|
Abstract:
|
The use of microprocessor-controlled data-acquisition systems allows the rapid collection of electrical-resistivity and induced-polarization data. Although this provides a method for cost-effective collection of multidimensional data sets, there can be significant sources of error introduced by these acquisition systems. These errors are systematic in nature and thus may be difficult to identify. Of particular concern are errors resulting from re-use of the same electrodes for transmitting and receiving and aging of those electrodes over time. In this paper, we consider results from laboratory sand-tank experiments and field studies at a number of electrical resistivity project sites. The field studies include measurements over an eight year period during the Drift-Scale Heater Test at Yucca Mountain, Nevada. The object of these studies was to understand the ability of the site to perform over a long period of time and to test a number of methods of mitigating both systematic and non-systematic errors for real-world applications. A number of different data strategies are being tested including the effects of varying the delay time of measurements using the same electrodes and the impact of using constant-current versus constant-voltage transmitter sources. The initial results show that the relationships between errors and measurement strategies are not always intuitive. For example, a longer delay between transmitting and receiving on a given electrode pair may not necessarily reduce the errors. In addition, electrodes show long-term changes in error levels that do not show a consistent pattern or correlation to other factors.
|
