Why Is Snow So Darn Hard to Forecast?

The just-ended snow storm was fairly well forecast in western Kansas, the Oklahoma Panhandle, the northern Texas Panhandle, northern Arkansas, Illinois, Kentucky and Indiana if preliminary reports are correct. 

The forecast wasn't very good in Oklahoma. And, the accumulating snow got into the Metroplex which is farther south than I thought. I'm really not surprised as I wrote this Saturday making the point that we need to do a better job of measuring the atmosphere over the eastern Pacific in these situations. 

There are at least two threads on Facebook and a blog post on an Oklahoma-based blog about how bad the forecasts were for the just-ended snow storm. I agree, they were not very good as to amount or location. About the only thing we got right in Oklahoma was that a snow storm would occur. 

click to enlarge

Here is a photo, taken at dawn this morning, that shows Dallas, OKC, Tulsa, Amarillo, and Lubbock as bright clumps of light and the snow on the ground in various shades of gray.

Without trying to make excuses, take a good look at the photo. Black is where no snow is on the ground. Look at the area of black in the southeast Panhandle almost totally surrounded by 3-4" snows! Look at the snow void between Norman (south extension of light clump around OKC) and Tulsa (below "no snow").

In the central U.S., I would much rather attempt to forecast tornadoes, flash floods and hurricanes than winter storms. Last week, we nailed the blizzard that extended from Denver to Des Moines to Green Bay. While I believe this poor quality of yesterday's forecast is the exception rather than rule these days, I don't think we are going to be able to consistently do a lot better until we start better measuring the atmosphere in the eastern Pacific.