Remote Sensing (Jul 2024)

Assessment of OMA Gap-Filling Performances for Multiple and Single Coastal HF Radar Systems: Validation with Drifter Data in the Ligurian Sea

  • Lorenzo Corgnati,
  • Maristella Berta,
  • Zoi Kokkini,
  • Carlo Mantovani,
  • Marcello G. Magaldi,
  • Anne Molcard,
  • Annalisa Griffa

DOI
https://doi.org/10.3390/rs16132458
Journal volume & issue
Vol. 16, no. 13
p. 2458

Abstract

Read online

High-frequency radars (HFRs) provide remote information on ocean surface velocity in extended coastal areas at high resolutions in space (O(km)) and time (O(h)). They directly produce radial velocities (in the radar antenna’s direction) combined to provide total vector velocities in areas covered by at least two radars. HFRs are a key element in ocean observing systems, with several important environmental applications. Here, we provide an assessment of the HFR-TirLig network in the NW Mediterranean Sea, including results from the gap-filling open-boundary modal analysis (OMA) using in situ velocity data from drifters. While the network consists of three radars, only two were active during the assessment experiment, so the test also includes an area where the radial velocities from only one radar system were available. The results, including several metrics, both Eulerian and Lagrangian, and configurations, show that the network performance is very satisfactory and compares well with the previous results in the literature in terms of both the radial and total combined vector velocities where the coverage is adequate, i.e., in the area sampled by two radars. Regarding the OMA results, not only do they perform equally well in the area sampled by the two radars but they also provide results in the area covered by one radar only. Even though obviously deteriorated with respect to the case of adequate coverage, the OMA results can still provide information regarding the velocity structure and speed as well as virtual trajectories, which can be of some use in practical applications. A general discussion on the implications of the results for the potential of remote sensing velocity estimation in terms of HFR network configurations and complementing gap-filling analysis is provided.

Keywords