A look ahead, but what to believe?

For sure, one thing that has been learned over the past few years is that looking, simply, at a model run is not enough. One must look at the ensembles. The same model with little tweaks of data upon the initialization of the model run that allows for subtle changes and thus, different solutions. This will give you a better idea of what might happen by seeing where the bulk of the soltuions end up. But then, sometimes, like this example, who knows what to believe? This is a 204 hour solution of the GFS (about 8 1/2 days out) for North Ameria. On the left is the ensemble means (the average of all the possible solutions of this model.) The right shows the "spaghetti plot" which is to say it shows every model solution thus creating a thing that looks like spaghetti. The white line on the right is representative of the operational GFS model (one member of the ensembles.) The ensembles really don't agree well at all here and the GFS operational solution is an outlier! So, what will happen in 8.5 days? Your guess is as good as mine!!

Thank you to PSU's E-Wall for this set of images. The E-Wall (just google PSU E-Wall) is an awesome site!