Intercomparison and validation of detected SST fronts based on CMEMS high-resolution reanalysis data and satellite observations in the South China Sea

Oceanic fronts, as important mesoscale phenomena, are transition zones between two water mass with different hydrological features, which has important significances in various fields, such as fishery, environment protection and marine military. In order to effectively obtain the ocean front information, a gradient-based Canny edge detection algorithm is applied to detect the ocean front, and post-processing procedure specifically designed for high resolution hindcast and reanalysis is utilized to extract primary ocean fronts, ensuring the balance of frontal continuity and positioning accuracy. Based on CMEMS reanlaysis and satellite SST data and regional model hindcast from NMEFC in China, a long-term intercomparison and validation is performed to evaluate the frontal detection method and model behavior in aspect of mesoscale forecasting/hindcasting. Frontal statistical product include mean status, seasonal and interannual analysis are utilized to intercompare. We also develop ‘CLASS 4’ validation metric to validate daily results in different datasets. Further more, forecast error, bias and skill of operational ocean front product are also shown for potential users.