Interpretation of C- and X-band radar images over an agricultural area, the Flevoland test site in the Agriscatt-87 campaign

C-band SAR and X-band SLAR radar backscattering data collected over an agricultural area in The Netherlands, during the Agriscatt 1987 campaign, were compared and analysed in relation to type and structural properties of soil cover. The C- and X-band data were found to be similar in major trends. The variation in field-average radar backscattering values, however, was smaller in the C-band than in the X-band. Some fields showed a strong look direction dependence while others showed a strong polarization dependence of the backscattering. From field observation and from a theoretical analysis, incorporating the different structures and spatial orientations of the canopies elements, it was shown that non-uniform azimuthal distribution was the main cause for this phenomenon. It was concluded that these different types of backscattering behaviour can be two aspects of a single physical mechanism. Signature interpretation problems may arise when these effects—non-uniform azimuthal distributions being caused by prevailing winds, lodging, harvesting or heliotropic behaviour-are not properly accounted for. This may be particularly important when monitoring systems like the ERS-I SAR are used, operating in two different look directions, i.e., in the ascending and descending nodes. For instance, in the data set investigated, the effect of look direction and state of polarization on the separation between classes of land cover type and condition was found to be large.