Ecological Background 1
The information on the ecology and biology of grizzly bears presented here calls into question the status quo approach to grizzly bear recovery in the contiguous US in two important ways: first by providing a science-based justification for recasting and reframing the Recovery Plan for grizzly bears in the Lower-48; and, second, by highlighting the immense body of relevant science that has emerged since the original Recovery Plan was last revised in 1993. The brief review of relevant science presented below is not meant to be exhaustive, but rather to be pointedly relevant and sufficient for illustrating the scientific merits of these two claims. Of particular relevance, fully 90% of the citations in this section post-date 1993, which is when the last Grizzly Bear Recovery Plan was finalized.
Image © Roger Hayden - all rights reserved
The Adaptability and Extensive Distribution of Brown and Grizzly Bears
Historically, brown bears were among the most widely distributed mammals on Earth. In the eastern hemisphere, brown bears inhabited most of Europe, including Great Britain in the west, northern Africa, Spain, and Turkey in the south, almost all of Russia, and parts of central Asia, China, and Japan. In the western hemisphere, the North American variant known as the grizzly bear was historically found from the tundra of Alaska in the north, down the Pacific coastal ranges to southern California, throughout western short- and mixed-grass prairie of the Great Plains, and the length of the Rocky Mountains from Canada south into Mexico (Figure 1). With the exception of the low deserts, grizzly bears were able to find food in almost every ecosystem from tundra to coniferous forest, deciduous forest, mixed woodland, prairie grassland, and savannah.
Although the digestive tract of grizzly bears is essentially that of a carnivore, grizzly bears are successful omnivores, and in some areas may be either almost entirely carnivorous or herbivorous. Grizzly bears are opportunistic feeders and will consume almost any available food including living or dead mammals or fish, and unfortunately, human garbage. In tundra ecosystems in the north, they feed mostly on ungulates, small mammals, roots, and berries. Along the West Coast, grizzly bears thrived on the great salmon and steelhead runs from Alaska to central California. On the Great Plains, grizzly bears consumed berries, followed the great herds of bison to scavenge carrion from winter die-offs and wolf-kills, and used their well-developed shoulder muscles and long claws to excavate roots. In California’s central valley, grizzly bears exploited salmon, acorns from the oak-savannah woodlands, and abundant game species to such an extent that they did not need to hibernate during the winter. In the Rocky Mountains, grizzly bears fed on abundant whitebark pine seeds, army cutworm moths in the alpine areas, and on cutthroat trout. And in the mountains of southern Arizona and Mexico, grizzly bears lived alongside jaguars, feeding on acorns, desert fruits, and carrion.
Figure 1. Historic grizzly bear range circa 1850 (light green), remaining range circa 1920 (dark green), and approximate dates of local extirpations, where known.
The Intrinsic Vulnerability of Grizzly Bears
A number of studies have identified factors that magnify risk of endangerment among carnivore species. The factor that is most often implicated is large body size, which is commonly understood to be correlated with, and a surrogate for, factors more directly relevant to conservation, including small litter sizes, slow reproductive rates, and low population densities. In addition, large body size among carnivores often translates into elevated levels of conflict with humans organized around depredation of livestock and perceived threats to human safety. Recent research has further implicated large body size in amplifying vulnerability to climate change, for much the same reasons, but additionally because of wide-ranging movements by individuals and large range-size requirements for populations.
These body-size-related considerations clearly apply to grizzly bears. Evolutionarily, ursids are bet-hedgers, to an extent greater than any other carnivore species; and, of the ursids, grizzly and polar bears are the most extreme in this regard.
As a life history and evolutionary strategy, bet-hedging comes down to a trade-off between adult female survival and reproductive output. Bet-hedgers invest fewer resources in the production of offspring than they do in behaviors and morphologic and physiologic adaptations evolved to insure the survival of adult females. In a nutshell, life-time reproductive success of a bet-hedging female is a "bet" on her long-term survival rather than on producing lots of offspring every year. Hence, annual survival of female grizzlies tends to be high (>90%), whereas average annual reproduction is low–roughly 2 cubs every 3-4 years, sexual maturity delayed until ages of 6-8, and lower reproductive rates in areas with lower habitat productivity.
