Reprocessing strategy to obtain quantitative early time data from historic VTEM surveys

Abstract

While data in AEM systems such as VTEM is continuously sampled at 5 or 10 microsecond intervals, conventional processing steps in the past have not provided useful data close to transmitter current turnoff. In historic VTEM data for example, delivered data usually starts many tens or even a hundred microseconds or so after turnoff. Experiments in deconvolution of early time data at high altitude identified that the symptoms of problem early-time data were that the underlying cause had a linear phase response, resulting in a consistent exponential response in time-domain. This unwanted or spurious response is experimentally determined to be of variable amplitude, and of either sign, usually reducing the data at early delays, but occasionally enhancing the combined response. A change in processing strategy to specifically identify and subtract this additive spurious response from a valid earth response has led to the extraction of quantitative AEM data at early delays in the 10 to 20 ?s range. The process can be applied to historic data. System bandwidth limitations do of course provide a limit to the accurate sampling of the earth response at extremely early delays. More recently, hardware changes have increased the system bandwidth and reduced the need for software corrections to acquired data.