A strengthening low-pressure system, eventually named Winter Storm Scott, brought accumulating snowfall to Colorado and the central Plains Saturday. The storm produced several feet of mountain snowfall, as well as up to 5-8 inches at lower elevations in Colorado and into Nebraska and Kansas. The system pushed eastward into The Ohio Valley on Sunday and brought a widespread swath of 1-3 inches of snowfall to the region.
Overnight Sunday night into Monday morning, Winter Storm Scott continued through the Mid-Atlantic and into New England, dropping up to 12-17 inches of snowfall across portions of Pennsylvania, New Jersey, New York, Connecticut, Rhode Island, and Massachusetts. The storm was able to pull significant amounts of moisture from the Atlantic Ocean, which aided in the significant snowfall. This was the only Nor'easter of the season so far, and in some places doubled or even tripled the seasonal snowfall to date.
Behind this storm, record cold weather settled in once again. As of 9:30 am EST Monday morning, 105 locations across the central and northern United States had broken or tied their record low temperatures for today, some breaking the monthly records, as well. High pressure at the surface reinforced the cold air and will help keep it around for the next couple of days.
Unfortunately, Winter Storm Scott had a devastating impact across the Southeast, as well.
Along the warm front of this system -- which stretched across Mississippi, Alabama, Georgia, and eventually into South Carolina -- several severe thunderstorms developed in a very ripe and unstable environment. These thunderstorms strengthened into supercells and produced a preliminary count of 43 tornadoes, several of which were large and destructive.
In particular, a long-track, tornado in Lee County, Ala., was rated EF-4 strength by the National Weather Service Birmingham. This tornado was estimated to have 170 mph winds (comparable to the strength of a Category 5 hurricane) and was 1 mile wide. Tragically, at last count 23 fatalities were confirmed from this tornado, with search and rescue efforts ongoing. Statistically, it is the deadliest tornado in the last six years, potentially more.
While snow forecasting is one of the more difficult aspects of meteorology, it becomes increasingly so during the late-winter season, namely March. This is because weather models and the atmosphere are going through several "transition" phases and gearing up for more spring-like weather.
One of the challenges of forecasting snowfall during this period is moisture. Models struggle with forecasting the amount of moisture that these systems will have with them, and they struggle by both under- and over-forecasting.
For example, with the recent minimal snowfall event Sunday, model data was very supportive of heavy snowfall for several days. However, as the event grew closer, it became evident that moisture was rapidly decreasing with each new run. This led to lesser snowfall amounts across southern portions of the Ohio Valley. On the other hand, warmer temperatures in March mean the air can hold more moisture, and sometimes these systems over-perform and bring higher snowfall amounts than anticipated due to a sudden influx of moisture from the Gulf of Mexico.
What follows is a brief introduction to model data and how we get it. Weather models are developed by highly skilled scientists and programmers who use climatological averages along with currently observed data and plug this into many physics’ equations. This then allows the model to "forecast" these values at various time periods into the future.
The problem with climate data is that these are averages and are not always representative of the current weather patterns. This leads to model data being biased, especially the further out the data is from the initial time. Recently, model data has been too cold in the medium- to longer-range forecasts, and has suggested more snow storms than have actually occurred. While there are several different models using various physics packages developed around the world, these model biases can significantly skew the data that meteorologists use.
It's important that we then take what we know about these biases, the current weather pattern, recent model performance, etc., and adjust as necessary for more accurate results.
Lastly, even when there is snowfall that verifies, it may not "look" like it. March sunshine is much stronger than December, January, and February, and often warms the ground, and especially paved surfaces, above freezing so that snow does not always stick and accumulate.
As forecasters who work in the professional snow and ice management industry, we know this can be particularly frustrating when planning for plowing versus salting. Unless snowfall rates are high during this time, accumulations are more likely to remain on grassy and elevated surfaces, and less likely to impact roadways.
Beth Carpenter is a co-founder and meteorologist at Thermodynamic Solutions, based in Indianapolis. You can reach Beth at email@example.com.