Event Summary
     National Weather Service, Raleigh NC


January 19, 2005 Winter Storm





Event Headlines

...Snow developed in a Northwest Flow aloft, a pattern that typically produces little if any measurable snow across central and eastern North Carolina...
...Only a trace to around 1 inch of snow fell across much of central North Carolina, while a few counties north and east of Raleigh received 2 inches...
...High snow to liquid equivalent water ratios, averaging around 16 – 22 to 1 were observed...
...The snowfall of an inch or less in Raleigh resulted in 8 hours of traffic gridlock...
...Unusually cold road surface temperatures, air temperatures in the lower 20’s, and a mass exit of traffic from early afternoon closures all resulted in very slick driving conditions, numerous accidents, and traffic jams...




Event Overview

A disturbance in the upper atmosphere moved southeastward across the Appalachians into North Carolina and Virginia on Wednesday, January 19, 2005. Typically, weather disturbances in a northwest flow aloft weaken and lose their limited moisture as they cross the mountains, resulting in little or no accumulating snow east of the mountains. Indeed in the Triad area, only a trace to a dusting of snow was reported. As the system neared the Triangle area, the area of snow briefly expanded in aerial coverage as well as intensified. Nearly an inch of snow accumulated in the city of Raleigh, while a few counties north and east of the city received two inches.



Snow Accumulation Map

A small portion of North Carolina received snow accumulations greater then an inch. These locations were generally extended from Raleigh north and northeast to the Virginia border. The vast majority of locations that experienced snow received only a dusting to a half of an inch of accumulation.

Snow Accumulation Map



Unusual High Impact Event

While it is not unusual for local businesses and schools to close early when there is accumulating snow, the extreme impact of such a small amount of snow was nearly unprecedented. Few, if any in the local community, would have anticipated so little snowfall causing city wide traffic jams. Motorists took several hours to travel commuting distances that normally take only tens of minutes.

This summary will focus on the factors that led to this small snowfall, yet high impact event. Also, we will highlight those meteorological features that likely played a role in producing the unexpected and sudden intensification of the snow just as it moved into Raleigh.

Photo of light snow cover in West Raleigh



Numerous Factors Considered in Forecasting a Winter Weather Event

As meteorologists apply the forecast process to project the impact of a potential winter weather event upon the local community, a great deal of complexity is involved calling for the integration of numerous factors that are dynamic and changing over both spatial and temporal scales. The forecast process typically begins with evaluating the synoptic scale meteorological patterns, especially regarding the potential for cold-air damming, the general flow pattern aloft, and the overall pattern of surface cyclogenesis. These factors are known to be well correlated with wintry precipitation types and the spatial distribution of precipitation types across Central North Carolina. Specific forecast tools are then applied to further resolve the precipitation type issues. Meteorological mechanisms for generating precipitation amounts are then reviewed. Since precipitation type can be influenced by the amount of precipitation, forecasters must then reconsider their precipitation type predictions, determining if any modifications are needed.



Key Meteorological Features

Forecasters routinely look upwind to monitor the evolution of an area of precipitation as it approaches and to diagnose the meteorological features that are producing the precipitation. Atmospheric disturbances come in a variety of forms and are indicative of differing mechanisms; however, they all share a common dominator which in large degree represents the “Holy Grail” for operational meteorologists. In short, they all have the capability of producing an upward component in the wind. It is this upward motion (i.e., lift) that gives rise (no pun intended) to clouds and precipitation.

Forecasters were monitoring conditions well upwind preceding the snow on the 19th. On the 18th, disturbances in the upper atmosphere were noted to be moving southeastward into the Ohio Valley. Despite weather radars showing widespread light snow, little snow had reached the ground in Kentucky during the early predawn hours of the 19th. The snow had been evaporating in the very dry air below the cloud bases.

By 700 AM (12Z) on the 19th, the stronger radar returns had moved into West Virginia and the western sections of Virginia and Pennsylvania. An analysis of upper level data at that time revealed a number of meteorological features. These features combined to produce sufficient upward motion resulting in the light snow as indicated by the stronger radar returns. The meteorological features noted included: 1) the nose of an upper level jet (note contour showing winds of at least 75 knots); 2) a 500 mb trough (note “u” shape height line contours); and 3) warm air advection at 850 mb (note -1 C air at Nashville moving toward much colder air, -9 C at Blacksburg, VA.).

