f Combined use of tem and RMT for the investigation of waste disposal sites

About 25 km northwest of Hannover at Mellendorf a former sand pit was filled with industrial waste between about 1960 and 1980. The contents is mainly magnesium drosses. In 1987 the top of the site was sealed with a 1.5 m thick cover to avoid penetration of water into the waste. This was important because the bottom of the disposal at a depth of about 15 m below surface was not sealed. 1996 and 1997 TEM- (Transient Electromagnetic) and RMT-(Radio Magnetotelluric) measurements were carried out over the deposit. Goal of the combined use of RMT and TEM was a complete investigation of the vertical and horizontal resistivity distribution within the deposit and the directly surrounding area. The RMT measurements covered the deposit with a 10 m by 10 m grid, using four frequencies between 16 kHz and 200 kHz. For each frequency a pair of transmitters was chosen perpendicular to each other and located in the directions of the long and short axis of the deposit. For the TEM measurements the Protem 47 unit (Geonics) was used in an inloop configuration, the receiver being located at the centre of 50 m * 50 m transmitter loops. The interpretation of the data was performed in three steps. In a first step a one-dimensional inversion of the TEM data was used based on a three layer starting model. In the same step the interpretation of the RMT-data was based on a two-dimensional model. Secondly, for a further explanation of the RMT and the TEM data three-dimensional models were applied, based on the results of the preliminary interpretation of each method separately. Finally a three-dimensional model fitting both data sets was developed. The basic principle for determining the optimum model for both methods was to use the RMT-results for the top layers down to a depth of approximately 17 m and below that the results from the TEM data. This led to a final model combining the high resolution of the RMT method for the top layers and the better resolution at greater depth penetration of the TEM system. This model was further optimized, using trial and error, the only possible way at the moment. Figure 1a shows the combined model, which uses two different resistivity distributions for the western and eastern part of the deposit. The interesting result is that the high conductivity of the deposit extends to a greater depth than that of the former sand pit. The most likely explanation of the extension of the low resistivity below the known deposit depth is a leakage of waste water from the deposit into the ground below. To prove the validity of this result, modeling tests were undertaken with a reduced depth extend of the conducting layers for both sections of the deposit, the western (fig. 1b) and the eastern part (fig. 1c). Both models could not satisfactorily explain the measured data. This verifies with increased reliability the resistivity distribution of the optimized model shown in fig. 1a.