ABSTRACT Since 2022, integrated surveys by researchers from the Sultan Qaboos University in Oman, the Technical University of Crete, Greece, and the University of Pisa in Italy combine geophysical and archaeological studies for the revealing and conservation of Early Bronze Age (EBA) human settlements. These settlements have become archaeogeophysical laboratories for the researchers involved, due to the different kinds of architectural features based on the period of their origin. 3D ground penetrating radar (GPR) data and AI techniques have served the scientific community in the latter years due to the increased computational power, which allows extracting 3D information on buried remains. Here, we present the results of the ongoing research at the Al‐Tikha site and describe a specific type of archaeological site–dedicated data processing methodology for the assessment of EBA mounds. We applied multipath summation on the GPR sections combined with topography corrections and enhanced the geophysical interpretation by diffraction velocity models based on the kinematic characteristics of diffractions mapped by a globally trained ML model. A diffraction velocity model can never be accurate as it is based on the number and position of the diffractions over the x‐t plane, making it not suitable for most cases for direct migration. Our methodology suggests the combination of multipath summation and topography corrections as an alternative. The latter proved more efficient than topographical migration in our examples. This methodology is implemented and tested here with the scope to efficiently image the subsurface of the mounds at EBA settlements. Two mounds were prospected, and buried walls and structural layers were mapped to serve as targets for future excavation, while both mounds were found to have different architectural phases of construction.
Economou et al. (Fri,) studied this question.