Abstract
Mesoscale eddies dominate energetics of the ocean, modify mass, heat and freshwater transport and primary production in the upper ocean. However, the forecast skill horizon for ocean mesoscales in current operational models is shorter than 10 days: eddy-resolving ocean models, with horizontal resolution finer than 10 km in mid-latitudes, represent mesoscale dynamics, but mesoscale initial conditions are hard to constrain with available observations. Here we analyze a suite of ocean model simulations at high (1/25°) and lower (1/12.5°) resolution and compare with an ensemble of lower-resolution simulations. We show that the ensemble forecast significantly extends the predictability of the ocean mesoscales to between 20 and 40 days. We find that the lack of predictive skill in data assimilative deterministic ocean models is due to high uncertainty in the initial location and forecast of mesoscale features. Ensemble simulations account for this uncertainty and filter-out unconstrained scales. We suggest that advancements in ensemble analysis and forecasting should complement the current focus on high-resolution modeling of the ocean.
Highlights
Mesoscale eddies dominate energetics of the ocean, modify mass, heat and freshwater transport and primary production in the upper ocean
We define the skill of the model as its ability to have lower root mean square error (RMSE) than a monthly mean climatology of the ocean observations
It is possible that a selection of a different error metric might highlight the differences between the two deterministic models
Summary
Mesoscale eddies dominate energetics of the ocean, modify mass, heat and freshwater transport and primary production in the upper ocean. The forecast skill horizon for ocean mesoscales in current operational models is shorter than 10 days: eddy-resolving ocean models, with horizontal resolution finer than 10 km in mid-latitudes, represent mesoscale dynamics, but mesoscale initial conditions are hard to constrain with available observations. Non-assimilative models are insufficient for accurate prediction of evolving mesoscale eddies and fronts due to mismatches in the initial locations of eddies and associated non-linear process, resulting in large forecast errors. This source of error could be diminished by accurate initialization of the ocean dynamic state through assimilation of ocean observations that are used to constrain the mesoscale features, thereby extending the forecast horizon[9,10]. We provide a quantitative analysis that shows how the ensemble mean selectively filter out scales that contribute to the forecast error
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
