Subsurface Ice Detection via Low Frequency Surface Electromagnetic Method

Abstract:

The geophysical detection of ice in the Cryosphere is typically conducted by measuring the absence of water. These interpretations can become non-unique in dry soils or in clay- and silt-rich soils that contain significant quantities of unfrozen water. Extensive laboratory measurements of electrical properties were made on permafrost samples as a function of frequency, temperature, and water content. These laboratory measurements show that the amount of ice can be uniquely obtained by measuring a frequency dependence of the electrical properties over a large frequency range (20 kHz - 10 Hz). In addition, the electrical properties of permafrost are temperature dependent, which can allow for an estimate of subsurface temperature. In order to test this approach in the field, we performed field surveys at four locations in Alaska. We used three low frequency electromagnetic methods: Spectral Induced Polarization (SIP: 20 kHz - 10 Hz), Capacively Coupled Resistivity (CCR: OhmMapper - 16.5 kHz), and DC Resistivity (Syscal ~ 8 Hz). At the Cold Regions Research and Engineering Laboratory permafrost tunnel near Fox, AK, we used SIP to measure the average ice concentration of 80 v% and determined the temperature to be –3±1°C by matching survey results to lab data. SIP data acquisition is very slow; therefore, at three sites near Tok, AK, we used CCR to perform reconnaissance of the area. Then SIP and DC resistivity were performed at anomalous areas. The three survey types give very similar absolute resistivity values. We found that while SIP gives the most quantitative results, the frequency dependence from the CCR and DC resistivity surveys is all that are needed to determine ice content in permafrost.