M. J. Martin

Forecasting the ocean state using NEMO:The new FOAM system

The Forecasting Ocean Assimilation Model (FOAM) deep ocean analysis and forecasting system has been running operationally at the Met Office for over 10 years.The system has recently been transitioned to use the Nucleus for European Modelling of the Ocean (NEMO) community model as its core ocean component. This paper gives an end-to-end description of the FOAM-NEMO operational system and presents some preliminary assessment of operational and hindcast integrations including verification statistics against observations and forecast verification against model best guess fields.Validation of the sea surface height fields is presented, which suggests that the system captures and tracks the major mesoscale features of the ocean circulation reasonably well, with some evidence of improvement in higher-resolution configurations.

The Forecasting Ocean Assimilation Model (Foam) System

We present a detailed technical description of the present FOAM system and discuss some representative examples of the scientific investigations we undertake to track-down problems within the system and to understand the importance (“impact”) of the various inputs to it. We also provide an historical perspective on the development of the system and the changing demands for it, and describe the way in which we are adapting to meet these demands.

Evaluating a new NEMO-based Persian/Arabian Gulf tidal operational model

A 3-D baroclinic pre-operational model, including tides of the Persian/Arabian Gulf, has been developed at the Met Office using the NEMO framework. The non-assimilative model is believed to represent a significant improvement over the existing POLCOMS based system, benefiting from: extended domain; improved resolution; more accurate representation of coasts and bathymetry; improved representation of tides; and improved representation of salinity. As expected, with sea surface temperature (SST) data assimilation, the accuracy of SST is significantly improved. However, data assimilation also appears to help reduce thermal biases throughout the water column, within the limited accuracy of a climatology comparison. Operational implementation occurred in late 2012.

Assessing equatorial surface currents in the FOAM Global and Indian Ocean models against observations from the global tropical moored buoy array

Surface currents from 2007–2008 hindcasts of the Forecast Ocean Assimilation Model (FOAM) Global and Indian Ocean models are assessed against observations at 46 global tropical moored buoy array sites. Zonal (u) currents are less challenging to model than meridional flows (v) due to their lower frequency variability. The assimilative global model has reasonable skill for zonal currents but less skill for meridional currents. The assimilative models have higher skill than the corresponding non-assimilative models. A too-strong westward bias of the order of 20cm/s is evident along the equator in all model versionsused in this study. No extra skill is evident in the high resolution (1/12°) regional model compared to the coarser resolution (1/4°) global model.