Hurricane Ike's Spaghetti Model
Category: Local... Nature... Science...
Posted Tuesday, September 09, 2008 at 08:14 PM [GMT-6]

Update (9/13/08): Ike has made landfall and is now moving fairly quickly through North and East Texas. Its path is well documented, and based on what actually happened, it's obvious that the two individual models shown below on the composite spaghetti model that were most accurate were the Hurricane Weather Research and Forecasting Model (HWRF; shown in yellow) and the Navy Operational Global Prediction (NGPS; shown in light green). Both are so-called dynamical models (see below for more details). And, in case you're wondering, Ike's path was too far to the east to affect our area at all.

I flipped over to The Weather Channel to get an update on Hurricane Ike, and arrived just in time to see a spaghetti model of the projected path. A spaghetti model is a visual representation of multiple possible paths, each path being derived from a different computer model. Ike's spaghetti model is shown below (image via Tampa Bay's Fox affiliate):

Local residents will note that three of the projected paths cut through or end up smack dab over our neck of the woods, giving us the silver lining of the cloud that our coastal neighbors are dreading.

I'd never heard of a spaghetti model before (and, by the way, the Weather Channel folks were quite impressed with this one, as all the models are fairly well aligned, at least until Ike makes landfall, indicating a high level of certainty about the projected path) so I clicked over to learn about the various models that comprise the aggregate picture. The three paths that come closest to West Texas are generated by the following forecast models: Navy Operational Global Prediction System (NOGAPS), shown in bright green and labeled "NGPS"; Beta and advection model (shallow layer), shown in teal and labeled "BAMS"; and Beta and advection model (deep layer), shown in white and labeled "BAMD."

As you might imagine, each of these models is quite complex, and a detailed explanation is beyond my capability (and, probably and understandably, your interest). But here are the summaries of how much reliability we might put in each of them, courtesy of the National Hurricane Center.

• NOGAPS - NOGAPS is a "dynamical model," one of the most computationally complex of the weather prediction models. "...the NOGAPS model cannot provide very skillful intensity forecasts but can provide skillful track forecasts."


• BAMS & BAMD - These models are in a class called "trajectory models" and they are actually variations of the same model, differing only in the altitude of the steering winds that influence the path of a hurricane. According to the NHC, trajectory models "...are much simpler than dynamical or statistical models as they merely move a TC [tropical cyclone] along a track based on the prevailing flow derived from a separate dynamical model. While trajectory models utilize information from dynamical models to represent the prevailing flow, they do not allow the cyclone to interact with the surrounding atmosphere." As with NOGAPS, the BAM predictions are more skillful in predicting the path of the storm than its intensity.

If these three predictors are accurate, West Texas can expect some potentially significant and definitely beneficial rainfall as Ike moves inland and finally peters out. Unfortunately for those of us who would welcome that outcome, the models that show the track heading north and then northeast immediately after landfall are also the more complex and potentially more accurate of the suite. On the other hand, those models by virtue of their complexity also suffer from the magnification of errors early in the computations, and so their later predictions may be less accurate. (NOGAPS also suffers from this shortcoming.)

The good news is that even if the storm tracks further north, it will still likely dump rainfall in central Texas, and the more water that drains into Lake O.H. Ivie, the happier we consumers of that water will be.

Still, all things considered, I think it's obvious that we hope that Ike takes a completely different path and spares the coast completely. We in West Texas will happily live with a few inches less rain if it means that our coastal neighbors will miss a Category 2 or 3 hurricane.

Technorati tags: Hurricane Ike | Spaghetti Model