Nondestructive Testing Resolution:

The practical limits on resolution of images from seismic or radar (electromagnetic) data depend on the spectrum of the source pulse, the incompleteness of the recording geometry, and the fidelity with which one can model the reference medium. These factors together prevent one from recovering a point image from data produced by a point diffractor in a given body. A simple imaging formula quantifies the effect of the source pulse spectrum, showing that spatial wavelet extent in an imaged volume is related to the product of the local velocity and the period of the propagating pulse. A general scheme for estimating the effect on resolution of the recording geometry is applied to various recording geometries in modeled concrete slabs for seismic and electromagnetic waves. The penetration of electromagnetic waves, generated from Radar instruments, into media like concrete, have similar kinematic properties to the seismic (ultrasonic) ones, differ in their dynamic and response behavior. Using finite difference numerical modeling, as well as physical observations, we realized that features in media, such as voids, appear differently on GPR sections versus seismic ones. Part of the difference is attributed to the wave velocity difference for the two waveforms for air and water filled voids, resulting in difference in appearance and resolution. Despite the much higher frequencies associated with Radar electromagnetic waves, the higher frequencies are well compensated by the high velocity of such waves resulting in wavelengths similar to those experienced in seismic experiments. Nevertheless, due to the single mode nature of electromagnetic waves, we obtain better scattering behavior (cleaner) using radar data than what can be obtained using seismic ones. The scattering response of seismic waves includes the conversion of part of the energy to other modes (i.e. shear modes). These modes can potentially provide additional information, but requires careful analysis.