Calibration and Validation of Direction-Finding High-Frequency Radar Ocean Surface Current Observations

This paper focuses on the validation of remotely sensed ocean surface currents from SeaSonde-type high-frequency (HF) radar systems. Hourly observations during the period July 22, 2003 through September 9, 2003 are used from four separate radar sites deployed around the shores of Monterey Bay, CA. Calibration of direction-finding techniques is addressed through the comparisons of results obtained using measured and ideal (i.e., perfect) antenna patterns. Radial currents are compared with observations from a moored current meter and from 16 surface drifter trajectories. In addition, four overwater baselines are used for radar-to-radar comparisons. Use of measured antenna patterns improves system performance in almost all cases. Antenna-pattern measurements repeated one year later at three of the four radar locations exhibit only minor changes indicating that pattern distortions are stable. Calibrated results show root-mean-square (rms) radial velocity differences in the range of 9.8-13.0 cm/s, which suggest radar observation error levels in the range of 6.9-9.2 cm/s. In most cases, clear evidence of bearing errors can be seen, which range up to 30deg for uncalibrated radar-derived radial currents and up to 15deg for currents obtained using measured antenna patterns. Bearing errors are not, however, constant with angle. The results recommend use of measured antenna patterns in all SeaSonde-type applications. They also recommend an expanded simulation effort to better describe the effects of antenna-pattern distortions on bearing determination under a variety of ocean conditions