Overlay Analysis

Abstract

One of the problems in presenting information from GPR datasets are that the targets of interest are often at different depths in the ground, or variable top soil topography can complicate the depth horizons to targets which are at the same absolute elevation. In addition, even in the case where structures are actually buried level plane in below a flat ground, variable velocity soils can make these structures appear at different reflection times on GPR radargrams. An example of this was shown in the simulation in Fig. 2.17 where strong velocity contrasts caused velocity pull-ups on horizontal features at depth. Standard imaging practices to make archaeological site plans from GPR, the simple flat time slices are the most commonly used imagery. One can imagine, that displaying a set of horizontal slices across a site, that if the targets of interest are recorded at different depths such as shown in Fig. 5.1, that any individual – thin horizontal time slices made across these kinds of radargrams would not completely image all the targets on any single image. Information on structures might be partially contained at different depths in the time slice dataset, making for connection of structures with different depth horizons difficult to visualize. In the unique case where the features have high contrast with surrounding reflections, isosurface rendering might be able to show a complete 3D image; and when projected from the top could give a more complete picture of all the important reflections at a site. These kinds of datasets however, are also far and few between in the GPR world, as target reflections are not always highly contrasted with surrounding reflections. Isosurface rendering of all the important reflections surfaces with depth can not be used to effectively illuminate the complete structure when the targets do not have uniformly high contrast with the matrix of recorded background reflections.