| Can't say I understand it all, but it's interesting. In this particular case perhaps, for at least that particular run and maybe some others, the Euro might not be as accurate since it wasn't/isn't getting the initial conditions right. The storm is stronger than what the Euro depicts, at least from what it depicts on that site. This just happens to be the case for this storm for the run they mention and might not be for future runs. (because they can see what the initial conditions are for the model and compare them to the advisory data) I should note that ECMWF also releases text data that the HurricaneCity model system displays. For that data I don't know if the model wind speeds I get are the highest winds it analyzes/forecasts for the storm or if it is maybe the wind speed for the best resolution it can resolve using whatever technique it is using or something else. If I look at the Euro data that comes directly from ECMWF, the high resolution Euro had winds of 32.1 knots for the initialization at 0Z on August 19th. In the HurricaneCity model system, I have that data under the identifier "ECMO". Henri text data: http://hurricanecity.com/models/models.cgi?basin=al&year=2021&storm=08&display=model_text&latestrun=1&show_zero_hour=1 If Euro data is not there you have to go back a run. (By default, the model system doesn't show zero hour data but I load it by default in that link above) Henri had winds of 60 knots according to the NHC during that time. But if you look at Grace, while the storm had winds of 70 knots the high resolution Euro text data initialized it at 49.0 knots. Grace: http://hurricanecity.com/models/models.cgi?basin=al&year=2021&storm=07&display=model_text&latestrun=1&show_zero_hour=1 So that particular text model data in the HurricaneCity model system might always be lower for the Euro. Wind speed for global models in the text data I get from the NHC usually, maybe always, are not close to the advisory winds for powerful storms. I assume, but don't know, that it relates to resolution. Maybe there is a definition for what is actually released somewhere. For hurricane models you do get intensity forecasts. For weak storms the global models will be close to all the others on intensity. For stronger storms though you have a small area where the winds are much more powerful and the pressure much lower. For that, I assume that the data released is based on something that can't resolve that more intense area of the storm because the area is too small. I just don't know. On Tropical Tidbits' site, Levi Cowan does seem to have the pressures resolved. He downloads the raw model data that can display everything graphically and processes it. I don't know if that massive amount of data comes with it or if he somehow resolves it from it. All beyond my knowledge, I just plot the simple text data that comes from the NHC and let others interpret it. I don't resolve anything from the data, I just present it as is. |