Earth observation and remote sensing: multispectral, LiDAR and SAR data for archaeological and cultural landscape analysis

This research explores the integration of Earth Observation (EO) and Remote Sensing (RS) technologies for archaeological and cultural landscape analysis, with a focus on the combined use of multispectral, LiDAR, and Synthetic Aperture Radar (SAR) data. The study aims to assess how different remote sensing datasets, characterised by distinct spatial, spectral, and temporal resolutions, can be exploited to detect, document, and interpret archaeological features and past landscapes, contributing to the development of innovative approaches for preventive archaeology and cultural heritage protection. The work is structured into two main sections. The first provides a theoretical and methodological framework that introduces EO and RS principles, research objectives, and working hypotheses. The second presents methodological applications and archaeological case studies based on multispectral, LiDAR, and SAR data. Each dataset is investigated according to its specific analytical potential and limitations, with attention to data pre-processing, enhancement techniques, and interpretation strategies. The multispectral component focuses on the use of Sentinel-2 imagery to identify spectral anomalies linked to buried archaeological structures and palaeoenvironmental features. Different enhancement and transformation techniques are tested, including linear and non-linear radiometric enhancement, colour composite combinations, spectral indices, M-statistics, and linear transformations such as Principal Component Analysis (PCA) and Tasseled Cap Transformation (TCT). Applications carried out in the archaeological site of Poderi Colli d’Agnano (Rieti, Lazio) and in the Rieti basin demonstrate how specific combinations of bands and indices can highlight anthropogenic traces otherwise invisible to the naked eye. The LiDAR section investigates airborne laser scanning data for both site-scale and landscape-scale analyses. High-resolution Digital Terrain Models (DTMs) are used to visualise and interpret topographic microreliefs associated with medieval castles such as Torre Palazzo, Fenuculus and Monte Caperno. The results show that advanced visualisation techniques (i.e. Visualisation for Archaeological Topography-VAT) enable a detailed understanding of site morphology, defensive structures, and settlement patterns. At the landscape scale, LiDAR-derived models support the reconstruction of geomorphological processes in the lower Calore valley, providing insights into the relationship between environmental evolution and human occupation. The SAR component examines the use of multitemporal COSMO-SkyMed data for archaeological prospection and Cultural Heritage monitoring at the Roman site of Telesia Vetere (Benevento, Italy). By analysing temporal backscatter variations, the study evaluates the potential of SAR data to detect subsurface structures, soil moisture dynamics, and anthropogenic modifications, underlining the complementarity between optical and radar observations. In the general conclusions, the research discusses the integration of multispectral, LiDAR, and SAR data as a synergistic framework capable of enhancing archaeological interpretation across different spatial scales. Critical reflections address dataset limitations, atmospheric and environmental constraints, and the need for cross-validation through ground surveys. The study highlights the implications of EO-based approaches for sustainable Cultural Heritage management, spatial planning, and risk assessment. Finally, future perspectives emphasise the role of automated analysis and artificial intelligence as key directions for advancing archaeological remote sensing in the era of big Earth Observation data.