Omnivory, hibernation, and large fat reserves are all behavioral, morphologic, and physiologic phenomena that increase the ability of individual females to buffer themselves against annual and seasonal variation in food resources. Fat stores exhibit multi-annual as well as seasonal cycles that allow individual bears to both survive and reproduce during prolonged food shortages, and omnivory allows grizzlies access to lower-quality foods, which further buffers individuals from times of dearth. Omnivory is imprinted on and, in turn, facilitated by molar dentition adapted to grinding coarse foods, an enlarged large intestine that facilitates digestion of fibrous material, and plantigrade posture, robust supra-scapular muscles, and elongate sturdy claws that all facilitate food manipulation, including access to under-ground foods.
Even though these adaptations allow individual females to survive extended periods of food shortage, female reproductive success is still affected by diet quality and body condition, whether reckoned for populations or for individuals. Such effects have been documented for polar bears, black bears, and grizzlies. At the population level, reproductive success of grizzlies has been positively correlated with greater consumption of spawning salmon and, to a lesser extent, with consumption of ungulates, and negatively correlated with colder, drier, and, overall, less productive environmental conditions. Among individual grizzly bears, reproductive success has been positively associated with greater consumption of fat-rich mast such as whitebark pine seeds and negatively associated with consumption of lower-quality foods such as roots. As a bottom line, even though individual grizzly bears are highly adapted to surviving frequent and sustained wide-amplitude variation in habitat conditions, females are not immune, reproductively, to the vicissitudes of their diet, whether reckoned for individuals or for populations. Such a conclusion is consistent with the strong correlation between grizzly bear densities and habitat productivity.
Of perhaps greater importance to conservation, the bet-hedging life history of grizzly bears makes populations extremely vulnerable to human-caused mortality. A number of demographic studies have shown that growth of brown bear populations is more sensitive to survival of adult females than to any other demographic parameter, in almost all cases accompanied by recommendations that managers focus on eliminating situations that lead humans to kill especially females. Although individual grizzly bears are resilient to seasonal and annual variation in habitat productivity, grizzly bear populations are not, primarily because of the sensitivity of population growth to survival of adult females and the degree to which female reproduction is affected by availability of high quality foods. More important, in all but the most productive coastal habitats, grizzly bear populations cannot sustain elevated levels of human-caused mortality, which highlights why sport hunting, poaching and removal of problem bears are all problematic in the management of grizzly bear populations, and reinforces the need for protections from human-caused mortality of any type.
The perhaps surprisingly limited ability of brown bears to overcome obstacles and colonize otherwise favorable habitats is a final important factor that amplifies the intrinsic vulnerability of this species to human persecution and long-term deterioration of habitat. Female brown bears tend to be philopatric, in some areas strongly so—which is to say that young females tend to settle in ranges that are near or overlapping with the ranges of their mothers. This translates into limited dispersal by young females, typically in the range of 10-30 km. together with a limited probability that they will move beyond their maternal range, at most on the order of 30-50%. Even though documented median dispersal distances for young males are 30-120 km, limited dispersal by females makes the establishment of reproducing populations in unoccupied areas >40 km from core ranges a very slow process. And colonization, as well as exchange of bears between established populations, is further impeded if not forestalled altogether by heavily-trafficked highways in settled valleys.
The Decline of Grizzly Bears in the Western U.S.
