Seasat and aircraft radar imagery of five areas of sand dunes in the southwestern United States and northwestern Mexico have been studied and compared to Landsat imagery and air photos. Radar imaging system parameters included two wavelengths (23.5 cm and 3.0 cm) and incidence angles ranging from 0 ¿ through 70 ¿. Horizontal polarization was used on both transmit and receive for all radar imagery. Resolution ranged from 12 m for the best aircraft imagery to about 40 m for Seasat. Sand dunes studied behave as smooth surfaces for both 23.5 cm and 3.0 cm wavlength radars and the reflection of the radar energy from the dunes is specular. Measured surface irregularities of sand dunes are inadequate to cause either Bragg or incoherent scattering of the incident radar beam in the 3.0 to 23.5 cm wavelenth region. This results in dunes that appear dark when imaged at incidence angles greater than the angle of repose because none of the dune faces are oriented properly (normal to the incident radar beam) to give a specular retun to the antenna. A scattering model based on specular return frm radomly oriented flat dune faces was used and experimentally verified using aircraft data. Imaging radar resolution resolution must be adequate to define the changing distribution of dune faces that characterize the dune morphology. Linear dune features with interdune spacing of less than 3--4 resolution elements are not recognizable. Distributed dune types such as barchans or stars must be more than about 10 resolution elements in each direction to be defined on the imagery. Illumination direction of the radar beam is important; directional dune features must be oriented within about 60 ¿ of perpendicular to the radar illumination direction in order to be imaged. If such features have considerable topographic relief, the direction of radar illumination is less important. Because directional trends in unexplored regions cannot be predicted, availability of radar imagery from two directions greatly facilitates interpretation of dune morphology and hence conclusions about causative wind regimes. |