Improving SeaSonde radial velocity accuracy and variance using radial metrics

The surface current observations made by HF radar systems are increasingly being relied on to support decision-making by coastal ocean users and managers. Thus, the need for high quality data with defined error characteristics has never been higher. Building on previous efforts, this paper examines the use of `radial metrics', the ancillary output of the direction finding algorithm used by the SeaSonde radar system, to improve both accuracy and error characterization. The radial metrics provide non-velocity based parameters that can be used to eliminate low-quality data and or weight the contribution of individual returns on the resulting spatially averaged radial velocity. While their effects on the data are different, both thresholding and weighted averaging based on the metrics are shown here to significantly increase the accuracy and reduce the error of the resulting velocity estimates. Thus, we suggest that operators should take advantage of these advancements to improve the accuracy of existing systems.