Conflicting swell height in models
-
Regularly there is a very marked difference in the swells reported
by the ECMWF, GFS and ICON. For instance, in the Atlantic Ocean east of the Caribbean Islands (17, -61), models show same wave height; however both models will show a significant difference in the swell heights. The ECMWF and ICON will show swell heights of 7 feet and the GFS will show swell heights if 0 feet. Is this an issue in the resolution or parametrization used between models? -
@ljos Hello, could you please enclose a screenshot?
-
Hi there sure. Sorry I just saw your post . The information may be slightly different because of model update. Notwithstanding, the discrepancy continues. Here is a screen shot of ECMWF
Here the GFS screenshot
Here is the ICON model screenshot
Please note that at the reference point the Swells for the
ECMWF is 5-8 feet
GFS is 0-3 feet
ICON is 6-8 feetClearly something is going on for such a large discrepancy between the GFS and the other models .
Will await your response
Ljos -
@ljos Hello, each forecast model can produce slightly different values, therefore such differences are not discrepancies.
-
@korina Hi Korina thanks for your response. I accept what what you mentioned; however the difference you point out is not a one day occurrence; it is consistent - model run after model run; so that's the reason why I describe it as such.
-
@ljos
It seems that GFS wave model shows weaker swell heights compared to ECMWF and to DWD wave models (both uses WAM model but with probably minor changes and forced by 2 different wind models, IFS and ICON).
These screenshots from yesterday show a big difference of swell heights in the overall tropical part of the Atlantic and in other areas.Is it a wrong visualisation from Windy? Probably not because at same time the NOAA shows the same swell map from GFS but with a different color scale (deep blue is less than 0.5m which corresponds to light blue in the Windy color scale)
Finally which the best model for predicting Swell? Difficult to say. But I remember good prediction of ECMWF in european seaside.