2026/04/13 by Salvatore Polverino, Hourakhsh Ahmad Nia, Rokhsaneh Rahbarianyazd · 1 voice
Engineering · Earth and Planetary Sciences · #Synthetic Aperture Radar (SAR) Applications and Techniques #Archaeological Research and Protection #Geophysical Methods and Applications
paper · doi:10.1002/arp.70044
openalex publication_date 2026/04/13 · openalex created_date 2026/04/15 · openalex updated_date 2026/07/14
ABSTRACT Burial mounds are key elements of Mediterranean funerary landscapes, but in intensively cultivated coastal plains their low‐relief expression is easily obscured by ploughing, levelling and rapidly changing surface conditions, making single‐date observations unreliable. This study develops a multiepoch Sentinel‐1 TOPS InSAR approach for repeatable mound‐domain recognition and disturbance‐hotspot screening at the Manicalunga–Timpone Nero necropolis near Selinunte, south‐western Sicily, Italy. Four C‐band interferometric epochs acquired between 2014 and 2025, spanning Sentinel‐1A, Sentinel‐1B and Sentinel‐1C, were analysed within DEM‐based visibility masks so that interpretation was restricted to coherence‐supported areas, while decorrelated sectors and layover/shadow zones were retained as explicit observational limits. Wrapped‐phase variability, coherence and terrain context were integrated into a covariance‐aware proxy, re‐expressed as an anchored ordinal ranking and fused across epochs to emphasise persistent rather than one‐off anomalies. The resulting multiepoch mean proxy, integrated with supervised machine‐learning‐based pixel classification (ilastik) to generate probability maps, delineates recurrent mound‐domain signatures and identifies deformation/disturbance hotspots both inside and outside the legally protected archaeological perimeter. The results show that a conservative, coherence‐gated InSAR workflow can separate persistent high‐rank anomaly structures from condition‐dependent noise in cultivated heritage landscapes. The approach also provides a transferable screening framework for archaeological prospection and conservation management, supporting targeted field verification, explicit reporting of non‐observable sectors and more robust prioritisation of monitoring actions.