An important video that describes how surveying archaeological sites at scale in India and Pakistan can be undertaken, using survey of India maps from the colonial period in part to help locate potential locations of interest. This is the task of the MAHSA – Mapping Archaeological Heritage in South Asia – project started at Cambridge University with numerous international collaborators. MAHSA cleverly uses these old maps, machine learning, decommissioned spy maps, academic publications, field research and more to retrieve archaeological information on a large scale, particularly the Indus River basin, that is revealing to archaeologists today.
Combined with modern digital tools and maps, this kind of investigation helps uncover ancient mounds since built over or otherwise ignored. Machine learning is also showing sites which under threat from agricultural and other forces. There is a fascinating bit on how Dholavira, Rakigarhi and Mohenjo-daro appeared on maps before archaeologists recognized them as sites worth excavating. Mounds not documented by archaeologists but illustrated by surveyors over a century ago like Agroha in Hissar are discussed, revealing new information encoded as it were on the old maps. The project also helps filter out the same site reported as different through different discovery methods. This large international collaborative effort documents archaeological sites in the subcontinent using modern, open-access tools like Arches, a geo-spatial mapping database to store and present data. Artificial intelligence (AI) is also used to match modern satellite imagery with known archaeological site locations and applies patterns to surveying new areas, "re-discovering" archaeological sites as the enterprise seems to have done in the Cholistan desert, near Ganweriwala, with new areas of potential ancient habitation appearing. In addition to all the remote sensing and digital endeavors, on-the-ground work using new tools developed by MAHSA for fieldworkers adds another level of information to the work. "A complex hydrological landscape," is how Dr. Petrie refers to the Punjabi landscape through history; MAHSA shows how very apt this description has been over thousands of years.
Most interesting are the mapped results from the pre to post-Harappan times and the distribution of sites since then, the potential role of the Ghaggar-Hakra river and settlements along it, and the varying distribution of sites through medieval times. In general, these seem to show a movement of peoples eastwards over time. The role of rivers and palaeorivers as "active participants" in the landscape becomes evident. As a special bonus, colonial policy and its role is also taken into account. Pre-Partition Indian surveyors and their names are being collected and highlighted as part of the effort, reading the fine and very fine print on old maps and documentation. Like in excavations from the period, the contribution and work of local archaeologists, photographers and scientists has been overshadowed by their colonial bosses whose names became ever-more inscribed and dominant in the record, and it seems as if the MAHSA project will ultimately contribute to restoring their important contributions.
"There are hundreds if not thousands of sites for which the data has not been recorded," says Dr. Petrie, "and our aim at the moment is to resolve that problem." A video that clearly proves that new technologies, old documentation and sophisticated researchers working together from all over the world can open up whole new reams and realms of archaeological discovery and insight. MAHSA is clearly an exceptionally sophisticated and promising enterprise.