To detect buried landmines, Planning Systems Incorporated (PSI) has developed a Ground Penetrat-ing Synthetic Aperture Radar (GPSAR) system. GPSAR can be used to generate three-dimensional images of mines. It has been shown that the SAR processing in the PSI GPSAR system can greatly improve the image quality and hence the mine (especially plastic mine) detection performance. Since the electromagnetic wave propagation velocity in the soil depends on many factors, inchlding the soil texture, bulk density, as well as water content, and it changes from one location to another, velocity uncertainty is inevitable. It is well-known that for the conventional airborne/spaceborne SAR processing, velocity uncertainty may result in phase errors that will severely blur the SAR image. However, we have observed from experimental examples that the PSI GPSAR system is very robust against the velocity uncertainty. More specifically, under mild conditions velocity uncertainty does not defoeus the image but only scales the image along the depth dimension, and hence will not affcct the mine detection performance. Theoretical analysis is provided to explain this observation. Although our discussion is based on the PSI GPSAR system, it applies to other GPR based landmine detection systems as well.