History of Surface Wave Methods

Surface-wave surveys commenced for the field of pavement engineering, to evaluate the pavement stiffness. The steady state vibration technique utilizing Rayleigh waves and employing single-receiver approach was used to find the thickness of the concrete pavement (Van der Poel 1951; Jones 1955). The pioneering steady state vibration technique was subsequently improved to make it more computationally efficient (Gordon 1997), and was, thereafter, mostly referred to as the Continuous Surface Wave (CSW) method (Matthews et al. 1996). The next advancement conforms to the two-receiver approach, commonly known as the Spectral Analysis of Surface Waves (SASW). Heisey et al. (1982) first introduced the SASW method for determining the moduli of materials used in pavements. Since then, various researchers applied the fundamental concepts of SASW for the evaluation of pavement thickness as well as for the geotechnical site characterization (Nazarian et al. 1987; Stokoe II et al. 1994). Although largely used, SASW method has several technical limitations. The primary shortcoming of the technique is its inability to identify the higher order modes-of vibration. Although SASW can provide shallow depth information, the resolution of the obtained Vs profile decreases beyond the depth of 10-15 m. It is insensitive to the sharp contrasts of density and velocity of the stratum. Difficulties also exist in evaluating and distinguishing the signal from noise with the usage of a single pair of receivers. The necessity of recording repeated shots for multiple field deployments at a given site increases the time and labour requirements to conduct a reliable SASW survey (Park et al. 1998).

Multichannel analysis of surface waves (MASW) overcomes many of the shortcomings of the SASW method (Park et al. 1999). In comparison to the SASW approach which is mostly based on the analysis of the fundamental mode (since the higher modes are nearly indistinguishable), MASW approach provides a dispersion image in which different modes can be easily identified.

Unlike SASW, which is based on an active source technique, MASW has a scope of using passive sources as well, which generates low frequency waves, and thus considerably enhances the depth of investigations. Moreover, in comparison to SASW that primarily uses a single pair of geophones, deployment of multiple receivers during the MASW survey saves the labour and the time associated with the field investigation.

MASW method became popular in early 2000s among the geotechnical engineers. The term

“MASW” was introduced by Park et al. (1999). The earliest usage of multichannel approach for surface waves was reported by Gabriels et al. (1987) for deducing the shear wave profile of tidal flats by analyzing the recorded surface waves using 24-channel acquisition system.

Subsequently, Park et al. (1999) highlighted the effectiveness of the technique by detailing the advantages of multichannel acquisition and processing concepts most appropriate for the geotechnical engineering applications. The convenience in field operation and robustness in the data processing provided by diverse multichannel seismic data processing techniques were also emphasized. Soon after the concept of MASW was developed and its effectiveness was realised, the application of MASW was initiated for pavement investigation (Ryden et al. 2004; Forbriger 2003a, 2003b; Ryden and Lowe 2004) and site characterizations (Kanli et al. 2006; Xia et al.

2000). For data processing, various techniques have been used over the time namely the f-k method (Capon 1969), the π-ω transformations (McMechan and Yedlin 1981), and the phase-

shift method (Park et al. 1998). Xia et al. (1999) presented an algorithm to carry out inversion analysis to obtain the 1D shear wave velocity profile. The 1D profile obtained from a single linear array was considered most representative of the subsurface directly beneath the middle of the geophone spread. Multiple 1D plots of S-wave velocity with depth can be generated by continuously varying the position of the array. A 2D vertical cross-section of S-wave velocity is constructed by combining and suitably interpolating all the obtained 1D S-wave velocity profiles (Miller et al. 1999; Xia et al. 2000). Utilizing the 2D Vs mapping, several MASW case studies have been undertaken worldwide (Xia et al. 2000, 2004; Park et al. 2005; Kaufmann et al. 2005).

As the surface wave techniques gradually gained popularity among the researchers and geophysicists, a growing demand was experienced to increase the depth of investigation. Until 2000, MASW survey was carried out using an active seismic source, such as hammers, weight drops, electromechanical shakers, propelled energy generators (PEG) and bulldozers, wherein the source and rate of loading can be controlled by the user. Literature reveals that, in general, the maximum depth of investigation (governed by the wavelengths) was obtained as 20-30 m in the case of active MASW survey, since the waves generated from impacts comprise of comparatively higher frequencies, greater than 30 Hz (Park et al. 2002a, 2007; Gosar et al.

2008). In order to acquire information of deeper strata, the depth of investigation had to be increased beyond 30 m, which was attained by using low frequency surface waves (5-7 Hz). To generate waves of longer wavelengths (i.e. low frequencies) using active source, heavy weights capable of producing high impact energies have to be used so that high-energy accumulation is achieved at the lower frequency range. Application of very heavy weight systems, such as bulldozer or PEG, restricts the portability of the MASW approach, and thus renders it infeasible

and impractical. On the other hand, passive surface waves generated from natural (e.g., tidal motion, microtremors) or cultural (e.g., traffic) sources usually comprises of low-frequency waves (1–30 Hz), with wavelengths ranging beyond 30 m to several kilometres (Okada and Sakajiri 1983; Okada et al. 1990; Okada 2003; Park et al. 2007). Passive MASW utilizes such waves for subsoil survey, thus attaining larger penetration depths and allowing for deeper subsurface investigation. Application based on a similar concept originated during 1950s in Japan and was called the Microtremor Survey Method (MSM) utilizing limited number of receivers (usually 10) aided by a processing technique known as the Spatial Auto Correlation, or SPAC (Aki 1957). Later, Linear Refraction Microtremor survey (ReMi method) was introduced by Louie (2001) using a linear array of geophones, which at a stage later gave rise to the Passive Roadside MASW methods (Park et al. 2007).

Based on the existing literature, it can be observed that there has been a systematic development of surface wave methods starting with the Steady State vibration approach utilizing a single receiver, Spectral Analysis of Surface Waves (SASW) utilizing a two-receiver system, and the most modern Multichannel Analysis of Surface Waves (MASW) utilizing a multi-receiver array system. With the advancement in technology, the techniques of analyses have gradually evolved from the principles of surface wave refraction and, at present, include the dispersion of waves for the analyses and identification of subsurface stratification in terms of the shear wave velocity profile.