Exploring ancient underground remains using the electrical resistivity method: A case study in Shangqiu, China

Shangqiu, regarded as the center of predynastic and dynastic Shang cultures, possesses a rich historical legacy and attracts significant attention in China. Excavating ancient underground remains in Shangqiu offers archaeologists valuable insights into ancient Chinese history and culture. In this study, electrical resistivity surveys were conducted to investigate potential ancient remains around an archaeological excavation pit in Shangqiu. Due to site constraints, only two approximately perpendicular 2-D survey lines were acquired around the archaeological excavation pit. To improve the reliability of interpretation, we compared 2-D inversion, 3-D inversion using the limited memory quasi-Newton (L-BFGS) method, and 3-D inversion based on broad learning (BL). The 2-D inversion results characterize the ancient city wall but fail to depict the ancient building components. In addition, false anomalies are produced in the deeper regions due to the influence of the archaeological excavation pit. The 3-D inversion results obtained using the L-BFGS method or the BL network depict the ancient city wall and building components beneath the survey lines. Compared to the 3-D inversion results obtained using the L-BFGS method, the BL inversion results reveal a broader anomaly range in the area of the ancient building components. Archaeological excavations confirm that the BL inversion results more closely reflect the actual subsurface conditions. Additionally, the area of the rammed-earth ancient city wall corresponds to lower radon gas values. This study provides valuable insights for decision-makers in archaeological excavations, and the proposed processing technology is helpful for the practical application of the electrical resistivity method in archaeological exploration.