Evaluation of trenching, ground penetrating radar (GPR) and electrical resistivity tomography (ERT) for sinkhole characterization

TLDR

The study investigates combining trenching with GPR and ERT to characterize sinkholes in a mantled karst setting. The authors applied the combined technique to two anthropogenically concealed sinkholes—one collapse, one sagging—using GPR to map boundaries and 3D geometry, ERT to image low‑resistivity fill, and trenching to record stratigraphy and strain, while evaluating factors affecting GPR quality and using numerical dating to estimate subsidence rates. The integrated data revealed the sinkhole edges, subsidence magnitude, and dominant mechanism, and demonstrated that such datasets can objectively forecast future sinkhole behavior and assess hazard and risk. © 2013 John Wiley & Sons, Ltd.

Abstract

ABSTRACT This paper explores the suitability and advantages of combining the trenching technique with geophysical surveys [ground penetrating radar (GPR) and electrical resistivity tomography (ERT)] for sinkhole characterization in a mantled karst area. The approach is applied to two active sinkholes concealed by anthropogenic deposits and formed by contrasting subsidence mechanisms; collapse and sagging. The ERT section acquired across the collapse sinkhole images the clayey fill of the depression as an obvious low resistivity area, showing the approximate location of the sinkhole edges. Spatially dense GPR surveys provide information on the position of the boundaries of the concealed subsidence structures and their three‐dimensional (3D) internal geometry, revealing the dominant subsidence mechanism. We illustrate the impact of several factors on the quality of the GPR data such as sinkhole size, nominal frequency of the antennas, antenna shielding, and the presence of backfilled excavations and above‐surface objects. Trenches provided detailed information on the subsurface structure of the sinkhole, subsidence magnitude, partitioning of the strain, and the position of the sinkhole edges, especially when they are deep enough and excavated across the central sector and perpendicular to the boundaries. The stratigraphic and structural relationships observed in the trench were then used to infer the spatial evolution of the sinkholes (e.g. enlargement), their kinematic behavior (episodic versus progressive), and to differentiate discrete subsidence events and their associated magnitude. Numerical dates were used to estimate average subsidence rates and the recurrence of subsidence events. Such integrated data sets may be used as an objective basis to forecast the future behavior of potentially damaging sinkholes and to assess the associated hazard and risk. Copyright © 2013 John Wiley & Sons, Ltd.