Spaghetti Plots

Sometimes we show selected model track forecasts, so called “spaghetti plots,” on TV or on our web page.  Based on some emails that I have received, there is some confusion as to why we do this.  My only reason in showing the various models is to show that there is a certain amount of uncertainty in the track forecast and there are various scenarios that could develop. 

Spaghetti plot of selected models on TS Danny from 8 am 28 Aug 2009

Spaghetti plot of selected models on TS Danny from 8 am 28 Aug 2009

The above graphic shows a spaghetti plot of selected models on TS Danny from 8:00 am 28 Aug 2009 available to NHC forecasters when making their most recent forecast.

Numerous models for which the NHC has permission to share are available to all on various websites (the NHC also has a few other models that they don’t have permission to put online).  There are “early” models (available when the NHC forecaster makes the forecast), “late” models (typically only available after the NHC forecast is made), simple “statistical” models, sophisticated “dynamical” models, “global” models, “regional” models, combinations of “statistical-dynamical” models, “ensemble” models, and “trajectory” models. 

In addition, the NHC has made good use of “consensus” forecasts which are obtained by combining the forecasts from a collection of models. 

The National Hurricane Center (NHC) verifies these model forecasts along with their official forecasts.  I only show the more credible models in the spaghetti plot along with the official NHC forecast on TV.

There is an excellent Technical Summary of the National Hurricane Center Track and Intensity Models on the NHC web site that has been recently updated.


2 thoughts on “Spaghetti Plots

  1. Hi Max. It’s Amy (again). We’ve been fortunate with Bill and Danny so I hope our “luck” holds out for 94L. I realize that it is over 3000 miles away from South Florida and I see that some of the computer models are starting to pick up on it. GFS and GFDL I think are the more reliable ones and they’re indicating a possible turn to the north like Danny and Bill? What’s your interpretation of where it may go, even though it’s over 5 days out. Thanks and have a great weekend. Always nice communicating with you. -Amy-

  2. Max,
    A general discussion of model reliability/credibility would certainly be appreciated. I have often heard it said (in agreement with Amy) by both amateur and professional meteorologists that the GFS and GFDL models have proven themselves over time to be more reliable than others. Some also add HWRF to that list. At given stages in the development of tropical disturbances, are some models more reliable than others? Given the many types of models you listed, is there a general pattern in their sequence of use (e.g., global, then regional, then early, then late)? Have some models shown themselves to be more reliable early or late in the hurricane season? Etc. Frankly, given the large set of variables that can potentially impact a model as well as the likelihood of the effects of unknowns, it amazes me sometimes that any model at all is reasonably accurate. Anyway, some further discussion of model credibility would be welcomed. Thank you. -Jeff

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s