HF radar two-station baseline bisector comparisons of radial components

One simple way to evaluate errors in the radial components of surface currents is to compare the values measured from two separate radars along the baseline joining them. This is best done at the midpoint of the baseline where the areas sampled by the radars are equal. This cannot be done if the baseline is close to the coast or over land. Here we compare radial components along the perpendicular bisector of the baseline and show that the RMS difference approaches the error as the point of observation approaches the midpoint. The application of this method to SeaSonde data showed rms differences decreasing until the observations were about 25 km off shore, and then increasing. We suspect that this increase near the coast is due to a known edge effect in the processing of radials that are derived from calibrated antenna patterns. If we exclude the near-coast data points then the rms differences extrapolate to a value of about 0.10 m s^-1 at the midpoint of the baseline.