% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@PHDTHESIS{Zhou:885923,
author = {Zhou, Zhen},
title = {{E}nhanced crosshole {GPR} full-waveform inversion to
improve aquifer characterization},
volume = {512},
school = {RWTH Aachen},
type = {Dr.},
address = {Jülich},
publisher = {Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag},
reportid = {FZJ-2020-04179},
isbn = {978-3-95806-500-0},
series = {Schriften des Forschungszentrums Jülich. Reihe Energie
$\&$ Umwelt / Energy $\&$ Environment},
pages = {VIII, 136 S.},
year = {2020},
note = {RWTH Aachen, Diss., 2020},
abstract = {Complex heterogeneities in the aquifers are critical and
challenging to be detected and can have a significant effect
on subsurface flow and transport. Thereby, reliable
prediction of groundwater flow and solute transport is
important for the protection of drinking water, and the
remediation of contaminants. Small-scale high resolution
images of the subsurface can help to improve the
understanding of complex and heterogeneous aquifer
structures that effect hydrological properties and
processes. To successfully obtain hydrological parameters
distributed in a 2D cross-section with high resolution, we
can apply crosshole ground penetrating radar (GPR).
Crosshole GPR uses high-frequency electromagnetic pulses
that are emitted from a dipole-type antenna in a borehole
and recorded by a receiver antenna in a second borehole. The
received electromagnetic wave with its arrival time and
amplitude contains information about the subsurface medium
properties through which it travelled. Thereby, crosshole
GPR is able to provide two electromagnetic parameters the
dielectric permittivity and the electrical conductivity at
the same time. Conventional inversion approaches for
crosshole GPR data are generally based on geometrical ray
theory, which provide relatively smooth images with a
resolution that scales approximately with the diameter of
the first Fresnel zone. In contrast, the crosshole GPR
full-waveform inversion (FWI) provides decimeter-scalehigh
resolution images, because it considers the fully recorded
waveform information and the inversion is based on solving
the full Maxwell’s equations. However, the crosshole GPR
FWI approach also includes some limiting factors and
requires several detailed processing and inversion steps. If
these steps are not carefully applied, the inversion can be
trapped in a local minimum. In order to minimize the
influence of at least some of these factors, appreciate FWI
starting models and an accurate estimation of the effective
source wavelet are required. To precisely describe aquifers,
high porosity layers and clay lenses, that can strongly
effect flow and transport processes, need to be considered.
In the framework of this thesis, we first extend this
amplitude analysis approach to identify two different types
of low-velocity waveguides, caused by an increased porosity
and/or by a higher electrical conductivity. The obtained
information about extension and dimension of such wave
guiding structures is considered to improve the starting
models of the FWI. Moreover, we estimate an updated
effective source wavelet based on the updated permittivity
starting model. To verify the presented scheme, nine GPR
cross-sections were measured and analyzed at the
Hermalle-sous-Argenteau test site near Liege in Belgium.
Consistent structures between different cross-sections show
the robustness of the updated amplitude analysis approach
and the FWI results. In addition, the aquifer structures
obtained from the new FWI results agree with the crosshole
electrical resistivity tomography (ERT) monitoring [...]},
cin = {IBG-3},
cid = {I:(DE-Juel1)IBG-3-20101118},
pnm = {255 - Terrestrial Systems: From Observation to Prediction
(POF3-255)},
pid = {G:(DE-HGF)POF3-255},
typ = {PUB:(DE-HGF)3 / PUB:(DE-HGF)11},
url = {https://juser.fz-juelich.de/record/885923},
}