Meanwhile due to dry air in place, the anticipation of the weather system weakening as it crossed the mountains, and the limited moisture available to the system, forecasters in Central North Carolina were anticipating only enough light snow to produce no more than a “dusting”. Indeed, the system did weaken as it crossed the mountains, as indicated by the diminishing radar returns at 9 am (14Z). As forecast, most counties west of Raleigh reported only a trace of snow as the light precipitation continued to move eastward.

As the precipitation neared the Raleigh area at noon (17Z), much of the snow was still not reaching the ground. Accumulating snow was light and generally confined to a narrow band of stronger radar returns indicated by the green shaded radar returns. This band was moving so fast that snow accumulations were well less than an inch. But by 1 PM (18Z), there had been a significant increase in both the intensity and the aerial coverage of the radar returns.

An analysis of real time data available early that evening, after the snow had moved eastward, provides insights. Note the twelve hour differences (7 AM vs. 7 PM on the 19th) in the meteorological features as they tracked from the Ohio Valley: 1) the nose of the upper level jet at 300 mb expanded southward through the Carolinas; 2) the 500 mb trough strengthen on its southern flank into the Carolina (height falls of 60 and 120 meters are shaded in red and represent a measure of the trough’s intensification); 3) moisture at the 850 mb level (shaded in green) also expanded. In short, the disturbances briefly strengthen farther southward than expected.



Raleigh - KRAX Radar Imagery

Raleigh, KRAX WSR-88D base reflectivity imagery from 1745Z (145 PM EST) Wednesday, January 19, 2005 is shown below. Note the area of enhanced reflectivity northeast of Raleigh. A second linear area of snow (stretching from south-central Virginia to near Greensboro early in the first loop below) "catches up" to the primary area of snow (stretching from Kerr Lake to Raleigh to near Fayetteville). The snow appears to redevelop and intensify across the region northeast of Raleigh between about 1715Z (1215 PM EST) and 1815Z (115 PM EST) as these feature "merge". The increase in coverage and intensity of the snow during this hour resulted in the heavier and more significant accumulations.

A Java Loop of KRAX Radar imagery from 1717Z (117 PM EST) to 1814Z (214 PM EST) shows the increase in coverage and intensity of the snow across the region northeast of Raleigh in the northern Piedmont of North Carolina.

A Java Loop of KRAX Radar imagery from 1458Z (958 AM EST) through 1900Z(300 PM EST) shows the evolution of the entire event across North Carolina.

Raleigh, KRAX Radar imagery from 1745Z (145 PM EST) Wednesday, January 19, 2005



Regional Radar Imagery

Regional composite reflectivity radar imagery from 16Z (1100 AM EST) Wednesday January 19, 2005 is shown below. The primary axis of precipitation is visible stretching from near Washington D.C. south through central Virginia to northern North Carolina. This area of precipitation was the primary snow producer across the region. Note the second area of scattered precipitation stretching from southwestern Virginia southwestward to northeast Tennessee. This second axis of precipitation quickly rotated east and "merged" with the first area of precipitation resulting in the increase in coverage and intensity of the snow across locations from Raleigh north and east.

A Java Loop of Regional Radar imagery from 10Z through 22Z Wednesday January 19, 2005 is available.

Regional Radar from 16Z Wednesday January 19, 2005



Forecasting Snow Amount

When the precipitation type is expected to be snow, there are still numerous other factors that must be considered before predicting how much snow will accumulate. These factors include: the potential for snow mixing with or changing to other wintry precipitation types (a frequent occurrence in Central North Carolina), the rate of snowfall, the impact of ground and road surface temperatures on snowfall accumulation, and the ratio of liquid precipitation to snow amount.



Snow to Liquid Equivalent Ratios

Measurements show that the ratio of snow to liquid equivalent ranges from very high numbers (e.g., 60:1) to very low numbers (e.g., 2:1). Meteorologists receive much in the way of forecast guidance from a variety of numerical prediction models; however, these objective models only forecast liquid amounts. It is left to the forecaster to determine the ratio of snow to liquid equivalent.

The snow to liquid equivalent ratio is generally a function of the temperature in the lower portion of the atmosphere (surface – 5000 feet), the amount of moisture in this layer, and the structure of the snow crystal present. Climatological studies for Greensboro, NC indicate that the average ratio there is 10:1, or 10 inches of snow to 1 inch of liquid equivalent. The most frequently observed ratio at Greensboro is 8:1, while 75% of all snow events showed a ratio less than 12:1.

