A Laboratory Seismoelectric Measurement for the Permafrost Model with a Frozen-unfrozen Interface

For the Qing-Cang railway line located in the permafrost region, the freeze-thaw cycling with the seasons and spring-thaw of the permafrost are main factors to weaken the railway bed. Therefore, the determination of the frozen-unfrozen interface depth below the railway bed is important for the railway operation, and moreover, it can contribute to the evaluation of the permafrost environment effected by the railway. Since the frozen-unfrozen interface is a contact of two media with various porosity and saturation, an electric double-layer can be formed at the interface by the absorption of electrical charge to it. When a seismic wave is incident at the interface, a relative motion of the charges in the electric double-layer would induce an electromagnetic (EM) wave, or a seismoeletric conversion signal that can be measured remotely, which is potential for determining the frost depth. A simple permafrost model with a frozen-unfrozen interface was built mainly by two parts: the upper part was a frozen sand block with a 7cm thickness and the lower one with the same material was in an unfrozen state saturated with water. And the contact of the two parts simulated the frozen-unfrozen interface. The interface model was placed in a freezer, while it was heated from the bottom with a heating sheet made by the electric heating wires laid under the unfrozen part. A P-wave source transducer with 48 kHz narrow band frequency was set on the top the frozen part and driven by a square electric pulse. The six electrodes with a 1 cm even interval were fixed inside the frozen part with 1 cm vertical distance to the interface. In the experiment, all the analog signals acquired from the temperature sensors, acoustic transducers, and electrodes were sent through preamplifiers and recorded digitally by computer-based virtual instruments (VIs). At the beginning of the experiment, the first arrivals of the seismoeletric signals observed from the six electrodes with minimum offset set to be 7cm were proportional to the distances between the acoustic sources to electrodes, and thus the EM signals are originated from the stationary electromagnetic field that moves along with the acoustic waves. After the eight hours, we recognized two new events of EM waves by their exactly identical arrive times from the six electrodes. The event A with identical arrival time being close to zero is the EM interference of the high-voltage pulse exciting the acoustic source transducer. The identical arrival time 23-25 microsecond of the event B roughly equates to that of the acoustic wave travel time from the source to the interface, and it is obviously the conversion EM signal originated from the electric double-layer in the interface. With a minimum 14cm offset, the event A arrived at the same time only with greatly reduced amplitude, and the event B had not able to be detected for its weak amplitude. Another event B' with an about 50 microsecond identical arriving time could, however, be recognized, and it should be a conversion EM wave from the interface exited by the second acoustic vibration cycle from the acoustic source wave with higher amplitude, as the arrival time just equates to that of the second cycle of the narrow band acoustic wave to travel to the interface. These measurements in the laboratory show that , the electric double-layer formed at the frozen-unfrozen interface can be polarized to generate EM waves by both an EM pulse and a vibration source, which imply that the frozen-unfrozen interface of the permafrost could be surveying by both EM, and seismoelectric methods. And the results also show that the electric double-layer needs several hours to be formed in a laboratory experiment under low tempreture.