Methods Study Of Helicopter-borne Towed Bird Time Domain Electromagnetic Data Processing

Time domain airborne electromagnetics (AEM) with excellent horizontal accuracy, ease of interpretation, effective cost, and good depth of effective exploration etc. has significant development. Time domain airborne electromagnetic data suffer large noise, especially late data. If this data is not preprocessed, analysis and interpretation will have a bad effect. Based on National 863 program sub-session, including methods study of data processing and basis map edit, and software research and development, methods of helicopter-borne time domain electromagnetic (HTEM) data processing were researched.In the paper, it is investigated and summarized that the kinds, sources, characters and methods for reduce of noise which has effects on airborne electromagnetic survey. Effects on time domain AEM survey by different noise were analyzed for finding the way to eliminate it. All those supply good suggestion to make rules about quality control (QC). There are three rules which are flight height, flight path and noise level. Checking deviation between measured data of height, path and fiducial data, an efficient algorithm of instantaneous signal-to-noise-ratio (SNR) estimation is used in evaluating noise levels which is effective and efficient. Reflights were required if any of the specifications were not met.Noise was a significant issue in time domain airborne electromagnetic survey. Median filter, wavelet filter and Vondrak smoothing method are optional. Median filter has the characters of nonlinear filters, especially in restraining impulse noise. Multi resolution and adaptability in wavelet filter, and db3 wavelet is chosen. The results from median and wavelet filtering are effective by analysis of deviation. And we also found the statistical distribution of noise from measured data is consistent with white noise's statistical distribution. Vondrak smoothing method was usually used in astronomic observation data processing. And this is first time applied in AEM data smoothing. An appropriate smoothing factor is the key to complete and achieve the whole de-noising process. Thus, standard deviation of observation and calculating frequency response are as effective methods to choose the factor. And the standard deviation described the smoothing result also verified that this chosen factor was reasonable and compatible. Therefore, these three methods can be applied in HTEM data de-noising.For the useful information as many as possible from HTEM data, Gabor method and Wigner-Ville distribution are also used in HTEM data processing. Three geo-electric sections were differentiated very well. And also time and frequency spectrum analysis present most noise in the range of high frequency, and energy of measured data focus on low frequency phase. Along with time axis the energy fast attenuate which show second field along depth weakening. For the time channel, anomaly is identified at the corresponding position. Above the method all provide basis in theory with the following geophysical simulation and analysis, and further geological interpreting.On all accounts, the paper based on a workflow of HTEM data preprocessing which supply not only an appropriate process for HTEM data processing, but also a theoretical basis for following geophysical analysis and geological interpreting.