ORONO — When 32 caribou were released in Maine’s forests in the early 1990s in an attempt to reintroduce them in the state, the assumption was that Maine had always been home to herds of caribou and could be once again. But within two years most were eaten by bears and decimated by brain worms, a parasite carried by deer but not fatal to them.

Whether a species can or should be reintroduced — indeed, whether a species’ presence was significant to begin with — is a challenging question and requires biological and historical data that are elusive.

The Zoogeography Project, a joint effort by state zooarchaeologists to consolidate all published data about animal bones found in prehistoric settlements in Maine, eventually could help answer those questions and others. Zooarchaeologists analyze animal remains in order to reconstruct ancient human culture and interrelationships among people, animals and the environment.

The project’s database, which already contains records on more than 71,000 prehistoric bone fragments found in 39 separate sites in Maine, will yield important clues about historical fauna, according to Kristin Sobolik, an associate professor with the department of anthropology and the Institute for Quaternary Studies at the University of Maine, who has been spearheading the project.

The database will be used to track the migration of species through time and geographical space since the Ice Age ended 12,000 years ago. Among the many questions it could help answer are why certain animals became extinct, why some proliferated while others declined, and how climatic warming and cooling affected animal populations.

For the time being, the project will limit data to Maine archaeological findings documented in scholarly journals, master’s theses and doctoral dissertations, Sobolik said. John Mosher, a UM graduate student working on the project, has entered about 80 percent of the information into the database.

The idea for the project came after Sobolik saw Faunmap, a database established by the Illinois State Museum and available online. Faunmap has information on mammal remains from the past 40,000 years from 48 states, but it has little information from Maine.

Arthur Spiess, an archaeologist with the Maine Historic Preservation Commission and a participant in the project, said that he would like to see unpublished bone findings also become part of the database.

Most of the bones excavated in Maine have been found near the coast in shell midden sites, ancient trash dumps made primarily of emptied seashells. But prehistoric Mainers also threw animal and fish bones into the pile, which have been preserved for millenniums by the chemically protective qualities of the shells.

Imagine that outside your home you had a garbage pail 100 feet deep and your family used it for generations. If, thousands of years later, you could look through the preserved mess, you likely would be able to judge when, say, a Kentucky Fried Chicken first opened in the area by the presence of bones and boxes — and likely when it went out of business.

A similar record of vanishing and burgeoning species is available in prehistoric settlements. Early inhabitants captured and killed whatever animals were available for food or clothing, then discarded bones and shells. Carbon dating is used to measure the age of bones.

Finding buried bones in Maine’s interior is rarer because the state’s soil is acidic and dissolves the bone in 50 to 100 years. But scientists have uncovered inland bones that have been “calcined,” which is when all organic matter is removed, thus preserving them from decay. Although scientists do not fully understand how the bones became calcined, the process appears to involve the heat used in a particular cooking method that burns out the organic matter, Sobolik said. Mosher has been conducting experiments in which he burns fresh bones by various methods, then looks for similarities in ancient bones.

Identifying ancient bone fragments is a painstaking process. During a recent short tour of the zooarchaeology lab in South Stevens Hall, Sobolik opened cabinet doors holding the disassembled skeletons of deer, raccoon, salmon and others. Holding a fragment of jaw and teeth, she pulled open a drawer holding deer head bones. She turned the piece until it matched a spot in the upper deer jaw. That was a relatively large fragment.

The database will be valuable to animal biologists and other scientists because it will give them a systematic look at the ebb and flow of any particular species, Sobolik said. For example, it could help in researching the timing and location of disappearances from Maine of the wolf, cougar, turkey, polar bear and walrus. It could also reveal the relationships between species occupying similar niches, such as martens and fishers or wolves, coyotes and foxes.

For each bone fragment, more than two dozen separate fields of information are entered into the computer about the specimen and the excavation site from which it came. The power of a computer database is that thousands of entries in disparate fields can be compared or eliminated almost instantly. Such queries could lead scientists to surprising correlations about post-ice age animals.

Some simple, initial queries have already suggested avenues of investigation, according to Sobolik. For example, caribou bones are rarely found, and when they are, they do not coexist with deer, she said. Caribou, preferring cold climates, appear to have followed the receding glaciers north. Deer flourished in their absence.

According to Sobolik, that dynamic is consistent with the failed attempt to reintroduce caribou in deer-heavy Maine: deer carried a parasite that was fatal to caribou.

And those who have worked hard to restore the stocks of Atlantic salmon in Maine rivers for the past half-century might be interested to learn that salmon are rarely found in the bone database, according to Sobolik. The exception was a period during the 1800s, which was also during the “little ice age,” a climatic trend that temporarily cooled the Northern Hemisphere and its rivers. That could mean that Atlantic salmon were a relative blip in Maine’s rivers during the past 10,000 years.