Isle Royale National Park is a 45-mile-long forested island in northwestern Lake Superior, accessible only by ferry from Houghton or Grand Portage. It has no roads, no vehicles, no permanent residents outside the park staff, and approximately 165 miles of trail. It is the least-visited national park in the contiguous United States, with about 25,000 visitors a year, all of them in summer.
The McCargoe Cove shelter sits on a small flat above the eastern shore of McCargoe Cove on the island's north side. It is one of thirty-six wooden Adirondack-style three-sided shelters maintained by the park. It has a roof, three plywood walls, an open front facing the water, a sleeping platform for two, and a metal bear box at the trail junction below.
Imogen Reece spent four nights at McCargoe Cove from June 3 to June 7, 2026, on a working trip for the magazine. She walked in from the Rock Harbor ranger station, eleven miles by way of the Greenstone Ridge and Lane Cove trails, and she walked out the same way.
She was traveling alone, with permission of the park, in connection with research she is doing on long-term wolf-moose study sites for a forthcoming feature.
The Isle Royale wolf-moose study is the longest-running predator-prey study in the world. It began in 1958 with the late Durward Allen of Purdue University. It has continued, without interruption, for sixty-eight years.
The wolf population on the island, after a near-collapse to two animals in 2018, was supplemented by the National Park Service through a controversial translocation program. Nineteen wolves were brought to the island from Minnesota, Michigan, and Ontario between 2018 and 2019. The current estimate, from the 2026 winter study, is thirty wolves in four packs.
Imogen arrived at the shelter at 4:20 p.m. on June 3. The cove was calm. Two loons were calling from the far shore. There was no one else at the camp.
She set up her stove on the rock below the shelter, filtered three liters of water from the cove, and ate a meal of instant rice and dehydrated black beans. She read until dark, which at 47 degrees north in early June means around 10:15 p.m.
She heard the wolves for the first time at approximately 2:14 a.m. on the morning of June 4. The howl came from somewhere on the north shore of the cove, perhaps half a mile distant, and consisted of three voices, one deeper than the others, in a chorus that lasted perhaps thirty seconds.
She lay in her sleeping bag and listened. She did not get up. She did not turn on her headlamp. She did not, in fact, move at all.
The chorus repeated, more faintly, perhaps a mile further west, six or seven minutes later. Then nothing.
She wrote in her notebook the next morning: 3 wolves, 2:14, NW shore. answered at 2:20 from further west. likely Middle Pack, possibly Chickenbone Pack returning from a kill.
She was guessing about the packs. She would learn, on her return to Rock Harbor, that the Middle Pack had been documented in the area earlier in the week.
She spent June 4 walking the eight-mile loop around Chickenbone Lake. She saw moose tracks in three places along the trail, fresh, perhaps a few hours old. She saw the moose in person at a meadow above Lake LeSage at about 11 a.m. — a cow and a calf of the year, the calf perhaps two weeks old, both feeding on aquatic vegetation in water to their shoulders.
She watched for twenty minutes from a distance of perhaps a hundred metres. The cow noticed her, lifted her head, kept feeding. The calf, who had not noticed her, continued to splash.
She heard wolves again on the morning of June 5, at 4:45 a.m., a single voice from somewhere up the ridge to the south. She heard them on the morning of June 6, four voices at 1:30 a.m., closer than the first night, perhaps a quarter mile across the cove.
She did not hear them on her last morning, June 7. She walked out that day, eleven miles back to Rock Harbor, and caught the afternoon ferry back to Houghton.
She filed a wolf-vocalization report with the park's wildlife biologist as part of the long-running study. Such reports from visitors are part of the data set, supplemental to the formal winter telemetry. Her three observations were added to the spring activity records for the McCargoe Cove area.
She wrote, on the ferry home: the wolves of Isle Royale were down to two related animals in 2018, an inbreeding spiral that had taken the population from fifty in 1980 to functional extinction in less than four decades. the decision to intervene was contested. the decision worked. thirty animals on the island in 2026, four packs, breeding successfully, a moose population that had grown to over 2,000 in the wolves' absence now back in some kind of relation with its predator.
She added, in pencil at the bottom of the page: I heard them three nights in four. they were not performing for me. they did not know I was there.
The Isle Royale ferry runs from late April to late October. The park is closed entirely from November through April, when the lake is too dangerous to cross by boat and the island belongs, for half the year, to its wolves and its moose and its small population of resident researchers.
It is, she writes, one of the better small experiments in restraint that the National Park Service has ever undertaken.
