Universiteit Leiden

nl en

Developing methods on remote sensing detection of archaeological features in Colombia with LDE grant

A Leiden-Delft-Erasmus research team has been awarded a LDE Global Support Grant to develop reusable algorithms in the remote detection of non-orthogonal architectural features, taking place in the archaeological context of the northern extremities of the Andean, part of the Istmo-Colombian Area.

Pre-Columbian structures

The initial focus lies on the Sierra Nevada de Santa Marta in Northern Colombia—an area characterized by snow -covered mountains, primordial rainforests, and high biodiversity with numerous endemic and endangered species. These mountains are also home to indigenous communities and harbor thousands of pre-Columbian structures, such as terraces, burials, retaining walls, paths, and staircases.

Collaborative effort

Sebastian Fajardo from the Leiden Institute of Advanced Computer Science (LIACS) at Leiden University is leading the team in collaboration with Alex Geurds  and Frank Takes (LIACS), This collaborative effort also involves Naomi Oosterman (Erasmus), Juan Carlos Vargas (Universidad del Magdalena), and Roderick Lindenbergh (TU Delft). This research is complementary to Geurds’ recently awarded NWO-Vici grant and will amplify its focus on monumental landscapes. The LDE grant also reinforces the MAPSHA-project at LIACS, led by Sebastian Fajardo, and supported by the Arcadia Fund. The project is embedded in Colombia through the local counterpart of the Universidad del Magdalena.

Machine learning and heritage approaches

Central to this project will be the training of MSc students at the intersection of machine learning and heritage approaches. UAV-LiDAR survey across a forested hillside region of the Sierra Nevada de Santa Marta will be combined with mapping vegetation differences related to archaeological surface features. Results and approaches will be discussed during a collaborative workshop with stakeholders in Colombia during the autumn of 2024.

This website uses cookies.