle Mississippian Presence in Minnesota. In Cahokia and the Hinterlands: Middle
Mississippian Cultures of the Midwest, edited by T.E. Emerson and R.B. Lewis, pp. 207-220.
University of Illinois Press, Urbana and Chicago.
1993 The Middle Missouri Tradition in Minnesota: A Review. In Prehistory and Human Ecology of
the Western Prairies and Northern Plains, edited by J. Tiffany. Plains Anthropologist Memoir
1994 Cultures of the Upper Mississippi River Valley and Adjacent Prairies in Iowa and Minnesota. In
Plains Indians, A.D. 500-1500: The Archaeological Past of Historic Groups, edited by Karl
Schlesier, pp. 128-148. University of Oklahoma Press, Norman.
1996a Modeling Paleoindian and Early Archaic Settlement in Minnesota. Paper presented at the 54th
Annual Plains Anthropological Conference. Iowa City, Iowa.
1996b Old Copper in Minnesota: A Review. In Great Lakes Archaeology: Papers Presented to
Ronald Mason, edited by R. Birmingham. Special issue of The Wisconsin Archeologist. In press.
Gibbon, G.E. (editor)
1990 The Woodland Tradition in the Western Great Lakes. University of Minnesota Publications in
Anthropology, No. 4. Minneapolis.
1982 Oneota Studies. University of Minnesota Publications in Anthropology, No. 1. Minneapolis.
Gibbon, G.E., and C.A.H. Caine.
1980 The Middle to Late Woodland Transition in Eastern Minnesota. Midcontinental Journal of
Gibbon, G.E., and C.A. Dobbs
1991 The Mississippian Presence in the Red Wing Area. In New Perspectives on Cahokia: Views
from the Periphery, edited by J.B. Stoltman, pp. 281-305. Monographs in World Archaeology,
No. 2. Madison.
Gibbon, G.E., and H. Hruby
1983 First-Step Settlement Subsistence Models for the Rock River Drainage in Southwestern
Minnesota. In Prairie Archaeology, edited by Guy Gibbon, pp. 131-150. Publications in
Anthropology No. 3. University of Minnesota, Minneapolis.
1982 Where Two Worlds Meet: The Great Lakes Fur Trade. Minnesota Historical Society, St.
1992 The Grand Portage Story. Minnesota Historical Society Press, St. Paul.
1957 The Development of Peatlands. Quarterly Review of Biology 32:145-16
1981 An Ecological and Paleoecological Study of the Vegetation in the Big Woods
Region of Minnesota. Unpublished Ph.D. dissertation, University of Minnesota, Minneapolis.
1983 Chronology and Dynamics of Vegetation Change in the Prairie-Woodland Region of Southern
Minnesota, USA. New Phytologist 93:311-350.
1984 Fire and Other Factors Controlling the Big Woods Vegetation of Minnesota in the Mid-
Nineteenth Century. Ecological Monographs 54:291-311.
1980 Bison and Man in North America. Canadian Journal of Anthropology 6:581-596.
Hall, S. P.
1987 Fort Snelling, Colussus in the Wilderness. Minnesota Historical Society, St. Paul.
Hanson, D.H. and B.C. Hargrave
1996 Development of a Mulitilevel Ecological Classification System for the State of Minnesota.
Environmental Monitoring and Assessment 39:75-84. Kluwer Academic Publishers.
Harington, C.R., and A.C. Ashworth.
1986 A Mammoth (Mammuthus primigenius) Tooth from Late Wisconsin Deposits Near Embden,
North Dakota, and Comments on the Distribution of Woolly Mammoths South of the Wisconsin
Ice Sheets. Canadian Journal of Earth Sciences 23:909--918.
Harrison, C., E. Redepenning, C.L.
Hill, G.(Rip) Rapp, Jr., S.E. Aschenbrenner, J.K. Huber, and S.C. Mulholland
1995 The Paleo-Indian of Southern St. Louis Co., Minnesota. Monograph 4. Interdisciplinary
Archaeological Studies, University of Minnesota. Kendall/Hunt Publishing Co., Dubuque.
