Which is the best Weather Model?

Chris Bedford, who launched Marine Weather University alongside 2-time America’s Cup winner Peter Isler, guides us through the assortment of weather models for our needs:

Back in “the day” when I was first studying meteorology at The University of Michigan (between sandwiches from Zingerman’s Deli), it was still early days in the application of weather models. Global models were only just becoming practical, with the Global Spectral Model entering service in 1982.

The model had a horizontal resolution roughly 2.5 degrees and just 12 vertical layers, with both the horizontal and vertical resolution decreasing at longer forecasts. For regional forecasts, the Limited Fine Mesh model was available over North America at a resolution of around 125 km – so coarse it didn’t even resolve the Great Lakes!

With steady improvement in computer and communication speeds over the years, model resolution, performance, and availability has improved steadily and dramatically. Today, I can count at least 12 reasonably good Global Models running in operational mode – some with under 10 km horizontal resolution and often well over 100 vertical layers.

While resolution and run frequency varies among the models, all bring something to the table when it comes to forecast guidance – especially over open ocean. In addition, multiple global models are run in an “ensemble” manner (reference Marine Weather University’s Advanced Course), providing hundreds of additional global model runs per day!

At the continental, regional, and local scales, the number of available models is even more staggering, with many models run for specific areas around the world by national meteorological agencies, universities, and private entities. These higher resolution models are extremely useful for providing guidance in forecasting near-coastal winds.

To help guide through all this information, Marine Weather University has a new course: Weather Models 201 which will be presented in two parts.

And good news for Scuttlebutt readers. Use the code SBUTTFAM to receive a 10% discount for all MWU class and full length courses….through May 31, 2021.

Lecture 1 of this two-part series will go through the many models currently available as guidance in predicting marine and coastal winds. We will look at coverage/domain, resolution, update frequency, and what areas the models are best applied. We’ll look at some of the differences between the models and why some might be better than others.

That said, we’ll also remind ourselves that even “bad” models can contain useful information, but you often don’t know that a models is “bad” until after the weather has occurred.

In Part 2 of the MWU course we will explore strategies for proper use and application of weather model “guidance”. At the end of the day, you need to make a prediction. This prediction can support a safe and efficient voyage, or find the winning course in a race. But getting to a prediction can be difficult when faced with an excess of model guidance.

You may suffer from the fear of missing out (“FOMO”) – of piling on more and more models because you don’t want to miss the one that provides “the answer.” Likewise, you can also suffer from the fear of a better option (“FOBO”) – looking at more and more models hoping for a “better answer”.
We will address the problems of FOMO and FOBO head-on. We’ll look at some idiosyncrasies of models and resolution – especially as it applies to routing in marine navigation software packages like Expedition.

There absolutely is a point of diminishing returns when it comes to the breadth of weather model information out there. Meteorologists have come up with some clever ways to deal with the amount of model data out there and minimize the problem of diminishing returns.

We will remind ourselves that weather models aren’t an “answer” – they are a “guide”. Our predictions must be consistent with our understanding of the weather and the underlying science.

For more information go to Weather Models 201 at Marine Weather University.