Complex Inversion Of Surface-Nmr Signals - Extending The Limits Of Model Resolution

The technique of Surface Nuclear Magnetic Resonance (SNMR, also known as MRS - Magnetic<br>Resonance Sounding) is used for direct groundwater determination and aquifer characterization. Among<br>other influences, the electrical conductivity of the subsurface leads to a complex-valued signal. However,<br>the standard interpretation scheme uses only the amplitude of the signal for determining the water content.<br>But real and imaginary parts of the signal are sensitive to different depth volumes. Generally, the imaginary<br>part is more sensitive to deep structures than the real part of the signal, i.e. in conductive media, signals<br>arising from deep layers have a significantly greater imaginary part than an equivalent signal from shallow<br>depths. For real data, physical effects additionally to the electromagnetic phase delay have to be considered<br>and adequately quantified to use the phase information for an enhanced data interpretation. This study<br>assesses the complex inversion using real and imaginary parts of the signal. Analyzes of synthetic and real<br>data with sufficient data quality show that the complex inversion is more reliable in terms of determining<br>deep structures, equivalence errors are reduced, and the depth resolution is increased.