Hocutt, C.H., and E.O. Wiley (editors)
1986 The Zoogeography of North American Freshwater Fishes. Wiley-Interscience, New York.
1974 The Pedersen Site (21LN2, Lincoln County, Minnesota). Unpublished Master's thesis,
Department of Anthropology, University of Nebraska, Lincoln.
1969 The Prehistoric Peoples of Minnesota. Minnesota Historical Society, St. Paul. (Revised in
1978 and 1987).
1988 The Prehistoric Peoples of Minnesota. Revised third edition. Minnesota Prehistoric
Archaeology Series, No. 3. Minnesota Historical Society Press, St. Paul.
Kohler, T.A., and S.C. Parker.
1986 Predictive Models for Archaeological Resource Location. In Advances in Archaeological
Method and Theory, Vol. 9, edited by M. B. Schiffer, pp. 397-452. Academic Press, New York.
Kurtén, B., and E. Anderson
1980 Pleistocene Mammals of North America. Columbia University Press, New York.
1985 Determining Empirical Relationships Between the Natural Environment and Prehistoric Site
Locations: a Hunter-Gatherer Example. In For Concordance in Archaeological
Analysis:Bridging Data Structure, Quantitative Technique, and Theory, edited by C. Carr, pp.
208-238. Westport Press, Kansas City.
1990 The Fundamental Principles and Practice of Predictive Archaeological Modeling. In
Mathematics and Information Science in Archaeology: A Flexible Framework, edited by A.
Voorrips, pp. 257-295. Studies in Modern Archaeology, vol. 3. Holos-Verlag, Bonn, Germany.
1992 A Predictive Site Location Model on the High Plains: An Example with an Independent Test.
Plains Anthropologist 37:19-40.
1994 Geology of the Lake Superior Region. Geoscience Press, Phoenix.
1980 The Boreal Ecosystem. Academic Press, New York.
Leusen, P. M. van.
1996 GIS and Locational Modeling in Dutch Archaeology: A Review of Current Approaches. In New
Methods, Old Problems: Geographic Information Systems in Modern Archaeological
Research, edited by H. Maschner, pp. 177-197. Occasional Paper No. 23, Center for
Archaeological Investigations, Southern Illinois University, Carbondale.
1976 The Archaeology of the Smith Site: A Study of the Ceramics and Culture History of
Minnesota Laurel and Blackduck. Unpublished Ph.D. dissertation, Department of Anthropology,
University of Wisconsin-Madison.
1994 Lanceolate Points from the Lake Agassiz Region of Northwestern Minnesota. Unpublished
Master’s thesis, Department of Anthropology, University of Illinois at Urbana-Champaign.
1974 The Original Vegetation of Minnesota. Compiled from U.S. General Land Office Survey
Notes. United States Department of Agriculture, North Central Forest Experiment Station, St.
1967 Prehistoric Overkill. In Pleistocene Extinctions: The Search for a Cause, edited by P.S.
Martin and H. Wright, Jr., pp. 75-120. Yale University Press, New Haven.
Martin, P.S., and R.G. Klein (editors)
1984 Quaternary Extinctions: A Prehistoric Revolution. University of Arizona Press, Tucson.
1981 Great Lakes Archaeology. Academic Press, New York.
1983 River Warren, the Southern Outlet of Glacial Lake Agassiz. In Glacial Lake Agassiz, edited by
J. T. Teller and L. Clayton, pp. 231-244. Geological Association of Canada Special Paper 26.
Department of Geology, Memorial University of Newfoundland, St. John’s.
