Electromagnetic induction methods reveal wetland hydrogeological
structure and properties
Abstract
Understanding sensitive wetlands often requires non-invasive methods to
characterize their complex geological structure and hydrogeological
parameters. Here, geoelectrical characterization is explored by
employing frequency-domain electromagnetic induction (EMI) at a site
previously characterized by extensive intrusive measurements and 3D
electrical resistivity tomography (ERT). This work investigates the
performance of several approaches to obtain structural information from
EMI data and sharp and smooth inversions. Additionally, the hydrological
information content of EMI data is investigated using correlation with
piezometric measurements, established petrophysical relationships, and
synthetic modeling. EMI measurements were dominated by peat thickness
and were relatively insensitive to both topography and depth to bedrock.
An iso-conductivity method for peat depth estimation had a normalized
mean absolute difference (NMAD) of 23.5%, and although this performed
better than the sharp inversion algorithm (NMAD = 73.5%), a
multi-linear regression approach achieved a more accurate prediction
with only 100 measurements (NMAD = 17.8%). In terms of hydrological
information content, it was not possible to unravel correlation
causation at the site, however, synthetic modeling demonstrates that the
EMI measurements are predominantly controlled by the electrical
conductivity of the upper peat pore-water and not the thickness of the
unsaturated zone or the lower peat pore-water conductivity.
Additionally, a priori information significantly improves the potential
for time-lapse applications in similar environments. This study provides
an objective overview and insights for future EMI applications in
similar environments. It also covers areas seldom investigated in EMI
studies, e.g. error quantification and the depth of investigation of ERT
models used for EMI calibration.