A Method For Integrating Ground And Heliborne Time-Domain Electromagnetic Data Using A Gradual Deformation Method Over St-Lawrence Lowlands (Quebec,Canada)
F.B.J. Yrro, M. Chouteau and E. Gloaguen
Event name: SAGEEP 2016 - 29th Annual Symposium on the Application of Geophysics to Engineering and Environmental Problems
Session: Near Surface Data Analysis
Publication date: 21 March 2016
Info: Extended abstract, PDF ( 422.89Kb )
We propose a new approach for integrating ground-based and heliborne time-domain electromagnetic data over the St-Lawrence Lowlands (Quebec, Canada) that allows recovering the resistivity model of the subsurface satisfying both collected data. The approach uses discrete and continuous geostatistical simulations and the gradual deformation method (GDM). First, 1D layered-earth inversion of VTEM data with 15 layers is computed in order to assess the vertical variability of the resistivity. The 1D output electrical model clearly shows that the first 3 layers are sufficient to explain the data. As expected over the study area, the first and the third layer of the 1D model are more resistive than the second layer. Based on a three facies-type geological interpretation of 1249 wells, we assume that the first 3 layers are respectively the alluvial sand, Champlain sea clay and the shale bedrock. Secondly, 3D geological model of the above 3 facies is built within the a 3D GIS software using the wells and a surface quaternary geological map. The 3D model is then used as a training image for conditional multipoint simulation (MPS) allowing to generate multiple geological scenarios. Each geological scenario is then populated with electrical resistivity from VTEM-plus inversion using conditional sequential gaussian simulation (SGS). Finally, these geo-electrical models are combined by a gradual deformation method in an optimization process to match the ground-based electromagnetic data.