Study of 3-D SNMR Inversion Modeling

Abstract

Motivated by the recent application of the surface nuclear magnetic resonance (SNMR) technique to detecting and mapping of subsurface groundwater, and making use of a recently developed theory of the method, we consider the resulting 3-D inversion modeling. SNMR is a new technology which just passed the experimental stage to become a promising surface measurement tool to investigate directly the existence, amount and productiveness of ground water. SNMR can be generalized to have two observables: initial amplitude and decay time. The aim of inversion is to extract the information, i.e., the value and distribution of two physical parameters of subsurface: water content and subsurface properties (pore size, grain size). Improved inversion is not only needed for having better estimates of the subsurface properties. It also serves as a guide to design measurements, layouts and defines technical requirements for new hardware capable to measure physical properties, which have been shown to exist in the data and can therefore be extracted. For a reliable and appropriate inversion, we do need to assess the requirements and the possibilities. We then present an improved 3-D inversion algorithm using Levenberg-Marquardt algorithm. The inversion results place the water content feature at the correct depth and volume.