Publication date: 15 December 2016
Source:Geoderma, Volume 284
Author(s): Olivier Guerrero, Jean-François Lataste, Antoine Marache
This paper presents an innovative system which adapts Electrical Resistivity Tomography (ERT) for fast investigation on long transects. The device consists in a set of on-the-go sensors allowing high sampling rate measurement for mobile prospections. The weak contact between the mobile sensors and the soil, to allow fast investigations, induces very noisy raw datasets. Kinematic Electrical Resistivity Tomography (named KERT) is then associated with a data treatment process. A two-step process reveals the information contained in the datasets. The first focuses on the procedure to remove noise through the use of a low-pass filter. The assessment of cutoff frequency for smoothing is deduced from a Fast Fourier Transform (FFT) analysis, thus the value is adapted to the site and the prospection. The second phase consists in calibration of the KERT on a conventional static ERT. At this point, a classical static ERT was used as a reference, and directed on a reduced part of the KERT transect. The estimation of a factor of calibration (Xc) to correct bias revealed the resistivity variations along the entire transect and in depth. In the last part of the paper, the validation of the investigation method and data treatment methodology is performed on a real case study: a 450 m transect is investigated in 5 min (mean speed: about 6 km/h), integrating 10 depths of investigation. The steps of data processing are detailed to show the final results, consisting in a zonation along the transect.
Source:Geoderma, Volume 284
Author(s): Olivier Guerrero, Jean-François Lataste, Antoine Marache