1986 The Archaeology of the Canning Site. The Minnesota Archaeologist 45(1):39-66.
1987 The Archaeology of the Mooney Site (21NR29). The Minnesota Archaeologist 46(2):39-66.
1983 The Red River Valley in the Prehistory of the Northern Plains. Plains Anthropologist 28:23-
31. Minnesota Department of Natural Resources
1995 A Guide to Minnesota’s Scientific and Natural Areas. Minnesota Department of Natural
Resources, Section of Wildlife, Scientific and Natural Areas Program, St. Paul. Minnesota
1981 Minnesota Statewide Archaeological Survey Summary: 1977-1980. Minnesota Historical
Society, St. Paul. Nute, G. L.
1978 Caesars of the Wilderness: Medard Chouart, Sieur des Groseilliers and Pierre Esprit
Radisson, 1618-1710 (reprint of 1943). Minnesota Historical Society, St. Paul.
Ojakangas, R.W., and C.L. Matsch.
1982 Minnesota’s Geology. University of Minnesota Press, Minneapolis.
1996 The King Coulee Site. Unpublished Masters Thesis, Interdisciplinary Archaeological Studies
Program, University of Minnesota.
Pettipas, L.F., and A.P. Buchner.
1983 Paleo-Indian Prehistory of the Glacial Lake Agassiz Region in Southern Manitoba, 11,500 to
6500 B.P. In Glacial Lake Agassiz, edited by J.T. Teller and L. Clayton, pp. 421-451.
Geological Association of Canada Special Paper 26. Department of Geology, Memorial University
of Newfoundland, St. John’s.
1993 A Time-Space Model for the Distribution of Shoreline Archaeological Sites in the Lake
Superior Basin. Geoarchaeology 8(2):87-107.
Phillips, J.L., and J.A. Brown (editors)
1983 Archaic Hunters and Gatherers in the American Midwest. New World Archaeological
Record. Academic Press, New York.
1988 The World of Northern Evergreens. Cornell University Press, Ithaca.
1991 After the Ice Age: The Return of Life to Glaciated North America. University of Chicago
Press, Chicago. Pike, Z.
1965 The Expeditions of Zebulon Montgomery Pike to the Headwaters of the Mississippi
Through Louisiana Territory, and in New Spain, During the Years 1805-6-7 (Reprinted from
the original 1810, with commentary and other additions by Elliott Coues). Ross and Haines,
1971 The Itasca Bison Kill Site: An Ecological Analysis. Minnesota Historical Society, St. Paul.
1970 Preliminary Statements on the Houska Point Site. Minnesota Archaeology Newsletter 15:4-5.
Department of Anthropology, University of Minnesota, Minneapolis.
1974 The Preceramic Archaeology of Northern Minnesota. In Aspects of Upper Great Lakes
Anthropology: Papers in Honor of Lloyd A. Wilford, edited by E. Johnson, pp. 64-73.
Minnesota Historical Society, St. Paul.
1975 Taxonomic and Associational Considerations of Copper Technology During the Archaic
Tradition. Unpublished Ph.D. dissertation, Department of Anthropology, University of Minnesota,
Stewart, K.W., and C.C. Lindsey
1983 Postglacial Dispersal of Lower Vertebrates in the Lake Agassiz Region. In Glacial Lake
Agassiz, edited by J.T. Teller and L. Clayton, pp. 391-419. Geological Association of Canada
Special Paper 26. Department of Geology, Memorial University of Newfoundland, St. John’s.
1973 The Laurel Culture in Minnesota, Minnesota Prehistoric Archaeology Series 8, Minnesota
Historical Society, St. Paul Teller, J.T., and L. Clayton (editors)
1983 Glacial Lake Agassiz. Geological Association of Canada Special Paper 26. Department of
Geology, Memorial University of Newfoundland, St. John’s.
Teller, J.T., and L.H. Thorleifson
1983 The Lake Agassiz-Lake Superior Connection. In Glacial Lake Agassiz, edited by J. T. Teller
and L. Clayton, pp. 261-290. Geological Association of Canada Special Paper 26. Department of
Geology, Memorial University of Newfoundland, St. John’s.
1995 Minnesota’s Natural Heritage: An Ecological Perspective. University of Minnesota Press,
Tomlin, C. C.
