TreasureIslandFLGal wrote:Doesn't a human set the starting coordinates and initial direction for the model to run off of? or is that automated (and continuously faulty)? -frustrating....since many of the models use teh GFS as their own starting points. So that means that once again, half of them will be way off and not even worthy to consider other than the synoptic setup they show AROUND Emily.
It doesn't quite work that way with the dynamical models. These models create their initial conditions through a data assimilation process where atmospheric observations are blended with the model's previous best guess in a statistically optimal manner. This means that where observations are sparse, the atmospheric state in the initial conditions of the model, including such things as strength and positions of tropical cyclones and other weather features may not match reality very well. Thus the sometimes poor initialization of the initial positions of hurricanes. This is something of an oversimplification, but you get the general idea. However, that said, there *are* techniques that are used to attempt to semi-manually correct erroneous locations and intensities of TC's. These are called "bogusing" techniques, and they are meant to help with some of these initial condition issues, but they don't completely eliminate the problem.
Ultimately, the pathway to improving the initial conditions is to have more and better observations, and better techniques for assimilating them into the models. Both are extremely active areas of research right now, motivated in no small part to improving hurricane forecasts.
