An Accuracy Assessment of Shoreline Data for Castle Bay Compiled from Synthetic Aperture Radar

CDR Grady H. Tuell
NOAA Remote Sensing research Engineer

James R. Lucas
Chief Photogrammetrist

Douglas B. Graham
Cartographer

National Oceanic and Atmospheric Administration (NOAA)
National Ocean Service (NOS)
National Geodetic Survey (NGS)
 

It is widely recognized that Synthetic Aperture Radar (SAR) technology holds great promise for shoreline mapping applications. Because SAR offers flexibility over weather and time of day restrictions, it may be especially valuable when mapping remote areas such as southwest Alaska. We have mapped a five nautical mile stretch of the coast of Castle Bay Alaska using both 1-meter airborne SAR and RADARSAT fine mode imagery. We have compared the resulting shoreline manuscripts with shoreline data produced using our conventional photogrammetric process. Although the different data sets are separated temporally by several months, and tidal differences exist, the resulting comparisons yield an opportunity to quantify the performance of SAR for this application. As part of our work, we have developed a semi-automated approach to remove positional biases from SAR imagery, and we use the residual vector from this approach as a metric for evaluating accuracy. We present results to date in the analysis of RADARSAT data.