Due to the proximity of the Gulf of Mexico and the Atlantic, it is more common for plentiful moisture to be found in snowfall events in central North Carolina. The higher moisture content results in a more compact, dense snow and hence lower ratios of snow to liquid equivalent. This is not typically the case for some other regions of the country (e.g., northern plains states) where moisture content is typically more limited and hence snow ratios are higher due to less compaction.

High ratios of snow to liquid equivalent in Central North Carolina are typically associated with disturbances moving into Central North Carolina from the northwest as was the case on January 19th. The snow ratios measured during the January 19, 2005 event are shown in the table below. The liquid equivalents ranged from 0.04" to 0.09" with snow accumulations of 0.7 to 2.0 inches. The resulting snow ratios ranged from around 16:1 to 22:1. Note that the measured snow ratio is inherently higher in lighter snow events, since compaction and settling of the accumulated snow is minimal.



Snow ratios measured during the January 19, 2005 snow event at selected locations



Snow Accumulation, Ground Temperatures and Snowfall Rates

In addition to the forecaster determining the overall meteorological pattern, the predominant precipitation type, the spatial distribution of precipitation types, the mechanisms for precipitation, the amount of precipitation, and the snow to liquid equivalent ratio; other factors still play an important role in the amount of snow accumulation. These factors included the rate of snowfall as well as the temperature and moisture characteristics of the ground and road surfaces.

The 60 hours preceding the 19 January snowfall were unusually cold. Temperatures remained near or below freezing during the entire period. High temperatures on Monday, January 17 were only in the 30s while the low temperatures on Monday night were in the teens. High temperatures on Tuesday, January 18 were in the mid 20s to mid 30s with overnight lows on Tuesday night between 8 and 15 degrees. The impact of this cold weather on ground temperatures is seen from those stations where soil temperatures are reported. The image below, provided by the State Climate Office of North Carolina, is a plot of the average hourly soil temperature at a depth of about 4 inches at the Lake Wheeler Road Field Lab in Raleigh, NC from 100 AM EST on January 16 through 1159 PM EST on January 20, 2005. Note that the ground temperature at the 4 inch depth level was only a bit above freezing in the hours just preceding and during when the snow fell in the Raleigh area on 19 January. With ground temperatures frozen to nearly the 4 inch soil level, there was no melting of the limited snow that fell. Also, the road surface temperatures were several degrees below freezing at the onset of the snow.

As light snow moved into the Raleigh area very early in the afternoon, there were numerous reports of snow blowing across the roads. The light snow then became occasional bursts of moderate snow (visibility reduced to ½ mile) in Raleigh between 100 and 200 PM . After the bursts of moderate snow developed, it took little time for road surfaces to turn white. With the moderate snow, temperatures in Raleigh fell from the upper 20s into the 23-25 degree range. The snow lasted only a little over 2 hours. Most of the snow was very light. But the occasional burst of moderate snow allowed for an efficient means to quickly cover the roads with a light coating.

Plot of the average hourly soil temperature at a depth of about 4 inches at at the Lake Wheeler Road Field Lab in Raleigh, NC




Sociological Factors Leading to a High Impact Event

In a “tongue and cheek” fashion, it has been reported in the local media that “it is the duty of southerners at the first mention of snow to rush to the grocery stores for bread and milk and to test their driving skills once the snow begins to fall.” While an exaggeration, there is never the less a "knee jerk" response to snow that is indeed engrained in southern culture. This response may well be founded upon the relatively infrequency of snow in the South. Relative to the “snow belt” locations in the north, there is certainly less experience in dealing with snow, and less equipment and resources for coping with snow. In the South, snow certainly draws more attention and that in itself results in more of a disruption of daily routines. There are also relatively infrequent heavy snows in the South (e.g., 22.5 inches in Raleigh on January 24-25, 2000) that are etched into the memories of those who experienced them and the anticipation of their reoccurrence.

Perhaps then, it should come with no surprise that there were so many early releases of schools and businesses during the early afternoon of Wednesday on January 19th. What was a surprise is how quickly and efficiently a light coating of snow on roads became so icy and slick. The mass exodus of many vehicles at nearly the same time clogged the primary roads. Road surface temperatures were too cold to promote melting. Instead, the snail paced traffic compressed the light snow while the heat from vehicular exhausts produce a slight melting effect. The combination of compression and slight melting changed the snow to an icy composition which the cold air temperatures maintained. Numerous accidents on the now slick roads prevented highway crews from spreading salt and sand on the gridlock primary roads. Many school buses were unable to make their rounds, while parents stuck in traffic were unable to get to the schools. What is normally a commuting trip consisting of just tens of minutes turned into an eight hour ordeal.



