For several years, biologists monitoring black bears in the Great Smoky Mountains National Park noticed a pattern in their data that they could not explain. Subadult male bears, typically two to four years old, were disappearing at rates higher than the baseline data predicted. The known causes of bear mortality in the park, such as vehicle collisions and human conflict, did not account for the losses. GPS collar data showed some collared bears going silent in ways that suggested the animals had died, rather than experiencing equipment failure.
The park is home to roughly 1,500 black bears and has been the subject of systematic scientific monitoring since the 1980s. The resulting dataset is one of the most detailed records of a wild bear population in North America. It includes population estimates, demographic breakdowns, mortality rates, and movement patterns. Because this baseline data was so comprehensive, the deviation among subadult males was noticeable. The trend persisted across multiple years, ruling out statistical noise or a single anomalous season.
The Great Smoky Mountains National Park spans approximately 500,000 acres along the Tennessee-North Carolina border. It is the most visited national park in the United States, receiving more than 12 million visitors annually. The park is also a significant biodiversity hotspot, containing more tree species than all of Northern Europe and supporting over 150 species of birds. Its long history of intensive study made it more likely that an unusual pattern in the bear population would be detected.
Before this investigation, the predator-prey relationships in the park were considered well understood. Gray wolves and cougars were eliminated from the region in the early 20th century due to hunting and habitat loss. In their absence, black bears functioned as the park’s largest native carnivores, though their diet is mostly vegetarian. Coyotes were present in the park, but they were not considered a significant threat to bears of any age. The Smokies were understood to be an environment with low predation pressure on black bears. The investigation would require a revision of that understanding.
The investigation was organized by the park service’s wildlife management division, working with researchers from the University of Tennessee and the Appalachian Bear Rescue. The collaboration combined the park’s monitoring infrastructure with independent expertise in bear ecology. The team reanalyzed existing GPS collar data, paying close attention to the spatial and temporal patterns of signal loss to distinguish between bears moving out of transmission range, equipment failure, and mortality. They supplemented the existing trail camera network with additional cameras in areas flagged by the collar data. They also initiated a targeted recapture program to collar a new group of subadult male bears for real-time tracking.
A forensic component of the investigation examined carcasses recovered from the park over the preceding five years. A wildlife veterinarian reviewed injury patterns and identified a subset of the carcasses, specifically those found in backcountry corridor areas, that showed injuries inconsistent with road mortality or disease but consistent with a large predator attack. This was a preliminary finding with a small sample size, but it gave the investigation its first specific hypothesis.
The supplemental trail cameras captured footage of large coyotes moving through the monitored movement corridors. Some sequences showed groups of coyotes moving together in patterns consistent with coordinated group hunting. When the locations where collared bears had gone silent were mapped against the areas of coyote activity, the team found a consistent overlap. The cameras did not directly capture predation events, but the pattern of bear mortality corresponded closely with the pattern of predator activity.
One sequence captured a subadult male bear at a stream crossing, identifiable as an animal whose collar had gone silent three days earlier. The footage showed the bear behaving with heightened alertness. At the edges of the camera’s field of view, multiple large coyotes approached from different directions simultaneously. The camera did not record the outcome, but the collar data had already indicated the result.
The investigation concluded that eastern coyotes were a significant and previously unrecognized cause of mortality for subadult male bears in the park. Coyotes killing bears was not unprecedented in the scientific literature, but it had generally been treated as exceptional rather than systematic. The investigation found that the eastern coyote is a different animal from the western coyote that most research has been based on. Through hybridization with gray wolves and domestic dogs during their colonization of the eastern United States, eastern coyotes have become larger and more behaviorally complex. Adult eastern coyotes in the southern Appalachians can weigh 40 to 50 pounds, significantly more than their western counterparts.
The investigation documented a hunting strategy that was more sophisticated than expected. The coyotes targeted subadult males in the life stage when they were physiologically stressed and inexperienced, navigating unfamiliar territory after dispersing from their mother’s home ranges. The approach strategy was consistent across multiple independently captured sequences, with animals approaching from different directions simultaneously. The coyotes exploited terrain features that limited the bear’s escape options, such as stream crossings and narrow valleys. The consistency of this behavior suggested learned group hunting rather than coincidental individual actions.
The investigation did not conclude that coyotes were the only factor in the bear disappearance pattern. Other pressures, including competition for food resources during masting failures and drought conditions, also contributed to elevated mortality. But the investigation concluded with high confidence that coyote predation was a significant contributor.
The management implications for the park are complex. The bear population is large enough that the current predation pressure is unlikely to threaten its long-term viability. However, the demographic effects of elevated mortality among subadult males will require monitoring with a more specific framework than the existing program provides. There is also a visitor communication challenge. The park receives millions of visitors annually, many of whom are familiar with guidance about bears. The finding that large, coordinated coyote groups are active in backcountry areas introduces a new element to that framework, and updating the park’s communication requires care to avoid overstating the risk to humans.
The findings also raise questions about eastern wilderness areas more broadly. The eastern coyote has colonized nearly the entire eastern United States over the past century. It has been understood as a significant predator of white-tailed deer, but not as a predator capable of exerting meaningful pressure on large mammals like black bears. The Smoky Mountains finding requires a reassessment of that understanding. Other major eastern wilderness areas with both coyote and bear populations, including the Adirondacks, the White Mountains, and the Appalachian corridor, may have similar dynamics. The question is whether their monitoring systems are designed to detect such a pattern if it is occurring.
The Smoky Mountains investigation succeeded because the biologists took anomalous data seriously and built an investigation specifically designed to explain it. The trail cameras, GPS collars, and spatial analysis tools were already in place. What was missing was the question that the investigation asked. The mountains have been reshaping themselves ecologically for a century, losing apex predators and gaining new ones. The monitoring systems built to track the old ecological model were not always equipped to detect the changes. The bear investigation is a demonstration that asking the right questions of existing data can reveal things that have been happening for years without scientific awareness.