Grizzly bears have, in fact, proven to be particularly vulnerable to human persecution in the western United States. Between 1800 and 1975 grizzly bear populations in the lower 48 States declined from an estimated 50,000-100,000 to perhaps less than 1,000 bears. As the mountainous areas of the western U.S. were settled, the burgeoning mining and logging industries contributed to the increase in human-caused mortality of grizzly bears. Livestock depredation control, habitat deterioration, commercial trapping, unregulated hunting, and protection of human life were leading causes of decline. Professional hunters/trappers hired by Federal and State agencies also greatly contributed to grizzly bear population exterminations—as a matter of formal government policy.
By 1922, only about 37 populations remained. In a number of these areas, identified in Figure 1, probably only one individual survived. Between 1850 and the 1970s grizzly bears tended to survive only where human densities were low and in mountainous areas where rough terrain and widely distributed food resources tended to keep bears out of harm’s way. Distributions of foods mattered, as exemplified by the accelerated extirpation of bears in areas where foods, such as salmon and bison, were concentrated along riparian areas near people, versus the persistence of bears in places where high-elevation foods such as whitebark pine seeds kept bears concentrated in remote habitats.
Populations of grizzly bears in the lower 48 states are currently relegated to areas of much lower human densities than typifies the joint distribution of brown bears and humans in Eurasia–largely as an artifact of levels of human lethality between 1850 and 1950 in the U.S. When bears are faced with lethal humans, the large-body advantage of fasting endurance is outweighed by the increased sensitivity to added increments of adult mortality. In fact, large bodies, low reproductive rates, and low densities may explain the comparative absence of grizzlies from eastern North America prior to European settlement. Even before the advent of firearms the distribution of grizzly bears was likely limited by the interacting effects of competition with black bears and mortality meted out by Native Americans.
At the time of passage of the US Endangered Species Act and the listing of the grizzly bear as a threatened species in 1975, bears were known to still be present in Montana, northern Idaho, and Wyoming. Four of the seven remaining populations lie along the border with Canada, where grizzly bears remain. Grizzly bears may have been present in remote areas of the North Cascades at the time of listing, and there have been continued reports up to the present time that grizzly bears occasionally disperse from the Canadian side of the Cascades into the U.S. A grizzly bear was shot in the San Juan National Forest in Colorado in 1979, but none have been found there since then. No resident grizzly bears have been found in the Selway-Bitteroot ecosystem since the time of listing, and were last documented in this ecosystem in 1938.
The Sensitivity of Grizzly Bear Densities to Habitat Productivity
Perhaps paradoxically, and despite the resourcefulness of individual grizzlies, grizzly bear population densities are determined by overall habitat productivity, with little or no mitigating effects of dietary flexibility. Grizzly bear densities in North America vary by orders of magnitude, and in ways that directly correlate with diet and overall habitat productivity. Mowat et al. (2013) compiled density estimates from 90 different interior North American study areas revealing a 26-fold difference in grizzly bear densities, from a minimum of 2.5 to a maximum of 65 bears/1000km2. Mattson and Merrill (In press) found a smaller but still substantial range of 3.25-fold (8-26 bears/1000km2) when an outlying high value of 64 bears/1000km2 was dropped (this anomalously high estimate was made for the North Fork of the Flathead River in British Columbia) . The range of variation in grizzly bear densities is even greater when coastal populations with access to salmon are considered. Mowat et al. (2013) included coastal grizzly bear population densities as high as 85 bears/1000km2 in their analysis, whereas Miller et al. (1997) documented local seasonal densities as high as 551 bears/1000km2. Overall, Miller et al. (1997) found that densities of grizzlies in coastal regions were 6-80 times higher than densities found in interior regions of Alaska.
