Recent advances in ground-probing radar instrumentation have allowed the collection of large volumes of digital data. Such data sets are amenable to modern data-processing techniques both to increase geological resolution and to enhance data presentation. The close similarity between ground-radar data and seismic data suggests that processing techniques that have been used in the seismic industry could be applied to radar data. As an example, a ground probing radar profile is deconvolved using the common prediction-error filter, which assumes a white power spectrum for the reflections, and a filter that assumes a spectrum proportional to spatial frequency. With the prediction-error filter we find three of four buried pipes which are not visible in the undeconvolved section; all four are found with the second filter. Key words: ground-penetrating radar, deconvolution, scaling geology, frozen-core dams, permafrost, containment dams, mill waste, Contwoyto Lake.