The basic idea behind this work is to find metallic objects in landfills that would be worthwhile for possible recycling processes. Using the non-invasive measurement methodology of geoelectric, large areas can be investigated from the surface and thus targeted individual areas can be located that are supposed to have a sufficiently high metal content. This means that large-scale test excavations in the landfill area can be dispensed with, as the areas of interest can already be locally delimited beforehand.
In the introduction, the theoretical background of geoelectrical measurements is given, with a special focus on the methodology of induced polarisation. Classical electrical resistivity tomography was included in parallel for each measurement carried out and is used to interpret the IP results. Apart from that, this data will not be discussed further as the focus is clearly on the induced polarisation results. Afterwards, literature examples and the key points of the Interreg North-West Europe RAWFILL project are briefly presented.
The main part of the paper deals with the field measurements carried out. The landfill in Allerheiligen in the Mürz valley, where mechanical-biological residual waste, so-called MBT material, is deposited, was chosen as the measurement site. The measurements were deliberately taken in the oldest part of the landfill, where since 1996 no new waste has been deposited. However, previous investigations showed a relatively low metal content in the landfill area, which is why a metallic gas dome was used as a measurement target. This is representative of larger metallic objects in the landfill subsoil. Two separate measurement runs were carried out on two different gas domes in May 2021 and November 2021. This should provide a broader picture of the influence of the landfill body on the measurement, as aspects such as soil moisture or the composition of the landfilled waste can have non-negligible effects on the IP data along the profile lines drawn. In addition, two different measuring devices were used to measure in the time-domain as well as in the frequency-domain. Furthermore, measurements were made with Wenner and Dipole-Dipole configurations as well as with classical steel electrodes and non-polarisable electrodes in order to investigate all aspects of the influence of certain factors on the measurement results. The main part of the thesis deals with the evaluation of the 100 field measurements carried out, which includes a quantitative part dealing with the statistical evaluation of the raw data and a more qualitative part dealing with the interpretation of the structures in the inversion models generated. In addition, comments are made on the aborted laboratory test.
Finally, an interpretation of the measurements and a presentation of the information obtained as well as the limitations of this measurement methodology and an outlook with regard to suggestions for improvement for future measurement experiments are given.
The evaluation of the phase angles, which have been used as a significant indicator for the IP effect, shows a consistent distribution between 10 and 50mrad on average for all measurements performed with the Lippmann 4 point light 10W in the frequency-domain. Those measurements with Wenner configuration show a more constant value distribution and a lower scatter compared to measurements with dipole-dipole configuration. The dU90 values, which were also analysed, allow conclusions to be drawn about the general measurement quality and also yield more favourable values for the Wenner configuration in a comparison of the two measurement configurations. In contrast, no statistical analyses were performed for the measurements in the time-domain that were measured with the Mangusta System 24/144E, since the raw data contain a low density of measurement information. However, the raw data had to be modified in the course of the inversion analysis because very high iteration errors were revealed in the first run. Therefore, all chargeability values above 100ms were deleted, as these are considered to be erroneous measurements, and thus the iteration error could be shifted into an acceptable range.
Compared with the evaluations from literature examples, there are similarities in the data. Therefore, their distribution can be classified as valid and applicable for landfills with a similar structure as the one in Allerheiligen in the Mürz valley.
By differentiating the measurements with respect to the measurement settings and electrode types used, the results can be used to make a recommendation for IP measurements using the Wenner configuration and the non-polarisable electrodes for measurements with similar environmental parameters.