Early prehistory: 70,000-10,000 years ago
This section of History covers the early prehistory of grizzly bears in North America, from roughly 70,000 to 10,000 years before the present. More specifically, the focus of this section is on changes in distributions of grizzly bears and the environments in which they lived during the late Pleistocene (the last Ice Age) and early Holocene (our contemporary Age). I don't address ancestral diets, as such, which I cover in another section devoted to the topic (see Ancestral diet). Even so, I present a lot of material here, probably because I am so interested in this critical transitional period between the harsh and cold conditions of the Ice Age to the, in many ways, more benevolent conditions of our current warm epoch. Yet many species of large mammals went extinct during this transition. The causes are debated, but there is no doubt about how much and how rapidly the world changed, and the extent to which these changes would have challenged all of the existing species to adapt in whatever ways they could. Clearly, many did not succeed and, of those, most were the largest-bodied of all (the megaherbivores and megacarnivores). But grizzlies seemed to flourish, probably because many of their competitors and predators disappeared. On the other hand, grizzlies increasingly shared this world with another extremely successful and rapidly expanding omnivore--humans. I cover early relations in North America between the two omnivores--grizzlies and humans--in a separate section because of the multi-faceted and complex nature of the topic (see The human factor).
In what follows, I start with the arrival of grizzlies in North America, the nature of the world within which they lived during the late Ice Age, the rapid changes that occurred at the terminal cusp of this Age, ending with the ensuing changes in distributions of brown bears.
Up until the early 2000s most scientists thought that grizzlies had not arrived in the middle part of North America until after the last continental Ice Sheets had melted enough to allow passage of bears from Beringia (which included all of modern-day Alaska) through an ice-free corridor along the eastern edge of the Rocky Mountains (see the maps below)--probably around 13,000 years ago (all of the ages here are corrected for biases in radiocarbon age). However, the discovery of grizzly bear remains near Edmonton, Alberta, dating to roughly 32,000 (or 32k) years ago turned this assumption on its head. Recent analyses of genetic material from North American grizzly bears have reinforced the idea that grizzlies arrived much earlier, probably during the last opening in the Alberta ice-free corridor, prior to closure at the height of the last Ice Age, roughly 55,000-18,000 years ago. Recent research by Chris Stokes and his colleagues suggests that a long-lasting opening between the Crodilleran (to the west) and Laurentide (to the east) continental ice sheets occurred between 80k and 65k years ago, with a possible brief opening again around 55k years ago. So, grizzlies probably arrived--and presumably then persisted--as early as 70k or as late as 55k years ago.
The map at left summarizes current thinking about the several migrations of brown bears from Asia into and through North America (parenthetically, grizzlies are brown bears, all Ursus arctos). An early wave of bears carrying genetic material identified with Clades 2a, 2c, 3c, and 4 (roughly the equivalent of subspecies) arrived in eastern Beringia around 70k years ago, having successfully crossed over the Bering Land Bridge. Of those bears, those of Clade 4 managed to successfully complete the journey into the middle part of the continent, presumably along the eastern edge of the Rocky Mountains. They were then isolated with closure of the ice-free corridor, intermingling with other grizzlies of the newly-arrived Clade 3b only after reopening of the ice-free corridor around 14-11k years ago (see the maps below). Shortly after, the Bering Land Bridge was closed to additional migrants by rising sea levels. Parenthetically, some researchers have theorized that grizzlies also got to the mid-continent by following ice-free coastal areas of modern-day Alaska and British Columbia--prior to the opening of Alberta's ice-free corridor. Land levels at these ice free margins were dramatically elevated as a compensatory ("forebulge") response to depression of the Earth's crust by the ice sheet farther inland. Migrating coastal grizzlies presumably subsisted largely on marine resources.
The Late Pleistocene world of North American grizzlies
The figure to the right gives you a sense of how brutally cold the last 400,000 years were (the bottom graph), the rapidity of warming around 14,000 years ago (top graph), and the dramatic fluctuations in temperature that occurred during the transition to warmer climates (also the top graph). The data come from ice cores drilled in the Greenland Icecap, represented as raw values (e.g., the concentration of deuterium isotopes) and as temperatures estimated from calibration of this isotope (in centigrade, as deviations from the present). The (more-or-less) current global average temperature is represented by a dashed line in the bottom graph. The gray bars in the upper graph mark major episodes of warming or cooling during the Pleistocene to Holocene transition.
The 400k year benchmark is relevant because this period encompasses much of the time that brown bears have been around as a species (see Evolution). The message is clear. Brown bears are largely creatures of the Ice Ages; they have existed in a much colder world almost their entire existence. The warm periods (represented by spikes that surpass current temperatures) were all relatively short.
