Archaeology Magazine - January/February 2018 - 36

University of Texas at Arlington archaeologist Ashley Lemke, who has worked on
the project for six seasons. She points out
that while they moved across the ridge,
Paleoindians would have encountered a
rocky landscape with thin soils, covered in
bogs and marshes. While the landscapes
on the mainland were slowly developing
into the grassland, savannah, and forest
ecosystems that exist there today, the
land bridge was covered in sparse clumps
of trees such as spruce, tamarack, and
aspen. "It was an Ice Age-like environment," says Lemke. "That's why caribou
would be there. This is a place where the
Ice Age lasted a little longer."
During the winter months, the ridge
would have been brutally cold, leading
O'Shea and his team to believe that the
Archaeologists investigate now-submerged stone caribou-hunting structures at the
Paleoindians would not have lived there
bottom of Lake Huron. They were built some 9,000 years ago, when this section of
year-round. In the autumn, when caribou
the lake bed was a land bridge that lay on the animals' migration route.
herds were healthiest, with the best meat
hat first field season in 2008 was grueling. On the
and hides, O'Shea believes small family groups would camp on
few days when the weather cooperated enough to get
the land bridge as the animals migrated across it toward the
out on the big lake, O'Shea and his team would make
southeast. The families would hunt in small groups, processing
the 60-mile journey in a small boat from the city of Alpena's
hides and drying and storing the meat in stone caches on the
port to the underwater ridge. There they identified three areas
ridge for the winter. Because the cold winters froze the lake
to scan, ranging from four to seven square miles. Day after day
solid, O'Shea believes the caribou hunters would have been
they followed a grid pattern, using side-scanning sonar to get
able to travel over the ice to the ridge in the winter and collect
an overview of the ridge, creating a map of the land below. This
meat when they needed to.
systematic approach paid off, and they were able to identify
In the spring, as the herds of caribou headed up the ridge to
ancient shorelines, prehistoric rivers, lakes, and bogs. O'Shea
the northwest, the hunters would have engaged in a different
hoped they could use that preliminary data to pinpoint posstyle of hunting, working in larger groups to herd the caribou
sible areas for investigation during a second field season. But
down through stone drive lanes and processing large amounts
then, something unexpected happened. During one pass, a
of meat. "Spring is the direst time in northern climates," says
bright line of rocks stood out. The team sent an ROV down
O'Shea, who notes that after enduring a harsh winter, the underand found, to their surprise, that it was a man-made drive lane,
nourished Paleoindians would have needed a way to get large
a long stone alignment used to herd caribou into a corral, and
amounts of meat quickly. "They used more complex hunting
a hunting blind where Paleoindians would have waited to kill
methods that took a lot of people to operate," he says. "Then
the animals. "It was just dumb luck," says O'Shea. If the boat
they would sit around for a couple of weeks eating caribou."
had been traveling at even a slightly different angle, the team
hile the teaM has successfully located a number
would have missed the drive lane.
of sites and even excavated a handful of stone
Since then, over almost a decade of field seasons, O'Shea's
tools, O'Shea knew that surveying the entire
team has identified more than 60 drive lanes and hunting
ridge would take years of mind-numbing radar scanning and
blinds, as well as structures that were possibly caribou meat
cost a lot of money. So he turned to computer scientist Robert
caches, all across the Alpena-Amberley Ridge. One reason
Reynolds, director of the Artificial Intelligence Laboratory at
they have found so many archaeological sites intact is that,
although the ridge's original layer of thin topsoil has washed
Wayne State University, for help in better understanding how
away, no new sediment covered it, leaving a perfectly prethe hunters might have exploited the land bridge and where
served record of the people who lived there. "That's what's
still more sites might be located. "Initially the team was using
nice about our research site," says O'Shea. "There's only a
their own intuition about the landscape to predict where they
2,000-year window where it was dry land. Then it was submight locate sites," says Reynolds. "They felt they reached a
limit of their intuition and called us in to create a 3-D virtual
merged and didn't reemerge again."
model of the ridge." By combining this virtual model of the
Dating of the remains of ancient trees on the ridge based
refugium with information about caribou migratory behavior
on wood samples and ancient pollen has shown that the land
and ethnographic and archaeological data, Reynolds was able
bridge would have indeed been a subarctic environment, says

ARCHAEOLOGY * January/February 2018

Table of Contents for the Digital Edition of Archaeology Magazine - January/February 2018