Six-component seismic land data acquired with geophones and rotation sensors: Wave-mode selectivity by application of multicomponent polarization filtering

Abstract

Multicomponent acquisition systems today record incomplete data because they do not measure rotations. Geophones or accelerometers provide linear motion and hydrophones provide pressure, but no current commercial acquisition system includes sensors that measure rotations. Without rotations, the data provide incomplete recording of the wavefield because in three dimensions, there are six degrees of freedom -three linear displacements and three rotations. In two small 2D seismic surveys recorded with six components, three-component geophones were deployed and three-component rotation sensors that measured pitch, roll, and yaw were deployed. Pitch was measured independently with closely spaced geophones. Measuring rotations with closely spaced geophones is not practical in production, but it can be used in research and development to validate rotationsensor data. A comparison of pitch measured by two independent methods finds that they fit after instrument designature. Data provided by rotation sensors have additional value because they can be used in analysis of singular value decomposition (SVD) to identify and separate ground roll and body waves.

A five component land seismic sensor for measuring lateral gradients of the wavefield

“…; Barak et al . ). The 5C seismic sensor data are dominated by the horizontal gradient of the vertical ground roll component.…”

“…In addition, the horizontal component data often contain strong shear-wave events. More recently it was established that gradient data can be used as a noise model instead (Edme et al 2014;Barak et al 2014). The 5C seismic sensor data are dominated by the horizontal gradient of the vertical ground roll component.…”

“…In recent years, single‐station ground roll noise attenuation methods were proposed by adaptive subtraction of gradient data (Edme, Muyzert and Kragh ), by singular value decomposition (Barak et al . ) and by polarization filtering (Sollberger et al . ).…”

“…Multichannel sampling methods, where both the wavefield and its spatial gradient are measured by a single device, are one such approach. In recent years, single-station ground roll noise attenuation methods were proposed by adaptive subtraction of gradient data (Edme, Muyzert and Kragh 2014), by singular value decomposition (Barak et al 2014) and by polarization filtering (Sollberger et al 2017). A multichannel, multi-station method for noise attenuation by wavefield reconstruction was proposed by Allouche, Ras andÖzbek (2015).…”

Recording active-seismic ground rotations using induction-coil magnetometers

“…Edme et al (2014) treat rotational data as a noise model for ground roll and use adaptive subtraction to remove ground roll from the vertical component of geophone land data. Barak et al (2014) show that rotation data are extra information, are independent of geophone data, and can be used in conjunction with geophone data to identify and separate wave modes on land using singular-value decomposition polarization analysis. Li and van der Baan (2015) show how to enhance microseismic event localization derived from picked P-and S-wave arrivals in borehole data using the translational and rotational wavefields.…”