Recent modeling techniques for fluid flow in a reservoir or an aquifer require characterizing the statistics of the spatial variability of physical properties such as porosity or reflectivity. The major interest of such modeling is in horizontal variations. We acquired a nontraditional seismic survey and analyzed it for the statistical parameters of correlation length and fractal dimension. Variograms and power spectra can extract these parameters from migrated sections. Statistical parameters are valuable for newly established hydrogeologic modeling techniques, with their definition greatly increasing the understanding of the aquifer under study. Applying similar techniques may also prove valuable for reservoir exploitation in the petroleum industry. We acquired our survey on top of a bench of an open-pit diatomite quarry and extracted the statistical parameters for an exposed vertical face. The imaged zone of interest is therefore physically accessible to future permeability studies that could validate our interpreted parameters. We can interpret horizontal, but not vertical, correlation lengths and fractal dimensions from either variograms or power spectra of our migrated seismic images. The horizontal correlation length and fractal dimension extracted from the seismic variograms match those extracted from photographs taken of the quarry face.