Using in situ eddy dissipation rate (EDR) observations for turbulence forecast verification

Turbulence forecasting algorithms are being developed by the FAA Aviation Weather Research Program (AWRP) Turbulence Product Development Team (TPDT). Because forecast verification has a critical role to play in the development of new forecast products, it is important to use the best observations possible for their evaluation. Currently, the verification of turbulence forecasts is based on pilot reports (PIREPs). Unfortunately, PIREPs are sporadic in space and time, provide only subjective measure of the aircraft's response to turbulence, and include relatively few negative reports. These features make one of the most important parts of the verification process, matching observations and forecasts, somewhat unreliable. Automated observations of turbulence conditions will soon become operationally available in adequate numbers for use in verification studies. The in situ turbulence algorithm that relates eddy dissipation rate (EDR) to aircraft vertical acceleration will provide data that should be very valuable for evaluating current and new systems for forecasting turbulence conditions.

This study investigates statistical characteristics of the in situ EDR data based on the latest quasi-real-time quality controlled observations. During a two-month sample period in January – February 2005, more than two million EDR observations, but only about thirty thousand turbulence PIREPs are available. This study also introduces a possible method of using EDR observations for turbulence forecast verification. The Graphical Turbulence Guidance 2 (GTG2) is evaluated using both PIREPs and EDR observations. This is the first time a turbulence forecast product has been verified by these two different and independent observational datasets.