This paper describes how operational radar, satellite and lightning data may be used in conjunction with numerical weather model data to provide remote detection and diagnosis of atmospheric turbulence in and around thunderstorms. In-cloud turbulence is measured with the NEXRAD Turbulence Detection Algorithm (NTDA) using extensively quality controlled, ground-based Doppler radar data. A real-time demonstration of the NTDA includes generation of a 3-D turbulence mosaic covering the CONUS east of the Rocky Mountains, a web-based display, and experimental uplinks of turbulence maps to en-route commercial aircraft. Near-cloud turbulence is inferred from thunderstorm morphology, intensity, growth rate and environment data provided by (1) satellite radiance measurements, rates of change, winds, and other derived features, (2) lightning strike measurements, (3) radar reflectivity measurements and (4) weather model data. These are combined via a machine learning technique trained using a database of in situ turbulence measurements from commercial aircraft to create a predictive model. This new capability is being developed under FAA and NASA funding to enhance current U.S. and international turbulence decision support systems, allowing rapid-update, high resolution,comprehensive assessments of atmospheric turbulence hazards for use by pilots, dispatchers, and air traffic controllers. It will also contribute to the comprehensive 4-D weather information database for NextGen.
