RESISTIVITY TENSOR PROBABILITY TOMOGRAPHY

The probability tomography approach developed for the scalar resistivitymethod is here extended to the 2D tensorial apparent resistivity acquisitionmode. The rotational invariant derived from the trace of the apparentresistivity tensor is considered, since it gives on the datum plane anomaliesconfined above the buried objects. Firstly, a departure function is introducedas the difference between the tensorial invariant measured over the realstructure and that computed for a reference uniform structure. Secondly, aresistivity anomaly occurrence probability (RAOP) function is defined as anormalised crosscorrelation involving the experimental departure function and ascanning function derived analytically using the Frechet derivative of theelectric potential for the reference uniform structure. The RAOP function canbe calculated in each cell of a 3D grid filling the investigated volume, andthe resulting values can then be contoured in order to obtain the 3Dtomographic image. Each non-vanishing value of the RAOP function is interpretedas the probability which a resistivity departure from the reference resistivityobtain in a cell as responsible of the observed tensorial apparent resistivitydataset on the datum plane. A synthetic case shows that the highest RAOP valuescorrectly indicate the position of the buried objects and a very high spacialresolution can be obtained even for adjacent objects with opposite resistivitycontrasts with respect to the resistivity of the hosting matrix. Finally, anexperimental field case dedicated to an archaeological application of theresistivity tensor method is presented as a proof of the high resolution powerof the probability tomography imaging, even when the data are collected innoisy open field conditions.Comment: 8 pages, 7 figure