Electromagnetic methods can be divided into two large classes. The first, called frequency-domain measurements, make use of a sinusoidally varying current in the transmitter. The secondary magnetic field produced by the induced currents is measured in the presence of the primary magnetic field. There is typically a phase shift between the primary and secondary fields that allows them to be separated for analysis. By varying the frequency of the transmitter current the depth of investigation can be changed with lower frequencies penetrating deeper into the ground. The second class of electromagnetic geophysical measurements are called time-domain or transient methods. With these measurements, the transmitter current is usually held constant for some length of time and then abruptly turned off. The rapid decrease in current and the accompanying collapse of the associated magnetic field induces current flow in the ground. A receiver coil measures the secondary magnetic field produced by this decaying current system in the ground. The measurement is made while the primary magnetic field is absent. In principle, time-domain and frequency-domain measurements should be equivalent, however, the way noise affects the two types of measurements is quite different, leading to advantages for each method (Kaufman and Keller, 1983).

While electromagnetic methods were originally developed for mineral exploration, they have been widely used over the past 20 years for ground-water investigations. Details can be found in a number of publications (Stewart, 1982; Stewart and Gay, 1986; Fitterman and Stewart, 1986; Fitterman, 1987; McNeill, 1990; Fitterman and Labson, 2005). In the work presented here we have used three electromagnetic geophysical techniques: 1) helicopter electromagnetic (HEM) surveying, time-domain electromagnetic (TEM) sounding, and borehole induction logging. The HEM and borehole techniques are frequency-domain methods, while TEM sounding is a time-domain method.

HEM surveying is done using electromagnetic transmitter and receiver coils that are housed in a 10-m long instrument assembly called a bird. The bird is slung below the helicopter on a 30-m long cable and it is flown at a nominal altitude of 30 m. The survey area is covered with flight lines oriented at N50°E and spaced 400 m apart. A measurement is made every second with spacing on the order of 4 to 15 m along flight lines. The birds commonly make measurements at five or more frequencies. Data are used to estimate layered earth models at each measurement point. The resulting conductivity-depth functions probe to depths varying from 15 to 80 m depending upon the average conductivity value. In conductive regions the depth of exploration is shallower than in resistive (less conductive) regions. HEM data serve as our primary source of subsurface information because of the high sampling density.

Transient soundings were used to obtain an independent estimate of conductivity-depth variation, as well as providing a means of calibrating the HEM instrumentation (Deszcz-Pan et al. 1998). Soundings were made using a central induction array where the 40-m by 40-m transmitter loop surrounds the receiver coil. Depth of exploration varies from 40 to 100 m (Fitterman et al., 1999). Inversion methods were used to convert the apparent resistivity-time data into interpreted resistivity-depth models.

The last electromagnetic method used is induction logging. This is a frequency-domain method that uses a small receiver and transmitter coil housed inside a non-conducting probe that is lowered down a PVC-cased borehole (Hearst et al., 2000). The recorded secondary field is converted to conductivity of the material surrounding the borehole and is measured continuously as the probe moves through the borehole. The induction tool senses about a meter into the surrounding formation. The induction logs were used to establish a relationship between the formation conductivity and the specific conductance (SC) of the pore water. Knowing this relationship and how SC is related to chloride content, we can estimate water quality from the HEM-derived conductivity models.