Call us: 818.886.4500
Toll Free  : 877.565.3590

“Specialists in Industrial, Military and Urban Settings”

Methods: Refraction Microtremor (ReMi)

    The Refraction Microtremor (ReMi) method is commonly used for site investigations where the following information is needed:

    The ReMi method is a seismic surface wave method (similar to SASW and MASW methods) that uses ambient noise and surface waves to generate a detailed vertical shear wave velocity (Sv) profile to depths of up to 100 meters. Prior to 1999, seismic shear wave profiles were obtained using shear wave refraction, seismic cone penetrometer, or downhole/crosshole techniques. In the case of seismic refraction techniques, the hidden layer or low-velocity layer case could arise, resulting in a false interpretation of shear velocities and associated depths. In the case of cone penetrometer testing, the rig can encounter refusal in coarse gravels and cobbles, glacial moraines, or competent bedrock, resulting in an incomplete vertical profile. Downhole and crosshole techniques tend to be more reliable but can be costly as a borehole or boreholes must be drilled to specifications prior to seismic data acquisition. The ReMi technique provides a non-invasive way of obtaining a vertical profile of the shear wave and, unlike borehole methods, this technique provides a shear wave sample of a greater volume of material, thereby allowing a more representative shear wave velocity column that is averaged over the length of the seismic array

    Unlike shear wave refraction, the ReMi method is capable of detecting thin layers and velocity inversions, and as such is highly reliable and commonly used to measure Vs30 for earthquake design ground motion determinations. ReMi can also be used at any site to measure the lateral variation of shear wave velocity in near surface materials. Because it thrives in noisy environments the ReMi method is ideal for shear wave profiling in urban environments where seismic refraction is precluded because of large amounts of ambient noise. Additionally, the ReMi method is also very useful for stratigraphic delineation in complex geologic environments where even shear wave refraction fails. Spectrum utilizes the Seistronix RAS-24 twenty-four channel signal enhancement seismograph along with geophones and associated cabling to collect ReMi data.

    The refraction microtremor (ReMi) technique, along with SASW and MASW methods, has significant advantages over traditional downhole and crosshole methods for shear wave velocity profiling. Like SASW and MASW methods, the ReMi technique takes advantage of the dispersive nature of surface Rayleigh waves, which causes different frequencies in the wavetrain to travel at different velocities. In a layered earth this causes higher frequencies to travel at shallow depths and typically lower velocities, and lower frequencies to travel at greater depths and typically higher velocities. Because the Rayleigh wave is coupled to the P-Sv (vertical component of the shear wave) system, a given shear wave velocity structure gives rise to a unique dispersion curve (curve that pairs a frequency with Rayleigh wave phase velocity).

    Unlike SASW and MASW methods, the hallmarks of the ReMi technique are 1) that it uses a linear array that does not require CMP-style shooting and 2) that it relies primarily on ambient noise to generate the Rayleigh waves. Louie (2001) has shown that a two-dimensional array is not needed to measure the dispersion curve of Rayleigh waves because the minimum phase velocity is correlated with waves traveling parallel to the array.


    In order to obtain the ReMi dispersion curve of the surface Rayleigh waves, a linear array of multiple (generally 24) equally-spaced geophones is established and connected to a seismograph. The length of the array depends on the depth of investigation; in order to obtain information to 30 meters (100 feet) a length of 300 feet or more is typically required. Once the linear array is established, Spectrum records both ambient and active-source noise records (shorter duration records for shallow information and longer duration records for deeper information). These noise records are then processed using the SeisOpt® ReMi™ software. First, a slowness-frequency wavefield transformation is applied to each of the noise records, yielding a slowness-frequency spectral image where the spectrum is normalized as the ratio of the power spectrum at a particular frequency and slowness (inverse velocity) over the average value for all slowness values at that frequency. Once this image is generated, the fundamental mode dispersion curve of the Rayleigh wave is then “picked” by selecting the minimum phase velocity of the envelope of Rayleigh wave energy, which has been shown to correlate with waves traveling parallel to the array. An iterative forward modeling process is then used to generate an optimal shear wave velocity-depth profile that would give rise to the selected dispersion curve. In order to minimize ambiguities Spectrum prefers to constrain the shear wave model with known geologic and refraction data for the area. The end product of the ReMi technique is a one-dimensional column of shear wave velocity variation for each seismic line established at a site.