Some considerations on electrical resistivity imaging for characterization of waterbed sediments

The paper focuses on defining the performance and limits of ERI in the detection and sedimentary characterization of near-bottom thin layers. The analysis of the resolution of floating and submerged cables, and the effect of the accuracy of a priori information (resistivity and thickness) in the data inversion, is based on theory, models and actual data. Theoretical models show that the actual reconstruction of the near water-bottom sediments, in terms of geometry and resistivity, can be obtained only with the submerged cable, however, the data, unlike that acquired with the floating cable, require a priori information on water resistivity and thickness for the data inversion. Theoretical forward models based on wrong a priori water thickness and resistivity information influence the inverted model in different ways, depending on the under- and over-estimation of water resistivity and thickness, and the resistivity contrast of the water–solid layer; however a water–solid resistivity contrast of less than 2 and within 10% of error in water resistivity has no effect. Overestimating water resistivity depicts a ground similar to the actual ground in terms of resistivity, more so than the underestimation of water resistivity. Moreover, the data inversion is less influenced by water parameter error in the case of low resistivity contrast in the water–solid layer, than it is for high resistivity contrast. Wenner and Schlumberger arrays give comparable results, while a dipole–dipole array seems to be more sensitive to the accuracy of apparent resistivity measurements and a priori information on water. The theoretical considerations were validated by actual data acquired with a submerged cable on the Tiber River. The study has shown that if highly accurate measurements are made of water thickness and resistivity, then electrical resistivity imaging from the submerged cable can be used in addition to, or even to substitute, seismic data for the reconstruction of the features and sedimentary characterization of near-bed sediments where seismic data fail to give a suitable resolution.