Abstract

Mesoscale eddy features are found ubiquitously throughout the world’s oceans. Many needs exist for numerical products from operational oceanographic systems to provide information on eddy properties. While numerous eddy identification and tracking methods have been developed for oceanic eddies, specific methods and metrics tailored to verify the skill of ocean analyses and forecasts in capturing these features are lacking. Here we introduce a novel feature-based verification methodology for operational oceanographic systems. This methodology builds on previous efforts at eddy tracking and applies open-source software to provide a robust method to evaluate the skill of operational oceanographic systems in terms of representing observed eddies. We demonstrate that an eddy tracking methodology can discern clear improvements in analyses produced using a regional analysis system (RIOPS; 1/12° grid-resolution) over a global system (GIOPS; 1/4° grid resolution). For eddies with amplitudes greater than 10 cm, RIOPS has a probability of detection 10%–30% higher than GIOPS with a false alarm ratio 5%–10% lower. A significant improvement in the spatial properties of simulated eddies in RIOPS is also found. In particular, results show a marked improvement in radius and separation distance errors (by 25% and 21% respectively), with fewer occurrences of errors above 20 km in radius and 40 km in separation distance. This basic demonstration opens the door for a more detailed examination of eddy features in ocean prediction systems.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call