Using 2D inversion of magnetic resonance soundings to locate a water-filled karst conduit

• Published in: Journal of hydrology (Amsterdam) Vol. 330; no. 3; pp. 413 - 421
• Main Authors: Boucher, M., Girard, J.-F., Legchenko, A., Baltassat, J.-M., Dörfliger, N., Chalikakis, K.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 15.11.2006; Elsevier Science; Elsevier
• Subjects: analytical methods; aquifers; Causse de Gramat; Earth Sciences; Earth, ocean, space; Environmental Sciences; Exact sciences and technology; Geophysical methods; geophysics; Global Changes; groundwater; Hydrology; Hydrology. Hydrogeology; Karst; karsts; magnetic resonance sounding; MRS; Sciences of the Universe; water-filled karst conduits; France; Western Europe; France, Causses; Causse de Gramat; Karst; Geophysical methods; MRS; models; inverse problem; orientation; carbonate rocks; Europe; caves; Causse de gramat; surveys; accuracy; ERTS; karst; resonance; sounding; heterogeneity; sedimentary rocks; limestone; interpretation; tomography; noise; data acquisition; geophysical methods; errors; electrical resistivity
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2006.03.034

A new methodology for magnetic resonance sounding (MRS) data acquisition and interpretation was developed for locating water-filled karst cavities. This methodology was used to investigate the Ouysse karst system in the Poumeyssens shaft in the Causse de Gramat (France). A new 2D numerical MRS response model was designed for improved accuracy over the previous 1D MRS approach. A special survey performed by cave divers confirmed the accuracy of the MRS results. Field results demonstrated that in favourable conditions (a low EM noise environment and a relatively shallow, large target) the MRS method, used with a coincident transmitter/receiver loop, can be an effective tool for locating a water-filled karst conduit. It was shown numerically that because an a priori orientation of the MRS profile with the karst conduit is used in the inversion scheme (perpendicular for instance), any error in this assumption introduces an additional error in locating the karst. However, the resulting error is within acceptable limits when the deviation is less than 30°. The MRS results were compared with an electrical resistivity tomography (ERT) survey. It was found that in Poumeyssens, ERT is not able to locate the water-filled karst. On the other hand, ERT provides additional information about heterogeneities in the limestone.