Validation of radar-based Lagrangian trajectories against surface-drogued drifters in the coral sea, Australia

Surface current velocity fields measured every 10 min by a high-frequency ocean radar system (HF radar) located at the southern Great Barrier Reef (GBR), Australia, were used to compute Lagrangian trajectories (radar-based trajectories). The radar-based trajectories were validated against surface-drogued satellite tracked drifting buoys released on the shelf inside the reef lagoon and on the continental slope. Current speeds estimated from the drifters were typically within 5% to 8% of the HF radar currents extracted at the exact position occupied by the drifters for each sampled time, but some large current velocity biases occurred over short periods of time in shallow areas. Maximum separation distances between the drifter and radar-based tracks ranged from 5 km to 14 km on the shelf and from 4 km to 35 km on the slope for 24 tracks tracked up to one week. The reduction of the predictability of the radar-based trajectories occurred mostly in the vicinity of the islands of the reef matrix, suggesting the influence of small scale processes in the drifter paths not seen at the HF radar spatial resolution. A filtering procedure was applied a posteriori to the u and v components of the current velocity vectors calculated from non-quality controlled radial data obtained from the IMOS archive. Interpolation was used to cover temporal gaps less than 3 h allowing radar-based trajectories to be followed for up to 7 days. The use of radial data reprocessed and quality controlled by the Australian Coastal Radar Network (ACORN) extended the tracking period up to 13 days, reprocessing allows retrieval of some information lost in the non-quality controlled radial data.