Insofar as the challenges of the last 25k years are concerned: Dramatic warming roughly 14.7k years ago, during a period called the Allerod-Bolling, magnified the melt of ice caps which had began roughly 3.5k years before. Dramatic cooling followed between roughly 13k and 11.8k years ago, during the Younger Dryas. This period of cold but variable climate is thought to have been driven by an interplay between melt and resurgence of the Laurentide Ice Sheet (which bestrode eastern North America) and massive outburst floods from Lake Agassiz, which lay along its southwest boundary. Periods of minor cooling--referred to as "events"-- occurred at around 9.3k and 8.2k years ago, also likely driven by oceanic cooling and desalination caused by outbursts of cold freshwater from Lake Agassiz. Most of the megaherbivores and carnivores were finished off during the Younger Dryas, although declines in populations of many had begun well before. (I cover fates of the megafauna is more detail under Ancestral diet, largely because the oversized herbivores were undoubtedly an important source of food for Ice Age grizzly bears)
The maps above offer a coarse-grain reconstruction of vegetation and ice at the peak and waning of the last Ice Age. The top map of the Northern Hemisphere provides a more-or-less speculative but evidence-based reconstruction of vegetation during the peak of last Ice Age (i.e., the Last Glacial Maximum, or LGM for short). If you look closely at North America south of the ice sheets, you will see that most of the eastern part of the continent was covered by boreal forest, and that most of the western part was covered by some mix of forest and non-forest vegetation, expressed either as "parkland" or "woodland." There is debate over the composition of vegetation in the middle part of the continent, but one reconstruction sees it as grassland (i.e., "steppe"). Unlike in Eurasia, only a relatively narrow band south of the ice sheets was typified by some sort of tundra--or polar desert--conditions.
Focusing in on the Yellowstone area, which centers on the small interior blob of ice bordering "polar desert": You can see that this region around the ice-covered Yellowstone highlands was typified by tundra to the north and park- or woodland to the south. This agrees with the results of a number of regional paleontological investigations as well as model-based reconstructions of vegetation. For more details, check out the page on Yellowstone paleontology. Overall, Pleistocene North America is thought to have been more arid than contemporary times.
The bottom series of panels are from research published by John Williams and his colleagues, and show two things: first, progressive melt of the continental ice sheets (see the maps below for more detail), and the extent of forest cover at different snapshots in time between 21k and 6k years ago. In agreement with the vegetation map above, forest cover was scant between 21k and 16k years ago--especially in the West--and increased to an approximation of current conditions by 9k years ago. Note that remnants of the Laurentide ice sheet persisted until between 7-6k years ago.
So...what does this mean for North American grizzly bears during the Last Glacial Maximum? It is hard to know for sure. Physical remains are virtually nil from south of the ice sheets. But grizzlies (and brown bears in general) have been associated with open conditions during most of their evolution. Pleistocene brown bears have been closely identified with steppe tundra by Paleontologists in Eurasia (see Evolutionary biogeography). An association with open vegetation is further reinforced by the fact that brown bears have lived in a much colder presumably less-well-forested world during most of their existence as a species. All of this suggests that Pleistocene grizzlies in what is now the contiguous United States would probably have been concentrated in the more open environments of the West, much like they are now. Recent history suggests that grizzlies are not particularly well-adapted to North American boreal forests (see The Holocene), especially in the face of competition from other omnivores...and, perhaps, the toll of predators. Which leads to the next point.
The Late Pleistocene beastiary
extinct species by
Sergio De la Rosa Martinez
The menagerie at left portrays most of the large carnivores that shared the world of grizzly bears living south of the continental ice sheets during the Pleistocene. Portrayals of all the extinct species are thanks to the passionate devotion of a single Mexican artist--Sergio De la Rosa Martinez. The species are grouped according to family: ursids at top left, canids at top right, and felids arrayed below. The grizzly bear is shown farthest to the right of the assemblage of bears.
The point here is pretty straight-forward. Grizzly bears had a lot of competition for meat, far more than under contemporary conditions. Moreover, they were probably routinely killed by the largest of the other carnivores. Prime suspects include the solitary giant short-faced bear (Arctodus simus, in the back) and pack-hunting lions (Panthera atrox), scimitar-toothed cats (Homotherium serum), and saber-toothed cats (Smilodon fatalis)--the three species arrayed at far left in the line-up of felids. (A scenario of predation is rendered all the more plausible by the fact that modern-day Siberian tigers kill and eat brown bears in the Amur region of Russia.)
With this in mind, it is instructive to look at the abundance and distribution of remains left by the largest of the large carnivores during the Pleistocene. The map at left summarizes this information, with each species of carnivore denoted by a letter explained by the legend at bottom. These locations are overlain on a base map showing the estimated distribution of forest cover during the LGM (in shades of green) along with a reconstruction of the ice sheet margins at 23 and 13.5k years ago. Glacial melt lakes are also shown in blue.
The pattern of remains suggests that Arctodus simus, Panthera atrox, and Homotherium serum were all most common in open country of the West. There also appears to have been a dearth of these largest carnivores north of the Yellowstone region in what is thought to have been tundra or polar desert (see above). That said, paleontologist Danny Walker documents the relatively common presence of Arctodus and Panthera leo on southerly edges of the Yellowstone region (see the map the left as well. Notice, too, that in contrast to a relative wealth of remains from other large carnivores, there are none currently known from this period for grizzlies other than the single find near Edmonton, Alberta.
