Verification of tropical cyclones (TC) wind structure forecasts from global NWP models and ensemble prediction systems (EPSs)

Abstract

Forecasting wind structure of tropical cyclone (TC) is vital in assessment of impact due to high winds using Numerical Weather Prediction (NWP) model. The usual verification technique on TC wind structure forecasts are based on grid-to-grid comparisons between forecast field and the actual field. However, precision of traditional verification measures is easily affected by small scale errors and thus cannot well discriminate the accuracy or effectiveness of NWP model forecast. In this study, the Method for Object-Based Diagnostic Evaluation (MODE), which has been widely adopted in verifying precipitation fields, is utilized in TC's wind field verification for the first time. The TC wind field forecast of deterministic NWP model and Ensemble Prediction System (EPS) of the European Centre for Medium-Range Weather Forecasts (ECMWF) over the western North Pacific and the South China Sea in 2020 were evaluated. A MODE score of 0.5 is used as a threshold value to represent a skillful (or good) forecast. It is found that the R34 (radius of 34 knots) wind field structure forecasts within 72 h are good regardless of DET or EPS. The performance of R50 and R64 is slightly worse but the R50 forecasts within 48 h remain good, with MODE exceeded 0.5. The R64 forecast within 48 h are worth for reference as well with MODE of around 0.5. This study states that the TC wind field structure forecast by ECMWF is skillful for TCs over the western North Pacific and the South China Sea.

Keywords

• Verification
• Tropical cyclones wind structure forecasts
• Numerical weather prediction models
• Ensemble prediction system