How to Acquire Data for Vs30m Evaluation
Most typical MASW data acquisition for seismic site characterization (SSC) requires a 24-channel seismic acquisition system consisting of 24-channel seismograph, seismic
cable with 24 hookups (takeouts), 24 low-frequency geophones (e.g., 4.5-Hz ones), and a heavy sledge hammer (e.g., 10 lb or heavier). The figure below illustrates a typical
configuration of all necessary components.
*Vertical stacking is applied whenever it is necessary to increase signal-to-noise ratio (S/N) to compensate for a noisy
environment during data acquisition. Instead of saving a set of recorded data from one hammer impact, multiple recordings
from successive multiple impacts can be summed (i.e., stacked on top of the previous recording data) in the control unit's
memory. This will enhance amplitudes of consistent signal surface waves through constructive overlapping, whereas those
inconsistent noise surface waves from ambient activities such as traffic will lose relative amplitudes via destructive
overlapping. Vertical stacking of "1" in seismograph's acquisition software means there will be only one impact for each
recording (therefore, "no" stacking). If it is set to 3, for example, then data from each impact will remain in the memory until the
third set of data is collected and then the stacked data will be saved onto the hard drive. In this case, each impact will have to
be separated by a time period longer than, at least, recording time (T).
*Receiver spacing can be approximately in this range of 1-2 m.
**Source offsets (X1's) are denoted by number of receiver spacings (dx), and can be applied on both sides of a given receiver
spread (i.e., forward and reverse shots, see figure below).
Proper setting of acquisition parameters is critical for a successful survey. They consist of recording and geometry parameters. The former relates to setting those to operate
seismograph properly, whereas the latter deals with proper spatial configuration of seismic source and receivers. Sampling interval (dt) and recording time (T) are the most
important recording parameters, and receiver spacing (dx) and source offset (X1) are the most critical geometry parameters. Tables 1 and 2 below summarize the most
optimum values of these two types of parameters when investigating the top 30 m to produce Vs30m values.
The geometry parameters (dx and X1) influence the maximum investigation depth (Zmax). In general, a longer receiver spread associated with a longer receiver spacing
ensures a deeper Zmax. Also, a longer source offset (X1) for a given receiver spread is necessary to ensure the high quality of surface waves for relatively longer wavelengths
by minimizing as much as possible some harmful effects, such as near-field effects. Longer wavelengths are necessary for the analysis of deeper depths. On the other hand,
a shorter X1 is also needed to ensure the same quality for relatively short wavelengths needed for the analysis at shallower depths. Therefore, it is usually a combination of
several different X1's that can achieve the highest quality ever possible for a broad range of wavelengths to cover as much depth range as possible for a given setting. In fact,
the same combination of different X1's can be applied on the front (forward) and back (reverse) sides of the receiver spread so that the possible change in subsurface property
in lateral direction (e.g., change in bedrock depth) can be properly accounted for during processing. This is illustrated in the figure below. Actual field records obtained at
different X1's are also displayed to illustrate how surface wave arrival patterns change as X1 and orientation of source change. Acquiring data from both ends of the receiver
spread can also provide redundancy in measurement that will eventually contribute to an increased signal-to-noise ratio (S/N). Table 2 below shows typical values of X1 and
dx used to perform a SSC survey for the Vs30m evaluation. It is also recommended collecting (saving) multiple files (e.g., 3 files) at a given X1 for quality control purposes and
also to further increase S/N if that is deemed necessary during data processing. File names, and all these recording and geometry parameters must be logged in a separate
field note, a sample of which is presented here.
On data analysis, all field records are processed to generate corresponding dispersion images, all of which are then stacked to generate one stacked dispersion image of the
highest quality as illustrated in the figure at the bottom. This stacked image will have the highest signal-to-noise (SN) ratio in comparison to all others because of the
incoherent noise effects suppressed, while coherent surface-wave dispersion effects enhanced, as much as possible. At the same time, the adverse near- and far-field
effects (if exist) will be attenuated through the same averaging effect. Finally, the most reliable fundamental-mode (M0) dispersion curve is extracted from this stacked image,
and the most reliable 1-D shear-wave velocity (Vs) profile is obtained from the inversion of this curve as illustrated in the figure at the bottom.