CO10 operational model use case statement

[ukmo-ansaulter]

Issue describes the primary operational short-medium range forecast use cases for Northwest European Shelf Models and key variables which should maintain or improve performance under the CO10 configuration.

Use case 1: Atmospheric model surface boundary layer
Primarily quality of SST (minimal bias, good description of geographic variation and rapid changes greater than 1 degree). Secondary, accurate depiction of (strong) surface currents and surface roughness (when including coupled wave model).

Use case 2: Search and Rescue/Marine Pollution response
Primarily quality of surface currents (strength, direction and timing of changes) from top 1m to 15m. Secondary variable water temperature (survivability).

Use case 3: Sub-surface T,S profile
Primarily quality of speed of sound profile and temperature at seabed. Also useful to understand quality of timing in onset of stratification across the NWS domain?

Use case 4: Sea-surface height/surge
Quality of total water level (tide+surge), particularly during extreme storms; intercompared with dedicated surge model.

Use case 5: Ocean colour/visibility
Quality of visibility versus secchi depth?

Use case 6: Marine ecosystem stress indicators
Ability of the model to detect extremes in: ocean temperature, dissolved oxygen, nutrients (eutrophication), algal blooms

[ukmo-ansaulter]

Proposed success criteria for the use cases:

Use case 1: Free running model has similar or reduced bias/RMSD against observed SSTs compared to control. Model with assimilation has similar or reduced SST innovations.

Use case 2: Model has similar or reduced errors for surface drift case studies (e.g. Bruciaferri et al., 2021) compared to control.

Use case 3: Free running model has similar or reduced errors for observed T,S profiles compared to control; onset and breakdown of stratification is compared with control model(s). Model with assimilation has similar or reduced T,S profile innovations.

Use case 4: Model has similar or reduced errors for total water level and/or storm surge against (at least) UK tide gauges compared to control model. Model has similar or reduced errors in sea-level anomaly bias/RMSD against observed SLA compared to control.

Use case 5: Model+BGC component has similar or reduced bias/RMSD versus observed secchi depth.

Use case 6: Model+BGC component detects an agreed set of marine ecosystem extreme events.