AEM noise reduction with remote referencing

Abstract

A number of sources contribute to noise in airborne electromagnetic (AEM) systems. The main source of noise at low frequencies (below about 1 kHz) is motion of the bird receiver coils in the earth's magnetic field. Generally, the main noise remaining after reduction of motion noise is caused by sferics spikes produced by lightning activity. Other external sources that produce noise in AEM systems include power lines and VLF transmitters. The characteristics of sferics noise recorded with a ground-based system with two receiver stations studied at a number of localities throughout Australia showed that a high degree of spatial correlation exists between the three vector components of the noise and also between noise measured simultaneously at two stations with a separation at least as large as 50 km. Sferics in the high-frequency range (~2 to 100 kHz) were found to correlate with data recorded at a remote station. Using simple subtraction of noise time series measured at the remote station from the corresponding local time series, a noise reduction factor of about 4 was obtained. Other strategies for noise reduction include prediction filtering using either local or remote referencing. By reducing noise in AEM data, target detection depth can be increased and stacking time can be decreased, improving spatial resolution. These two strategies have important implications for AEM surveys, since stacking time is always limited because of the continual travel of the receiver along a flight line. Noise produced by other external sources, in particular, noise produced by VLF transmitters and 50 Hz powerline noise can also be reduced through referencing. The continual monitoring of noise with remote referencing permits changes in amplitude or frequency of noise from these sources to be accommodated in the noise reduction process.