Our network

Did The Models Fail With Sandy? | Weather

Title (Max 100 Characters)

Did The Models Fail With Sandy?
Did The Models Fail With Sandy?

With Hurricane Sandy something of the past now, it is brought to our attention that our technological advancement here in the U.S. is far behind that of other countries. Well over a week prior to Sandy made landfall in the Northeast, some of the weather computer models were predicting “Sandy” (unnamed at the time) to be an unusual, yet very strong storm to the very region it hit. Some of the other computer models predicted that the storm would develop, but would head out to the Atlantic and spare the U.S. entirely. While all of the models ended up with the same consensus, including the National Hurricane Center (NHC) 4 days before it hit the New Jersey/New York, was that enough time for all residents, businesses, and governments to prepare? It also limits the time to rush supplies to that area that they will need once the storm hits.

While the actual weather computer models themselves were invented here in the U.S., the level of accuracy and ability of those models have fallen substantially behind those of other nations’ models. In fact, Kerry Emanuel, a writer for the Wall Street Journal, believes we may actually be in third or even fourth place compared to other nations. So who is the leader of the pack? What is commonly referred to as the Euro Model, is definitely at the top. In fact, this year alone, the Euro correctly predicted the tracks of both Hurricane Isaac and Sandy well before the other models did. This is important since these were the two big storms that actually hit the U.S. The Washington Post noted that the NHC’s parent company, NOAA had mentioned the European model’s scenario on its website, saying that if it proved true, it would create “a powerhouse capable of whipping the Atlantic into a frenzy and churning up dangerous tides.” However, it did not go as far as to agree with the model, just to say that basically “this is what the Euro model is predicting”.

“[The U.S. has] tremendous weather forecasting capacity, but we’re still losing to the Europeans,” said Heidi Cullen, chief climatologist at Climate Central, a nonprofit that communicates science to the public.

Why is the Euro so much better than the others? Well, for starters, the Euro model tends to focus more on numerical weather prediction and run their models on larger and faster computers.

In fact, Judith Curry, chairwoman of the School of Earth and Atmospheric Sciences at Georgia  Tech, said about the Euro, “[It’s] better overall”.  She also went on to say that the model has a higher resolution than its U.S. equivalent, makes more frequent forecasts, and is better at assimilating satellite data.

However, one area where we excel over them is that we recognize that this information is a public good, and since it is funded largely by the taxpayer, it should be available free of charge to the public. Why is that so important? Well, ironically, some European businesses prefer to use U.S. forecast models because they are free, whereas some U.S. businesses will actually spend very large amounts of money for European forecasts because they are better. So if only we could provide the same accuracy that the Euro does, and free like we currently do, we certainly would get back on top again.

Marshall Shepherd, director of the atmospheric sciences program at the University of Georgia, blames lack of funding for many of the problems facing NOAA. “That only gets worse going forward as we talk about sequestration and cutting agencies’ budgets,” he said.

But is more money necessarily the solution to fixing this problem? What if the real solution lies with restructuring the current NOAA system. Why not do like the European model does, and instead of only having one country’s model be the end all be all, use multiple countries to provide that model data? Conceivably we could have Canada and Mexico join in with us, providing us with a larger blend of government, media, academic, and private sector talent. Perhaps then could we get back on top again, and not have to rely so much on other computer models, simply because they are better than ours.