Snow Forecasting Experiment??

Strange considerations can crop up in the SFE. In previous years we have forecasted in areas of low radar coverage such as the mountain west, determined which side of the U.S.-Mexico border a storm would form on, and dealt with the severity of convection coming onshore from the Gulf. However, remnants of last weekend's storm threw a highly unusual wrinkle in the forecast....

That's right, a large swath of snow from last weekend's winter storm was still lingering just north of our forecast area of concern yesterday. The scenario we were envisioning was an MCS over northern Oklahoma, initiating in the Panhandle and propagating eastward late in the period. This scenario occurred in many of the high-resolution CAM ensembles, including the SSEO, HREFv2, and the NCAR ensemble. The HRRRE and the GSI EnKF ensemble produced by OU were outliers yesterday, showing less organized storms at 0600 UTC. What actually happened? The reports tell the tale:

Multiple clusters of storms formed, producing first hail (brown) and then wind (pink) reports while growing upscale and moving east across Oklahoma:

As you can see in the 04 UTC reflectivity above, strong storms were present across eastern Kansas, near that snowpack. While this snow was melting quickly (as seen in the gif below), as of this evening (3 May 2017) some snow is still present in the satellite imagery, meaning that it is entirely possible that some Kansas residents had a thunderstorm with snow on the ground.

While many CAMs picked up on the snowpack, the most evident signal we saw was during the NEWS-e activity in the afternoon. The NEWS-e is the NSSL Experimental Warn-on Forecast System for Ensembles, and is a preliminary Warn-on-Forecast system that is being tested for short-range forecasting. The NEWS-e provides high temporal and spatial resolution forecasts (forecast fields are given at five minute intervals and the ensemble is run at 3 km on a small domain chosen in the morning) and assimilates data every fifteen minutes. The desk lead forecaster on the Innovation Desk (formerly the "Total Severe" or NSSL desk) issues forecasts at 2030 UTC (3:30 PM CDT) along with participants for 21Z-22Z and 22Z-23Z, and then updates those forecasts based on a new run of the NEWS-e.

As the system we were forecasting was later than the hourly forecasts using the NEWS-e, participants had plenty of time to explore the fields provided by the NEWS-e. And sure enough, when we looked at the temperature field, the snowpack showed up quite nicely:

Although the main impact of the snowpack on our forecasts was to provide a northern and western boundary to our probabilistic forecasts of severe convective hazards, it's not every day that you have to consider snowfall when forecasting severe convective thunderstorms in May.