Using empirical mode decomposition to process marine magnetotelluric data

Chen, Jin; Heincke, Björn; Jegen, Marion; Moorkamp, Max

doi:10.1111/j.1365-246x.2012.05470.x

Cited by 36 publications

(37 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We can find the receiver transfer function by analysing this relation between the electric and magnetic time series, even if they are all non stationary (Chen et al, 2012). Per definition, each time step of the IMF has well-defined values for frequency, phase and amplitude.…”

Section: Theory and Methodsmentioning

confidence: 99%

Non Stationary, Broad-band Waveforms for CSEM - An Analysis with Synthetic Data

Neukirch¹,

García²

2014

Proceedings

View full text Add to dashboard Cite

SUMMARYControlled source electromagnetic (CSEM) methods are sensitive to the subsurface conductivity structure and thus had led to its use in resource exploration. Since the frequency for peak sensitivity and the exact location of an exploration target is normally unknown prior exploration, it is desirable to acquire the transfer functions for a broad range of frequencies and in a wide area. Investigations in both directions have been driven by optimising properties of the Fourier transform in order to enhance the frequency range and the source-receiver distances. Recent research on non-stationary (NS) time series analysis tools significantly enhanced processing of NS time series. This work assesses the possibility of NS source waveforms by presenting a chirp source that is highly customisable in amplitude and frequency range in order to accommodate any frequency range in combination with virtually any amplitude for each frequency (e.g. in order to counter attenuation by decreasing power from low frequencies to increase high frequency power assuming constant energy supply). A numeric example illustrates the NS waveform and that robust NS time series processing may lead reliably to the transfer functions of a 1D conductivity model. Lastly, the advantages of a freely customisable waveform design are discussed.

show abstract

Section: Theory and Methodsmentioning

confidence: 99%

Non Stationary, Broad-band Waveforms for CSEM - An Analysis with Synthetic Data

Neukirch¹,

García²

2014

Proceedings

View full text Add to dashboard Cite

show abstract

“…Decomposing signal by singular-value decomposition has been used in airborne TEM data processing (Reninger et al, 2011). Chen et al (2012) propose empiricalmode decomposition (Huang et al, 1998;Chen and Ma, 2014;Chen, 2016) based on a denoising algorithm for marine magnetotelluric data and obtain acceptable results. Denoising methods that are used in seismic data can also be applied in processing CSEM data, such as some decomposition methods (Chen, 2017(Chen, , 2018.…”

Section: Introductionmentioning

confidence: 99%

Denoising controlled-source electromagnetic data using least-squares inversion

Yang

Tiegang

et al. 2018

GEOPHYSICS

View full text Add to dashboard Cite

Strong noise is one of the toughest problems in the controlled-source electromagnetic (CSEM) method, which highly affects the quality of recorded data. The three main types of noise existing in CSEM data are periodic noise, Gaussian white noise, and nonperiodic noise, among which the nonperiodic noise is thought to be the most difficult to remove. We have developed a novel and effective method for removing such nonperiodic noise by formulating an inverse problem that is based on inverse discrete Fourier transform and several time windows in which only Gaussian white noise exists. These critical locations, which we call reconstruction locations, can be found by taking advantage of the continuous wavelet transform (CWT) and the temporal derivative of the scalogram generated by CWT. The coefficients of the nonperiodic noise are first estimated using the new least-squares method, and then they are subtracted from the coefficients of the raw data to produce denoised data. Together with the nonperiodic noise, we also remove Gaussian noise using the proposed method. We validate the methodology using real-world CSEM data.

show abstract

“…The ensemble empirical mode decomposition (EEMD) is an improved method based on EMD, which identifies the IMFs components using an ensemble of trials, each involving the signal plus a white noise of finite amplitude . EEMD largely eliminates the mode mixing (scale-mixing) problem and the end effect caused by intermittency signal noise in the EMD method (Huang and Wu 2008), although EMD has demonstrated its applicability in a wide range of geoscience studies over the last decade (Vasudevan and Cook 2000;Huang and Wu 2008;Jackson and Mound 2010;Chen et al 2012). When EEMD analysis is used to decompose nonlinear and non-stationary data the added white noise is averaged out over a sufficient number of trials.…”

Section: Introductionmentioning

confidence: 99%

Noise Reduction, Atmospheric Pressure Admittance Estimation and Long-Period Component Extraction in Time-Varying Gravity Signals Using Ensemble Empirical Mode Decomposition

Wang¹,

Chen²,

Du³

et al. 2015

Terr. Atmos. Ocean. Sci.

View full text Add to dashboard Cite

Time-varying gravity signals, with their nonlinear, non-stationary and multi-scale characteristics, record the physical responses of various geodynamic processes and consist of a blend of signals with various periods and amplitudes, corresponding to numerous phenomena. Superconducting gravimeter (SG) records are processed in this study using a multi-scale analytical method and corrected for known effects to reduce noise, to study geodynamic phenomena using their gravimetric signatures. Continuous SG (GWR-C032) gravity and barometric data are decomposed into a series of intrinsic mode functions (IMFs) using the ensemble empirical mode decomposition (EEMD) method, which is proposed to alleviate some unresolved issues (the mode mixing problem and the end effect) of the empirical mode decomposition (EMD). Further analysis of the variously scaled signals is based on a dyadic filter bank of the IMFs. The results indicate that removing the high-frequency IMFs can reduce the natural and man-made noise in the data, which are caused by electronic device noise, Earth background noise and the residual effects of pre-processing. The atmospheric admittances based on frequency changes are estimated from the gravity and the atmospheric pressure IMFs in various frequency bands. These time-and frequency-dependent admittance values can be used effectively to improve the atmospheric correction. Using the EEMD method as a filter, the long-period IMFs are extracted from the SG time-varying gravity signals spanning 7 years. The resulting gravity residuals are well correlated with the gravity effect caused by the Earth's polar motion after correcting for atmospheric effects.

show abstract

Using empirical mode decomposition to process marine magnetotelluric data

Cited by 36 publications

References 44 publications

Non Stationary, Broad-band Waveforms for CSEM - An Analysis with Synthetic Data

Non Stationary, Broad-band Waveforms for CSEM - An Analysis with Synthetic Data

Denoising controlled-source electromagnetic data using least-squares inversion

Noise Reduction, Atmospheric Pressure Admittance Estimation and Long-Period Component Extraction in Time-Varying Gravity Signals Using Ensemble Empirical Mode Decomposition

Contact Info

Product

Resources

About