Archived Text Data from the Winter Storm

Select the desired product along with the date and click "Get Archive Data."
Date and time should be selected based on issuance time in GMT (Greenwich Mean Time which equals EST time + 5 hours).

Product ID information for the most frequently used products...

RDUAFDRAH - Area Forecast Discussion
RDUAFMRAH - Area Forecast Matrices
RDUHWORAH - Hazardous Weather Outlook
RDUNOWRAH - Short Term Forecast
RDUPFMRAH - Point Forecast Matrices
RDUPNSRAH - Public Information Statements (snow/ice reports among other items.)
RDUWRKDRT - Soil Temperature Data from the NC State Climate Office
RDUWSWRAH - Winter Storm Watch/Warning/Advisory
RDUZFPRAH - Zone Forecast Products


 from 



Selected Photographs of the Winter Weather Event

Photos courtesy of Jonathan Blaes.
(Click the image to enlarge.)


Trees on NC State’s Centennial Campus - Click to enlarge                      Research III Building on NC State’s Centennial Campus - Click to enlarge                 Water Fountain on NC State’s Centennial Campus - Click to enlarge

Lake Raleigh on NC State’s Centennial Campus - Click to enlarge           Planter Next to Research III on NC State’s Centennial Campus - Click to enlarge           Snow covered Parking Lot Next to Research III on NC State’s Centennial Campus - Click to enlarge

Engineering Graduate Research Center on NC State’s Centennial Campus - Click to enlarge           Engineering Graduate Research Center on NC State’s Centennial Campus - Click to enlarge           Water Fountain on NC State’s Centennial Campus - Click to enlarge

Bridge on NC State’s Centennial Campus - Click to enlarge           Bridge on NC State’s Centennial Campus - Click to enlarge           College of Textiles on NC State’s Centennial Campus - Click to enlarge


Final Thoughts

The small snowfall in Raleigh NC on 19 January 2005 will be long remembered for the unprecedented high impact it had on the local community. It is an event with as many sociological implications as meteorological.

Regarding the meteorological implications, the forecast calling for only a trace of snow to no more than a dusting (i.e., less than an inch) was correct for 22 of the 32 counties in Central North Carolina. The sudden increase in aerial coverage and intensity of the light snow as it quickly moved across the Raleigh area and points just to the north and east can be explained after the fact; however, consistently and accurately forecasting such a small scale (i.e., across several counties) change well in advance is still beyond the state of the science.

Forecasters were certainly well aware of the relatively high liquid equivalent to snow ratios associated with a northwest flow driven event, and they intuitively knew that cold ground and road surfaces would not contribute to melting. Still a number of factors had to come together to produce 8 hours of gridlock from so little snowfall. Chief among them was the occasional bursts of moderate snow. Though the bursts occurred only several times during the 2.5 hour snowfall period, they did feature larger flakes which efficiently coated the roads. During the period of light snow, the very small and dry flakes had simply been blowing across the roads. Once the snow was on the roads, the impact of heavy traffic was realized. Vehicles clogging the primary roads in a mass exodus of the city compacted the light coating into an icy composition which lead to numerous accidents and more gridlock.

Perhaps, the bottom line for forecasters is that a marginal snow event such as this one, calls for more than meteorological considerations. If factors are combining to present a threat of high impact from even a small amount of snow, then additional consideration can be given to issuing a forecast that accounts for that potential. Those factors include light snow with occasional moderate bursts moving across a large metropolitan area during the afternoon of a business day, preceded by very cold temperatures, and associated with a northwest flow driven event where liquid equivalent to snow ratios can be high. The only ingredient needed to again turn a ½ - 1 inch snowfall into traffic gridlock is the human response to the fallen snow.



Case Study Team

Kermit Keeter
Phillip Badgett
Michael Brennan
Gail Hartfield
Jeff Orrock
Douglas Schneider
Scott Sharp
Michael Strickler
Jonathan Blaes


For questions regarding the web site, please contact Jonathan Blaes.


  • NWS Disclaimer.