Possibilities and limitations of cave detection with ERT

Abstract

About 15 % of the global ice-free continental surface is characterized by the presence of karstifiable carbonate rock. In Austria, about 20 % of the surface area consists of water-soluble rocks such as limestone or dolomite. Karstification may result in the formation of various cavities with varying size and distribution. Karst massifs are highly complex systems in which the water moves along different geological discontinuities that range in a broad spatial scale, from intragranular cracks (in the micro- to centimeter range) to fractures and cave systems, which can extend over kilometers. Electrical resistivity tomography (ERT) is widely used for cave detection, yet associated limitations have not been addressed so far. This study investigates subsurface conditions favoring the detection of cavities with ERT as well as the geological settings limiting the method. In a first step, we use a numerical study to evaluate the ability of the inversion method to recover the geometry of well-known resistivity anomalies, by varying the resistivity values of the host rock. Synthetic data is used to understand the contrasts required between the anomaly and the host rock. In a second step, we present extensive ERT data collected to delineate the geometry of the “Forststraß eneinbruch” cavity. This area is located near Lunz am See in the Northern Calcareous Alps and was first mentioned in December 2016 due to a sinkhole formed on a forest road. The geometry of the accessible parts of the cave was recorded using a speleological survey and could be used to evaluate the field results. Moreover, ERT result can be used to detect other cavities not found during the speleological survey. By using the speleological survey data we could create realistic models of the subsurface, from which synthetic data were derived to understand the ability of the inversion to retrieve a well-known model. The ERT images show in general a good agreement with the known geometry of the cave and could be validated through the synthetic models and the further available information. However, the size of the anomalies point out to the presence of further caves. The synthetic results revealed that the resistive anomaly related to the cave is better resolved in presence of a resistive host rock. While decreasing the resistivity of the host rock leads to a decrease also in the sensitivity of the method to sense the cave, likely due to the ability of the current to easily move around the resistive anomaly.