Cloud-to-ground (CG) lightning is one of the leading causes of weather related fatalities in the United States (U.S.). In heavily forested areas such as Alaska, CG lightning also is a major contributor to the initiation of forest fires. Graphical, probabilistic guidance for thunderstorms over Alaska would allow the fire weather community and other users to better assess the CG lightning threat, and thereby aid in the protection of life and property.

For many years, National Weather Service (NWS) forecasters have used Model Output Statistics (MOS) guidance produced by the Meteorological Development Laboratory (MDL) as an aid in generating text forecast products issued to the user community. However, forecasters now are required to produce forecasts on a high-resolution grid in support of the National Digital Forecast Database (NDFD). Recently, updated thunderstorm probability guidance based on output from the Global Forecast System (GFS) was implemented for the contiguous U.S. to satisfy NDFD grid requirements. Efforts now are underway to expand this new gridded thunderstorm guidance system to cover the state of Alaska.

Observations of CG lightning from the Bureau of Land Management (BLM) network and GFS model output are used to develop objective probabilistic thunderstorm guidance for Alaska and adjacent waters. Linear screening regression is used to statistically relate lightning data to GFS model forecast fields on a 40 x 40 km grid. Generalized operator equations are developed giving the probability of thunderstorms at 3-h, 6-h, 12-h, and 24-h intervals for each GFS model cycle. Seventeen warm seasons of lightning observations are used to generate relative frequencies for each month and time period. The relative frequencies are used to study the climatological characteristics of lightning over Alaska, and are available as potential predictors in the statistical forecast system.

A vast majority of lighting strikes in Alaska occur during the warm season. Due to limited data availability, and the rareness of the event in the cool season, the MOS thunderstorm guidance is developed for May through September. The extended abstract will describe the development of the product, and will also present some objective verification scores for an independent data sample. Examples of the guidance also will be presented. It is anticipated that the Alaska thunderstorm guidance will be available for the 2008 warm season.