Because the Rayleigh waves generally have the largest amplitude of all waves generated by a vertical surface impact and because the near-surface shear-wave velocity primarily determines the Rayleigh wave velocity, the Rayleigh waves may be used to image shallow shear-wave structures. The Rayleigh wave group velocity can be measured from records of surface waves that have traversed a study area, typically, with a surface source on one side and an array of geophones along the opposite side. The multiple-filter technique is used for group velocity determination. First, each trace is narrow-band filtered and the instantaneous amplitude of its complex trace is computed. Second, the time of the peak in the instantaneous amplitude of the complex trace consistent with the mean group travel time is identified. Third, all the observed group travel times are corrected for group delays introduced by instrument response, geophone ground coupling and the source time function. Finally, the effect of noise can be minimized by constraining the observed group arrival times to be consistent among arrivals that follow similar paths. By obtaining data from different source and receiver locations, the resulting group travel times can be used in a tomographic inversion to image the distribution of group velocity within the study area. The tomographic inversion is repeated for many frequencies giving a dispersion curve for each point in the study area. Then, vertical shearwave velocity structure at any point can be interpreted from its dispersion curve. In this study we obtained data covering a square area that was 30meters on a side. Usable group velocities were obtained for frequencies from 16to50Hz using a sledgehammer source. The resulting tomographic image and velocity anomaly was sufficient to delineate a suspected 4-meter deep burial trench.