Short-Term Wind Forecasts for Aviation

Winds are a leading cause of air traffic delay due to their impact on aircraft spacing and runway use, yet air traffic management lacks adequate short-term real-time wind forecasts and wind-based decision support. In response, MIT Lincoln Laboratory is developing technology to generate improved short-term (0-1 hour) wind forecasts through data fusion and machine learning with the goal of providing an initial platform where wind forecast products can be developed, assessed, and demonstrated. A forecast system has been developed that combines numerical model forecasts (e.g., High-Resolution Rapid Refresh), heuristic advection wind forecasts, and wind observations from surface stations as well as aircraft observations from Mode-S radar surveillance. The data sources are combined using Gaussian Process Regression, which is a non-parametric Bayesian model that provides forecasts as well as uncertainty information.

This talk will describe an initial proof-of-concept system that produces wind forecasts at airports and along flight paths, potentially useful for compression events. Performance statistics from operationally significant cases will be presented, and potential uses in air traffic operations will also be discussed.

*DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited.
This material is based upon work supported by the United States Air Force under Air Force Contract No. FA8702-15-D-0001. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the United States Air Force.