High Resolution Model Forecasts Predict Convective Mode for May 3, 2009 Event

SPoRT participates in the annual Spring Experiments that take place at the National Severe Storms Laboratory in Norman, Oklahoma.  One component of the experiment examines the usefulness of high resolution, cloud-permitting forecasts with model grid spacings of 4 km or less.  Rather than predicting the bulk effects of clouds, these models permit a forecast of cloud characteristics, such as quantities of rain, snow, hail and other cloud constituents.  SPoRT is examining the value of unique NASA capabilities and contributions, such as land information systems to enhance the quality of soil moisture estimations (key to interactions in the lowest portion of the atmosphere), as well as the realism of cloud parameterizations, as confirmed by radar characteristics from CloudSat and ground-based systems.

For this event, the model forecast was 2 to 3 hours slow in that it had not progressed the thunderstorms eastward as fast as nature’s solution.  This may have happened for a variety of reasons related to model physics and data used to initialize the forecast.  However, the model performed admirably by advertising to forecasters that a bow echo should be expected across southern Mississippi.  In addition, hourly maximum forecast fields (developed by SPoRT member Scott Dembek) track the most intense values within each hourly segment of the forecast and identifies a period in which the model predicts severe straightline winds at the surface.  This forecast provides a 12-18 hour lead time, confirming to forecasters that the expected convective mode (or type of storm) would be bowing, linear segments capable of damaging winds.  The Storm Prediction Center issued a moderate risk outlook for the area, highlighting damaging winds as the primary threat.  Below are some images demonstrating a comparison between the model forecast and reality.

Simulated radar reflectivity for the WRF model forecast, valid at 19 UTC on 3 May 2009. Although radar imagery below indicates that the model was "too slow" -- the simulated convective line is lagging to the west -- the bow shaped nature is apparent.

Hourly maximum surface winds predicted by the model, given in meters per second. Note that the reflectivity above is a depiction at exactly 19 UTC, while the wind speed data represents the maximum value between 18-19 UTC, or as the line moved through. Areas of greens indicate winds beyond 30 meters per second, or 65 miles per hour, while severe thunderstorm wind criteria is 58 miles per hour.

Base reflectivity from the Jackson, MS radar with active warnings and example text overlaid, highlighting the threat of damaging straightline winds as predicted by the WRF model forecast.

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