In addition to a strong positive effect of salmon on grizzly bear densities, Boyce and Waller (2003), Mattson and Merrill (2004), Merrill (2005), and Mowat et al. (2005, 2013), found strong positive relationships between densities of interior grizzly bear populations and indices of overall habitat productivity, including NDVI, precipitation, average annual temperature, and tasseled-cap transformed Wetness and Greenness. Mowat et al. (2013) additionally found a negative correlation with amounts of terrestrial meat in the grizzly bear diet and Mattson and Merrill (In press) found positive correlations with estimated net dietary energy and extent of whitebark pine range. This latter positive relationship with whitebark pine range is consistent with a positive effect of whitebark pine on persistence of grizzly bear populations between 1850 and 1970 in the face of persecution by Europeans. Similarly, Mowat et al. (2013) found a positive relationship between terrain ruggedness and grizzly bear densities in coastal and interior regions, consistent with the positive effect of mountainous topography on survival of grizzly bear populations during 1850-1970 found by Mattson and Merrill (2002).
As a bottom line, despite the fact that grizzly bears are adaptable omnivores capable of exhibiting considerable dietary flexibility, maximum densities of grizzly bears are strongly governed by the quality and quantity of food, in turn correlated in interior regions with overall primary productivity. Although terrestrial meat is a high quality bear food, bears tend to eat more ungulate meat in areas of lower primary productivity, which is reflected in lower maximum potential densities. In other words, omnivory and dietary flexibility are better understood as adaptations that allow individual bears to cope with high intra- and inter-annual variation in food availability rather than as mechanisms by which bear populations neutralize the effects of long-term trends or broad-scale differences in habitat productivity. For these reasons, it is imperative that grizzly bears be recovered across a diversity of habitats to buffer against any changes in productivity in one ecosystem versus another.
The Vulnerability of Grizzly Bears to Climate Change—Forecasts and Productivity
The strong links between grizzly bear densities and indices of primary productivity suggest that there will be major negative effects of foreseeable climate change. There is little doubt that dramatic and rapid climate change is happening at a pace not seen in thousands of years, largely due to anthropogenic forcing. Increases in temperature have accelerated during the last 40 years and are projected to increase in virtually all regions globally. Projections regarding precipitation, especially at a regional level, have remained more uncertain than projections regarding temperatures. Nonetheless, regional climate models have proliferated and improved to the point where researchers have been able to reach increasingly robust conclusions about not only precipitation, but also drought and related effects on vegetation. In North America much of this advance has been driven by the North American Regional Climate Change Assessment Program (NARCCAP) consortium.
Of relevance to grizzly bear range in the contiguous US, regional models are in consensus that summer-time temperatures will increase substantially in the northern Rocky Mountains over the next 100 years–not as much as in some regions to the south and east, but substantially nonetheless. Moreover, even though projections of growing season (June-August) precipitation vary, there is perhaps surprising consensus about the incidence of drought, largely driven by increases in growing season temperatures and earlier snow-melt. Recent multi-model forecasts project a substantial increase in drought frequency and severity throughout the northern Rockies, with demonstrable and projected effects on productivity and ecosystems accentuated by potentially dramatic changes in fire, insect, and disease regimes affecting already drought-stressed vegetation.
The Vulnerability of Grizzly Bears to Climate Change—The Greater Yellowstone Ecosystem
The Greater Yellowstone Ecosystem (GYE) offers important examples of how climate change, mediated through concrete natural and anthropogenic mechanisms, has or will foreseeably impact grizzly bear habitat productivity. Of the four energetically most important foods in the GYE—army cutworm moths, whitebark pine seeds, ungulates, and cutthroat trout —two, whitebark pine and trout, have already experienced catastrophic declines and a third, moths, are almost certain to suffer a similar fate. Cutthroat trout have been functionally eliminated as a grizzly bear food due to reduced stream-flows and predation by an introduced non-native predator, Lake trout. The effects of predation have been, and will continue to be exacerbated by the deteriorating condition of spawning streams driven by projected increases in the frequency of forest fires, ambient temperatures, drought, and rapidity of snow melt. Loss of the trout that fed roughly 80 grizzly bears during late spring and early summer may have already caused a compensatory increase in bear predation on elk calves, with resulting negative impacts on regional elk populations.
