Real-time LIS shows detailed snow water equivalent of recent SE U.S. snowstorm

On 29-30 January 2010, a major storm impacted the Southeastern U.S. with significant amounts of frozen precipitation.  Heavy snow fell from Oklahoma across Arkansas, Tennessee, and into North Carolina and Virginia.  Snowfall of over a foot occurred on the northern fringe of the storm, while freezing rain and sleet affected portions of the Tennessee Valley.  Mainly rain occurred further south.  The morning surface analysis from the daily weather map series is shown in Figure 1.   The 24-hour accumulated precipitation ending Saturday 1200 UTC 30 January is given in Figure 2.

Figure 3 displays a sample snow water equivalent image from the real-time Land Information System (LIS) run at SPoRT.   The graphic depicts widespread areas of > 0.25″ liquid equivalent of the snowpack extending in a swath from Oklahoma into the Carolinas.  The LIS shows that portions of NE Arkansas and western Tennessee north of Memphis had the highest water equivalent amounts of 1-2″ or more.   This type of output could aid forecasters in diagnosing the risks for flooding during a rapid snowmelt situation.

A little background on the LIS:  The Land Information System is a NASA-developed software framework that can run a variety of Land Surface Models (LSMs) with a user-specified atmospheric input dataset.  The atmospheric input or “forcings” provide the variables required to integrate the LSMs.  In the SPoRT implementation of LIS, the Noah LSM (same scheme as used in the operational NAM model) is run at 3-km grid spacing 4x per day on the domain shown in Figure 3.   The atmospheric forcings come from a combination of short-term forecasts from the 0.47 degree Global Data Assimilation System, hourly 1/8th degree North American Land Data Assimilation System analyses created at NCEP, and hourly 4.8 km radar+gauge Stage IV precipitation analyses.  The LIS is currently configured to output hourly GRIB-1 files that are posted to an anonymous ftp site.  The data can be used for local modeling applications, and could also be used for diagnostic purposes.

Figure 1. Daily surface weather map at 1200 UTC 30 JAN 2010

Figure 2. QPF analysis for 24 hours ending 1200 UTC 30 JAN 2010

Figure 3. Land Information System (LIS) snow water equivalent over SE U.S. in inches.

3 thoughts on “Real-time LIS shows detailed snow water equivalent of recent SE U.S. snowstorm

  1. Interesting analysis but it is unclear what input data is driving the spatial distribution and how snow water equiv versus rain water is determined.

  2. The 4.8-km NCEP Stage IV radar+gauge hourly analyses drive the quantitative precipitation amounts for the LSM within the CONUS. The precipitaton is then partitioned into rain versus snow versus freezing rain, based on whether the 2-m/surface temperatures are above or below freezing. The assumption in the LIS/Noah LSM is that if the 2-m temperature and surface (skin) temperature are both below freezing, then snow is occurring. If the 2-m Temp is above freezing, but the surface/skin Temp is below freezing, then freezing rain is assumed. In both instances of snow and freezing rain, the precipitation adds to the snow water equivalent field. To summarize, the snow water equivalent includes both the accumulated snow depth and ice accumulation.

    Good question!

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s