"Mammoth Snow Storm" Underwhelms

mi (197448) writes You heard the scare-mongering, you heard the governors and mayors closing public transit and declaring driving on public roads a crime. But it turned out to have been a mistake. Boston may have been hit somewhat, but further South — NYC and Philadelphia — the snowfall was rather underwhelming. Promised "2-3 feet" of snow, NYC got only a few inches. Is this an example of "better safe than sorry," or is government's overreach justified by questionable weather models exceeding the threshold of an honest mistake?

Comments

Problem was underinvestment

[gurps_npc]

Up until recently, the US weather prediction was SIGNIFICANTLY inferior to European. They talked about the American Model vs the European Model, and the European Model was consistently correct.

People have finally begun to realize this problem, and created a new American Model. The predictions of large NYC and Philly snowfalls came from the Old American Model. The new American Model, along with the European Model, both correctly predicted the snowfalls.

The New American Model requires significantly more computer power to use. It has not been thoroughly tested. But expect to see it being used more often after this success.

Re:Problem was underinvestment

[Anonymous Coward]

Actually the European model also predicted more snow. From the Wunderground blog,

The 7 am EST (12 UTC) Monday run of what is usually our top forecast model, the European model, predicted that the storm would track about 100 miles farther west than it actually did. The American GFS model, which just underwent a significant upgrade over the past month to give it increased horizontal resolution, performed better, putting the storm farther to the east. Forecasts that relied too heavily on the European model put too much snow over New York City.

Re:Communicating probabilities

[BenFranske]

I'm guessing that you do not live in a location which regularly sees substantial snowfall. If you did you would realize that, at least with current models, this would be pretty much impossible. Snowfall amounts are one of the most difficult things to model and are notoriously incorrect.

Unlike precipitation like rain, where the density is always the same, with snow the ambient temperature and humidity level play a huge role in determining how dense the snowfall is (heavy wet snow vs light fluffy snow). We can predict the amount of water which will fall from the sky during a snowfall with the same probabilities, amounts, and accuracy as with summer rains (which we're reasonably good at). The problem is that depending on the density of the snow (which is much harder to predict) that same amount of water can give a snowfall of between 5 and 20 inches.

Re:Not underwhelms, a little off predicted target.

[SternisheFan]

Long Island N.Y. has over 5 million people living there, and is car-centric. The predicted warnings of heavy snow caused almost all businesses and schools to shut down today, which probably saved many lives. About ten years ago two inches of snow paralyzed N.Y.City because they weren't prepared for it (upstaters had a good laugh then).

News for nerds. Stuff that matters.

[MobyDisk]

News for nerds. Stuff that matters.

Bad example: Whining about how a snowstorm wasn't big enough.

Good example: A discussion about the mathematical modeling used to predict the snowstorms, and a historical graph comparing predictions to actuals. Oooh, how about graphing the delta between the two by color and overlay it onto a map so we can see where predictions are more or less accurate.

Re:Boston Representing

[houghi]

http://imgur.com/gallery/ILX0g... That is boston.