1990 Geographic Information Systems and Cartographic Modeling. Prentice-Hall, New York.
Treur, A. S.
1994 Ojibwe-Dakota Relations: Diplomacy, War and Social Union, 1679-1862. Unpublished
Master’s thesis, History Department, University of Minnesota, Minneapolis. Van Zant, K.
1979 Late Glacial and Postglacial Pollen and Plant Macrofossils from Lake West Okoboji,
Northwestern Iowa. Quaternary Research 12:358-380.
1990 Predictive Modeling in Archaeology: A Primer. In Interpreting Space: GIS and Archaeology,
edited by K.M.S. Allen, S. W. Green, and E.B.W. Zubrow, pp. 90-111. Taylor and Francis,
Watts, W.A., and R C. Bright.
1968 Pollen, Seed, and Mollusk Analysis of a Sediment Core from Pickerel Lake, Northeastern
South Dakota. Geological Society of America Bulletin 79:855-876.
Webb, T., III.
1981 The Past 11,000 Years of Vegetational Change in Eastern North America. BioScience 31:501-
Webb, T., III, E.J. Cushing, and
H.E. Wright, Jr.
1983 Holocene Changes in the Vegetation of the Midwest. In Late-Quaternary Environments of
the United States, vol. 2, The Holocene, edited by H. E. Wright, Jr., pp. 142-165. University of
Minnesota Press, Minneapolis.
Whitlock, C., P.J. Bartlein, and
1993 Vegetation History of Elk Lake. In Elk Lake, Minnesota: Evidence for Rapid Climate
Change in the North-Central United States, edited by J. P. Bradbury and W. E. Dean, pp. 251-
274. Special Paper 276, Geological Society of America, U.S. Geological Survey, Denver.
1941 A Tentative Classification of the Prehistoric Cultures of Minnesota. American Antiquity 6:231-
1945 Three Village Sites of the Mississippi Pattern in Minnesota. American Antiquity 11:32-40.
1955 A Revised Classification of the Prehistoric Cultures of Minnesota. American Antiquity 21:130-
1960 The First Minnesotans. In Minnesota Heritage, edited by L.M. Brings, pp 40-79. T.S.
Wright, H.E., Jr.
1971 Retreat of the Laurentide Ice Sheet from 14,000 to 9,000 Years Ago. Quaternary Research
1972a Quaternary History of Minnesota. In Geology of Minnesota: A Centennial Volume, edited
by P. K. Sims and G. B. Morey, pp. 515-545. Minnesota Geological Survey, St. Paul.
1976a Ice Retreat and Revegetation of the Western Great Lakes Area. In Quaternary Stratigraphy
of North America, edited by W. C. Malaney, pp. 119-132. Dowden, Hutchison, and Ross,
1976b The Dynamic Nature of Holocene Vegetation. Quaternary Research 6:581--596.
Wright, H.E., Jr., and P.H. Glaser
1983 Postglacial Peatlands of the Lake Agassiz Plain, Northern Minnesota. In Glacial Lake Agassiz,
edited by J. T. Teller and L. Clayton, pp. 375-389. Geological Association of Canada Special
Paper 26. Department of Geology, Memorial University of Newfoundland, St. John’s.
1995 Eastern Deciduous Forest. University of Minnesota Press, Minneapolis.
Ziebarth, M. and A. Ominsky
1970 Fort Snelling, Anchor Post of the Northwest. Minnesota Historical Society, St. Paul.
1952 The Lakes of Minnesota: Their Origin and Classification. Minnesota Geological Survey
Bulletin 35, St. Paul.
The Mn/Model Final Report (Phases 1-3) is available on CD-ROM. Copies may be requested by visiting the contact page.
Mn/Model was financed by the Minnesota Department of Transportation using funds set aside by the Federal Highway Administration's Intermodal Surface Transportation Efficiency Act.
The Mn/Model process and the predictive models it produced are copyrighted by the Minnesota Department of Transportation (MnDOT), 2000. They may not be used without MnDOT's Consent.