Investigations of electromagnetic methods applied to the in-river and near-river environment along the Murray River, Australia.

Abstract

In the last decade more than 2500 km of the Murray River in southeastern Australia and its surrounding floodplains have been surveyed using a variety of geophysical techniques to assess the environmental impact of changing salinity conditions along the river. These have included a combination of ground-based and airborne electromagnetic (EM) techniques. Along the Murray River, particularly in the lower reaches, the near-surface conductivity ranges from being moderately resistive to highly conductive, with groundwater conductivities approaching that of seawater in many areas. Such conditions present challenges for the application of most EM techniques, particularly as the signal attenuates rapidly resulting in little signal penetration. For all EM and DC-resistivity techniques, careful selection of frequency range and/or system geometry is therefore necessary to collect information over depth ranges of interest. An understanding of tradeoffs between frequency range and/or system geometry, resolution, and penetration depth is arguably critical to technique selection at the outset of the survey, and to the application of appropriate data processing and interpretation procedures post-survey. This thesis examines and evaluates a number of geophysical techniques for their suitability in characterising the shallow subsurface beneath and adjacent to the Murray River. Results for three inductively and galvanically based geophysical techniques used to collect data in the in-river setting are compared over the same stretch of river. Two are variations on ground-based systems, adapted to collect data nearly continuously while being towed behind a boat. These are a time domain EM (TEM) system and a DC-resistivity system. The third is a high-resolution, helicopter-based, airborne EM (AEM) system. In this study, the AEM system was demonstrated to provide a similar horizontal resolution as that provided by river-based systems. In addition, although the river-based techniques were able to image to depths of approximately 20 m, the AEM system was able to image to approximately 25 m depth in highly conductive areas, and to approximately 60 m depth in more resistive areas. The AEM system was also able to collect data over floodplains adjacent to the river, providing contextual information where needed. A new methodology for processing ground penetrating radar (GPR) data collected over conductive ground is developed. As part of this development, the transition band in the EM spectrum is defined, using the loss tangent to define parameter boundaries. In the transition band (ranging from approximately 110 kHz to 57 MHz for typical soil conditions), both conductive and displacement currents are important as conduction mechanisms, and need to be accounted for when calculating phase velocity and skin depth. Data from three 25 MHz GPR traverses, varying in length from less than 3 km to more than 16 km, were collected over conductive floodplains. These data are processed using this new methodology, and interpreted in conjunction with ground TEM data. After processing, the GPR is able to provide useful information about the depth to the top of the capillary fringe, as well as the location of nearsurface flush zones. This information correlates well with other data sets collected in conjunction with the GPR.