Recent interest in mapping the topography of basal surfaces of polar ice sheets motivated the development and deployment of a low-frequency imaging radar. The data available from this system can also be used to provide surface roughness properties at the upper surface of the ice sheet. The reflectivity of the basal surface is also of interest for detecting water bodies that could be present beneath kilometers of ice.
Analyzing measured data from this system requires a proper calibration and a model that can predict the shape and the power level of the measured radar waveforms for given surface roughness parameters. This dissertation presents the development of a model to describe the averaged pulse return waveforms from the upper and basal surfaces of the ice sheet.
The model takes the form of a convolution product of an impulse response of the flat surface, whose cross section per unit area is the same as for the actual rough surface, and other functions describing the surface roughness and the system impulse response. For low-frequency applications, this model (commonly referred to as the “Brown model” in high-frequency altimetry applications) after required modifications is justified through comparisons with a complete Physical Optics electromagnetic model. The convolution model, which is simpler to formulate, is then extended to two-dimensionally rough surfaces to study pulse returns from the ice sheet’s upper surface and the returns from the basal surface of the ice sheet.
The study of scattering from the basal surface also requires a good estimate of the effective normalized radar cross section (NRCS) that also accounts for the layered structure of the surface. The formulation of the surface NRCS is also presented. With the model and NRCS formula presented, the visibility of the basal returns for given parameters – including radar platform heights, surface roughness, and attenuation caused by the ice volume – are examined. The results can be useful for further development of the radar system and for measurement planning. In addition, the pulse return model developed can be used to estimate the surface roughness of the ice sheet. The results consistent with those obtained from separated airborne laser altimeter measurement are obtained.