A Dimensionally Adapted Method for the Efficient Simulations of Geophysical Electromagnetic Measurements

No photo
BCAM principal investigator: David Pardo
BCAM research area(s) involved:
Reference: MTM2016-81697-ERC
Coordinator: UPV/EHU - University of the Basque Country
Duration: 2017 - 2018
Funding agency: MINECO - Projects R&D&i - Excellence Europe
Type: National Project
Status: Closed

Objective:

This Project (EMEARTHSIM) aims at developing more efficient simulation methods of geophysical electromagnetic (EM) measurements for the characterization of the materials composing the Earth’s subsurface. We will focus on the development of fast and accurate forward simulations that can be efficiently integrated in a gradient-based inversion method. The main feature of the proposed simulation method is that it will be able to efficiently deal with geological models composed of various subdomains exhibiting different spatial dimensionality (e.g., a one-dimensional (1D) subdomain combined with a three-dimensional (3D) one on a different zone of the domain). For subdomains expressed in terms of a 1D model, we will employ a Finite Element (FE) method that selects Bessel basis functions combined with multiscale basis functions along the direction where the material properties vary. For higher dimensional simulations, we well employ a recently developed “refined Isogeometric Analysis” (rIGA) method. The discretization error will be controlled via a novel goal-oriented p-adaptive algorithm that employs unconventional error representations. We shall consider magnetotellurics (MT), controlled source electromagnetics (CSEM) and borehole resistivity measurements (acquired with the most advanced logging-while-drilling, and geosteering deep azimuthal tools). MT measurements of Lorca (earthquake assessment) and Hontomin (CO2 sequestration) are available to us via our collaboration with the group of P. Queralt (Univ. of Barcelona). We also collaborate with the oil company TOTAL S.A. in the context of a RISE European Project that we coordinate. During this Project we shall intensify our current conversations with REPSOL in order to collaborate more closely with this company. The successful development of this project entails an interdisciplinary knowledge on applied mathematics, high performance computing (HPC), and geophysics.