Probabilistic Lightning Forecast Guidance for Aviation

To provide tactical support for air traffic management, more graphical and probabilistic weather forecast guidance is needed at higher resolution and shorter time periods, along with more frequent updates. Traditional Model Output Statistics (MOS) products support aviation weather operations by providing site-specific forecast guidance for wind, ceiling height, sky cover, visibility, obstruction to vision, and thunderstorms for over 1500 locations, including most airports. The equations to predict the probability of thunderstorms are developed for the contiguous U.S. (CONUS) by using observed lightning data and output from the numerical weather prediction models. Observations of cloud-to-ground lightning data from the National Lightning Detection Network are used to define thunderstorm events for the MOS system. Relative frequencies of lightning, generated from the lightning observations, are available as potential predictors in the statistical forecast system. The relative frequencies are also valuable as a tool for comparing the accuracy of the forecast guidance to forecasts based on climate.

The Meteorological Development Laboratory currently produces thunderstorm probability forecast guidance on a 40-km grid for 6-, 12-, and 24-h periods using output from both the Global Forecast System (GFS) and the Eta model. To meet the needs of the aviation community, our latest development concentrates on issues related to shorter temporal and smaller spatial resolutions. We are developing a set of GFS-based probability forecast equations for 3-h time periods on a 20-km grid over the CONUS. The forecast projections are valid for 6-9, 9-12, 12-15, … and 33-36 hours in advance, and are generated 4 times daily. Statistical techniques are also used to develop Localized Aviation MOS Program (LAMP) products, which are updates to the MOS products and provide forecast guidance more frequently for shorter time periods. Development is now underway to produce LAMP thunderstorm products, which will be derived from the MOS guidance, observations of lightning and radar data. The LAMP thunderstorm probabilities will be available for 2-h time periods, including 0-2, 2-4, 4-6, and 6-8 hours in advance, and will update more frequently.

Probabilistic forecasts will become an increasingly valuable tool for decision makers. We believe the increase in temporal and spatial resolution of our lightning-based thunderstorm guidance will assist the aviation industry in making critical decisions in the earliest forecast periods. With feedback from the aviation weather community, the thunderstorm probability guidance can be tailored to meet the needs of the users. As an example, we can derive categorical forecasts from the probability guidance if there is a need. In this paper, we present objective verification scores of the MOS thunderstorm forecasts, show a sample of the lightning relative frequencies, and present an example of the guidance during a thunderstorm outbreak.