(An) approach for quality control and quality assurance of high-frequency radar-derived surface radial velocity maps off the Korean coast

Abstract

We present a statistical analysis to estimate uncertainty of high-frequency radar (HFR)-derived surface radial velocities based on concurrently and closely sampled radial velocity pairs from individual HFRs instead of comparing the currents sampled from individual HFRs instead of comparing the currents sampled from independent observational platforms (e.g., drifters or ADCPs), which allow us to report the observational uncertainty of the regional HFR system. We compute temporal correlations of the radial velocity pairs and standard deviations of their sum, which are theoretically given as a function of uncertainty, bearing angle of radial velocities, and anisotropic variance ratio and correlation of vector components. This approach is applied to the radial velocities sampled from (1) simple spectral model and (2) HFR radial velocity maps for a period of two years (2015 to 2016) in two Korean coastal re- gions (Yeosu and Busan) to evaluate the proposed method and to estimate the uncertainty. We identify outliers which are not following the expected correlations and standard deviations and evaluate the uncertainty in the two coastal regions.