Dépôt numérique
RECHERCHER

Integrating Geophysics and Soil Sampling for Site Characterization: A Kernel Approach.

Fabien-Ouellet, Gabriel; Gloaguen, Erwan et Plassart, Gaël (2017). Integrating Geophysics and Soil Sampling for Site Characterization: A Kernel Approach. Journal of Environmental & Engineering Geophysics , vol. 22 , nº 3. pp. 305-308. DOI: 10.2113/JEEG22.3.305.

Ce document n'est pas hébergé sur EspaceINRS.

Résumé

Reconstructing the variation of contaminant concentration with a limited number of soil samples is more or less the norm, even though it fails more often than not for problems of even moderate complexity. To overcome the limits inherent to discrete measurements, we propose to integrate soil sampling with continuous surface geophysical measurements in a geostatistical framework. We present this integrated analysis for a PAH contaminated site in France. For the study site, two 3D surveys were acquired: an electrical resistivity tomography survey and a seismic travel time tomography survey. Those two surveys permitted us to infer two spatially continuous physical properties on the whole volume, namely the electrical resistivity and P-wave velocity. The probability density function relating the velocity-resistivity pairs with each of the 75 lab measurements of PAH concentration was modeled using a Gaussian kernel. This probability density function combined with the 3D volumes of resistivity and P-wave velocity provided a means to translate the latter into a 3D map of PAH concentration. This 3D map of concentration was then used as a secondary variable in a cokriging simulation of the 75 lab samples, thus reintroducing the spatial correlation of the initial dataset. Comparing this final 3D PAH concentration model with the simple kriging of the PAH samples, the geophysical integrated model reproduce much better the distribution of measured concentration, shows a much more realistic spatial pattern of the contamination, and lowers the estimated contaminated volume.

Type de document: Article
Mots-clés libres: computer simulation; electrical resistivity; Gaussian method; geophysics; geostatistics; integrated approach; kriging; P-wave; PAH; physical property; probability density function; site characterization; soil analysis; soil pollution; tomography; travel time; wave velocity
Centre: Centre Eau Terre Environnement
Date de dépôt: 08 mai 2018 14:30
Dernière modification: 08 mai 2018 14:30
URI: https://espace.inrs.ca/id/eprint/6553

Gestion Actions (Identification requise)

Modifier la notice Modifier la notice