Status of ENSEMBLE GLOBAL OCEAN WAVE FORECAST SYSTEM

Status of ENSEMBLE GLOBAL OCEAN WAVE FORECAST SYSTEM (EGOWa. FS) For 2 nd NAEFS Workshop November 16 -18, 2004 H. S. Chen NCEP/EMC/MMAB

1. Introduction. - Meteorological Forecast System (FS ) vs. Ocean Wave FS. - NCEP ocean wave model, NWW 3 : Directional, spectral wave model, http: //polar. ncep. noaa. gov/waves - Deterministic FS vs. Ensemble FS. 2. The Ensemble Global Ocean Wave FS (EGOWa. FS). 3. Preliminary Results.

1. Introduction. Initial Condition Meteorological FS: - chaotically nonlinear, - the butterfly effect. Sea surface forecast wind Initial Condition Ocean Wave FS: - weakly nonlinear, - highly dissipative. Forecast wave field

Figure 1. 1 A schematic illustration of the growth of an isopleth of the forecast error probability distribution function, from (a) initial phase, to (b) linear growth phase, to (c) nonlinear growth phase, to (d) loss of predictability. (borrow from ? ? ? )

Remark: - The initial wave field has a significant impact on the forecast waves only up to the first 24 hours and the wind forcing is the most dominant factor throughout the forecast period.

2. The Ensemble Global Ocean Wave FS (EGOWa. FS). - 10 ensemble wind fields + 1 control wind field (operational), - NWW 3 wave model, - 10 ensemble wave fields + 1 control wave field (operational). - Ensemble parameters: Ens Mean, Ens Spread and Ens Probability. - Output only up to 84 hour forecast, will increase in the future. - In Parallel Run. Hopefully, operational next spring. - use a considerable amount of computer resource. 3. Some results for 2004 June.

Green thin lines: ensembles, Red dash line with + sign: control, Blue line with + sign: ensemble mean Black dot with o sign: observed data.

Some Remarks: - The ensemble FS is more reliable and realistic than the deterministic FS. - A long list of works need to be done, such as verifications, postprocessors, score analysis, …etc.