The virtual reality of the land bridge was created by computer scientist Bob Reynolds and a group of students at Wayne State University.
(Photo courtesy of John O’Shea, University of Michigan)
In a previous story, Life Before Lake Huron, we recounted how researcher John O’Shea and his team shattered Western science’s view of life in the Great Lakes region during the Mesolithic era. Now, those same researchers are working with computer scientists to create herds of ancient AI caribou to find new archaeological sites and allow modern-day people to explore this ancient virtual landscape.
Archaeological sites sitting at the bottom of Canada’s second-largest lake, Lake Huron, are difficult to sample, and even more difficult to identify. Using sonar, O’Shea and his research team were able to identify a series of stone structures used to bottleneck herds of caribou when hunting. At this site, they recovered a stone tool that originated more than 4,000 kilometres away, indicating that extensive trade networks existed across the continent much earlier than previously thought. However, many similarly rich archaeological sites do not have such salient features literally pointing to their location.
“Underwater research is always like a needle in a haystack. So, any clues you can get that help you narrow down and focus the kind of places you might look at is a real help to us,” said O’Shea.
That’s where artificial intelligence comes in. Facing the dilemma of where to dig next, O’Shea teamed up with Robert Reynolds, professor of computer science at Wayne State University, to use computer modelling in the search for new archaeological sites.
Using this same bathymetric data collected by O’Shea that identified the previous site, the research team is now modelling the topography and land cover of the region and tracking AI caribou across this ancient landscape.
“Because of the slopes, you could simulate the flow of the water. … So we were able to establish, sort of, marshes,” said Reynolds. Once the hydrology of the region was modelled, they were able to work with biologists to populate the model with plants and land cover. “And then what we had to do is to populate the landscape, and we did that by developing a model of caribou.”
From this, the researchers have already been able to identify new sites with significant artifacts that were previously overlooked using traditional sonar techniques.