What does this all mean for Pleistocene grizzlies? Probably that they were pretty scarce, very likely holding out in a restricted range under conditions that were not favorable for other carnivores. Most likely in the fringe of tundra and polar desert just to the south of the ice sheets--habitat that was particularly well-represented immediately north of the Yellowstone ice cap. Such scarcity would be consistent with the probable levels of competition from and predation by other carnivores and the virtual non-existence of grizzly bear remains. This last point might be contested by suggesting that grizzly bear bones are intrinsically unlikely to be preserved. This doesn't seem to be a plausible explanation given that there are ample remains of other carnivores--presumably equally unlikely to be preserved--and that remains of grizzly bears do begin to show up after the mass extinctions of 13-11k years ago, albeit still in small numbers. But present nonetheless.
The Pleistocene-Holocene transition
The world of grizzly bears began to rapidly change roughly 15k years ago when the Pleistocene ice sheets started to collapse. This was followed (as described above) by major oscillations in temperature and moisture. The figure immediately above shows the extent of North American ice sheets at four intervals, beginning with near the LGM and progressing to when, around 14(-11?)k years ago, the ice-free corridor in Alberta emerged, cleanly differentiating the Cordilleran ice sheet to the west from the much larger Laurentide ice sheet to the east. At this point, grizzlies south of the ice sheets could begin moving north and grizzlies in Beringia could begin moving south. Perhaps more importantly, though: virtually all of the grizzly's competitors and predators disappeared, with the notable exceptions of black bears (Ursus americanus) and burgeoning numbers of humans.
The maps to the right show the distributions of grizzly bear remains from two different periods: In panel A, from 32-10k years ago, with all but the ones near Edmonton dating to less than 15.5k years ago; and, in panel B, from 10k to 200 years ago--up to essentially historical times. In panel A the extent of the ice sheets and bordering glacial melt lakes are shown at 13.5 and 11.5k years ago. Note Lake Agassiz (see the discussion above about climate). Most of the remains are not radiocarbon-dated.
The map in panel A constitutes pretty conclusive evidence that grizzly bears occurred in eastern North America between roughly 15.5 and 11k years ago. But by 10k years grizzlies seem to have been gone in the East. All of the specimens post-10k roughly coincide with the historical distribution of grizzlies (see The Holocene). The one exception is the Utz site in Missouri. This specimen was apparently found during excavations of a prehistoric Indian settlement. I tried to track down the original publication documenting this find, but without success. My current suspicions are that either: (1) the specimen was misidentified or, if authentic, then (2) the result of a series of trades among tribes that transported grizzly bear remains from farther west to this site in Missouri.
So, how is the prehistoric but post-Pleistocene pattern of grizzly bear remains explained? Moreover, how does one explain the isolated, highly anomalous, and recently extirpated population of grizzly bears in the Ungava area of Labrador? I would argue that the answers to both questions are interrelated.
The map at left shows a highly speculative reconstruction of the widest plausible distribution of grizzly bears during the LGM--Clade 4 south of the ice sheets and Clade 3b to the north in Beringia. I also show the likely distribution of Ungava grizzly bears along with the only three reliably-dated grizzly bear remains from eastern North America. This is all juxtaposed with a nested series of isopleths showing the margin of the Laurentide ice sheet at different intervals from 13k to 6k years ago. All of the ice margin reconstructions come from a more-or-less definitive compilation put together by the Canadian geologist Arthur Dyke.
But getting to the point. Notice that the grizzly bear locations in eastern North America are progressively younger the farther north you go, and that the two most recent ones (13.6k and 11.3k years ago) are near the margins of the ice sheet reconstructed for that particular point in time. This margin very likely remained mostly non-forested simply because of the lag times in colonization by trees, with spruce in the vanguard. If grizzlies were, indeed, competitively favored over humans and black bears in open country (see The Holocene), they may have been following the northward-advancing nonforested verge of the Laurentide ice sheet as it melted. Given the rapid spread of forest in eastern North America (see the maps above), these margin-hugging grizzlies may have ended up isolated and essentially stranded in the far northern tundra of Labrador. Another possibility is that grizzlies more recently spread east from Beringia to Labrador. Possibly. But the pattern of ice melt suggests that this route would have been blocked longer than the route north would have been. A definitive verdict on these competing explanations will be if any physical remains from Ungava can be tested to determine whether these grizzlies were descended from Clade 4 or Clade 3 bears.
Where did this leave grizzlies, and why? I speculate that they were pushed out of the heavily-forested regions of eastern North America by competition from forest-adapted black bears and increasingly ubiquitous humans. On top of this, I suspect that First Peoples were killing some number of grizzlies, which could have been the figurative nail in the coffin for a large-bodied slow-reproducing species (see Consequences of large size) that was already struggling in the face of scramble competition for dispersed food resources. This would be in sharp contrast to highly concentrated spawning salmon that brown bears readily dominate along the heavily-forested Pacific coast. Much the same likely held for large-bodied bison carcasses in the Great Plains. The point being that grizzlies probably fare better in open country and where rich food resources tend to be concentrated rather than dispersed. Their range thus ended up being restricted to western parts of the continent. I cover all of this in more detail under The Human, Black bear, & Bison factors, as well as under Ancestral diet.