The recent and projected fate of whitebark pine is as bleak as that of cutthroat trout. An estimated 80-90% of current whitebark pine range is expected to be lost over the next 100 years owing to climate warming, with losses catalyzed by disease, insects, fire, and failed recruitment. Whitebark pine forests have already undergone major declines during the last decade due primarily to an unprecedented climate-driven outbreak of native mountain pine beetles, exacerbated by an on-going warming-enhanced epidemic of a non-native fungal pathogen called white pine blister rust. These two agents synergistically contribute to tree mortality, with blister rust more immediately lethal to small trees and beetles lethal to trees >6 inches dbh. Loss of whitebark pine is consequential because of its demonstrable effects on the reproduction and survival of Yellowstone grizzly bears. Female bears ate twice as many pine seeds as did males, and produced more cubs following good compared to poor whitebark pine seed crops. All bears also tended to survive at a higher rate during good seed crops because they were less exposed to human-related risks while exploiting this food, which occurs in remote high-elevation areas.
Looking to the future, there are no foods of a quality comparable to those being lost that are likely to successfully colonize or increase in abundance in the GYE during the next century. Some research has suggested that consumption of berries by Yellowstone grizzly bears has increased during recent years, with the prospect of more to come. However, key berry-producing species such as serviceberry, chokecherry, and buffaloberry are projected to decline substantially if not catastrophically during the next century. Even possible immigrants such as Gambel’s oak will not likely arrive and proliferate in time to help grizzly bears, despite the projected northward expansion of a favorable climate, primarily because migration of tree species is increasingly understood to be curtailed by limited dispersal abilities. And existing Yellowstone bear foods such as ants, hornets, mushrooms, and false truffles, which tend to be consumed more by bears during warmer drier growing seasons or when pine seed crops are small, are either not as abundant or of as high a quality as the foods being lost.
At the same time, the alpine habitats that support aggregations of army cutworm moth are, like whitebark pine habitats, projected to decline by roughly 90% during the next 100 years. Alpine moth aggregation sites have been used by increasing numbers of bears during July-August since the mid-1980s in the GYE, with the likelihood that use has increased during recent years in compensation for losses of whitebark pine and cutthroat trout, although exploitation of alpine moth aggregations by bears in other northern Rocky Mountain ecosystems is known to, or likely did, predate this activity in the Yellowstone area. Little is known comprehensively about the ecology of army cutworm moths while in mountainous areas during the summer, however, there is no doubt that the moths consumed by bears aggregate in alpine talus slopes during the day and subsist on nectar from alpine flowers obtained while foraging at night. With loss of alpine tundra, there is a good chance that army cutworm moth populations will either decline or that summer aggregations will shift to sites less likely to be used by bears.
A different perspective on the prospective effects of climate change is offered by Servheen and Cross (2010) and Roberts et al. (2014), who emphasize that grizzly bears are adaptable omnivores and that, although some foods will decrease in abundance, others will increase. The limitations of these analyses are several. For one, they fail to consider the evidence of relations between grizzly bear densities and primary productivity. For another, they fail to consider the comparative quality of foods likely to be lost with those likely to be gained. Related to this second point, they also tend to conflate the fact that bears are omnivores with the indefensible tacit assumptions that all bear foods are of equal quality, which is not the case, and that differences in the relative abundance and quality of foods have little effect on demography and density (see above). As a bottom line, these two publications offer little credible basis for concluding that climate change will have little effect on grizzly bear populations.
There is little doubt that grizzly bear range in the northern Rocky Mountains will be subjected to increasing warming and drying during the next century, with concomitant declines in overall productivity. Given the correlation of grizzly bear densities with productivity and the more concrete declines in abundance of high-quality foods already documented or foreseeable in the GYE, average bear densities will likely decline as well. There is no doubt that we are seeing just the beginning of climate impacts on grizzly bears and their habitat. The question is not whether grizzly bear densities will decline, but to what extent, which increases the imperative to establish and maintain large connected populations as a buffer against these climate forced changes.