Thousands of cultural sites that span Illinois’ history are preserved across the state. These places were the homes of indigenous groups as early as 13,000 years ago and, more recently, of Euroamericans, both free and enslaved Black Americans, and other immigrants. These sites contain the material record of past human behavior—buried features like storage pits or house foundations and artifacts like stone tools and animal bone. This record speaks to the lives of the people that called this landscape home for millennia, and is an invaluable and non-replaceable resource both for archaeologists trying to answer questions about how human groups adapted to (or created) changes in their physical and social environments and for descendant communities seeking physical and historical connections to their past.
More than 70,000 archaeological sites of all time periods have been documented during more than a century of archaeological work in Illinois. More than 9,500 of these sites are located in Cook County and the surrounding collar counties (Lake, McHenry, Kane, DuPage, Will). This includes over 600 registered Paleoindian or Early Archaic sites (13,000-8,000 years ago) in our study area that were occupied during the Late Pleistocene or Early Holocene. 200 more of these earliest sites have been identified thus far as part of the First Peoples project based on analysis of material in private collections. These sites and associated artifacts make up a critical baseline for what we know about the first people to call Illinois home.
However, Paleo and Early Archaic sites are actually quite rare compared to sites from the more recent past and there is still much we have to learn about the lives of Illinois’ earliest peoples. In part, the earliest sites simply harder to find—they are smaller, more ephemeral, less visible on the landscape (they don’t have visually distinctive features like mounds), and contain more organic materials (for example wood or bone tools) that are more vulnerable to natural decay processes than materials which are common in post-contact sites like glass and metal. The oldest sites are also less likely to survive to the present in places that have experienced erosion, flooding, or more recent development.
Given the scrappy nature of this early archaeological record, how can we fill in the gaps in our knowledge and begin to flesh out patterns that may not be immediately obvious? Revisiting known sites and conducting surface survey and shovel testing to find new ones will obviously be important, but limitations on time, funding, and other resources mean we need to look for other avenues that use existing data. As much as archaeologists would love to, it’s just not possible to look everywhere.
Another way to make informed, evidence-based decisions about where Paleoindian and Early Archaic sites are likely to be found is through predictive modeling. As part of the First Peoples project, we developed a model that considers the known locations early indigenous sites, landforms that lack early sites, and other environmental variables such as elevation, distance to rivers and streams, and native vegetation. Data sampled from these locations are analyzed via GIS (Geographic Information Systems) and used in a TensorFlow neural network (a machine learning algorithm) to predict the probability of encountering a Paleoindian or Early Archaic site in every 30 x 30 meter cell across Cook County and the collar counties.
