Simultaneous inversion of magnetic resonance sounding in terms of water content, resistivity and decay times

Magnetic resonance sounding (MRS) or surface nuclear magnetic resonance (SNMR) is used for direct groundwater exploration and for an improved aquifer characterization. Currently, it is the only geophysical method that is capable of directly determining the free water content and estimating the pore sizes of the aquifer in the subsurface. However, MRS is basically an electromagnetic method. Therefore, it is sensitive to the resistivity of the subsurface. The water content is the main target of investigation, therefore first inversion routines focused on the water content. Later on, inversion routines determining water content and decay times became available. Very recently, MRS inversion for water content and resistivity has been realized. We present here a simultaneous inversion of MRS in terms of determining the three inversion parameters – water content, resistivity and decay time – within one single inversion routine. Within the iterative inversion scheme, the extrapolated initial values are determined on the basis of the physical effective decay times in the subsurface, which are estimated within the inversion scheme. Due to an instrumental dead time, the initial values for amplitude and phase, which are related to water content and resistivity, cannot be measured directly. Therefore, the initial amplitude must be extrapolated using the decay time of the signal. The standard approach is a mono-exponential decay curve; implicitly, the phase is assumed to be time-invariant. However, multi-exponential signals are natural when considering relaxation behaviour in the underground. It originates from multi-modal pore size distributions or simply a number of differently relaxing signal contributions from the various lithological units.

[1]  Ugur Yaramanci,et al.  Inversion of resistivity in Magnetic Resonance Sounding , 2008 .

[2]  Anatoly Legchenko,et al.  A review of the basic principles for proton magnetic resonance sounding measurements , 2002 .

[3]  G. Ghiglieri,et al.  Geophysical approach to the environmental study of a coastal plain , 2003 .

[4]  Anatoly Legchenko,et al.  Processing of surface proton magnetic resonance signals using non-linear fitting , 1998 .

[5]  Weichman,et al.  Theory of surface nuclear magnetic resonance with applications to geophysical imaging problems , 2000, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[6]  Ugur Yaramanci,et al.  Forward modeling and inversion of MRS relaxation signals using multi-exponential decomposition , 2005 .

[7]  Thomas Günther,et al.  Surface Nuclear Magnetic Resonance Tomography , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[8]  D. V. Trushkin,et al.  Surface NMR applied to an electroconductive medium1 , 1995 .

[9]  A. Legchenko.  Magnetic resonance sounding: Enhanced modeling of a phase shift , 2004 .

[10]  Ugur Yaramanci,et al.  Smooth and block inversion of surface NMR amplitudes and decay times using simulated annealing , 2002 .

[11]  Sandro Ridella,et al.  Minimizing multimodal functions of continuous variables with the “simulated annealing” algorithmCorrigenda for this article is available here , 1987, TOMS.

[12]  J. Roy,et al.  The magnetic resonance sounding technique and its use for groundwater investigations , 2003 .

[13]  Ugur Yaramanci,et al.  Pore size distributions and hydraulic conductivities of rocks derived from Magnetic Resonance Sounding relaxation data using multi-exponential decay time inversion , 2008 .