Spatial Wind Variability from Different SAR Imagery Received at CSTARS

CSTARS (Center for Southeastern Tropical Advanced Remote Sensing) is the University of Miami's facility conducting research with remotely sensed data received from earth-orbiting satellite systems. CSTARS is a new state-of-the-art real-time satellite tracking facility located at the former U.S. Naval Observatory/Alternate Time Tracking Station site at Richmond, in southern Miami-Dade County, Florida. CSTARS was designed and developed as a highly automated, near-real time, multi-satellite reception and processing facility that provides data over the Equatorial Atlantic region, northern South America, Central America, the Caribbean Basin, Gulf of Mexico and the majority of North America. The facility is ideally situated to provide rapid data access for many time sensitive applications such as environmental monitoring, storm prediction, volcanic eruptions, pollution, natural hazards, and change detection in these regions. CSTARS is capable of receiving and processing radar data from three different space-borne SAR systems: RadarSat-1, ERS-2 and ENVISAT ASAR. All three SARs operate in C-band, but have different imaging modes and spatial resolution as well as polarization options.

For the last several years, there has been increasing interest in obtaining high-resolution (sub-kilometer) marine wind speed measurements from SAR imagery. The need for higher resolution wind fields in coastal areas, in frontal systems and tropical storms cannot be obtained with wind fields measured onboard conventional space-based satellites such as scatterometers (e.g., QuikSCAT) or passive microwave radiometers (e.g., SSM/I and the recently launched WindSat). These sensors are best suited for global measurements, because with 25-km spatial resolution many important high-resolution spatial phenomena are missed.

The normalized radar cross-section (NRCS) at the incidence angles in the range of 20° to 70° is dominated by Bragg resonant scattering from centimetric waves traveling in the direction of the incident microwave energy. Consequently the radar reflectivity is strongly dependent on direction of propagation of these short waves with respect to that of the radar energy. These short waves are known to propagate centered on the wind direction. The SAR images can be used to determine a spatial average of the instantaneous backscatter, i.e. the roughness of the ocean surface. Thus the accurate retrieval of wind speed and direction are linked. First, the estimation of wind direction from SAR images is determined from surface features like wind streaks. The calibrated NRCS is then taken together with this direction to determine wind speed by applying a CMOD-type algorithm originally developed for the ERS scatterometer.

The spatial resolution of the derived wind fields from the different SAR systems is about 500 m. The results will be compared with in-situ measurements such as buoys which provide a temporal average at a location from anemometers or sonic measurements at a specific height.