Determination of the Primary Diagnostics for the CAT (Clear-Air Turbulence) Forecast in Korea

In order to improve the CAT (clear-air turbulence) forecasts of KITFA (Korean Integrated Turbulence Forecasting Algorithm), analyses and forecast fields of 12 CAT diagnostics included in KITFA, in which 5-㎞ RDAPS (Regional Data Assimilation and Prediction System) forecasts routinely ingested, are analyzed in comparison with PIREPs reported in Korea for Spring 2002. The ROC (Receiver Operating Characteristics) curves shown as the relationship between 1-PODn and PODy represent that NGM1, vertical shear (VWS), and the squared vorticy (Vort²) suggest somewhat meaningful statistics for detecting the turbulence potential area in the analysis fields. The areas under the ROC curves suggested that the significance of the CAT forecasts for the 12 indices decreased with lead times. Other statistical values such as FAR (False Alarm Ratio), Critical Success Index (CSI), and TSS (True Skill Statistics) for some specified thresholds of the indices suggest more validated statistics for the performance of the turbulence forecasts. Although the statistical results obtained by matching turbulence forecasts of the CAT diagnostics to the observations can be somewhat uncertain because of a limitation of the NWP model forecast and insufficient PIREPs, this intercomparison exercise contributes not only for the improvement of the CAT forecast, but also for the robustness of the verification methods for the KITFA.