An exploration of blending techniques for use in production of automated forecasts of ceiling height and visibility

The forecasting of adverse ceiling and visibility (C&V) conditions continues to be a difficult challenge that impacts the safety and operations of the aviation community. Resources used in forecasting C&V conditions include numerical weather prediction models, MOS guidance, METAR observations, GOES satellite data, rule-based forecasts and others. The value of forecast guidance from these resources can vary widely depending upon forecast length, synoptic situation, season, location and other factors. Thus, to formulate an optimum automated forecast, the selection and blending of input forecasts and the use of related forecast aids must be guided by the specific strengths of each input component.

This paper will outline the steps taken by the National Ceiling and Visibility Product Development Team (NCV PDT) to explore means to selectively blend relevant observational and numerical forecast information for use in an automated C&V forecast system producing forecasts out to 12 hours. This research is being carried out under funding from the FAA's Aviation Weather Research Program.