HF RADAR COMPARISONS WITH MOORED ESTIMATES OF CURRENT SPEED AND DIRECTION - EXPECTED DIFFERENCES AND IMPLICATIONS

The validation of estimates of ocean surface current speed and directi on from high-frequency (HF) Doppler radars can be obtained through com parisons with measurements from moored near-surface current meters, ac oustic Doppler current profilers, or drifters. Expected differences be tween current meter (CM) and HF radar estimates of ocean surface vecto r currents depend on numerous sources of errors and differences such a s instrument and sensor limitations, sampling characteristics, mooring response, and geophysical variability. We classify these sources of e rrors and differences as being associated exclusively with the current meter, as being associated exclusively with the HF radar, or as a res ult of differing methodologies in which current meters and HF radars s ample the spatially and temporally varying ocean surface current vecto r field. In this latter context we consider three geophysical processe s, namely, the Stokes drift, Ekman drift, and baroclinicity, which con tribute to the differences between surface and nearsurface vector curr ent measurements. The performance of the HF radar is evaluated on the basis of these expected differences. Vector currents were collected du ring the High Resolution Remote Sensing Experiment II off the coast of Cape Hatteras, North Carolina, in June 1993. The results of this anal ysis suggest that 40%-60% of the observed differences between near-sur face CM and HF radar velocity measurements can be explained in terms o f contributions from instrument noise, collocation and concurrence dif ferences, and geophysical processes. The rms magnitude difference rang ed from 11 to 20 cm s(-1) at the four mooring sites. The average angul ar difference ranged between 15 degrees and 25 degrees of which about 10 degrees is attributed to the directional error of the radar current vector estimates due to the alignment of the radial beams.