American forecasters work to catch up to European model

It's an ongoing battle of sorts in the weather forecasting world, although if accuracy is the key – as it always is – the European model has quite often had a leg up on its American counterpart.

One of the more recent examples of the discrepancies between the two came last year after Hurricane Joaquin formed in the Atlantic Ocean.

Computer models created dozens of possibilities for the storm, with the American model predicting it could make landfall in the Tar Heel State.

The European model was drastically different, and in the end, right. Joaquin didn't make landfall anywhere along the East Coast and eventually turned north and east away from the U.S.

Dominque Marbouty, the former director of the group responsible for the European model, says the model has consistently outperformed its American counterpart.

Founded in 1975, the European model is a collaborative effort by 22 member states and 12 cooperating states to produce forecasts using real-time data. The "medium-range" forecasts, which were first issued in 1979, go out to about two weeks.

Hurricane names for the 2018 Atlantic season

ECMWF employs about 300 people from more than 30 countries, and Marbouty says it has one goal – to make the best global weather model there is.

"Numerical weather prediction, nothing else," Marbouty said of the focus of the model.

Marbouty says the diversity of the scientists producing the forecasts also contributes to its success.

"We bring together people with different cultures, different ways of working, etc.," he said.

The pay isn't bad, either, Marbouty said.

"You can attract some of the best scientists from all over Europe," he said.

The American model, also known as the Global Forecast System, is produced by the National Centers for Environmental Prediction, a division of the National Oceanic and Atmospheric Administration.

In an effort to catch up to the European Model, NOAA recently made significant updates to the American Model (GFS) computer systems to help create more accurate models.

The updates include a 4-D scheme for ingesting new data, with time now becoming the key fourth dimension in, NOAA said. This improves the initial conditions — the data the model uses to begin making its forecasts — and should result in better forecasts from the model.