January 20, 2007 Non-Snow Event

The NWS in Tulsa issued a Winter Storm Watch early in the morning on January 18th, based on a forecast for snowfall amounts of greater than 4 inches and/or sleet accumulations greater then 1/2 inch over all of Eastern Oklahoma and Northwest Arkansas. Winter Storm Warnings were issued on Friday afternoon for the northern half of the watch. Heavy Snow was expected generally north of a line from Eufaula to Fayetteville mainly during the day Saturday.


On Saturday morning, forecasters noted that a layer of warm air off the surface was not being eroded away as expected and the rain that was falling may not change to snow. During the day it became evident that the rain/snow line would remain farther north then was indicated in earlier forecasts and warnings were cancelled. Measurable snow fell near the Kansas border with rain elsewere.


The forecast of the fate of the warm layer around 6000 feet above ground seems to be the key to the forecast of this event. Point Temperature/Dewpoint Profiles (Soundings) indicated a better chance for snow then larger scale model fields (850-700mb Thickness) did. The NAM(WRF) model was the model of choice during the event and its data is shown below. GFS model data is also shown in some images for comparison.


Observed Rain/Snow Radar Image with Temperature profiles inserted. The snow profile shows no melting layer. The rain profile has a melting layer. The forecast of this melting layer determined the rain vs. snow forecast for this event. Model and human forecasts had the rain/snow line too far south.


The evolution of the forecast of the model sounding at Tulsa and the melting layer forecast are shown below.


The observed sounding and thickness valid Saturday evening are at the bottom of this table. The Day 3 (72 Hr) forecast is at the top followed by successive forecasts.

Tulsa forecast Temperature Profile

Melting Layer Forecast (850-700 mb thickness)

Melting favored in red shades and south of there (rain area). Little melting in blue shades and north (snow area)




72 Hr


Fcst made






Snow indicated at Tulsa



60 Hr

Fcst made



Slight warm nose on sounding, but not enough to melt falling snow. Thickness leaning towards more melting however (red shade). Snow/Sleet best forecast.




48 Hr


Fcst made




Deeper melting layer indicated on sounding and thickness shows good melting over tulsa county. Rain/Sleet best forecast.




36 Hr


Fcst made




Sounding indicates no melting layer and thickness only slight melting. Snow best forecast.




24 Hr


Fcst made




Slightly cooler and less melting shown then previous forecast. Snow best forecast.




18 Hr


Fcst made




No change from previous forecast. Snow best forecast.




12 Hr


Fcst made




Sounding warmer and thickness indicating some melting. Rain/Snow best forecast.




6 Hr


Fcst made




Sounding warmer and a tiny area above freezing around 800 mb. Thickness indicates good melting. Snow/Rain best forecast.




00HR Analysis (Observed)




850-700mb layer was warmer and the melting layer more pronounced then in any forecast.


NAM model forecasts of 800 mb 0 deg isotherm (nose of melting level)

The 00 Hr line is considered the observed position of the isotherm at 01/21/00Z, the 3HR RUC is the RUC 3 hour forecast, the 6HR is the NAM 6 Hour Fcst, 12HR is NAM 12 Hour Fcst, 18 HR is NAM 18 Hour Fcst. The model cycle and forecast hours are also shown in the legend at the bottom of the image.


A Time Series of the GFS and NAM 850-700 MB thickness at TUL which begin with the 78 Hour forecast (which was made on the 17th at 18z) and ends with the 6 Hour forecast (which was made on the 20th at 18z).

The thresholds for melting in this layer are indicated by the red and blue lines. The GFS is in green and the NAM is orange.

At 78,72 and 60 hours, the GFS was forecasting no melting layer (snow), while the NAM had values in the mid to lower range (some melting, rain/snow mix). After 60 hours, both models showed warming. From 48 to 24 hours the NAM showed cooling (better chance for snow), but the GFS was warmer (better chance for rain/snow mix). Both the NAM and GFS were in good agreement in their 18,12 and 6 hour forecasts, which trended towards the solution of an "ALL MELTED" melting layer (mainly rain).


The 1/21/07 00z OUN Raob Sounding, the RUC 00HR Analysis and LAPS Analysis are shown below.


The melting layer is not present in the Raob, but there is a thin layer above freezing in the RUC and LAPS soundings.


Snow was falling near OUN at 23z and 00z, but the precipitation was ending.



USA.gov is the U.S. government's official web portal to all federal, state and local government web resources and services.