Köstel, Johannes: Quantitative characterization of solute transport processes in an undisturbed unsaturated soil by means of electrical resistivity tomography (ERT). - Bonn, 2010. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5N-22163
@phdthesis{handle:20.500.11811/4220,
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5N-22163,
author = {{Johannes Köstel}},
title = {Quantitative characterization of solute transport processes in an undisturbed unsaturated soil by means of electrical resistivity tomography (ERT)},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2010,
month = jul,

note = {Improved understanding of water flow and solute transport through the unsaturated zone is important for the sustainable management of soils. As soils are complex and heterogeneous systems, quantification of the transport processes is difficult. More knowledge on the relationship between solute transport process, soil structure, hydrologic initial and boundary conditions, and observation scale is needed here. Therefore, non-invasive quantitative 3-D spatio-temporal imaging of solute displacement is necessary. Such imaging may be provided by hydro-geophysical methods, for example electrical resistivity tomography (ERT). Recent studies have shown that ERT is able to visualize solute transport through the subsurface with 3-D resolution, but quantitative interpretations of the ERT-image data are difficult. This thesis explores the potential of ERT to quantitatively characterize vadose zone solute displacement with 3-D spatio-temporal resolution. Four inert solute transport experiments in a large undisturbed unsaturated soil column were imaged using 3-D ERT. Two experiments were conducted with an irrigation rate of 1.5 cm/d, the remaining two with an irrigation rate of 6.5 cm/d. An improved ERT-data-error estimation approach was introduced. Time-domain reflectometry (TDR) was applied to quantitatively ground-truth the ERT-image data. The experimental setup allowed translating the ERT-derived bulk electrical conductivity to solute concentration without neglecting the 3-D structure of the petrophysical properties of the soil. Using these innovations, ERT recovered the mass of the applied tracer with an error of approximately 5% for all four displacement experiments. The solute transport was characterized by means of ERT-derived apparent convection-dispersion equation (CDE) parameters. The apparent velocities were barely affected by the choice of the error level used in the ERT inversion but the apparent dispersivities were. The apparent velocities were less impaired by temporal smearing than the apparent dispersivities. The mean apparent velocity appeared to be not biased by ERT-sensitivity in contrast the velocity variability. The apparent dispersivities were barely correlated with decreasing ERT sensitivity. In an additional displacement experiment, the food dye Brilliant Blue was directly imaged using ERT. Here, advantage was taken of the negative ionic charge of Brilliant Blue molecule under weakly acid and basic conditions. A comparison of photographically obtained Brilliant Blue staining patterns with respective ERT-derived solute electrical conductivity patterns showed that both patterns coincided as long as the staining features were not cut off from the electrodes. The ERT-derived 3-D solute concentration images were useful to quantitatively investigate and compare the solute transport of the four inert solute displacement experiments. It was found that the main velocity patterns remained invariable for all four experiments despite a considerable increase in water content between the low and the high flux experiments. Another outcome was that all transport velocity patterns were aligned to a structural feature in the topsoil which, in turn, was aligned to the plowing direction and, therefore, identified as a man-made structure. The mixing regime of all displacement experiments was found to be convective-dispersive despite considerable lateral variations in the local transport velocity. The voxel-scale ERT-derived apparent dispersivity was successfully laterally up-scaled to the column scale using a newly introduced up-scaling approach. Furthermore, the up-scaling approach was useful to investigating the relationship between lateral observation scale and apparent dispersivity.},
url = {https://hdl.handle.net/20.500.11811/4220}
}

Die folgenden Nutzungsbestimmungen sind mit dieser Ressource verbunden:

InCopyright