Mapping endangered cultural soilscapes in post-conflict regions through Big Earth Observation Data, artificial intelligence and geoarchaeological signatures

Informations projet

SIGNATURE

N° de convention de subvention: 101067100

DOI

10.3030/101067100

Projet clôturé

Date de signature de la CE 4 Juillet 2022

Date de début 1 Août 2023

Date de fin 31 Juillet 2025

Financé au titre de

• Marie Skłodowska-Curie Actions (MSCA)

Coût total

Aucune donnée

Contribution de l’UE

€ 181 152,96

Coordonné par

Institut Català d'Arqueologia Clàssica
Spain

Periodic Reporting for period 1 - SIGNATURE (Mapping endangered cultural soilscapes in post-conflict regions through Big Earth Observation Data, artificial intelligence and geoarchaeological signatures)

Période du rapport: 2023-08-01 au 2025-07-31

Résumé du contexte et des objectifs généraux du projet

The SIGNATURE project has focused on preserving endangered cultural soilscapes—the long-term human imprint in landscapes—across the Levant, particularly in the fertile alluvial plains of Lebanon’s Beqaa Valley and the southern Jordan River Valley. These landscapes have been shaped by millennia of human activity but face increasing threats from climate change, desertification, erosion, modern agriculture, settlement expansion, and ongoing political instability.

SIGNATURE aimed to document, characterise, and protect these fragile heritage sites by combining archaeological legacy data, historical maps, and multi-temporal satellite imagery with geoarchaeological analysis and advanced computational workflows. By investigating anthropogenic soils and landscape changes, the project provides evidence to monitor vulnerable archaeological locations, understand past land-use practices, and inform sustainable heritage management policies.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

SIGNATURE integrated legacy archaeological datasets, historical topographic maps, and remote sensing imagery from CORONA, HEXAGON, Sentinel, and Landsat satellites. Key achievements include:
Revising Leon Marfoe’s 1979 survey of the Beqaa Valley, identifying 119 mound-like archaeological sites, and geo-correcting 75 locations for further analysis.


Developing R and JavaScript codes for automated correction, analysis, and integration of multi-source Earth Observation data in cloud-based platforms such as Google Earth Engine.


Conducting field campaigns in Lebanon (2023) and Jordan (2024–2025) to validate remote sensing results, relocate and reassess archaeological mounds, and collect geoarchaeological samples.


Detecting long-term land-use patterns and distinctive anthropogenic soil signatures through multi-temporal analyses, enabling the identification of archaeological features even when they are eroded or partially destroyed.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

SIGNATURE has advanced archaeological remote sensing by integrating multi-sensor and multi-temporal datasets to detect subtle geoarchaeological features and monitor endangered landscapes. Results demonstrate the potential of combined Earth Observation and soil analyses to identify and preserve fragile heritage assets in post-conflict and semi-arid regions.

The project has established three international collaborations with the Directorate of Antiquities of Lebanon, the Department of Antiquities of Jordan, and the EAMENA project (Universities of Oxford and Durham, UK). These partnerships facilitate future joint field surveys and capacity-building in post-conflict regions.

Fieldwork and geoarchaeological sampling of tell archaeological sites in the Jordan River Valley.

Risk assessment of Leon Marfoe’s (1970s) tell locations in the Beqaa Valley, Lebanon, identifying ma

MODIS multitemporal soil signatures (EVI spectral index, yearly average for 2010-2023) showing mound

Reassessment of Leon Marfoe’s (1970s) tell locations in the Beqaa Valley, Lebanon, using geospatial