Semiquantitative evaluation of massive rock quality using ground penetrating radar

This work studies a methodology starting from georadar data that allows a semiquantitative evaluation of massive rock quality. The method is based on the concept that in good quality rock, most of the energy is transmitted, while in low quality rock, the energy is backscattered from fractures, strata joints, cavities, etc. When the energy loss due to spherical divergence and attenuation can be recovered by applying a constant spherical/exponential gain, the resulting energy function observed in the georadar section depends only on the backscattered energy. In such cases, it can be assumed that the amount of energy is an index of rock quality. Radar section interpretation is usually based on the reconstruction of reflected high-energy organized events. Thus, no consideration is given to backscattered not-organized energy produced by microfractures that greatly influences the geotechnical characteristics of the rock mass. In order to take into consideration all the backscattered energy, we propose a method based on the calculation of the average energy relative to a portion of predefined rock. The method allows a synthetic representation of the energy distributed throughout the section. The energy is computed as the sum of the square of amplitude of samplings contained inside cells of appropriate dimensions. The resultant section gives a synthetic and immediate mapping of rock quality. The consistency of the method has been tested by comparing georadar data acquired in travertine and limestone quarries, with seismic tomography and images of actual geological sections. The comparison highlights how effectively the energy calculated inside the cells give synthetic representation of the quality of rocks; this can result in maps where the high-energy values correspond to rock of poor quality and the low energy values correspond to a good quality region. The results obtained in this way can, in this case, be partly superimposed onto those of seismic tomography. (C) 2002 Elsevier Science B.V All rights reserved.