This project will map subterranean water supply systems—qanats—across arid regions of the ancient world, from North Africa to Western China. Recognizing increasing aridity and water shortages, international organizations like UNESCO, along with national heritage and development agencies, have expressed interest in these ancient water extraction and conveyance systems, which have underpinned production and habitation since the mid-first millennium CE. Despite substantial public and academic interest, the identification and study of qanats remains limited; a comprehensive understanding of this extensive human heritage is impeded by the slow pace of traditional detection methods.
To address this challenge, we propose to use the latest advances in Deep Learning to create an AI-assisted detection application for archaeological mapping, establishing a partnership between UChicago, Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE) in France, and the Endangered Archaeology in the Middle East and North Africa (EAMENA) project, along with its network of UK-based universities. Through a series of hackathons, this international collaboration will map all qanat systems from North Africa to Central Asia, paving the way for a more comprehensive understanding of ancient water management infrastructures as well as faster adoption of AI in the archaeology and cultural heritage sectors.