A Space View of Radar Archaeological Marks: First Applications of COSMO-SkyMed X-Band Data

摘要

With the development of Synthetic Aperture Radar (SAR) in terms of multi-band, multi-polarization and high-resolution data, space radar remote sensing for archaeology has become a potential field for research. Nevertheless, the archaeological detection capability of this technology has so far not been fully assessed. This paper is a pioneering effort to assess the potential of satellite SAR X-band data in the detection of archaeological marks. We focus on the results obtained from a collaborative contribution jointly carried out by archaeologists and remote sensing experts in order to test the use of COSMO-SkyMed data in different contexts and environmental conditions. The methodological approaches we adopted are based on two different feature-enhancement procedures: (i) multi-temporal analysis performed to reduce noise and highlight archaeological marks; (ii) single-date analysis to assess the ability of the single SAR scene to detect archaeological features like with optical remote sensing. Results from multi-temporal data analysis, conducted using 40 scenes from COSMO-SkyMed X-band Stripmap data (27 February to 17 October 2013), enable us to detect unknown archaeological crop, soil, and shadow marks representing Luoyang city, dating from the Eastern-Han to Northern-Wei Dynasties. Single-date analyses were conducted using COSMO-SkyMed Spotlight scenes acquired for Sabratha (Libya) and Metapontum (southern Italy). These case studies were selected because they are characterized by diverse superficial conditions (desert and Mediterranean area) and archaeological marks (crop, soil and shadow). The results we obtained for both of them show that even a single SAR X-band acquisition is a feasible and effective approach for archaeological prospection. Overall, the methodological approach adopted demonstrated that both multi-temporal and single-date analysis are suitable for the enhancement of archaeological and palaeoenvironmental features.